redonkable
/
alistair23-linux
1
0
Fork 0
Code Releases Activity
alistair23-linux/drivers/scsi/ipr.c

10868 lines
301 KiB
C
Raw Normal View History

treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 156 Based on 1 normalized pattern(s): this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version this program is distributed in the hope that it will be useful but without any warranty without even the implied warranty of merchantability or fitness for a particular purpose see the gnu general public license for more details you should have received a copy of the gnu general public license along with this program if not write to the free software foundation inc 59 temple place suite 330 boston ma 02111 1307 usa extracted by the scancode license scanner the SPDX license identifier GPL-2.0-or-later has been chosen to replace the boilerplate/reference in 1334 file(s). Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Allison Randal <allison@lohutok.net> Reviewed-by: Richard Fontana <rfontana@redhat.com> Cc: linux-spdx@vger.kernel.org Link: https://lkml.kernel.org/r/20190527070033.113240726@linutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-05-27 00:55:05 -06:00
// SPDX-License-Identifier: GPL-2.0-or-later
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/*
* ipr.c -- driver for IBM Power Linux RAID adapters
*
* Written By: Brian King <brking@us.ibm.com>, IBM Corporation
*
* Copyright (C) 2003, 2004 IBM Corporation
*/
/*
* Notes:
*
* This driver is used to control the following SCSI adapters:
*
* IBM iSeries: 5702, 5703, 2780, 5709, 570A, 570B
*
* IBM pSeries: PCI-X Dual Channel Ultra 320 SCSI RAID Adapter
* PCI-X Dual Channel Ultra 320 SCSI Adapter
* PCI-X Dual Channel Ultra 320 SCSI RAID Enablement Card
* Embedded SCSI adapter on p615 and p655 systems
*
* Supported Hardware Features:
* - Ultra 320 SCSI controller
* - PCI-X host interface
* - Embedded PowerPC RISC Processor and Hardware XOR DMA Engine
* - Non-Volatile Write Cache
* - Supports attachment of non-RAID disks, tape, and optical devices
* - RAID Levels 0, 5, 10
* - Hot spare
* - Background Parity Checking
* - Background Data Scrubbing
* - Ability to increase the capacity of an existing RAID 5 disk array
* by adding disks
*
* Driver Features:
* - Tagged command queuing
* - Adapter microcode download
* - PCI hot plug
* - SCSI device hot plug
*
*/
#include <linux/fs.h>
#include <linux/init.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/kernel.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 02:04:11 -06:00
#include <linux/slab.h>
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
#include <linux/vmalloc.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
#include <linux/ioport.h>
#include <linux/delay.h>
#include <linux/pci.h>
#include <linux/wait.h>
#include <linux/spinlock.h>
#include <linux/sched.h>
#include <linux/interrupt.h>
#include <linux/blkdev.h>
#include <linux/firmware.h>
#include <linux/module.h>
#include <linux/moduleparam.h>
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
#include <linux/libata.h>
[SCSI] ipr: Fix HDIO_GET_IDENTITY oops for SATA devices Currently, ipr does not support HDIO_GET_IDENTITY to SATA devices. An oops occurs if userspace attempts to send the command. Since hald issues the command, ensure we fail the ioctl in ipr. This is a temporary solution to the oops. Once the ipr libata EH conversion is upstream, ipr will fully support HDIO_GET_IDENTITY. Tested-by: Milton Miller <miltonm@bga.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-11 12:37:50 -06:00
#include <linux/hdreg.h>
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
#include <linux/reboot.h>
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
#include <linux/stringify.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
#include <asm/io.h>
#include <asm/irq.h>
#include <asm/processor.h>
#include <scsi/scsi.h>
#include <scsi/scsi_host.h>
#include <scsi/scsi_tcq.h>
#include <scsi/scsi_eh.h>
#include <scsi/scsi_cmnd.h>
#include "ipr.h"
/*
* Global Data
*/
[SCSI] ipr: use LIST_HEAD instead of LIST_HEAD_INIT Signed-off-by: Denis Cheng <crquan@gmail.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2007-12-13 17:14:27 -07:00
static LIST_HEAD(ipr_ioa_head);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
static unsigned int ipr_log_level = IPR_DEFAULT_LOG_LEVEL;
static unsigned int ipr_max_speed = 1;
static int ipr_testmode = 0;
static unsigned int ipr_fastfail = 0;
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
static unsigned int ipr_transop_timeout = 0;
[SCSI] ipr: Runtime debugging options Make some compile time debugging options runtime module options. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:34 -07:00
static unsigned int ipr_debug = 0;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
static unsigned int ipr_max_devs = IPR_DEFAULT_SIS64_DEVS;
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
static unsigned int ipr_dual_ioa_raid = 1;
ipr: Increase MSIX vectors number Increase MSIX vectors from 2 to 16 by default. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-07-12 15:02:08 -06:00
static unsigned int ipr_number_of_msix = 16;
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
static unsigned int ipr_fast_reboot;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
static DEFINE_SPINLOCK(ipr_driver_lock);
/* This table describes the differences between DMA controller chips */
static const struct ipr_chip_cfg_t ipr_chip_cfg[] = {
[SCSI] ipr: PCI IDs for new SAS adapters Adds PCI IDs for some new ipr SAS adapters. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:10 -07:00
{ /* Gemstone, Citrine, Obsidian, and Obsidian-E */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.mailbox = 0x0042C,
[SCSI] ipr: Increase max concurrent oustanding commands Increase the total number of max concurrent outstanding commands for the most recent family of adapters in order to improve overall adapter performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:10 -06:00
.max_cmds = 100,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.cache_line_size = 0x20,
[SCSI] ipr: Remove unnecessary interrupt clearing on new adapters The latest ipr hardware no longer requires the driver to issue any MMIOs to clear the interrupt so remove this to optimize performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:08 -06:00
.clear_isr = 1,
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
.iopoll_weight = 0,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
.set_interrupt_mask_reg = 0x0022C,
.clr_interrupt_mask_reg = 0x00230,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.clr_interrupt_mask_reg32 = 0x00230,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.sense_interrupt_mask_reg = 0x0022C,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_interrupt_mask_reg32 = 0x0022C,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.clr_interrupt_reg = 0x00228,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.clr_interrupt_reg32 = 0x00228,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.sense_interrupt_reg = 0x00224,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_interrupt_reg32 = 0x00224,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.ioarrin_reg = 0x00404,
.sense_uproc_interrupt_reg = 0x00214,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_uproc_interrupt_reg32 = 0x00214,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.set_uproc_interrupt_reg = 0x00214,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.set_uproc_interrupt_reg32 = 0x00214,
.clr_uproc_interrupt_reg = 0x00218,
.clr_uproc_interrupt_reg32 = 0x00218
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
},
{ /* Snipe and Scamp */
.mailbox = 0x0052C,
[SCSI] ipr: Increase max concurrent oustanding commands Increase the total number of max concurrent outstanding commands for the most recent family of adapters in order to improve overall adapter performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:10 -06:00
.max_cmds = 100,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.cache_line_size = 0x20,
[SCSI] ipr: Remove unnecessary interrupt clearing on new adapters The latest ipr hardware no longer requires the driver to issue any MMIOs to clear the interrupt so remove this to optimize performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:08 -06:00
.clear_isr = 1,
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
.iopoll_weight = 0,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
.set_interrupt_mask_reg = 0x00288,
.clr_interrupt_mask_reg = 0x0028C,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.clr_interrupt_mask_reg32 = 0x0028C,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.sense_interrupt_mask_reg = 0x00288,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_interrupt_mask_reg32 = 0x00288,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.clr_interrupt_reg = 0x00284,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.clr_interrupt_reg32 = 0x00284,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.sense_interrupt_reg = 0x00280,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_interrupt_reg32 = 0x00280,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.ioarrin_reg = 0x00504,
.sense_uproc_interrupt_reg = 0x00290,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_uproc_interrupt_reg32 = 0x00290,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.set_uproc_interrupt_reg = 0x00290,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.set_uproc_interrupt_reg32 = 0x00290,
.clr_uproc_interrupt_reg = 0x00294,
.clr_uproc_interrupt_reg32 = 0x00294
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
},
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
{ /* CRoC */
[SCSI] ipr: fix mailbox register definition and add a delay before reading The definition for the mailbox register for new adapters was incorrect. The value has been updated to the correct offset. After an adapter reset, the mailbox register on the new adapters takes a number of seconds to stabilize. A delay has been added before reading the register. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-11-04 10:36:16 -06:00
.mailbox = 0x00044,
[SCSI] ipr: Increase max concurrent oustanding commands Increase the total number of max concurrent outstanding commands for the most recent family of adapters in order to improve overall adapter performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:10 -06:00
.max_cmds = 1000,
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
.cache_line_size = 0x20,
[SCSI] ipr: Remove unnecessary interrupt clearing on new adapters The latest ipr hardware no longer requires the driver to issue any MMIOs to clear the interrupt so remove this to optimize performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:08 -06:00
.clear_isr = 0,
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
.iopoll_weight = 64,
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
{
.set_interrupt_mask_reg = 0x00010,
.clr_interrupt_mask_reg = 0x00018,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.clr_interrupt_mask_reg32 = 0x0001C,
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
.sense_interrupt_mask_reg = 0x00010,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_interrupt_mask_reg32 = 0x00014,
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
.clr_interrupt_reg = 0x00008,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.clr_interrupt_reg32 = 0x0000C,
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
.sense_interrupt_reg = 0x00000,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_interrupt_reg32 = 0x00004,
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
.ioarrin_reg = 0x00070,
.sense_uproc_interrupt_reg = 0x00020,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.sense_uproc_interrupt_reg32 = 0x00024,
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
.set_uproc_interrupt_reg = 0x00020,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.set_uproc_interrupt_reg32 = 0x00024,
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
.clr_uproc_interrupt_reg = 0x00028,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
.clr_uproc_interrupt_reg32 = 0x0002C,
.init_feedback_reg = 0x0005C,
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
.dump_addr_reg = 0x00064,
[SCSI] ipr: add endian swap enablement for 64 bit adapters A change in the hardware design of the chip for the new adapters changes the default endianness of MMIO operations. This patch adds a register definition which when written to with a predefined value will change the endianness back to what the driver expects. This patch also fixes two problems found during testing. First, the first reserved field in the ipr_hostrcb64_fabirc_desc structure only reserved one byte. The correct amount to reserve is 2 bytes. Second, the reserved field of the ipr_hostrcb64_error structure only reserved 2 bytes. The correct amount to reserve is 16 bytes. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-04 11:26:50 -06:00
.dump_data_reg = 0x00068,
.endian_swap_reg = 0x00084
[SCSI] ipr: define new offsets to registers for the next generation chip This patch adds the entry to the ipr_chip_cfg array that defines the register offsets for the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:51 -07:00
}
},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
};
static const struct ipr_chip_t ipr_chip[] = {
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
{ PCI_VENDOR_ID_MYLEX, PCI_DEVICE_ID_IBM_GEMSTONE, false, IPR_SIS32, IPR_PCI_CFG, &ipr_chip_cfg[0] },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CITRINE, false, IPR_SIS32, IPR_PCI_CFG, &ipr_chip_cfg[0] },
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_OBSIDIAN, false, IPR_SIS32, IPR_PCI_CFG, &ipr_chip_cfg[0] },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN, false, IPR_SIS32, IPR_PCI_CFG, &ipr_chip_cfg[0] },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN_E, true, IPR_SIS32, IPR_PCI_CFG, &ipr_chip_cfg[0] },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_SNIPE, false, IPR_SIS32, IPR_PCI_CFG, &ipr_chip_cfg[1] },
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_SCAMP, false, IPR_SIS32, IPR_PCI_CFG, &ipr_chip_cfg[1] },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2, true, IPR_SIS64, IPR_MMIO, &ipr_chip_cfg[2] },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE, true, IPR_SIS64, IPR_MMIO, &ipr_chip_cfg[2] },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_RATTLESNAKE, true, IPR_SIS64, IPR_MMIO, &ipr_chip_cfg[2] }
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
};
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
static int ipr_max_bus_speeds[] = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
IPR_80MBs_SCSI_RATE, IPR_U160_SCSI_RATE, IPR_U320_SCSI_RATE
};
MODULE_AUTHOR("Brian King <brking@us.ibm.com>");
MODULE_DESCRIPTION("IBM Power RAID SCSI Adapter Driver");
module_param_named(max_speed, ipr_max_speed, uint, 0);
MODULE_PARM_DESC(max_speed, "Maximum bus speed (0-2). Default: 1=U160. Speeds: 0=80 MB/s, 1=U160, 2=U320");
module_param_named(log_level, ipr_log_level, uint, 0);
MODULE_PARM_DESC(log_level, "Set to 0 - 4 for increasing verbosity of device driver");
module_param_named(testmode, ipr_testmode, int, 0);
MODULE_PARM_DESC(testmode, "DANGEROUS!!! Allows unsupported configurations");
[SCSI] ipr: Expose debug and fastfail parameters Expose the debug and fastfail parameters to /sys/module/ipr/parameters such that they can be enabled/disabled at run time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-02-24 12:36:00 -07:00
module_param_named(fastfail, ipr_fastfail, int, S_IRUGO | S_IWUSR);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
MODULE_PARM_DESC(fastfail, "Reduce timeouts and retries");
module_param_named(transop_timeout, ipr_transop_timeout, int, 0);
MODULE_PARM_DESC(transop_timeout, "Time in seconds to wait for adapter to come operational (default: 300)");
[SCSI] ipr: Expose debug and fastfail parameters Expose the debug and fastfail parameters to /sys/module/ipr/parameters such that they can be enabled/disabled at run time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-02-24 12:36:00 -07:00
module_param_named(debug, ipr_debug, int, S_IRUGO | S_IWUSR);
[SCSI] ipr: Runtime debugging options Make some compile time debugging options runtime module options. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:34 -07:00
MODULE_PARM_DESC(debug, "Enable device driver debugging logging. Set to 1 to enable. (default: 0)");
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
module_param_named(dual_ioa_raid, ipr_dual_ioa_raid, int, 0);
MODULE_PARM_DESC(dual_ioa_raid, "Enable dual adapter RAID support. Set to 1 to enable. (default: 1)");
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
module_param_named(max_devs, ipr_max_devs, int, 0);
MODULE_PARM_DESC(max_devs, "Specify the maximum number of physical devices. "
"[Default=" __stringify(IPR_DEFAULT_SIS64_DEVS) "]");
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
module_param_named(number_of_msix, ipr_number_of_msix, int, 0);
ipr: Increase MSIX vectors number Increase MSIX vectors from 2 to 16 by default. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-07-12 15:02:08 -06:00
MODULE_PARM_DESC(number_of_msix, "Specify the number of MSIX interrupts to use on capable adapters (1 - 16). (default:16)");
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
module_param_named(fast_reboot, ipr_fast_reboot, int, S_IRUGO | S_IWUSR);
MODULE_PARM_DESC(fast_reboot, "Skip adapter shutdown during reboot. Set to 1 to enable. (default: 0)");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
MODULE_LICENSE("GPL");
MODULE_VERSION(IPR_DRIVER_VERSION);
/* A constant array of IOASCs/URCs/Error Messages */
static const
struct ipr_error_table_t ipr_error_table[] = {
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x00000000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"8155: An unknown error was received"},
{0x00330000, 0, 0,
"Soft underlength error"},
{0x005A0000, 0, 0,
"Command to be cancelled not found"},
{0x00808000, 0, 0,
"Qualified success"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01080000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFFE: Soft device bus error recovered by the IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01088100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4101: Soft device bus fabric error"},
[SCSI] ipr: add support for new IOASCs This patch adds support for new errors that can be received from adapters using the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:32 -07:00
{0x01100100, 0, IPR_DEFAULT_LOG_LEVEL,
"FFFC: Logical block guard error recovered by the device"},
{0x01100300, 0, IPR_DEFAULT_LOG_LEVEL,
"FFFC: Logical block reference tag error recovered by the device"},
{0x01108300, 0, IPR_DEFAULT_LOG_LEVEL,
"4171: Recovered scatter list tag / sequence number error"},
{0x01109000, 0, IPR_DEFAULT_LOG_LEVEL,
"FF3D: Recovered logical block CRC error on IOA to Host transfer"},
{0x01109200, 0, IPR_DEFAULT_LOG_LEVEL,
"4171: Recovered logical block sequence number error on IOA to Host transfer"},
{0x0110A000, 0, IPR_DEFAULT_LOG_LEVEL,
"FFFD: Recovered logical block reference tag error detected by the IOA"},
{0x0110A100, 0, IPR_DEFAULT_LOG_LEVEL,
"FFFD: Logical block guard error recovered by the IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01170600, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF9: Device sector reassign successful"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01170900, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF7: Media error recovered by device rewrite procedures"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01180200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"7001: IOA sector reassignment successful"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01180500, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF9: Soft media error. Sector reassignment recommended"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01180600, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF7: Media error recovered by IOA rewrite procedures"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01418000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FF3D: Soft PCI bus error recovered by the IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01440000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF6: Device hardware error recovered by the IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01448100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF6: Device hardware error recovered by the device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01448200, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FF3D: Soft IOA error recovered by the IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x01448300, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFFA: Undefined device response recovered by the IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x014A0000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF6: Device bus error, message or command phase"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x014A8000, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
"FFFE: Task Management Function failed"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x015D0000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF6: Failure prediction threshold exceeded"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x015D9200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"8009: Impending cache battery pack failure"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x02040100, 0, 0,
"Logical Unit in process of becoming ready"},
{0x02040200, 0, 0,
"Initializing command required"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x02040400, 0, 0,
"34FF: Disk device format in progress"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x02040C00, 0, 0,
"Logical unit not accessible, target port in unavailable state"},
[SCSI] ipr: Handle IOA reset request In ipr dual adapter configurations, the ipr adapter firmware may require an adapter reset for various reasons. The reset is requested by the adapter firmware logging an error with an IOASC of 0x02048000. Add support to log this error, and reset the adapter. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:12 -06:00
{0x02048000, 0, IPR_DEFAULT_LOG_LEVEL,
"9070: IOA requested reset"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x023F0000, 0, 0,
"Synchronization required"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x02408500, 0, 0,
"IOA microcode download required"},
{0x02408600, 0, 0,
"Device bus connection is prohibited by host"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x024E0000, 0, 0,
"No ready, IOA shutdown"},
{0x025A0000, 0, 0,
"Not ready, IOA has been shutdown"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x02670100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3020: Storage subsystem configuration error"},
{0x03110B00, 0, 0,
"FFF5: Medium error, data unreadable, recommend reassign"},
{0x03110C00, 0, 0,
"7000: Medium error, data unreadable, do not reassign"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x03310000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF3: Disk media format bad"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04050000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3002: Addressed device failed to respond to selection"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04080000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3100: Device bus error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04080100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3109: IOA timed out a device command"},
{0x04088000, 0, 0,
"3120: SCSI bus is not operational"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04088100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4100: Hard device bus fabric error"},
[SCSI] ipr: add support for new IOASCs This patch adds support for new errors that can be received from adapters using the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:32 -07:00
{0x04100100, 0, IPR_DEFAULT_LOG_LEVEL,
"310C: Logical block guard error detected by the device"},
{0x04100300, 0, IPR_DEFAULT_LOG_LEVEL,
"310C: Logical block reference tag error detected by the device"},
{0x04108300, 1, IPR_DEFAULT_LOG_LEVEL,
"4170: Scatter list tag / sequence number error"},
{0x04109000, 1, IPR_DEFAULT_LOG_LEVEL,
"8150: Logical block CRC error on IOA to Host transfer"},
{0x04109200, 1, IPR_DEFAULT_LOG_LEVEL,
"4170: Logical block sequence number error on IOA to Host transfer"},
{0x0410A000, 0, IPR_DEFAULT_LOG_LEVEL,
"310D: Logical block reference tag error detected by the IOA"},
{0x0410A100, 0, IPR_DEFAULT_LOG_LEVEL,
"310D: Logical block guard error detected by the IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04118000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9000: IOA reserved area data check"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04118100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9001: IOA reserved area invalid data pattern"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04118200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9002: IOA reserved area LRC error"},
[SCSI] ipr: add support for new IOASCs This patch adds support for new errors that can be received from adapters using the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:32 -07:00
{0x04118300, 1, IPR_DEFAULT_LOG_LEVEL,
"Hardware Error, IOA metadata access error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04320000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"102E: Out of alternate sectors for disk storage"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04330000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF4: Data transfer underlength error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04338000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF4: Data transfer overlength error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x043E0100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3400: Logical unit failure"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04408500, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF4: Device microcode is corrupt"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04418000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"8150: PCI bus error"},
{0x04430000, 1, 0,
"Unsupported device bus message received"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04440000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF4: Disk device problem"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04448200, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"8150: Permanent IOA failure"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04448300, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3010: Disk device returned wrong response to IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04448400, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"8151: IOA microcode error"},
{0x04448500, 0, 0,
"Device bus status error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04448600, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"8157: IOA error requiring IOA reset to recover"},
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
{0x04448700, 0, 0,
"ATA device status error"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x04490000, 0, 0,
"Message reject received from the device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04449200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"8008: A permanent cache battery pack failure occurred"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x0444A000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9090: Disk unit has been modified after the last known status"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x0444A200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9081: IOA detected device error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x0444A300, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9082: IOA detected device error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x044A0000, 1, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3110: Device bus error, message or command phase"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x044A8000, 1, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
"3110: SAS Command / Task Management Function failed"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04670400, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9091: Incorrect hardware configuration change has been detected"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04678000, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
"9073: Invalid multi-adapter configuration"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04678100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4010: Incorrect connection between cascaded expanders"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04678200, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4020: Connections exceed IOA design limits"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04678300, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4030: Incorrect multipath connection"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x04679000, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4110: Unsupported enclosure function"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x04679800, 0, IPR_DEFAULT_LOG_LEVEL,
"4120: SAS cable VPD cannot be read"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x046E0000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFF4: Command to logical unit failed"},
{0x05240000, 1, 0,
"Illegal request, invalid request type or request packet"},
{0x05250000, 0, 0,
"Illegal request, invalid resource handle"},
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
{0x05258000, 0, 0,
"Illegal request, commands not allowed to this device"},
{0x05258100, 0, 0,
"Illegal request, command not allowed to a secondary adapter"},
[SCSI] ipr: add support for new IOASCs This patch adds support for new errors that can be received from adapters using the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:32 -07:00
{0x05258200, 0, 0,
"Illegal request, command not allowed to a non-optimized resource"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x05260000, 0, 0,
"Illegal request, invalid field in parameter list"},
{0x05260100, 0, 0,
"Illegal request, parameter not supported"},
{0x05260200, 0, 0,
"Illegal request, parameter value invalid"},
{0x052C0000, 0, 0,
"Illegal request, command sequence error"},
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
{0x052C8000, 1, 0,
"Illegal request, dual adapter support not enabled"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x052C8100, 1, 0,
"Illegal request, another cable connector was physically disabled"},
{0x054E8000, 1, 0,
"Illegal request, inconsistent group id/group count"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06040500, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9031: Array protection temporarily suspended, protection resuming"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06040600, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9040: Array protection temporarily suspended, protection resuming"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x060B0100, 0, IPR_DEFAULT_LOG_LEVEL,
"4080: IOA exceeded maximum operating temperature"},
{0x060B8000, 0, IPR_DEFAULT_LOG_LEVEL,
"4085: Service required"},
scsi: ipr: new IOASC update This patch adds new adapter error log for P9 system with the new AZ SAS cable. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-05-09 12:47:54 -06:00
{0x060B8100, 0, IPR_DEFAULT_LOG_LEVEL,
"4086: SAS Adapter Hardware Configuration Error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06288000, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"3140: Device bus not ready to ready transition"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06290000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFFB: SCSI bus was reset"},
{0x06290500, 0, 0,
"FFFE: SCSI bus transition to single ended"},
{0x06290600, 0, 0,
"FFFE: SCSI bus transition to LVD"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06298000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"FFFB: SCSI bus was reset by another initiator"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x063F0300, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3029: A device replacement has occurred"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x063F8300, 0, IPR_DEFAULT_LOG_LEVEL,
"4102: Device bus fabric performance degradation"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x064C8000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9051: IOA cache data exists for a missing or failed device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x064C8100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
"9055: Auxiliary cache IOA contains cache data needed by the primary IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06670100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9025: Disk unit is not supported at its physical location"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06670600, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3020: IOA detected a SCSI bus configuration error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06678000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"3150: SCSI bus configuration error"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06678100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
"9074: Asymmetric advanced function disk configuration"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06678300, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4040: Incomplete multipath connection between IOA and enclosure"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06678400, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4041: Incomplete multipath connection between enclosure and device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06678500, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"9075: Incomplete multipath connection between IOA and remote IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06678600, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"9076: Configuration error, missing remote IOA"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06679100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4050: Enclosure does not support a required multipath function"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x06679800, 0, IPR_DEFAULT_LOG_LEVEL,
"4121: Configuration error, required cable is missing"},
{0x06679900, 0, IPR_DEFAULT_LOG_LEVEL,
"4122: Cable is not plugged into the correct location on remote IOA"},
{0x06679A00, 0, IPR_DEFAULT_LOG_LEVEL,
"4123: Configuration error, invalid cable vital product data"},
{0x06679B00, 0, IPR_DEFAULT_LOG_LEVEL,
"4124: Configuration error, both cable ends are plugged into the same IOA"},
[SCSI] ipr: add message to error table Adds a message to the error table for an error that wasn't previously handled. In some cases the I/O Adapter will detect an error condition and mark a block as "logically bad". Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-01-28 09:24:50 -07:00
{0x06690000, 0, IPR_DEFAULT_LOG_LEVEL,
"4070: Logically bad block written on device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06690200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9041: Array protection temporarily suspended"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x06698200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9042: Corrupt array parity detected on specified device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x066B0200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9030: Array no longer protected due to missing or failed disk unit"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x066B8000, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
"9071: Link operational transition"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x066B8100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
"9072: Link not operational transition"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x066B8200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9032: Array exposed but still protected"},
scsi: ipr: Don't log unnecessary 9084 error details A 9084 error gets logged by the ipr adapter when adapter raw mode gets enabled. A bunch of unformatted hex data also gets logged for this error, which is of little use, so let's avoid logging it by default in order to avoid the log getting polluted with useless data. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-16 15:51:36 -06:00
{0x066B8300, 0, IPR_DEBUG_LOG_LEVEL,
[SCSI] ipr: Disrupt device error Add entry in ipr error translation table for an error received when a device is forced into the failed state by the user. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:37 -06:00
"70DD: Device forced failed by disrupt device command"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x066B9100, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4061: Multipath redundancy level got better"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x066B9200, 0, IPR_DEFAULT_LOG_LEVEL,
[SCSI] ipr: Handle new SAS error codes Add definitions for some SAS error codes that can be logged by ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:44 -06:00
"4060: Multipath redundancy level got worse"},
scsi: ipr: Don't log unnecessary 9084 error details A 9084 error gets logged by the ipr adapter when adapter raw mode gets enabled. A bunch of unformatted hex data also gets logged for this error, which is of little use, so let's avoid logging it by default in order to avoid the log getting polluted with useless data. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-16 15:51:36 -06:00
{0x06808100, 0, IPR_DEBUG_LOG_LEVEL,
ipr: AF DASD raw mode implementation in ipr driver This patch implements raw mode support for AF DASD in ipr driver which allows for tools to send commands directly to physical devices which are members of RAID arrays when enabled in the firmware. [jejb: fix up whitespace] Signed-off-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:58 -06:00
"9083: Device raw mode enabled"},
scsi: ipr: Don't log unnecessary 9084 error details A 9084 error gets logged by the ipr adapter when adapter raw mode gets enabled. A bunch of unformatted hex data also gets logged for this error, which is of little use, so let's avoid logging it by default in order to avoid the log getting polluted with useless data. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-16 15:51:36 -06:00
{0x06808200, 0, IPR_DEBUG_LOG_LEVEL,
ipr: AF DASD raw mode implementation in ipr driver This patch implements raw mode support for AF DASD in ipr driver which allows for tools to send commands directly to physical devices which are members of RAID arrays when enabled in the firmware. [jejb: fix up whitespace] Signed-off-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:58 -06:00
"9084: Device raw mode disabled"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x07270000, 0, 0,
"Failure due to other device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278000, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9008: IOA does not support functions expected by devices"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9010: Cache data associated with attached devices cannot be found"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9011: Cache data belongs to devices other than those attached"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278400, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9020: Array missing 2 or more devices with only 1 device present"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278500, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9021: Array missing 2 or more devices with 2 or more devices present"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278600, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9022: Exposed array is missing a required device"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278700, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9023: Array member(s) not at required physical locations"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278800, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9024: Array not functional due to present hardware configuration"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278900, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9026: Array not functional due to present hardware configuration"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278A00, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9027: Array is missing a device and parity is out of sync"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278B00, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9028: Maximum number of arrays already exist"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278C00, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9050: Required cache data cannot be located for a disk unit"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278D00, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9052: Cache data exists for a device that has been modified"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07278F00, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9054: IOA resources not available due to previous problems"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07279100, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9092: Disk unit requires initialization before use"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07279200, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9029: Incorrect hardware configuration change has been detected"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07279600, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9060: One or more disk pairs are missing from an array"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07279700, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9061: One or more disks are missing from an array"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07279800, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9062: One or more disks are missing from an array"},
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
{0x07279900, 0, IPR_DEFAULT_LOG_LEVEL,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
"9063: Maximum number of functional arrays has been exceeded"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x07279A00, 0, 0,
"Data protect, other volume set problem"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x0B260000, 0, 0,
"Aborted command, invalid descriptor"},
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
{0x0B3F9000, 0, 0,
"Target operating conditions have changed, dual adapter takeover"},
{0x0B530200, 0, 0,
"Aborted command, medium removal prevented"},
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{0x0B5A0000, 0, 0,
[SCSI] ipr: IOA Status Code(IOASC) update Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-06-04 17:57:35 -06:00
"Command terminated by host"},
{0x0B5B8000, 0, 0,
"Aborted command, command terminated by host"}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
};
static const struct ipr_ses_table_entry ipr_ses_table[] = {
{ "2104-DL1 ", "XXXXXXXXXXXXXXXX", 80 },
{ "2104-TL1 ", "XXXXXXXXXXXXXXXX", 80 },
{ "HSBP07M P U2SCSI", "XXXXXXXXXXXXXXXX", 80 }, /* Hidive 7 slot */
{ "HSBP05M P U2SCSI", "XXXXXXXXXXXXXXXX", 80 }, /* Hidive 5 slot */
{ "HSBP05M S U2SCSI", "XXXXXXXXXXXXXXXX", 80 }, /* Bowtie */
{ "HSBP06E ASU2SCSI", "XXXXXXXXXXXXXXXX", 80 }, /* MartinFenning */
{ "2104-DU3 ", "XXXXXXXXXXXXXXXX", 160 },
{ "2104-TU3 ", "XXXXXXXXXXXXXXXX", 160 },
{ "HSBP04C RSU2SCSI", "XXXXXXX*XXXXXXXX", 160 },
{ "HSBP06E RSU2SCSI", "XXXXXXX*XXXXXXXX", 160 },
{ "St V1S2 ", "XXXXXXXXXXXXXXXX", 160 },
{ "HSBPD4M PU3SCSI", "XXXXXXX*XXXXXXXX", 160 },
{ "VSBPD1H U3SCSI", "XXXXXXX*XXXXXXXX", 160 }
};
/*
* Function Prototypes
*/
static int ipr_reset_alert(struct ipr_cmnd *);
static void ipr_process_ccn(struct ipr_cmnd *);
static void ipr_process_error(struct ipr_cmnd *);
static void ipr_reset_ioa_job(struct ipr_cmnd *);
static void ipr_initiate_ioa_reset(struct ipr_ioa_cfg *,
enum ipr_shutdown_type);
#ifdef CONFIG_SCSI_IPR_TRACE
/**
* ipr_trc_hook - Add a trace entry to the driver trace
* @ipr_cmd: ipr command struct
* @type: trace type
* @add_data: additional data
*
* Return value:
* none
**/
static void ipr_trc_hook(struct ipr_cmnd *ipr_cmd,
u8 type, u32 add_data)
{
struct ipr_trace_entry *trace_entry;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ipr: Fix incorrect trace indexing When ipr's internal driver trace was changed to an atomic, a signed/unsigned bug slipped in which results in us indexing backwards in our memory buffer writing on memory that does not belong to us. This patch fixes this by removing the modulo and instead just mask off the low bits. Cc: <stable@vger.kernel.org> Tested-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:31 -06:00
unsigned int trace_index;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: Fix incorrect trace indexing When ipr's internal driver trace was changed to an atomic, a signed/unsigned bug slipped in which results in us indexing backwards in our memory buffer writing on memory that does not belong to us. This patch fixes this by removing the modulo and instead just mask off the low bits. Cc: <stable@vger.kernel.org> Tested-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:31 -06:00
trace_index = atomic_add_return(1, &ioa_cfg->trace_index) & IPR_TRACE_INDEX_MASK;
trace_entry = &ioa_cfg->trace[trace_index];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
trace_entry->time = jiffies;
trace_entry->op_code = ipr_cmd->ioarcb.cmd_pkt.cdb[0];
trace_entry->type = type;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ipr_cmd->ioa_cfg->sis64)
trace_entry->ata_op_code = ipr_cmd->i.ata_ioadl.regs.command;
else
trace_entry->ata_op_code = ipr_cmd->ioarcb.u.add_data.u.regs.command;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
trace_entry->cmd_index = ipr_cmd->cmd_index & 0xff;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
trace_entry->res_handle = ipr_cmd->ioarcb.res_handle;
trace_entry->u.add_data = add_data;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
wmb();
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
#else
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
#define ipr_trc_hook(ipr_cmd, type, add_data) do { } while (0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
#endif
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
/**
* ipr_lock_and_done - Acquire lock and complete command
* @ipr_cmd: ipr command struct
*
* Return value:
* none
**/
static void ipr_lock_and_done(struct ipr_cmnd *ipr_cmd)
{
unsigned long lock_flags;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ipr_cmd->done(ipr_cmd);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_reinit_ipr_cmnd - Re-initialize an IPR Cmnd block for reuse
* @ipr_cmd: ipr command struct
*
* Return value:
* none
**/
static void ipr_reinit_ipr_cmnd(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
struct ipr_ioasa *ioasa = &ipr_cmd->s.ioasa;
struct ipr_ioasa64 *ioasa64 = &ipr_cmd->s.ioasa64;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
dma_addr_t dma_addr = ipr_cmd->dma_addr;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
int hrrq_id;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
hrrq_id = ioarcb->cmd_pkt.hrrq_id;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
memset(&ioarcb->cmd_pkt, 0, sizeof(struct ipr_cmd_pkt));
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ioarcb->cmd_pkt.hrrq_id = hrrq_id;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->data_transfer_length = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->read_data_transfer_length = 0;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->ioadl_len = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->read_ioadl_len = 0;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
if (ipr_cmd->ioa_cfg->sis64) {
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->u.sis64_addr_data.data_ioadl_addr =
cpu_to_be64(dma_addr + offsetof(struct ipr_cmnd, i.ioadl64));
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasa64->u.gata.status = 0;
} else {
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->write_ioadl_addr =
cpu_to_be32(dma_addr + offsetof(struct ipr_cmnd, i.ioadl));
ioarcb->read_ioadl_addr = ioarcb->write_ioadl_addr;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasa->u.gata.status = 0;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
}
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasa->hdr.ioasc = 0;
ioasa->hdr.residual_data_len = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->scsi_cmd = NULL;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ipr_cmd->qc = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->sense_buffer[0] = 0;
ipr_cmd->dma_use_sg = 0;
}
/**
* ipr_init_ipr_cmnd - Initialize an IPR Cmnd block
* @ipr_cmd: ipr command struct
*
* Return value:
* none
**/
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
static void ipr_init_ipr_cmnd(struct ipr_cmnd *ipr_cmd,
void (*fast_done) (struct ipr_cmnd *))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
ipr_reinit_ipr_cmnd(ipr_cmd);
ipr_cmd->u.scratch = 0;
ipr_cmd->sibling = NULL;
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
ipr_cmd->eh_comp = NULL;
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
ipr_cmd->fast_done = fast_done;
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
timer_setup(&ipr_cmd->timer, NULL, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
* __ipr_get_free_ipr_cmnd - Get a free IPR Cmnd block
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @ioa_cfg: ioa config struct
*
* Return value:
* pointer to ipr command struct
**/
static
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_cmnd *__ipr_get_free_ipr_cmnd(struct ipr_hrr_queue *hrrq)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_cmnd *ipr_cmd = NULL;
if (likely(!list_empty(&hrrq->hrrq_free_q))) {
ipr_cmd = list_entry(hrrq->hrrq_free_q.next,
struct ipr_cmnd, queue);
list_del(&ipr_cmd->queue);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return ipr_cmd;
}
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
/**
* ipr_get_free_ipr_cmnd - Get a free IPR Cmnd block and initialize it
* @ioa_cfg: ioa config struct
*
* Return value:
* pointer to ipr command struct
**/
static
struct ipr_cmnd *ipr_get_free_ipr_cmnd(struct ipr_ioa_cfg *ioa_cfg)
{
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_cmnd *ipr_cmd =
__ipr_get_free_ipr_cmnd(&ioa_cfg->hrrq[IPR_INIT_HRRQ]);
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
ipr_init_ipr_cmnd(ipr_cmd, ipr_lock_and_done);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
return ipr_cmd;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_mask_and_clear_interrupts - Mask all and clear specified interrupts
* @ioa_cfg: ioa config struct
* @clr_ints: interrupts to clear
*
* This function masks all interrupts on the adapter, then clears the
* interrupts specified in the mask
*
* Return value:
* none
**/
static void ipr_mask_and_clear_interrupts(struct ipr_ioa_cfg *ioa_cfg,
u32 clr_ints)
{
volatile u32 int_reg;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
int i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Stop new interrupts */
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].allow_interrupts = 0;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Set interrupt mask to stop all new interrupts */
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
if (ioa_cfg->sis64)
writeq(~0, ioa_cfg->regs.set_interrupt_mask_reg);
else
writel(~0, ioa_cfg->regs.set_interrupt_mask_reg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Clear any pending interrupts */
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
if (ioa_cfg->sis64)
writel(~0, ioa_cfg->regs.clr_interrupt_reg);
writel(clr_ints, ioa_cfg->regs.clr_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg);
}
/**
* ipr_save_pcix_cmd_reg - Save PCI-X command register
* @ioa_cfg: ioa config struct
*
* Return value:
* 0 on success / -EIO on failure
**/
static int ipr_save_pcix_cmd_reg(struct ipr_ioa_cfg *ioa_cfg)
{
int pcix_cmd_reg = pci_find_capability(ioa_cfg->pdev, PCI_CAP_ID_PCIX);
[SCSI] ipr: Tolerate not finding PCI-X registers Don't fail initialization of an adapter if the PCI-X registers cannot be found since it may be a PCI-E adapter. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:30 -07:00
if (pcix_cmd_reg == 0)
return 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (pci_read_config_word(ioa_cfg->pdev, pcix_cmd_reg + PCI_X_CMD,
&ioa_cfg->saved_pcix_cmd_reg) != PCIBIOS_SUCCESSFUL) {
dev_err(&ioa_cfg->pdev->dev, "Failed to save PCI-X command register\n");
return -EIO;
}
ioa_cfg->saved_pcix_cmd_reg |= PCI_X_CMD_DPERR_E | PCI_X_CMD_ERO;
return 0;
}
/**
* ipr_set_pcix_cmd_reg - Setup PCI-X command register
* @ioa_cfg: ioa config struct
*
* Return value:
* 0 on success / -EIO on failure
**/
static int ipr_set_pcix_cmd_reg(struct ipr_ioa_cfg *ioa_cfg)
{
int pcix_cmd_reg = pci_find_capability(ioa_cfg->pdev, PCI_CAP_ID_PCIX);
if (pcix_cmd_reg) {
if (pci_write_config_word(ioa_cfg->pdev, pcix_cmd_reg + PCI_X_CMD,
ioa_cfg->saved_pcix_cmd_reg) != PCIBIOS_SUCCESSFUL) {
dev_err(&ioa_cfg->pdev->dev, "Failed to setup PCI-X command register\n");
return -EIO;
}
}
return 0;
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
* __ipr_sata_eh_done - done function for aborted SATA commands
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
* @ipr_cmd: ipr command struct
*
* This function is invoked for ops generated to SATA
* devices which are being aborted.
*
* Return value:
* none
**/
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
static void __ipr_sata_eh_done(struct ipr_cmnd *ipr_cmd)
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
{
struct ata_queued_cmd *qc = ipr_cmd->qc;
struct ipr_sata_port *sata_port = qc->ap->private_data;
qc->err_mask |= AC_ERR_OTHER;
sata_port->ioasa.status |= ATA_BUSY;
ata_qc_complete(qc);
scsi: ipr: Fix missed EH wakeup Following a command abort or device reset, ipr's EH handlers wait for the commands getting aborted to get sent back from the adapter prior to returning from the EH handler. This fixes up some cases where the completion handler was not getting called, which would have resulted in the EH thread waiting until it timed out, greatly extending EH time. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:36 -06:00
if (ipr_cmd->eh_comp)
complete(ipr_cmd->eh_comp);
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
* ipr_sata_eh_done - done function for aborted SATA commands
* @ipr_cmd: ipr command struct
*
* This function is invoked for ops generated to SATA
* devices which are being aborted.
*
* Return value:
* none
**/
static void ipr_sata_eh_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
unsigned long hrrq_flags;
spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
__ipr_sata_eh_done(ipr_cmd);
spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
}
/**
* __ipr_scsi_eh_done - mid-layer done function for aborted ops
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @ipr_cmd: ipr command struct
*
* This function is invoked by the interrupt handler for
* ops generated by the SCSI mid-layer which are being aborted.
*
* Return value:
* none
**/
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
static void __ipr_scsi_eh_done(struct ipr_cmnd *ipr_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
scsi_cmd->result |= (DID_ERROR << 16);
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
scsi_dma_unmap(ipr_cmd->scsi_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi_cmd->scsi_done(scsi_cmd);
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
if (ipr_cmd->eh_comp)
complete(ipr_cmd->eh_comp);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
/**
* ipr_scsi_eh_done - mid-layer done function for aborted ops
* @ipr_cmd: ipr command struct
*
* This function is invoked by the interrupt handler for
* ops generated by the SCSI mid-layer which are being aborted.
*
* Return value:
* none
**/
static void ipr_scsi_eh_done(struct ipr_cmnd *ipr_cmd)
{
unsigned long hrrq_flags;
struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
__ipr_scsi_eh_done(ipr_cmd);
spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_fail_all_ops - Fails all outstanding ops.
* @ioa_cfg: ioa config struct
*
* This function fails all outstanding ops.
*
* Return value:
* none
**/
static void ipr_fail_all_ops(struct ipr_ioa_cfg *ioa_cfg)
{
struct ipr_cmnd *ipr_cmd, *temp;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for_each_hrrq(hrrq, ioa_cfg) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock(&hrrq->_lock);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_for_each_entry_safe(ipr_cmd,
temp, &hrrq->hrrq_pending_q, queue) {
list_del(&ipr_cmd->queue);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_cmd->s.ioasa.hdr.ioasc =
cpu_to_be32(IPR_IOASC_IOA_WAS_RESET);
ipr_cmd->s.ioasa.hdr.ilid =
cpu_to_be32(IPR_DRIVER_ILID);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
if (ipr_cmd->scsi_cmd)
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
ipr_cmd->done = __ipr_scsi_eh_done;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
else if (ipr_cmd->qc)
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
ipr_cmd->done = __ipr_sata_eh_done;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_trc_hook(ipr_cmd, IPR_TRACE_FINISH,
IPR_IOASC_IOA_WAS_RESET);
del_timer(&ipr_cmd->timer);
ipr_cmd->done(ipr_cmd);
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&hrrq->_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
LEAVE;
}
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
/**
* ipr_send_command - Send driver initiated requests.
* @ipr_cmd: ipr command struct
*
* This function sends a command to the adapter using the correct write call.
* In the case of sis64, calculate the ioarcb size required. Then or in the
* appropriate bits.
*
* Return value:
* none
**/
static void ipr_send_command(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
dma_addr_t send_dma_addr = ipr_cmd->dma_addr;
if (ioa_cfg->sis64) {
/* The default size is 256 bytes */
send_dma_addr |= 0x1;
/* If the number of ioadls * size of ioadl > 128 bytes,
then use a 512 byte ioarcb */
if (ipr_cmd->dma_use_sg * sizeof(struct ipr_ioadl64_desc) > 128 )
send_dma_addr |= 0x4;
writeq(send_dma_addr, ioa_cfg->regs.ioarrin_reg);
} else
writel(send_dma_addr, ioa_cfg->regs.ioarrin_reg);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_do_req - Send driver initiated requests.
* @ipr_cmd: ipr command struct
* @done: done function
* @timeout_func: timeout function
* @timeout: timeout value
*
* This function sends the specified command to the adapter with the
* timeout given. The done function is invoked on command completion.
*
* Return value:
* none
**/
static void ipr_do_req(struct ipr_cmnd *ipr_cmd,
void (*done) (struct ipr_cmnd *),
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
void (*timeout_func) (struct timer_list *), u32 timeout)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_pending_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->done = done;
ipr_cmd->timer.expires = jiffies + timeout;
treewide: Remove TIMER_FUNC_TYPE and TIMER_DATA_TYPE casts With all callbacks converted, and the timer callback prototype switched over, the TIMER_FUNC_TYPE cast is no longer needed, so remove it. Conversion was done with the following scripts: perl -pi -e 's|\(TIMER_FUNC_TYPE\)||g' \ $(git grep TIMER_FUNC_TYPE | cut -d: -f1 | sort -u) perl -pi -e 's|\(TIMER_DATA_TYPE\)||g' \ $(git grep TIMER_DATA_TYPE | cut -d: -f1 | sort -u) The now unused macros are also dropped from include/linux/timer.h. Signed-off-by: Kees Cook <keescook@chromium.org>
2017-10-23 01:40:42 -06:00
ipr_cmd->timer.function = timeout_func;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
add_timer(&ipr_cmd->timer);
ipr_trc_hook(ipr_cmd, IPR_TRACE_START, 0);
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_send_command(ipr_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_internal_cmd_done - Op done function for an internally generated op.
* @ipr_cmd: ipr command struct
*
* This function is the op done function for an internally generated,
* blocking op. It simply wakes the sleeping thread.
*
* Return value:
* none
**/
static void ipr_internal_cmd_done(struct ipr_cmnd *ipr_cmd)
{
if (ipr_cmd->sibling)
ipr_cmd->sibling = NULL;
else
complete(&ipr_cmd->completion);
}
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
/**
* ipr_init_ioadl - initialize the ioadl for the correct SIS type
* @ipr_cmd: ipr command struct
* @dma_addr: dma address
* @len: transfer length
* @flags: ioadl flag value
*
* This function initializes an ioadl in the case where there is only a single
* descriptor.
*
* Return value:
* nothing
**/
static void ipr_init_ioadl(struct ipr_cmnd *ipr_cmd, dma_addr_t dma_addr,
u32 len, int flags)
{
struct ipr_ioadl_desc *ioadl = ipr_cmd->i.ioadl;
struct ipr_ioadl64_desc *ioadl64 = ipr_cmd->i.ioadl64;
ipr_cmd->dma_use_sg = 1;
if (ipr_cmd->ioa_cfg->sis64) {
ioadl64->flags = cpu_to_be32(flags);
ioadl64->data_len = cpu_to_be32(len);
ioadl64->address = cpu_to_be64(dma_addr);
ipr_cmd->ioarcb.ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl64_desc));
ipr_cmd->ioarcb.data_transfer_length = cpu_to_be32(len);
} else {
ioadl->flags_and_data_len = cpu_to_be32(flags | len);
ioadl->address = cpu_to_be32(dma_addr);
if (flags == IPR_IOADL_FLAGS_READ_LAST) {
ipr_cmd->ioarcb.read_ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl_desc));
ipr_cmd->ioarcb.read_data_transfer_length = cpu_to_be32(len);
} else {
ipr_cmd->ioarcb.ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl_desc));
ipr_cmd->ioarcb.data_transfer_length = cpu_to_be32(len);
}
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_send_blocking_cmd - Send command and sleep on its completion.
* @ipr_cmd: ipr command struct
* @timeout_func: function to invoke if command times out
* @timeout: timeout
*
* Return value:
* none
**/
static void ipr_send_blocking_cmd(struct ipr_cmnd *ipr_cmd,
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
void (*timeout_func) (struct timer_list *),
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
u32 timeout)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
init_completion(&ipr_cmd->completion);
ipr_do_req(ipr_cmd, ipr_internal_cmd_done, timeout_func, timeout);
spin_unlock_irq(ioa_cfg->host->host_lock);
wait_for_completion(&ipr_cmd->completion);
spin_lock_irq(ioa_cfg->host->host_lock);
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
static int ipr_get_hrrq_index(struct ipr_ioa_cfg *ioa_cfg)
{
ipr: Fix invalid array indexing for HRRQ Fixes another signed / unsigned array indexing bug in the ipr driver. Currently, when hrrq_index wraps, it becomes a negative number. We do the modulo, but still have a negative number, so we end up indexing backwards in the array. Given where the hrrq array is located in memory, we probably won't actually reference memory we don't own, but nonetheless ipr is still looking at data within struct ipr_ioa_cfg and interpreting it as struct ipr_hrr_queue data, so bad things could certainly happen. Each ipr adapter has anywhere from 1 to 16 HRRQs. By default, we use 2 on new adapters. Let's take an example: Assume ioa_cfg->hrrq_index=0x7fffffffe and ioa_cfg->hrrq_num=4: The atomic_add_return will then return -1. We mod this with 3 and get -2, add one and get -1 for an array index. On adapters which support more than a single HRRQ, we dedicate HRRQ to adapter initialization and error interrupts so that we can optimize the other queues for fast path I/O. So all normal I/O uses HRRQ 1-15. So we want to spread the I/O requests across those HRRQs. With the default module parameter settings, this bug won't hit, only when someone sets the ipr.number_of_msix parameter to a value larger than 3 is when bad things start to happen. Cc: <stable@vger.kernel.org> Tested-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:33 -06:00
unsigned int hrrq;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
if (ioa_cfg->hrrq_num == 1)
ipr: Fix invalid array indexing for HRRQ Fixes another signed / unsigned array indexing bug in the ipr driver. Currently, when hrrq_index wraps, it becomes a negative number. We do the modulo, but still have a negative number, so we end up indexing backwards in the array. Given where the hrrq array is located in memory, we probably won't actually reference memory we don't own, but nonetheless ipr is still looking at data within struct ipr_ioa_cfg and interpreting it as struct ipr_hrr_queue data, so bad things could certainly happen. Each ipr adapter has anywhere from 1 to 16 HRRQs. By default, we use 2 on new adapters. Let's take an example: Assume ioa_cfg->hrrq_index=0x7fffffffe and ioa_cfg->hrrq_num=4: The atomic_add_return will then return -1. We mod this with 3 and get -2, add one and get -1 for an array index. On adapters which support more than a single HRRQ, we dedicate HRRQ to adapter initialization and error interrupts so that we can optimize the other queues for fast path I/O. So all normal I/O uses HRRQ 1-15. So we want to spread the I/O requests across those HRRQs. With the default module parameter settings, this bug won't hit, only when someone sets the ipr.number_of_msix parameter to a value larger than 3 is when bad things start to happen. Cc: <stable@vger.kernel.org> Tested-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:33 -06:00
hrrq = 0;
else {
hrrq = atomic_add_return(1, &ioa_cfg->hrrq_index);
hrrq = (hrrq % (ioa_cfg->hrrq_num - 1)) + 1;
}
return hrrq;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_send_hcam - Send an HCAM to the adapter.
* @ioa_cfg: ioa config struct
* @type: HCAM type
* @hostrcb: hostrcb struct
*
* This function will send a Host Controlled Async command to the adapter.
* If HCAMs are currently not allowed to be issued to the adapter, it will
* place the hostrcb on the free queue.
*
* Return value:
* none
**/
static void ipr_send_hcam(struct ipr_ioa_cfg *ioa_cfg, u8 type,
struct ipr_hostrcb *hostrcb)
{
struct ipr_cmnd *ipr_cmd;
struct ipr_ioarcb *ioarcb;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].allow_cmds) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd = ipr_get_free_ipr_cmnd(ioa_cfg);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_pending_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_add_tail(&hostrcb->queue, &ioa_cfg->hostrcb_pending_q);
ipr_cmd->u.hostrcb = hostrcb;
ioarcb = &ipr_cmd->ioarcb;
ioarcb->res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_HCAM;
ioarcb->cmd_pkt.cdb[0] = IPR_HOST_CONTROLLED_ASYNC;
ioarcb->cmd_pkt.cdb[1] = type;
ioarcb->cmd_pkt.cdb[7] = (sizeof(hostrcb->hcam) >> 8) & 0xff;
ioarcb->cmd_pkt.cdb[8] = sizeof(hostrcb->hcam) & 0xff;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_init_ioadl(ipr_cmd, hostrcb->hostrcb_dma,
sizeof(hostrcb->hcam), IPR_IOADL_FLAGS_READ_LAST);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (type == IPR_HCAM_CDB_OP_CODE_CONFIG_CHANGE)
ipr_cmd->done = ipr_process_ccn;
else
ipr_cmd->done = ipr_process_error;
ipr_trc_hook(ipr_cmd, IPR_TRACE_START, IPR_IOA_RES_ADDR);
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_send_command(ipr_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
} else {
list_add_tail(&hostrcb->queue, &ioa_cfg->hostrcb_free_q);
}
}
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
/**
* ipr_update_ata_class - Update the ata class in the resource entry
* @res: resource entry struct
* @proto: cfgte device bus protocol value
*
* Return value:
* none
**/
static void ipr_update_ata_class(struct ipr_resource_entry *res, unsigned int proto)
{
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
switch (proto) {
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
case IPR_PROTO_SATA:
case IPR_PROTO_SAS_STP:
res->ata_class = ATA_DEV_ATA;
break;
case IPR_PROTO_SATA_ATAPI:
case IPR_PROTO_SAS_STP_ATAPI:
res->ata_class = ATA_DEV_ATAPI;
break;
default:
res->ata_class = ATA_DEV_UNKNOWN;
break;
};
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_init_res_entry - Initialize a resource entry struct.
* @res: resource entry struct
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
* @cfgtew: config table entry wrapper struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* none
**/
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
static void ipr_init_res_entry(struct ipr_resource_entry *res,
struct ipr_config_table_entry_wrapper *cfgtew)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
int found = 0;
unsigned int proto;
struct ipr_ioa_cfg *ioa_cfg = res->ioa_cfg;
struct ipr_resource_entry *gscsi_res = NULL;
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
res->needs_sync_complete = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
res->in_erp = 0;
res->add_to_ml = 0;
res->del_from_ml = 0;
res->resetting_device = 0;
[SCSI] ipr: Remove extended delay bit on GSCSI reads/writes ops This patch removes extended delay bit on GSCSI reads/writes ops, the performance will be significanly better. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:39 -07:00
res->reset_occurred = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
res->sdev = NULL;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
res->sata_port = NULL;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (ioa_cfg->sis64) {
proto = cfgtew->u.cfgte64->proto;
ipr: Endian / sparse fixes Some misc fixes for endianness checking with sparse so sparse with endian checking now runs clean. Fixes a minor bug in the process which was uncovered by sparse which would result in unnecessary error recovery for check conditions. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-06-11 19:45:20 -06:00
res->flags = be16_to_cpu(cfgtew->u.cfgte64->flags);
res->res_flags = be16_to_cpu(cfgtew->u.cfgte64->res_flags);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
res->qmodel = IPR_QUEUEING_MODEL64(res);
[SCSI] ipr: implement fixes for 64 bit adapter support Implement some small fixes for 64 bit support that were preventing the adapter from becoming operational. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:00 -06:00
res->type = cfgtew->u.cfgte64->res_type;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
memcpy(res->res_path, &cfgtew->u.cfgte64->res_path,
sizeof(res->res_path));
res->bus = 0;
[SCSI] ipr: fix lun assignment and comparison The lun value was not getting set up correctly for all devices attached to the new 64 bit adapters. The fix is to move the logic to earlier in the ipr_init_res_entry routine such that the value does get set correctly for all devices. Then the ipr_is_same_device comparison function was using the wrong lun value in the logic for the new adapters. Change this to use the correct lun value. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-11-04 10:35:58 -06:00
memcpy(&res->dev_lun.scsi_lun, &cfgtew->u.cfgte64->lun,
sizeof(res->dev_lun.scsi_lun));
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
res->lun = scsilun_to_int(&res->dev_lun);
if (res->type == IPR_RES_TYPE_GENERIC_SCSI) {
list_for_each_entry(gscsi_res, &ioa_cfg->used_res_q, queue) {
if (gscsi_res->dev_id == cfgtew->u.cfgte64->dev_id) {
found = 1;
res->target = gscsi_res->target;
break;
}
}
if (!found) {
res->target = find_first_zero_bit(ioa_cfg->target_ids,
ioa_cfg->max_devs_supported);
set_bit(res->target, ioa_cfg->target_ids);
}
} else if (res->type == IPR_RES_TYPE_IOAFP) {
res->bus = IPR_IOAFP_VIRTUAL_BUS;
res->target = 0;
} else if (res->type == IPR_RES_TYPE_ARRAY) {
res->bus = IPR_ARRAY_VIRTUAL_BUS;
res->target = find_first_zero_bit(ioa_cfg->array_ids,
ioa_cfg->max_devs_supported);
set_bit(res->target, ioa_cfg->array_ids);
} else if (res->type == IPR_RES_TYPE_VOLUME_SET) {
res->bus = IPR_VSET_VIRTUAL_BUS;
res->target = find_first_zero_bit(ioa_cfg->vset_ids,
ioa_cfg->max_devs_supported);
set_bit(res->target, ioa_cfg->vset_ids);
} else {
res->target = find_first_zero_bit(ioa_cfg->target_ids,
ioa_cfg->max_devs_supported);
set_bit(res->target, ioa_cfg->target_ids);
}
} else {
proto = cfgtew->u.cfgte->proto;
res->qmodel = IPR_QUEUEING_MODEL(res);
res->flags = cfgtew->u.cfgte->flags;
if (res->flags & IPR_IS_IOA_RESOURCE)
res->type = IPR_RES_TYPE_IOAFP;
else
res->type = cfgtew->u.cfgte->rsvd_subtype & 0x0f;
res->bus = cfgtew->u.cfgte->res_addr.bus;
res->target = cfgtew->u.cfgte->res_addr.target;
res->lun = cfgtew->u.cfgte->res_addr.lun;
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
res->lun_wwn = get_unaligned_be64(cfgtew->u.cfgte->lun_wwn);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
}
ipr_update_ata_class(res, proto);
}
/**
* ipr_is_same_device - Determine if two devices are the same.
* @res: resource entry struct
* @cfgtew: config table entry wrapper struct
*
* Return value:
* 1 if the devices are the same / 0 otherwise
**/
static int ipr_is_same_device(struct ipr_resource_entry *res,
struct ipr_config_table_entry_wrapper *cfgtew)
{
if (res->ioa_cfg->sis64) {
if (!memcmp(&res->dev_id, &cfgtew->u.cfgte64->dev_id,
sizeof(cfgtew->u.cfgte64->dev_id)) &&
[SCSI] ipr: fix lun assignment and comparison The lun value was not getting set up correctly for all devices attached to the new 64 bit adapters. The fix is to move the logic to earlier in the ipr_init_res_entry routine such that the value does get set correctly for all devices. Then the ipr_is_same_device comparison function was using the wrong lun value in the logic for the new adapters. Change this to use the correct lun value. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-11-04 10:35:58 -06:00
!memcmp(&res->dev_lun.scsi_lun, &cfgtew->u.cfgte64->lun,
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
sizeof(cfgtew->u.cfgte64->lun))) {
return 1;
}
} else {
if (res->bus == cfgtew->u.cfgte->res_addr.bus &&
res->target == cfgtew->u.cfgte->res_addr.target &&
res->lun == cfgtew->u.cfgte->res_addr.lun)
return 1;
}
return 0;
}
/**
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
* __ipr_format_res_path - Format the resource path for printing.
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
* @res_path: resource path
* @buf: buffer
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
* @len: length of buffer provided
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
*
* Return value:
* pointer to buffer
**/
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
static char *__ipr_format_res_path(u8 *res_path, char *buffer, int len)
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
{
int i;
[SCSI] ipr: fix resource path display and formatting It was possible to overflow the buffer used to print out the formatted version of the resource path. The fix is to limit the number of bytes that get formatted. This patch also updates the ipr_show_resource_path function to display the resource address for devices that are attached to adapters that don't support resource paths. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-03 17:02:21 -06:00
char *p = buffer;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
*p = '\0';
[SCSI] ipr: fix resource path display and formatting It was possible to overflow the buffer used to print out the formatted version of the resource path. The fix is to limit the number of bytes that get formatted. This patch also updates the ipr_show_resource_path function to display the resource address for devices that are attached to adapters that don't support resource paths. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-03 17:02:21 -06:00
p += snprintf(p, buffer + len - p, "%02X", res_path[0]);
for (i = 1; res_path[i] != 0xff && ((i * 3) < len); i++)
p += snprintf(p, buffer + len - p, "-%02X", res_path[i]);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
return buffer;
}
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
/**
* ipr_format_res_path - Format the resource path for printing.
* @ioa_cfg: ioa config struct
* @res_path: resource path
* @buf: buffer
* @len: length of buffer provided
*
* Return value:
* pointer to buffer
**/
static char *ipr_format_res_path(struct ipr_ioa_cfg *ioa_cfg,
u8 *res_path, char *buffer, int len)
{
char *p = buffer;
*p = '\0';
p += snprintf(p, buffer + len - p, "%d/", ioa_cfg->host->host_no);
__ipr_format_res_path(res_path, p, len - (buffer - p));
return buffer;
}
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
/**
* ipr_update_res_entry - Update the resource entry.
* @res: resource entry struct
* @cfgtew: config table entry wrapper struct
*
* Return value:
* none
**/
static void ipr_update_res_entry(struct ipr_resource_entry *res,
struct ipr_config_table_entry_wrapper *cfgtew)
{
char buffer[IPR_MAX_RES_PATH_LENGTH];
unsigned int proto;
int new_path = 0;
if (res->ioa_cfg->sis64) {
ipr: Endian / sparse fixes Some misc fixes for endianness checking with sparse so sparse with endian checking now runs clean. Fixes a minor bug in the process which was uncovered by sparse which would result in unnecessary error recovery for check conditions. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-06-11 19:45:20 -06:00
res->flags = be16_to_cpu(cfgtew->u.cfgte64->flags);
res->res_flags = be16_to_cpu(cfgtew->u.cfgte64->res_flags);
[SCSI] ipr: fix resource type update and add sdev and shost attributes Setting the resource type in the ipr_update_res_entry function was incorrect in that the top 4 bits were masked off. The assignment has been updated to no longer mask those bits. Then, two new attributes were added to allow the user space utilities to more easily get information. The resource_type sdev attribute is set for all devices in the adapter's configuration table and indicates the type of device. The fw_type shost attribute indicates the firmware type supported by the adapter. Finally, the resource_path attribute was changed to be mode S_IRUGO. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:50:14 -06:00
res->type = cfgtew->u.cfgte64->res_type;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
memcpy(&res->std_inq_data, &cfgtew->u.cfgte64->std_inq_data,
sizeof(struct ipr_std_inq_data));
res->qmodel = IPR_QUEUEING_MODEL64(res);
proto = cfgtew->u.cfgte64->proto;
res->res_handle = cfgtew->u.cfgte64->res_handle;
res->dev_id = cfgtew->u.cfgte64->dev_id;
memcpy(&res->dev_lun.scsi_lun, &cfgtew->u.cfgte64->lun,
sizeof(res->dev_lun.scsi_lun));
if (memcmp(res->res_path, &cfgtew->u.cfgte64->res_path,
sizeof(res->res_path))) {
memcpy(res->res_path, &cfgtew->u.cfgte64->res_path,
sizeof(res->res_path));
new_path = 1;
}
if (res->sdev && new_path)
sdev_printk(KERN_INFO, res->sdev, "Resource path: %s\n",
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(res->ioa_cfg,
res->res_path, buffer, sizeof(buffer)));
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
} else {
res->flags = cfgtew->u.cfgte->flags;
if (res->flags & IPR_IS_IOA_RESOURCE)
res->type = IPR_RES_TYPE_IOAFP;
else
res->type = cfgtew->u.cfgte->rsvd_subtype & 0x0f;
memcpy(&res->std_inq_data, &cfgtew->u.cfgte->std_inq_data,
sizeof(struct ipr_std_inq_data));
res->qmodel = IPR_QUEUEING_MODEL(res);
proto = cfgtew->u.cfgte->proto;
res->res_handle = cfgtew->u.cfgte->res_handle;
}
ipr_update_ata_class(res, proto);
}
/**
* ipr_clear_res_target - Clear the bit in the bit map representing the target
* for the resource.
* @res: resource entry struct
* @cfgtew: config table entry wrapper struct
*
* Return value:
* none
**/
static void ipr_clear_res_target(struct ipr_resource_entry *res)
{
struct ipr_resource_entry *gscsi_res = NULL;
struct ipr_ioa_cfg *ioa_cfg = res->ioa_cfg;
if (!ioa_cfg->sis64)
return;
if (res->bus == IPR_ARRAY_VIRTUAL_BUS)
clear_bit(res->target, ioa_cfg->array_ids);
else if (res->bus == IPR_VSET_VIRTUAL_BUS)
clear_bit(res->target, ioa_cfg->vset_ids);
else if (res->bus == 0 && res->type == IPR_RES_TYPE_GENERIC_SCSI) {
list_for_each_entry(gscsi_res, &ioa_cfg->used_res_q, queue)
if (gscsi_res->dev_id == res->dev_id && gscsi_res != res)
return;
clear_bit(res->target, ioa_cfg->target_ids);
} else if (res->bus == 0)
clear_bit(res->target, ioa_cfg->target_ids);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_handle_config_change - Handle a config change from the adapter
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb
*
* Return value:
* none
**/
static void ipr_handle_config_change(struct ipr_ioa_cfg *ioa_cfg,
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
struct ipr_hostrcb *hostrcb)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_resource_entry *res = NULL;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
struct ipr_config_table_entry_wrapper cfgtew;
__be32 cc_res_handle;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
u32 is_ndn = 1;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (ioa_cfg->sis64) {
cfgtew.u.cfgte64 = &hostrcb->hcam.u.ccn.u.cfgte64;
cc_res_handle = cfgtew.u.cfgte64->res_handle;
} else {
cfgtew.u.cfgte = &hostrcb->hcam.u.ccn.u.cfgte;
cc_res_handle = cfgtew.u.cfgte->res_handle;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (res->res_handle == cc_res_handle) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
is_ndn = 0;
break;
}
}
if (is_ndn) {
if (list_empty(&ioa_cfg->free_res_q)) {
ipr_send_hcam(ioa_cfg,
IPR_HCAM_CDB_OP_CODE_CONFIG_CHANGE,
hostrcb);
return;
}
res = list_entry(ioa_cfg->free_res_q.next,
struct ipr_resource_entry, queue);
list_del(&res->queue);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_init_res_entry(res, &cfgtew);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_add_tail(&res->queue, &ioa_cfg->used_res_q);
}
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_update_res_entry(res, &cfgtew);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (hostrcb->hcam.notify_type == IPR_HOST_RCB_NOTIF_TYPE_REM_ENTRY) {
if (res->sdev) {
res->del_from_ml = 1;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
res->res_handle = IPR_INVALID_RES_HANDLE;
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
schedule_work(&ioa_cfg->work_q);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
} else {
ipr_clear_res_target(res);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_move_tail(&res->queue, &ioa_cfg->free_res_q);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
}
[SCSI] ipr: Fix a race on multiple configuration changes In a multiple configuration change scenario a remove notification can be followed by an immediate add notification for the same device, which will cause the device to be removed but never added back. This patch fixes the problem by ensuring that in such situations the device will be added back. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-02-14 15:19:31 -07:00
} else if (!res->sdev || res->del_from_ml) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
res->add_to_ml = 1;
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
schedule_work(&ioa_cfg->work_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ipr_send_hcam(ioa_cfg, IPR_HCAM_CDB_OP_CODE_CONFIG_CHANGE, hostrcb);
}
/**
* ipr_process_ccn - Op done function for a CCN.
* @ipr_cmd: ipr command struct
*
* This function is the op done function for a configuration
* change notification host controlled async from the adapter.
*
* Return value:
* none
**/
static void ipr_process_ccn(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_hostrcb *hostrcb = ipr_cmd->u.hostrcb;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
list_del_init(&hostrcb->queue);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioasc) {
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
if (ioasc != IPR_IOASC_IOA_WAS_RESET &&
ioasc != IPR_IOASC_ABORTED_CMD_TERM_BY_HOST)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_err(&ioa_cfg->pdev->dev,
"Host RCB failed with IOASC: 0x%08X\n", ioasc);
ipr_send_hcam(ioa_cfg, IPR_HCAM_CDB_OP_CODE_CONFIG_CHANGE, hostrcb);
} else {
ipr_handle_config_change(ioa_cfg, hostrcb);
}
}
[SCSI] ipr: Improved dual adapter errors Formats ipr dual adapter errors so that they are more compact. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:14 -06:00
/**
* strip_and_pad_whitespace - Strip and pad trailing whitespace.
* @i: index into buffer
* @buf: string to modify
*
* This function will strip all trailing whitespace, pad the end
* of the string with a single space, and NULL terminate the string.
*
* Return value:
* new length of string
**/
static int strip_and_pad_whitespace(int i, char *buf)
{
while (i && buf[i] == ' ')
i--;
buf[i+1] = ' ';
buf[i+2] = '\0';
return i + 2;
}
/**
* ipr_log_vpd_compact - Log the passed extended VPD compactly.
* @prefix: string to print at start of printk
* @hostrcb: hostrcb pointer
* @vpd: vendor/product id/sn struct
*
* Return value:
* none
**/
static void ipr_log_vpd_compact(char *prefix, struct ipr_hostrcb *hostrcb,
struct ipr_vpd *vpd)
{
char buffer[IPR_VENDOR_ID_LEN + IPR_PROD_ID_LEN + IPR_SERIAL_NUM_LEN + 3];
int i = 0;
memcpy(buffer, vpd->vpids.vendor_id, IPR_VENDOR_ID_LEN);
i = strip_and_pad_whitespace(IPR_VENDOR_ID_LEN - 1, buffer);
memcpy(&buffer[i], vpd->vpids.product_id, IPR_PROD_ID_LEN);
i = strip_and_pad_whitespace(i + IPR_PROD_ID_LEN - 1, buffer);
memcpy(&buffer[i], vpd->sn, IPR_SERIAL_NUM_LEN);
buffer[IPR_SERIAL_NUM_LEN + i] = '\0';
ipr_hcam_err(hostrcb, "%s VPID/SN: %s\n", prefix, buffer);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_log_vpd - Log the passed VPD to the error log.
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
* @vpd: vendor/product id/sn struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* none
**/
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
static void ipr_log_vpd(struct ipr_vpd *vpd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
char buffer[IPR_VENDOR_ID_LEN + IPR_PROD_ID_LEN
+ IPR_SERIAL_NUM_LEN];
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
memcpy(buffer, vpd->vpids.vendor_id, IPR_VENDOR_ID_LEN);
memcpy(buffer + IPR_VENDOR_ID_LEN, vpd->vpids.product_id,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
IPR_PROD_ID_LEN);
buffer[IPR_VENDOR_ID_LEN + IPR_PROD_ID_LEN] = '\0';
ipr_err("Vendor/Product ID: %s\n", buffer);
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
memcpy(buffer, vpd->sn, IPR_SERIAL_NUM_LEN);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
buffer[IPR_SERIAL_NUM_LEN] = '\0';
ipr_err(" Serial Number: %s\n", buffer);
}
[SCSI] ipr: Improved dual adapter errors Formats ipr dual adapter errors so that they are more compact. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:14 -06:00
/**
* ipr_log_ext_vpd_compact - Log the passed extended VPD compactly.
* @prefix: string to print at start of printk
* @hostrcb: hostrcb pointer
* @vpd: vendor/product id/sn/wwn struct
*
* Return value:
* none
**/
static void ipr_log_ext_vpd_compact(char *prefix, struct ipr_hostrcb *hostrcb,
struct ipr_ext_vpd *vpd)
{
ipr_log_vpd_compact(prefix, hostrcb, &vpd->vpd);
ipr_hcam_err(hostrcb, "%s WWN: %08X%08X\n", prefix,
be32_to_cpu(vpd->wwid[0]), be32_to_cpu(vpd->wwid[1]));
}
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
/**
* ipr_log_ext_vpd - Log the passed extended VPD to the error log.
* @vpd: vendor/product id/sn/wwn struct
*
* Return value:
* none
**/
static void ipr_log_ext_vpd(struct ipr_ext_vpd *vpd)
{
ipr_log_vpd(&vpd->vpd);
ipr_err(" WWN: %08X%08X\n", be32_to_cpu(vpd->wwid[0]),
be32_to_cpu(vpd->wwid[1]));
}
/**
* ipr_log_enhanced_cache_error - Log a cache error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_enhanced_cache_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
struct ipr_hostrcb_type_12_error *error;
if (ioa_cfg->sis64)
error = &hostrcb->hcam.u.error64.u.type_12_error;
else
error = &hostrcb->hcam.u.error.u.type_12_error;
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
ipr_err("-----Current Configuration-----\n");
ipr_err("Cache Directory Card Information:\n");
ipr_log_ext_vpd(&error->ioa_vpd);
ipr_err("Adapter Card Information:\n");
ipr_log_ext_vpd(&error->cfc_vpd);
ipr_err("-----Expected Configuration-----\n");
ipr_err("Cache Directory Card Information:\n");
ipr_log_ext_vpd(&error->ioa_last_attached_to_cfc_vpd);
ipr_err("Adapter Card Information:\n");
ipr_log_ext_vpd(&error->cfc_last_attached_to_ioa_vpd);
ipr_err("Additional IOA Data: %08X %08X %08X\n",
be32_to_cpu(error->ioa_data[0]),
be32_to_cpu(error->ioa_data[1]),
be32_to_cpu(error->ioa_data[2]));
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_log_cache_error - Log a cache error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_cache_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
struct ipr_hostrcb_type_02_error *error =
&hostrcb->hcam.u.error.u.type_02_error;
ipr_err("-----Current Configuration-----\n");
ipr_err("Cache Directory Card Information:\n");
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&error->ioa_vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Adapter Card Information:\n");
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&error->cfc_vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("-----Expected Configuration-----\n");
ipr_err("Cache Directory Card Information:\n");
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&error->ioa_last_attached_to_cfc_vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Adapter Card Information:\n");
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&error->cfc_last_attached_to_ioa_vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Additional IOA Data: %08X %08X %08X\n",
be32_to_cpu(error->ioa_data[0]),
be32_to_cpu(error->ioa_data[1]),
be32_to_cpu(error->ioa_data[2]));
}
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
/**
* ipr_log_enhanced_config_error - Log a configuration error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_enhanced_config_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
int errors_logged, i;
struct ipr_hostrcb_device_data_entry_enhanced *dev_entry;
struct ipr_hostrcb_type_13_error *error;
error = &hostrcb->hcam.u.error.u.type_13_error;
errors_logged = be32_to_cpu(error->errors_logged);
ipr_err("Device Errors Detected/Logged: %d/%d\n",
be32_to_cpu(error->errors_detected), errors_logged);
dev_entry = error->dev;
for (i = 0; i < errors_logged; i++, dev_entry++) {
ipr_err_separator;
ipr_phys_res_err(ioa_cfg, dev_entry->dev_res_addr, "Device %d", i + 1);
ipr_log_ext_vpd(&dev_entry->vpd);
ipr_err("-----New Device Information-----\n");
ipr_log_ext_vpd(&dev_entry->new_vpd);
ipr_err("Cache Directory Card Information:\n");
ipr_log_ext_vpd(&dev_entry->ioa_last_with_dev_vpd);
ipr_err("Adapter Card Information:\n");
ipr_log_ext_vpd(&dev_entry->cfc_last_with_dev_vpd);
}
}
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
/**
* ipr_log_sis64_config_error - Log a device error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_sis64_config_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
int errors_logged, i;
struct ipr_hostrcb64_device_data_entry_enhanced *dev_entry;
struct ipr_hostrcb_type_23_error *error;
char buffer[IPR_MAX_RES_PATH_LENGTH];
error = &hostrcb->hcam.u.error64.u.type_23_error;
errors_logged = be32_to_cpu(error->errors_logged);
ipr_err("Device Errors Detected/Logged: %d/%d\n",
be32_to_cpu(error->errors_detected), errors_logged);
dev_entry = error->dev;
for (i = 0; i < errors_logged; i++, dev_entry++) {
ipr_err_separator;
ipr_err("Device %d : %s", i + 1,
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
__ipr_format_res_path(dev_entry->res_path,
buffer, sizeof(buffer)));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
ipr_log_ext_vpd(&dev_entry->vpd);
ipr_err("-----New Device Information-----\n");
ipr_log_ext_vpd(&dev_entry->new_vpd);
ipr_err("Cache Directory Card Information:\n");
ipr_log_ext_vpd(&dev_entry->ioa_last_with_dev_vpd);
ipr_err("Adapter Card Information:\n");
ipr_log_ext_vpd(&dev_entry->cfc_last_with_dev_vpd);
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_log_config_error - Log a configuration error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_config_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
int errors_logged, i;
struct ipr_hostrcb_device_data_entry *dev_entry;
struct ipr_hostrcb_type_03_error *error;
error = &hostrcb->hcam.u.error.u.type_03_error;
errors_logged = be32_to_cpu(error->errors_logged);
ipr_err("Device Errors Detected/Logged: %d/%d\n",
be32_to_cpu(error->errors_detected), errors_logged);
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
dev_entry = error->dev;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
for (i = 0; i < errors_logged; i++, dev_entry++) {
ipr_err_separator;
[SCSI] ipr: Physical resource error logging macro Adds a macro in the ipr driver for logging a physical device location. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:27 -07:00
ipr_phys_res_err(ioa_cfg, dev_entry->dev_res_addr, "Device %d", i + 1);
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&dev_entry->vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("-----New Device Information-----\n");
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&dev_entry->new_vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Cache Directory Card Information:\n");
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&dev_entry->ioa_last_with_dev_vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Adapter Card Information:\n");
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&dev_entry->cfc_last_with_dev_vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Additional IOA Data: %08X %08X %08X %08X %08X\n",
be32_to_cpu(dev_entry->ioa_data[0]),
be32_to_cpu(dev_entry->ioa_data[1]),
be32_to_cpu(dev_entry->ioa_data[2]),
be32_to_cpu(dev_entry->ioa_data[3]),
be32_to_cpu(dev_entry->ioa_data[4]));
}
}
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
/**
* ipr_log_enhanced_array_error - Log an array configuration error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_enhanced_array_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
int i, num_entries;
struct ipr_hostrcb_type_14_error *error;
struct ipr_hostrcb_array_data_entry_enhanced *array_entry;
const u8 zero_sn[IPR_SERIAL_NUM_LEN] = { [0 ... IPR_SERIAL_NUM_LEN-1] = '0' };
error = &hostrcb->hcam.u.error.u.type_14_error;
ipr_err_separator;
ipr_err("RAID %s Array Configuration: %d:%d:%d:%d\n",
error->protection_level,
ioa_cfg->host->host_no,
error->last_func_vset_res_addr.bus,
error->last_func_vset_res_addr.target,
error->last_func_vset_res_addr.lun);
ipr_err_separator;
array_entry = error->array_member;
num_entries = min_t(u32, be32_to_cpu(error->num_entries),
[SCSI] ipr: fix array error logging The structure definitions for reporting array errors did not have the correct size for the Array WWID field. This patch fixes those definitions. It also fixes part of the output formatting that did not have newlines and fixes size calculations. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-09-27 11:45:28 -06:00
ARRAY_SIZE(error->array_member));
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
for (i = 0; i < num_entries; i++, array_entry++) {
if (!memcmp(array_entry->vpd.vpd.sn, zero_sn, IPR_SERIAL_NUM_LEN))
continue;
if (be32_to_cpu(error->exposed_mode_adn) == i)
ipr_err("Exposed Array Member %d:\n", i);
else
ipr_err("Array Member %d:\n", i);
ipr_log_ext_vpd(&array_entry->vpd);
ipr_phys_res_err(ioa_cfg, array_entry->dev_res_addr, "Current Location");
ipr_phys_res_err(ioa_cfg, array_entry->expected_dev_res_addr,
"Expected Location");
ipr_err_separator;
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_log_array_error - Log an array configuration error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_array_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
int i;
struct ipr_hostrcb_type_04_error *error;
struct ipr_hostrcb_array_data_entry *array_entry;
const u8 zero_sn[IPR_SERIAL_NUM_LEN] = { [0 ... IPR_SERIAL_NUM_LEN-1] = '0' };
error = &hostrcb->hcam.u.error.u.type_04_error;
ipr_err_separator;
ipr_err("RAID %s Array Configuration: %d:%d:%d:%d\n",
error->protection_level,
ioa_cfg->host->host_no,
error->last_func_vset_res_addr.bus,
error->last_func_vset_res_addr.target,
error->last_func_vset_res_addr.lun);
ipr_err_separator;
array_entry = error->array_member;
for (i = 0; i < 18; i++) {
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
if (!memcmp(array_entry->vpd.sn, zero_sn, IPR_SERIAL_NUM_LEN))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
continue;
[SCSI] ipr: Physical resource error logging macro Adds a macro in the ipr driver for logging a physical device location. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:27 -07:00
if (be32_to_cpu(error->exposed_mode_adn) == i)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Exposed Array Member %d:\n", i);
[SCSI] ipr: Physical resource error logging macro Adds a macro in the ipr driver for logging a physical device location. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:27 -07:00
else
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("Array Member %d:\n", i);
[SCSI] ipr: Cleanup error structures Simplify the ipr error structures a bit by removing some duplication. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:18 -07:00
ipr_log_vpd(&array_entry->vpd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Physical resource error logging macro Adds a macro in the ipr driver for logging a physical device location. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:27 -07:00
ipr_phys_res_err(ioa_cfg, array_entry->dev_res_addr, "Current Location");
ipr_phys_res_err(ioa_cfg, array_entry->expected_dev_res_addr,
"Expected Location");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err_separator;
if (i == 9)
array_entry = error->array_member2;
else
array_entry++;
}
}
/**
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
* ipr_log_hex_data - Log additional hex IOA error data.
[SCSI] ipr: Reduce default error log size Since the default error log size has increased on SAS adapters, prevent ipr from logging this additional data unless requested to do so by the user set log level in order to prevent flooding the logs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:42 -07:00
* @ioa_cfg: ioa config struct
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
* @data: IOA error data
* @len: data length
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* none
**/
ipr: Endian / sparse fixes Some misc fixes for endianness checking with sparse so sparse with endian checking now runs clean. Fixes a minor bug in the process which was uncovered by sparse which would result in unnecessary error recovery for check conditions. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-06-11 19:45:20 -06:00
static void ipr_log_hex_data(struct ipr_ioa_cfg *ioa_cfg, __be32 *data, int len)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
int i;
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
if (len == 0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
[SCSI] ipr: Reduce default error log size Since the default error log size has increased on SAS adapters, prevent ipr from logging this additional data unless requested to do so by the user set log level in order to prevent flooding the logs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:42 -07:00
if (ioa_cfg->log_level <= IPR_DEFAULT_LOG_LEVEL)
len = min_t(int, len, IPR_DEFAULT_MAX_ERROR_DUMP);
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
for (i = 0; i < len / 4; i += 4) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_err("%08X: %08X %08X %08X %08X\n", i*4,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
be32_to_cpu(data[i]),
be32_to_cpu(data[i+1]),
be32_to_cpu(data[i+2]),
be32_to_cpu(data[i+3]));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
}
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
/**
* ipr_log_enhanced_dual_ioa_error - Log an enhanced dual adapter error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_enhanced_dual_ioa_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
struct ipr_hostrcb_type_17_error *error;
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
if (ioa_cfg->sis64)
error = &hostrcb->hcam.u.error64.u.type_17_error;
else
error = &hostrcb->hcam.u.error.u.type_17_error;
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
error->failure_reason[sizeof(error->failure_reason) - 1] = '\0';
Subject: Re: [PATCH] strstrip incorrectly marked __must_check Recently, We marked strstrip() as must_check. because it was frequently misused and it should be checked. However, we found one exception. scsi/ipr.c intentionally ignore return value of strstrip. Because it wishes to keep the whitespace at the beginning. Thus we need to keep with and without checked whitespace trim function. This patch adds a new strim() and changes ipr.c to use it. [akpm@linux-foundation.org: coding-style fixes] Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-14 19:01:15 -07:00
strim(error->failure_reason);
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
[SCSI] ipr: Improved dual adapter errors Formats ipr dual adapter errors so that they are more compact. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:14 -06:00
ipr_hcam_err(hostrcb, "%s [PRC: %08X]\n", error->failure_reason,
be32_to_cpu(hostrcb->hcam.u.error.prc));
ipr_log_ext_vpd_compact("Remote IOA", hostrcb, &error->vpd);
[SCSI] ipr: Reduce default error log size Since the default error log size has increased on SAS adapters, prevent ipr from logging this additional data unless requested to do so by the user set log level in order to prevent flooding the logs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:42 -07:00
ipr_log_hex_data(ioa_cfg, error->data,
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
be32_to_cpu(hostrcb->hcam.length) -
(offsetof(struct ipr_hostrcb_error, u) +
offsetof(struct ipr_hostrcb_type_17_error, data)));
}
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
/**
* ipr_log_dual_ioa_error - Log a dual adapter error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_dual_ioa_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
struct ipr_hostrcb_type_07_error *error;
error = &hostrcb->hcam.u.error.u.type_07_error;
error->failure_reason[sizeof(error->failure_reason) - 1] = '\0';
Subject: Re: [PATCH] strstrip incorrectly marked __must_check Recently, We marked strstrip() as must_check. because it was frequently misused and it should be checked. However, we found one exception. scsi/ipr.c intentionally ignore return value of strstrip. Because it wishes to keep the whitespace at the beginning. Thus we need to keep with and without checked whitespace trim function. This patch adds a new strim() and changes ipr.c to use it. [akpm@linux-foundation.org: coding-style fixes] Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-12-14 19:01:15 -07:00
strim(error->failure_reason);
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
[SCSI] ipr: Improved dual adapter errors Formats ipr dual adapter errors so that they are more compact. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:14 -06:00
ipr_hcam_err(hostrcb, "%s [PRC: %08X]\n", error->failure_reason,
be32_to_cpu(hostrcb->hcam.u.error.prc));
ipr_log_vpd_compact("Remote IOA", hostrcb, &error->vpd);
[SCSI] ipr: Reduce default error log size Since the default error log size has increased on SAS adapters, prevent ipr from logging this additional data unless requested to do so by the user set log level in order to prevent flooding the logs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:42 -07:00
ipr_log_hex_data(ioa_cfg, error->data,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
be32_to_cpu(hostrcb->hcam.length) -
(offsetof(struct ipr_hostrcb_error, u) +
offsetof(struct ipr_hostrcb_type_07_error, data)));
}
[SCSI] ipr: Add support for logging SAS fabric errors Adds support for logging SAS fabric errors logged by the ipr firmware. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:35 -07:00
static const struct {
u8 active;
char *desc;
} path_active_desc[] = {
{ IPR_PATH_NO_INFO, "Path" },
{ IPR_PATH_ACTIVE, "Active path" },
{ IPR_PATH_NOT_ACTIVE, "Inactive path" }
};
static const struct {
u8 state;
char *desc;
} path_state_desc[] = {
{ IPR_PATH_STATE_NO_INFO, "has no path state information available" },
{ IPR_PATH_HEALTHY, "is healthy" },
{ IPR_PATH_DEGRADED, "is degraded" },
{ IPR_PATH_FAILED, "is failed" }
};
/**
* ipr_log_fabric_path - Log a fabric path error
* @hostrcb: hostrcb struct
* @fabric: fabric descriptor
*
* Return value:
* none
**/
static void ipr_log_fabric_path(struct ipr_hostrcb *hostrcb,
struct ipr_hostrcb_fabric_desc *fabric)
{
int i, j;
u8 path_state = fabric->path_state;
u8 active = path_state & IPR_PATH_ACTIVE_MASK;
u8 state = path_state & IPR_PATH_STATE_MASK;
for (i = 0; i < ARRAY_SIZE(path_active_desc); i++) {
if (path_active_desc[i].active != active)
continue;
for (j = 0; j < ARRAY_SIZE(path_state_desc); j++) {
if (path_state_desc[j].state != state)
continue;
if (fabric->cascaded_expander == 0xff && fabric->phy == 0xff) {
ipr_hcam_err(hostrcb, "%s %s: IOA Port=%d\n",
path_active_desc[i].desc, path_state_desc[j].desc,
fabric->ioa_port);
} else if (fabric->cascaded_expander == 0xff) {
ipr_hcam_err(hostrcb, "%s %s: IOA Port=%d, Phy=%d\n",
path_active_desc[i].desc, path_state_desc[j].desc,
fabric->ioa_port, fabric->phy);
} else if (fabric->phy == 0xff) {
ipr_hcam_err(hostrcb, "%s %s: IOA Port=%d, Cascade=%d\n",
path_active_desc[i].desc, path_state_desc[j].desc,
fabric->ioa_port, fabric->cascaded_expander);
} else {
ipr_hcam_err(hostrcb, "%s %s: IOA Port=%d, Cascade=%d, Phy=%d\n",
path_active_desc[i].desc, path_state_desc[j].desc,
fabric->ioa_port, fabric->cascaded_expander, fabric->phy);
}
return;
}
}
ipr_err("Path state=%02X IOA Port=%d Cascade=%d Phy=%d\n", path_state,
fabric->ioa_port, fabric->cascaded_expander, fabric->phy);
}
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
/**
* ipr_log64_fabric_path - Log a fabric path error
* @hostrcb: hostrcb struct
* @fabric: fabric descriptor
*
* Return value:
* none
**/
static void ipr_log64_fabric_path(struct ipr_hostrcb *hostrcb,
struct ipr_hostrcb64_fabric_desc *fabric)
{
int i, j;
u8 path_state = fabric->path_state;
u8 active = path_state & IPR_PATH_ACTIVE_MASK;
u8 state = path_state & IPR_PATH_STATE_MASK;
char buffer[IPR_MAX_RES_PATH_LENGTH];
for (i = 0; i < ARRAY_SIZE(path_active_desc); i++) {
if (path_active_desc[i].active != active)
continue;
for (j = 0; j < ARRAY_SIZE(path_state_desc); j++) {
if (path_state_desc[j].state != state)
continue;
ipr_hcam_err(hostrcb, "%s %s: Resource Path=%s\n",
path_active_desc[i].desc, path_state_desc[j].desc,
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(hostrcb->ioa_cfg,
fabric->res_path,
buffer, sizeof(buffer)));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
return;
}
}
ipr_err("Path state=%02X Resource Path=%s\n", path_state,
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(hostrcb->ioa_cfg, fabric->res_path,
buffer, sizeof(buffer)));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
}
[SCSI] ipr: Add support for logging SAS fabric errors Adds support for logging SAS fabric errors logged by the ipr firmware. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:35 -07:00
static const struct {
u8 type;
char *desc;
} path_type_desc[] = {
{ IPR_PATH_CFG_IOA_PORT, "IOA port" },
{ IPR_PATH_CFG_EXP_PORT, "Expander port" },
{ IPR_PATH_CFG_DEVICE_PORT, "Device port" },
{ IPR_PATH_CFG_DEVICE_LUN, "Device LUN" }
};
static const struct {
u8 status;
char *desc;
} path_status_desc[] = {
{ IPR_PATH_CFG_NO_PROB, "Functional" },
{ IPR_PATH_CFG_DEGRADED, "Degraded" },
{ IPR_PATH_CFG_FAILED, "Failed" },
{ IPR_PATH_CFG_SUSPECT, "Suspect" },
{ IPR_PATH_NOT_DETECTED, "Missing" },
{ IPR_PATH_INCORRECT_CONN, "Incorrectly connected" }
};
static const char *link_rate[] = {
"unknown",
"disabled",
"phy reset problem",
"spinup hold",
"port selector",
"unknown",
"unknown",
"unknown",
"1.5Gbps",
"3.0Gbps",
"unknown",
"unknown",
"unknown",
"unknown",
"unknown",
"unknown"
};
/**
* ipr_log_path_elem - Log a fabric path element.
* @hostrcb: hostrcb struct
* @cfg: fabric path element struct
*
* Return value:
* none
**/
static void ipr_log_path_elem(struct ipr_hostrcb *hostrcb,
struct ipr_hostrcb_config_element *cfg)
{
int i, j;
u8 type = cfg->type_status & IPR_PATH_CFG_TYPE_MASK;
u8 status = cfg->type_status & IPR_PATH_CFG_STATUS_MASK;
if (type == IPR_PATH_CFG_NOT_EXIST)
return;
for (i = 0; i < ARRAY_SIZE(path_type_desc); i++) {
if (path_type_desc[i].type != type)
continue;
for (j = 0; j < ARRAY_SIZE(path_status_desc); j++) {
if (path_status_desc[j].status != status)
continue;
if (type == IPR_PATH_CFG_IOA_PORT) {
ipr_hcam_err(hostrcb, "%s %s: Phy=%d, Link rate=%s, WWN=%08X%08X\n",
path_status_desc[j].desc, path_type_desc[i].desc,
cfg->phy, link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]), be32_to_cpu(cfg->wwid[1]));
} else {
if (cfg->cascaded_expander == 0xff && cfg->phy == 0xff) {
ipr_hcam_err(hostrcb, "%s %s: Link rate=%s, WWN=%08X%08X\n",
path_status_desc[j].desc, path_type_desc[i].desc,
link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]), be32_to_cpu(cfg->wwid[1]));
} else if (cfg->cascaded_expander == 0xff) {
ipr_hcam_err(hostrcb, "%s %s: Phy=%d, Link rate=%s, "
"WWN=%08X%08X\n", path_status_desc[j].desc,
path_type_desc[i].desc, cfg->phy,
link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]), be32_to_cpu(cfg->wwid[1]));
} else if (cfg->phy == 0xff) {
ipr_hcam_err(hostrcb, "%s %s: Cascade=%d, Link rate=%s, "
"WWN=%08X%08X\n", path_status_desc[j].desc,
path_type_desc[i].desc, cfg->cascaded_expander,
link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]), be32_to_cpu(cfg->wwid[1]));
} else {
ipr_hcam_err(hostrcb, "%s %s: Cascade=%d, Phy=%d, Link rate=%s "
"WWN=%08X%08X\n", path_status_desc[j].desc,
path_type_desc[i].desc, cfg->cascaded_expander, cfg->phy,
link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]), be32_to_cpu(cfg->wwid[1]));
}
}
return;
}
}
ipr_hcam_err(hostrcb, "Path element=%02X: Cascade=%d Phy=%d Link rate=%s "
"WWN=%08X%08X\n", cfg->type_status, cfg->cascaded_expander, cfg->phy,
link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]), be32_to_cpu(cfg->wwid[1]));
}
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
/**
* ipr_log64_path_elem - Log a fabric path element.
* @hostrcb: hostrcb struct
* @cfg: fabric path element struct
*
* Return value:
* none
**/
static void ipr_log64_path_elem(struct ipr_hostrcb *hostrcb,
struct ipr_hostrcb64_config_element *cfg)
{
int i, j;
u8 desc_id = cfg->descriptor_id & IPR_DESCRIPTOR_MASK;
u8 type = cfg->type_status & IPR_PATH_CFG_TYPE_MASK;
u8 status = cfg->type_status & IPR_PATH_CFG_STATUS_MASK;
char buffer[IPR_MAX_RES_PATH_LENGTH];
if (type == IPR_PATH_CFG_NOT_EXIST || desc_id != IPR_DESCRIPTOR_SIS64)
return;
for (i = 0; i < ARRAY_SIZE(path_type_desc); i++) {
if (path_type_desc[i].type != type)
continue;
for (j = 0; j < ARRAY_SIZE(path_status_desc); j++) {
if (path_status_desc[j].status != status)
continue;
ipr_hcam_err(hostrcb, "%s %s: Resource Path=%s, Link rate=%s, WWN=%08X%08X\n",
path_status_desc[j].desc, path_type_desc[i].desc,
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(hostrcb->ioa_cfg,
cfg->res_path, buffer, sizeof(buffer)),
link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]),
be32_to_cpu(cfg->wwid[1]));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
return;
}
}
ipr_hcam_err(hostrcb, "Path element=%02X: Resource Path=%s, Link rate=%s "
"WWN=%08X%08X\n", cfg->type_status,
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(hostrcb->ioa_cfg,
cfg->res_path, buffer, sizeof(buffer)),
link_rate[cfg->link_rate & IPR_PHY_LINK_RATE_MASK],
be32_to_cpu(cfg->wwid[0]), be32_to_cpu(cfg->wwid[1]));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
}
[SCSI] ipr: Add support for logging SAS fabric errors Adds support for logging SAS fabric errors logged by the ipr firmware. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:35 -07:00
/**
* ipr_log_fabric_error - Log a fabric error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_fabric_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
struct ipr_hostrcb_type_20_error *error;
struct ipr_hostrcb_fabric_desc *fabric;
struct ipr_hostrcb_config_element *cfg;
int i, add_len;
error = &hostrcb->hcam.u.error.u.type_20_error;
error->failure_reason[sizeof(error->failure_reason) - 1] = '\0';
ipr_hcam_err(hostrcb, "%s\n", error->failure_reason);
add_len = be32_to_cpu(hostrcb->hcam.length) -
(offsetof(struct ipr_hostrcb_error, u) +
offsetof(struct ipr_hostrcb_type_20_error, desc));
for (i = 0, fabric = error->desc; i < error->num_entries; i++) {
ipr_log_fabric_path(hostrcb, fabric);
for_each_fabric_cfg(fabric, cfg)
ipr_log_path_elem(hostrcb, cfg);
add_len -= be16_to_cpu(fabric->length);
fabric = (struct ipr_hostrcb_fabric_desc *)
((unsigned long)fabric + be16_to_cpu(fabric->length));
}
ipr: Endian / sparse fixes Some misc fixes for endianness checking with sparse so sparse with endian checking now runs clean. Fixes a minor bug in the process which was uncovered by sparse which would result in unnecessary error recovery for check conditions. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-06-11 19:45:20 -06:00
ipr_log_hex_data(ioa_cfg, (__be32 *)fabric, add_len);
[SCSI] ipr: Add support for logging SAS fabric errors Adds support for logging SAS fabric errors logged by the ipr firmware. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:35 -07:00
}
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
/**
* ipr_log_sis64_array_error - Log a sis64 array error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_sis64_array_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
int i, num_entries;
struct ipr_hostrcb_type_24_error *error;
struct ipr_hostrcb64_array_data_entry *array_entry;
char buffer[IPR_MAX_RES_PATH_LENGTH];
const u8 zero_sn[IPR_SERIAL_NUM_LEN] = { [0 ... IPR_SERIAL_NUM_LEN-1] = '0' };
error = &hostrcb->hcam.u.error64.u.type_24_error;
ipr_err_separator;
ipr_err("RAID %s Array Configuration: %s\n",
error->protection_level,
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(ioa_cfg, error->last_res_path,
buffer, sizeof(buffer)));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
ipr_err_separator;
array_entry = error->array_member;
[SCSI] ipr: fix array error logging The structure definitions for reporting array errors did not have the correct size for the Array WWID field. This patch fixes those definitions. It also fixes part of the output formatting that did not have newlines and fixes size calculations. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-09-27 11:45:28 -06:00
num_entries = min_t(u32, error->num_entries,
ARRAY_SIZE(error->array_member));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
for (i = 0; i < num_entries; i++, array_entry++) {
if (!memcmp(array_entry->vpd.vpd.sn, zero_sn, IPR_SERIAL_NUM_LEN))
continue;
if (error->exposed_mode_adn == i)
ipr_err("Exposed Array Member %d:\n", i);
else
ipr_err("Array Member %d:\n", i);
ipr_err("Array Member %d:\n", i);
ipr_log_ext_vpd(&array_entry->vpd);
[SCSI] ipr: fix array error logging The structure definitions for reporting array errors did not have the correct size for the Array WWID field. This patch fixes those definitions. It also fixes part of the output formatting that did not have newlines and fixes size calculations. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-09-27 11:45:28 -06:00
ipr_err("Current Location: %s\n",
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(ioa_cfg, array_entry->res_path,
buffer, sizeof(buffer)));
[SCSI] ipr: fix array error logging The structure definitions for reporting array errors did not have the correct size for the Array WWID field. This patch fixes those definitions. It also fixes part of the output formatting that did not have newlines and fixes size calculations. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-09-27 11:45:28 -06:00
ipr_err("Expected Location: %s\n",
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(ioa_cfg,
array_entry->expected_res_path,
buffer, sizeof(buffer)));
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
ipr_err_separator;
}
}
/**
* ipr_log_sis64_fabric_error - Log a sis64 fabric error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_sis64_fabric_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
struct ipr_hostrcb_type_30_error *error;
struct ipr_hostrcb64_fabric_desc *fabric;
struct ipr_hostrcb64_config_element *cfg;
int i, add_len;
error = &hostrcb->hcam.u.error64.u.type_30_error;
error->failure_reason[sizeof(error->failure_reason) - 1] = '\0';
ipr_hcam_err(hostrcb, "%s\n", error->failure_reason);
add_len = be32_to_cpu(hostrcb->hcam.length) -
(offsetof(struct ipr_hostrcb64_error, u) +
offsetof(struct ipr_hostrcb_type_30_error, desc));
for (i = 0, fabric = error->desc; i < error->num_entries; i++) {
ipr_log64_fabric_path(hostrcb, fabric);
for_each_fabric_cfg(fabric, cfg)
ipr_log64_path_elem(hostrcb, cfg);
add_len -= be16_to_cpu(fabric->length);
fabric = (struct ipr_hostrcb64_fabric_desc *)
((unsigned long)fabric + be16_to_cpu(fabric->length));
}
ipr: Endian / sparse fixes Some misc fixes for endianness checking with sparse so sparse with endian checking now runs clean. Fixes a minor bug in the process which was uncovered by sparse which would result in unnecessary error recovery for check conditions. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-06-11 19:45:20 -06:00
ipr_log_hex_data(ioa_cfg, (__be32 *)fabric, add_len);
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
}
scsi: ipr: Format HCAM overlay ID 0x41 This patch adds formatting error overlay 0x41 to improve debug capabilities. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-06-06 09:01:36 -06:00
/**
* ipr_log_sis64_service_required_error - Log a sis64 service required error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_sis64_service_required_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
struct ipr_hostrcb_type_41_error *error;
error = &hostrcb->hcam.u.error64.u.type_41_error;
error->failure_reason[sizeof(error->failure_reason) - 1] = '\0';
ipr_err("Primary Failure Reason: %s\n", error->failure_reason);
ipr_log_hex_data(ioa_cfg, error->data,
be32_to_cpu(hostrcb->hcam.length) -
(offsetof(struct ipr_hostrcb_error, u) +
offsetof(struct ipr_hostrcb_type_41_error, data)));
}
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
/**
* ipr_log_generic_error - Log an adapter error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_generic_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
[SCSI] ipr: Reduce default error log size Since the default error log size has increased on SAS adapters, prevent ipr from logging this additional data unless requested to do so by the user set log level in order to prevent flooding the logs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:42 -07:00
ipr_log_hex_data(ioa_cfg, hostrcb->hcam.u.raw.data,
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
be32_to_cpu(hostrcb->hcam.length));
}
[SCSI] ipr: Format HCAM overlay ID 0x21 This patch adds formatting error overlay 0x21 to improve debug capabilities. [jejb: checkpatch fixes] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-03-12 15:08:51 -06:00
/**
* ipr_log_sis64_device_error - Log a cache error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* Return value:
* none
**/
static void ipr_log_sis64_device_error(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
struct ipr_hostrcb_type_21_error *error;
char buffer[IPR_MAX_RES_PATH_LENGTH];
error = &hostrcb->hcam.u.error64.u.type_21_error;
ipr_err("-----Failing Device Information-----\n");
ipr_err("World Wide Unique ID: %08X%08X%08X%08X\n",
be32_to_cpu(error->wwn[0]), be32_to_cpu(error->wwn[1]),
be32_to_cpu(error->wwn[2]), be32_to_cpu(error->wwn[3]));
ipr_err("Device Resource Path: %s\n",
__ipr_format_res_path(error->res_path,
buffer, sizeof(buffer)));
error->primary_problem_desc[sizeof(error->primary_problem_desc) - 1] = '\0';
error->second_problem_desc[sizeof(error->second_problem_desc) - 1] = '\0';
ipr_err("Primary Problem Description: %s\n", error->primary_problem_desc);
ipr_err("Secondary Problem Description: %s\n", error->second_problem_desc);
ipr_err("SCSI Sense Data:\n");
ipr_log_hex_data(ioa_cfg, error->sense_data, sizeof(error->sense_data));
ipr_err("SCSI Command Descriptor Block: \n");
ipr_log_hex_data(ioa_cfg, error->cdb, sizeof(error->cdb));
ipr_err("Additional IOA Data:\n");
ipr_log_hex_data(ioa_cfg, error->ioa_data, be32_to_cpu(error->length_of_error));
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_get_error - Find the specfied IOASC in the ipr_error_table.
* @ioasc: IOASC
*
* This function will return the index of into the ipr_error_table
* for the specified IOASC. If the IOASC is not in the table,
* 0 will be returned, which points to the entry used for unknown errors.
*
* Return value:
* index into the ipr_error_table
**/
static u32 ipr_get_error(u32 ioasc)
{
int i;
for (i = 0; i < ARRAY_SIZE(ipr_error_table); i++)
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (ipr_error_table[i].ioasc == (ioasc & IPR_IOASC_IOASC_MASK))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return i;
return 0;
}
/**
* ipr_handle_log_data - Log an adapter error.
* @ioa_cfg: ioa config struct
* @hostrcb: hostrcb struct
*
* This function logs an adapter error to the system.
*
* Return value:
* none
**/
static void ipr_handle_log_data(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_hostrcb *hostrcb)
{
u32 ioasc;
int error_index;
ipr: don't log error messages when applications issues illegal requests Failing Device information are logged when IOA firmware detected these illegal request such as IOA firmware doesn't support inquiry with page code 2. The patch fixes the issue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Tested-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-09-24 15:25:47 -06:00
struct ipr_hostrcb_type_21_error *error;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (hostrcb->hcam.notify_type != IPR_HOST_RCB_NOTIF_TYPE_ERROR_LOG_ENTRY)
return;
if (hostrcb->hcam.notifications_lost == IPR_HOST_RCB_NOTIFICATIONS_LOST)
dev_err(&ioa_cfg->pdev->dev, "Error notifications lost\n");
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
if (ioa_cfg->sis64)
ioasc = be32_to_cpu(hostrcb->hcam.u.error64.fd_ioasc);
else
ioasc = be32_to_cpu(hostrcb->hcam.u.error.fd_ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
if (!ioa_cfg->sis64 && (ioasc == IPR_IOASC_BUS_WAS_RESET ||
ioasc == IPR_IOASC_BUS_WAS_RESET_BY_OTHER)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Tell the midlayer we had a bus reset so it will handle the UA properly */
scsi_report_bus_reset(ioa_cfg->host,
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
hostrcb->hcam.u.error.fd_res_addr.bus);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
error_index = ipr_get_error(ioasc);
if (!ipr_error_table[error_index].log_hcam)
return;
ipr: don't log error messages when applications issues illegal requests Failing Device information are logged when IOA firmware detected these illegal request such as IOA firmware doesn't support inquiry with page code 2. The patch fixes the issue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Tested-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-09-24 15:25:47 -06:00
if (ioasc == IPR_IOASC_HW_CMD_FAILED &&
hostrcb->hcam.overlay_id == IPR_HOST_RCB_OVERLAY_ID_21) {
error = &hostrcb->hcam.u.error64.u.type_21_error;
if (((be32_to_cpu(error->sense_data[0]) & 0x0000ff00) >> 8) == ILLEGAL_REQUEST &&
ioa_cfg->log_level <= IPR_DEFAULT_LOG_LEVEL)
return;
}
[SCSI] ipr: Add support for logging SAS fabric errors Adds support for logging SAS fabric errors logged by the ipr firmware. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:35 -07:00
ipr_hcam_err(hostrcb, "%s\n", ipr_error_table[error_index].error);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Set indication we have logged an error */
ioa_cfg->errors_logged++;
[SCSI] ipr: Improve async error logging level control Add the ability to control how much error data gets logged on a per error basis. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:30 -06:00
if (ioa_cfg->log_level < ipr_error_table[error_index].log_hcam)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
[SCSI] ipr: Error logging cleanup Simplify error logging path, sanitize error length returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:47 -07:00
if (be32_to_cpu(hostrcb->hcam.length) > sizeof(hostrcb->hcam.u.raw))
hostrcb->hcam.length = cpu_to_be32(sizeof(hostrcb->hcam.u.raw));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
switch (hostrcb->hcam.overlay_id) {
case IPR_HOST_RCB_OVERLAY_ID_2:
ipr_log_cache_error(ioa_cfg, hostrcb);
break;
case IPR_HOST_RCB_OVERLAY_ID_3:
ipr_log_config_error(ioa_cfg, hostrcb);
break;
case IPR_HOST_RCB_OVERLAY_ID_4:
case IPR_HOST_RCB_OVERLAY_ID_6:
ipr_log_array_error(ioa_cfg, hostrcb);
break;
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
case IPR_HOST_RCB_OVERLAY_ID_7:
ipr_log_dual_ioa_error(ioa_cfg, hostrcb);
break;
[SCSI] ipr: New adapter error types Handle some new types of ipr errors that can be returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:15 -07:00
case IPR_HOST_RCB_OVERLAY_ID_12:
ipr_log_enhanced_cache_error(ioa_cfg, hostrcb);
break;
case IPR_HOST_RCB_OVERLAY_ID_13:
ipr_log_enhanced_config_error(ioa_cfg, hostrcb);
break;
case IPR_HOST_RCB_OVERLAY_ID_14:
case IPR_HOST_RCB_OVERLAY_ID_16:
ipr_log_enhanced_array_error(ioa_cfg, hostrcb);
break;
case IPR_HOST_RCB_OVERLAY_ID_17:
ipr_log_enhanced_dual_ioa_error(ioa_cfg, hostrcb);
break;
[SCSI] ipr: Add support for logging SAS fabric errors Adds support for logging SAS fabric errors logged by the ipr firmware. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:35 -07:00
case IPR_HOST_RCB_OVERLAY_ID_20:
ipr_log_fabric_error(ioa_cfg, hostrcb);
break;
[SCSI] ipr: Format HCAM overlay ID 0x21 This patch adds formatting error overlay 0x21 to improve debug capabilities. [jejb: checkpatch fixes] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-03-12 15:08:51 -06:00
case IPR_HOST_RCB_OVERLAY_ID_21:
ipr_log_sis64_device_error(ioa_cfg, hostrcb);
break;
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
case IPR_HOST_RCB_OVERLAY_ID_23:
ipr_log_sis64_config_error(ioa_cfg, hostrcb);
break;
case IPR_HOST_RCB_OVERLAY_ID_24:
case IPR_HOST_RCB_OVERLAY_ID_26:
ipr_log_sis64_array_error(ioa_cfg, hostrcb);
break;
case IPR_HOST_RCB_OVERLAY_ID_30:
ipr_log_sis64_fabric_error(ioa_cfg, hostrcb);
break;
scsi: ipr: Format HCAM overlay ID 0x41 This patch adds formatting error overlay 0x41 to improve debug capabilities. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-06-06 09:01:36 -06:00
case IPR_HOST_RCB_OVERLAY_ID_41:
ipr_log_sis64_service_required_error(ioa_cfg, hostrcb);
break;
[SCSI] ipr: Error logging cleanup Simplify error logging path, sanitize error length returned by the adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:47 -07:00
case IPR_HOST_RCB_OVERLAY_ID_1:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
case IPR_HOST_RCB_OVERLAY_ID_DEFAULT:
default:
[SCSI] ipr: Handle unknown errors Better handle errors received which are not known to the device driver. Just dump the hex data so that we have a hope of figuring out what went wrong. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:00:41 -07:00
ipr_log_generic_error(ioa_cfg, hostrcb);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
break;
}
}
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
static struct ipr_hostrcb *ipr_get_free_hostrcb(struct ipr_ioa_cfg *ioa)
{
struct ipr_hostrcb *hostrcb;
hostrcb = list_first_entry_or_null(&ioa->hostrcb_free_q,
struct ipr_hostrcb, queue);
if (unlikely(!hostrcb)) {
dev_info(&ioa->pdev->dev, "Reclaiming async error buffers.");
hostrcb = list_first_entry_or_null(&ioa->hostrcb_report_q,
struct ipr_hostrcb, queue);
}
list_del_init(&hostrcb->queue);
return hostrcb;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_process_error - Op done function for an adapter error log.
* @ipr_cmd: ipr command struct
*
* This function is the op done function for an error log host
* controlled async from the adapter. It will log the error and
* send the HCAM back to the adapter.
*
* Return value:
* none
**/
static void ipr_process_error(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_hostrcb *hostrcb = ipr_cmd->u.hostrcb;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
u32 fd_ioasc;
if (ioa_cfg->sis64)
fd_ioasc = be32_to_cpu(hostrcb->hcam.u.error64.fd_ioasc);
else
fd_ioasc = be32_to_cpu(hostrcb->hcam.u.error.fd_ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
list_del_init(&hostrcb->queue);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ioasc) {
ipr_handle_log_data(ioa_cfg, hostrcb);
[SCSI] ipr: Handle IOA reset request In ipr dual adapter configurations, the ipr adapter firmware may require an adapter reset for various reasons. The reset is requested by the adapter firmware logging an error with an IOASC of 0x02048000. Add support to log this error, and reset the adapter. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:12 -06:00
if (fd_ioasc == IPR_IOASC_NR_IOA_RESET_REQUIRED)
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_ABBREV);
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
} else if (ioasc != IPR_IOASC_IOA_WAS_RESET &&
ioasc != IPR_IOASC_ABORTED_CMD_TERM_BY_HOST) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_err(&ioa_cfg->pdev->dev,
"Host RCB failed with IOASC: 0x%08X\n", ioasc);
}
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
list_add_tail(&hostrcb->queue, &ioa_cfg->hostrcb_report_q);
scsi: ipr: Fix async error WARN_ON Commit afc3f83cb4a5 ("scsi: ipr: Add asynchronous error notification") introduced the warn on shown below. To fix this, rather than attempting to send the KOBJ_CHANGE uevent from interrupt context, which is what is causing the WARN_ON, just wake the ipr worker thread which will send a KOBJ_CHANGE uevent. [ 142.278120] WARNING: CPU: 15 PID: 0 at kernel/softirq.c:161 __local_bh_enable_ip+0x7c/0xd0 [ 142.278124] Modules linked in: ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw iptable_filter ses enclosure scsi_transport_sas sg pseries_rng nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables xfs libcrc32c sr_mod sd_mod cdrom ipr libata ibmvscsi scsi_transport_srp ibmveth dm_mirror dm_region_hash dm_log dm_mod [ 142.278208] CPU: 15 PID: 0 Comm: swapper/15 Not tainted 4.8.0.ipr+ #21 [ 142.278213] task: c00000010cf24480 task.stack: c00000010cfec000 [ 142.278217] NIP: c0000000000c0c7c LR: c000000000881778 CTR: c0000000003c5bf0 [ 142.278221] REGS: c00000010cfef080 TRAP: 0700 Not tainted (4.8.0.ipr+) [ 142.278224] MSR: 8000000000029033 <SF,EE,ME,IR,DR,RI,LE> CR: 28008022 XER: 2000000f [ 142.278236] CFAR: c0000000000c0c20 SOFTE: 0 GPR00: c000000000706c78 c00000010cfef300 c000000000f91d00 c000000000706c78 GPR04: 0000000000000200 c000000000f7bc80 0000000000000000 00000000024000c0 GPR08: 0000000000000000 0000000000000001 c000000000ee1d00 c000000000a9bdd0 GPR12: c0000000003c5bf0 c00000000eb22d00 c000000100ca3880 c00000020ed38400 GPR16: 0000000000000000 0000000000000000 c000000100940508 0000000000000000 GPR20: 0000000000000000 0000000000000000 0000000000000000 00000000024000c0 GPR24: c0000000004588e0 c00000010863bd00 c00000010863bd00 c0000000013773f8 GPR28: c000000000f7bc80 0000000000000000 ffffffffffffffff c000000000f7bcd8 [ 142.278290] NIP [c0000000000c0c7c] __local_bh_enable_ip+0x7c/0xd0 [ 142.278296] LR [c000000000881778] _raw_spin_unlock_bh+0x38/0x60 [ 142.278299] Call Trace: [ 142.278303] [c00000010cfef300] [c000000000f7bc80] init_net+0x0/0x1900 (unreliable) [ 142.278310] [c00000010cfef320] [c000000000706c78] peernet2id+0x58/0x80 [ 142.278316] [c00000010cfef370] [c00000000075caec] netlink_broadcast_filtered+0x30c/0x550 [ 142.278323] [c00000010cfef430] [c000000000459078] kobject_uevent_env+0x588/0x780 [ 142.278331] [c00000010cfef510] [d000000003163a6c] ipr_process_error+0x11c/0x240 [ipr] [ 142.278337] [c00000010cfef5c0] [d000000003152298] ipr_fail_all_ops+0x108/0x220 [ipr] [ 142.278343] [c00000010cfef670] [d0000000031643f8] ipr_reset_restore_cfg_space+0xa8/0x240 [ipr] [ 142.278350] [c00000010cfef6f0] [d000000003158a00] ipr_reset_ioa_job+0x80/0xe0 [ipr] [ 142.278356] [c00000010cfef720] [d000000003153f78] ipr_reset_timer_done+0xa8/0xe0 [ipr] [ 142.278363] [c00000010cfef770] [c000000000149c88] call_timer_fn+0x58/0x1c0 [ 142.278368] [c00000010cfef800] [c000000000149f60] expire_timers+0x140/0x200 [ 142.278373] [c00000010cfef870] [c00000000014a0e8] run_timer_softirq+0xc8/0x230 [ 142.278379] [c00000010cfef900] [c0000000000c0844] __do_softirq+0x164/0x3c0 [ 142.278384] [c00000010cfef9f0] [c0000000000c0f18] irq_exit+0x1a8/0x1c0 [ 142.278389] [c00000010cfefa20] [c000000000020b54] timer_interrupt+0xa4/0xe0 [ 142.278394] [c00000010cfefa50] [c000000000002414] decrementer_common+0x114/0x180 Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-10-13 13:45:24 -06:00
schedule_work(&ioa_cfg->work_q);
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
hostrcb = ipr_get_free_hostrcb(ioa_cfg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_send_hcam(ioa_cfg, IPR_HCAM_CDB_OP_CODE_LOG_DATA, hostrcb);
}
/**
* ipr_timeout - An internally generated op has timed out.
* @ipr_cmd: ipr command struct
*
* This function blocks host requests and initiates an
* adapter reset.
*
* Return value:
* none
**/
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
static void ipr_timeout(struct timer_list *t)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
struct ipr_cmnd *ipr_cmd = from_timer(ipr_cmd, t, timer);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
unsigned long lock_flags = 0;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ioa_cfg->errors_logged++;
dev_err(&ioa_cfg->pdev->dev,
"Adapter being reset due to command timeout.\n");
if (WAIT_FOR_DUMP == ioa_cfg->sdt_state)
ioa_cfg->sdt_state = GET_DUMP;
if (!ioa_cfg->in_reset_reload || ioa_cfg->reset_cmd == ipr_cmd)
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
LEAVE;
}
/**
* ipr_oper_timeout - Adapter timed out transitioning to operational
* @ipr_cmd: ipr command struct
*
* This function blocks host requests and initiates an
* adapter reset.
*
* Return value:
* none
**/
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
static void ipr_oper_timeout(struct timer_list *t)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
struct ipr_cmnd *ipr_cmd = from_timer(ipr_cmd, t, timer);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
unsigned long lock_flags = 0;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ioa_cfg->errors_logged++;
dev_err(&ioa_cfg->pdev->dev,
"Adapter timed out transitioning to operational.\n");
if (WAIT_FOR_DUMP == ioa_cfg->sdt_state)
ioa_cfg->sdt_state = GET_DUMP;
if (!ioa_cfg->in_reset_reload || ioa_cfg->reset_cmd == ipr_cmd) {
if (ipr_fastfail)
ioa_cfg->reset_retries += IPR_NUM_RESET_RELOAD_RETRIES;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
LEAVE;
}
/**
* ipr_find_ses_entry - Find matching SES in SES table
* @res: resource entry struct of SES
*
* Return value:
* pointer to SES table entry / NULL on failure
**/
static const struct ipr_ses_table_entry *
ipr_find_ses_entry(struct ipr_resource_entry *res)
{
int i, j, matches;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
struct ipr_std_inq_vpids *vpids;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
const struct ipr_ses_table_entry *ste = ipr_ses_table;
for (i = 0; i < ARRAY_SIZE(ipr_ses_table); i++, ste++) {
for (j = 0, matches = 0; j < IPR_PROD_ID_LEN; j++) {
if (ste->compare_product_id_byte[j] == 'X') {
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
vpids = &res->std_inq_data.vpids;
if (vpids->product_id[j] == ste->product_id[j])
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
matches++;
else
break;
} else
matches++;
}
if (matches == IPR_PROD_ID_LEN)
return ste;
}
return NULL;
}
/**
* ipr_get_max_scsi_speed - Determine max SCSI speed for a given bus
* @ioa_cfg: ioa config struct
* @bus: SCSI bus
* @bus_width: bus width
*
* Return value:
* SCSI bus speed in units of 100KHz, 1600 is 160 MHz
* For a 2-byte wide SCSI bus, the maximum transfer speed is
* twice the maximum transfer rate (e.g. for a wide enabled bus,
* max 160MHz = max 320MB/sec).
**/
static u32 ipr_get_max_scsi_speed(struct ipr_ioa_cfg *ioa_cfg, u8 bus, u8 bus_width)
{
struct ipr_resource_entry *res;
const struct ipr_ses_table_entry *ste;
u32 max_xfer_rate = IPR_MAX_SCSI_RATE(bus_width);
/* Loop through each config table entry in the config table buffer */
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (!(IPR_IS_SES_DEVICE(res->std_inq_data)))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
continue;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (bus != res->bus)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
continue;
if (!(ste = ipr_find_ses_entry(res)))
continue;
max_xfer_rate = (ste->max_bus_speed_limit * 10) / (bus_width / 8);
}
return max_xfer_rate;
}
/**
* ipr_wait_iodbg_ack - Wait for an IODEBUG ACK from the IOA
* @ioa_cfg: ioa config struct
* @max_delay: max delay in micro-seconds to wait
*
* Waits for an IODEBUG ACK from the IOA, doing busy looping.
*
* Return value:
* 0 on success / other on failure
**/
static int ipr_wait_iodbg_ack(struct ipr_ioa_cfg *ioa_cfg, int max_delay)
{
volatile u32 pcii_reg;
int delay = 1;
/* Read interrupt reg until IOA signals IO Debug Acknowledge */
while (delay < max_delay) {
pcii_reg = readl(ioa_cfg->regs.sense_interrupt_reg);
if (pcii_reg & IPR_PCII_IO_DEBUG_ACKNOWLEDGE)
return 0;
/* udelay cannot be used if delay is more than a few milliseconds */
if ((delay / 1000) > MAX_UDELAY_MS)
mdelay(delay / 1000);
else
udelay(delay);
delay += delay;
}
return -EIO;
}
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
/**
* ipr_get_sis64_dump_data_section - Dump IOA memory
* @ioa_cfg: ioa config struct
* @start_addr: adapter address to dump
* @dest: destination kernel buffer
* @length_in_words: length to dump in 4 byte words
*
* Return value:
* 0 on success
**/
static int ipr_get_sis64_dump_data_section(struct ipr_ioa_cfg *ioa_cfg,
u32 start_addr,
__be32 *dest, u32 length_in_words)
{
int i;
for (i = 0; i < length_in_words; i++) {
writel(start_addr+(i*4), ioa_cfg->regs.dump_addr_reg);
*dest = cpu_to_be32(readl(ioa_cfg->regs.dump_data_reg));
dest++;
}
return 0;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_get_ldump_data_section - Dump IOA memory
* @ioa_cfg: ioa config struct
* @start_addr: adapter address to dump
* @dest: destination kernel buffer
* @length_in_words: length to dump in 4 byte words
*
* Return value:
* 0 on success / -EIO on failure
**/
static int ipr_get_ldump_data_section(struct ipr_ioa_cfg *ioa_cfg,
u32 start_addr,
__be32 *dest, u32 length_in_words)
{
volatile u32 temp_pcii_reg;
int i, delay = 0;
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
if (ioa_cfg->sis64)
return ipr_get_sis64_dump_data_section(ioa_cfg, start_addr,
dest, length_in_words);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Write IOA interrupt reg starting LDUMP state */
writel((IPR_UPROCI_RESET_ALERT | IPR_UPROCI_IO_DEBUG_ALERT),
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
ioa_cfg->regs.set_uproc_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Wait for IO debug acknowledge */
if (ipr_wait_iodbg_ack(ioa_cfg,
IPR_LDUMP_MAX_LONG_ACK_DELAY_IN_USEC)) {
dev_err(&ioa_cfg->pdev->dev,
"IOA dump long data transfer timeout\n");
return -EIO;
}
/* Signal LDUMP interlocked - clear IO debug ack */
writel(IPR_PCII_IO_DEBUG_ACKNOWLEDGE,
ioa_cfg->regs.clr_interrupt_reg);
/* Write Mailbox with starting address */
writel(start_addr, ioa_cfg->ioa_mailbox);
/* Signal address valid - clear IOA Reset alert */
writel(IPR_UPROCI_RESET_ALERT,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
ioa_cfg->regs.clr_uproc_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
for (i = 0; i < length_in_words; i++) {
/* Wait for IO debug acknowledge */
if (ipr_wait_iodbg_ack(ioa_cfg,
IPR_LDUMP_MAX_SHORT_ACK_DELAY_IN_USEC)) {
dev_err(&ioa_cfg->pdev->dev,
"IOA dump short data transfer timeout\n");
return -EIO;
}
/* Read data from mailbox and increment destination pointer */
*dest = cpu_to_be32(readl(ioa_cfg->ioa_mailbox));
dest++;
/* For all but the last word of data, signal data received */
if (i < (length_in_words - 1)) {
/* Signal dump data received - Clear IO debug Ack */
writel(IPR_PCII_IO_DEBUG_ACKNOWLEDGE,
ioa_cfg->regs.clr_interrupt_reg);
}
}
/* Signal end of block transfer. Set reset alert then clear IO debug ack */
writel(IPR_UPROCI_RESET_ALERT,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
ioa_cfg->regs.set_uproc_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
writel(IPR_UPROCI_IO_DEBUG_ALERT,
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
ioa_cfg->regs.clr_uproc_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Signal dump data received - Clear IO debug Ack */
writel(IPR_PCII_IO_DEBUG_ACKNOWLEDGE,
ioa_cfg->regs.clr_interrupt_reg);
/* Wait for IOA to signal LDUMP exit - IOA reset alert will be cleared */
while (delay < IPR_LDUMP_MAX_SHORT_ACK_DELAY_IN_USEC) {
temp_pcii_reg =
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
readl(ioa_cfg->regs.sense_uproc_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!(temp_pcii_reg & IPR_UPROCI_RESET_ALERT))
return 0;
udelay(10);
delay += 10;
}
return 0;
}
#ifdef CONFIG_SCSI_IPR_DUMP
/**
* ipr_sdt_copy - Copy Smart Dump Table to kernel buffer
* @ioa_cfg: ioa config struct
* @pci_address: adapter address
* @length: length of data to copy
*
* Copy data from PCI adapter to kernel buffer.
* Note: length MUST be a 4 byte multiple
* Return value:
* 0 on success / other on failure
**/
static int ipr_sdt_copy(struct ipr_ioa_cfg *ioa_cfg,
unsigned long pci_address, u32 length)
{
int bytes_copied = 0;
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
int cur_len, rc, rem_len, rem_page_len, max_dump_size;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
__be32 *page;
unsigned long lock_flags = 0;
struct ipr_ioa_dump *ioa_dump = &ioa_cfg->dump->ioa_dump;
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (ioa_cfg->sis64)
max_dump_size = IPR_FMT3_MAX_IOA_DUMP_SIZE;
else
max_dump_size = IPR_FMT2_MAX_IOA_DUMP_SIZE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
while (bytes_copied < length &&
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
(ioa_dump->hdr.len + bytes_copied) < max_dump_size) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioa_dump->page_offset >= PAGE_SIZE ||
ioa_dump->page_offset == 0) {
page = (__be32 *)__get_free_page(GFP_ATOMIC);
if (!page) {
ipr_trace;
return bytes_copied;
}
ioa_dump->page_offset = 0;
ioa_dump->ioa_data[ioa_dump->next_page_index] = page;
ioa_dump->next_page_index++;
} else
page = ioa_dump->ioa_data[ioa_dump->next_page_index - 1];
rem_len = length - bytes_copied;
rem_page_len = PAGE_SIZE - ioa_dump->page_offset;
cur_len = min(rem_len, rem_page_len);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (ioa_cfg->sdt_state == ABORT_DUMP) {
rc = -EIO;
} else {
rc = ipr_get_ldump_data_section(ioa_cfg,
pci_address + bytes_copied,
&page[ioa_dump->page_offset / 4],
(cur_len / sizeof(u32)));
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
if (!rc) {
ioa_dump->page_offset += cur_len;
bytes_copied += cur_len;
} else {
ipr_trace;
break;
}
schedule();
}
return bytes_copied;
}
/**
* ipr_init_dump_entry_hdr - Initialize a dump entry header.
* @hdr: dump entry header struct
*
* Return value:
* nothing
**/
static void ipr_init_dump_entry_hdr(struct ipr_dump_entry_header *hdr)
{
hdr->eye_catcher = IPR_DUMP_EYE_CATCHER;
hdr->num_elems = 1;
hdr->offset = sizeof(*hdr);
hdr->status = IPR_DUMP_STATUS_SUCCESS;
}
/**
* ipr_dump_ioa_type_data - Fill in the adapter type in the dump.
* @ioa_cfg: ioa config struct
* @driver_dump: driver dump struct
*
* Return value:
* nothing
**/
static void ipr_dump_ioa_type_data(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_driver_dump *driver_dump)
{
struct ipr_inquiry_page3 *ucode_vpd = &ioa_cfg->vpd_cbs->page3_data;
ipr_init_dump_entry_hdr(&driver_dump->ioa_type_entry.hdr);
driver_dump->ioa_type_entry.hdr.len =
sizeof(struct ipr_dump_ioa_type_entry) -
sizeof(struct ipr_dump_entry_header);
driver_dump->ioa_type_entry.hdr.data_type = IPR_DUMP_DATA_TYPE_BINARY;
driver_dump->ioa_type_entry.hdr.id = IPR_DUMP_DRIVER_TYPE_ID;
driver_dump->ioa_type_entry.type = ioa_cfg->type;
driver_dump->ioa_type_entry.fw_version = (ucode_vpd->major_release << 24) |
(ucode_vpd->card_type << 16) | (ucode_vpd->minor_release[0] << 8) |
ucode_vpd->minor_release[1];
driver_dump->hdr.num_entries++;
}
/**
* ipr_dump_version_data - Fill in the driver version in the dump.
* @ioa_cfg: ioa config struct
* @driver_dump: driver dump struct
*
* Return value:
* nothing
**/
static void ipr_dump_version_data(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_driver_dump *driver_dump)
{
ipr_init_dump_entry_hdr(&driver_dump->version_entry.hdr);
driver_dump->version_entry.hdr.len =
sizeof(struct ipr_dump_version_entry) -
sizeof(struct ipr_dump_entry_header);
driver_dump->version_entry.hdr.data_type = IPR_DUMP_DATA_TYPE_ASCII;
driver_dump->version_entry.hdr.id = IPR_DUMP_DRIVER_VERSION_ID;
strcpy(driver_dump->version_entry.version, IPR_DRIVER_VERSION);
driver_dump->hdr.num_entries++;
}
/**
* ipr_dump_trace_data - Fill in the IOA trace in the dump.
* @ioa_cfg: ioa config struct
* @driver_dump: driver dump struct
*
* Return value:
* nothing
**/
static void ipr_dump_trace_data(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_driver_dump *driver_dump)
{
ipr_init_dump_entry_hdr(&driver_dump->trace_entry.hdr);
driver_dump->trace_entry.hdr.len =
sizeof(struct ipr_dump_trace_entry) -
sizeof(struct ipr_dump_entry_header);
driver_dump->trace_entry.hdr.data_type = IPR_DUMP_DATA_TYPE_BINARY;
driver_dump->trace_entry.hdr.id = IPR_DUMP_TRACE_ID;
memcpy(driver_dump->trace_entry.trace, ioa_cfg->trace, IPR_TRACE_SIZE);
driver_dump->hdr.num_entries++;
}
/**
* ipr_dump_location_data - Fill in the IOA location in the dump.
* @ioa_cfg: ioa config struct
* @driver_dump: driver dump struct
*
* Return value:
* nothing
**/
static void ipr_dump_location_data(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_driver_dump *driver_dump)
{
ipr_init_dump_entry_hdr(&driver_dump->location_entry.hdr);
driver_dump->location_entry.hdr.len =
sizeof(struct ipr_dump_location_entry) -
sizeof(struct ipr_dump_entry_header);
driver_dump->location_entry.hdr.data_type = IPR_DUMP_DATA_TYPE_ASCII;
driver_dump->location_entry.hdr.id = IPR_DUMP_LOCATION_ID;
[SCSI] struct device - replace bus_id with dev_name(), dev_set_name() [jejb: limit ioctl to returning 20 characters to avoid overrun on long device names and add a few more conversions] Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-12-03 14:41:36 -07:00
strcpy(driver_dump->location_entry.location, dev_name(&ioa_cfg->pdev->dev));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
driver_dump->hdr.num_entries++;
}
/**
* ipr_get_ioa_dump - Perform a dump of the driver and adapter.
* @ioa_cfg: ioa config struct
* @dump: dump struct
*
* Return value:
* nothing
**/
static void ipr_get_ioa_dump(struct ipr_ioa_cfg *ioa_cfg, struct ipr_dump *dump)
{
unsigned long start_addr, sdt_word;
unsigned long lock_flags = 0;
struct ipr_driver_dump *driver_dump = &dump->driver_dump;
struct ipr_ioa_dump *ioa_dump = &dump->ioa_dump;
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
u32 num_entries, max_num_entries, start_off, end_off;
u32 max_dump_size, bytes_to_copy, bytes_copied, rc;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_sdt *sdt;
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
int valid = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
int i;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Stop reading adapter dump prematurely When the ipr driver decides to dump the adapter, it changes the sdt_state to GET_DUMP, then prepares the adapter so that the dump can be read. However, if the ipr worker thread wakes up for some reason before the driver has put the adapter in a state where it can succesfully dump the adapter, the driver will start dumping the adapter too early, which can potentially trigger a BUG check in the pci config blocking API. Fix this by adding a new sdt_state to differentiate between the ipr driver wanting to dump the adapter in the near future and wanting to dump the adapter now. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-09-21 07:51:11 -06:00
if (ioa_cfg->sdt_state != READ_DUMP) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
[SCSI] ipr: fix mailbox register definition and add a delay before reading The definition for the mailbox register for new adapters was incorrect. The value has been updated to the correct offset. After an adapter reset, the mailbox register on the new adapters takes a number of seconds to stabilize. A delay has been added before reading the register. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-11-04 10:36:16 -06:00
if (ioa_cfg->sis64) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
ssleep(IPR_DUMP_DELAY_SECONDS);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
start_addr = readl(ioa_cfg->ioa_mailbox);
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
if (!ioa_cfg->sis64 && !ipr_sdt_is_fmt2(start_addr)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_err(&ioa_cfg->pdev->dev,
"Invalid dump table format: %lx\n", start_addr);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
dev_err(&ioa_cfg->pdev->dev, "Dump of IOA initiated\n");
driver_dump->hdr.eye_catcher = IPR_DUMP_EYE_CATCHER;
/* Initialize the overall dump header */
driver_dump->hdr.len = sizeof(struct ipr_driver_dump);
driver_dump->hdr.num_entries = 1;
driver_dump->hdr.first_entry_offset = sizeof(struct ipr_dump_header);
driver_dump->hdr.status = IPR_DUMP_STATUS_SUCCESS;
driver_dump->hdr.os = IPR_DUMP_OS_LINUX;
driver_dump->hdr.driver_name = IPR_DUMP_DRIVER_NAME;
ipr_dump_version_data(ioa_cfg, driver_dump);
ipr_dump_location_data(ioa_cfg, driver_dump);
ipr_dump_ioa_type_data(ioa_cfg, driver_dump);
ipr_dump_trace_data(ioa_cfg, driver_dump);
/* Update dump_header */
driver_dump->hdr.len += sizeof(struct ipr_dump_entry_header);
/* IOA Dump entry */
ipr_init_dump_entry_hdr(&ioa_dump->hdr);
ioa_dump->hdr.len = 0;
ioa_dump->hdr.data_type = IPR_DUMP_DATA_TYPE_BINARY;
ioa_dump->hdr.id = IPR_DUMP_IOA_DUMP_ID;
/* First entries in sdt are actually a list of dump addresses and
lengths to gather the real dump data. sdt represents the pointer
to the ioa generated dump table. Dump data will be extracted based
on entries in this table */
sdt = &ioa_dump->sdt;
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (ioa_cfg->sis64) {
max_num_entries = IPR_FMT3_NUM_SDT_ENTRIES;
max_dump_size = IPR_FMT3_MAX_IOA_DUMP_SIZE;
} else {
max_num_entries = IPR_FMT2_NUM_SDT_ENTRIES;
max_dump_size = IPR_FMT2_MAX_IOA_DUMP_SIZE;
}
bytes_to_copy = offsetof(struct ipr_sdt, entry) +
(max_num_entries * sizeof(struct ipr_sdt_entry));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
rc = ipr_get_ldump_data_section(ioa_cfg, start_addr, (__be32 *)sdt,
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
bytes_to_copy / sizeof(__be32));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Smart Dump table is ready to use and the first entry is valid */
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
if (rc || ((be32_to_cpu(sdt->hdr.state) != IPR_FMT3_SDT_READY_TO_USE) &&
(be32_to_cpu(sdt->hdr.state) != IPR_FMT2_SDT_READY_TO_USE))) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_err(&ioa_cfg->pdev->dev,
"Dump of IOA failed. Dump table not valid: %d, %X.\n",
rc, be32_to_cpu(sdt->hdr.state));
driver_dump->hdr.status = IPR_DUMP_STATUS_FAILED;
ioa_cfg->sdt_state = DUMP_OBTAINED;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
num_entries = be32_to_cpu(sdt->hdr.num_entries_used);
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (num_entries > max_num_entries)
num_entries = max_num_entries;
/* Update dump length to the actual data to be copied */
dump->driver_dump.hdr.len += sizeof(struct ipr_sdt_header);
if (ioa_cfg->sis64)
dump->driver_dump.hdr.len += num_entries * sizeof(struct ipr_sdt_entry);
else
dump->driver_dump.hdr.len += max_num_entries * sizeof(struct ipr_sdt_entry);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
for (i = 0; i < num_entries; i++) {
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (ioa_dump->hdr.len > max_dump_size) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
driver_dump->hdr.status = IPR_DUMP_STATUS_QUAL_SUCCESS;
break;
}
if (sdt->entry[i].flags & IPR_SDT_VALID_ENTRY) {
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
sdt_word = be32_to_cpu(sdt->entry[i].start_token);
if (ioa_cfg->sis64)
bytes_to_copy = be32_to_cpu(sdt->entry[i].end_token);
else {
start_off = sdt_word & IPR_FMT2_MBX_ADDR_MASK;
end_off = be32_to_cpu(sdt->entry[i].end_token);
if (ipr_sdt_is_fmt2(sdt_word) && sdt_word)
bytes_to_copy = end_off - start_off;
else
valid = 0;
}
if (valid) {
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (bytes_to_copy > max_dump_size) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sdt->entry[i].flags &= ~IPR_SDT_VALID_ENTRY;
continue;
}
/* Copy data from adapter to driver buffers */
bytes_copied = ipr_sdt_copy(ioa_cfg, sdt_word,
bytes_to_copy);
ioa_dump->hdr.len += bytes_copied;
if (bytes_copied != bytes_to_copy) {
driver_dump->hdr.status = IPR_DUMP_STATUS_QUAL_SUCCESS;
break;
}
}
}
}
dev_err(&ioa_cfg->pdev->dev, "Dump of IOA completed.\n");
/* Update dump_header */
driver_dump->hdr.len += ioa_dump->hdr.len;
wmb();
ioa_cfg->sdt_state = DUMP_OBTAINED;
LEAVE;
}
#else
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
#define ipr_get_ioa_dump(ioa_cfg, dump) do { } while (0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
#endif
/**
* ipr_release_dump - Free adapter dump memory
* @kref: kref struct
*
* Return value:
* nothing
**/
static void ipr_release_dump(struct kref *kref)
{
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
struct ipr_dump *dump = container_of(kref, struct ipr_dump, kref);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = dump->ioa_cfg;
unsigned long lock_flags = 0;
int i;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ioa_cfg->dump = NULL;
ioa_cfg->sdt_state = INACTIVE;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
for (i = 0; i < dump->ioa_dump.next_page_index; i++)
free_page((unsigned long) dump->ioa_dump.ioa_data[i]);
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
vfree(dump->ioa_dump.ioa_data);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
kfree(dump);
LEAVE;
}
scsi: ipr: System hung while dlpar adding primary ipr adapter back While dlpar adding primary ipr adapter back, driver goes through adapter initialization then schedule ipr_worker_thread to start te disk scan by dropping the host lock, calling scsi_add_device. Then get the adapter reset request again, so driver does scsi_block_requests, this will cause the scsi_add_device get hung until we unblock. But we can't run ipr_worker_thread to do the unblock because its stuck in scsi_add_device. This patch fixes the issue. [mkp: typo and whitespace fixes] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-09-20 18:32:12 -06:00
static void ipr_add_remove_thread(struct work_struct *work)
{
unsigned long lock_flags;
struct ipr_resource_entry *res;
struct scsi_device *sdev;
struct ipr_ioa_cfg *ioa_cfg =
container_of(work, struct ipr_ioa_cfg, scsi_add_work_q);
u8 bus, target, lun;
int did_work;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
restart:
do {
did_work = 0;
if (!ioa_cfg->hrrq[IPR_INIT_HRRQ].allow_cmds) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
if (res->del_from_ml && res->sdev) {
did_work = 1;
sdev = res->sdev;
if (!scsi_device_get(sdev)) {
if (!res->add_to_ml)
list_move_tail(&res->queue, &ioa_cfg->free_res_q);
else
res->del_from_ml = 0;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
scsi_remove_device(sdev);
scsi_device_put(sdev);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
}
break;
}
}
} while (did_work);
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
if (res->add_to_ml) {
bus = res->bus;
target = res->target;
lun = res->lun;
res->add_to_ml = 0;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
scsi_add_device(ioa_cfg->host, bus, target, lun);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
goto restart;
}
}
ioa_cfg->scan_done = 1;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
kobject_uevent(&ioa_cfg->host->shost_dev.kobj, KOBJ_CHANGE);
LEAVE;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_worker_thread - Worker thread
WorkStruct: make allyesconfig Fix up for make allyesconfig. Signed-Off-By: David Howells <dhowells@redhat.com>
2006-11-22 07:57:56 -07:00
* @work: ioa config struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Called at task level from a work thread. This function takes care
* of adding and removing device from the mid-layer as configuration
* changes are detected by the adapter.
*
* Return value:
* nothing
**/
WorkStruct: make allyesconfig Fix up for make allyesconfig. Signed-Off-By: David Howells <dhowells@redhat.com>
2006-11-22 07:57:56 -07:00
static void ipr_worker_thread(struct work_struct *work)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
unsigned long lock_flags;
struct ipr_dump *dump;
WorkStruct: make allyesconfig Fix up for make allyesconfig. Signed-Off-By: David Howells <dhowells@redhat.com>
2006-11-22 07:57:56 -07:00
struct ipr_ioa_cfg *ioa_cfg =
container_of(work, struct ipr_ioa_cfg, work_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Stop reading adapter dump prematurely When the ipr driver decides to dump the adapter, it changes the sdt_state to GET_DUMP, then prepares the adapter so that the dump can be read. However, if the ipr worker thread wakes up for some reason before the driver has put the adapter in a state where it can succesfully dump the adapter, the driver will start dumping the adapter too early, which can potentially trigger a BUG check in the pci config blocking API. Fix this by adding a new sdt_state to differentiate between the ipr driver wanting to dump the adapter in the near future and wanting to dump the adapter now. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-09-21 07:51:11 -06:00
if (ioa_cfg->sdt_state == READ_DUMP) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dump = ioa_cfg->dump;
if (!dump) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
kref_get(&dump->kref);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
ipr_get_ioa_dump(ioa_cfg, dump);
kref_put(&dump->kref, ipr_release_dump);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Fix BUG on adapter dump timeout If an adapter dump times out, the ipr driver will abort the dump and proceed to reset and recover the adapter. When an adapter dump completes, the work thread which is reading the adapter dump will initiate an adapter reset to recover the adapter. However, when the adapter dump gets aborted, the work thread should not initiate an adapter reset, since an adapter reset is already in progress. This fixes a case of calling pci_block_user_cfg_access overlapped, which results in a BUG. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-10-15 08:08:56 -06:00
if (ioa_cfg->sdt_state == DUMP_OBTAINED && !ioa_cfg->dump_timeout)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
scsi: ipr: Fix scsi-mq lockdep issue Fixes the following lockdep warning that can occur when scsi-mq is enabled with ipr due to ipr calling scsi_unblock_requests from irq context. The fix is to move the call to scsi_unblock_requests to ipr's existing workqueue. stack backtrace: CPU: 28 PID: 0 Comm: swapper/28 Not tainted 4.13.0-rc2-gcc6x-gf74c89b #1 Call Trace: [c000001fffe97550] [c000000000b50818] dump_stack+0xe8/0x160 (unreliable) [c000001fffe97590] [c0000000001586d0] print_usage_bug+0x2d0/0x390 [c000001fffe97640] [c000000000158f34] mark_lock+0x7a4/0x8e0 [c000001fffe976f0] [c00000000015a000] __lock_acquire+0x6a0/0x1a70 [c000001fffe97860] [c00000000015befc] lock_acquire+0xec/0x2e0 [c000001fffe97930] [c000000000b71514] _raw_spin_lock+0x44/0x70 [c000001fffe97960] [c0000000005b60f4] blk_mq_sched_dispatch_requests+0xa4/0x2a0 [c000001fffe979c0] [c0000000005acac0] __blk_mq_run_hw_queue+0x100/0x2c0 [c000001fffe97a00] [c0000000005ad478] __blk_mq_delay_run_hw_queue+0x118/0x130 [c000001fffe97a40] [c0000000005ad61c] blk_mq_start_hw_queues+0x6c/0xa0 [c000001fffe97a80] [c000000000797aac] scsi_kick_queue+0x2c/0x60 [c000001fffe97aa0] [c000000000797cf0] scsi_run_queue+0x210/0x360 [c000001fffe97b10] [c00000000079b888] scsi_run_host_queues+0x48/0x80 [c000001fffe97b40] [c0000000007b6090] ipr_ioa_bringdown_done+0x70/0x1e0 [c000001fffe97bc0] [c0000000007bc860] ipr_reset_ioa_job+0x80/0xf0 [c000001fffe97bf0] [c0000000007b4d50] ipr_reset_timer_done+0xd0/0x100 [c000001fffe97c30] [c0000000001937bc] call_timer_fn+0xdc/0x4b0 [c000001fffe97cf0] [c000000000193d08] expire_timers+0x178/0x330 [c000001fffe97d60] [c0000000001940c8] run_timer_softirq+0xb8/0x120 [c000001fffe97de0] [c000000000b726a8] __do_softirq+0x168/0x6d8 [c000001fffe97ef0] [c0000000000df2c8] irq_exit+0x108/0x150 [c000001fffe97f10] [c000000000017bf4] __do_irq+0x2a4/0x4a0 [c000001fffe97f90] [c00000000002da50] call_do_irq+0x14/0x24 [c0000007fad93aa0] [c000000000017e8c] do_IRQ+0x9c/0x140 [c0000007fad93af0] [c000000000008b98] hardware_interrupt_common+0x138/0x140 Reported-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-01 09:21:30 -06:00
if (ioa_cfg->scsi_unblock) {
ioa_cfg->scsi_unblock = 0;
ioa_cfg->scsi_blocked = 0;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
scsi_unblock_requests(ioa_cfg->host);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (ioa_cfg->scsi_blocked)
scsi_block_requests(ioa_cfg->host);
}
ipr: Wait to do async scan until scsi host is initialized When performing an async scan, make sure the kthread doing scanning doesn't start before the scsi host is fully initialized. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-07-15 13:48:03 -06:00
if (!ioa_cfg->scan_enabled) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
scsi: ipr: System hung while dlpar adding primary ipr adapter back While dlpar adding primary ipr adapter back, driver goes through adapter initialization then schedule ipr_worker_thread to start te disk scan by dropping the host lock, calling scsi_add_device. Then get the adapter reset request again, so driver does scsi_block_requests, this will cause the scsi_add_device get hung until we unblock. But we can't run ipr_worker_thread to do the unblock because its stuck in scsi_add_device. This patch fixes the issue. [mkp: typo and whitespace fixes] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-09-20 18:32:12 -06:00
schedule_work(&ioa_cfg->scsi_add_work_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
LEAVE;
}
#ifdef CONFIG_SCSI_IPR_TRACE
/**
* ipr_read_trace - Dump the adapter trace
sysfs: add struct file* to bin_attr callbacks This allows bin_attr->read,write,mmap callbacks to check file specific data (such as inode owner) as part of any privilege validation. Signed-off-by: Chris Wright <chrisw@sous-sol.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2010-05-12 19:28:57 -06:00
* @filp: open sysfs file
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @kobj: kobject struct
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 23:57:22 -06:00
* @bin_attr: bin_attribute struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @buf: buffer
* @off: offset
* @count: buffer size
*
* Return value:
* number of bytes printed to buffer
**/
sysfs: add struct file* to bin_attr callbacks This allows bin_attr->read,write,mmap callbacks to check file specific data (such as inode owner) as part of any privilege validation. Signed-off-by: Chris Wright <chrisw@sous-sol.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2010-05-12 19:28:57 -06:00
static ssize_t ipr_read_trace(struct file *filp, struct kobject *kobj,
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 23:57:22 -06:00
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct device *dev = container_of(kobj, struct device, kobj);
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags = 0;
[SCSI] ipr: use memory_read_from_buffer() Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-09-22 15:56:47 -06:00
ssize_t ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: use memory_read_from_buffer() Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-09-22 15:56:47 -06:00
ret = memory_read_from_buffer(buf, count, &off, ioa_cfg->trace,
IPR_TRACE_SIZE);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: use memory_read_from_buffer() Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-09-22 15:56:47 -06:00
return ret;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
static struct bin_attribute ipr_trace_attr = {
.attr = {
.name = "trace",
.mode = S_IRUGO,
},
.size = 0,
.read = ipr_read_trace,
};
#endif
/**
* ipr_show_fw_version - Show the firmware version
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @dev: class device struct
* @buf: buffer
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* number of bytes printed to buffer
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_show_fw_version(struct device *dev,
struct device_attribute *attr, char *buf)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
struct ipr_inquiry_page3 *ucode_vpd = &ioa_cfg->vpd_cbs->page3_data;
unsigned long lock_flags = 0;
int len;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
len = snprintf(buf, PAGE_SIZE, "%02X%02X%02X%02X\n",
ucode_vpd->major_release, ucode_vpd->card_type,
ucode_vpd->minor_release[0],
ucode_vpd->minor_release[1]);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static struct device_attribute ipr_fw_version_attr = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.attr = {
.name = "fw_version",
.mode = S_IRUGO,
},
.show = ipr_show_fw_version,
};
/**
* ipr_show_log_level - Show the adapter's error logging level
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @dev: class device struct
* @buf: buffer
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* number of bytes printed to buffer
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_show_log_level(struct device *dev,
struct device_attribute *attr, char *buf)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags = 0;
int len;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
len = snprintf(buf, PAGE_SIZE, "%d\n", ioa_cfg->log_level);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
/**
* ipr_store_log_level - Change the adapter's error logging level
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @dev: class device struct
* @buf: buffer
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* number of bytes printed to buffer
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_store_log_level(struct device *dev,
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
struct device_attribute *attr,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
const char *buf, size_t count)
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags = 0;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ioa_cfg->log_level = simple_strtoul(buf, NULL, 10);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return strlen(buf);
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static struct device_attribute ipr_log_level_attr = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.attr = {
.name = "log_level",
.mode = S_IRUGO | S_IWUSR,
},
.show = ipr_show_log_level,
.store = ipr_store_log_level
};
/**
* ipr_store_diagnostics - IOA Diagnostics interface
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @dev: device struct
* @buf: buffer
* @count: buffer size
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* This function will reset the adapter and wait a reasonable
* amount of time for any errors that the adapter might log.
*
* Return value:
* count on success / other on failure
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_store_diagnostics(struct device *dev,
struct device_attribute *attr,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
const char *buf, size_t count)
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags = 0;
int rc = count;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
while (ioa_cfg->in_reset_reload) {
[SCSI] ipr: Prevent overlapped adapter resets This patch fixes some scenarios where an ipr adapter could get reset overlapped, which could cause very long timeouts to occur, or PCI bus errors. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:06 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->errors_logged = 0;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NORMAL);
if (ioa_cfg->in_reset_reload) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
/* Wait for a second for any errors to be logged */
msleep(1000);
} else {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return -EIO;
}
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (ioa_cfg->in_reset_reload || ioa_cfg->errors_logged)
rc = -EIO;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return rc;
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static struct device_attribute ipr_diagnostics_attr = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.attr = {
.name = "run_diagnostics",
.mode = S_IWUSR,
},
.store = ipr_store_diagnostics
};
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
/**
* ipr_show_adapter_state - Show the adapter's state
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @class_dev: device struct
* @buf: buffer
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
*
* Return value:
* number of bytes printed to buffer
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_show_adapter_state(struct device *dev,
struct device_attribute *attr, char *buf)
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags = 0;
int len;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead)
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
len = snprintf(buf, PAGE_SIZE, "offline\n");
else
len = snprintf(buf, PAGE_SIZE, "online\n");
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
/**
* ipr_store_adapter_state - Change adapter state
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @dev: device struct
* @buf: buffer
* @count: buffer size
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
*
* This function will change the adapter's state.
*
* Return value:
* count on success / other on failure
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_store_adapter_state(struct device *dev,
struct device_attribute *attr,
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
const char *buf, size_t count)
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
int result = count, i;
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead &&
!strncmp(buf, "online", 6)) {
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].ioa_is_dead = 0;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
wmb();
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
ioa_cfg->reset_retries = 0;
ioa_cfg->in_ioa_bringdown = 0;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
return result;
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static struct device_attribute ipr_ioa_state_attr = {
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
.attr = {
[SCSI] ipr: Rename ipr's state scsi host attribute to prevent collisions Due to recent device model changes it now no longer tolerates name collisions. This causes a problem for ipr whose "state" attribute collides with an identically named one in the SCSI mid-layer. Rename the ipr driver attribute to be more specific. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-04-28 16:36:20 -06:00
.name = "online_state",
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
.mode = S_IRUGO | S_IWUSR,
},
.show = ipr_show_adapter_state,
.store = ipr_store_adapter_state
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_store_reset_adapter - Reset the adapter
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @dev: device struct
* @buf: buffer
* @count: buffer size
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* This function will reset the adapter.
*
* Return value:
* count on success / other on failure
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_store_reset_adapter(struct device *dev,
struct device_attribute *attr,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
const char *buf, size_t count)
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags;
int result = count;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (!ioa_cfg->in_reset_reload)
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NORMAL);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
return result;
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static struct device_attribute ipr_ioa_reset_attr = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.attr = {
.name = "reset_host",
.mode = S_IWUSR,
},
.store = ipr_store_reset_adapter
};
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
static int ipr_iopoll(struct irq_poll *iop, int budget);
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
/**
* ipr_show_iopoll_weight - Show ipr polling mode
* @dev: class device struct
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_show_iopoll_weight(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct Scsi_Host *shost = class_to_shost(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags = 0;
int len;
spin_lock_irqsave(shost->host_lock, lock_flags);
len = snprintf(buf, PAGE_SIZE, "%d\n", ioa_cfg->iopoll_weight);
spin_unlock_irqrestore(shost->host_lock, lock_flags);
return len;
}
/**
* ipr_store_iopoll_weight - Change the adapter's polling mode
* @dev: class device struct
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_store_iopoll_weight(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct Scsi_Host *shost = class_to_shost(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long user_iopoll_weight;
unsigned long lock_flags = 0;
int i;
if (!ioa_cfg->sis64) {
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
dev_info(&ioa_cfg->pdev->dev, "irq_poll not supported on this adapter\n");
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
return -EINVAL;
}
if (kstrtoul(buf, 10, &user_iopoll_weight))
return -EINVAL;
if (user_iopoll_weight > 256) {
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
dev_info(&ioa_cfg->pdev->dev, "Invalid irq_poll weight. It must be less than 256\n");
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
return -EINVAL;
}
if (user_iopoll_weight == ioa_cfg->iopoll_weight) {
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
dev_info(&ioa_cfg->pdev->dev, "Current irq_poll weight has the same weight\n");
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
return strlen(buf);
}
block: remove old blk_iopoll_enabled variable This was a debugging measure to toggle enabled/disabled when testing. But for real production setups, it's not safe to toggle this setting without either reloading drivers of quiescing IO first. Neither of which the toggle enforces. Additionally, it makes drivers deal with the conditional state. Remove it completely. It's up to the driver whether iopoll is enabled or not. Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-13 09:38:42 -06:00
if (ioa_cfg->iopoll_weight && ioa_cfg->sis64 && ioa_cfg->nvectors > 1) {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
for (i = 1; i < ioa_cfg->hrrq_num; i++)
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
irq_poll_disable(&ioa_cfg->hrrq[i].iopoll);
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
}
spin_lock_irqsave(shost->host_lock, lock_flags);
ioa_cfg->iopoll_weight = user_iopoll_weight;
block: remove old blk_iopoll_enabled variable This was a debugging measure to toggle enabled/disabled when testing. But for real production setups, it's not safe to toggle this setting without either reloading drivers of quiescing IO first. Neither of which the toggle enforces. Additionally, it makes drivers deal with the conditional state. Remove it completely. It's up to the driver whether iopoll is enabled or not. Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-13 09:38:42 -06:00
if (ioa_cfg->iopoll_weight && ioa_cfg->sis64 && ioa_cfg->nvectors > 1) {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
for (i = 1; i < ioa_cfg->hrrq_num; i++) {
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
irq_poll_init(&ioa_cfg->hrrq[i].iopoll,
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
ioa_cfg->iopoll_weight, ipr_iopoll);
}
}
spin_unlock_irqrestore(shost->host_lock, lock_flags);
return strlen(buf);
}
static struct device_attribute ipr_iopoll_weight_attr = {
.attr = {
.name = "iopoll_weight",
.mode = S_IRUGO | S_IWUSR,
},
.show = ipr_show_iopoll_weight,
.store = ipr_store_iopoll_weight
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_alloc_ucode_buffer - Allocates a microcode download buffer
* @buf_len: buffer length
*
* Allocates a DMA'able buffer in chunks and assembles a scatter/gather
* list to use for microcode download
*
* Return value:
* pointer to sglist / NULL on failure
**/
static struct ipr_sglist *ipr_alloc_ucode_buffer(int buf_len)
{
scsi: ipr: Use sgl_alloc_order() and sgl_free_order() Use the sgl_alloc_order() and sgl_free_order() functions instead of open coding these functions. Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-02-12 09:58:17 -07:00
int sg_size, order;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_sglist *sglist;
/* Get the minimum size per scatter/gather element */
sg_size = buf_len / (IPR_MAX_SGLIST - 1);
/* Get the actual size per element */
order = get_order(sg_size);
/* Allocate a scatter/gather list for the DMA */
scsi: ipr: Use sgl_alloc_order() and sgl_free_order() Use the sgl_alloc_order() and sgl_free_order() functions instead of open coding these functions. Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-02-12 09:58:17 -07:00
sglist = kzalloc(sizeof(struct ipr_sglist), GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (sglist == NULL) {
ipr_trace;
return NULL;
}
sglist->order = order;
scsi: ipr: Use sgl_alloc_order() and sgl_free_order() Use the sgl_alloc_order() and sgl_free_order() functions instead of open coding these functions. Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-02-12 09:58:17 -07:00
sglist->scatterlist = sgl_alloc_order(buf_len, order, false, GFP_KERNEL,
&sglist->num_sg);
if (!sglist->scatterlist) {
kfree(sglist);
return NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
return sglist;
}
/**
* ipr_free_ucode_buffer - Frees a microcode download buffer
* @p_dnld: scatter/gather list pointer
*
* Free a DMA'able ucode download buffer previously allocated with
* ipr_alloc_ucode_buffer
*
* Return value:
* nothing
**/
static void ipr_free_ucode_buffer(struct ipr_sglist *sglist)
{
scsi: ipr: Use sgl_alloc_order() and sgl_free_order() Use the sgl_alloc_order() and sgl_free_order() functions instead of open coding these functions. Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Reviewed-by: Hannes Reinecke <hare@suse.com> Cc: Martin K. Petersen <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-02-12 09:58:17 -07:00
sgl_free_order(sglist->scatterlist, sglist->order);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
kfree(sglist);
}
/**
* ipr_copy_ucode_buffer - Copy user buffer to kernel buffer
* @sglist: scatter/gather list pointer
* @buffer: buffer pointer
* @len: buffer length
*
* Copy a microcode image from a user buffer into a buffer allocated by
* ipr_alloc_ucode_buffer
*
* Return value:
* 0 on success / other on failure
**/
static int ipr_copy_ucode_buffer(struct ipr_sglist *sglist,
u8 *buffer, u32 len)
{
int bsize_elem, i, result = 0;
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
struct scatterlist *sg;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
void *kaddr;
/* Determine the actual number of bytes per element */
bsize_elem = PAGE_SIZE * (1 << sglist->order);
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
sg = sglist->scatterlist;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
for (i = 0; i < (len / bsize_elem); i++, sg = sg_next(sg),
buffer += bsize_elem) {
struct page *page = sg_page(sg);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-22 13:19:53 -06:00
kaddr = kmap(page);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
memcpy(kaddr, buffer, bsize_elem);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-22 13:19:53 -06:00
kunmap(page);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
sg->length = bsize_elem;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (result != 0) {
ipr_trace;
return result;
}
}
if (len % bsize_elem) {
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
struct page *page = sg_page(sg);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-22 13:19:53 -06:00
kaddr = kmap(page);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
memcpy(kaddr, buffer, len % bsize_elem);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-22 13:19:53 -06:00
kunmap(page);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
sg->length = len % bsize_elem;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
sglist->buffer_len = len;
return result;
}
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
/**
* ipr_build_ucode_ioadl64 - Build a microcode download IOADL
* @ipr_cmd: ipr command struct
* @sglist: scatter/gather list
*
* Builds a microcode download IOA data list (IOADL).
*
**/
static void ipr_build_ucode_ioadl64(struct ipr_cmnd *ipr_cmd,
struct ipr_sglist *sglist)
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
struct ipr_ioadl64_desc *ioadl64 = ipr_cmd->i.ioadl64;
struct scatterlist *scatterlist = sglist->scatterlist;
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
struct scatterlist *sg;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
int i;
ipr_cmd->dma_use_sg = sglist->num_dma_sg;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
ioarcb->data_transfer_length = cpu_to_be32(sglist->buffer_len);
ioarcb->ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl64_desc) * ipr_cmd->dma_use_sg);
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
for_each_sg(scatterlist, sg, ipr_cmd->dma_use_sg, i) {
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioadl64[i].flags = cpu_to_be32(IPR_IOADL_FLAGS_WRITE);
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
ioadl64[i].data_len = cpu_to_be32(sg_dma_len(sg));
ioadl64[i].address = cpu_to_be64(sg_dma_address(sg));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
}
ioadl64[i-1].flags |= cpu_to_be32(IPR_IOADL_FLAGS_LAST);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
* ipr_build_ucode_ioadl - Build a microcode download IOADL
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @ipr_cmd: ipr command struct
* @sglist: scatter/gather list
*
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
* Builds a microcode download IOA data list (IOADL).
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
**/
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
static void ipr_build_ucode_ioadl(struct ipr_cmnd *ipr_cmd,
struct ipr_sglist *sglist)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
struct ipr_ioadl_desc *ioadl = ipr_cmd->i.ioadl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct scatterlist *scatterlist = sglist->scatterlist;
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
struct scatterlist *sg;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
int i;
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
ipr_cmd->dma_use_sg = sglist->num_dma_sg;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->data_transfer_length = cpu_to_be32(sglist->buffer_len);
ioarcb->ioadl_len =
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
cpu_to_be32(sizeof(struct ipr_ioadl_desc) * ipr_cmd->dma_use_sg);
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
for_each_sg(scatterlist, sg, ipr_cmd->dma_use_sg, i) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioadl[i].flags_and_data_len =
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
cpu_to_be32(IPR_IOADL_FLAGS_WRITE | sg_dma_len(sg));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioadl[i].address =
scsi: ipr: use sg helper to iterate over scatterlist Unlike the legacy I/O path, scsi-mq preallocates a large array to hold the scatterlist for each request. This static allocation can consume substantial amounts of memory on modern controllers which support a large number of concurrently outstanding requests. To facilitate a switch to a smaller static allocation combined with a dynamic allocation for requests that need it, we need to make sure all SCSI drivers handle chained scatterlists correctly. Convert remaining drivers that directly dereference the scatterlist array to using the iterator functions. [mkp: clarified commit message] Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-06-17 19:37:46 -06:00
cpu_to_be32(sg_dma_address(sg));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
ioadl[i-1].flags_and_data_len |=
cpu_to_be32(IPR_IOADL_FLAGS_LAST);
}
/**
* ipr_update_ioa_ucode - Update IOA's microcode
* @ioa_cfg: ioa config struct
* @sglist: scatter/gather list
*
* Initiate an adapter reset to update the IOA's microcode
*
* Return value:
* 0 on success / -EIO on failure
**/
static int ipr_update_ioa_ucode(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_sglist *sglist)
{
unsigned long lock_flags;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
while (ioa_cfg->in_reset_reload) {
[SCSI] ipr: Prevent overlapped adapter resets This patch fixes some scenarios where an ipr adapter could get reset overlapped, which could cause very long timeouts to occur, or PCI bus errors. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:06 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
}
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
if (ioa_cfg->ucode_sglist) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
dev_err(&ioa_cfg->pdev->dev,
"Microcode download already in progress\n");
return -EIO;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
sglist->num_dma_sg = dma_map_sg(&ioa_cfg->pdev->dev,
sglist->scatterlist, sglist->num_sg,
DMA_TO_DEVICE);
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
if (!sglist->num_dma_sg) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
dev_err(&ioa_cfg->pdev->dev,
"Failed to map microcode download buffer!\n");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return -EIO;
}
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
ioa_cfg->ucode_sglist = sglist;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NORMAL);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ioa_cfg->ucode_sglist = NULL;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return 0;
}
/**
* ipr_store_update_fw - Update the firmware on the adapter
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
* @class_dev: device struct
* @buf: buffer
* @count: buffer size
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* This function will update the firmware on the adapter.
*
* Return value:
* count on success / other on failure
**/
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static ssize_t ipr_store_update_fw(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
struct ipr_ucode_image_header *image_hdr;
const struct firmware *fw_entry;
struct ipr_sglist *sglist;
char fname[100];
char *src;
ipr: Fix regression when loading firmware Commit d63c7dd5bcb9 ("ipr: Fix out-of-bounds null overwrite") removed the end of line handling when storing the update_fw sysfs attribute. This changed the userpace API because it started refusing writes terminated by a line feed, which broke the update tools we already have. This patch re-adds that handling, so both a write terminated by a line feed or not can make it through with the update. Fixes: d63c7dd5bcb9 ("ipr: Fix out-of-bounds null overwrite") Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Cc: Insu Yun <wuninsu@gmail.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-02-25 09:54:20 -07:00
char *endline;
ipr: Fix out-of-bounds null overwrite Return value of snprintf is not bound by size value, 2nd argument. (https://www.kernel.org/doc/htmldocs/kernel-api/API-snprintf.html). Return value is number of printed chars, can be larger than 2nd argument. Therefore, it can write null byte out of bounds ofbuffer. Since snprintf puts null, it does not need to put additional null byte. Signed-off-by: Insu Yun <wuninsu@gmail.com> Reviewed-by: Shane Seymour <shane.seymour@hpe.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-01-06 10:44:01 -07:00
int result, dnld_size;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
ipr: Fix out-of-bounds null overwrite Return value of snprintf is not bound by size value, 2nd argument. (https://www.kernel.org/doc/htmldocs/kernel-api/API-snprintf.html). Return value is number of printed chars, can be larger than 2nd argument. Therefore, it can write null byte out of bounds ofbuffer. Since snprintf puts null, it does not need to put additional null byte. Signed-off-by: Insu Yun <wuninsu@gmail.com> Reviewed-by: Shane Seymour <shane.seymour@hpe.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-01-06 10:44:01 -07:00
snprintf(fname, sizeof(fname), "%s", buf);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: Fix regression when loading firmware Commit d63c7dd5bcb9 ("ipr: Fix out-of-bounds null overwrite") removed the end of line handling when storing the update_fw sysfs attribute. This changed the userpace API because it started refusing writes terminated by a line feed, which broke the update tools we already have. This patch re-adds that handling, so both a write terminated by a line feed or not can make it through with the update. Fixes: d63c7dd5bcb9 ("ipr: Fix out-of-bounds null overwrite") Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Cc: Insu Yun <wuninsu@gmail.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-02-25 09:54:20 -07:00
endline = strchr(fname, '\n');
if (endline)
*endline = '\0';
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
if (request_firmware(&fw_entry, fname, &ioa_cfg->pdev->dev)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_err(&ioa_cfg->pdev->dev, "Firmware file %s not found\n", fname);
return -EIO;
}
image_hdr = (struct ipr_ucode_image_header *)fw_entry->data;
src = (u8 *)image_hdr + be32_to_cpu(image_hdr->header_length);
dnld_size = fw_entry->size - be32_to_cpu(image_hdr->header_length);
sglist = ipr_alloc_ucode_buffer(dnld_size);
if (!sglist) {
dev_err(&ioa_cfg->pdev->dev, "Microcode buffer allocation failed\n");
release_firmware(fw_entry);
return -ENOMEM;
}
result = ipr_copy_ucode_buffer(sglist, src, dnld_size);
if (result) {
dev_err(&ioa_cfg->pdev->dev,
"Microcode buffer copy to DMA buffer failed\n");
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Add support to flash FPGA and flash back DRAM images The write buffer command is used to download and burn new IOA FW images. The same interface can now be used to flash FPGA and flash back DRAM images. To download and flash the new images takes more than 15 minutes, so increase the write buffer command timeout to 30 minutes. The FPGA and flash back DRAM images don't have the same card_type as the IOA FW image. So, remove the sanity checking from the driver. The adapter has sanity checking and will only accept a valid image. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-10-03 21:54:37 -06:00
ipr_info("Updating microcode, please be patient. This may take up to 30 minutes.\n");
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
result = ipr_update_ioa_ucode(ioa_cfg, sglist);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
if (!result)
result = count;
out:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_free_ucode_buffer(sglist);
release_firmware(fw_entry);
[SCSI] ipr: Fix adapter microcode update DMA mapping leak If the write buffer command that is issued to the ipr adapter to update its microcode fails for some reason, the DMA buffer will never get unmapped. Move the pci_map/unmap out of the IOA reset job so that the buffer is always clearly mapped and unmapped. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:27 -07:00
return result;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static struct device_attribute ipr_update_fw_attr = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.attr = {
.name = "update_fw",
.mode = S_IWUSR,
},
.store = ipr_store_update_fw
};
[SCSI] ipr: fix resource type update and add sdev and shost attributes Setting the resource type in the ipr_update_res_entry function was incorrect in that the top 4 bits were masked off. The assignment has been updated to no longer mask those bits. Then, two new attributes were added to allow the user space utilities to more easily get information. The resource_type sdev attribute is set for all devices in the adapter's configuration table and indicates the type of device. The fw_type shost attribute indicates the firmware type supported by the adapter. Finally, the resource_path attribute was changed to be mode S_IRUGO. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:50:14 -06:00
/**
* ipr_show_fw_type - Show the adapter's firmware type.
* @dev: class device struct
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_show_fw_type(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct Scsi_Host *shost = class_to_shost(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
unsigned long lock_flags = 0;
int len;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
len = snprintf(buf, PAGE_SIZE, "%d\n", ioa_cfg->sis64);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
static struct device_attribute ipr_ioa_fw_type_attr = {
.attr = {
.name = "fw_type",
.mode = S_IRUGO,
},
.show = ipr_show_fw_type
};
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
static ssize_t ipr_read_async_err_log(struct file *filep, struct kobject *kobj,
struct bin_attribute *bin_attr, char *buf,
loff_t off, size_t count)
{
struct device *cdev = container_of(kobj, struct device, kobj);
struct Scsi_Host *shost = class_to_shost(cdev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
struct ipr_hostrcb *hostrcb;
unsigned long lock_flags = 0;
int ret;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
hostrcb = list_first_entry_or_null(&ioa_cfg->hostrcb_report_q,
struct ipr_hostrcb, queue);
if (!hostrcb) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return 0;
}
ret = memory_read_from_buffer(buf, count, &off, &hostrcb->hcam,
sizeof(hostrcb->hcam));
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return ret;
}
static ssize_t ipr_next_async_err_log(struct file *filep, struct kobject *kobj,
struct bin_attribute *bin_attr, char *buf,
loff_t off, size_t count)
{
struct device *cdev = container_of(kobj, struct device, kobj);
struct Scsi_Host *shost = class_to_shost(cdev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
struct ipr_hostrcb *hostrcb;
unsigned long lock_flags = 0;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
hostrcb = list_first_entry_or_null(&ioa_cfg->hostrcb_report_q,
struct ipr_hostrcb, queue);
if (!hostrcb) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return count;
}
/* Reclaim hostrcb before exit */
list_move_tail(&hostrcb->queue, &ioa_cfg->hostrcb_free_q);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return count;
}
static struct bin_attribute ipr_ioa_async_err_log = {
.attr = {
.name = "async_err_log",
.mode = S_IRUGO | S_IWUSR,
},
.size = 0,
.read = ipr_read_async_err_log,
.write = ipr_next_async_err_log
};
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
static struct device_attribute *ipr_ioa_attrs[] = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
&ipr_fw_version_attr,
&ipr_log_level_attr,
&ipr_diagnostics_attr,
[SCSI] ipr: Provide reset_adapter retry method for offlined adapters If an ipr adapter repeatedly fails its initialization the ipr driver will take the adapter offline and never talk to it again. This provides a method for the user to manually try the initialization again through sysfs. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:40 -07:00
&ipr_ioa_state_attr,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
&ipr_ioa_reset_attr,
&ipr_update_fw_attr,
[SCSI] ipr: fix resource type update and add sdev and shost attributes Setting the resource type in the ipr_update_res_entry function was incorrect in that the top 4 bits were masked off. The assignment has been updated to no longer mask those bits. Then, two new attributes were added to allow the user space utilities to more easily get information. The resource_type sdev attribute is set for all devices in the adapter's configuration table and indicates the type of device. The fw_type shost attribute indicates the firmware type supported by the adapter. Finally, the resource_path attribute was changed to be mode S_IRUGO. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:50:14 -06:00
&ipr_ioa_fw_type_attr,
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
&ipr_iopoll_weight_attr,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
NULL,
};
#ifdef CONFIG_SCSI_IPR_DUMP
/**
* ipr_read_dump - Dump the adapter
sysfs: add struct file* to bin_attr callbacks This allows bin_attr->read,write,mmap callbacks to check file specific data (such as inode owner) as part of any privilege validation. Signed-off-by: Chris Wright <chrisw@sous-sol.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2010-05-12 19:28:57 -06:00
* @filp: open sysfs file
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @kobj: kobject struct
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 23:57:22 -06:00
* @bin_attr: bin_attribute struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @buf: buffer
* @off: offset
* @count: buffer size
*
* Return value:
* number of bytes printed to buffer
**/
sysfs: add struct file* to bin_attr callbacks This allows bin_attr->read,write,mmap callbacks to check file specific data (such as inode owner) as part of any privilege validation. Signed-off-by: Chris Wright <chrisw@sous-sol.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2010-05-12 19:28:57 -06:00
static ssize_t ipr_read_dump(struct file *filp, struct kobject *kobj,
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 23:57:22 -06:00
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct device *cdev = container_of(kobj, struct device, kobj);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct Scsi_Host *shost = class_to_shost(cdev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
struct ipr_dump *dump;
unsigned long lock_flags = 0;
char *src;
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
int len, sdt_end;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
size_t rc = count;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
dump = ioa_cfg->dump;
if (ioa_cfg->sdt_state != DUMP_OBTAINED || !dump) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return 0;
}
kref_get(&dump->kref);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
if (off > dump->driver_dump.hdr.len) {
kref_put(&dump->kref, ipr_release_dump);
return 0;
}
if (off + count > dump->driver_dump.hdr.len) {
count = dump->driver_dump.hdr.len - off;
rc = count;
}
if (count && off < sizeof(dump->driver_dump)) {
if (off + count > sizeof(dump->driver_dump))
len = sizeof(dump->driver_dump) - off;
else
len = count;
src = (u8 *)&dump->driver_dump + off;
memcpy(buf, src, len);
buf += len;
off += len;
count -= len;
}
off -= sizeof(dump->driver_dump);
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (ioa_cfg->sis64)
sdt_end = offsetof(struct ipr_ioa_dump, sdt.entry) +
(be32_to_cpu(dump->ioa_dump.sdt.hdr.num_entries_used) *
sizeof(struct ipr_sdt_entry));
else
sdt_end = offsetof(struct ipr_ioa_dump, sdt.entry) +
(IPR_FMT2_NUM_SDT_ENTRIES * sizeof(struct ipr_sdt_entry));
if (count && off < sdt_end) {
if (off + count > sdt_end)
len = sdt_end - off;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
else
len = count;
src = (u8 *)&dump->ioa_dump + off;
memcpy(buf, src, len);
buf += len;
off += len;
count -= len;
}
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
off -= sdt_end;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
while (count) {
if ((off & PAGE_MASK) != ((off + count) & PAGE_MASK))
len = PAGE_ALIGN(off) - off;
else
len = count;
src = (u8 *)dump->ioa_dump.ioa_data[(off & PAGE_MASK) >> PAGE_SHIFT];
src += off & ~PAGE_MASK;
memcpy(buf, src, len);
buf += len;
off += len;
count -= len;
}
kref_put(&dump->kref, ipr_release_dump);
return rc;
}
/**
* ipr_alloc_dump - Prepare for adapter dump
* @ioa_cfg: ioa config struct
*
* Return value:
* 0 on success / other on failure
**/
static int ipr_alloc_dump(struct ipr_ioa_cfg *ioa_cfg)
{
struct ipr_dump *dump;
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
__be32 **ioa_data;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
unsigned long lock_flags = 0;
[SCSI] ipr: Convert to use kzalloc Convert appropriate kmalloc/memset calls to use kzalloc. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:20 -07:00
dump = kzalloc(sizeof(struct ipr_dump), GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!dump) {
ipr_err("Dump memory allocation failed\n");
return -ENOMEM;
}
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (ioa_cfg->sis64)
treewide: Use array_size() in vmalloc() The vmalloc() function has no 2-factor argument form, so multiplication factors need to be wrapped in array_size(). This patch replaces cases of: vmalloc(a * b) with: vmalloc(array_size(a, b)) as well as handling cases of: vmalloc(a * b * c) with: vmalloc(array3_size(a, b, c)) This does, however, attempt to ignore constant size factors like: vmalloc(4 * 1024) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( vmalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | vmalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( vmalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(char) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(u8) * COUNT + COUNT , ...) | vmalloc( - sizeof(__u8) * COUNT + COUNT , ...) | vmalloc( - sizeof(char) * COUNT + COUNT , ...) | vmalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( vmalloc( - sizeof(TYPE) * (COUNT_ID) + array_size(COUNT_ID, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT_ID + array_size(COUNT_ID, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * (COUNT_CONST) + array_size(COUNT_CONST, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT_CONST + array_size(COUNT_CONST, sizeof(TYPE)) , ...) | vmalloc( - sizeof(THING) * (COUNT_ID) + array_size(COUNT_ID, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT_ID + array_size(COUNT_ID, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * (COUNT_CONST) + array_size(COUNT_CONST, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT_CONST + array_size(COUNT_CONST, sizeof(THING)) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ vmalloc( - SIZE * COUNT + array_size(COUNT, SIZE) , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( vmalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( vmalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | vmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | vmalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | vmalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | vmalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | vmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( vmalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( vmalloc(C1 * C2 * C3, ...) | vmalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants. @@ expression E1, E2; constant C1, C2; @@ ( vmalloc(C1 * C2, ...) | vmalloc( - E1 * E2 + array_size(E1, E2) , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 15:27:11 -06:00
ioa_data = vmalloc(array_size(IPR_FMT3_MAX_NUM_DUMP_PAGES,
sizeof(__be32 *)));
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
else
treewide: Use array_size() in vmalloc() The vmalloc() function has no 2-factor argument form, so multiplication factors need to be wrapped in array_size(). This patch replaces cases of: vmalloc(a * b) with: vmalloc(array_size(a, b)) as well as handling cases of: vmalloc(a * b * c) with: vmalloc(array3_size(a, b, c)) This does, however, attempt to ignore constant size factors like: vmalloc(4 * 1024) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( vmalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | vmalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( vmalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(char) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | vmalloc( - sizeof(u8) * COUNT + COUNT , ...) | vmalloc( - sizeof(__u8) * COUNT + COUNT , ...) | vmalloc( - sizeof(char) * COUNT + COUNT , ...) | vmalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( vmalloc( - sizeof(TYPE) * (COUNT_ID) + array_size(COUNT_ID, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT_ID + array_size(COUNT_ID, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * (COUNT_CONST) + array_size(COUNT_CONST, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT_CONST + array_size(COUNT_CONST, sizeof(TYPE)) , ...) | vmalloc( - sizeof(THING) * (COUNT_ID) + array_size(COUNT_ID, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT_ID + array_size(COUNT_ID, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * (COUNT_CONST) + array_size(COUNT_CONST, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT_CONST + array_size(COUNT_CONST, sizeof(THING)) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ vmalloc( - SIZE * COUNT + array_size(COUNT, SIZE) , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( vmalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | vmalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | vmalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( vmalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | vmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | vmalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | vmalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | vmalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | vmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( vmalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | vmalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( vmalloc(C1 * C2 * C3, ...) | vmalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants. @@ expression E1, E2; constant C1, C2; @@ ( vmalloc(C1 * C2, ...) | vmalloc( - E1 * E2 + array_size(E1, E2) , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 15:27:11 -06:00
ioa_data = vmalloc(array_size(IPR_FMT2_MAX_NUM_DUMP_PAGES,
sizeof(__be32 *)));
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
if (!ioa_data) {
ipr_err("Dump memory allocation failed\n");
kfree(dump);
return -ENOMEM;
}
dump->ioa_dump.ioa_data = ioa_data;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
kref_init(&dump->kref);
dump->ioa_cfg = ioa_cfg;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (INACTIVE != ioa_cfg->sdt_state) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: increase the dump size for 64 bit adapters Currently the size of the dump generated by the driver is limited in 4MB, which is insufficient to gather much useful data from the new 64 bit adapters. This patch makes the needed changes to increase the dump limit for the 64 bit adapters to 32MB, or even to a bigger value in the future, but keeping the current limitations for the legacy 32 bit adapters. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-26 16:23:29 -06:00
vfree(dump->ioa_dump.ioa_data);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
kfree(dump);
return 0;
}
ioa_cfg->dump = dump;
ioa_cfg->sdt_state = WAIT_FOR_DUMP;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead && !ioa_cfg->dump_taken) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->dump_taken = 1;
schedule_work(&ioa_cfg->work_q);
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return 0;
}
/**
* ipr_free_dump - Free adapter dump memory
* @ioa_cfg: ioa config struct
*
* Return value:
* 0 on success / other on failure
**/
static int ipr_free_dump(struct ipr_ioa_cfg *ioa_cfg)
{
struct ipr_dump *dump;
unsigned long lock_flags = 0;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
dump = ioa_cfg->dump;
if (!dump) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return 0;
}
ioa_cfg->dump = NULL;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
kref_put(&dump->kref, ipr_release_dump);
LEAVE;
return 0;
}
/**
* ipr_write_dump - Setup dump state of adapter
sysfs: add struct file* to bin_attr callbacks This allows bin_attr->read,write,mmap callbacks to check file specific data (such as inode owner) as part of any privilege validation. Signed-off-by: Chris Wright <chrisw@sous-sol.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2010-05-12 19:28:57 -06:00
* @filp: open sysfs file
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @kobj: kobject struct
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 23:57:22 -06:00
* @bin_attr: bin_attribute struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @buf: buffer
* @off: offset
* @count: buffer size
*
* Return value:
* number of bytes printed to buffer
**/
sysfs: add struct file* to bin_attr callbacks This allows bin_attr->read,write,mmap callbacks to check file specific data (such as inode owner) as part of any privilege validation. Signed-off-by: Chris Wright <chrisw@sous-sol.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2010-05-12 19:28:57 -06:00
static ssize_t ipr_write_dump(struct file *filp, struct kobject *kobj,
sysfs: add parameter "struct bin_attribute *" in .read/.write methods for sysfs binary attributes Well, first of all, I don't want to change so many files either. What I do: Adding a new parameter "struct bin_attribute *" in the .read/.write methods for the sysfs binary attributes. In fact, only the four lines change in fs/sysfs/bin.c and include/linux/sysfs.h do the real work. But I have to update all the files that use binary attributes to make them compatible with the new .read and .write methods. I'm not sure if I missed any. :( Why I do this: For a sysfs attribute, we can get a pointer pointing to the struct attribute in the .show/.store method, while we can't do this for the binary attributes. I don't know why this is different, but this does make it not so handy to use the binary attributes as the regular ones. So I think this patch is reasonable. :) Who benefits from it: The patch that exposes ACPI tables in sysfs requires such an improvement. All the table binary attributes share the same .read method. Parameter "struct bin_attribute *" is used to get the table signature and instance number which are used to distinguish different ACPI table binary attributes. Without this parameter, we need to offer different .read methods for different ACPI table binary attributes. This is impossible as there are various ACPI tables on different platforms, and we don't know what they are until they are loaded. Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 23:57:22 -06:00
struct bin_attribute *bin_attr,
char *buf, loff_t off, size_t count)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
struct device *cdev = container_of(kobj, struct device, kobj);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct Scsi_Host *shost = class_to_shost(cdev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
int rc;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if (buf[0] == '1')
rc = ipr_alloc_dump(ioa_cfg);
else if (buf[0] == '0')
rc = ipr_free_dump(ioa_cfg);
else
return -EINVAL;
if (rc)
return rc;
else
return count;
}
static struct bin_attribute ipr_dump_attr = {
.attr = {
.name = "dump",
.mode = S_IRUSR | S_IWUSR,
},
.size = 0,
.read = ipr_read_dump,
.write = ipr_write_dump
};
#else
static int ipr_free_dump(struct ipr_ioa_cfg *ioa_cfg) { return 0; };
#endif
/**
* ipr_change_queue_depth - Change the device's queue depth
* @sdev: scsi device struct
* @qdepth: depth to set
[SCSI] modify change_queue_depth to take in reason why it is being called This patch modifies scsi_host_template->change_queue_depth so that it takes an argument indicating why it is being called. This will be used so that if a LLD needs to do some extra processing when handling queue fulls or later ramp ups, it can do so. This is a simple port of the drivers setting a change_queue_depth callback. In the patch I just have these LLDs adjust the queue depth if the user was requesting it. Signed-off-by: Mike Christie <michaelc@cs.wisc.edu> [Vasu.Dev: v2 Also converted pmcraid_change_queue_depth and then verified all modules compile using "make allmodconfig" for any new build warnings on X86_64. Updated original description after combing two original patches from Mike to make this patch git bisectable.] Signed-off-by: Vasu Dev <vasu.dev@intel.com> [jejb: fixed up 53c700] Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-15 18:46:39 -06:00
* @reason: calling context
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* actual depth set
**/
scsi: drop reason argument from ->change_queue_depth Drop the now unused reason argument from the ->change_queue_depth method. Also add a return value to scsi_adjust_queue_depth, and rename it to scsi_change_queue_depth now that it can be used as the default ->change_queue_depth implementation. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Reviewed-by: Hannes Reinecke <hare@suse.de>
2014-11-13 07:08:42 -07:00
static int ipr_change_queue_depth(struct scsi_device *sdev, int qdepth)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *)sdev->hostdata;
if (res && ipr_is_gata(res) && qdepth > IPR_MAX_CMD_PER_ATA_LUN)
qdepth = IPR_MAX_CMD_PER_ATA_LUN;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
scsi: drop reason argument from ->change_queue_depth Drop the now unused reason argument from the ->change_queue_depth method. Also add a return value to scsi_adjust_queue_depth, and rename it to scsi_change_queue_depth now that it can be used as the default ->change_queue_depth implementation. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Reviewed-by: Hannes Reinecke <hare@suse.de>
2014-11-13 07:08:42 -07:00
scsi_change_queue_depth(sdev, qdepth);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return sdev->queue_depth;
}
/**
* ipr_show_adapter_handle - Show the adapter's resource handle for this device
* @dev: device struct
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
* @attr: device attribute structure
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
[PATCH] Driver Core: drivers/s390/net/qeth_sys.c - drivers/usb/gadget/pxa2xx_udc.c: update device attribute callbacks Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-05-17 04:43:37 -06:00
static ssize_t ipr_show_adapter_handle(struct device *dev, struct device_attribute *attr, char *buf)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct scsi_device *sdev = to_scsi_device(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
ssize_t len = -ENXIO;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *)sdev->hostdata;
if (res)
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
len = snprintf(buf, PAGE_SIZE, "%08X\n", res->res_handle);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
static struct device_attribute ipr_adapter_handle_attr = {
.attr = {
.name = "adapter_handle",
.mode = S_IRUSR,
},
.show = ipr_show_adapter_handle
};
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
/**
[SCSI] ipr: fix resource path display and formatting It was possible to overflow the buffer used to print out the formatted version of the resource path. The fix is to limit the number of bytes that get formatted. This patch also updates the ipr_show_resource_path function to display the resource address for devices that are attached to adapters that don't support resource paths. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-03 17:02:21 -06:00
* ipr_show_resource_path - Show the resource path or the resource address for
* this device.
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
* @dev: device struct
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
* @attr: device attribute structure
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_show_resource_path(struct device *dev, struct device_attribute *attr, char *buf)
{
struct scsi_device *sdev = to_scsi_device(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
ssize_t len = -ENXIO;
char buffer[IPR_MAX_RES_PATH_LENGTH];
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *)sdev->hostdata;
[SCSI] ipr: fix resource path display and formatting It was possible to overflow the buffer used to print out the formatted version of the resource path. The fix is to limit the number of bytes that get formatted. This patch also updates the ipr_show_resource_path function to display the resource address for devices that are attached to adapters that don't support resource paths. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-03 17:02:21 -06:00
if (res && ioa_cfg->sis64)
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
len = snprintf(buf, PAGE_SIZE, "%s\n",
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
__ipr_format_res_path(res->res_path, buffer,
sizeof(buffer)));
[SCSI] ipr: fix resource path display and formatting It was possible to overflow the buffer used to print out the formatted version of the resource path. The fix is to limit the number of bytes that get formatted. This patch also updates the ipr_show_resource_path function to display the resource address for devices that are attached to adapters that don't support resource paths. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-03 17:02:21 -06:00
else if (res)
len = snprintf(buf, PAGE_SIZE, "%d:%d:%d:%d\n", ioa_cfg->host->host_no,
res->bus, res->target, res->lun);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
static struct device_attribute ipr_resource_path_attr = {
.attr = {
.name = "resource_path",
[SCSI] ipr: fix resource type update and add sdev and shost attributes Setting the resource type in the ipr_update_res_entry function was incorrect in that the top 4 bits were masked off. The assignment has been updated to no longer mask those bits. Then, two new attributes were added to allow the user space utilities to more easily get information. The resource_type sdev attribute is set for all devices in the adapter's configuration table and indicates the type of device. The fw_type shost attribute indicates the firmware type supported by the adapter. Finally, the resource_path attribute was changed to be mode S_IRUGO. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:50:14 -06:00
.mode = S_IRUGO,
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
},
.show = ipr_show_resource_path
};
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
/**
* ipr_show_device_id - Show the device_id for this device.
* @dev: device struct
* @attr: device attribute structure
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_show_device_id(struct device *dev, struct device_attribute *attr, char *buf)
{
struct scsi_device *sdev = to_scsi_device(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
ssize_t len = -ENXIO;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *)sdev->hostdata;
if (res && ioa_cfg->sis64)
ipr: Byte swapping for device_id attribute in sysfs On LE system, users see the wrong device_id attribute. This patch does necessary byte swapping for device_id attribute and works on both of LE and BE systems. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-06-11 19:45:18 -06:00
len = snprintf(buf, PAGE_SIZE, "0x%llx\n", be64_to_cpu(res->dev_id));
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
else if (res)
len = snprintf(buf, PAGE_SIZE, "0x%llx\n", res->lun_wwn);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
static struct device_attribute ipr_device_id_attr = {
.attr = {
.name = "device_id",
.mode = S_IRUGO,
},
.show = ipr_show_device_id
};
[SCSI] ipr: fix resource type update and add sdev and shost attributes Setting the resource type in the ipr_update_res_entry function was incorrect in that the top 4 bits were masked off. The assignment has been updated to no longer mask those bits. Then, two new attributes were added to allow the user space utilities to more easily get information. The resource_type sdev attribute is set for all devices in the adapter's configuration table and indicates the type of device. The fw_type shost attribute indicates the firmware type supported by the adapter. Finally, the resource_path attribute was changed to be mode S_IRUGO. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:50:14 -06:00
/**
* ipr_show_resource_type - Show the resource type for this device.
* @dev: device struct
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
* @attr: device attribute structure
[SCSI] ipr: fix resource type update and add sdev and shost attributes Setting the resource type in the ipr_update_res_entry function was incorrect in that the top 4 bits were masked off. The assignment has been updated to no longer mask those bits. Then, two new attributes were added to allow the user space utilities to more easily get information. The resource_type sdev attribute is set for all devices in the adapter's configuration table and indicates the type of device. The fw_type shost attribute indicates the firmware type supported by the adapter. Finally, the resource_path attribute was changed to be mode S_IRUGO. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:50:14 -06:00
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_show_resource_type(struct device *dev, struct device_attribute *attr, char *buf)
{
struct scsi_device *sdev = to_scsi_device(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
ssize_t len = -ENXIO;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *)sdev->hostdata;
if (res)
len = snprintf(buf, PAGE_SIZE, "%x\n", res->type);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
static struct device_attribute ipr_resource_type_attr = {
.attr = {
.name = "resource_type",
.mode = S_IRUGO,
},
.show = ipr_show_resource_type
};
ipr: AF DASD raw mode implementation in ipr driver This patch implements raw mode support for AF DASD in ipr driver which allows for tools to send commands directly to physical devices which are members of RAID arrays when enabled in the firmware. [jejb: fix up whitespace] Signed-off-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:58 -06:00
/**
* ipr_show_raw_mode - Show the adapter's raw mode
* @dev: class device struct
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_show_raw_mode(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct scsi_device *sdev = to_scsi_device(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
ssize_t len;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *)sdev->hostdata;
if (res)
len = snprintf(buf, PAGE_SIZE, "%d\n", res->raw_mode);
else
len = -ENXIO;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
/**
* ipr_store_raw_mode - Change the adapter's raw mode
* @dev: class device struct
* @buf: buffer
*
* Return value:
* number of bytes printed to buffer
**/
static ssize_t ipr_store_raw_mode(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct scsi_device *sdev = to_scsi_device(dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
ssize_t len;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *)sdev->hostdata;
if (res) {
ipr: Enable SIS pipe commands for SIS-32 devices. Remove unnecessary check that disabled SIS pipe commands for SIS-32 devices. This change was sufficient to enable raw mode and send SIS pipe commands for a 57B3 device. Fixes: f8ee25d7d239 ("ipr: AF DASD raw mode implementation in ipr driver") Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-08-19 08:47:06 -06:00
if (ipr_is_af_dasd_device(res)) {
ipr: AF DASD raw mode implementation in ipr driver This patch implements raw mode support for AF DASD in ipr driver which allows for tools to send commands directly to physical devices which are members of RAID arrays when enabled in the firmware. [jejb: fix up whitespace] Signed-off-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:58 -06:00
res->raw_mode = simple_strtoul(buf, NULL, 10);
len = strlen(buf);
if (res->sdev)
sdev_printk(KERN_INFO, res->sdev, "raw mode is %s\n",
res->raw_mode ? "enabled" : "disabled");
} else
len = -EINVAL;
} else
len = -ENXIO;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return len;
}
static struct device_attribute ipr_raw_mode_attr = {
.attr = {
.name = "raw_mode",
.mode = S_IRUGO | S_IWUSR,
},
.show = ipr_show_raw_mode,
.store = ipr_store_raw_mode
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
static struct device_attribute *ipr_dev_attrs[] = {
&ipr_adapter_handle_attr,
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
&ipr_resource_path_attr,
[SCSI] ipr: fix resource address formatting and add attribute for device ID This patch fixes a resource address formatting problem where the first byte was being zeroed out. Also, the device ID is now made available as a sysfs attribute. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-11 08:15:17 -06:00
&ipr_device_id_attr,
[SCSI] ipr: fix resource type update and add sdev and shost attributes Setting the resource type in the ipr_update_res_entry function was incorrect in that the top 4 bits were masked off. The assignment has been updated to no longer mask those bits. Then, two new attributes were added to allow the user space utilities to more easily get information. The resource_type sdev attribute is set for all devices in the adapter's configuration table and indicates the type of device. The fw_type shost attribute indicates the firmware type supported by the adapter. Finally, the resource_path attribute was changed to be mode S_IRUGO. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:50:14 -06:00
&ipr_resource_type_attr,
ipr: AF DASD raw mode implementation in ipr driver This patch implements raw mode support for AF DASD in ipr driver which allows for tools to send commands directly to physical devices which are members of RAID arrays when enabled in the firmware. [jejb: fix up whitespace] Signed-off-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:58 -06:00
&ipr_raw_mode_attr,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
NULL,
};
/**
* ipr_biosparam - Return the HSC mapping
* @sdev: scsi device struct
* @block_device: block device pointer
* @capacity: capacity of the device
* @parm: Array containing returned HSC values.
*
* This function generates the HSC parms that fdisk uses.
* We want to make sure we return something that places partitions
* on 4k boundaries for best performance with the IOA.
*
* Return value:
* 0 on success
**/
static int ipr_biosparam(struct scsi_device *sdev,
struct block_device *block_device,
sector_t capacity, int *parm)
{
int heads, sectors;
sector_t cylinders;
heads = 128;
sectors = 32;
cylinders = capacity;
sector_div(cylinders, (128 * 32));
/* return result */
parm[0] = heads;
parm[1] = sectors;
parm[2] = cylinders;
return 0;
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
* ipr_find_starget - Find target based on bus/target.
* @starget: scsi target struct
*
* Return value:
* resource entry pointer if found / NULL if not found
**/
static struct ipr_resource_entry *ipr_find_starget(struct scsi_target *starget)
{
struct Scsi_Host *shost = dev_to_shost(&starget->dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *) shost->hostdata;
struct ipr_resource_entry *res;
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if ((res->bus == starget->channel) &&
[SCSI] ipr: Fix target id allocation re-use problem For the latest ipr SAS adapters, target id's are a completely logical construct that are managed in the ipr driver. This fixes an issue that can arise if a device is deleted via sysfs. If a new device is then physically added, it will use the previous device's target id. If the host is then rescanned, the device that had been deleted, since it is using the same target id as the new device is using, will never be found, resulting in a missing device. Fix this by only freeing the target id only if the resource is actually gone. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:06 -06:00
(res->target == starget->id)) {
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
return res;
}
}
return NULL;
}
static struct ata_port_info sata_port_info;
/**
* ipr_target_alloc - Prepare for commands to a SCSI target
* @starget: scsi target struct
*
* If the device is a SATA device, this function allocates an
* ATA port with libata, else it does nothing.
*
* Return value:
* 0 on success / non-0 on failure
**/
static int ipr_target_alloc(struct scsi_target *starget)
{
struct Scsi_Host *shost = dev_to_shost(&starget->dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *) shost->hostdata;
struct ipr_sata_port *sata_port;
struct ata_port *ap;
struct ipr_resource_entry *res;
unsigned long lock_flags;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = ipr_find_starget(starget);
starget->hostdata = NULL;
if (res && ipr_is_gata(res)) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
sata_port = kzalloc(sizeof(*sata_port), GFP_KERNEL);
if (!sata_port)
return -ENOMEM;
ap = ata_sas_port_alloc(&ioa_cfg->ata_host, &sata_port_info, shost);
if (ap) {
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
sata_port->ioa_cfg = ioa_cfg;
sata_port->ap = ap;
sata_port->res = res;
res->sata_port = sata_port;
ap->private_data = sata_port;
starget->hostdata = sata_port;
} else {
kfree(sata_port);
return -ENOMEM;
}
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return 0;
}
/**
* ipr_target_destroy - Destroy a SCSI target
* @starget: scsi target struct
*
* If the device was a SATA device, this function frees the libata
* ATA port, else it does nothing.
*
**/
static void ipr_target_destroy(struct scsi_target *starget)
{
struct ipr_sata_port *sata_port = starget->hostdata;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
struct Scsi_Host *shost = dev_to_shost(&starget->dev);
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *) shost->hostdata;
if (ioa_cfg->sis64) {
[SCSI] ipr: Fix target id allocation re-use problem For the latest ipr SAS adapters, target id's are a completely logical construct that are managed in the ipr driver. This fixes an issue that can arise if a device is deleted via sysfs. If a new device is then physically added, it will use the previous device's target id. If the host is then rescanned, the device that had been deleted, since it is using the same target id as the new device is using, will never be found, resulting in a missing device. Fix this by only freeing the target id only if the resource is actually gone. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:06 -06:00
if (!ipr_find_starget(starget)) {
if (starget->channel == IPR_ARRAY_VIRTUAL_BUS)
clear_bit(starget->id, ioa_cfg->array_ids);
else if (starget->channel == IPR_VSET_VIRTUAL_BUS)
clear_bit(starget->id, ioa_cfg->vset_ids);
else if (starget->channel == 0)
clear_bit(starget->id, ioa_cfg->target_ids);
}
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (sata_port) {
starget->hostdata = NULL;
ata_sas_port_destroy(sata_port->ap);
kfree(sata_port);
}
}
/**
* ipr_find_sdev - Find device based on bus/target/lun.
* @sdev: scsi device struct
*
* Return value:
* resource entry pointer if found / NULL if not found
**/
static struct ipr_resource_entry *ipr_find_sdev(struct scsi_device *sdev)
{
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *) sdev->host->hostdata;
struct ipr_resource_entry *res;
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if ((res->bus == sdev->channel) &&
(res->target == sdev->id) &&
(res->lun == sdev->lun))
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
return res;
}
return NULL;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_slave_destroy - Unconfigure a SCSI device
* @sdev: scsi device struct
*
* Return value:
* nothing
**/
static void ipr_slave_destroy(struct scsi_device *sdev)
{
struct ipr_resource_entry *res;
struct ipr_ioa_cfg *ioa_cfg;
unsigned long lock_flags = 0;
ioa_cfg = (struct ipr_ioa_cfg *) sdev->host->hostdata;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = (struct ipr_resource_entry *) sdev->hostdata;
if (res) {
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (res->sata_port)
libata: kill ATA_FLAG_DISABLED ATA_FLAG_DISABLED is only used by drivers which don't use ->error_handler framework and is largely broken. Its only meaningful function is to make irq handlers skip processing if the flag is set, which is largely useless and even harmful as it makes those ports more likely to cause IRQ storms. Kill ATA_FLAG_DISABLED and makes the callers disable attached devices instead. ata_port_probe() and ata_port_disable() which manipulate the flag are also killed. This simplifies condition check in IRQ handlers. While updating IRQ handlers, remove ap NULL check as libata guarantees consecutive port allocation (unoccupied ports are initialized with dummies) and long-obsolete ATA_QCFLAG_ACTIVE check (checked by ata_qc_from_tag()). Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2010-05-10 13:41:30 -06:00
res->sata_port->ap->link.device[0].class = ATA_DEV_NONE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sdev->hostdata = NULL;
res->sdev = NULL;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
res->sata_port = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
}
/**
* ipr_slave_configure - Configure a SCSI device
* @sdev: scsi device struct
*
* This function configures the specified scsi device.
*
* Return value:
* 0 on success
**/
static int ipr_slave_configure(struct scsi_device *sdev)
{
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *) sdev->host->hostdata;
struct ipr_resource_entry *res;
[SCSI] ipr: Fix sleeping function called with interrupts disabled The ata_sas_slave_configure was changed such that it now allocates some memory for a drain buffer for ATAPI devices. Fixup the ipr driver such that we no longer make this call with interrupts disabled. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-04-22 07:58:02 -06:00
struct ata_port *ap = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
unsigned long lock_flags = 0;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
char buffer[IPR_MAX_RES_PATH_LENGTH];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
res = sdev->hostdata;
if (res) {
if (ipr_is_af_dasd_device(res))
sdev->type = TYPE_RAID;
[SCSI] ipr: Prevent upper layer driver binding Set the no_uld_attach for devices ipr does not want upper layer drivers to attach to. These devices are only reported for RAID management and only sg should be used to talk to them. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:01 -07:00
if (ipr_is_af_dasd_device(res) || ipr_is_ioa_resource(res)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sdev->scsi_level = 4;
[SCSI] ipr: Prevent upper layer driver binding Set the no_uld_attach for devices ipr does not want upper layer drivers to attach to. These devices are only reported for RAID management and only sg should be used to talk to them. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:01 -07:00
sdev->no_uld_attach = 1;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ipr_is_vset_device(res)) {
ipr: set scsi_level correctly for disk arrays Set the scsi_level correctly for disk arrays such that things like the rotational field get set properly by sd.c. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:46 -07:00
sdev->scsi_level = SCSI_SPC_3;
scsi: ipr: Set no_report_opcodes for RAID arrays Since ipr RAID arrays do not support the MAINTENANCE_IN / MI_REPORT_SUPPORTED_OPERATION_CODES, set no_report_opcodes to prevent it from being sent. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 15:17:32 -06:00
sdev->no_report_opcodes = 1;
block: unify request timeout handling Right now SCSI and others do their own command timeout handling. Move those bits to the block layer. Instead of having a timer per command, we try to be a bit more clever and simply have one per-queue. This avoids the overhead of having to tear down and setup a timer for each command, so it will result in a lot less timer fiddling. Signed-off-by: Mike Anderson <andmike@linux.vnet.ibm.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2008-09-14 06:55:09 -06:00
blk_queue_rq_timeout(sdev->request_queue,
IPR_VSET_RW_TIMEOUT);
block: Rename blk_queue_max_sectors to blk_queue_max_hw_sectors The block layer calling convention is blk_queue_<limit name>. blk_queue_max_sectors predates this practice, leading to some confusion. Rename the function to appropriately reflect that its intended use is to set max_hw_sectors. Also introduce a temporary wrapper for backwards compability. This can be removed after the merge window is closed. Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-02-25 22:20:38 -07:00
blk_queue_max_hw_sectors(sdev->request_queue, IPR_VSET_MAX_SECTORS);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Fix sleeping function called with interrupts disabled The ata_sas_slave_configure was changed such that it now allocates some memory for a drain buffer for ATAPI devices. Fixup the ipr driver such that we no longer make this call with interrupts disabled. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-04-22 07:58:02 -06:00
if (ipr_is_gata(res) && res->sata_port)
ap = res->sata_port->ap;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
if (ap) {
scsi: drop reason argument from ->change_queue_depth Drop the now unused reason argument from the ->change_queue_depth method. Also add a return value to scsi_adjust_queue_depth, and rename it to scsi_change_queue_depth now that it can be used as the default ->change_queue_depth implementation. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Reviewed-by: Hannes Reinecke <hare@suse.de>
2014-11-13 07:08:42 -07:00
scsi_change_queue_depth(sdev, IPR_MAX_CMD_PER_ATA_LUN);
[SCSI] ipr: Fix sleeping function called with interrupts disabled The ata_sas_slave_configure was changed such that it now allocates some memory for a drain buffer for ATAPI devices. Fixup the ipr driver such that we no longer make this call with interrupts disabled. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-04-22 07:58:02 -06:00
ata_sas_slave_configure(sdev, ap);
scsi: don't set tagging state from scsi_adjust_queue_depth Remove the tagged argument from scsi_adjust_queue_depth, and just let it handle the queue depth. For most drivers those two are fairly separate, given that most modern drivers don't care about the SCSI "tagged" status of a command at all, and many old drivers allow queuing of multiple untagged commands in the driver. Instead we start out with the ->simple_tags flag set before calling ->slave_configure, which is how all drivers actually looking at ->simple_tags except for one worke anyway. The one other case looks broken, but I've kept the behavior as-is for now. Except for that we only change ->simple_tags from the ->change_queue_type, and when rejecting a tag message in a single driver, so keeping this churn out of scsi_adjust_queue_depth is a clear win. Now that the usage of scsi_adjust_queue_depth is more obvious we can also remove all the trivial instances in ->slave_alloc or ->slave_configure that just set it to the cmd_per_lun default. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2014-11-03 12:15:14 -07:00
}
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (ioa_cfg->sis64)
sdev_printk(KERN_INFO, sdev, "Resource path: %s\n",
[SCSI] ipr: Resource path error logging cleanup The resource path as displayed by the ipr driver is the location string identifying a location on the SAS fabric. This patch adds the SCSI host number such that error logs can be more easily correlated in multiple adapter configurations. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:49 -07:00
ipr_format_res_path(ioa_cfg,
res->res_path, buffer, sizeof(buffer)));
[SCSI] ipr: Fix sleeping function called with interrupts disabled The ata_sas_slave_configure was changed such that it now allocates some memory for a drain buffer for ATAPI devices. Fixup the ipr driver such that we no longer make this call with interrupts disabled. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-04-22 07:58:02 -06:00
return 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return 0;
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
* ipr_ata_slave_alloc - Prepare for commands to a SATA device
* @sdev: scsi device struct
*
* This function initializes an ATA port so that future commands
* sent through queuecommand will work.
*
* Return value:
* 0 on success
**/
static int ipr_ata_slave_alloc(struct scsi_device *sdev)
{
struct ipr_sata_port *sata_port = NULL;
int rc = -ENXIO;
ENTER;
if (sdev->sdev_target)
sata_port = sdev->sdev_target->hostdata;
[SCSI] libsas, libata: fix start of life for a sas ata_port This changes the ordering of initialization and probing events from: 1/ allocate rphy in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 2/ allocate ata_port and schedule port probe in DISCE_PROBE ...to: 1/ allocate ata_port in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 2/ allocate rphy in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 3/ schedule port probe in DISCE_PROBE This ordering prevents PHYE_SIGNAL_LOSS_EVENTS from sneaking in to destrory ata devices before they have been fully initialized: BUG: unable to handle kernel paging request at 0000000000003b10 IP: [<ffffffffa0053d7e>] sas_ata_end_eh+0x12/0x5e [libsas] ... [<ffffffffa004d1af>] sas_unregister_common_dev+0x78/0xc9 [libsas] [<ffffffffa004d4d4>] sas_unregister_dev+0x4f/0xad [libsas] [<ffffffffa004d5b1>] sas_unregister_domain_devices+0x7f/0xbf [libsas] [<ffffffffa004c487>] sas_deform_port+0x61/0x1b8 [libsas] [<ffffffffa004bed0>] sas_phye_loss_of_signal+0x29/0x2b [libsas] ...and kills the awkward "sata domain_device briefly existing in the domain without an ata_port" state. Reported-by: Michal Kosciowski <michal.kosciowski@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Acked-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-21 22:09:07 -06:00
if (sata_port) {
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
rc = ata_sas_port_init(sata_port->ap);
[SCSI] libsas, libata: fix start of life for a sas ata_port This changes the ordering of initialization and probing events from: 1/ allocate rphy in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 2/ allocate ata_port and schedule port probe in DISCE_PROBE ...to: 1/ allocate ata_port in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 2/ allocate rphy in PORTE_BYTES_DMAED, DISCE_REVALIDATE_DOMAIN 3/ schedule port probe in DISCE_PROBE This ordering prevents PHYE_SIGNAL_LOSS_EVENTS from sneaking in to destrory ata devices before they have been fully initialized: BUG: unable to handle kernel paging request at 0000000000003b10 IP: [<ffffffffa0053d7e>] sas_ata_end_eh+0x12/0x5e [libsas] ... [<ffffffffa004d1af>] sas_unregister_common_dev+0x78/0xc9 [libsas] [<ffffffffa004d4d4>] sas_unregister_dev+0x4f/0xad [libsas] [<ffffffffa004d5b1>] sas_unregister_domain_devices+0x7f/0xbf [libsas] [<ffffffffa004c487>] sas_deform_port+0x61/0x1b8 [libsas] [<ffffffffa004bed0>] sas_phye_loss_of_signal+0x29/0x2b [libsas] ...and kills the awkward "sata domain_device briefly existing in the domain without an ata_port" state. Reported-by: Michal Kosciowski <michal.kosciowski@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Acked-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-21 22:09:07 -06:00
if (rc == 0)
rc = ata_sas_sync_probe(sata_port->ap);
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (rc)
ipr_slave_destroy(sdev);
LEAVE;
return rc;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_slave_alloc - Prepare for commands to a device.
* @sdev: scsi device struct
*
* This function saves a pointer to the resource entry
* in the scsi device struct if the device exists. We
* can then use this pointer in ipr_queuecommand when
* handling new commands.
*
* Return value:
[SCSI] ipr: slave_alloc optimization Optimize ipr's slave_alloc to return -ENXIO for devices that do not exist. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:07 -07:00
* 0 on success / -ENXIO if device does not exist
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
**/
static int ipr_slave_alloc(struct scsi_device *sdev)
{
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *) sdev->host->hostdata;
struct ipr_resource_entry *res;
unsigned long lock_flags;
[SCSI] ipr: slave_alloc optimization Optimize ipr's slave_alloc to return -ENXIO for devices that do not exist. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:07 -07:00
int rc = -ENXIO;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sdev->hostdata = NULL;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
res = ipr_find_sdev(sdev);
if (res) {
res->sdev = sdev;
res->add_to_ml = 0;
res->in_erp = 0;
sdev->hostdata = res;
if (!ipr_is_naca_model(res))
res->needs_sync_complete = 1;
rc = 0;
if (ipr_is_gata(res)) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return ipr_ata_slave_alloc(sdev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: slave_alloc optimization Optimize ipr's slave_alloc to return -ENXIO for devices that do not exist. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:07 -07:00
return rc;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
/**
* ipr_match_lun - Match function for specified LUN
* @ipr_cmd: ipr command struct
* @device: device to match (sdev)
*
* Returns:
* 1 if command matches sdev / 0 if command does not match sdev
**/
static int ipr_match_lun(struct ipr_cmnd *ipr_cmd, void *device)
{
if (ipr_cmd->scsi_cmd && ipr_cmd->scsi_cmd->device == device)
return 1;
return 0;
}
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
/**
* ipr_cmnd_is_free - Check if a command is free or not
* @ipr_cmd ipr command struct
*
* Returns:
* true / false
**/
static bool ipr_cmnd_is_free(struct ipr_cmnd *ipr_cmd)
{
struct ipr_cmnd *loop_cmd;
list_for_each_entry(loop_cmd, &ipr_cmd->hrrq->hrrq_free_q, queue) {
if (loop_cmd == ipr_cmd)
return true;
}
return false;
}
scsi: ipr: Fix SATA EH hang This patch fixes a hang that can occur in ATA EH with ipr. With ipr's usage of libata, commands should never end up on ap->eh_done_q. The timeout function we use for ipr, even for SATA devices, is scsi_times_out, so ATA_QCFLAG_EH_SCHEDULED never gets set for ipr and EH is driven completely by ipr and SCSI. The SCSI EH thread ends up calling ipr's eh_device_reset_handler, which then calls ata_std_error_handler. This ends up calling ipr_sata_reset, which issues a reset to the device. This should result in all pending commands getting failed back and having ata_qc_complete called for them, which should end up clearing ATA_QCFLAG_FAILED as qc->flags gets zeroed in ata_qc_free. This ensures that when we end up in ata_eh_finish, we don't do anything more with the command. On adapters that only support a single interrupt and when running with two MSI-X vectors or less, the adapter firmware guarantees that responses to all outstanding commands are sent back prior to sending the response to the SATA reset command. On newer adapters supporting multiple HRRQs, however, this can no longer be guaranteed, since the command responses and reset response may be processed on different HRRQs. If ipr returns from ipr_sata_reset before the outstanding command was returned, this sends us down the path of __ata_eh_qc_complete which then moves the associated scsi_cmd from the work_q in scsi_eh_bus_device_reset to ap->eh_done_q, which then will sit there forever and we will be wedged. This patch fixes this up by ensuring that any outstanding commands are flushed before returning from eh_device_reset_handler for a SATA device. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:41 -06:00
/**
* ipr_match_res - Match function for specified resource entry
* @ipr_cmd: ipr command struct
* @resource: resource entry to match
*
* Returns:
* 1 if command matches sdev / 0 if command does not match sdev
**/
static int ipr_match_res(struct ipr_cmnd *ipr_cmd, void *resource)
{
struct ipr_resource_entry *res = resource;
if (res && ipr_cmd->ioarcb.res_handle == res->res_handle)
return 1;
return 0;
}
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
/**
* ipr_wait_for_ops - Wait for matching commands to complete
* @ipr_cmd: ipr command struct
* @device: device to match (sdev)
* @match: match function to use
*
* Returns:
* SUCCESS / FAILED
**/
static int ipr_wait_for_ops(struct ipr_ioa_cfg *ioa_cfg, void *device,
int (*match)(struct ipr_cmnd *, void *))
{
struct ipr_cmnd *ipr_cmd;
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
int wait, i;
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
unsigned long flags;
struct ipr_hrr_queue *hrrq;
signed long timeout = IPR_ABORT_TASK_TIMEOUT;
DECLARE_COMPLETION_ONSTACK(comp);
ENTER;
do {
wait = 0;
for_each_hrrq(hrrq, ioa_cfg) {
spin_lock_irqsave(hrrq->lock, flags);
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
ipr_cmd = ioa_cfg->ipr_cmnd_list[i];
if (!ipr_cmnd_is_free(ipr_cmd)) {
if (match(ipr_cmd, device)) {
ipr_cmd->eh_comp = &comp;
wait++;
}
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
}
}
spin_unlock_irqrestore(hrrq->lock, flags);
}
if (wait) {
timeout = wait_for_completion_timeout(&comp, timeout);
if (!timeout) {
wait = 0;
for_each_hrrq(hrrq, ioa_cfg) {
spin_lock_irqsave(hrrq->lock, flags);
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
ipr_cmd = ioa_cfg->ipr_cmnd_list[i];
if (!ipr_cmnd_is_free(ipr_cmd)) {
if (match(ipr_cmd, device)) {
ipr_cmd->eh_comp = NULL;
wait++;
}
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
}
}
spin_unlock_irqrestore(hrrq->lock, flags);
}
if (wait)
dev_err(&ioa_cfg->pdev->dev, "Timed out waiting for aborted commands\n");
LEAVE;
return wait ? FAILED : SUCCESS;
}
}
} while (wait);
LEAVE;
return SUCCESS;
}
[SCSI] ipr: Fix sparse error in ipr driver This patch fixes the following sparse error: CHECK drivers/scsi/ipr.c spinlock.h:147:9: warning: context imbalance in 'ipr_reset_reload' - unexpected unlock Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:54 -07:00
static int ipr_eh_host_reset(struct scsi_cmnd *cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioa_cfg *ioa_cfg;
[SCSI] ipr: Fix sparse error in ipr driver This patch fixes the following sparse error: CHECK drivers/scsi/ipr.c spinlock.h:147:9: warning: context imbalance in 'ipr_reset_reload' - unexpected unlock Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:54 -07:00
unsigned long lock_flags = 0;
int rc = SUCCESS;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: Fix sparse error in ipr driver This patch fixes the following sparse error: CHECK drivers/scsi/ipr.c spinlock.h:147:9: warning: context imbalance in 'ipr_reset_reload' - unexpected unlock Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:54 -07:00
ioa_cfg = (struct ipr_ioa_cfg *) cmd->device->host->hostdata;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Need to reset adapter after the 6th EEH error Add reset adapter after the 6th EEH errors in ipr driver. This triggers the adapter reset via the PCI config space even when the slot is frozen. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-04-17 08:34:06 -06:00
if (!ioa_cfg->in_reset_reload && !ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead) {
[SCSI] ipr: Fix sparse error in ipr driver This patch fixes the following sparse error: CHECK drivers/scsi/ipr.c spinlock.h:147:9: warning: context imbalance in 'ipr_reset_reload' - unexpected unlock Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:54 -07:00
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_ABBREV);
[SCSI] ipr: fix eeh recovery for 64-bit adapters In some scenarios, an EEH error can take a long time to be detected, since the driver issues an MMIO read only after a device reset command times out and we try to reset the adapter. This patch adds some code in ipr_cancel_op() to read a hardware register so we detect the error earlier in case the op is being aborted because of a timeout caused by a frozen adapter slot. Another problem in such scenarios is that in __ipr_eh_host_reset() we change the dump state flag from WAIT_FOR_DUMP to GET_DUMP, and the flag is later changed from GET_DUMP to READ_DUMP in ipr_reset_restore_cfg_space(). However, if when __ipr_eh_host_reset() is called by the SCSI error handling the function ipr_reset_restore_cfg_space() has already been called by the PCI EEH code, we end up with the flag in an inconsistent state. This patch also prevents this problem. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-01-16 14:30:25 -07:00
dev_err(&ioa_cfg->pdev->dev,
"Adapter being reset as a result of error recovery.\n");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: fix eeh recovery for 64-bit adapters In some scenarios, an EEH error can take a long time to be detected, since the driver issues an MMIO read only after a device reset command times out and we try to reset the adapter. This patch adds some code in ipr_cancel_op() to read a hardware register so we detect the error earlier in case the op is being aborted because of a timeout caused by a frozen adapter slot. Another problem in such scenarios is that in __ipr_eh_host_reset() we change the dump state flag from WAIT_FOR_DUMP to GET_DUMP, and the flag is later changed from GET_DUMP to READ_DUMP in ipr_reset_restore_cfg_space(). However, if when __ipr_eh_host_reset() is called by the SCSI error handling the function ipr_reset_restore_cfg_space() has already been called by the PCI EEH code, we end up with the flag in an inconsistent state. This patch also prevents this problem. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-01-16 14:30:25 -07:00
if (WAIT_FOR_DUMP == ioa_cfg->sdt_state)
ioa_cfg->sdt_state = GET_DUMP;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Fix sparse error in ipr driver This patch fixes the following sparse error: CHECK drivers/scsi/ipr.c spinlock.h:147:9: warning: context imbalance in 'ipr_reset_reload' - unexpected unlock Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:54 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
[SCSI] allow sleeping in ->eh_host_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:57:14 -06:00
[SCSI] ipr: Fix sparse error in ipr driver This patch fixes the following sparse error: CHECK drivers/scsi/ipr.c spinlock.h:147:9: warning: context imbalance in 'ipr_reset_reload' - unexpected unlock Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:54 -07:00
/* If we got hit with a host reset while we were already resetting
the adapter for some reason, and the reset failed. */
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead) {
ipr_trace;
rc = FAILED;
}
[SCSI] allow sleeping in ->eh_host_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:57:14 -06:00
[SCSI] ipr: Fix sparse error in ipr driver This patch fixes the following sparse error: CHECK drivers/scsi/ipr.c spinlock.h:147:9: warning: context imbalance in 'ipr_reset_reload' - unexpected unlock Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:54 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
LEAVE;
[SCSI] allow sleeping in ->eh_host_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:57:14 -06:00
return rc;
}
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
/**
* ipr_device_reset - Reset the device
* @ioa_cfg: ioa config struct
* @res: resource entry struct
*
* This function issues a device reset to the affected device.
* If the device is a SCSI device, a LUN reset will be sent
* to the device first. If that does not work, a target reset
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
* will be sent. If the device is a SATA device, a PHY reset will
* be sent.
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
*
* Return value:
* 0 on success / non-zero on failure
**/
static int ipr_device_reset(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_resource_entry *res)
{
struct ipr_cmnd *ipr_cmd;
struct ipr_ioarcb *ioarcb;
struct ipr_cmd_pkt *cmd_pkt;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
struct ipr_ioarcb_ata_regs *regs;
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
u32 ioasc;
ENTER;
ipr_cmd = ipr_get_free_ipr_cmnd(ioa_cfg);
ioarcb = &ipr_cmd->ioarcb;
cmd_pkt = &ioarcb->cmd_pkt;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ipr_cmd->ioa_cfg->sis64) {
regs = &ipr_cmd->i.ata_ioadl.regs;
ioarcb->add_cmd_parms_offset = cpu_to_be16(sizeof(*ioarcb));
} else
regs = &ioarcb->u.add_data.u.regs;
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ioarcb->res_handle = res->res_handle;
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
cmd_pkt->request_type = IPR_RQTYPE_IOACMD;
cmd_pkt->cdb[0] = IPR_RESET_DEVICE;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (ipr_is_gata(res)) {
cmd_pkt->cdb[2] = IPR_ATA_PHY_RESET;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->add_cmd_parms_len = cpu_to_be16(sizeof(regs->flags));
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
regs->flags |= IPR_ATA_FLAG_STATUS_ON_GOOD_COMPLETION;
}
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
ipr_send_blocking_cmd(ipr_cmd, ipr_timeout, IPR_DEVICE_RESET_TIMEOUT);
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
if (ipr_is_gata(res) && res->sata_port && ioasc != IPR_IOASC_IOA_WAS_RESET) {
if (ipr_cmd->ioa_cfg->sis64)
memcpy(&res->sata_port->ioasa, &ipr_cmd->s.ioasa64.u.gata,
sizeof(struct ipr_ioasa_gata));
else
memcpy(&res->sata_port->ioasa, &ipr_cmd->s.ioasa.u.gata,
sizeof(struct ipr_ioasa_gata));
}
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
LEAVE;
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
return IPR_IOASC_SENSE_KEY(ioasc) ? -EIO : 0;
[SCSI] ipr: Reset device cleanup Encapsulate some more of the device reset processing in preparation for SATA support. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:43 -07:00
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
* ipr_sata_reset - Reset the SATA port
libata-link: linkify reset Make reset methods and related functions deal with ata_link instead of ata_port. * ata_do_reset() * ata_eh_reset() * all prereset/reset/postreset methods and related functions This patch introduces no behavior change. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-08-06 03:36:23 -06:00
* @link: SATA link to reset
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
* @classes: class of the attached device
*
libata-link: linkify reset Make reset methods and related functions deal with ata_link instead of ata_port. * ata_do_reset() * ata_eh_reset() * all prereset/reset/postreset methods and related functions This patch introduces no behavior change. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-08-06 03:36:23 -06:00
* This function issues a SATA phy reset to the affected ATA link.
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
*
* Return value:
* 0 on success / non-zero on failure
**/
libata-link: linkify reset Make reset methods and related functions deal with ata_link instead of ata_port. * ata_do_reset() * ata_eh_reset() * all prereset/reset/postreset methods and related functions This patch introduces no behavior change. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-08-06 03:36:23 -06:00
static int ipr_sata_reset(struct ata_link *link, unsigned int *classes,
git-libata-all-ipr-fix drivers/scsi/ipr.c: In function '__ipr_eh_dev_reset': drivers/scsi/ipr.c:3865: warning: passing argument 4 of 'ata_do_eh' from incompatible pointer type Cc: Tejun Heo <htejun@gmail.com> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-03-26 03:17:43 -06:00
unsigned long deadline)
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
{
libata-link: linkify reset Make reset methods and related functions deal with ata_link instead of ata_port. * ata_do_reset() * ata_eh_reset() * all prereset/reset/postreset methods and related functions This patch introduces no behavior change. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-08-06 03:36:23 -06:00
struct ipr_sata_port *sata_port = link->ap->private_data;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
struct ipr_ioa_cfg *ioa_cfg = sata_port->ioa_cfg;
struct ipr_resource_entry *res;
unsigned long lock_flags = 0;
scsi: ipr: Fix SATA EH hang This patch fixes a hang that can occur in ATA EH with ipr. With ipr's usage of libata, commands should never end up on ap->eh_done_q. The timeout function we use for ipr, even for SATA devices, is scsi_times_out, so ATA_QCFLAG_EH_SCHEDULED never gets set for ipr and EH is driven completely by ipr and SCSI. The SCSI EH thread ends up calling ipr's eh_device_reset_handler, which then calls ata_std_error_handler. This ends up calling ipr_sata_reset, which issues a reset to the device. This should result in all pending commands getting failed back and having ata_qc_complete called for them, which should end up clearing ATA_QCFLAG_FAILED as qc->flags gets zeroed in ata_qc_free. This ensures that when we end up in ata_eh_finish, we don't do anything more with the command. On adapters that only support a single interrupt and when running with two MSI-X vectors or less, the adapter firmware guarantees that responses to all outstanding commands are sent back prior to sending the response to the SATA reset command. On newer adapters supporting multiple HRRQs, however, this can no longer be guaranteed, since the command responses and reset response may be processed on different HRRQs. If ipr returns from ipr_sata_reset before the outstanding command was returned, this sends us down the path of __ata_eh_qc_complete which then moves the associated scsi_cmd from the work_q in scsi_eh_bus_device_reset to ap->eh_done_q, which then will sit there forever and we will be wedged. This patch fixes this up by ensuring that any outstanding commands are flushed before returning from eh_device_reset_handler for a SATA device. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:41 -06:00
int rc = -ENXIO, ret;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
while (ioa_cfg->in_reset_reload) {
[SCSI] ipr: SATA reset - wait for host reset completion If an ipr adapter hits a fatal microcode error requiring a reset while a SATA device is going through EH, it can result in a command getting issued to the ipr adapter while it is getting reset, which can cause PCI bus errors. Wait for any outstanding adapter reset to finish prior to issuing a SATA device reset. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:27:58 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
res = sata_port->res;
if (res) {
rc = ipr_device_reset(ioa_cfg, res);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
*classes = res->ata_class;
scsi: ipr: Fix SATA EH hang This patch fixes a hang that can occur in ATA EH with ipr. With ipr's usage of libata, commands should never end up on ap->eh_done_q. The timeout function we use for ipr, even for SATA devices, is scsi_times_out, so ATA_QCFLAG_EH_SCHEDULED never gets set for ipr and EH is driven completely by ipr and SCSI. The SCSI EH thread ends up calling ipr's eh_device_reset_handler, which then calls ata_std_error_handler. This ends up calling ipr_sata_reset, which issues a reset to the device. This should result in all pending commands getting failed back and having ata_qc_complete called for them, which should end up clearing ATA_QCFLAG_FAILED as qc->flags gets zeroed in ata_qc_free. This ensures that when we end up in ata_eh_finish, we don't do anything more with the command. On adapters that only support a single interrupt and when running with two MSI-X vectors or less, the adapter firmware guarantees that responses to all outstanding commands are sent back prior to sending the response to the SATA reset command. On newer adapters supporting multiple HRRQs, however, this can no longer be guaranteed, since the command responses and reset response may be processed on different HRRQs. If ipr returns from ipr_sata_reset before the outstanding command was returned, this sends us down the path of __ata_eh_qc_complete which then moves the associated scsi_cmd from the work_q in scsi_eh_bus_device_reset to ap->eh_done_q, which then will sit there forever and we will be wedged. This patch fixes this up by ensuring that any outstanding commands are flushed before returning from eh_device_reset_handler for a SATA device. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:41 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
ret = ipr_wait_for_ops(ioa_cfg, res, ipr_match_res);
if (ret != SUCCESS) {
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_ABBREV);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
}
} else
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
LEAVE;
return rc;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_eh_dev_reset - Reset the device
* @scsi_cmd: scsi command struct
*
* This function issues a device reset to the affected device.
* A LUN reset will be sent to the device first. If that does
* not work, a target reset will be sent.
*
* Return value:
* SUCCESS / FAILED
**/
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
static int __ipr_eh_dev_reset(struct scsi_cmnd *scsi_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_cmnd *ipr_cmd;
struct ipr_ioa_cfg *ioa_cfg;
struct ipr_resource_entry *res;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
struct ata_port *ap;
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
int rc = 0, i;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
ioa_cfg = (struct ipr_ioa_cfg *) scsi_cmd->device->host->hostdata;
res = scsi_cmd->device->hostdata;
/*
* If we are currently going through reset/reload, return failed. This will force the
* mid-layer to call ipr_eh_host_reset, which will then go to sleep and wait for the
* reset to complete
*/
if (ioa_cfg->in_reset_reload)
return FAILED;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return FAILED;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for_each_hrrq(hrrq, ioa_cfg) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock(&hrrq->_lock);
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
ipr_cmd = ioa_cfg->ipr_cmnd_list[i];
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
if (ipr_cmd->ioarcb.res_handle == res->res_handle) {
scsi: ipr: Remove redundant initialization Removes some code in __ipr_eh_dev_reset which was modifying the ipr_cmd done function. This should have already been setup at command allocation time and if its since been changed, it means we are in the ipr_erp* functions and need to wait for them to complete and don't want to override that here. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:37 -06:00
if (!ipr_cmd->qc)
continue;
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
if (ipr_cmnd_is_free(ipr_cmd))
continue;
scsi: ipr: Remove redundant initialization Removes some code in __ipr_eh_dev_reset which was modifying the ipr_cmd done function. This should have already been setup at command allocation time and if its since been changed, it means we are in the ipr_erp* functions and need to wait for them to complete and don't want to override that here. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:37 -06:00
ipr_cmd->done = ipr_sata_eh_done;
if (!(ipr_cmd->qc->flags & ATA_QCFLAG_FAILED)) {
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_cmd->qc->err_mask |= AC_ERR_TIMEOUT;
ipr_cmd->qc->flags |= ATA_QCFLAG_FAILED;
}
[SCSI] ipr: Remove ipr_scsi_timed_out Remove ipr's usage of the scsi transport eh_timed_out for handling SATA timeouts. This was only needed in order to set some flags on the qc prior to calling ata_do_eh. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:23 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&hrrq->_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
res->resetting_device = 1;
[SCSI] ipr: printk macro cleanup/removal Remove some unused printk macros, make some more robust, and convert some to use standard printk macros when possible. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:37 -07:00
scmd_printk(KERN_ERR, scsi_cmd, "Resetting device\n");
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (ipr_is_gata(res) && res->sata_port) {
ap = res->sata_port->ap;
spin_unlock_irq(scsi_cmd->device->host->host_lock);
libata: make reset related methods proper port operations Currently reset methods are not specified directly in the ata_port_operations table. If a LLD wants to use custom reset methods, it should construct and use a error_handler which uses those reset methods. It's done this way for two reasons. First, the ops table already contained too many methods and adding four more of them would noticeably increase the amount of necessary boilerplate code all over low level drivers. Second, as ->error_handler uses those reset methods, it can get confusing. ie. By overriding ->error_handler, those reset ops can be made useless making layering a bit hazy. Now that ops table uses inheritance, the first problem doesn't exist anymore. The second isn't completely solved but is relieved by providing default values - most drivers can just override what it has implemented and don't have to concern itself about higher level callbacks. In fact, there currently is no driver which actually modifies error handling behavior. Drivers which override ->error_handler just wraps the standard error handler only to prepare the controller for EH. I don't think making ops layering strict has any noticeable benefit. This patch makes ->prereset, ->softreset, ->hardreset, ->postreset and their PMP counterparts propoer ops. Default ops are provided in the base ops tables and drivers are converted to override individual reset methods instead of creating custom error_handler. * ata_std_error_handler() doesn't use sata_std_hardreset() if SCRs aren't accessible. sata_promise doesn't need to use separate error_handlers for PATA and SATA anymore. * softreset is broken for sata_inic162x and sata_sx4. As libata now always prefers hardreset, this doesn't really matter but the ops are forced to NULL using ATA_OP_NULL for documentation purpose. * pata_hpt374 needs to use different prereset for the first and second PCI functions. This used to be done by branching from hpt374_error_handler(). The proper way to do this is to use separate ops and port_info tables for each function. Converted. Signed-off-by: Tejun Heo <htejun@gmail.com>
2008-03-24 21:22:50 -06:00
ata_std_error_handler(ap);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
spin_lock_irq(scsi_cmd->device->host->host_lock);
} else
rc = ipr_device_reset(ioa_cfg, res);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
res->resetting_device = 0;
[SCSI] ipr: Remove extended delay bit on GSCSI reads/writes ops This patch removes extended delay bit on GSCSI reads/writes ops, the performance will be significanly better. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:39 -07:00
res->reset_occurred = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
return rc ? FAILED : SUCCESS;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
static int ipr_eh_dev_reset(struct scsi_cmnd *cmd)
[SCSI] allow sleeping in ->eh_device_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:55:48 -06:00
{
int rc;
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
struct ipr_ioa_cfg *ioa_cfg;
scsi: ipr: Fix SATA EH hang This patch fixes a hang that can occur in ATA EH with ipr. With ipr's usage of libata, commands should never end up on ap->eh_done_q. The timeout function we use for ipr, even for SATA devices, is scsi_times_out, so ATA_QCFLAG_EH_SCHEDULED never gets set for ipr and EH is driven completely by ipr and SCSI. The SCSI EH thread ends up calling ipr's eh_device_reset_handler, which then calls ata_std_error_handler. This ends up calling ipr_sata_reset, which issues a reset to the device. This should result in all pending commands getting failed back and having ata_qc_complete called for them, which should end up clearing ATA_QCFLAG_FAILED as qc->flags gets zeroed in ata_qc_free. This ensures that when we end up in ata_eh_finish, we don't do anything more with the command. On adapters that only support a single interrupt and when running with two MSI-X vectors or less, the adapter firmware guarantees that responses to all outstanding commands are sent back prior to sending the response to the SATA reset command. On newer adapters supporting multiple HRRQs, however, this can no longer be guaranteed, since the command responses and reset response may be processed on different HRRQs. If ipr returns from ipr_sata_reset before the outstanding command was returned, this sends us down the path of __ata_eh_qc_complete which then moves the associated scsi_cmd from the work_q in scsi_eh_bus_device_reset to ap->eh_done_q, which then will sit there forever and we will be wedged. This patch fixes this up by ensuring that any outstanding commands are flushed before returning from eh_device_reset_handler for a SATA device. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:41 -06:00
struct ipr_resource_entry *res;
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
ioa_cfg = (struct ipr_ioa_cfg *) cmd->device->host->hostdata;
scsi: ipr: Fix SATA EH hang This patch fixes a hang that can occur in ATA EH with ipr. With ipr's usage of libata, commands should never end up on ap->eh_done_q. The timeout function we use for ipr, even for SATA devices, is scsi_times_out, so ATA_QCFLAG_EH_SCHEDULED never gets set for ipr and EH is driven completely by ipr and SCSI. The SCSI EH thread ends up calling ipr's eh_device_reset_handler, which then calls ata_std_error_handler. This ends up calling ipr_sata_reset, which issues a reset to the device. This should result in all pending commands getting failed back and having ata_qc_complete called for them, which should end up clearing ATA_QCFLAG_FAILED as qc->flags gets zeroed in ata_qc_free. This ensures that when we end up in ata_eh_finish, we don't do anything more with the command. On adapters that only support a single interrupt and when running with two MSI-X vectors or less, the adapter firmware guarantees that responses to all outstanding commands are sent back prior to sending the response to the SATA reset command. On newer adapters supporting multiple HRRQs, however, this can no longer be guaranteed, since the command responses and reset response may be processed on different HRRQs. If ipr returns from ipr_sata_reset before the outstanding command was returned, this sends us down the path of __ata_eh_qc_complete which then moves the associated scsi_cmd from the work_q in scsi_eh_bus_device_reset to ap->eh_done_q, which then will sit there forever and we will be wedged. This patch fixes this up by ensuring that any outstanding commands are flushed before returning from eh_device_reset_handler for a SATA device. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:41 -06:00
res = cmd->device->hostdata;
if (!res)
return FAILED;
[SCSI] allow sleeping in ->eh_device_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:55:48 -06:00
spin_lock_irq(cmd->device->host->host_lock);
rc = __ipr_eh_dev_reset(cmd);
spin_unlock_irq(cmd->device->host->host_lock);
scsi: ipr: Fix SATA EH hang This patch fixes a hang that can occur in ATA EH with ipr. With ipr's usage of libata, commands should never end up on ap->eh_done_q. The timeout function we use for ipr, even for SATA devices, is scsi_times_out, so ATA_QCFLAG_EH_SCHEDULED never gets set for ipr and EH is driven completely by ipr and SCSI. The SCSI EH thread ends up calling ipr's eh_device_reset_handler, which then calls ata_std_error_handler. This ends up calling ipr_sata_reset, which issues a reset to the device. This should result in all pending commands getting failed back and having ata_qc_complete called for them, which should end up clearing ATA_QCFLAG_FAILED as qc->flags gets zeroed in ata_qc_free. This ensures that when we end up in ata_eh_finish, we don't do anything more with the command. On adapters that only support a single interrupt and when running with two MSI-X vectors or less, the adapter firmware guarantees that responses to all outstanding commands are sent back prior to sending the response to the SATA reset command. On newer adapters supporting multiple HRRQs, however, this can no longer be guaranteed, since the command responses and reset response may be processed on different HRRQs. If ipr returns from ipr_sata_reset before the outstanding command was returned, this sends us down the path of __ata_eh_qc_complete which then moves the associated scsi_cmd from the work_q in scsi_eh_bus_device_reset to ap->eh_done_q, which then will sit there forever and we will be wedged. This patch fixes this up by ensuring that any outstanding commands are flushed before returning from eh_device_reset_handler for a SATA device. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:41 -06:00
if (rc == SUCCESS) {
if (ipr_is_gata(res) && res->sata_port)
rc = ipr_wait_for_ops(ioa_cfg, res, ipr_match_res);
else
rc = ipr_wait_for_ops(ioa_cfg, cmd->device, ipr_match_lun);
}
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
[SCSI] allow sleeping in ->eh_device_reset_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:55:48 -06:00
return rc;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_bus_reset_done - Op done function for bus reset.
* @ipr_cmd: ipr command struct
*
* This function is the op done function for a bus reset
*
* Return value:
* none
**/
static void ipr_bus_reset_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_resource_entry *res;
ENTER;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (!ioa_cfg->sis64)
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
if (res->res_handle == ipr_cmd->ioarcb.res_handle) {
scsi_report_bus_reset(ioa_cfg->host, res->bus);
break;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/*
* If abort has not completed, indicate the reset has, else call the
* abort's done function to wake the sleeping eh thread
*/
if (ipr_cmd->sibling->sibling)
ipr_cmd->sibling->sibling = NULL;
else
ipr_cmd->sibling->done(ipr_cmd->sibling);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
}
/**
* ipr_abort_timeout - An abort task has timed out
* @ipr_cmd: ipr command struct
*
* This function handles when an abort task times out. If this
* happens we issue a bus reset since we have resources tied
* up that must be freed before returning to the midlayer.
*
* Return value:
* none
**/
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
static void ipr_abort_timeout(struct timer_list *t)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
struct ipr_cmnd *ipr_cmd = from_timer(ipr_cmd, t, timer);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_cmnd *reset_cmd;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_cmd_pkt *cmd_pkt;
unsigned long lock_flags = 0;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (ipr_cmd->completion.done || ioa_cfg->in_reset_reload) {
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return;
}
[SCSI] ipr: printk macro cleanup/removal Remove some unused printk macros, make some more robust, and convert some to use standard printk macros when possible. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:37 -07:00
sdev_printk(KERN_ERR, ipr_cmd->u.sdev, "Abort timed out. Resetting bus.\n");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
reset_cmd = ipr_get_free_ipr_cmnd(ioa_cfg);
ipr_cmd->sibling = reset_cmd;
reset_cmd->sibling = ipr_cmd;
reset_cmd->ioarcb.res_handle = ipr_cmd->ioarcb.res_handle;
cmd_pkt = &reset_cmd->ioarcb.cmd_pkt;
cmd_pkt->request_type = IPR_RQTYPE_IOACMD;
cmd_pkt->cdb[0] = IPR_RESET_DEVICE;
cmd_pkt->cdb[2] = IPR_RESET_TYPE_SELECT | IPR_BUS_RESET;
ipr_do_req(reset_cmd, ipr_bus_reset_done, ipr_timeout, IPR_DEVICE_RESET_TIMEOUT);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
LEAVE;
}
/**
* ipr_cancel_op - Cancel specified op
* @scsi_cmd: scsi command struct
*
* This function cancels specified op.
*
* Return value:
* SUCCESS / FAILED
**/
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
static int ipr_cancel_op(struct scsi_cmnd *scsi_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_cmnd *ipr_cmd;
struct ipr_ioa_cfg *ioa_cfg;
struct ipr_resource_entry *res;
struct ipr_cmd_pkt *cmd_pkt;
[SCSI] ipr: fix eeh recovery for 64-bit adapters In some scenarios, an EEH error can take a long time to be detected, since the driver issues an MMIO read only after a device reset command times out and we try to reset the adapter. This patch adds some code in ipr_cancel_op() to read a hardware register so we detect the error earlier in case the op is being aborted because of a timeout caused by a frozen adapter slot. Another problem in such scenarios is that in __ipr_eh_host_reset() we change the dump state flag from WAIT_FOR_DUMP to GET_DUMP, and the flag is later changed from GET_DUMP to READ_DUMP in ipr_reset_restore_cfg_space(). However, if when __ipr_eh_host_reset() is called by the SCSI error handling the function ipr_reset_restore_cfg_space() has already been called by the PCI EEH code, we end up with the flag in an inconsistent state. This patch also prevents this problem. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-01-16 14:30:25 -07:00
u32 ioasc, int_reg;
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
int i, op_found = 0;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
ioa_cfg = (struct ipr_ioa_cfg *)scsi_cmd->device->host->hostdata;
res = scsi_cmd->device->hostdata;
[SCSI] allow sleeping in ->eh_abort_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:54:40 -06:00
/* If we are currently going through reset/reload, return failed.
* This will force the mid-layer to call ipr_eh_host_reset,
* which will then go to sleep and wait for the reset to complete
*/
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ioa_cfg->in_reset_reload ||
ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead)
[SCSI] allow sleeping in ->eh_abort_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:54:40 -06:00
return FAILED;
[SCSI] ipr: fix eeh recovery for 64-bit adapters In some scenarios, an EEH error can take a long time to be detected, since the driver issues an MMIO read only after a device reset command times out and we try to reset the adapter. This patch adds some code in ipr_cancel_op() to read a hardware register so we detect the error earlier in case the op is being aborted because of a timeout caused by a frozen adapter slot. Another problem in such scenarios is that in __ipr_eh_host_reset() we change the dump state flag from WAIT_FOR_DUMP to GET_DUMP, and the flag is later changed from GET_DUMP to READ_DUMP in ipr_reset_restore_cfg_space(). However, if when __ipr_eh_host_reset() is called by the SCSI error handling the function ipr_reset_restore_cfg_space() has already been called by the PCI EEH code, we end up with the flag in an inconsistent state. This patch also prevents this problem. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-01-16 14:30:25 -07:00
if (!res)
return FAILED;
/*
* If we are aborting a timed out op, chances are that the timeout was caused
* by a still not detected EEH error. In such cases, reading a register will
* trigger the EEH recovery infrastructure.
*/
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg);
if (!ipr_is_gscsi(res))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return FAILED;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for_each_hrrq(hrrq, ioa_cfg) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock(&hrrq->_lock);
scsi: ipr: Fix abort path race condition This fixes a race condition in the error handlomg paths of ipr. While a command is outstanding to the adapter, it is placed on a pending queue for the hrrq it is associated with, while holding the HRRQ lock. When a command is completed, it is removed from the pending queue, under HRRQ lock, and placed on a local list. This list is then iterated through without any locks and each command's done function is invoked, inside of which, the command gets returned to the free list while grabbing the HRRQ lock. This fixes two race conditions when commands have been removed from the pending list but have not yet been added to the free list. Both of these changes fix race conditions that could result in returning success from eh_abort_handler and then later calling scsi_done for the same request. The first race condition is in ipr_cancel_op. It looks through each pending queue to see if the command to be aborted is still outstanding or not. Rather than looking on the pending queue, reverse the logic to check to look for commands that are NOT on the free queue. The second race condition can occur when in ipr_wait_for_ops where we are waiting for responses for commands we've aborted. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
if (ioa_cfg->ipr_cmnd_list[i]->scsi_cmd == scsi_cmd) {
if (!ipr_cmnd_is_free(ioa_cfg->ipr_cmnd_list[i])) {
op_found = 1;
break;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&hrrq->_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
if (!op_found)
return SUCCESS;
ipr_cmd = ipr_get_free_ipr_cmnd(ioa_cfg);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_cmd->ioarcb.res_handle = res->res_handle;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
cmd_pkt = &ipr_cmd->ioarcb.cmd_pkt;
cmd_pkt->request_type = IPR_RQTYPE_IOACMD;
cmd_pkt->cdb[0] = IPR_CANCEL_ALL_REQUESTS;
ipr_cmd->u.sdev = scsi_cmd->device;
[SCSI] ipr: printk macro cleanup/removal Remove some unused printk macros, make some more robust, and convert some to use standard printk macros when possible. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:37 -07:00
scmd_printk(KERN_ERR, scsi_cmd, "Aborting command: %02X\n",
scsi_cmd->cmnd[0]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_send_blocking_cmd(ipr_cmd, ipr_abort_timeout, IPR_CANCEL_ALL_TIMEOUT);
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/*
* If the abort task timed out and we sent a bus reset, we will get
* one the following responses to the abort
*/
if (ioasc == IPR_IOASC_BUS_WAS_RESET || ioasc == IPR_IOASC_SYNC_REQUIRED) {
ioasc = 0;
ipr_trace;
}
[SCSI] ipr: fix addition of abort command to HRRQ free queue The abort command issued by ipr_cancel_op() is being added to the wrong HRRQ free queue after the command returns. Fix it by using the HRRQ pointer in the ipr command struct itself. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-03-14 12:52:23 -06:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (!ipr_is_naca_model(res))
res->needs_sync_complete = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
return IPR_IOASC_SENSE_KEY(ioasc) ? FAILED : SUCCESS;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_eh_abort - Abort a single op
* @scsi_cmd: scsi command struct
*
* Return value:
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
* 0 if scan in progress / 1 if scan is complete
**/
static int ipr_scan_finished(struct Scsi_Host *shost, unsigned long elapsed_time)
{
unsigned long lock_flags;
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *) shost->hostdata;
int rc = 0;
spin_lock_irqsave(shost->host_lock, lock_flags);
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead || ioa_cfg->scan_done)
rc = 1;
if ((elapsed_time/HZ) > (ioa_cfg->transop_timeout * 2))
rc = 1;
spin_unlock_irqrestore(shost->host_lock, lock_flags);
return rc;
}
/**
* ipr_eh_host_reset - Reset the host adapter
* @scsi_cmd: scsi command struct
*
* Return value:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* SUCCESS / FAILED
**/
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
static int ipr_eh_abort(struct scsi_cmnd *scsi_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[SCSI] allow sleeping in ->eh_abort_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:54:40 -06:00
unsigned long flags;
int rc;
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
struct ipr_ioa_cfg *ioa_cfg;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
ioa_cfg = (struct ipr_ioa_cfg *) scsi_cmd->device->host->hostdata;
[SCSI] allow sleeping in ->eh_abort_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:54:40 -06:00
spin_lock_irqsave(scsi_cmd->device->host->host_lock, flags);
rc = ipr_cancel_op(scsi_cmd);
spin_unlock_irqrestore(scsi_cmd->device->host->host_lock, flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: wait for aborted command responses Fixes a race condition in abort handling that was injected when multiple interrupt support was added. When only a single interrupt is present, the adapter guarantees it will send responses for aborted commands prior to the response for the abort command itself. With multiple interrupts, these responses generally come back on different interrupts, so we need to ensure the abort thread waits until the aborted command is complete so we don't perform a double completion. This race condition was being hit frequently in environments which were triggering command timeouts, which was resulting in a double completion causing a kernel oops. Cc: <stable@vger.kernel.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:10 -06:00
if (rc == SUCCESS)
rc = ipr_wait_for_ops(ioa_cfg, scsi_cmd->device, ipr_match_lun);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
[SCSI] allow sleeping in ->eh_abort_handler() Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-28 05:54:40 -06:00
return rc;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_handle_other_interrupt - Handle "other" interrupts
* @ioa_cfg: ioa config struct
[SCSI] ipr: reverse the isr optimization changes The isr optimization patch that was submitted a few months ago exposed problems with receiving and handling spurious HRRQ interrutps. commit 64ffdb762241c0a9c0c8fac7ea92aa0ba1529334 Author: Wayne Boyer <wayneb@linux.vnet.ibm.com> Date: Wed May 19 11:56:13 2010 -0700 [SCSI] ipr: improve interrupt service routine performance That patch is reverted with this one. A new patch will be submitted once the issue is better understood and properly handled in the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-27 15:45:07 -06:00
* @int_reg: interrupt register
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* IRQ_NONE / IRQ_HANDLED
**/
[SCSI] ipr: reverse the isr optimization changes The isr optimization patch that was submitted a few months ago exposed problems with receiving and handling spurious HRRQ interrutps. commit 64ffdb762241c0a9c0c8fac7ea92aa0ba1529334 Author: Wayne Boyer <wayneb@linux.vnet.ibm.com> Date: Wed May 19 11:56:13 2010 -0700 [SCSI] ipr: improve interrupt service routine performance That patch is reverted with this one. A new patch will be submitted once the issue is better understood and properly handled in the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-27 15:45:07 -06:00
static irqreturn_t ipr_handle_other_interrupt(struct ipr_ioa_cfg *ioa_cfg,
[SCSI] ipr: remove unneeded volatile declarations This patch removes three volatile declarations based on some feedback and code analysis. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-07 13:12:30 -06:00
u32 int_reg)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
irqreturn_t rc = IRQ_HANDLED;
[SCSI] ipr: improve interrupt service routine performance During performance testing on P7 machines it was observed that the interrupt service routine was doing unnecessary MMIO operations. This patch rearranges the logic of the routine and moves some of the code out of the main routine. The result is that there are now fewer MMIO operations in the performance path of the code. As a result of the above change, an existing condition was exposed where the driver could get an "unexpected" hrrq interrupt. The original code would flag the interrupt as unexpected and then reset the adapter. After further analysis it was confirmed that this condition can occasionally occur and that the interrupt can safely be ignored. Additional code in this patch detects this condition, clears the interrupt and allows the driver to continue without resetting the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-12 11:29:02 -06:00
u32 int_mask_reg;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
[SCSI] ipr: improve interrupt service routine performance During performance testing on P7 machines it was observed that the interrupt service routine was doing unnecessary MMIO operations. This patch rearranges the logic of the routine and moves some of the code out of the main routine. The result is that there are now fewer MMIO operations in the performance path of the code. As a result of the above change, an existing condition was exposed where the driver could get an "unexpected" hrrq interrupt. The original code would flag the interrupt as unexpected and then reset the adapter. After further analysis it was confirmed that this condition can occasionally occur and that the interrupt can safely be ignored. Additional code in this patch detects this condition, clears the interrupt and allows the driver to continue without resetting the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-12 11:29:02 -06:00
int_mask_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg32);
int_reg &= ~int_mask_reg;
/* If an interrupt on the adapter did not occur, ignore it.
* Or in the case of SIS 64, check for a stage change interrupt.
*/
if ((int_reg & IPR_PCII_OPER_INTERRUPTS) == 0) {
if (ioa_cfg->sis64) {
int_mask_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg);
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg) & ~int_mask_reg;
if (int_reg & IPR_PCII_IPL_STAGE_CHANGE) {
/* clear stage change */
writel(IPR_PCII_IPL_STAGE_CHANGE, ioa_cfg->regs.clr_interrupt_reg);
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg) & ~int_mask_reg;
list_del(&ioa_cfg->reset_cmd->queue);
del_timer(&ioa_cfg->reset_cmd->timer);
ipr_reset_ioa_job(ioa_cfg->reset_cmd);
return IRQ_HANDLED;
}
}
return IRQ_NONE;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (int_reg & IPR_PCII_IOA_TRANS_TO_OPER) {
/* Mask the interrupt */
writel(IPR_PCII_IOA_TRANS_TO_OPER, ioa_cfg->regs.set_interrupt_mask_reg);
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg);
list_del(&ioa_cfg->reset_cmd->queue);
del_timer(&ioa_cfg->reset_cmd->timer);
ipr_reset_ioa_job(ioa_cfg->reset_cmd);
[SCSI] ipr: improve interrupt service routine performance During performance testing on P7 machines it was observed that the interrupt service routine was doing unnecessary MMIO operations. This patch rearranges the logic of the routine and moves some of the code out of the main routine. The result is that there are now fewer MMIO operations in the performance path of the code. As a result of the above change, an existing condition was exposed where the driver could get an "unexpected" hrrq interrupt. The original code would flag the interrupt as unexpected and then reset the adapter. After further analysis it was confirmed that this condition can occasionally occur and that the interrupt can safely be ignored. Additional code in this patch detects this condition, clears the interrupt and allows the driver to continue without resetting the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-12 11:29:02 -06:00
} else if ((int_reg & IPR_PCII_HRRQ_UPDATED) == int_reg) {
[SCSI] ipr: Remove unnecessary interrupt clearing on new adapters The latest ipr hardware no longer requires the driver to issue any MMIOs to clear the interrupt so remove this to optimize performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:08 -06:00
if (ioa_cfg->clear_isr) {
if (ipr_debug && printk_ratelimit())
dev_err(&ioa_cfg->pdev->dev,
"Spurious interrupt detected. 0x%08X\n", int_reg);
writel(IPR_PCII_HRRQ_UPDATED, ioa_cfg->regs.clr_interrupt_reg32);
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg32);
return IRQ_NONE;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
} else {
if (int_reg & IPR_PCII_IOA_UNIT_CHECKED)
ioa_cfg->ioa_unit_checked = 1;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
else if (int_reg & IPR_PCII_NO_HOST_RRQ)
dev_err(&ioa_cfg->pdev->dev,
"No Host RRQ. 0x%08X\n", int_reg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
else
dev_err(&ioa_cfg->pdev->dev,
"Permanent IOA failure. 0x%08X\n", int_reg);
if (WAIT_FOR_DUMP == ioa_cfg->sdt_state)
ioa_cfg->sdt_state = GET_DUMP;
ipr_mask_and_clear_interrupts(ioa_cfg, ~0);
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return rc;
}
[SCSI] ipr: add workaround for MSI interrupts on P7 This patch adds some additional logic to the interrupt service routine to fix a potential problem where an MSI interrupt does not get cleared the first time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-20 12:09:00 -06:00
/**
* ipr_isr_eh - Interrupt service routine error handler
* @ioa_cfg: ioa config struct
* @msg: message to log
*
* Return value:
* none
**/
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
static void ipr_isr_eh(struct ipr_ioa_cfg *ioa_cfg, char *msg, u16 number)
[SCSI] ipr: add workaround for MSI interrupts on P7 This patch adds some additional logic to the interrupt service routine to fix a potential problem where an MSI interrupt does not get cleared the first time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-20 12:09:00 -06:00
{
ioa_cfg->errors_logged++;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
dev_err(&ioa_cfg->pdev->dev, "%s %d\n", msg, number);
[SCSI] ipr: add workaround for MSI interrupts on P7 This patch adds some additional logic to the interrupt service routine to fix a potential problem where an MSI interrupt does not get cleared the first time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-20 12:09:00 -06:00
if (WAIT_FOR_DUMP == ioa_cfg->sdt_state)
ioa_cfg->sdt_state = GET_DUMP;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
}
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
static int ipr_process_hrrq(struct ipr_hrr_queue *hrr_queue, int budget,
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct list_head *doneq)
{
u32 ioasc;
u16 cmd_index;
struct ipr_cmnd *ipr_cmd;
struct ipr_ioa_cfg *ioa_cfg = hrr_queue->ioa_cfg;
int num_hrrq = 0;
/* If interrupts are disabled, ignore the interrupt */
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (!hrr_queue->allow_interrupts)
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
return 0;
while ((be32_to_cpu(*hrr_queue->hrrq_curr) & IPR_HRRQ_TOGGLE_BIT) ==
hrr_queue->toggle_bit) {
cmd_index = (be32_to_cpu(*hrr_queue->hrrq_curr) &
IPR_HRRQ_REQ_RESP_HANDLE_MASK) >>
IPR_HRRQ_REQ_RESP_HANDLE_SHIFT;
if (unlikely(cmd_index > hrr_queue->max_cmd_id ||
cmd_index < hrr_queue->min_cmd_id)) {
ipr_isr_eh(ioa_cfg,
"Invalid response handle from IOA: ",
cmd_index);
break;
}
ipr_cmd = ioa_cfg->ipr_cmnd_list[cmd_index];
ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
ipr_trc_hook(ipr_cmd, IPR_TRACE_FINISH, ioasc);
list_move_tail(&ipr_cmd->queue, doneq);
if (hrr_queue->hrrq_curr < hrr_queue->hrrq_end) {
hrr_queue->hrrq_curr++;
} else {
hrr_queue->hrrq_curr = hrr_queue->hrrq_start;
hrr_queue->toggle_bit ^= 1u;
}
num_hrrq++;
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
if (budget > 0 && num_hrrq >= budget)
break;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
return num_hrrq;
}
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
static int ipr_iopoll(struct irq_poll *iop, int budget)
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
{
struct ipr_ioa_cfg *ioa_cfg;
struct ipr_hrr_queue *hrrq;
struct ipr_cmnd *ipr_cmd, *temp;
unsigned long hrrq_flags;
int completed_ops;
LIST_HEAD(doneq);
hrrq = container_of(iop, struct ipr_hrr_queue, iopoll);
ioa_cfg = hrrq->ioa_cfg;
spin_lock_irqsave(hrrq->lock, hrrq_flags);
completed_ops = ipr_process_hrrq(hrrq, budget, &doneq);
if (completed_ops < budget)
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
irq_poll_complete(iop);
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
list_for_each_entry_safe(ipr_cmd, temp, &doneq, queue) {
list_del(&ipr_cmd->queue);
del_timer(&ipr_cmd->timer);
ipr_cmd->fast_done(ipr_cmd);
}
return completed_ops;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_isr - Interrupt service routine
* @irq: irq number
* @devp: pointer to ioa config struct
*
* Return value:
* IRQ_NONE / IRQ_HANDLED
**/
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 07:55:46 -06:00
static irqreturn_t ipr_isr(int irq, void *devp)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq = (struct ipr_hrr_queue *)devp;
struct ipr_ioa_cfg *ioa_cfg = hrrq->ioa_cfg;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
unsigned long hrrq_flags = 0;
[SCSI] ipr: improve interrupt service routine performance During performance testing on P7 machines it was observed that the interrupt service routine was doing unnecessary MMIO operations. This patch rearranges the logic of the routine and moves some of the code out of the main routine. The result is that there are now fewer MMIO operations in the performance path of the code. As a result of the above change, an existing condition was exposed where the driver could get an "unexpected" hrrq interrupt. The original code would flag the interrupt as unexpected and then reset the adapter. After further analysis it was confirmed that this condition can occasionally occur and that the interrupt can safely be ignored. Additional code in this patch detects this condition, clears the interrupt and allows the driver to continue without resetting the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-12 11:29:02 -06:00
u32 int_reg = 0;
[SCSI] ipr: add workaround for MSI interrupts on P7 This patch adds some additional logic to the interrupt service routine to fix a potential problem where an MSI interrupt does not get cleared the first time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-20 12:09:00 -06:00
int num_hrrq = 0;
[SCSI] ipr: improve interrupt service routine performance During performance testing on P7 machines it was observed that the interrupt service routine was doing unnecessary MMIO operations. This patch rearranges the logic of the routine and moves some of the code out of the main routine. The result is that there are now fewer MMIO operations in the performance path of the code. As a result of the above change, an existing condition was exposed where the driver could get an "unexpected" hrrq interrupt. The original code would flag the interrupt as unexpected and then reset the adapter. After further analysis it was confirmed that this condition can occasionally occur and that the interrupt can safely be ignored. Additional code in this patch detects this condition, clears the interrupt and allows the driver to continue without resetting the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-12 11:29:02 -06:00
int irq_none = 0;
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
struct ipr_cmnd *ipr_cmd, *temp;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
irqreturn_t rc = IRQ_NONE;
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
LIST_HEAD(doneq);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock_irqsave(hrrq->lock, hrrq_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* If interrupts are disabled, ignore the interrupt */
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (!hrrq->allow_interrupts) {
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return IRQ_NONE;
}
while (1) {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
if (ipr_process_hrrq(hrrq, -1, &doneq)) {
rc = IRQ_HANDLED;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
if (!ioa_cfg->clear_isr)
break;
[SCSI] ipr: Remove unnecessary interrupt clearing on new adapters The latest ipr hardware no longer requires the driver to issue any MMIOs to clear the interrupt so remove this to optimize performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:08 -06:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Clear the PCI interrupt */
[SCSI] ipr: fix possible false positive detection of stuck interrupt If the driver is getting flooded with interrupts, there's a possibility that the interrupt service routine could falsely detect a stuck interrupt condition and reset the adapter. This patch changes the logic such that the routine will loop back into the command processing code one more time after detecting the stuck interrupt signature. If there are no commands to process after that pass, and the interrupt is still not cleared, then the driver will print the "Error clearing HRRQ" message and reset the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <jbottomley@parallels.com>
2011-05-17 10:18:53 -06:00
num_hrrq = 0;
[SCSI] ipr: add workaround for MSI interrupts on P7 This patch adds some additional logic to the interrupt service routine to fix a potential problem where an MSI interrupt does not get cleared the first time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-20 12:09:00 -06:00
do {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
writel(IPR_PCII_HRRQ_UPDATED,
ioa_cfg->regs.clr_interrupt_reg32);
[SCSI] ipr: improve interrupt service routine performance During performance testing on P7 machines it was observed that the interrupt service routine was doing unnecessary MMIO operations. This patch rearranges the logic of the routine and moves some of the code out of the main routine. The result is that there are now fewer MMIO operations in the performance path of the code. As a result of the above change, an existing condition was exposed where the driver could get an "unexpected" hrrq interrupt. The original code would flag the interrupt as unexpected and then reset the adapter. After further analysis it was confirmed that this condition can occasionally occur and that the interrupt can safely be ignored. Additional code in this patch detects this condition, clears the interrupt and allows the driver to continue without resetting the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-12 11:29:02 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg32);
[SCSI] ipr: add workaround for MSI interrupts on P7 This patch adds some additional logic to the interrupt service routine to fix a potential problem where an MSI interrupt does not get cleared the first time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-20 12:09:00 -06:00
} while (int_reg & IPR_PCII_HRRQ_UPDATED &&
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
num_hrrq++ < IPR_MAX_HRRQ_RETRIES);
[SCSI] ipr: add workaround for MSI interrupts on P7 This patch adds some additional logic to the interrupt service routine to fix a potential problem where an MSI interrupt does not get cleared the first time. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-10-20 12:09:00 -06:00
[SCSI] ipr: improve interrupt service routine performance During performance testing on P7 machines it was observed that the interrupt service routine was doing unnecessary MMIO operations. This patch rearranges the logic of the routine and moves some of the code out of the main routine. The result is that there are now fewer MMIO operations in the performance path of the code. As a result of the above change, an existing condition was exposed where the driver could get an "unexpected" hrrq interrupt. The original code would flag the interrupt as unexpected and then reset the adapter. After further analysis it was confirmed that this condition can occasionally occur and that the interrupt can safely be ignored. Additional code in this patch detects this condition, clears the interrupt and allows the driver to continue without resetting the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-12 11:29:02 -06:00
} else if (rc == IRQ_NONE && irq_none == 0) {
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg32);
irq_none++;
[SCSI] ipr: fix possible false positive detection of stuck interrupt If the driver is getting flooded with interrupts, there's a possibility that the interrupt service routine could falsely detect a stuck interrupt condition and reset the adapter. This patch changes the logic such that the routine will loop back into the command processing code one more time after detecting the stuck interrupt signature. If there are no commands to process after that pass, and the interrupt is still not cleared, then the driver will print the "Error clearing HRRQ" message and reset the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <jbottomley@parallels.com>
2011-05-17 10:18:53 -06:00
} else if (num_hrrq == IPR_MAX_HRRQ_RETRIES &&
int_reg & IPR_PCII_HRRQ_UPDATED) {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
ipr_isr_eh(ioa_cfg,
"Error clearing HRRQ: ", num_hrrq);
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
rc = IRQ_HANDLED;
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
} else
break;
}
if (unlikely(rc == IRQ_NONE))
[SCSI] ipr: reverse the isr optimization changes The isr optimization patch that was submitted a few months ago exposed problems with receiving and handling spurious HRRQ interrutps. commit 64ffdb762241c0a9c0c8fac7ea92aa0ba1529334 Author: Wayne Boyer <wayneb@linux.vnet.ibm.com> Date: Wed May 19 11:56:13 2010 -0700 [SCSI] ipr: improve interrupt service routine performance That patch is reverted with this one. A new patch will be submitted once the issue is better understood and properly handled in the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-08-27 15:45:07 -06:00
rc = ipr_handle_other_interrupt(ioa_cfg, int_reg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
list_for_each_entry_safe(ipr_cmd, temp, &doneq, queue) {
list_del(&ipr_cmd->queue);
del_timer(&ipr_cmd->timer);
ipr_cmd->fast_done(ipr_cmd);
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
return rc;
}
/**
* ipr_isr_mhrrq - Interrupt service routine
* @irq: irq number
* @devp: pointer to ioa config struct
*
* Return value:
* IRQ_NONE / IRQ_HANDLED
**/
static irqreturn_t ipr_isr_mhrrq(int irq, void *devp)
{
struct ipr_hrr_queue *hrrq = (struct ipr_hrr_queue *)devp;
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
struct ipr_ioa_cfg *ioa_cfg = hrrq->ioa_cfg;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
unsigned long hrrq_flags = 0;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_cmnd *ipr_cmd, *temp;
irqreturn_t rc = IRQ_NONE;
LIST_HEAD(doneq);
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock_irqsave(hrrq->lock, hrrq_flags);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
/* If interrupts are disabled, ignore the interrupt */
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (!hrrq->allow_interrupts) {
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
return IRQ_NONE;
}
block: remove old blk_iopoll_enabled variable This was a debugging measure to toggle enabled/disabled when testing. But for real production setups, it's not safe to toggle this setting without either reloading drivers of quiescing IO first. Neither of which the toggle enforces. Additionally, it makes drivers deal with the conditional state. Remove it completely. It's up to the driver whether iopoll is enabled or not. Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-13 09:38:42 -06:00
if (ioa_cfg->iopoll_weight && ioa_cfg->sis64 && ioa_cfg->nvectors > 1) {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
if ((be32_to_cpu(*hrrq->hrrq_curr) & IPR_HRRQ_TOGGLE_BIT) ==
hrrq->toggle_bit) {
irq_poll: fold irq_poll_sched_prep into irq_poll_sched There is no good reason to keep them apart, and this makes using the API a bit simpler. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-12-07 07:41:11 -07:00
irq_poll_sched(&hrrq->iopoll);
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
return IRQ_HANDLED;
}
} else {
if ((be32_to_cpu(*hrrq->hrrq_curr) & IPR_HRRQ_TOGGLE_BIT) ==
hrrq->toggle_bit)
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
if (ipr_process_hrrq(hrrq, -1, &doneq))
rc = IRQ_HANDLED;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_for_each_entry_safe(ipr_cmd, temp, &doneq, queue) {
list_del(&ipr_cmd->queue);
del_timer(&ipr_cmd->timer);
ipr_cmd->fast_done(ipr_cmd);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return rc;
}
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
/**
* ipr_build_ioadl64 - Build a scatter/gather list and map the buffer
* @ioa_cfg: ioa config struct
* @ipr_cmd: ipr command struct
*
* Return value:
* 0 on success / -1 on failure
**/
static int ipr_build_ioadl64(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_cmnd *ipr_cmd)
{
int i, nseg;
struct scatterlist *sg;
u32 length;
u32 ioadl_flags = 0;
struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
struct ipr_ioadl64_desc *ioadl64 = ipr_cmd->i.ioadl64;
length = scsi_bufflen(scsi_cmd);
if (!length)
return 0;
nseg = scsi_dma_map(scsi_cmd);
if (nseg < 0) {
[SCSI] ipr: Rate limit DMA mapping errors I noticed a stream of errors from the IPR driver while doing IOMMU fault injection. Rate limit the errors so we don't clog up the console and logfiles. Signed-off-by: Anton Blanchard <anton@samba.org> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <jbottomley@parallels.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-05-08 18:07:40 -06:00
if (printk_ratelimit())
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dev_err(&ioa_cfg->pdev->dev, "scsi_dma_map failed!\n");
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
return -1;
}
ipr_cmd->dma_use_sg = nseg;
[SCSI] ipr: implement fixes for 64 bit adapter support Implement some small fixes for 64 bit support that were preventing the adapter from becoming operational. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:00 -06:00
ioarcb->data_transfer_length = cpu_to_be32(length);
[SCSI] ipr: set the data list length in the request control block In bring up testing for the new 64 bit adapters, the first read command failed after loading the driver. The cause was that the command requires more than one scatter gather element and the corresponding code to set the data list length in the request control block was missing. This patch adds the correct assignment. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-14 09:55:13 -06:00
ioarcb->ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl64_desc) * ipr_cmd->dma_use_sg);
[SCSI] ipr: implement fixes for 64 bit adapter support Implement some small fixes for 64 bit support that were preventing the adapter from becoming operational. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:00 -06:00
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (scsi_cmd->sc_data_direction == DMA_TO_DEVICE) {
ioadl_flags = IPR_IOADL_FLAGS_WRITE;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
} else if (scsi_cmd->sc_data_direction == DMA_FROM_DEVICE)
ioadl_flags = IPR_IOADL_FLAGS_READ;
scsi_for_each_sg(scsi_cmd, sg, ipr_cmd->dma_use_sg, i) {
ioadl64[i].flags = cpu_to_be32(ioadl_flags);
ioadl64[i].data_len = cpu_to_be32(sg_dma_len(sg));
ioadl64[i].address = cpu_to_be64(sg_dma_address(sg));
}
ioadl64[i-1].flags |= cpu_to_be32(IPR_IOADL_FLAGS_LAST);
return 0;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_build_ioadl - Build a scatter/gather list and map the buffer
* @ioa_cfg: ioa config struct
* @ipr_cmd: ipr command struct
*
* Return value:
* 0 on success / -1 on failure
**/
static int ipr_build_ioadl(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_cmnd *ipr_cmd)
{
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
int i, nseg;
struct scatterlist *sg;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
u32 length;
u32 ioadl_flags = 0;
struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
struct ipr_ioadl_desc *ioadl = ipr_cmd->i.ioadl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
length = scsi_bufflen(scsi_cmd);
if (!length)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return 0;
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
nseg = scsi_dma_map(scsi_cmd);
if (nseg < 0) {
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dev_err(&ioa_cfg->pdev->dev, "scsi_dma_map failed!\n");
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
return -1;
}
[SCSI] ipr: Faster sg list fetch Improve overall command performance by embedding the scatterlist in the command block used by the adapter. This decreases the overall number of DMAs required for a single command. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:50 -06:00
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
ipr_cmd->dma_use_sg = nseg;
if (scsi_cmd->sc_data_direction == DMA_TO_DEVICE) {
ioadl_flags = IPR_IOADL_FLAGS_WRITE;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->data_transfer_length = cpu_to_be32(length);
ioarcb->ioadl_len =
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
cpu_to_be32(sizeof(struct ipr_ioadl_desc) * ipr_cmd->dma_use_sg);
} else if (scsi_cmd->sc_data_direction == DMA_FROM_DEVICE) {
ioadl_flags = IPR_IOADL_FLAGS_READ;
ioarcb->read_data_transfer_length = cpu_to_be32(length);
ioarcb->read_ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl_desc) * ipr_cmd->dma_use_sg);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ipr_cmd->dma_use_sg <= ARRAY_SIZE(ioarcb->u.add_data.u.ioadl)) {
ioadl = ioarcb->u.add_data.u.ioadl;
ioarcb->write_ioadl_addr = cpu_to_be32((ipr_cmd->dma_addr) +
offsetof(struct ipr_ioarcb, u.add_data));
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
ioarcb->read_ioadl_addr = ioarcb->write_ioadl_addr;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
scsi_for_each_sg(scsi_cmd, sg, ipr_cmd->dma_use_sg, i) {
ioadl[i].flags_and_data_len =
cpu_to_be32(ioadl_flags | sg_dma_len(sg));
ioadl[i].address = cpu_to_be32(sg_dma_address(sg));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
ioadl[i-1].flags_and_data_len |= cpu_to_be32(IPR_IOADL_FLAGS_LAST);
return 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
* __ipr_erp_done - Process completion of ERP for a device
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @ipr_cmd: ipr command struct
*
* This function copies the sense buffer into the scsi_cmd
* struct and pushes the scsi_done function.
*
* Return value:
* nothing
**/
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
static void __ipr_erp_done(struct ipr_cmnd *ipr_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
struct ipr_resource_entry *res = scsi_cmd->device->hostdata;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (IPR_IOASC_SENSE_KEY(ioasc) > 0) {
scsi_cmd->result |= (DID_ERROR << 16);
[SCSI] ipr: printk macro cleanup/removal Remove some unused printk macros, make some more robust, and convert some to use standard printk macros when possible. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:37 -07:00
scmd_printk(KERN_ERR, scsi_cmd,
"Request Sense failed with IOASC: 0x%08X\n", ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
} else {
memcpy(scsi_cmd->sense_buffer, ipr_cmd->sense_buffer,
SCSI_SENSE_BUFFERSIZE);
}
if (res) {
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (!ipr_is_naca_model(res))
res->needs_sync_complete = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
res->in_erp = 0;
}
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
scsi_dma_unmap(ipr_cmd->scsi_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi_cmd->scsi_done(scsi_cmd);
scsi: ipr: Fix missed EH wakeup Following a command abort or device reset, ipr's EH handlers wait for the commands getting aborted to get sent back from the adapter prior to returning from the EH handler. This fixes up some cases where the completion handler was not getting called, which would have resulted in the EH thread waiting until it timed out, greatly extending EH time. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:36 -06:00
if (ipr_cmd->eh_comp)
complete(ipr_cmd->eh_comp);
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
/**
* ipr_erp_done - Process completion of ERP for a device
* @ipr_cmd: ipr command struct
*
* This function copies the sense buffer into the scsi_cmd
* struct and pushes the scsi_done function.
*
* Return value:
* nothing
**/
static void ipr_erp_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
unsigned long hrrq_flags;
spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
__ipr_erp_done(ipr_cmd);
spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_reinit_ipr_cmnd_for_erp - Re-initialize a cmnd block to be used for ERP
* @ipr_cmd: ipr command struct
*
* Return value:
* none
**/
static void ipr_reinit_ipr_cmnd_for_erp(struct ipr_cmnd *ipr_cmd)
{
[SCSI] ipr: Faster sg list fetch Improve overall command performance by embedding the scatterlist in the command block used by the adapter. This decreases the overall number of DMAs required for a single command. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:50 -06:00
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
struct ipr_ioasa *ioasa = &ipr_cmd->s.ioasa;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
dma_addr_t dma_addr = ipr_cmd->dma_addr;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
memset(&ioarcb->cmd_pkt, 0, sizeof(struct ipr_cmd_pkt));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->data_transfer_length = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->read_data_transfer_length = 0;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->ioadl_len = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->read_ioadl_len = 0;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasa->hdr.ioasc = 0;
ioasa->hdr.residual_data_len = 0;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ipr_cmd->ioa_cfg->sis64)
ioarcb->u.sis64_addr_data.data_ioadl_addr =
cpu_to_be64(dma_addr + offsetof(struct ipr_cmnd, i.ioadl64));
else {
ioarcb->write_ioadl_addr =
cpu_to_be32(dma_addr + offsetof(struct ipr_cmnd, i.ioadl));
ioarcb->read_ioadl_addr = ioarcb->write_ioadl_addr;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
* __ipr_erp_request_sense - Send request sense to a device
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @ipr_cmd: ipr command struct
*
* This function sends a request sense to a device as a result
* of a check condition.
*
* Return value:
* nothing
**/
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
static void __ipr_erp_request_sense(struct ipr_cmnd *ipr_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_cmd_pkt *cmd_pkt = &ipr_cmd->ioarcb.cmd_pkt;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (IPR_IOASC_SENSE_KEY(ioasc) > 0) {
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
__ipr_erp_done(ipr_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
}
ipr_reinit_ipr_cmnd_for_erp(ipr_cmd);
cmd_pkt->request_type = IPR_RQTYPE_SCSICDB;
cmd_pkt->cdb[0] = REQUEST_SENSE;
cmd_pkt->cdb[4] = SCSI_SENSE_BUFFERSIZE;
cmd_pkt->flags_hi |= IPR_FLAGS_HI_SYNC_OVERRIDE;
cmd_pkt->flags_hi |= IPR_FLAGS_HI_NO_ULEN_CHK;
cmd_pkt->timeout = cpu_to_be16(IPR_REQUEST_SENSE_TIMEOUT / HZ);
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_init_ioadl(ipr_cmd, ipr_cmd->sense_buffer_dma,
SCSI_SENSE_BUFFERSIZE, IPR_IOADL_FLAGS_READ_LAST);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_do_req(ipr_cmd, ipr_erp_done, ipr_timeout,
IPR_REQUEST_SENSE_TIMEOUT * 2);
}
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
/**
* ipr_erp_request_sense - Send request sense to a device
* @ipr_cmd: ipr command struct
*
* This function sends a request sense to a device as a result
* of a check condition.
*
* Return value:
* nothing
**/
static void ipr_erp_request_sense(struct ipr_cmnd *ipr_cmd)
{
struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
unsigned long hrrq_flags;
spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
__ipr_erp_request_sense(ipr_cmd);
spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_erp_cancel_all - Send cancel all to a device
* @ipr_cmd: ipr command struct
*
* This function sends a cancel all to a device to clear the
* queue. If we are running TCQ on the device, QERR is set to 1,
* which means all outstanding ops have been dropped on the floor.
* Cancel all will return them to us.
*
* Return value:
* nothing
**/
static void ipr_erp_cancel_all(struct ipr_cmnd *ipr_cmd)
{
struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
struct ipr_resource_entry *res = scsi_cmd->device->hostdata;
struct ipr_cmd_pkt *cmd_pkt;
res->in_erp = 1;
ipr_reinit_ipr_cmnd_for_erp(ipr_cmd);
scsi: remove scsi_get_tag_type Both remaining users are better of just checking sdev->simple_tags directly. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com Reviewed-by: Hannes Reinecke <hare@suse.de>
2014-11-24 07:36:20 -07:00
if (!scsi_cmd->device->simple_tags) {
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
__ipr_erp_request_sense(ipr_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
}
cmd_pkt = &ipr_cmd->ioarcb.cmd_pkt;
cmd_pkt->request_type = IPR_RQTYPE_IOACMD;
cmd_pkt->cdb[0] = IPR_CANCEL_ALL_REQUESTS;
ipr_do_req(ipr_cmd, ipr_erp_request_sense, ipr_timeout,
IPR_CANCEL_ALL_TIMEOUT);
}
/**
* ipr_dump_ioasa - Dump contents of IOASA
* @ioa_cfg: ioa config struct
* @ipr_cmd: ipr command struct
[SCSI] ipr: Simplify status area dumping Simplify the dumping of the command status area by removing some device specific information that has proven to not be worthwhile. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:29 -07:00
* @res: resource entry struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* This function is invoked by the interrupt handler when ops
* fail. It will log the IOASA if appropriate. Only called
* for GPDD ops.
*
* Return value:
* none
**/
static void ipr_dump_ioasa(struct ipr_ioa_cfg *ioa_cfg,
[SCSI] ipr: Simplify status area dumping Simplify the dumping of the command status area by removing some device specific information that has proven to not be worthwhile. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:29 -07:00
struct ipr_cmnd *ipr_cmd, struct ipr_resource_entry *res)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
int i;
u16 data_len;
[SCSI] ipr: Log error for SAS dual path switch For ipr SAS adapters that support dual pathing, this patch modifies ipr to log an error when a path fails. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:09 -06:00
u32 ioasc, fd_ioasc;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
struct ipr_ioasa *ioasa = &ipr_cmd->s.ioasa;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
__be32 *ioasa_data = (__be32 *)ioasa;
int error_index;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasc = be32_to_cpu(ioasa->hdr.ioasc) & IPR_IOASC_IOASC_MASK;
fd_ioasc = be32_to_cpu(ioasa->hdr.fd_ioasc) & IPR_IOASC_IOASC_MASK;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (0 == ioasc)
return;
if (ioa_cfg->log_level < IPR_DEFAULT_LOG_LEVEL)
return;
[SCSI] ipr: Log error for SAS dual path switch For ipr SAS adapters that support dual pathing, this patch modifies ipr to log an error when a path fails. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:09 -06:00
if (ioasc == IPR_IOASC_BUS_WAS_RESET && fd_ioasc)
error_index = ipr_get_error(fd_ioasc);
else
error_index = ipr_get_error(ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioa_cfg->log_level < IPR_MAX_LOG_LEVEL) {
/* Don't log an error if the IOA already logged one */
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
if (ioasa->hdr.ilid != 0)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
[SCSI] ipr: Enable logging of debug error data for all devices The ipr driver has a sysfs attribute which can be used to adjust the logging level of the driver for error events. The error response data for commands can be dumped by increasing the logging level of the ipr driver. This currently only works for JBOD passthrough devices. This patch enables this function for all devices, including RAID devices, to aid in debugging problems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:01 -06:00
if (!ipr_is_gscsi(res))
return;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ipr_error_table[error_index].log_ioasa == 0)
return;
}
[SCSI] ipr: Simplify status area dumping Simplify the dumping of the command status area by removing some device specific information that has proven to not be worthwhile. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:29 -07:00
ipr_res_err(ioa_cfg, res, "%s\n", ipr_error_table[error_index].error);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
data_len = be16_to_cpu(ioasa->hdr.ret_stat_len);
if (ioa_cfg->sis64 && sizeof(struct ipr_ioasa64) < data_len)
data_len = sizeof(struct ipr_ioasa64);
else if (!ioa_cfg->sis64 && sizeof(struct ipr_ioasa) < data_len)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
data_len = sizeof(struct ipr_ioasa);
ipr_err("IOASA Dump:\n");
for (i = 0; i < data_len / 4; i += 4) {
ipr_err("%08X: %08X %08X %08X %08X\n", i*4,
be32_to_cpu(ioasa_data[i]),
be32_to_cpu(ioasa_data[i+1]),
be32_to_cpu(ioasa_data[i+2]),
be32_to_cpu(ioasa_data[i+3]));
}
}
/**
* ipr_gen_sense - Generate SCSI sense data from an IOASA
* @ioasa: IOASA
* @sense_buf: sense data buffer
*
* Return value:
* none
**/
static void ipr_gen_sense(struct ipr_cmnd *ipr_cmd)
{
u32 failing_lba;
u8 *sense_buf = ipr_cmd->scsi_cmd->sense_buffer;
struct ipr_resource_entry *res = ipr_cmd->scsi_cmd->device->hostdata;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
struct ipr_ioasa *ioasa = &ipr_cmd->s.ioasa;
u32 ioasc = be32_to_cpu(ioasa->hdr.ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
memset(sense_buf, 0, SCSI_SENSE_BUFFERSIZE);
if (ioasc >= IPR_FIRST_DRIVER_IOASC)
return;
ipr_cmd->scsi_cmd->result = SAM_STAT_CHECK_CONDITION;
if (ipr_is_vset_device(res) &&
ioasc == IPR_IOASC_MED_DO_NOT_REALLOC &&
ioasa->u.vset.failing_lba_hi != 0) {
sense_buf[0] = 0x72;
sense_buf[1] = IPR_IOASC_SENSE_KEY(ioasc);
sense_buf[2] = IPR_IOASC_SENSE_CODE(ioasc);
sense_buf[3] = IPR_IOASC_SENSE_QUAL(ioasc);
sense_buf[7] = 12;
sense_buf[8] = 0;
sense_buf[9] = 0x0A;
sense_buf[10] = 0x80;
failing_lba = be32_to_cpu(ioasa->u.vset.failing_lba_hi);
sense_buf[12] = (failing_lba & 0xff000000) >> 24;
sense_buf[13] = (failing_lba & 0x00ff0000) >> 16;
sense_buf[14] = (failing_lba & 0x0000ff00) >> 8;
sense_buf[15] = failing_lba & 0x000000ff;
failing_lba = be32_to_cpu(ioasa->u.vset.failing_lba_lo);
sense_buf[16] = (failing_lba & 0xff000000) >> 24;
sense_buf[17] = (failing_lba & 0x00ff0000) >> 16;
sense_buf[18] = (failing_lba & 0x0000ff00) >> 8;
sense_buf[19] = failing_lba & 0x000000ff;
} else {
sense_buf[0] = 0x70;
sense_buf[2] = IPR_IOASC_SENSE_KEY(ioasc);
sense_buf[12] = IPR_IOASC_SENSE_CODE(ioasc);
sense_buf[13] = IPR_IOASC_SENSE_QUAL(ioasc);
/* Illegal request */
if ((IPR_IOASC_SENSE_KEY(ioasc) == 0x05) &&
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
(be32_to_cpu(ioasa->hdr.ioasc_specific) & IPR_FIELD_POINTER_VALID)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sense_buf[7] = 10; /* additional length */
/* IOARCB was in error */
if (IPR_IOASC_SENSE_CODE(ioasc) == 0x24)
sense_buf[15] = 0xC0;
else /* Parameter data was invalid */
sense_buf[15] = 0x80;
sense_buf[16] =
((IPR_FIELD_POINTER_MASK &
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
be32_to_cpu(ioasa->hdr.ioasc_specific)) >> 8) & 0xff;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sense_buf[17] =
(IPR_FIELD_POINTER_MASK &
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
be32_to_cpu(ioasa->hdr.ioasc_specific)) & 0xff;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
} else {
if (ioasc == IPR_IOASC_MED_DO_NOT_REALLOC) {
if (ipr_is_vset_device(res))
failing_lba = be32_to_cpu(ioasa->u.vset.failing_lba_lo);
else
failing_lba = be32_to_cpu(ioasa->u.dasd.failing_lba);
sense_buf[0] |= 0x80; /* Or in the Valid bit */
sense_buf[3] = (failing_lba & 0xff000000) >> 24;
sense_buf[4] = (failing_lba & 0x00ff0000) >> 16;
sense_buf[5] = (failing_lba & 0x0000ff00) >> 8;
sense_buf[6] = failing_lba & 0x000000ff;
}
sense_buf[7] = 6; /* additional length */
}
}
}
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
/**
* ipr_get_autosense - Copy autosense data to sense buffer
* @ipr_cmd: ipr command struct
*
* This function copies the autosense buffer to the buffer
* in the scsi_cmd, if there is autosense available.
*
* Return value:
* 1 if autosense was available / 0 if not
**/
static int ipr_get_autosense(struct ipr_cmnd *ipr_cmd)
{
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
struct ipr_ioasa *ioasa = &ipr_cmd->s.ioasa;
struct ipr_ioasa64 *ioasa64 = &ipr_cmd->s.ioasa64;
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
if ((be32_to_cpu(ioasa->hdr.ioasc_specific) & IPR_AUTOSENSE_VALID) == 0)
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
return 0;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
if (ipr_cmd->ioa_cfg->sis64)
memcpy(ipr_cmd->scsi_cmd->sense_buffer, ioasa64->auto_sense.data,
min_t(u16, be16_to_cpu(ioasa64->auto_sense.auto_sense_len),
SCSI_SENSE_BUFFERSIZE));
else
memcpy(ipr_cmd->scsi_cmd->sense_buffer, ioasa->auto_sense.data,
min_t(u16, be16_to_cpu(ioasa->auto_sense.auto_sense_len),
SCSI_SENSE_BUFFERSIZE));
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
return 1;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_erp_start - Process an error response for a SCSI op
* @ioa_cfg: ioa config struct
* @ipr_cmd: ipr command struct
*
* This function determines whether or not to initiate ERP
* on the affected device.
*
* Return value:
* nothing
**/
static void ipr_erp_start(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_cmnd *ipr_cmd)
{
struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
struct ipr_resource_entry *res = scsi_cmd->device->hostdata;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
[SCSI] ipr: Handle check condition status from disk array device On newer levels of microcode for ipr RAID adapters supporting multi-initiator configurations, the disk array, or VSET, resources are capable of generating a check condition. This patch prevents ipr from generating sense data in this scenario and retrieving it from the logical device instead. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:10 -06:00
u32 masked_ioasc = ioasc & IPR_IOASC_IOASC_MASK;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!res) {
scsi: ipr: Error path locking fixes This patch closes up some potential race conditions observed in the error handling paths in ipr while debugging an issue resulting in a hang with SATA error handling. These patches ensure we are holding the correct lock when adding and removing commands from the free and pending queues in some error scenarios. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:39 -06:00
__ipr_scsi_eh_done(ipr_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
}
[SCSI] ipr: Handle check condition status from disk array device On newer levels of microcode for ipr RAID adapters supporting multi-initiator configurations, the disk array, or VSET, resources are capable of generating a check condition. This patch prevents ipr from generating sense data in this scenario and retrieving it from the logical device instead. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:10 -06:00
if (!ipr_is_gscsi(res) && masked_ioasc != IPR_IOASC_HW_DEV_BUS_STATUS)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_gen_sense(ipr_cmd);
[SCSI] ipr: Enable logging of debug error data for all devices The ipr driver has a sysfs attribute which can be used to adjust the logging level of the driver for error events. The error response data for commands can be dumped by increasing the logging level of the ipr driver. This currently only works for JBOD passthrough devices. This patch enables this function for all devices, including RAID devices, to aid in debugging problems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:01 -06:00
ipr_dump_ioasa(ioa_cfg, ipr_cmd, res);
[SCSI] ipr: Handle check condition status from disk array device On newer levels of microcode for ipr RAID adapters supporting multi-initiator configurations, the disk array, or VSET, resources are capable of generating a check condition. This patch prevents ipr from generating sense data in this scenario and retrieving it from the logical device instead. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:10 -06:00
switch (masked_ioasc) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
case IPR_IOASC_ABORTED_CMD_TERM_BY_HOST:
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (ipr_is_naca_model(res))
scsi_cmd->result |= (DID_ABORT << 16);
else
scsi_cmd->result |= (DID_IMM_RETRY << 16);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
break;
case IPR_IOASC_IR_RESOURCE_HANDLE:
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
case IPR_IOASC_IR_NO_CMDS_TO_2ND_IOA:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi_cmd->result |= (DID_NO_CONNECT << 16);
break;
case IPR_IOASC_HW_SEL_TIMEOUT:
scsi_cmd->result |= (DID_NO_CONNECT << 16);
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (!ipr_is_naca_model(res))
res->needs_sync_complete = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
break;
case IPR_IOASC_SYNC_REQUIRED:
if (!res->in_erp)
res->needs_sync_complete = 1;
scsi_cmd->result |= (DID_IMM_RETRY << 16);
break;
case IPR_IOASC_MED_DO_NOT_REALLOC: /* prevent retries */
[SCSI] ipr: handle new adapter errors Add support for handling some new errors that may be returned by ipr adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:47 -07:00
case IPR_IOASA_IR_DUAL_IOA_DISABLED:
scsi: ipr: do not set DID_PASSTHROUGH on CHECK CONDITION On a dual controller setup with multipath enabled, some MEDIUM ERRORs caused both paths to be failed, thus I/O got queued/blocked since the 'queue_if_no_path' feature is enabled by default on IPR controllers. This example disabled 'queue_if_no_path' so the I/O failure is seen at the sg_dd program. Notice that after the sg_dd test-case, both paths are in 'failed' state, and both path/priority groups are in 'enabled' state (not 'active') -- which would block I/O with 'queue_if_no_path'. # sg_dd if=/dev/dm-2 bs=4096 count=1 dio=1 verbose=4 blk_sgio=0 <...> read(unix): count=4096, res=-1 sg_dd: reading, skip=0 : Input/output error <...> # dmesg [...] sd 2:2:16:0: [sds] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE [...] sd 2:2:16:0: [sds] Sense Key : Medium Error [current] [...] sd 2:2:16:0: [sds] Add. Sense: Unrecovered read error - recommend rewrite the data [...] sd 2:2:16:0: [sds] CDB: Read(10) 28 00 00 00 00 00 00 00 20 00 [...] blk_update_request: I/O error, dev sds, sector 0 [...] device-mapper: multipath: Failing path 65:32. <...> [...] device-mapper: multipath: Failing path 65:224. # multipath -l 1IBM_IPR-0_59C2AE0000001F80 dm-2 IBM ,IPR-0 59C2AE00 size=5.2T features='0' hwhandler='1 alua' wp=rw |-+- policy='service-time 0' prio=0 status=enabled | `- 2:2:16:0 sds 65:32 failed undef running `-+- policy='service-time 0' prio=0 status=enabled `- 1:2:7:0 sdae 65:224 failed undef running This is not the desired behavior. The dm-multipath explicitly checks for the MEDIUM ERROR case (and a few others) so not to fail the path (e.g., I/O to other sectors could potentially happen without problems). See dm-mpath.c :: do_end_io_bio() -> noretry_error() !->! fail_path(). The problem trace is: 1) ipr_scsi_done() // SENSE KEY/CHECK CONDITION detected, go to.. 2) ipr_erp_start() // ipr_is_gscsi() and masked_ioasc OK, go to.. 3) ipr_gen_sense() // masked_ioasc is IPR_IOASC_MED_DO_NOT_REALLOC, // so set DID_PASSTHROUGH. 4) scsi_decide_disposition() // check for DID_PASSTHROUGH and return // early on, faking a DID_OK.. *instead* // of reaching scsi_check_sense(). // Had it reached the latter, that would // set host_byte to DID_MEDIUM_ERROR. 5) scsi_finish_command() 6) scsi_io_completion() 7) __scsi_error_from_host_byte() // That would be converted to -ENODATA <...> 8) dm_softirq_done() 9) multipath_end_io() 10) do_end_io() 11) noretry_error() // And that is checked in dm-mpath :: noretry_error() // which would cause fail_path() not to be called. With this patch applied, the I/O is failed but the paths are not. This multipath device continues accepting more I/O requests without blocking. (and notice the different host byte/driver byte handling per SCSI layer). # dmesg [...] sd 2:2:7:0: [sdaf] Done: SUCCESS Result: hostbyte=0x13 driverbyte=DRIVER_OK [...] sd 2:2:7:0: [sdaf] CDB: Read(10) 28 00 00 00 00 00 00 00 40 00 [...] sd 2:2:7:0: [sdaf] Sense Key : Medium Error [current] [...] sd 2:2:7:0: [sdaf] Add. Sense: Unrecovered read error - recommend rewrite the data [...] blk_update_request: critical medium error, dev sdaf, sector 0 [...] blk_update_request: critical medium error, dev dm-6, sector 0 [...] sd 2:2:7:0: [sdaf] Done: SUCCESS Result: hostbyte=0x13 driverbyte=DRIVER_OK [...] sd 2:2:7:0: [sdaf] CDB: Read(10) 28 00 00 00 00 00 00 00 10 00 [...] sd 2:2:7:0: [sdaf] Sense Key : Medium Error [current] [...] sd 2:2:7:0: [sdaf] Add. Sense: Unrecovered read error - recommend rewrite the data [...] blk_update_request: critical medium error, dev sdaf, sector 0 [...] blk_update_request: critical medium error, dev dm-6, sector 0 [...] Buffer I/O error on dev dm-6, logical block 0, async page read # multipath -l 1IBM_IPR-0_59C2AE0000001F80 1IBM_IPR-0_59C2AE0000001F80 dm-6 IBM ,IPR-0 59C2AE00 size=5.2T features='1 queue_if_no_path' hwhandler='1 alua' wp=rw |-+- policy='service-time 0' prio=0 status=active | `- 2:2:7:0 sdaf 65:240 active undef running `-+- policy='service-time 0' prio=0 status=enabled `- 1:2:7:0 sdh 8:112 active undef running Signed-off-by: Mauricio Faria de Oliveira <mauricfo@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-04-11 08:46:04 -06:00
/*
* exception: do not set DID_PASSTHROUGH on CHECK CONDITION
* so SCSI mid-layer and upper layers handle it accordingly.
*/
if (scsi_cmd->result != SAM_STAT_CHECK_CONDITION)
scsi_cmd->result |= (DID_PASSTHROUGH << 16);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
break;
case IPR_IOASC_BUS_WAS_RESET:
case IPR_IOASC_BUS_WAS_RESET_BY_OTHER:
/*
* Report the bus reset and ask for a retry. The device
* will give CC/UA the next command.
*/
if (!res->resetting_device)
scsi_report_bus_reset(ioa_cfg->host, scsi_cmd->device->channel);
scsi_cmd->result |= (DID_ERROR << 16);
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (!ipr_is_naca_model(res))
res->needs_sync_complete = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
break;
case IPR_IOASC_HW_DEV_BUS_STATUS:
scsi_cmd->result |= IPR_IOASC_SENSE_STATUS(ioasc);
if (IPR_IOASC_SENSE_STATUS(ioasc) == SAM_STAT_CHECK_CONDITION) {
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (!ipr_get_autosense(ipr_cmd)) {
if (!ipr_is_naca_model(res)) {
ipr_erp_cancel_all(ipr_cmd);
return;
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (!ipr_is_naca_model(res))
res->needs_sync_complete = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
break;
case IPR_IOASC_NR_INIT_CMD_REQUIRED:
break;
ipr: AF DASD raw mode implementation in ipr driver This patch implements raw mode support for AF DASD in ipr driver which allows for tools to send commands directly to physical devices which are members of RAID arrays when enabled in the firmware. [jejb: fix up whitespace] Signed-off-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:58 -06:00
case IPR_IOASC_IR_NON_OPTIMIZED:
if (res->raw_mode) {
res->raw_mode = 0;
scsi_cmd->result |= (DID_IMM_RETRY << 16);
} else
scsi_cmd->result |= (DID_ERROR << 16);
break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
default:
[SCSI] ipr: Properly handle IOA recovered errors The ipr driver currently translates adapter recovered errors to DID_ERROR. This patch fixes this to translate these errors to success instead. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-08-02 13:57:51 -06:00
if (IPR_IOASC_SENSE_KEY(ioasc) > RECOVERED_ERROR)
scsi_cmd->result |= (DID_ERROR << 16);
[SCSI] ipr: Support new device queueing model New ipr adapters support a new device queueing model in the adapter firmware. The queueing model is the NACA queueing model, but it does not mean use of NACA is required. The new model removes some of the adapter firmware queue state that made handling QERR=0 almost impossible. The queueing model on older adapters included the concept of a queue frozen state, which would freeze the response queue in the adapter when a check condition occurred, requiring a a primitive to resume the queue. The new queueing model removes this complexity. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:22 -07:00
if (!ipr_is_vset_device(res) && !ipr_is_naca_model(res))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
res->needs_sync_complete = 1;
break;
}
[SCSI] ipr: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-25 09:26:59 -06:00
scsi_dma_unmap(ipr_cmd->scsi_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi_cmd->scsi_done(scsi_cmd);
scsi: ipr: Fix missed EH wakeup Following a command abort or device reset, ipr's EH handlers wait for the commands getting aborted to get sent back from the adapter prior to returning from the EH handler. This fixes up some cases where the completion handler was not getting called, which would have resulted in the EH thread waiting until it timed out, greatly extending EH time. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:36 -06:00
if (ipr_cmd->eh_comp)
complete(ipr_cmd->eh_comp);
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_scsi_done - mid-layer done function
* @ipr_cmd: ipr command struct
*
* This function is invoked by the interrupt handler for
* ops generated by the SCSI mid-layer
*
* Return value:
* none
**/
static void ipr_scsi_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
ipr: Fix locking for unit attention handling Make sure we have the host lock held when calling scsi_report_bus_reset. Fixes a crash seen as the __devices list in the scsi host was changing as we were iterating through it. Cc: <stable@vger.kernel.org> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:29 -06:00
unsigned long lock_flags;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
scsi_set_resid(scsi_cmd, be32_to_cpu(ipr_cmd->s.ioasa.hdr.residual_data_len));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (likely(IPR_IOASC_SENSE_KEY(ioasc) == 0)) {
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
scsi_dma_unmap(scsi_cmd);
ipr: Fix locking for unit attention handling Make sure we have the host lock held when calling scsi_report_bus_reset. Fixes a crash seen as the __devices list in the scsi host was changing as we were iterating through it. Cc: <stable@vger.kernel.org> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:29 -06:00
spin_lock_irqsave(ipr_cmd->hrrq->lock, lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi_cmd->scsi_done(scsi_cmd);
scsi: ipr: Fix missed EH wakeup Following a command abort or device reset, ipr's EH handlers wait for the commands getting aborted to get sent back from the adapter prior to returning from the EH handler. This fixes up some cases where the completion handler was not getting called, which would have resulted in the EH thread waiting until it timed out, greatly extending EH time. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Tested-by: Wendy Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-03-15 15:58:36 -06:00
if (ipr_cmd->eh_comp)
complete(ipr_cmd->eh_comp);
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
ipr: Fix locking for unit attention handling Make sure we have the host lock held when calling scsi_report_bus_reset. Fixes a crash seen as the __devices list in the scsi host was changing as we were iterating through it. Cc: <stable@vger.kernel.org> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:29 -06:00
spin_unlock_irqrestore(ipr_cmd->hrrq->lock, lock_flags);
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
} else {
ipr: Fix locking for unit attention handling Make sure we have the host lock held when calling scsi_report_bus_reset. Fixes a crash seen as the __devices list in the scsi host was changing as we were iterating through it. Cc: <stable@vger.kernel.org> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:29 -06:00
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
spin_lock(&ipr_cmd->hrrq->_lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_erp_start(ioa_cfg, ipr_cmd);
ipr: Fix locking for unit attention handling Make sure we have the host lock held when calling scsi_report_bus_reset. Fixes a crash seen as the __devices list in the scsi host was changing as we were iterating through it. Cc: <stable@vger.kernel.org> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-07-14 10:41:29 -06:00
spin_unlock(&ipr_cmd->hrrq->_lock);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_queuecommand - Queue a mid-layer request
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
* @shost: scsi host struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @scsi_cmd: scsi command struct
*
* This function queues a request generated by the mid-layer.
*
* Return value:
* 0 on success
* SCSI_MLQUEUE_DEVICE_BUSY if device is busy
* SCSI_MLQUEUE_HOST_BUSY if host is busy
**/
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
static int ipr_queuecommand(struct Scsi_Host *shost,
struct scsi_cmnd *scsi_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioa_cfg *ioa_cfg;
struct ipr_resource_entry *res;
struct ipr_ioarcb *ioarcb;
struct ipr_cmnd *ipr_cmd;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
unsigned long hrrq_flags, lock_flags;
[SCSI] ipr: remove an unneeded check "rc" is always zero here, so there is no need to check. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-30 02:18:22 -06:00
int rc;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq;
int hrrq_id;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi_cmd->result = (DID_OK << 16);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
res = scsi_cmd->device->hostdata;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ipr_is_gata(res) && res->sata_port) {
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
rc = ata_sas_queuecmd(scsi_cmd, res->sata_port->ap);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return rc;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
hrrq_id = ipr_get_hrrq_index(ioa_cfg);
hrrq = &ioa_cfg->hrrq[hrrq_id];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock_irqsave(hrrq->lock, hrrq_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/*
* We are currently blocking all devices due to a host reset
* We have told the host to stop giving us new requests, but
* ERP ops don't count. FIXME
*/
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
if (unlikely(!hrrq->allow_cmds && !hrrq->ioa_is_dead && !hrrq->removing_ioa)) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return SCSI_MLQUEUE_HOST_BUSY;
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/*
* FIXME - Create scsi_set_host_offline interface
* and the ioa_is_dead check can be removed
*/
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
if (unlikely(hrrq->ioa_is_dead || hrrq->removing_ioa || !res)) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
goto err_nodev;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_cmd = __ipr_get_free_ipr_cmnd(hrrq);
if (ipr_cmd == NULL) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
return SCSI_MLQUEUE_HOST_BUSY;
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
ipr_init_ipr_cmnd(ipr_cmd, ipr_scsi_done);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb = &ipr_cmd->ioarcb;
memcpy(ioarcb->cmd_pkt.cdb, scsi_cmd->cmnd, scsi_cmd->cmd_len);
ipr_cmd->scsi_cmd = scsi_cmd;
[SCSI] ipr: Reduce interrupt lock time Reduce the amount of time the host lock is held in the interrupt handler for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:14:40 -06:00
ipr_cmd->done = ipr_scsi_eh_done;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: Don't set NO_ULEN_CHK bit when resource is a vset. According to the IPR specification, Inhibit Underlength Checking bit must be disabled when issuing commands to vsets. Enabling it in this case might cause SCSI commands to fail with an Illegal Request, so make sure we keep this bit cleared when resource is a vset. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:07 -07:00
if (ipr_is_gscsi(res)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (scsi_cmd->underflow == 0)
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_ULEN_CHK;
ipr: Don't set NO_ULEN_CHK bit when resource is a vset. According to the IPR specification, Inhibit Underlength Checking bit must be disabled when issuing commands to vsets. Enabling it in this case might cause SCSI commands to fail with an Illegal Request, so make sure we keep this bit cleared when resource is a vset. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:07 -07:00
if (res->reset_occurred) {
[SCSI] ipr: Remove extended delay bit on GSCSI reads/writes ops This patch removes extended delay bit on GSCSI reads/writes ops, the performance will be significanly better. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:39 -07:00
res->reset_occurred = 0;
[SCSI] ipr: fix synchronous request flags for better performance In testing it was noticed that Extended Delay after Reset flag was being set for gscsi and volume set devices. This had a negative effect on performance for volume sets. The fix is to only set the flag for gscsi devices. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-03-31 10:56:10 -06:00
ioarcb->cmd_pkt.flags_lo |= IPR_FLAGS_LO_DELAY_AFTER_RST;
[SCSI] ipr: Remove extended delay bit on GSCSI reads/writes ops This patch removes extended delay bit on GSCSI reads/writes ops, the performance will be significanly better. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:39 -07:00
}
ipr: Don't set NO_ULEN_CHK bit when resource is a vset. According to the IPR specification, Inhibit Underlength Checking bit must be disabled when issuing commands to vsets. Enabling it in this case might cause SCSI commands to fail with an Illegal Request, so make sure we keep this bit cleared when resource is a vset. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:07 -07:00
}
if (ipr_is_gscsi(res) || ipr_is_vset_device(res)) {
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_LINK_DESC;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->cmd_pkt.flags_lo |= IPR_FLAGS_LO_ALIGNED_BFR;
scsi: remove abuses of scsi_populate_tag Unless we want to build a SPI tag message we should just check SCMD_TAGGED instead of reverse engineering a tag type through the use of scsi_populate_tag_msg. Also rename the function to spi_populate_tag_msg, make it behave like the other spi message helpers, and move it to the spi transport class. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Reviewed-by: Hannes Reinecke <hare@suse.de>
2014-10-30 07:30:06 -06:00
if (scsi_cmd->flags & SCMD_TAGGED)
ioarcb->cmd_pkt.flags_lo |= IPR_FLAGS_LO_SIMPLE_TASK;
else
ioarcb->cmd_pkt.flags_lo |= IPR_FLAGS_LO_UNTAGGED_TASK;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
if (scsi_cmd->cmnd[0] >= 0xC0 &&
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
(!ipr_is_gscsi(res) || scsi_cmd->cmnd[0] == IPR_QUERY_RSRC_STATE)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
ipr: Inhibit underlength data check for AFDASD in raw mode. Disable underlength error verification based on count of bytes actually transferred for AF DASD devices when SIS pipe mode is enabled. This avoids unexpected underlength errors when issuing some commands in raw mode. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-08-19 08:47:05 -06:00
if (res->raw_mode && ipr_is_af_dasd_device(res)) {
ipr: AF DASD raw mode implementation in ipr driver This patch implements raw mode support for AF DASD in ipr driver which allows for tools to send commands directly to physical devices which are members of RAID arrays when enabled in the firmware. [jejb: fix up whitespace] Signed-off-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:58 -06:00
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_PIPE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: Inhibit underlength data check for AFDASD in raw mode. Disable underlength error verification based on count of bytes actually transferred for AF DASD devices when SIS pipe mode is enabled. This avoids unexpected underlength errors when issuing some commands in raw mode. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-08-19 08:47:05 -06:00
if (scsi_cmd->underflow == 0)
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_ULEN_CHK;
}
[SCSI] ipr: remove an unneeded check "rc" is always zero here, so there is no need to check. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-30 02:18:22 -06:00
if (ioa_cfg->sis64)
rc = ipr_build_ioadl64(ioa_cfg, ipr_cmd);
else
rc = ipr_build_ioadl(ioa_cfg, ipr_cmd);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock_irqsave(hrrq->lock, hrrq_flags);
if (unlikely(rc || (!hrrq->allow_cmds && !hrrq->ioa_is_dead))) {
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &hrrq->hrrq_free_q);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
if (!rc)
scsi_dma_unmap(scsi_cmd);
[SCSI] ipr: Remove unnecessary memory barriers The ipr driver added some memory barriers in order to ensure a PowerPC sync instruction was executed prior to sending a command to the adapter to ensure the command block was coherent with respect to the PCI bus's view of memory. However, some time ago, the powerpc architecture writel macros were changed to include the sync since most drivers don't properly handle this. So remove these memory barriers since they are not needed and result in executing twice as many sync instructions, which has a significant performance penalty. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:09 -06:00
return SCSI_MLQUEUE_HOST_BUSY;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (unlikely(hrrq->ioa_is_dead)) {
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &hrrq->hrrq_free_q);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
scsi_dma_unmap(scsi_cmd);
goto err_nodev;
}
ioarcb->res_handle = res->res_handle;
if (res->needs_sync_complete) {
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_SYNC_COMPLETE;
res->needs_sync_complete = 0;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &hrrq->hrrq_pending_q);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
ipr_trc_hook(ipr_cmd, IPR_TRACE_START, IPR_GET_RES_PHYS_LOC(res));
[SCSI] ipr: Remove unnecessary memory barriers The ipr driver added some memory barriers in order to ensure a PowerPC sync instruction was executed prior to sending a command to the adapter to ensure the command block was coherent with respect to the PCI bus's view of memory. However, some time ago, the powerpc architecture writel macros were changed to include the sync since most drivers don't properly handle this. So remove these memory barriers since they are not needed and result in executing twice as many sync instructions, which has a significant performance penalty. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:09 -06:00
ipr_send_command(ipr_cmd);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return 0;
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
err_nodev:
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock_irqsave(hrrq->lock, hrrq_flags);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
memset(scsi_cmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE);
scsi_cmd->result = (DID_NO_CONNECT << 16);
scsi_cmd->scsi_done(scsi_cmd);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(hrrq->lock, hrrq_flags);
[SCSI] ipr: Reduce queuecommand lock time Reduce the amount of time the host lock is held in queuecommand for improved performance. [jejb: fix up checkpatch noise] Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-17 07:13:52 -06:00
return 0;
}
SCSI host lock push-down Move the mid-layer's ->queuecommand() invocation from being locked with the host lock to being unlocked to facilitate speeding up the critical path for drivers who don't need this lock taken anyway. The patch below presents a simple SCSI host lock push-down as an equivalent transformation. No locking or other behavior should change with this patch. All existing bugs and locking orders are preserved. Additionally, add one parameter to queuecommand, struct Scsi_Host * and remove one parameter from queuecommand, void (*done)(struct scsi_cmnd *) Scsi_Host* is a convenient pointer that most host drivers need anyway, and 'done' is redundant to struct scsi_cmnd->scsi_done. Minimal code disturbance was attempted with this change. Most drivers needed only two one-line modifications for their host lock push-down. Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Acked-by: James Bottomley <James.Bottomley@suse.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-11-16 00:10:29 -07:00
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
* ipr_ioctl - IOCTL handler
* @sdev: scsi device struct
* @cmd: IOCTL cmd
* @arg: IOCTL arg
*
* Return value:
* 0 on success / other on failure
**/
scsi: ata: Use unsigned int for cmd's type in ioctls in scsi_host_template Clang warns several times in the scsi subsystem (trimmed for brevity): drivers/scsi/hpsa.c:6209:7: warning: overflow converting case value to switch condition type (2147762695 to 18446744071562347015) [-Wswitch] case CCISS_GETBUSTYPES: ^ drivers/scsi/hpsa.c:6208:7: warning: overflow converting case value to switch condition type (2147762694 to 18446744071562347014) [-Wswitch] case CCISS_GETHEARTBEAT: ^ The root cause is that the _IOC macro can generate really large numbers, which don't fit into type 'int', which is used for the cmd parameter in the ioctls in scsi_host_template. My research into how GCC and Clang are handling this at a low level didn't prove fruitful. However, looking at the rest of the kernel tree, all ioctls use an 'unsigned int' for the cmd parameter, which will fit all of the _IOC values in the scsi/ata subsystems. Make that change because none of the ioctls expect a negative value for any command, it brings the ioctls inline with the reset of the kernel, and it removes ambiguity, which is never good when dealing with compilers. Link: https://github.com/ClangBuiltLinux/linux/issues/85 Link: https://github.com/ClangBuiltLinux/linux/issues/154 Link: https://github.com/ClangBuiltLinux/linux/issues/157 Signed-off-by: Nathan Chancellor <natechancellor@gmail.com> Acked-by: Bradley Grove <bgrove@attotech.com> Acked-by: Don Brace <don.brace@microsemi.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Tested-by: Nick Desaulniers <ndesaulniers@google.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2019-02-07 09:07:20 -07:00
static int ipr_ioctl(struct scsi_device *sdev, unsigned int cmd,
void __user *arg)
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
{
struct ipr_resource_entry *res;
res = (struct ipr_resource_entry *)sdev->hostdata;
[SCSI] ipr: Fix HDIO_GET_IDENTITY oops for SATA devices Currently, ipr does not support HDIO_GET_IDENTITY to SATA devices. An oops occurs if userspace attempts to send the command. Since hald issues the command, ensure we fail the ioctl in ipr. This is a temporary solution to the oops. Once the ipr libata EH conversion is upstream, ipr will fully support HDIO_GET_IDENTITY. Tested-by: Milton Miller <miltonm@bga.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-11 12:37:50 -06:00
if (res && ipr_is_gata(res)) {
if (cmd == HDIO_GET_IDENTITY)
return -ENOTTY;
[libata] get-identity ioctl: Fix use of invalid memory pointer for SAS drivers. Caught by Ke Wei (and team?) at Marvell. Also, move the ata_scsi_ioctl export to libata-scsi.c, as that seems to be the general trend. Acked-by: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2009-01-16 08:17:09 -07:00
return ata_sas_scsi_ioctl(res->sata_port->ap, sdev, cmd, arg);
[SCSI] ipr: Fix HDIO_GET_IDENTITY oops for SATA devices Currently, ipr does not support HDIO_GET_IDENTITY to SATA devices. An oops occurs if userspace attempts to send the command. Since hald issues the command, ensure we fail the ioctl in ipr. This is a temporary solution to the oops. Once the ipr libata EH conversion is upstream, ipr will fully support HDIO_GET_IDENTITY. Tested-by: Milton Miller <miltonm@bga.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-11 12:37:50 -06:00
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
return -EINVAL;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_info - Get information about the card/driver
* @scsi_host: scsi host struct
*
* Return value:
* pointer to buffer with description string
**/
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
static const char *ipr_ioa_info(struct Scsi_Host *host)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
static char buffer[512];
struct ipr_ioa_cfg *ioa_cfg;
unsigned long lock_flags = 0;
ioa_cfg = (struct ipr_ioa_cfg *) host->hostdata;
spin_lock_irqsave(host->host_lock, lock_flags);
sprintf(buffer, "IBM %X Storage Adapter", ioa_cfg->type);
spin_unlock_irqrestore(host->host_lock, lock_flags);
return buffer;
}
static struct scsi_host_template driver_template = {
.module = THIS_MODULE,
.name = "IPR",
.info = ipr_ioa_info,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
.ioctl = ipr_ioctl,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.queuecommand = ipr_queuecommand,
.eh_abort_handler = ipr_eh_abort,
.eh_device_reset_handler = ipr_eh_dev_reset,
.eh_host_reset_handler = ipr_eh_host_reset,
.slave_alloc = ipr_slave_alloc,
.slave_configure = ipr_slave_configure,
.slave_destroy = ipr_slave_destroy,
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
.scan_finished = ipr_scan_finished,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
.target_alloc = ipr_target_alloc,
.target_destroy = ipr_target_destroy,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
.change_queue_depth = ipr_change_queue_depth,
.bios_param = ipr_biosparam,
.can_queue = IPR_MAX_COMMANDS,
.this_id = -1,
.sg_tablesize = IPR_MAX_SGLIST,
.max_sectors = IPR_IOA_MAX_SECTORS,
.cmd_per_lun = IPR_MAX_CMD_PER_LUN,
.shost_attrs = ipr_ioa_attrs,
.sdev_attrs = ipr_dev_attrs,
[SCSI] Disable WRITE SAME for RAID and virtual host adapter drivers Some host adapters do not pass commands through to the target disk directly. Instead they provide an emulated target which may or may not accurately report its capabilities. In some cases the physical device characteristics are reported even when the host adapter is processing commands on the device's behalf. This can lead to adapter firmware hangs or excessive I/O errors. This patch disables WRITE SAME for devices connected to host adapters that provide an emulated target. Driver writers can disable WRITE SAME by setting the no_write_same flag in the host adapter template. [jejb: fix up rejections due to eh_deadline patch] Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Cc: stable@kernel.org Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-10-23 04:25:40 -06:00
.proc_name = IPR_NAME,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
};
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
* ipr_ata_phy_reset - libata phy_reset handler
* @ap: ata port to reset
*
**/
static void ipr_ata_phy_reset(struct ata_port *ap)
{
unsigned long flags;
struct ipr_sata_port *sata_port = ap->private_data;
struct ipr_resource_entry *res = sata_port->res;
struct ipr_ioa_cfg *ioa_cfg = sata_port->ioa_cfg;
int rc;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
while (ioa_cfg->in_reset_reload) {
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (!ioa_cfg->hrrq[IPR_INIT_HRRQ].allow_cmds)
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
goto out_unlock;
rc = ipr_device_reset(ioa_cfg, res);
if (rc) {
libata: kill ATA_FLAG_DISABLED ATA_FLAG_DISABLED is only used by drivers which don't use ->error_handler framework and is largely broken. Its only meaningful function is to make irq handlers skip processing if the flag is set, which is largely useless and even harmful as it makes those ports more likely to cause IRQ storms. Kill ATA_FLAG_DISABLED and makes the callers disable attached devices instead. ata_port_probe() and ata_port_disable() which manipulate the flag are also killed. This simplifies condition check in IRQ handlers. While updating IRQ handlers, remove ap NULL check as libata guarantees consecutive port allocation (unoccupied ports are initialized with dummies) and long-obsolete ATA_QCFLAG_ACTIVE check (checked by ata_qc_from_tag()). Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2010-05-10 13:41:30 -06:00
ap->link.device[0].class = ATA_DEV_NONE;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
goto out_unlock;
}
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ap->link.device[0].class = res->ata_class;
if (ap->link.device[0].class == ATA_DEV_UNKNOWN)
libata: kill ATA_FLAG_DISABLED ATA_FLAG_DISABLED is only used by drivers which don't use ->error_handler framework and is largely broken. Its only meaningful function is to make irq handlers skip processing if the flag is set, which is largely useless and even harmful as it makes those ports more likely to cause IRQ storms. Kill ATA_FLAG_DISABLED and makes the callers disable attached devices instead. ata_port_probe() and ata_port_disable() which manipulate the flag are also killed. This simplifies condition check in IRQ handlers. While updating IRQ handlers, remove ap NULL check as libata guarantees consecutive port allocation (unoccupied ports are initialized with dummies) and long-obsolete ATA_QCFLAG_ACTIVE check (checked by ata_qc_from_tag()). Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2010-05-10 13:41:30 -06:00
ap->link.device[0].class = ATA_DEV_NONE;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
out_unlock:
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
LEAVE;
}
/**
* ipr_ata_post_internal - Cleanup after an internal command
* @qc: ATA queued command
*
* Return value:
* none
**/
static void ipr_ata_post_internal(struct ata_queued_cmd *qc)
{
struct ipr_sata_port *sata_port = qc->ap->private_data;
struct ipr_ioa_cfg *ioa_cfg = sata_port->ioa_cfg;
struct ipr_cmnd *ipr_cmd;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
unsigned long flags;
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
while (ioa_cfg->in_reset_reload) {
[SCSI] ipr: SATA reset - wait for host reset completion If an ipr adapter hits a fatal microcode error requiring a reset while a SATA device is going through EH, it can result in a command getting issued to the ipr adapter while it is getting reset, which can cause PCI bus errors. Wait for any outstanding adapter reset to finish prior to issuing a SATA device reset. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:27:58 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for_each_hrrq(hrrq, ioa_cfg) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock(&hrrq->_lock);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_for_each_entry(ipr_cmd, &hrrq->hrrq_pending_q, queue) {
if (ipr_cmd->qc == qc) {
ipr_device_reset(ioa_cfg, sata_port->res);
break;
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&hrrq->_lock);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
}
/**
* ipr_copy_sata_tf - Copy a SATA taskfile to an IOA data structure
* @regs: destination
* @tf: source ATA taskfile
*
* Return value:
* none
**/
static void ipr_copy_sata_tf(struct ipr_ioarcb_ata_regs *regs,
struct ata_taskfile *tf)
{
regs->feature = tf->feature;
regs->nsect = tf->nsect;
regs->lbal = tf->lbal;
regs->lbam = tf->lbam;
regs->lbah = tf->lbah;
regs->device = tf->device;
regs->command = tf->command;
regs->hob_feature = tf->hob_feature;
regs->hob_nsect = tf->hob_nsect;
regs->hob_lbal = tf->hob_lbal;
regs->hob_lbam = tf->hob_lbam;
regs->hob_lbah = tf->hob_lbah;
regs->ctl = tf->ctl;
}
/**
* ipr_sata_done - done function for SATA commands
* @ipr_cmd: ipr command struct
*
* This function is invoked by the interrupt handler for
* ops generated by the SCSI mid-layer to SATA devices
*
* Return value:
* none
**/
static void ipr_sata_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ata_queued_cmd *qc = ipr_cmd->qc;
struct ipr_sata_port *sata_port = qc->ap->private_data;
struct ipr_resource_entry *res = sata_port->res;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock(&ipr_cmd->hrrq->_lock);
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
if (ipr_cmd->ioa_cfg->sis64)
memcpy(&sata_port->ioasa, &ipr_cmd->s.ioasa64.u.gata,
sizeof(struct ipr_ioasa_gata));
else
memcpy(&sata_port->ioasa, &ipr_cmd->s.ioasa.u.gata,
sizeof(struct ipr_ioasa_gata));
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ipr_dump_ioasa(ioa_cfg, ipr_cmd, res);
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
if (be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc_specific) & IPR_ATA_DEVICE_WAS_RESET)
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
scsi_report_device_reset(ioa_cfg->host, res->bus, res->target);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (IPR_IOASC_SENSE_KEY(ioasc) > RECOVERED_ERROR)
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
qc->err_mask |= __ac_err_mask(sata_port->ioasa.status);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
else
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
qc->err_mask |= ac_err_mask(sata_port->ioasa.status);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&ipr_cmd->hrrq->_lock);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ata_qc_complete(qc);
}
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
/**
* ipr_build_ata_ioadl64 - Build an ATA scatter/gather list
* @ipr_cmd: ipr command struct
* @qc: ATA queued command
*
**/
static void ipr_build_ata_ioadl64(struct ipr_cmnd *ipr_cmd,
struct ata_queued_cmd *qc)
{
u32 ioadl_flags = 0;
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: SATA DVD probing failed with 64bit adapter Driver passed the wrong IOADL address to IOA adapter. The patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-04-18 20:32:48 -06:00
struct ipr_ioadl64_desc *ioadl64 = ipr_cmd->i.ata_ioadl.ioadl64;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
struct ipr_ioadl64_desc *last_ioadl64 = NULL;
int len = qc->nbytes;
struct scatterlist *sg;
unsigned int si;
dma_addr_t dma_addr = ipr_cmd->dma_addr;
if (len == 0)
return;
if (qc->dma_dir == DMA_TO_DEVICE) {
ioadl_flags = IPR_IOADL_FLAGS_WRITE;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
} else if (qc->dma_dir == DMA_FROM_DEVICE)
ioadl_flags = IPR_IOADL_FLAGS_READ;
ioarcb->data_transfer_length = cpu_to_be32(len);
ioarcb->ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl64_desc) * ipr_cmd->dma_use_sg);
ioarcb->u.sis64_addr_data.data_ioadl_addr =
[SCSI] ipr: SATA DVD probing failed with 64bit adapter Driver passed the wrong IOADL address to IOA adapter. The patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-04-18 20:32:48 -06:00
cpu_to_be64(dma_addr + offsetof(struct ipr_cmnd, i.ata_ioadl.ioadl64));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
for_each_sg(qc->sg, sg, qc->n_elem, si) {
ioadl64->flags = cpu_to_be32(ioadl_flags);
ioadl64->data_len = cpu_to_be32(sg_dma_len(sg));
ioadl64->address = cpu_to_be64(sg_dma_address(sg));
last_ioadl64 = ioadl64;
ioadl64++;
}
if (likely(last_ioadl64))
last_ioadl64->flags |= cpu_to_be32(IPR_IOADL_FLAGS_LAST);
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
* ipr_build_ata_ioadl - Build an ATA scatter/gather list
* @ipr_cmd: ipr command struct
* @qc: ATA queued command
*
**/
static void ipr_build_ata_ioadl(struct ipr_cmnd *ipr_cmd,
struct ata_queued_cmd *qc)
{
u32 ioadl_flags = 0;
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
struct ipr_ioadl_desc *ioadl = ipr_cmd->i.ioadl;
[libata] kill ata_sg_is_last() Short term, this works around a bug introduced by early sg-chaining work. Long term, removing this function eliminates a branch from a hot path loop in each scatter/gather table build. Also, as this code demonstrates, we don't need to _track_ the end of the s/g list, as long as we mark it in some way. And doing so programatically is nice. So its a useful cleanup, regardless of its short term effects. Based conceptually on a quick patch by Jens Axboe. Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2007-10-18 14:21:18 -06:00
struct ipr_ioadl_desc *last_ioadl = NULL;
libata: eliminate the home grown dma padding in favour of that provided by the block layer ATA requires that all DMA transfers begin and end on word boundaries. Because of this, a large amount of machinery grew up in ide to adjust scatterlists on this basis. However, as of 2.5, the block layer has a dma_alignment variable which ensures both the beginning and length of a DMA transfer are aligned on the dma_alignment boundary. Although the block layer does adjust the beginning of the transfer to ensure this happens, it doesn't actually adjust the length, it merely makes sure that space is allocated for transfers beyond the declared length. The upshot of this is that scatterlists may be padded to any size between the actual length and the length adjusted to the dma_alignment safely knowing that memory is allocated in this region. Right at the moment, SCSI takes the default dma_aligment which is on a 512 byte boundary. Note that this aligment only applies to transfers coming in from user space. However, since all kernel allocations are automatically aligned on a minimum of 32 byte boundaries, it is safe to adjust them in this manner as well. tj: * Adjusting sg after padding is done in block layer. Make libata set queue alignment correctly for ATAPI devices and drop broken sg mangling from ata_sg_setup(). * Use request->raw_data_len for ATAPI transfer chunk size. * Killed qc->raw_nbytes. * Separated out killing qc->n_iter. Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2008-02-19 03:36:56 -07:00
int len = qc->nbytes;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
struct scatterlist *sg;
libata: convert to chained sg libata used private sg iterator to handle padding sg. Now that sg can be chained, padding can be handled using standard sg ops. Convert to chained sg. * s/qc->__sg/qc->sg/ * s/qc->pad_sgent/qc->extra_sg[]/. Because chaining consumes one sg entry. There need to be two extra sg entries. The renaming is also for future addition of other extra sg entries. * Padding setup is moved into ata_sg_setup_extra() which is organized in a way that future addition of other extra sg entries is easy. * qc->orig_n_elem is unused and removed. * qc->n_elem now contains the number of sg entries that LLDs should map. qc->mapped_n_elem is added to carry the original number of mapped sgs for unmapping. * The last sg of the original sg list is used to chain to extra sg list. The original last sg is pointed to by qc->last_sg and the content is stored in qc->saved_last_sg. It's restored during ata_sg_clean(). * All sg walking code has been updated. Unnecessary assertions and checks for conditions the core layer already guarantees are removed. Signed-off-by: Tejun Heo <htejun@gmail.com> Cc: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-12-05 00:43:11 -07:00
unsigned int si;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
if (len == 0)
return;
if (qc->dma_dir == DMA_TO_DEVICE) {
ioadl_flags = IPR_IOADL_FLAGS_WRITE;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ioarcb->data_transfer_length = cpu_to_be32(len);
ioarcb->ioadl_len =
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
cpu_to_be32(sizeof(struct ipr_ioadl_desc) * ipr_cmd->dma_use_sg);
} else if (qc->dma_dir == DMA_FROM_DEVICE) {
ioadl_flags = IPR_IOADL_FLAGS_READ;
ioarcb->read_data_transfer_length = cpu_to_be32(len);
ioarcb->read_ioadl_len =
cpu_to_be32(sizeof(struct ipr_ioadl_desc) * ipr_cmd->dma_use_sg);
}
libata: convert to chained sg libata used private sg iterator to handle padding sg. Now that sg can be chained, padding can be handled using standard sg ops. Convert to chained sg. * s/qc->__sg/qc->sg/ * s/qc->pad_sgent/qc->extra_sg[]/. Because chaining consumes one sg entry. There need to be two extra sg entries. The renaming is also for future addition of other extra sg entries. * Padding setup is moved into ata_sg_setup_extra() which is organized in a way that future addition of other extra sg entries is easy. * qc->orig_n_elem is unused and removed. * qc->n_elem now contains the number of sg entries that LLDs should map. qc->mapped_n_elem is added to carry the original number of mapped sgs for unmapping. * The last sg of the original sg list is used to chain to extra sg list. The original last sg is pointed to by qc->last_sg and the content is stored in qc->saved_last_sg. It's restored during ata_sg_clean(). * All sg walking code has been updated. Unnecessary assertions and checks for conditions the core layer already guarantees are removed. Signed-off-by: Tejun Heo <htejun@gmail.com> Cc: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2007-12-05 00:43:11 -07:00
for_each_sg(qc->sg, sg, qc->n_elem, si) {
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ioadl->flags_and_data_len = cpu_to_be32(ioadl_flags | sg_dma_len(sg));
ioadl->address = cpu_to_be32(sg_dma_address(sg));
[libata] kill ata_sg_is_last() Short term, this works around a bug introduced by early sg-chaining work. Long term, removing this function eliminates a branch from a hot path loop in each scatter/gather table build. Also, as this code demonstrates, we don't need to _track_ the end of the s/g list, as long as we mark it in some way. And doing so programatically is nice. So its a useful cleanup, regardless of its short term effects. Based conceptually on a quick patch by Jens Axboe. Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2007-10-18 14:21:18 -06:00
last_ioadl = ioadl;
ioadl++;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
}
[libata] kill ata_sg_is_last() Short term, this works around a bug introduced by early sg-chaining work. Long term, removing this function eliminates a branch from a hot path loop in each scatter/gather table build. Also, as this code demonstrates, we don't need to _track_ the end of the s/g list, as long as we mark it in some way. And doing so programatically is nice. So its a useful cleanup, regardless of its short term effects. Based conceptually on a quick patch by Jens Axboe. Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2007-10-18 14:21:18 -06:00
if (likely(last_ioadl))
last_ioadl->flags_and_data_len |= cpu_to_be32(IPR_IOADL_FLAGS_LAST);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
/**
* ipr_qc_defer - Get a free ipr_cmd
* @qc: queued command
*
* Return value:
* 0 if success
**/
static int ipr_qc_defer(struct ata_queued_cmd *qc)
{
struct ata_port *ap = qc->ap;
struct ipr_sata_port *sata_port = ap->private_data;
struct ipr_ioa_cfg *ioa_cfg = sata_port->ioa_cfg;
struct ipr_cmnd *ipr_cmd;
struct ipr_hrr_queue *hrrq;
int hrrq_id;
hrrq_id = ipr_get_hrrq_index(ioa_cfg);
hrrq = &ioa_cfg->hrrq[hrrq_id];
qc->lldd_task = NULL;
spin_lock(&hrrq->_lock);
if (unlikely(hrrq->ioa_is_dead)) {
spin_unlock(&hrrq->_lock);
return 0;
}
if (unlikely(!hrrq->allow_cmds)) {
spin_unlock(&hrrq->_lock);
return ATA_DEFER_LINK;
}
ipr_cmd = __ipr_get_free_ipr_cmnd(hrrq);
if (ipr_cmd == NULL) {
spin_unlock(&hrrq->_lock);
return ATA_DEFER_LINK;
}
qc->lldd_task = ipr_cmd;
spin_unlock(&hrrq->_lock);
return 0;
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
/**
* ipr_qc_issue - Issue a SATA qc to a device
* @qc: queued command
*
* Return value:
* 0 if success
**/
static unsigned int ipr_qc_issue(struct ata_queued_cmd *qc)
{
struct ata_port *ap = qc->ap;
struct ipr_sata_port *sata_port = ap->private_data;
struct ipr_resource_entry *res = sata_port->res;
struct ipr_ioa_cfg *ioa_cfg = sata_port->ioa_cfg;
struct ipr_cmnd *ipr_cmd;
struct ipr_ioarcb *ioarcb;
struct ipr_ioarcb_ata_regs *regs;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (qc->lldd_task == NULL)
ipr_qc_defer(qc);
ipr_cmd = qc->lldd_task;
if (ipr_cmd == NULL)
[SCSI] ipr: Return better qc_issue errors If qc_issue fails for some reason, return a better error to libata. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:43 -06:00
return AC_ERR_SYSTEM;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
qc->lldd_task = NULL;
spin_lock(&ipr_cmd->hrrq->_lock);
if (unlikely(!ipr_cmd->hrrq->allow_cmds ||
ipr_cmd->hrrq->ioa_is_dead)) {
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
spin_unlock(&ipr_cmd->hrrq->_lock);
return AC_ERR_SYSTEM;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_init_ipr_cmnd(ipr_cmd, ipr_lock_and_done);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ioa_cfg->sis64) {
regs = &ipr_cmd->i.ata_ioadl.regs;
ioarcb->add_cmd_parms_offset = cpu_to_be16(sizeof(*ioarcb));
} else
regs = &ioarcb->u.add_data.u.regs;
memset(regs, 0, sizeof(*regs));
ioarcb->add_cmd_parms_len = cpu_to_be16(sizeof(*regs));
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_pending_q);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ipr_cmd->qc = qc;
ipr_cmd->done = ipr_sata_done;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_cmd->ioarcb.res_handle = res->res_handle;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_ATA_PASSTHRU;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_LINK_DESC;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_ULEN_CHK;
libata: eliminate the home grown dma padding in favour of that provided by the block layer ATA requires that all DMA transfers begin and end on word boundaries. Because of this, a large amount of machinery grew up in ide to adjust scatterlists on this basis. However, as of 2.5, the block layer has a dma_alignment variable which ensures both the beginning and length of a DMA transfer are aligned on the dma_alignment boundary. Although the block layer does adjust the beginning of the transfer to ensure this happens, it doesn't actually adjust the length, it merely makes sure that space is allocated for transfers beyond the declared length. The upshot of this is that scatterlists may be padded to any size between the actual length and the length adjusted to the dma_alignment safely knowing that memory is allocated in this region. Right at the moment, SCSI takes the default dma_aligment which is on a 512 byte boundary. Note that this aligment only applies to transfers coming in from user space. However, since all kernel allocations are automatically aligned on a minimum of 32 byte boundaries, it is safe to adjust them in this manner as well. tj: * Adjusting sg after padding is done in block layer. Make libata set queue alignment correctly for ATAPI devices and drop broken sg mangling from ata_sg_setup(). * Use request->raw_data_len for ATAPI transfer chunk size. * Killed qc->raw_nbytes. * Separated out killing qc->n_iter. Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2008-02-19 03:36:56 -07:00
ipr_cmd->dma_use_sg = qc->n_elem;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ioa_cfg->sis64)
ipr_build_ata_ioadl64(ipr_cmd, qc);
else
ipr_build_ata_ioadl(ipr_cmd, qc);
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
regs->flags |= IPR_ATA_FLAG_STATUS_ON_GOOD_COMPLETION;
ipr_copy_sata_tf(regs, &qc->tf);
memcpy(ioarcb->cmd_pkt.cdb, qc->cdb, IPR_MAX_CDB_LEN);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_trc_hook(ipr_cmd, IPR_TRACE_START, IPR_GET_RES_PHYS_LOC(res));
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
switch (qc->tf.protocol) {
case ATA_PROT_NODATA:
case ATA_PROT_PIO:
break;
case ATA_PROT_DMA:
regs->flags |= IPR_ATA_FLAG_XFER_TYPE_DMA;
break;
libata: rename ATA_PROT_ATAPI_* to ATAPI_PROT_* ATA_PROT_ATAPI_* are ugly and naming schemes between ATA_PROT_* and ATA_PROT_ATAPI_* are inconsistent causing confusion. Rename them to ATAPI_PROT_* and make them consistent with ATA counterpart. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2007-12-18 14:34:43 -07:00
case ATAPI_PROT_PIO:
case ATAPI_PROT_NODATA:
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
regs->flags |= IPR_ATA_FLAG_PACKET_CMD;
break;
libata: rename ATA_PROT_ATAPI_* to ATAPI_PROT_* ATA_PROT_ATAPI_* are ugly and naming schemes between ATA_PROT_* and ATA_PROT_ATAPI_* are inconsistent causing confusion. Rename them to ATAPI_PROT_* and make them consistent with ATA counterpart. Signed-off-by: Tejun Heo <htejun@gmail.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2007-12-18 14:34:43 -07:00
case ATAPI_PROT_DMA:
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
regs->flags |= IPR_ATA_FLAG_PACKET_CMD;
regs->flags |= IPR_ATA_FLAG_XFER_TYPE_DMA;
break;
default:
WARN_ON(1);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&ipr_cmd->hrrq->_lock);
[SCSI] ipr: Return better qc_issue errors If qc_issue fails for some reason, return a better error to libata. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:43:43 -06:00
return AC_ERR_INVALID;
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
}
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_send_command(ipr_cmd);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&ipr_cmd->hrrq->_lock);
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
return 0;
}
libata: replace tf_read with qc_fill_rtf for non-SFF drivers Now that all SFF stuff is separated out of core layer, core layer doesn't call ops->tf_read directly. It gets called only via ops->qc_fill_rtf() for non-SFF drivers. This patch directly implements private ops->qc_fill_rtf() for non-SFF controllers and kill ops->tf_read(). This is much cleaner for non-SFF controllers as some of them have to cache SFF register values in private data structure and report the cached values via ops->tf_read(). Also, ops->tf_read() gets nasty for controllers which don't have clear notion of TF registers when operation is not in progress. As this change makes default ops->qc_fill_rtf unnecessary, move ata_sff_qc_fill_rtf() form ata_base_port_ops to ata_sff_port_ops where it belongs. Signed-off-by: Tejun Heo <htejun@gmail.com>
2008-04-07 07:47:20 -06:00
/**
* ipr_qc_fill_rtf - Read result TF
* @qc: ATA queued command
*
* Return value:
* true
**/
static bool ipr_qc_fill_rtf(struct ata_queued_cmd *qc)
{
struct ipr_sata_port *sata_port = qc->ap->private_data;
struct ipr_ioasa_gata *g = &sata_port->ioasa;
struct ata_taskfile *tf = &qc->result_tf;
tf->feature = g->error;
tf->nsect = g->nsect;
tf->lbal = g->lbal;
tf->lbam = g->lbam;
tf->lbah = g->lbah;
tf->device = g->device;
tf->command = g->status;
tf->hob_nsect = g->hob_nsect;
tf->hob_lbal = g->hob_lbal;
tf->hob_lbam = g->hob_lbam;
tf->hob_lbah = g->hob_lbah;
return true;
}
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
static struct ata_port_operations ipr_sata_ops = {
.phy_reset = ipr_ata_phy_reset,
libata: make reset related methods proper port operations Currently reset methods are not specified directly in the ata_port_operations table. If a LLD wants to use custom reset methods, it should construct and use a error_handler which uses those reset methods. It's done this way for two reasons. First, the ops table already contained too many methods and adding four more of them would noticeably increase the amount of necessary boilerplate code all over low level drivers. Second, as ->error_handler uses those reset methods, it can get confusing. ie. By overriding ->error_handler, those reset ops can be made useless making layering a bit hazy. Now that ops table uses inheritance, the first problem doesn't exist anymore. The second isn't completely solved but is relieved by providing default values - most drivers can just override what it has implemented and don't have to concern itself about higher level callbacks. In fact, there currently is no driver which actually modifies error handling behavior. Drivers which override ->error_handler just wraps the standard error handler only to prepare the controller for EH. I don't think making ops layering strict has any noticeable benefit. This patch makes ->prereset, ->softreset, ->hardreset, ->postreset and their PMP counterparts propoer ops. Default ops are provided in the base ops tables and drivers are converted to override individual reset methods instead of creating custom error_handler. * ata_std_error_handler() doesn't use sata_std_hardreset() if SCRs aren't accessible. sata_promise doesn't need to use separate error_handlers for PATA and SATA anymore. * softreset is broken for sata_inic162x and sata_sx4. As libata now always prefers hardreset, this doesn't really matter but the ops are forced to NULL using ATA_OP_NULL for documentation purpose. * pata_hpt374 needs to use different prereset for the first and second PCI functions. This used to be done by branching from hpt374_error_handler(). The proper way to do this is to use separate ops and port_info tables for each function. Converted. Signed-off-by: Tejun Heo <htejun@gmail.com>
2008-03-24 21:22:50 -06:00
.hardreset = ipr_sata_reset,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
.post_internal_cmd = ipr_ata_post_internal,
.qc_prep = ata_noop_qc_prep,
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
.qc_defer = ipr_qc_defer,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
.qc_issue = ipr_qc_issue,
libata: replace tf_read with qc_fill_rtf for non-SFF drivers Now that all SFF stuff is separated out of core layer, core layer doesn't call ops->tf_read directly. It gets called only via ops->qc_fill_rtf() for non-SFF drivers. This patch directly implements private ops->qc_fill_rtf() for non-SFF controllers and kill ops->tf_read(). This is much cleaner for non-SFF controllers as some of them have to cache SFF register values in private data structure and report the cached values via ops->tf_read(). Also, ops->tf_read() gets nasty for controllers which don't have clear notion of TF registers when operation is not in progress. As this change makes default ops->qc_fill_rtf unnecessary, move ata_sff_qc_fill_rtf() form ata_base_port_ops to ata_sff_port_ops where it belongs. Signed-off-by: Tejun Heo <htejun@gmail.com>
2008-04-07 07:47:20 -06:00
.qc_fill_rtf = ipr_qc_fill_rtf,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
.port_start = ata_sas_port_start,
.port_stop = ata_sas_port_stop
};
static struct ata_port_info sata_port_info = {
ata: Add a new flag to destinguish sas controller SAS controller has its own tag allocation, which doesn't directly match to ATA tag, so SAS and SATA have different code path for ata tags. Originally we use port->scsi_host (98bd4be1) to destinguish SAS controller, but libsas set ->scsi_host too, so we can't use it for the destinguish, we add a new flag for this purpose. Without this patch, the following oops can happen because scsi-mq uses a host-wide tag map shared among all devices with some integer tag values >= ATA_MAX_QUEUE. These unexpectedly high tag values cause __ata_qc_from_tag() to return NULL, which is then dereferenced in ata_qc_new_init(). BUG: unable to handle kernel NULL pointer dereference at 0000000000000058 IP: [<ffffffff804fd46e>] ata_qc_new_init+0x3e/0x120 PGD 32adf0067 PUD 32adf1067 PMD 0 Oops: 0002 [#1] SMP DEBUG_PAGEALLOC Modules linked in: iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi igb i2c_algo_bit ptp pps_core pm80xx libsas scsi_transport_sas sg coretemp eeprom w83795 i2c_i801 CPU: 4 PID: 1450 Comm: cydiskbench Not tainted 4.0.0-rc3 #1 Hardware name: Supermicro X8DTH-i/6/iF/6F/X8DTH, BIOS 2.1b 05/04/12 task: ffff8800ba86d500 ti: ffff88032a064000 task.ti: ffff88032a064000 RIP: 0010:[<ffffffff804fd46e>] [<ffffffff804fd46e>] ata_qc_new_init+0x3e/0x120 RSP: 0018:ffff88032a067858 EFLAGS: 00010046 RAX: 0000000000000000 RBX: ffff8800ba0d2230 RCX: 000000000000002a RDX: ffffffff80505ae0 RSI: 0000000000000020 RDI: ffff8800ba0d2230 RBP: ffff88032a067868 R08: 0000000000000201 R09: 0000000000000001 R10: 0000000000000000 R11: 0000000000000000 R12: ffff8800ba0d0000 R13: ffff8800ba0d2230 R14: ffffffff80505ae0 R15: ffff8800ba0d0000 FS: 0000000041223950(0063) GS:ffff88033e480000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000058 CR3: 000000032a0a3000 CR4: 00000000000006e0 Stack: ffff880329eee758 ffff880329eee758 ffff88032a0678a8 ffffffff80502dad ffff8800ba167978 ffff880329eee758 ffff88032bf9c520 ffff8800ba167978 ffff88032bf9c520 ffff88032bf9a290 ffff88032a0678b8 ffffffff80506909 Call Trace: [<ffffffff80502dad>] ata_scsi_translate+0x3d/0x1b0 [<ffffffff80506909>] ata_sas_queuecmd+0x149/0x2a0 [<ffffffffa0046650>] sas_queuecommand+0xa0/0x1f0 [libsas] [<ffffffff804ea544>] scsi_dispatch_cmd+0xd4/0x1a0 [<ffffffff804eb50f>] scsi_queue_rq+0x66f/0x7f0 [<ffffffff803e5098>] __blk_mq_run_hw_queue+0x208/0x3f0 [<ffffffff803e54b8>] blk_mq_run_hw_queue+0x88/0xc0 [<ffffffff803e5c74>] blk_mq_insert_request+0xc4/0x130 [<ffffffff803e0b63>] blk_execute_rq_nowait+0x73/0x160 [<ffffffffa0023fca>] sg_common_write+0x3da/0x720 [sg] [<ffffffffa0025100>] sg_new_write+0x250/0x360 [sg] [<ffffffffa0025feb>] sg_write+0x13b/0x450 [sg] [<ffffffff8032ec91>] vfs_write+0xd1/0x1b0 [<ffffffff8032ee54>] SyS_write+0x54/0xc0 [<ffffffff80689932>] system_call_fastpath+0x12/0x17 tj: updated description. Fixes: 12cb5ce101ab ("libata: use blk taging") Reported-and-tested-by: Tony Battersby <tonyb@cybernetics.com> Signed-off-by: Shaohua Li <shli@fb.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2015-03-12 11:32:18 -06:00
.flags = ATA_FLAG_SATA | ATA_FLAG_PIO_DMA |
ATA_FLAG_SAS_HOST,
ipr/sas_ata: use mode mask macros from <linux/ata.h> Commit 14bdef982caeda19afe34010482867c18217c641 ([libata] convert drivers to use ata.h mode mask defines) didn't convert these two libata driver outside drivers/ata/... Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
2011-01-21 10:32:01 -07:00
.pio_mask = ATA_PIO4_ONLY,
.mwdma_mask = ATA_MWDMA2,
.udma_mask = ATA_UDMA6,
[SCSI] ipr: Support attaching SATA devices Adds support to attach SATA devices to ipr SAS adapters. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 11:39:20 -06:00
.port_ops = &ipr_sata_ops
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
#ifdef CONFIG_PPC_PSERIES
static const u16 ipr_blocked_processors[] = {
powerpc: Rename 64-bit PVR constants to PVR_foo We have an old FIXME in reg.h which points out that we should standardise on PVR_foo for our PVR #defines. Currently we use PVR_ on 32-bit and PV_ on 64-bit. So do that rename and remove the FIXME. Seeing as we're touching all but one usage of __is_processor(), rename it to something less ugly and more indicative of what it does, which is simply to check the PVR version. Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2012-08-19 15:44:01 -06:00
PVR_NORTHSTAR,
PVR_PULSAR,
PVR_POWER4,
PVR_ICESTAR,
PVR_SSTAR,
PVR_POWER4p,
PVR_630,
PVR_630p
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
};
/**
* ipr_invalid_adapter - Determine if this adapter is supported on this hardware
* @ioa_cfg: ioa cfg struct
*
* Adapters that use Gemstone revision < 3.1 do not work reliably on
* certain pSeries hardware. This function determines if the given
* adapter is in one of these confgurations or not.
*
* Return value:
* 1 if adapter is not supported / 0 if adapter is supported
**/
static int ipr_invalid_adapter(struct ipr_ioa_cfg *ioa_cfg)
{
int i;
PCI: Change all drivers to use pci_device->revision Instead of all drivers reading pci config space to get the revision ID, they can now use the pci_device->revision member. This exposes some issues where drivers where reading a word or a dword for the revision number, and adding useless error-handling around the read. Some drivers even just read it for no purpose of all. In devices where the revision ID is being copied over and used in what appears to be the equivalent of hotpath, I have left the copy code and the cached copy as not to influence the driver's performance. Compile tested with make all{yes,mod}config on x86_64 and i386. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Acked-by: Dave Jones <davej@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 16:46:36 -06:00
if ((ioa_cfg->type == 0x5702) && (ioa_cfg->pdev->revision < 4)) {
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
for (i = 0; i < ARRAY_SIZE(ipr_blocked_processors); i++) {
powerpc: Rename 64-bit PVR constants to PVR_foo We have an old FIXME in reg.h which points out that we should standardise on PVR_foo for our PVR #defines. Currently we use PVR_ on 32-bit and PV_ on 64-bit. So do that rename and remove the FIXME. Seeing as we're touching all but one usage of __is_processor(), rename it to something less ugly and more indicative of what it does, which is simply to check the PVR version. Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2012-08-19 15:44:01 -06:00
if (pvr_version_is(ipr_blocked_processors[i]))
PCI: Change all drivers to use pci_device->revision Instead of all drivers reading pci config space to get the revision ID, they can now use the pci_device->revision member. This exposes some issues where drivers where reading a word or a dword for the revision number, and adding useless error-handling around the read. Some drivers even just read it for no purpose of all. In devices where the revision ID is being copied over and used in what appears to be the equivalent of hotpath, I have left the copy code and the cached copy as not to influence the driver's performance. Compile tested with make all{yes,mod}config on x86_64 and i386. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Acked-by: Dave Jones <davej@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 16:46:36 -06:00
return 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
}
return 0;
}
#else
#define ipr_invalid_adapter(ioa_cfg) 0
#endif
/**
* ipr_ioa_bringdown_done - IOA bring down completion.
* @ipr_cmd: ipr command struct
*
* This function processes the completion of an adapter bring down.
* It wakes any reset sleepers.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioa_bringdown_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
[SCSI] ipr: Need to reset adapter after the 6th EEH error Add reset adapter after the 6th EEH errors in ipr driver. This triggers the adapter reset via the PCI config space even when the slot is frozen. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-04-17 08:34:06 -06:00
int i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
if (!ioa_cfg->hrrq[IPR_INIT_HRRQ].removing_ioa) {
ipr_trace;
scsi: ipr: Fix scsi-mq lockdep issue Fixes the following lockdep warning that can occur when scsi-mq is enabled with ipr due to ipr calling scsi_unblock_requests from irq context. The fix is to move the call to scsi_unblock_requests to ipr's existing workqueue. stack backtrace: CPU: 28 PID: 0 Comm: swapper/28 Not tainted 4.13.0-rc2-gcc6x-gf74c89b #1 Call Trace: [c000001fffe97550] [c000000000b50818] dump_stack+0xe8/0x160 (unreliable) [c000001fffe97590] [c0000000001586d0] print_usage_bug+0x2d0/0x390 [c000001fffe97640] [c000000000158f34] mark_lock+0x7a4/0x8e0 [c000001fffe976f0] [c00000000015a000] __lock_acquire+0x6a0/0x1a70 [c000001fffe97860] [c00000000015befc] lock_acquire+0xec/0x2e0 [c000001fffe97930] [c000000000b71514] _raw_spin_lock+0x44/0x70 [c000001fffe97960] [c0000000005b60f4] blk_mq_sched_dispatch_requests+0xa4/0x2a0 [c000001fffe979c0] [c0000000005acac0] __blk_mq_run_hw_queue+0x100/0x2c0 [c000001fffe97a00] [c0000000005ad478] __blk_mq_delay_run_hw_queue+0x118/0x130 [c000001fffe97a40] [c0000000005ad61c] blk_mq_start_hw_queues+0x6c/0xa0 [c000001fffe97a80] [c000000000797aac] scsi_kick_queue+0x2c/0x60 [c000001fffe97aa0] [c000000000797cf0] scsi_run_queue+0x210/0x360 [c000001fffe97b10] [c00000000079b888] scsi_run_host_queues+0x48/0x80 [c000001fffe97b40] [c0000000007b6090] ipr_ioa_bringdown_done+0x70/0x1e0 [c000001fffe97bc0] [c0000000007bc860] ipr_reset_ioa_job+0x80/0xf0 [c000001fffe97bf0] [c0000000007b4d50] ipr_reset_timer_done+0xd0/0x100 [c000001fffe97c30] [c0000000001937bc] call_timer_fn+0xdc/0x4b0 [c000001fffe97cf0] [c000000000193d08] expire_timers+0x178/0x330 [c000001fffe97d60] [c0000000001940c8] run_timer_softirq+0xb8/0x120 [c000001fffe97de0] [c000000000b726a8] __do_softirq+0x168/0x6d8 [c000001fffe97ef0] [c0000000000df2c8] irq_exit+0x108/0x150 [c000001fffe97f10] [c000000000017bf4] __do_irq+0x2a4/0x4a0 [c000001fffe97f90] [c00000000002da50] call_do_irq+0x14/0x24 [c0000007fad93aa0] [c000000000017e8c] do_IRQ+0x9c/0x140 [c0000007fad93af0] [c000000000008b98] hardware_interrupt_common+0x138/0x140 Reported-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-01 09:21:30 -06:00
ioa_cfg->scsi_unblock = 1;
schedule_work(&ioa_cfg->work_q);
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->in_reset_reload = 0;
ioa_cfg->reset_retries = 0;
[SCSI] ipr: Need to reset adapter after the 6th EEH error Add reset adapter after the 6th EEH errors in ipr driver. This triggers the adapter reset via the PCI config space even when the slot is frozen. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-04-17 08:34:06 -06:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].ioa_is_dead = 1;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
wmb();
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
wake_up_all(&ioa_cfg->reset_wait_q);
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_ioa_reset_done - IOA reset completion.
* @ipr_cmd: ipr command struct
*
* This function processes the completion of an adapter reset.
* It schedules any necessary mid-layer add/removes and
* wakes any reset sleepers.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioa_reset_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_resource_entry *res;
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
int j;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
ioa_cfg->in_reset_reload = 0;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
for (j = 0; j < ioa_cfg->hrrq_num; j++) {
spin_lock(&ioa_cfg->hrrq[j]._lock);
ioa_cfg->hrrq[j].allow_cmds = 1;
spin_unlock(&ioa_cfg->hrrq[j]._lock);
}
wmb();
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->reset_cmd = NULL;
[SCSI] ipr: Runtime reset Some IPR RAID adapter will automatically create single device RAID arrays for all attached devices when the card is initialized. Setting the RUNTIME_RESET doorbell bit will prevent this from occurring, since we only want this behavior the first time the card is initialized and not each time the card happens to get reset. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:54 -07:00
ioa_cfg->doorbell |= IPR_RUNTIME_RESET;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_for_each_entry(res, &ioa_cfg->used_res_q, queue) {
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
if (res->add_to_ml || res->del_from_ml) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_trace;
break;
}
}
schedule_work(&ioa_cfg->work_q);
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
for (j = 0; j < IPR_NUM_HCAMS; j++) {
list_del_init(&ioa_cfg->hostrcb[j]->queue);
if (j < IPR_NUM_LOG_HCAMS)
ipr_send_hcam(ioa_cfg,
IPR_HCAM_CDB_OP_CODE_LOG_DATA,
ioa_cfg->hostrcb[j]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
else
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
ipr_send_hcam(ioa_cfg,
IPR_HCAM_CDB_OP_CODE_CONFIG_CHANGE,
ioa_cfg->hostrcb[j]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Handle UA on disk array following an adapter reset On certain ipr RAID adapters, which are capable of multi-initiator configurations, the disk array, or VSET, resources will be in a POR Unit Attention state following an adapter reset. In order for the midlayer to handle the UA, I must report a bus reset to the logical disk array bus at the end of an adapter reset. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:08 -06:00
scsi_report_bus_reset(ioa_cfg->host, IPR_VSET_BUS);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_info(&ioa_cfg->pdev->dev, "IOA initialized.\n");
ioa_cfg->reset_retries = 0;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
wake_up_all(&ioa_cfg->reset_wait_q);
scsi: ipr: Fix scsi-mq lockdep issue Fixes the following lockdep warning that can occur when scsi-mq is enabled with ipr due to ipr calling scsi_unblock_requests from irq context. The fix is to move the call to scsi_unblock_requests to ipr's existing workqueue. stack backtrace: CPU: 28 PID: 0 Comm: swapper/28 Not tainted 4.13.0-rc2-gcc6x-gf74c89b #1 Call Trace: [c000001fffe97550] [c000000000b50818] dump_stack+0xe8/0x160 (unreliable) [c000001fffe97590] [c0000000001586d0] print_usage_bug+0x2d0/0x390 [c000001fffe97640] [c000000000158f34] mark_lock+0x7a4/0x8e0 [c000001fffe976f0] [c00000000015a000] __lock_acquire+0x6a0/0x1a70 [c000001fffe97860] [c00000000015befc] lock_acquire+0xec/0x2e0 [c000001fffe97930] [c000000000b71514] _raw_spin_lock+0x44/0x70 [c000001fffe97960] [c0000000005b60f4] blk_mq_sched_dispatch_requests+0xa4/0x2a0 [c000001fffe979c0] [c0000000005acac0] __blk_mq_run_hw_queue+0x100/0x2c0 [c000001fffe97a00] [c0000000005ad478] __blk_mq_delay_run_hw_queue+0x118/0x130 [c000001fffe97a40] [c0000000005ad61c] blk_mq_start_hw_queues+0x6c/0xa0 [c000001fffe97a80] [c000000000797aac] scsi_kick_queue+0x2c/0x60 [c000001fffe97aa0] [c000000000797cf0] scsi_run_queue+0x210/0x360 [c000001fffe97b10] [c00000000079b888] scsi_run_host_queues+0x48/0x80 [c000001fffe97b40] [c0000000007b6090] ipr_ioa_bringdown_done+0x70/0x1e0 [c000001fffe97bc0] [c0000000007bc860] ipr_reset_ioa_job+0x80/0xf0 [c000001fffe97bf0] [c0000000007b4d50] ipr_reset_timer_done+0xd0/0x100 [c000001fffe97c30] [c0000000001937bc] call_timer_fn+0xdc/0x4b0 [c000001fffe97cf0] [c000000000193d08] expire_timers+0x178/0x330 [c000001fffe97d60] [c0000000001940c8] run_timer_softirq+0xb8/0x120 [c000001fffe97de0] [c000000000b726a8] __do_softirq+0x168/0x6d8 [c000001fffe97ef0] [c0000000000df2c8] irq_exit+0x108/0x150 [c000001fffe97f10] [c000000000017bf4] __do_irq+0x2a4/0x4a0 [c000001fffe97f90] [c00000000002da50] call_do_irq+0x14/0x24 [c0000007fad93aa0] [c000000000017e8c] do_IRQ+0x9c/0x140 [c0000007fad93af0] [c000000000008b98] hardware_interrupt_common+0x138/0x140 Reported-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-01 09:21:30 -06:00
ioa_cfg->scsi_unblock = 1;
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
schedule_work(&ioa_cfg->work_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_set_sup_dev_dflt - Initialize a Set Supported Device buffer
* @supported_dev: supported device struct
* @vpids: vendor product id struct
*
* Return value:
* none
**/
static void ipr_set_sup_dev_dflt(struct ipr_supported_device *supported_dev,
struct ipr_std_inq_vpids *vpids)
{
memset(supported_dev, 0, sizeof(struct ipr_supported_device));
memcpy(&supported_dev->vpids, vpids, sizeof(struct ipr_std_inq_vpids));
supported_dev->num_records = 1;
supported_dev->data_length =
cpu_to_be16(sizeof(struct ipr_supported_device));
supported_dev->reserved = 0;
}
/**
* ipr_set_supported_devs - Send Set Supported Devices for a device
* @ipr_cmd: ipr command struct
*
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
* This function sends a Set Supported Devices to the adapter
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_set_supported_devs(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_supported_device *supp_dev = &ioa_cfg->vpd_cbs->supp_dev;
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
struct ipr_resource_entry *res = ipr_cmd->u.res;
ipr_cmd->job_step = ipr_ioa_reset_done;
list_for_each_entry_continue(res, &ioa_cfg->used_res_q, queue) {
[SCSI] ipr: Fixup device type check Fixup a check used by the ipr driver to determine if a given device is a SCSI disk. Due to the addition of support for attaching SATA devices, this check needs to be more robust. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-03-29 08:37:22 -07:00
if (!ipr_is_scsi_disk(res))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
continue;
ipr_cmd->u.res = res;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_set_sup_dev_dflt(supp_dev, &res->std_inq_data.vpids);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
ioarcb->cmd_pkt.cdb[0] = IPR_SET_SUPPORTED_DEVICES;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ioarcb->cmd_pkt.cdb[1] = IPR_SET_ALL_SUPPORTED_DEVICES;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->cmd_pkt.cdb[7] = (sizeof(struct ipr_supported_device) >> 8) & 0xff;
ioarcb->cmd_pkt.cdb[8] = sizeof(struct ipr_supported_device) & 0xff;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_init_ioadl(ipr_cmd,
ioa_cfg->vpd_cbs_dma +
offsetof(struct ipr_misc_cbs, supp_dev),
sizeof(struct ipr_supported_device),
IPR_IOADL_FLAGS_WRITE_LAST);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout,
IPR_SET_SUP_DEVICE_TIMEOUT);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (!ioa_cfg->sis64)
ipr_cmd->job_step = ipr_set_supported_devs;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
LEAVE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return IPR_RC_JOB_RETURN;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
LEAVE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return IPR_RC_JOB_CONTINUE;
}
/**
* ipr_get_mode_page - Locate specified mode page
* @mode_pages: mode page buffer
* @page_code: page code to find
* @len: minimum required length for mode page
*
* Return value:
* pointer to mode page / NULL on failure
**/
static void *ipr_get_mode_page(struct ipr_mode_pages *mode_pages,
u32 page_code, u32 len)
{
struct ipr_mode_page_hdr *mode_hdr;
u32 page_length;
u32 length;
if (!mode_pages || (mode_pages->hdr.length == 0))
return NULL;
length = (mode_pages->hdr.length + 1) - 4 - mode_pages->hdr.block_desc_len;
mode_hdr = (struct ipr_mode_page_hdr *)
(mode_pages->data + mode_pages->hdr.block_desc_len);
while (length) {
if (IPR_GET_MODE_PAGE_CODE(mode_hdr) == page_code) {
if (mode_hdr->page_length >= (len - sizeof(struct ipr_mode_page_hdr)))
return mode_hdr;
break;
} else {
page_length = (sizeof(struct ipr_mode_page_hdr) +
mode_hdr->page_length);
length -= page_length;
mode_hdr = (struct ipr_mode_page_hdr *)
((unsigned long)mode_hdr + page_length);
}
}
return NULL;
}
/**
* ipr_check_term_power - Check for term power errors
* @ioa_cfg: ioa config struct
* @mode_pages: IOAFP mode pages buffer
*
* Check the IOAFP's mode page 28 for term power errors
*
* Return value:
* nothing
**/
static void ipr_check_term_power(struct ipr_ioa_cfg *ioa_cfg,
struct ipr_mode_pages *mode_pages)
{
int i;
int entry_length;
struct ipr_dev_bus_entry *bus;
struct ipr_mode_page28 *mode_page;
mode_page = ipr_get_mode_page(mode_pages, 0x28,
sizeof(struct ipr_mode_page28));
entry_length = mode_page->entry_length;
bus = mode_page->bus;
for (i = 0; i < mode_page->num_entries; i++) {
if (bus->flags & IPR_SCSI_ATTR_NO_TERM_PWR) {
dev_err(&ioa_cfg->pdev->dev,
"Term power is absent on scsi bus %d\n",
bus->res_addr.bus);
}
bus = (struct ipr_dev_bus_entry *)((char *)bus + entry_length);
}
}
/**
* ipr_scsi_bus_speed_limit - Limit the SCSI speed based on SES table
* @ioa_cfg: ioa config struct
*
* Looks through the config table checking for SES devices. If
* the SES device is in the SES table indicating a maximum SCSI
* bus speed, the speed is limited for the bus.
*
* Return value:
* none
**/
static void ipr_scsi_bus_speed_limit(struct ipr_ioa_cfg *ioa_cfg)
{
u32 max_xfer_rate;
int i;
for (i = 0; i < IPR_MAX_NUM_BUSES; i++) {
max_xfer_rate = ipr_get_max_scsi_speed(ioa_cfg, i,
ioa_cfg->bus_attr[i].bus_width);
if (max_xfer_rate < ioa_cfg->bus_attr[i].max_xfer_rate)
ioa_cfg->bus_attr[i].max_xfer_rate = max_xfer_rate;
}
}
/**
* ipr_modify_ioafp_mode_page_28 - Modify IOAFP Mode Page 28
* @ioa_cfg: ioa config struct
* @mode_pages: mode page 28 buffer
*
* Updates mode page 28 based on driver configuration
*
* Return value:
* none
**/
static void ipr_modify_ioafp_mode_page_28(struct ipr_ioa_cfg *ioa_cfg,
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
struct ipr_mode_pages *mode_pages)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
int i, entry_length;
struct ipr_dev_bus_entry *bus;
struct ipr_bus_attributes *bus_attr;
struct ipr_mode_page28 *mode_page;
mode_page = ipr_get_mode_page(mode_pages, 0x28,
sizeof(struct ipr_mode_page28));
entry_length = mode_page->entry_length;
/* Loop for each device bus entry */
for (i = 0, bus = mode_page->bus;
i < mode_page->num_entries;
i++, bus = (struct ipr_dev_bus_entry *)((u8 *)bus + entry_length)) {
if (bus->res_addr.bus > IPR_MAX_NUM_BUSES) {
dev_err(&ioa_cfg->pdev->dev,
"Invalid resource address reported: 0x%08X\n",
IPR_GET_PHYS_LOC(bus->res_addr));
continue;
}
bus_attr = &ioa_cfg->bus_attr[i];
bus->extended_reset_delay = IPR_EXTENDED_RESET_DELAY;
bus->bus_width = bus_attr->bus_width;
bus->max_xfer_rate = cpu_to_be32(bus_attr->max_xfer_rate);
bus->flags &= ~IPR_SCSI_ATTR_QAS_MASK;
if (bus_attr->qas_enabled)
bus->flags |= IPR_SCSI_ATTR_ENABLE_QAS;
else
bus->flags |= IPR_SCSI_ATTR_DISABLE_QAS;
}
}
/**
* ipr_build_mode_select - Build a mode select command
* @ipr_cmd: ipr command struct
* @res_handle: resource handle to send command to
* @parm: Byte 2 of Mode Sense command
* @dma_addr: DMA buffer address
* @xfer_len: data transfer length
*
* Return value:
* none
**/
static void ipr_build_mode_select(struct ipr_cmnd *ipr_cmd,
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
__be32 res_handle, u8 parm,
dma_addr_t dma_addr, u8 xfer_len)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
ioarcb->res_handle = res_handle;
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_SCSICDB;
ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_WRITE_NOT_READ;
ioarcb->cmd_pkt.cdb[0] = MODE_SELECT;
ioarcb->cmd_pkt.cdb[1] = parm;
ioarcb->cmd_pkt.cdb[4] = xfer_len;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_init_ioadl(ipr_cmd, dma_addr, xfer_len, IPR_IOADL_FLAGS_WRITE_LAST);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_ioafp_mode_select_page28 - Issue Mode Select Page 28 to IOA
* @ipr_cmd: ipr command struct
*
* This function sets up the SCSI bus attributes and sends
* a Mode Select for Page 28 to activate them.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_mode_select_page28(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_mode_pages *mode_pages = &ioa_cfg->vpd_cbs->mode_pages;
int length;
ENTER;
[SCSI] ipr: Fix adapter initialization failure Since scsi core is always sending scatterlists now, remove some code which was written with the bad assumption that a small transfer would not be sent down in a scatterlist. Without this fix, the ipr driver ends up sending garbage data to the adapter following a reset, causing it to fail the reset and take the adapter offline. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-08 19:57:42 -07:00
ipr_scsi_bus_speed_limit(ioa_cfg);
ipr_check_term_power(ioa_cfg, mode_pages);
ipr_modify_ioafp_mode_page_28(ioa_cfg, mode_pages);
length = mode_pages->hdr.length + 1;
mode_pages->hdr.length = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_build_mode_select(ipr_cmd, cpu_to_be32(IPR_IOA_RES_HANDLE), 0x11,
ioa_cfg->vpd_cbs_dma + offsetof(struct ipr_misc_cbs, mode_pages),
length);
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
ipr_cmd->job_step = ipr_set_supported_devs;
ipr_cmd->u.res = list_entry(ioa_cfg->used_res_q.next,
struct ipr_resource_entry, queue);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout, IPR_INTERNAL_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_build_mode_sense - Builds a mode sense command
* @ipr_cmd: ipr command struct
* @res: resource entry struct
* @parm: Byte 2 of mode sense command
* @dma_addr: DMA address of mode sense buffer
* @xfer_len: Size of DMA buffer
*
* Return value:
* none
**/
static void ipr_build_mode_sense(struct ipr_cmnd *ipr_cmd,
__be32 res_handle,
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
u8 parm, dma_addr_t dma_addr, u8 xfer_len)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
ioarcb->res_handle = res_handle;
ioarcb->cmd_pkt.cdb[0] = MODE_SENSE;
ioarcb->cmd_pkt.cdb[2] = parm;
ioarcb->cmd_pkt.cdb[4] = xfer_len;
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_SCSICDB;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_init_ioadl(ipr_cmd, dma_addr, xfer_len, IPR_IOADL_FLAGS_READ_LAST);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
/**
* ipr_reset_cmd_failed - Handle failure of IOA reset command
* @ipr_cmd: ipr command struct
*
* This function handles the failure of an IOA bringup command.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_reset_cmd_failed(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
dev_err(&ioa_cfg->pdev->dev,
"0x%02X failed with IOASC: 0x%08X\n",
ipr_cmd->ioarcb.cmd_pkt.cdb[0], ioasc);
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
return IPR_RC_JOB_RETURN;
}
/**
* ipr_reset_mode_sense_failed - Handle failure of IOAFP mode sense
* @ipr_cmd: ipr command struct
*
* This function handles the failure of a Mode Sense to the IOAFP.
* Some adapters do not handle all mode pages.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_mode_sense_failed(struct ipr_cmnd *ipr_cmd)
{
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
if (ioasc == IPR_IOASC_IR_INVALID_REQ_TYPE_OR_PKT) {
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
ipr_cmd->job_step = ipr_set_supported_devs;
ipr_cmd->u.res = list_entry(ioa_cfg->used_res_q.next,
struct ipr_resource_entry, queue);
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
return IPR_RC_JOB_CONTINUE;
}
return ipr_reset_cmd_failed(ipr_cmd);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_ioafp_mode_sense_page28 - Issue Mode Sense Page 28 to IOA
* @ipr_cmd: ipr command struct
*
* This function send a Page 28 mode sense to the IOA to
* retrieve SCSI bus attributes.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_mode_sense_page28(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
ipr_build_mode_sense(ipr_cmd, cpu_to_be32(IPR_IOA_RES_HANDLE),
0x28, ioa_cfg->vpd_cbs_dma +
offsetof(struct ipr_misc_cbs, mode_pages),
sizeof(struct ipr_mode_pages));
ipr_cmd->job_step = ipr_ioafp_mode_select_page28;
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
ipr_cmd->job_step_failed = ipr_reset_mode_sense_failed;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout, IPR_INTERNAL_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
/**
* ipr_ioafp_mode_select_page24 - Issue Mode Select to IOA
* @ipr_cmd: ipr command struct
*
* This function enables dual IOA RAID support if possible.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_mode_select_page24(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_mode_pages *mode_pages = &ioa_cfg->vpd_cbs->mode_pages;
struct ipr_mode_page24 *mode_page;
int length;
ENTER;
mode_page = ipr_get_mode_page(mode_pages, 0x24,
sizeof(struct ipr_mode_page24));
if (mode_page)
mode_page->flags |= IPR_ENABLE_DUAL_IOA_AF;
length = mode_pages->hdr.length + 1;
mode_pages->hdr.length = 0;
ipr_build_mode_select(ipr_cmd, cpu_to_be32(IPR_IOA_RES_HANDLE), 0x11,
ioa_cfg->vpd_cbs_dma + offsetof(struct ipr_misc_cbs, mode_pages),
length);
ipr_cmd->job_step = ipr_ioafp_mode_sense_page28;
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout, IPR_INTERNAL_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_reset_mode_sense_page24_failed - Handle failure of IOAFP mode sense
* @ipr_cmd: ipr command struct
*
* This function handles the failure of a Mode Sense to the IOAFP.
* Some adapters do not handle all mode pages.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_mode_sense_page24_failed(struct ipr_cmnd *ipr_cmd)
{
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
if (ioasc == IPR_IOASC_IR_INVALID_REQ_TYPE_OR_PKT) {
ipr_cmd->job_step = ipr_ioafp_mode_sense_page28;
return IPR_RC_JOB_CONTINUE;
}
return ipr_reset_cmd_failed(ipr_cmd);
}
/**
* ipr_ioafp_mode_sense_page24 - Issue Page 24 Mode Sense to IOA
* @ipr_cmd: ipr command struct
*
* This function send a mode sense to the IOA to retrieve
* the IOA Advanced Function Control mode page.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_mode_sense_page24(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
ipr_build_mode_sense(ipr_cmd, cpu_to_be32(IPR_IOA_RES_HANDLE),
0x24, ioa_cfg->vpd_cbs_dma +
offsetof(struct ipr_misc_cbs, mode_pages),
sizeof(struct ipr_mode_pages));
ipr_cmd->job_step = ipr_ioafp_mode_select_page24;
ipr_cmd->job_step_failed = ipr_reset_mode_sense_page24_failed;
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout, IPR_INTERNAL_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_init_res_table - Initialize the resource table
* @ipr_cmd: ipr command struct
*
* This function looks through the existing resource table, comparing
* it with the config table. This function will take care of old/new
* devices and schedule adding/removing them from the mid-layer
* as appropriate.
*
* Return value:
* IPR_RC_JOB_CONTINUE
**/
static int ipr_init_res_table(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_resource_entry *res, *temp;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
struct ipr_config_table_entry_wrapper cfgtew;
int entries, found, flag, i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LIST_HEAD(old_res);
ENTER;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (ioa_cfg->sis64)
flag = ioa_cfg->u.cfg_table64->hdr64.flags;
else
flag = ioa_cfg->u.cfg_table->hdr.flags;
if (flag & IPR_UCODE_DOWNLOAD_REQ)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_err(&ioa_cfg->pdev->dev, "Microcode download required\n");
list_for_each_entry_safe(res, temp, &ioa_cfg->used_res_q, queue)
list_move_tail(&res->queue, &old_res);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (ioa_cfg->sis64)
[SCSI] ipr: implement fixes for 64 bit adapter support Implement some small fixes for 64 bit support that were preventing the adapter from becoming operational. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:00 -06:00
entries = be16_to_cpu(ioa_cfg->u.cfg_table64->hdr64.num_entries);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
else
entries = ioa_cfg->u.cfg_table->hdr.num_entries;
for (i = 0; i < entries; i++) {
if (ioa_cfg->sis64)
cfgtew.u.cfgte64 = &ioa_cfg->u.cfg_table64->dev[i];
else
cfgtew.u.cfgte = &ioa_cfg->u.cfg_table->dev[i];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
found = 0;
list_for_each_entry_safe(res, temp, &old_res, queue) {
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (ipr_is_same_device(res, &cfgtew)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_move_tail(&res->queue, &ioa_cfg->used_res_q);
found = 1;
break;
}
}
if (!found) {
if (list_empty(&ioa_cfg->free_res_q)) {
dev_err(&ioa_cfg->pdev->dev, "Too many devices attached\n");
break;
}
found = 1;
res = list_entry(ioa_cfg->free_res_q.next,
struct ipr_resource_entry, queue);
list_move_tail(&res->queue, &ioa_cfg->used_res_q);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_init_res_entry(res, &cfgtew);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
res->add_to_ml = 1;
[SCSI] ipr: move setting of the allow_restart flag for vsets and disks A problem was found where the call to scsi_add_device() fails intermittently for an adapter. This is caused when __scsi_add_device() returns -ENODEV as a result of not calling scsi_probe_and_add_lun() since the call to scsi_host_scan_allowed() fails. scsi_host_scan_allowed() fails because the adapter state is set to SHOST_RECOVERY instead of SHOST_RUNNING. The state of the adapter is being set to SHOST_RECOVERY by scsi_eh_scmd_add() during error handling. This problem is avoided by moving the setting of the allow_restart flag to later in the device initialization sequence. This prevents further error handling if we get a NOT_READY response from a TUR command by causing scsi_check_sense() to return SUCCESS. Therefore, scsi_eh_scmd_add() will not run and the adapter state will remain as SHOST_RUNNING. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-10 15:46:34 -06:00
} else if (res->sdev && (ipr_is_vset_device(res) || ipr_is_scsi_disk(res)))
res->sdev->allow_restart = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (found)
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_update_res_entry(res, &cfgtew);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
list_for_each_entry_safe(res, temp, &old_res, queue) {
if (res->sdev) {
res->del_from_ml = 1;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
res->res_handle = IPR_INVALID_RES_HANDLE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_move_tail(&res->queue, &ioa_cfg->used_res_q);
}
}
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
list_for_each_entry_safe(res, temp, &old_res, queue) {
ipr_clear_res_target(res);
list_move_tail(&res->queue, &ioa_cfg->free_res_q);
}
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
if (ioa_cfg->dual_raid && ipr_dual_ioa_raid)
ipr_cmd->job_step = ipr_ioafp_mode_sense_page24;
else
ipr_cmd->job_step = ipr_ioafp_mode_sense_page28;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
return IPR_RC_JOB_CONTINUE;
}
/**
* ipr_ioafp_query_ioa_cfg - Send a Query IOA Config to the adapter.
* @ipr_cmd: ipr command struct
*
* This function sends a Query IOA Configuration command
* to the adapter to retrieve the IOA configuration table.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_query_ioa_cfg(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
struct ipr_inquiry_page3 *ucode_vpd = &ioa_cfg->vpd_cbs->page3_data;
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
struct ipr_inquiry_cap *cap = &ioa_cfg->vpd_cbs->cap;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
if (cap->cap & IPR_CAP_DUAL_IOA_RAID)
ioa_cfg->dual_raid = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_info(&ioa_cfg->pdev->dev, "Adapter firmware version: %02X%02X%02X%02X\n",
ucode_vpd->major_release, ucode_vpd->card_type,
ucode_vpd->minor_release[0], ucode_vpd->minor_release[1]);
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
ioarcb->res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ioarcb->cmd_pkt.cdb[0] = IPR_QUERY_IOA_CONFIG;
[SCSI] ipr: implement fixes for 64 bit adapter support Implement some small fixes for 64 bit support that were preventing the adapter from becoming operational. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:00 -06:00
ioarcb->cmd_pkt.cdb[6] = (ioa_cfg->cfg_table_size >> 16) & 0xff;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ioarcb->cmd_pkt.cdb[7] = (ioa_cfg->cfg_table_size >> 8) & 0xff;
ioarcb->cmd_pkt.cdb[8] = ioa_cfg->cfg_table_size & 0xff;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ipr_init_ioadl(ipr_cmd, ioa_cfg->cfg_table_dma, ioa_cfg->cfg_table_size,
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
IPR_IOADL_FLAGS_READ_LAST);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->job_step = ipr_init_res_table;
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout, IPR_INTERNAL_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
ipr: Issue Configure Cache Parameters command. Some new adapters require a special Configure Cache Parameters command to enable the adapter write cache, so send this during the adapter initialization if the adapter requires it. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:09 -07:00
static int ipr_ioa_service_action_failed(struct ipr_cmnd *ipr_cmd)
{
u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
if (ioasc == IPR_IOASC_IR_INVALID_REQ_TYPE_OR_PKT)
return IPR_RC_JOB_CONTINUE;
return ipr_reset_cmd_failed(ipr_cmd);
}
static void ipr_build_ioa_service_action(struct ipr_cmnd *ipr_cmd,
__be32 res_handle, u8 sa_code)
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
ioarcb->res_handle = res_handle;
ioarcb->cmd_pkt.cdb[0] = IPR_IOA_SERVICE_ACTION;
ioarcb->cmd_pkt.cdb[1] = sa_code;
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
}
/**
* ipr_ioafp_set_caching_parameters - Issue Set Cache parameters service
* action
*
* Return value:
* none
**/
static int ipr_ioafp_set_caching_parameters(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_inquiry_pageC4 *pageC4 = &ioa_cfg->vpd_cbs->pageC4_data;
ENTER;
ipr_cmd->job_step = ipr_ioafp_query_ioa_cfg;
if (pageC4->cache_cap[0] & IPR_CAP_SYNC_CACHE) {
ipr_build_ioa_service_action(ipr_cmd,
cpu_to_be32(IPR_IOA_RES_HANDLE),
IPR_IOA_SA_CHANGE_CACHE_PARAMS);
ioarcb->cmd_pkt.cdb[2] = 0x40;
ipr_cmd->job_step_failed = ipr_ioa_service_action_failed;
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout,
IPR_SET_SUP_DEVICE_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
LEAVE;
return IPR_RC_JOB_CONTINUE;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_ioafp_inquiry - Send an Inquiry to the adapter.
* @ipr_cmd: ipr command struct
*
* This utility function sends an inquiry to the adapter.
*
* Return value:
* none
**/
static void ipr_ioafp_inquiry(struct ipr_cmnd *ipr_cmd, u8 flags, u8 page,
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
dma_addr_t dma_addr, u8 xfer_len)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
ENTER;
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_SCSICDB;
ioarcb->res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ioarcb->cmd_pkt.cdb[0] = INQUIRY;
ioarcb->cmd_pkt.cdb[1] = flags;
ioarcb->cmd_pkt.cdb[2] = page;
ioarcb->cmd_pkt.cdb[4] = xfer_len;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_init_ioadl(ipr_cmd, dma_addr, xfer_len, IPR_IOADL_FLAGS_READ_LAST);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout, IPR_INTERNAL_TIMEOUT);
LEAVE;
}
[SCSI] ipr: Write caching state host attribute Adds a scsi_host sysfs attribute and module parm to enable/disable the write cache on an ipr adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:14 -07:00
/**
* ipr_inquiry_page_supported - Is the given inquiry page supported
* @page0: inquiry page 0 buffer
* @page: page code.
*
* This function determines if the specified inquiry page is supported.
*
* Return value:
* 1 if page is supported / 0 if not
**/
static int ipr_inquiry_page_supported(struct ipr_inquiry_page0 *page0, u8 page)
{
int i;
for (i = 0; i < min_t(u8, page0->len, IPR_INQUIRY_PAGE0_ENTRIES); i++)
if (page0->page[i] == page)
return 1;
return 0;
}
ipr: Inquiry IOA page 0xC4 during initialization. Add an IOA Inquiry command for Page 0xC4 during IOA initialization to collect cache capabilities, particularly to check if Sync IOA Write Cache is supported. Inquiry will happen right after Cap Inquiry on page 0xD0; and will execute only if the "Supported Pages" field in Inquiry Page 0x0 shows support for Page 0xC4. Otherwise, assume Sync IOA Write Cache is not supported. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:08 -07:00
/**
* ipr_ioafp_pageC4_inquiry - Send a Page 0xC4 Inquiry to the adapter.
* @ipr_cmd: ipr command struct
*
* This function sends a Page 0xC4 inquiry to the adapter
* to retrieve software VPD information.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_pageC4_inquiry(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_inquiry_page0 *page0 = &ioa_cfg->vpd_cbs->page0_data;
struct ipr_inquiry_pageC4 *pageC4 = &ioa_cfg->vpd_cbs->pageC4_data;
ENTER;
ipr: Issue Configure Cache Parameters command. Some new adapters require a special Configure Cache Parameters command to enable the adapter write cache, so send this during the adapter initialization if the adapter requires it. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:09 -07:00
ipr_cmd->job_step = ipr_ioafp_set_caching_parameters;
ipr: Inquiry IOA page 0xC4 during initialization. Add an IOA Inquiry command for Page 0xC4 during IOA initialization to collect cache capabilities, particularly to check if Sync IOA Write Cache is supported. Inquiry will happen right after Cap Inquiry on page 0xD0; and will execute only if the "Supported Pages" field in Inquiry Page 0x0 shows support for Page 0xC4. Otherwise, assume Sync IOA Write Cache is not supported. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:08 -07:00
memset(pageC4, 0, sizeof(*pageC4));
if (ipr_inquiry_page_supported(page0, 0xC4)) {
ipr_ioafp_inquiry(ipr_cmd, 1, 0xC4,
(ioa_cfg->vpd_cbs_dma
+ offsetof(struct ipr_misc_cbs,
pageC4_data)),
sizeof(struct ipr_inquiry_pageC4));
return IPR_RC_JOB_RETURN;
}
LEAVE;
return IPR_RC_JOB_CONTINUE;
}
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
/**
* ipr_ioafp_cap_inquiry - Send a Page 0xD0 Inquiry to the adapter.
* @ipr_cmd: ipr command struct
*
* This function sends a Page 0xD0 inquiry to the adapter
* to retrieve adapter capabilities.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_cap_inquiry(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_inquiry_page0 *page0 = &ioa_cfg->vpd_cbs->page0_data;
struct ipr_inquiry_cap *cap = &ioa_cfg->vpd_cbs->cap;
ENTER;
ipr: Inquiry IOA page 0xC4 during initialization. Add an IOA Inquiry command for Page 0xC4 during IOA initialization to collect cache capabilities, particularly to check if Sync IOA Write Cache is supported. Inquiry will happen right after Cap Inquiry on page 0xD0; and will execute only if the "Supported Pages" field in Inquiry Page 0x0 shows support for Page 0xC4. Otherwise, assume Sync IOA Write Cache is not supported. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:08 -07:00
ipr_cmd->job_step = ipr_ioafp_pageC4_inquiry;
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
memset(cap, 0, sizeof(*cap));
if (ipr_inquiry_page_supported(page0, 0xD0)) {
ipr_ioafp_inquiry(ipr_cmd, 1, 0xD0,
ioa_cfg->vpd_cbs_dma + offsetof(struct ipr_misc_cbs, cap),
sizeof(struct ipr_inquiry_cap));
return IPR_RC_JOB_RETURN;
}
LEAVE;
return IPR_RC_JOB_CONTINUE;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_ioafp_page3_inquiry - Send a Page 3 Inquiry to the adapter.
* @ipr_cmd: ipr command struct
*
* This function sends a Page 3 inquiry to the adapter
* to retrieve software VPD information.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_page3_inquiry(struct ipr_cmnd *ipr_cmd)
[SCSI] ipr: Write caching state host attribute Adds a scsi_host sysfs attribute and module parm to enable/disable the write cache on an ipr adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:14 -07:00
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
ipr_cmd->job_step = ipr_ioafp_cap_inquiry;
[SCSI] ipr: Write caching state host attribute Adds a scsi_host sysfs attribute and module parm to enable/disable the write cache on an ipr adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:14 -07:00
ipr_ioafp_inquiry(ipr_cmd, 1, 3,
ioa_cfg->vpd_cbs_dma + offsetof(struct ipr_misc_cbs, page3_data),
sizeof(struct ipr_inquiry_page3));
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_ioafp_page0_inquiry - Send a Page 0 Inquiry to the adapter.
* @ipr_cmd: ipr command struct
*
* This function sends a Page 0 inquiry to the adapter
* to retrieve supported inquiry pages.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_page0_inquiry(struct ipr_cmnd *ipr_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
char type[5];
ENTER;
/* Grab the type out of the VPD and store it away */
memcpy(type, ioa_cfg->vpd_cbs->ioa_vpd.std_inq_data.vpids.product_id, 4);
type[4] = '\0';
ioa_cfg->type = simple_strtoul((char *)type, NULL, 16);
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
if (ipr_invalid_adapter(ioa_cfg)) {
dev_err(&ioa_cfg->pdev->dev,
"Adapter not supported in this hardware configuration.\n");
if (!ipr_testmode) {
ioa_cfg->reset_retries += IPR_NUM_RESET_RELOAD_RETRIES;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
list_add_tail(&ipr_cmd->queue,
&ioa_cfg->hrrq->hrrq_free_q);
return IPR_RC_JOB_RETURN;
}
}
[SCSI] ipr: Write caching state host attribute Adds a scsi_host sysfs attribute and module parm to enable/disable the write cache on an ipr adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:14 -07:00
ipr_cmd->job_step = ipr_ioafp_page3_inquiry;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Write caching state host attribute Adds a scsi_host sysfs attribute and module parm to enable/disable the write cache on an ipr adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:14 -07:00
ipr_ioafp_inquiry(ipr_cmd, 1, 0,
ioa_cfg->vpd_cbs_dma + offsetof(struct ipr_misc_cbs, page0_data),
sizeof(struct ipr_inquiry_page0));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_ioafp_std_inquiry - Send a Standard Inquiry to the adapter.
* @ipr_cmd: ipr command struct
*
* This function sends a standard inquiry to the adapter.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_ioafp_std_inquiry(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
[SCSI] ipr: Write caching state host attribute Adds a scsi_host sysfs attribute and module parm to enable/disable the write cache on an ipr adapter. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:14 -07:00
ipr_cmd->job_step = ipr_ioafp_page0_inquiry;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_ioafp_inquiry(ipr_cmd, 0, 0,
ioa_cfg->vpd_cbs_dma + offsetof(struct ipr_misc_cbs, ioa_vpd),
sizeof(struct ipr_ioa_vpd));
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
* ipr_ioafp_identify_hrrq - Send Identify Host RRQ.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @ipr_cmd: ipr command struct
*
* This function send an Identify Host Request Response Queue
* command to establish the HRRQ with the adapter.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
static int ipr_ioafp_identify_hrrq(struct ipr_cmnd *ipr_cmd)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_cmd->job_step = ipr_ioafp_std_inquiry;
scsi: ipr: Remove redundant messages at adapter init time Whenever multiple HRRQs are enabled, which is the default setting now, we end up seeing the following message logged prior to initialization of each HRRQ: Starting IOA initialization sequence This results in 16 of these messages on most adapters, which serves little purpose. Change to just log this once. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-16 15:51:37 -06:00
if (ioa_cfg->identify_hrrq_index == 0)
dev_info(&ioa_cfg->pdev->dev, "Starting IOA initialization sequence.\n");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (ioa_cfg->identify_hrrq_index < ioa_cfg->hrrq_num) {
hrrq = &ioa_cfg->hrrq[ioa_cfg->identify_hrrq_index];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ioarcb->cmd_pkt.cdb[0] = IPR_ID_HOST_RR_Q;
ioarcb->res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ioarcb->cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
if (ioa_cfg->sis64)
ioarcb->cmd_pkt.cdb[1] = 0x1;
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
if (ioa_cfg->nvectors == 1)
ioarcb->cmd_pkt.cdb[1] &= ~IPR_ID_HRRQ_SELE_ENABLE;
else
ioarcb->cmd_pkt.cdb[1] |= IPR_ID_HRRQ_SELE_ENABLE;
ioarcb->cmd_pkt.cdb[2] =
((u64) hrrq->host_rrq_dma >> 24) & 0xff;
ioarcb->cmd_pkt.cdb[3] =
((u64) hrrq->host_rrq_dma >> 16) & 0xff;
ioarcb->cmd_pkt.cdb[4] =
((u64) hrrq->host_rrq_dma >> 8) & 0xff;
ioarcb->cmd_pkt.cdb[5] =
((u64) hrrq->host_rrq_dma) & 0xff;
ioarcb->cmd_pkt.cdb[7] =
((sizeof(u32) * hrrq->size) >> 8) & 0xff;
ioarcb->cmd_pkt.cdb[8] =
(sizeof(u32) * hrrq->size) & 0xff;
if (ioarcb->cmd_pkt.cdb[1] & IPR_ID_HRRQ_SELE_ENABLE)
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
ioarcb->cmd_pkt.cdb[9] =
ioa_cfg->identify_hrrq_index;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
if (ioa_cfg->sis64) {
ioarcb->cmd_pkt.cdb[10] =
((u64) hrrq->host_rrq_dma >> 56) & 0xff;
ioarcb->cmd_pkt.cdb[11] =
((u64) hrrq->host_rrq_dma >> 48) & 0xff;
ioarcb->cmd_pkt.cdb[12] =
((u64) hrrq->host_rrq_dma >> 40) & 0xff;
ioarcb->cmd_pkt.cdb[13] =
((u64) hrrq->host_rrq_dma >> 32) & 0xff;
}
if (ioarcb->cmd_pkt.cdb[1] & IPR_ID_HRRQ_SELE_ENABLE)
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
ioarcb->cmd_pkt.cdb[14] =
ioa_cfg->identify_hrrq_index;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout,
IPR_INTERNAL_TIMEOUT);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
if (++ioa_cfg->identify_hrrq_index < ioa_cfg->hrrq_num)
ipr_cmd->job_step = ipr_ioafp_identify_hrrq;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
LEAVE;
return IPR_RC_JOB_RETURN;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
return IPR_RC_JOB_CONTINUE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_reset_timer_done - Adapter reset timer function
* @ipr_cmd: ipr command struct
*
* Description: This function is used in adapter reset processing
* for timing events. If the reset_cmd pointer in the IOA
* config struct is not this adapter's we are doing nested
* resets and fail_all_ops will take care of freeing the
* command block.
*
* Return value:
* none
**/
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
static void ipr_reset_timer_done(struct timer_list *t)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
scsi: ipr: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Brian King <brking@us.ibm.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-18 17:53:24 -06:00
struct ipr_cmnd *ipr_cmd = from_timer(ipr_cmd, t, timer);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
unsigned long lock_flags = 0;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (ioa_cfg->reset_cmd == ipr_cmd) {
list_del(&ipr_cmd->queue);
ipr_cmd->done(ipr_cmd);
}
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
}
/**
* ipr_reset_start_timer - Start a timer for adapter reset job
* @ipr_cmd: ipr command struct
* @timeout: timeout value
*
* Description: This function is used in adapter reset processing
* for timing events. If the reset_cmd pointer in the IOA
* config struct is not this adapter's we are doing nested
* resets and fail_all_ops will take care of freeing the
* command block.
*
* Return value:
* none
**/
static void ipr_reset_start_timer(struct ipr_cmnd *ipr_cmd,
unsigned long timeout)
{
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ENTER;
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_pending_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->done = ipr_reset_ioa_job;
ipr_cmd->timer.expires = jiffies + timeout;
treewide: Remove TIMER_FUNC_TYPE and TIMER_DATA_TYPE casts With all callbacks converted, and the timer callback prototype switched over, the TIMER_FUNC_TYPE cast is no longer needed, so remove it. Conversion was done with the following scripts: perl -pi -e 's|\(TIMER_FUNC_TYPE\)||g' \ $(git grep TIMER_FUNC_TYPE | cut -d: -f1 | sort -u) perl -pi -e 's|\(TIMER_DATA_TYPE\)||g' \ $(git grep TIMER_DATA_TYPE | cut -d: -f1 | sort -u) The now unused macros are also dropped from include/linux/timer.h. Signed-off-by: Kees Cook <keescook@chromium.org>
2017-10-23 01:40:42 -06:00
ipr_cmd->timer.function = ipr_reset_timer_done;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
add_timer(&ipr_cmd->timer);
}
/**
* ipr_init_ioa_mem - Initialize ioa_cfg control block
* @ioa_cfg: ioa cfg struct
*
* Return value:
* nothing
**/
static void ipr_init_ioa_mem(struct ipr_ioa_cfg *ioa_cfg)
{
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
struct ipr_hrr_queue *hrrq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for_each_hrrq(hrrq, ioa_cfg) {
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock(&hrrq->_lock);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
memset(hrrq->host_rrq, 0, sizeof(u32) * hrrq->size);
/* Initialize Host RRQ pointers */
hrrq->hrrq_start = hrrq->host_rrq;
hrrq->hrrq_end = &hrrq->host_rrq[hrrq->size - 1];
hrrq->hrrq_curr = hrrq->hrrq_start;
hrrq->toggle_bit = 1;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock(&hrrq->_lock);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
wmb();
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
ioa_cfg->identify_hrrq_index = 0;
if (ioa_cfg->hrrq_num == 1)
atomic_set(&ioa_cfg->hrrq_index, 0);
else
atomic_set(&ioa_cfg->hrrq_index, 1);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/* Zero out config table */
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
memset(ioa_cfg->u.cfg_table, 0, ioa_cfg->cfg_table_size);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
/**
* ipr_reset_next_stage - Process IPL stage change based on feedback register.
* @ipr_cmd: ipr command struct
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_next_stage(struct ipr_cmnd *ipr_cmd)
{
unsigned long stage, stage_time;
u32 feedback;
volatile u32 int_reg;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
u64 maskval = 0;
feedback = readl(ioa_cfg->regs.init_feedback_reg);
stage = feedback & IPR_IPL_INIT_STAGE_MASK;
stage_time = feedback & IPR_IPL_INIT_STAGE_TIME_MASK;
ipr_dbg("IPL stage = 0x%lx, IPL stage time = %ld\n", stage, stage_time);
/* sanity check the stage_time value */
[SCSI] ipr: implement fixes for 64 bit adapter support Implement some small fixes for 64 bit support that were preventing the adapter from becoming operational. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:00 -06:00
if (stage_time == 0)
stage_time = IPR_IPL_INIT_DEFAULT_STAGE_TIME;
else if (stage_time < IPR_IPL_INIT_MIN_STAGE_TIME)
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
stage_time = IPR_IPL_INIT_MIN_STAGE_TIME;
else if (stage_time > IPR_LONG_OPERATIONAL_TIMEOUT)
stage_time = IPR_LONG_OPERATIONAL_TIMEOUT;
if (stage == IPR_IPL_INIT_STAGE_UNKNOWN) {
writel(IPR_PCII_IPL_STAGE_CHANGE, ioa_cfg->regs.set_interrupt_mask_reg);
int_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg);
stage_time = ioa_cfg->transop_timeout;
ipr_cmd->job_step = ipr_ioafp_identify_hrrq;
} else if (stage == IPR_IPL_INIT_STAGE_TRANSOP) {
[SCSI] ipr: fix transition to operational for new adapters The method of transitioning to operational for new adapters includes using initialization stages. The current stage is indicated via a register read. The final good stage in the sequence is "operational" but does not necessarily indicate that the driver can proceed. There is another bit that gets set in the adapter->host interrupt register when the adapter has completed enough of its bringup such that it can accept commands. The driver was not checking that bit before proceeding which led to intermittent errors and adapter resets. The fix is to check the "transition to operational" bit in the interrupt register after detecting that the initialization stage is "operational" and only proceed if both are set. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-07-14 11:49:43 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg32);
if (int_reg & IPR_PCII_IOA_TRANS_TO_OPER) {
ipr_cmd->job_step = ipr_ioafp_identify_hrrq;
maskval = IPR_PCII_IPL_STAGE_CHANGE;
maskval = (maskval << 32) | IPR_PCII_IOA_TRANS_TO_OPER;
writeq(maskval, ioa_cfg->regs.set_interrupt_mask_reg);
int_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg);
return IPR_RC_JOB_CONTINUE;
}
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
}
ipr_cmd->timer.expires = jiffies + stage_time * HZ;
treewide: Remove TIMER_FUNC_TYPE and TIMER_DATA_TYPE casts With all callbacks converted, and the timer callback prototype switched over, the TIMER_FUNC_TYPE cast is no longer needed, so remove it. Conversion was done with the following scripts: perl -pi -e 's|\(TIMER_FUNC_TYPE\)||g' \ $(git grep TIMER_FUNC_TYPE | cut -d: -f1 | sort -u) perl -pi -e 's|\(TIMER_DATA_TYPE\)||g' \ $(git grep TIMER_DATA_TYPE | cut -d: -f1 | sort -u) The now unused macros are also dropped from include/linux/timer.h. Signed-off-by: Kees Cook <keescook@chromium.org>
2017-10-23 01:40:42 -06:00
ipr_cmd->timer.function = ipr_oper_timeout;
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
ipr_cmd->done = ipr_reset_ioa_job;
add_timer(&ipr_cmd->timer);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_pending_q);
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
return IPR_RC_JOB_RETURN;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_reset_enable_ioa - Enable the IOA following a reset.
* @ipr_cmd: ipr command struct
*
* This function reinitializes some control blocks and
* enables destructive diagnostics on the adapter.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_reset_enable_ioa(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
volatile u32 int_reg;
[SCSI] ipr: fix a register read to use the correct address for 64 bit adapters Fix ipr_reset_enable_ioa() to read the correct IOA to host interrupt register address for 64 bit adapters. We need to read the lower 32 bits, not the upper 32 bits. Also change the write of the 64 bit mask value to a single writeq instead of two writel calls. Finally, use the correct u8 type for the type field in the ipr_resource_entry structure. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:14:07 -06:00
volatile u64 maskval;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
int i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
ipr_cmd->job_step = ipr_ioafp_identify_hrrq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_init_ioa_mem(ioa_cfg);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].allow_interrupts = 1;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
[SCSI] ipr: add endian swap enablement for 64 bit adapters A change in the hardware design of the chip for the new adapters changes the default endianness of MMIO operations. This patch adds a register definition which when written to with a predefined value will change the endianness back to what the driver expects. This patch also fixes two problems found during testing. First, the first reserved field in the ipr_hostrcb64_fabirc_desc structure only reserved one byte. The correct amount to reserve is 2 bytes. Second, the reserved field of the ipr_hostrcb64_error structure only reserved 2 bytes. The correct amount to reserve is 16 bytes. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-04 11:26:50 -06:00
if (ioa_cfg->sis64) {
/* Set the adapter to the correct endian mode. */
writel(IPR_ENDIAN_SWAP_KEY, ioa_cfg->regs.endian_swap_reg);
int_reg = readl(ioa_cfg->regs.endian_swap_reg);
}
[SCSI] ipr: fix a register read to use the correct address for 64 bit adapters Fix ipr_reset_enable_ioa() to read the correct IOA to host interrupt register address for 64 bit adapters. We need to read the lower 32 bits, not the upper 32 bits. Also change the write of the 64 bit mask value to a single writeq instead of two writel calls. Finally, use the correct u8 type for the type field in the ipr_resource_entry structure. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:14:07 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (int_reg & IPR_PCII_IOA_TRANS_TO_OPER) {
writel((IPR_PCII_ERROR_INTERRUPTS | IPR_PCII_HRRQ_UPDATED),
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
ioa_cfg->regs.clr_interrupt_mask_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg);
return IPR_RC_JOB_CONTINUE;
}
/* Enable destructive diagnostics on IOA */
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
writel(ioa_cfg->doorbell, ioa_cfg->regs.set_uproc_interrupt_reg32);
[SCSI] ipr: fix a register read to use the correct address for 64 bit adapters Fix ipr_reset_enable_ioa() to read the correct IOA to host interrupt register address for 64 bit adapters. We need to read the lower 32 bits, not the upper 32 bits. Also change the write of the 64 bit mask value to a single writeq instead of two writel calls. Finally, use the correct u8 type for the type field in the ipr_resource_entry structure. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:14:07 -06:00
if (ioa_cfg->sis64) {
maskval = IPR_PCII_IPL_STAGE_CHANGE;
maskval = (maskval << 32) | IPR_PCII_OPER_INTERRUPTS;
writeq(maskval, ioa_cfg->regs.clr_interrupt_mask_reg);
} else
writel(IPR_PCII_OPER_INTERRUPTS, ioa_cfg->regs.clr_interrupt_mask_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg);
dev_info(&ioa_cfg->pdev->dev, "Initializing IOA.\n");
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
if (ioa_cfg->sis64) {
ipr_cmd->job_step = ipr_reset_next_stage;
return IPR_RC_JOB_CONTINUE;
}
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
ipr_cmd->timer.expires = jiffies + (ioa_cfg->transop_timeout * HZ);
treewide: Remove TIMER_FUNC_TYPE and TIMER_DATA_TYPE casts With all callbacks converted, and the timer callback prototype switched over, the TIMER_FUNC_TYPE cast is no longer needed, so remove it. Conversion was done with the following scripts: perl -pi -e 's|\(TIMER_FUNC_TYPE\)||g' \ $(git grep TIMER_FUNC_TYPE | cut -d: -f1 | sort -u) perl -pi -e 's|\(TIMER_DATA_TYPE\)||g' \ $(git grep TIMER_DATA_TYPE | cut -d: -f1 | sort -u) The now unused macros are also dropped from include/linux/timer.h. Signed-off-by: Kees Cook <keescook@chromium.org>
2017-10-23 01:40:42 -06:00
ipr_cmd->timer.function = ipr_oper_timeout;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->done = ipr_reset_ioa_job;
add_timer(&ipr_cmd->timer);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_pending_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_reset_wait_for_dump - Wait for a dump to timeout.
* @ipr_cmd: ipr command struct
*
* This function is invoked when an adapter dump has run out
* of processing time.
*
* Return value:
* IPR_RC_JOB_CONTINUE
**/
static int ipr_reset_wait_for_dump(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
if (ioa_cfg->sdt_state == GET_DUMP)
[SCSI] ipr: Stop reading adapter dump prematurely When the ipr driver decides to dump the adapter, it changes the sdt_state to GET_DUMP, then prepares the adapter so that the dump can be read. However, if the ipr worker thread wakes up for some reason before the driver has put the adapter in a state where it can succesfully dump the adapter, the driver will start dumping the adapter too early, which can potentially trigger a BUG check in the pci config blocking API. Fix this by adding a new sdt_state to differentiate between the ipr driver wanting to dump the adapter in the near future and wanting to dump the adapter now. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-09-21 07:51:11 -06:00
ioa_cfg->sdt_state = WAIT_FOR_DUMP;
else if (ioa_cfg->sdt_state == READ_DUMP)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->sdt_state = ABORT_DUMP;
[SCSI] ipr: Fix BUG on adapter dump timeout If an adapter dump times out, the ipr driver will abort the dump and proceed to reset and recover the adapter. When an adapter dump completes, the work thread which is reading the adapter dump will initiate an adapter reset to recover the adapter. However, when the adapter dump gets aborted, the work thread should not initiate an adapter reset, since an adapter reset is already in progress. This fixes a case of calling pci_block_user_cfg_access overlapped, which results in a BUG. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-10-15 08:08:56 -06:00
ioa_cfg->dump_timeout = 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->job_step = ipr_reset_alert;
return IPR_RC_JOB_CONTINUE;
}
/**
* ipr_unit_check_no_data - Log a unit check/no data error log
* @ioa_cfg: ioa config struct
*
* Logs an error indicating the adapter unit checked, but for some
* reason, we were unable to fetch the unit check buffer.
*
* Return value:
* nothing
**/
static void ipr_unit_check_no_data(struct ipr_ioa_cfg *ioa_cfg)
{
ioa_cfg->errors_logged++;
dev_err(&ioa_cfg->pdev->dev, "IOA unit check with no data\n");
}
/**
* ipr_get_unit_check_buffer - Get the unit check buffer from the IOA
* @ioa_cfg: ioa config struct
*
* Fetches the unit check buffer from the adapter by clocking the data
* through the mailbox register.
*
* Return value:
* nothing
**/
static void ipr_get_unit_check_buffer(struct ipr_ioa_cfg *ioa_cfg)
{
unsigned long mailbox;
struct ipr_hostrcb *hostrcb;
struct ipr_uc_sdt sdt;
int rc, length;
[SCSI] ipr: Handle IOA reset request In ipr dual adapter configurations, the ipr adapter firmware may require an adapter reset for various reasons. The reset is requested by the adapter firmware logging an error with an IOASC of 0x02048000. Add support to log this error, and reset the adapter. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:12 -06:00
u32 ioasc;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
mailbox = readl(ioa_cfg->ioa_mailbox);
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
if (!ioa_cfg->sis64 && !ipr_sdt_is_fmt2(mailbox)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_unit_check_no_data(ioa_cfg);
return;
}
memset(&sdt, 0, sizeof(struct ipr_uc_sdt));
rc = ipr_get_ldump_data_section(ioa_cfg, mailbox, (__be32 *) &sdt,
(sizeof(struct ipr_uc_sdt)) / sizeof(__be32));
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
if (rc || !(sdt.entry[0].flags & IPR_SDT_VALID_ENTRY) ||
((be32_to_cpu(sdt.hdr.state) != IPR_FMT3_SDT_READY_TO_USE) &&
(be32_to_cpu(sdt.hdr.state) != IPR_FMT2_SDT_READY_TO_USE))) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_unit_check_no_data(ioa_cfg);
return;
}
/* Find length of the first sdt entry (UC buffer) */
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
if (be32_to_cpu(sdt.hdr.state) == IPR_FMT3_SDT_READY_TO_USE)
length = be32_to_cpu(sdt.entry[0].end_token);
else
length = (be32_to_cpu(sdt.entry[0].end_token) -
be32_to_cpu(sdt.entry[0].start_token)) &
IPR_FMT2_MBX_ADDR_MASK;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
hostrcb = list_entry(ioa_cfg->hostrcb_free_q.next,
struct ipr_hostrcb, queue);
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
list_del_init(&hostrcb->queue);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
memset(&hostrcb->hcam, 0, sizeof(hostrcb->hcam));
rc = ipr_get_ldump_data_section(ioa_cfg,
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
be32_to_cpu(sdt.entry[0].start_token),
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
(__be32 *)&hostrcb->hcam,
min(length, (int)sizeof(hostrcb->hcam)) / sizeof(__be32));
[SCSI] ipr: Handle IOA reset request In ipr dual adapter configurations, the ipr adapter firmware may require an adapter reset for various reasons. The reset is requested by the adapter firmware logging an error with an IOASC of 0x02048000. Add support to log this error, and reset the adapter. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:12 -06:00
if (!rc) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_handle_log_data(ioa_cfg, hostrcb);
[SCSI] ipr: add error handling updates for the next generation chip Add support for the new log data notification and overlay IDs. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:07 -07:00
ioasc = be32_to_cpu(hostrcb->hcam.u.error.fd_ioasc);
[SCSI] ipr: Handle IOA reset request In ipr dual adapter configurations, the ipr adapter firmware may require an adapter reset for various reasons. The reset is requested by the adapter firmware logging an error with an IOASC of 0x02048000. Add support to log this error, and reset the adapter. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:12 -06:00
if (ioasc == IPR_IOASC_NR_IOA_RESET_REQUIRED &&
ioa_cfg->sdt_state == GET_DUMP)
ioa_cfg->sdt_state = WAIT_FOR_DUMP;
} else
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_unit_check_no_data(ioa_cfg);
list_add_tail(&hostrcb->queue, &ioa_cfg->hostrcb_free_q);
}
[SCSI] ipr: fix mailbox register definition and add a delay before reading The definition for the mailbox register for new adapters was incorrect. The value has been updated to the correct offset. After an adapter reset, the mailbox register on the new adapters takes a number of seconds to stabilize. A delay has been added before reading the register. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-11-04 10:36:16 -06:00
/**
* ipr_reset_get_unit_check_job - Call to get the unit check buffer.
* @ipr_cmd: ipr command struct
*
* Description: This function will call to get the unit check buffer.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_reset_get_unit_check_job(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
ioa_cfg->ioa_unit_checked = 0;
ipr_get_unit_check_buffer(ioa_cfg);
ipr_cmd->job_step = ipr_reset_alert;
ipr_reset_start_timer(ipr_cmd, 0);
LEAVE;
return IPR_RC_JOB_RETURN;
}
ipr: Add delay to ensure coherent dumps. Add a holding pattern prior to collecting dump data, to wait for the IOA indication that the Mailbox register is stable and won't change without an explicit reset. This ensures we'll be collecting meaningful dump data, even when dumping right after an adapter reset. In the event of a timeout, we still force the dump, since a partial dump still might be useful. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:06 -07:00
static int ipr_dump_mailbox_wait(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
if (ioa_cfg->sdt_state != GET_DUMP)
return IPR_RC_JOB_RETURN;
if (!ioa_cfg->sis64 || !ipr_cmd->u.time_left ||
(readl(ioa_cfg->regs.sense_interrupt_reg) &
IPR_PCII_MAILBOX_STABLE)) {
if (!ipr_cmd->u.time_left)
dev_err(&ioa_cfg->pdev->dev,
"Timed out waiting for Mailbox register.\n");
ioa_cfg->sdt_state = READ_DUMP;
ioa_cfg->dump_timeout = 0;
if (ioa_cfg->sis64)
ipr_reset_start_timer(ipr_cmd, IPR_SIS64_DUMP_TIMEOUT);
else
ipr_reset_start_timer(ipr_cmd, IPR_SIS32_DUMP_TIMEOUT);
ipr_cmd->job_step = ipr_reset_wait_for_dump;
schedule_work(&ioa_cfg->work_q);
} else {
ipr_cmd->u.time_left -= IPR_CHECK_FOR_RESET_TIMEOUT;
ipr_reset_start_timer(ipr_cmd,
IPR_CHECK_FOR_RESET_TIMEOUT);
}
LEAVE;
return IPR_RC_JOB_RETURN;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_reset_restore_cfg_space - Restore PCI config space.
* @ipr_cmd: ipr command struct
*
* Description: This function restores the saved PCI config space of
* the adapter, fails all outstanding ops back to the callers, and
* fetches the dump/unit check if applicable to this reset.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_restore_cfg_space(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
[SCSI] ipr: remove unneeded volatile declarations This patch removes three volatile declarations based on some feedback and code analysis. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-07 13:12:30 -06:00
u32 int_reg;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: fix EEH recovery After commits c82f63e411f1b58427c103bd95af2863b1c96dd1 (PCI: check saved state before restore) and 4b77b0a2ba27d64f58f16d8d4d48d8319dda36ff (PCI: Clear saved_state after the state has been restored) PCI drivers are prevented from restoring the device standard configuration registers twice in a row. These changes introduced a regression on ipr EEH recovery. The ipr device driver saves the PCI state only during the device probe and restores it on ipr_reset_restore_cfg_space() during IOA resets. This behavior is causing the EEH recovery to fail after the second error detected, since the registers are not being restored. One possible solution would be saving the registers after restoring them. The problem with this approach is that while recovering from an EEH error if pci_save_state() results in an EEH error, the adapter/slot will be reset, and end up back in ipr_reset_restore_cfg_space(), but it won't have a valid saved state to restore, so pci_restore_state() will fail. The following patch introduces a workaround for this problem, hacking around the PCI API by setting pdev->state_saved = true before we do the restore. It fixes the EEH regression and prevents that we hit another EEH error during EEH recovery. [jejb: fix is a hack ... Jesse and Rafael will fix properly] Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Cc: Jesse Barnes <jbarnes@virtuousgeek.org> Cc: stable@kernel.org Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2009-11-25 15:13:43 -07:00
ioa_cfg->pdev->state_saved = true;
PCI: make pci_restore_state return void pci_restore_state only ever returns 0, thus there is no benefit in having it return any value. Also, a large majority of the callers do not check the return code of pci_restore_state. Make the pci_restore_state a void return and avoid the overhead. Acked-by: Mauro Carvalho Chehab <mchehab@redhat.com> Signed-off-by: Jon Mason <jon.mason@exar.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2010-11-30 16:43:26 -07:00
pci_restore_state(ioa_cfg->pdev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ipr_set_pcix_cmd_reg(ioa_cfg)) {
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ipr_cmd->s.ioasa.hdr.ioasc = cpu_to_be32(IPR_IOASC_PCI_ACCESS_ERROR);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return IPR_RC_JOB_CONTINUE;
}
ipr_fail_all_ops(ioa_cfg);
[SCSI] ipr: add endian swap enablement for 64 bit adapters A change in the hardware design of the chip for the new adapters changes the default endianness of MMIO operations. This patch adds a register definition which when written to with a predefined value will change the endianness back to what the driver expects. This patch also fixes two problems found during testing. First, the first reserved field in the ipr_hostrcb64_fabirc_desc structure only reserved one byte. The correct amount to reserve is 2 bytes. Second, the reserved field of the ipr_hostrcb64_error structure only reserved 2 bytes. The correct amount to reserve is 16 bytes. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-04 11:26:50 -06:00
if (ioa_cfg->sis64) {
/* Set the adapter to the correct endian mode. */
writel(IPR_ENDIAN_SWAP_KEY, ioa_cfg->regs.endian_swap_reg);
int_reg = readl(ioa_cfg->regs.endian_swap_reg);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioa_cfg->ioa_unit_checked) {
[SCSI] ipr: fix mailbox register definition and add a delay before reading The definition for the mailbox register for new adapters was incorrect. The value has been updated to the correct offset. After an adapter reset, the mailbox register on the new adapters takes a number of seconds to stabilize. A delay has been added before reading the register. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-11-04 10:36:16 -06:00
if (ioa_cfg->sis64) {
ipr_cmd->job_step = ipr_reset_get_unit_check_job;
ipr_reset_start_timer(ipr_cmd, IPR_DUMP_DELAY_TIMEOUT);
return IPR_RC_JOB_RETURN;
} else {
ioa_cfg->ioa_unit_checked = 0;
ipr_get_unit_check_buffer(ioa_cfg);
ipr_cmd->job_step = ipr_reset_alert;
ipr_reset_start_timer(ipr_cmd, 0);
return IPR_RC_JOB_RETURN;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
if (ioa_cfg->in_ioa_bringdown) {
ipr_cmd->job_step = ipr_ioa_bringdown_done;
ipr: Add delay to ensure coherent dumps. Add a holding pattern prior to collecting dump data, to wait for the IOA indication that the Mailbox register is stable and won't change without an explicit reset. This ensures we'll be collecting meaningful dump data, even when dumping right after an adapter reset. In the event of a timeout, we still force the dump, since a partial dump still might be useful. Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2015-11-03 11:26:06 -07:00
} else if (ioa_cfg->sdt_state == GET_DUMP) {
ipr_cmd->job_step = ipr_dump_mailbox_wait;
ipr_cmd->u.time_left = IPR_WAIT_FOR_MAILBOX;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
} else {
ipr_cmd->job_step = ipr_reset_enable_ioa;
}
[SCSI] ipr: implement fixes for 64 bit adapter support Implement some small fixes for 64 bit support that were preventing the adapter from becoming operational. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:00 -06:00
LEAVE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return IPR_RC_JOB_CONTINUE;
}
[SCSI] ipr: PCI error recovery fix Since the pci_block_user_cfg_access API was modified to track block/unblocks, it was discovered that the ipr driver had a path through its code (in PCI error recovery) which would unblock when not previously blocked. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:37 -07:00
/**
* ipr_reset_bist_done - BIST has completed on the adapter.
* @ipr_cmd: ipr command struct
*
* Description: Unblock config space and resume the reset process.
*
* Return value:
* IPR_RC_JOB_CONTINUE
**/
static int ipr_reset_bist_done(struct ipr_cmnd *ipr_cmd)
{
PCI: Rework config space blocking services pci_block_user_cfg_access was designed for the use case that a single context, the IPR driver, temporarily delays user space accesses to the config space via sysfs. This assumption became invalid by the time pci_dev_reset was added as locking instance. Today, if you run two loops in parallel that reset the same device via sysfs, you end up with a kernel BUG as pci_block_user_cfg_access detect the broken assumption. This reworks the pci_block_user_cfg_access to a sleeping service pci_cfg_access_lock and an atomic-compatible variant called pci_cfg_access_trylock. The former not only blocks user space access as before but also waits if access was already locked. The latter service just returns false in this case, allowing the caller to resolve the conflict instead of raising a BUG. Adaptions of the ipr driver were originally written by Brian King. Acked-by: Brian King <brking@linux.vnet.ibm.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-11-04 02:45:59 -06:00
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
[SCSI] ipr: PCI error recovery fix Since the pci_block_user_cfg_access API was modified to track block/unblocks, it was discovered that the ipr driver had a path through its code (in PCI error recovery) which would unblock when not previously blocked. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:37 -07:00
ENTER;
PCI: Rework config space blocking services pci_block_user_cfg_access was designed for the use case that a single context, the IPR driver, temporarily delays user space accesses to the config space via sysfs. This assumption became invalid by the time pci_dev_reset was added as locking instance. Today, if you run two loops in parallel that reset the same device via sysfs, you end up with a kernel BUG as pci_block_user_cfg_access detect the broken assumption. This reworks the pci_block_user_cfg_access to a sleeping service pci_cfg_access_lock and an atomic-compatible variant called pci_cfg_access_trylock. The former not only blocks user space access as before but also waits if access was already locked. The latter service just returns false in this case, allowing the caller to resolve the conflict instead of raising a BUG. Adaptions of the ipr driver were originally written by Brian King. Acked-by: Brian King <brking@linux.vnet.ibm.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-11-04 02:45:59 -06:00
if (ioa_cfg->cfg_locked)
pci_cfg_access_unlock(ioa_cfg->pdev);
ioa_cfg->cfg_locked = 0;
[SCSI] ipr: PCI error recovery fix Since the pci_block_user_cfg_access API was modified to track block/unblocks, it was discovered that the ipr driver had a path through its code (in PCI error recovery) which would unblock when not previously blocked. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:37 -07:00
ipr_cmd->job_step = ipr_reset_restore_cfg_space;
LEAVE;
return IPR_RC_JOB_CONTINUE;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_reset_start_bist - Run BIST on the adapter.
* @ipr_cmd: ipr command struct
*
* Description: This function runs BIST on the adapter, then delays 2 seconds.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_start_bist(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
[SCSI] ipr: add MMIO write to perform BIST for 64 bit adapters The 64 bit chip used in new adapters does not properly support the BIST register in PCI config space. This patch implements an alternative MMIO write reset method. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-17 12:51:40 -06:00
int rc = PCIBIOS_SUCCESSFUL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
[SCSI] ipr: add MMIO write to perform BIST for 64 bit adapters The 64 bit chip used in new adapters does not properly support the BIST register in PCI config space. This patch implements an alternative MMIO write reset method. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-17 12:51:40 -06:00
if (ioa_cfg->ipr_chip->bist_method == IPR_MMIO)
writel(IPR_UPROCI_SIS64_START_BIST,
ioa_cfg->regs.set_uproc_interrupt_reg32);
else
rc = pci_write_config_byte(ioa_cfg->pdev, PCI_BIST, PCI_BIST_START);
if (rc == PCIBIOS_SUCCESSFUL) {
[SCSI] ipr: PCI error recovery fix Since the pci_block_user_cfg_access API was modified to track block/unblocks, it was discovered that the ipr driver had a path through its code (in PCI error recovery) which would unblock when not previously blocked. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:37 -07:00
ipr_cmd->job_step = ipr_reset_bist_done;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_reset_start_timer(ipr_cmd, IPR_WAIT_FOR_BIST_TIMEOUT);
rc = IPR_RC_JOB_RETURN;
[SCSI] ipr: add MMIO write to perform BIST for 64 bit adapters The 64 bit chip used in new adapters does not properly support the BIST register in PCI config space. This patch implements an alternative MMIO write reset method. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-17 12:51:40 -06:00
} else {
PCI: Rework config space blocking services pci_block_user_cfg_access was designed for the use case that a single context, the IPR driver, temporarily delays user space accesses to the config space via sysfs. This assumption became invalid by the time pci_dev_reset was added as locking instance. Today, if you run two loops in parallel that reset the same device via sysfs, you end up with a kernel BUG as pci_block_user_cfg_access detect the broken assumption. This reworks the pci_block_user_cfg_access to a sleeping service pci_cfg_access_lock and an atomic-compatible variant called pci_cfg_access_trylock. The former not only blocks user space access as before but also waits if access was already locked. The latter service just returns false in this case, allowing the caller to resolve the conflict instead of raising a BUG. Adaptions of the ipr driver were originally written by Brian King. Acked-by: Brian King <brking@linux.vnet.ibm.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-11-04 02:45:59 -06:00
if (ioa_cfg->cfg_locked)
pci_cfg_access_unlock(ipr_cmd->ioa_cfg->pdev);
ioa_cfg->cfg_locked = 0;
[SCSI] ipr: add MMIO write to perform BIST for 64 bit adapters The 64 bit chip used in new adapters does not properly support the BIST register in PCI config space. This patch implements an alternative MMIO write reset method. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-17 12:51:40 -06:00
ipr_cmd->s.ioasa.hdr.ioasc = cpu_to_be32(IPR_IOASC_PCI_ACCESS_ERROR);
rc = IPR_RC_JOB_CONTINUE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
LEAVE;
return rc;
}
[SCSI] ipr: Use PCI-E reset API for new ipr adapter Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-07 16:09:05 -06:00
/**
* ipr_reset_slot_reset_done - Clear PCI reset to the adapter
* @ipr_cmd: ipr command struct
*
* Description: This clears PCI reset to the adapter and delays two seconds.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_reset_slot_reset_done(struct ipr_cmnd *ipr_cmd)
{
ENTER;
ipr_cmd->job_step = ipr_reset_bist_done;
ipr_reset_start_timer(ipr_cmd, IPR_WAIT_FOR_BIST_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
/**
* ipr_reset_reset_work - Pulse a PCIe fundamental reset
* @work: work struct
*
* Description: This pulses warm reset to a slot.
*
**/
static void ipr_reset_reset_work(struct work_struct *work)
{
struct ipr_cmnd *ipr_cmd = container_of(work, struct ipr_cmnd, work);
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct pci_dev *pdev = ioa_cfg->pdev;
unsigned long lock_flags = 0;
ENTER;
pci_set_pcie_reset_state(pdev, pcie_warm_reset);
msleep(jiffies_to_msecs(IPR_PCI_RESET_TIMEOUT));
pci_set_pcie_reset_state(pdev, pcie_deassert_reset);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
if (ioa_cfg->reset_cmd == ipr_cmd)
ipr_reset_ioa_job(ipr_cmd);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
LEAVE;
}
[SCSI] ipr: Use PCI-E reset API for new ipr adapter Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-07 16:09:05 -06:00
/**
* ipr_reset_slot_reset - Reset the PCI slot of the adapter.
* @ipr_cmd: ipr command struct
*
* Description: This asserts PCI reset to the adapter.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_reset_slot_reset(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
INIT_WORK(&ipr_cmd->work, ipr_reset_reset_work);
queue_work(ioa_cfg->reset_work_q, &ipr_cmd->work);
[SCSI] ipr: Use PCI-E reset API for new ipr adapter Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-07 16:09:05 -06:00
ipr_cmd->job_step = ipr_reset_slot_reset_done;
LEAVE;
return IPR_RC_JOB_RETURN;
}
PCI: Rework config space blocking services pci_block_user_cfg_access was designed for the use case that a single context, the IPR driver, temporarily delays user space accesses to the config space via sysfs. This assumption became invalid by the time pci_dev_reset was added as locking instance. Today, if you run two loops in parallel that reset the same device via sysfs, you end up with a kernel BUG as pci_block_user_cfg_access detect the broken assumption. This reworks the pci_block_user_cfg_access to a sleeping service pci_cfg_access_lock and an atomic-compatible variant called pci_cfg_access_trylock. The former not only blocks user space access as before but also waits if access was already locked. The latter service just returns false in this case, allowing the caller to resolve the conflict instead of raising a BUG. Adaptions of the ipr driver were originally written by Brian King. Acked-by: Brian King <brking@linux.vnet.ibm.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-11-04 02:45:59 -06:00
/**
* ipr_reset_block_config_access_wait - Wait for permission to block config access
* @ipr_cmd: ipr command struct
*
* Description: This attempts to block config access to the IOA.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_block_config_access_wait(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
int rc = IPR_RC_JOB_CONTINUE;
if (pci_cfg_access_trylock(ioa_cfg->pdev)) {
ioa_cfg->cfg_locked = 1;
ipr_cmd->job_step = ioa_cfg->reset;
} else {
if (ipr_cmd->u.time_left) {
rc = IPR_RC_JOB_RETURN;
ipr_cmd->u.time_left -= IPR_CHECK_FOR_RESET_TIMEOUT;
ipr_reset_start_timer(ipr_cmd,
IPR_CHECK_FOR_RESET_TIMEOUT);
} else {
ipr_cmd->job_step = ioa_cfg->reset;
dev_err(&ioa_cfg->pdev->dev,
"Timed out waiting to lock config access. Resetting anyway.\n");
}
}
return rc;
}
/**
* ipr_reset_block_config_access - Block config access to the IOA
* @ipr_cmd: ipr command struct
*
* Description: This attempts to block config access to the IOA
*
* Return value:
* IPR_RC_JOB_CONTINUE
**/
static int ipr_reset_block_config_access(struct ipr_cmnd *ipr_cmd)
{
ipr_cmd->ioa_cfg->cfg_locked = 0;
ipr_cmd->job_step = ipr_reset_block_config_access_wait;
ipr_cmd->u.time_left = IPR_WAIT_FOR_RESET_TIMEOUT;
return IPR_RC_JOB_CONTINUE;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_reset_allowed - Query whether or not IOA can be reset
* @ioa_cfg: ioa config struct
*
* Return value:
* 0 if reset not allowed / non-zero if reset is allowed
**/
static int ipr_reset_allowed(struct ipr_ioa_cfg *ioa_cfg)
{
volatile u32 temp_reg;
temp_reg = readl(ioa_cfg->regs.sense_interrupt_reg);
return ((temp_reg & IPR_PCII_CRITICAL_OPERATION) == 0);
}
/**
* ipr_reset_wait_to_start_bist - Wait for permission to reset IOA.
* @ipr_cmd: ipr command struct
*
* Description: This function waits for adapter permission to run BIST,
* then runs BIST. If the adapter does not give permission after a
* reasonable time, we will reset the adapter anyway. The impact of
* resetting the adapter without warning the adapter is the risk of
* losing the persistent error log on the adapter. If the adapter is
* reset while it is writing to the flash on the adapter, the flash
* segment will have bad ECC and be zeroed.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_wait_to_start_bist(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
int rc = IPR_RC_JOB_RETURN;
if (!ipr_reset_allowed(ioa_cfg) && ipr_cmd->u.time_left) {
ipr_cmd->u.time_left -= IPR_CHECK_FOR_RESET_TIMEOUT;
ipr_reset_start_timer(ipr_cmd, IPR_CHECK_FOR_RESET_TIMEOUT);
} else {
PCI: Rework config space blocking services pci_block_user_cfg_access was designed for the use case that a single context, the IPR driver, temporarily delays user space accesses to the config space via sysfs. This assumption became invalid by the time pci_dev_reset was added as locking instance. Today, if you run two loops in parallel that reset the same device via sysfs, you end up with a kernel BUG as pci_block_user_cfg_access detect the broken assumption. This reworks the pci_block_user_cfg_access to a sleeping service pci_cfg_access_lock and an atomic-compatible variant called pci_cfg_access_trylock. The former not only blocks user space access as before but also waits if access was already locked. The latter service just returns false in this case, allowing the caller to resolve the conflict instead of raising a BUG. Adaptions of the ipr driver were originally written by Brian King. Acked-by: Brian King <brking@linux.vnet.ibm.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-11-04 02:45:59 -06:00
ipr_cmd->job_step = ipr_reset_block_config_access;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
rc = IPR_RC_JOB_CONTINUE;
}
return rc;
}
/**
[SCSI] ipr: add endian swap enablement for 64 bit adapters A change in the hardware design of the chip for the new adapters changes the default endianness of MMIO operations. This patch adds a register definition which when written to with a predefined value will change the endianness back to what the driver expects. This patch also fixes two problems found during testing. First, the first reserved field in the ipr_hostrcb64_fabirc_desc structure only reserved one byte. The correct amount to reserve is 2 bytes. Second, the reserved field of the ipr_hostrcb64_error structure only reserved 2 bytes. The correct amount to reserve is 16 bytes. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-04 11:26:50 -06:00
* ipr_reset_alert - Alert the adapter of a pending reset
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @ipr_cmd: ipr command struct
*
* Description: This function alerts the adapter that it will be reset.
* If memory space is not currently enabled, proceed directly
* to running BIST on the adapter. The timer must always be started
* so we guarantee we do not run BIST from ipr_isr.
*
* Return value:
* IPR_RC_JOB_RETURN
**/
static int ipr_reset_alert(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
u16 cmd_reg;
int rc;
ENTER;
rc = pci_read_config_word(ioa_cfg->pdev, PCI_COMMAND, &cmd_reg);
if ((rc == PCIBIOS_SUCCESSFUL) && (cmd_reg & PCI_COMMAND_MEMORY)) {
ipr_mask_and_clear_interrupts(ioa_cfg, ~0);
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
writel(IPR_UPROCI_RESET_ALERT, ioa_cfg->regs.set_uproc_interrupt_reg32);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->job_step = ipr_reset_wait_to_start_bist;
} else {
PCI: Rework config space blocking services pci_block_user_cfg_access was designed for the use case that a single context, the IPR driver, temporarily delays user space accesses to the config space via sysfs. This assumption became invalid by the time pci_dev_reset was added as locking instance. Today, if you run two loops in parallel that reset the same device via sysfs, you end up with a kernel BUG as pci_block_user_cfg_access detect the broken assumption. This reworks the pci_block_user_cfg_access to a sleeping service pci_cfg_access_lock and an atomic-compatible variant called pci_cfg_access_trylock. The former not only blocks user space access as before but also waits if access was already locked. The latter service just returns false in this case, allowing the caller to resolve the conflict instead of raising a BUG. Adaptions of the ipr driver were originally written by Brian King. Acked-by: Brian King <brking@linux.vnet.ibm.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-11-04 02:45:59 -06:00
ipr_cmd->job_step = ipr_reset_block_config_access;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ipr_cmd->u.time_left = IPR_WAIT_FOR_RESET_TIMEOUT;
ipr_reset_start_timer(ipr_cmd, IPR_CHECK_FOR_RESET_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
/**
* ipr_reset_quiesce_done - Complete IOA disconnect
* @ipr_cmd: ipr command struct
*
* Description: Freeze the adapter to complete quiesce processing
*
* Return value:
* IPR_RC_JOB_CONTINUE
**/
static int ipr_reset_quiesce_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
ENTER;
ipr_cmd->job_step = ipr_ioa_bringdown_done;
ipr_mask_and_clear_interrupts(ioa_cfg, ~IPR_PCII_IOA_TRANS_TO_OPER);
LEAVE;
return IPR_RC_JOB_CONTINUE;
}
/**
* ipr_reset_cancel_hcam_done - Check for outstanding commands
* @ipr_cmd: ipr command struct
*
* Description: Ensure nothing is outstanding to the IOA and
* proceed with IOA disconnect. Otherwise reset the IOA.
*
* Return value:
* IPR_RC_JOB_RETURN / IPR_RC_JOB_CONTINUE
**/
static int ipr_reset_cancel_hcam_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_cmnd *loop_cmd;
struct ipr_hrr_queue *hrrq;
int rc = IPR_RC_JOB_CONTINUE;
int count = 0;
ENTER;
ipr_cmd->job_step = ipr_reset_quiesce_done;
for_each_hrrq(hrrq, ioa_cfg) {
spin_lock(&hrrq->_lock);
list_for_each_entry(loop_cmd, &hrrq->hrrq_pending_q, queue) {
count++;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
rc = IPR_RC_JOB_RETURN;
break;
}
spin_unlock(&hrrq->_lock);
if (count)
break;
}
LEAVE;
return rc;
}
/**
* ipr_reset_cancel_hcam - Cancel outstanding HCAMs
* @ipr_cmd: ipr command struct
*
* Description: Cancel any oustanding HCAMs to the IOA.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_cancel_hcam(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
int rc = IPR_RC_JOB_CONTINUE;
struct ipr_cmd_pkt *cmd_pkt;
struct ipr_cmnd *hcam_cmd;
struct ipr_hrr_queue *hrrq = &ioa_cfg->hrrq[IPR_INIT_HRRQ];
ENTER;
ipr_cmd->job_step = ipr_reset_cancel_hcam_done;
if (!hrrq->ioa_is_dead) {
if (!list_empty(&ioa_cfg->hostrcb_pending_q)) {
list_for_each_entry(hcam_cmd, &hrrq->hrrq_pending_q, queue) {
if (hcam_cmd->ioarcb.cmd_pkt.cdb[0] != IPR_HOST_CONTROLLED_ASYNC)
continue;
ipr_cmd->ioarcb.res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ipr_cmd->ioarcb.cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
cmd_pkt = &ipr_cmd->ioarcb.cmd_pkt;
cmd_pkt->request_type = IPR_RQTYPE_IOACMD;
cmd_pkt->cdb[0] = IPR_CANCEL_REQUEST;
cmd_pkt->cdb[1] = IPR_CANCEL_64BIT_IOARCB;
cmd_pkt->cdb[10] = ((u64) hcam_cmd->dma_addr >> 56) & 0xff;
cmd_pkt->cdb[11] = ((u64) hcam_cmd->dma_addr >> 48) & 0xff;
cmd_pkt->cdb[12] = ((u64) hcam_cmd->dma_addr >> 40) & 0xff;
cmd_pkt->cdb[13] = ((u64) hcam_cmd->dma_addr >> 32) & 0xff;
cmd_pkt->cdb[2] = ((u64) hcam_cmd->dma_addr >> 24) & 0xff;
cmd_pkt->cdb[3] = ((u64) hcam_cmd->dma_addr >> 16) & 0xff;
cmd_pkt->cdb[4] = ((u64) hcam_cmd->dma_addr >> 8) & 0xff;
cmd_pkt->cdb[5] = ((u64) hcam_cmd->dma_addr) & 0xff;
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout,
IPR_CANCEL_TIMEOUT);
rc = IPR_RC_JOB_RETURN;
ipr_cmd->job_step = ipr_reset_cancel_hcam;
break;
}
}
} else
ipr_cmd->job_step = ipr_reset_alert;
LEAVE;
return rc;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_reset_ucode_download_done - Microcode download completion
* @ipr_cmd: ipr command struct
*
* Description: This function unmaps the microcode download buffer.
*
* Return value:
* IPR_RC_JOB_CONTINUE
**/
static int ipr_reset_ucode_download_done(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_sglist *sglist = ioa_cfg->ucode_sglist;
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_unmap_sg(&ioa_cfg->pdev->dev, sglist->scatterlist,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sglist->num_sg, DMA_TO_DEVICE);
ipr_cmd->job_step = ipr_reset_alert;
return IPR_RC_JOB_CONTINUE;
}
/**
* ipr_reset_ucode_download - Download microcode to the adapter
* @ipr_cmd: ipr command struct
*
* Description: This function checks to see if it there is microcode
* to download to the adapter. If there is, a download is performed.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_ucode_download(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
struct ipr_sglist *sglist = ioa_cfg->ucode_sglist;
ENTER;
ipr_cmd->job_step = ipr_reset_alert;
if (!sglist)
return IPR_RC_JOB_CONTINUE;
ipr_cmd->ioarcb.res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ipr_cmd->ioarcb.cmd_pkt.request_type = IPR_RQTYPE_SCSICDB;
ipr_cmd->ioarcb.cmd_pkt.cdb[0] = WRITE_BUFFER;
ipr_cmd->ioarcb.cmd_pkt.cdb[1] = IPR_WR_BUF_DOWNLOAD_AND_SAVE;
ipr_cmd->ioarcb.cmd_pkt.cdb[6] = (sglist->buffer_len & 0xff0000) >> 16;
ipr_cmd->ioarcb.cmd_pkt.cdb[7] = (sglist->buffer_len & 0x00ff00) >> 8;
ipr_cmd->ioarcb.cmd_pkt.cdb[8] = sglist->buffer_len & 0x0000ff;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ioa_cfg->sis64)
ipr_build_ucode_ioadl64(ipr_cmd, sglist);
else
ipr_build_ucode_ioadl(ipr_cmd, sglist);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->job_step = ipr_reset_ucode_download_done;
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout,
IPR_WRITE_BUFFER_TIMEOUT);
LEAVE;
return IPR_RC_JOB_RETURN;
}
/**
* ipr_reset_shutdown_ioa - Shutdown the adapter
* @ipr_cmd: ipr command struct
*
* Description: This function issues an adapter shutdown of the
* specified type to the specified adapter as part of the
* adapter reset job.
*
* Return value:
* IPR_RC_JOB_CONTINUE / IPR_RC_JOB_RETURN
**/
static int ipr_reset_shutdown_ioa(struct ipr_cmnd *ipr_cmd)
{
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
enum ipr_shutdown_type shutdown_type = ipr_cmd->u.shutdown_type;
unsigned long timeout;
int rc = IPR_RC_JOB_CONTINUE;
ENTER;
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
if (shutdown_type == IPR_SHUTDOWN_QUIESCE)
ipr_cmd->job_step = ipr_reset_cancel_hcam;
else if (shutdown_type != IPR_SHUTDOWN_NONE &&
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
!ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd->ioarcb.res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ipr_cmd->ioarcb.cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
ipr_cmd->ioarcb.cmd_pkt.cdb[0] = IPR_IOA_SHUTDOWN;
ipr_cmd->ioarcb.cmd_pkt.cdb[1] = shutdown_type;
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
if (shutdown_type == IPR_SHUTDOWN_NORMAL)
timeout = IPR_SHUTDOWN_TIMEOUT;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
else if (shutdown_type == IPR_SHUTDOWN_PREPARE_FOR_NORMAL)
timeout = IPR_INTERNAL_TIMEOUT;
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
else if (ioa_cfg->dual_raid && ipr_dual_ioa_raid)
timeout = IPR_DUAL_IOA_ABBR_SHUTDOWN_TO;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
else
[SCSI] ipr: Enable multi-initator RAID support Enables multi-initiator support on ipr RAID adapters that support it. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:16 -06:00
timeout = IPR_ABBREV_SHUTDOWN_TIMEOUT;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_do_req(ipr_cmd, ipr_reset_ioa_job, ipr_timeout, timeout);
rc = IPR_RC_JOB_RETURN;
ipr_cmd->job_step = ipr_reset_ucode_download;
} else
ipr_cmd->job_step = ipr_reset_alert;
LEAVE;
return rc;
}
/**
* ipr_reset_ioa_job - Adapter reset job
* @ipr_cmd: ipr command struct
*
* Description: This function is the job router for the adapter reset job.
*
* Return value:
* none
**/
static void ipr_reset_ioa_job(struct ipr_cmnd *ipr_cmd)
{
u32 rc, ioasc;
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
do {
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioa_cfg->reset_cmd != ipr_cmd) {
/*
* We are doing nested adapter resets and this is
* not the current reset job.
*/
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue,
&ipr_cmd->hrrq->hrrq_free_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
}
if (IPR_IOASC_SENSE_KEY(ioasc)) {
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
rc = ipr_cmd->job_step_failed(ipr_cmd);
if (rc == IPR_RC_JOB_RETURN)
return;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ipr_reinit_ipr_cmnd(ipr_cmd);
[SCSI] ipr: Better handle failure of adapter bringup commands Some new ipr adapters do not support some of the initialization commands currently sent to it from the driver. Handle these commands failing and continue on with the adapter initialization. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:55 -07:00
ipr_cmd->job_step_failed = ipr_reset_cmd_failed;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
rc = ipr_cmd->job_step(ipr_cmd);
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
} while (rc == IPR_RC_JOB_CONTINUE);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* _ipr_initiate_ioa_reset - Initiate an adapter reset
* @ioa_cfg: ioa config struct
* @job_step: first job step of reset job
* @shutdown_type: shutdown type
*
* Description: This function will initiate the reset of the given adapter
* starting at the selected job step.
* If the caller needs to wait on the completion of the reset,
* the caller must sleep on the reset_wait_q.
*
* Return value:
* none
**/
static void _ipr_initiate_ioa_reset(struct ipr_ioa_cfg *ioa_cfg,
int (*job_step) (struct ipr_cmnd *),
enum ipr_shutdown_type shutdown_type)
{
struct ipr_cmnd *ipr_cmd;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
int i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->in_reset_reload = 1;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].allow_cmds = 0;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
wmb();
scsi: ipr: Fix scsi-mq lockdep issue Fixes the following lockdep warning that can occur when scsi-mq is enabled with ipr due to ipr calling scsi_unblock_requests from irq context. The fix is to move the call to scsi_unblock_requests to ipr's existing workqueue. stack backtrace: CPU: 28 PID: 0 Comm: swapper/28 Not tainted 4.13.0-rc2-gcc6x-gf74c89b #1 Call Trace: [c000001fffe97550] [c000000000b50818] dump_stack+0xe8/0x160 (unreliable) [c000001fffe97590] [c0000000001586d0] print_usage_bug+0x2d0/0x390 [c000001fffe97640] [c000000000158f34] mark_lock+0x7a4/0x8e0 [c000001fffe976f0] [c00000000015a000] __lock_acquire+0x6a0/0x1a70 [c000001fffe97860] [c00000000015befc] lock_acquire+0xec/0x2e0 [c000001fffe97930] [c000000000b71514] _raw_spin_lock+0x44/0x70 [c000001fffe97960] [c0000000005b60f4] blk_mq_sched_dispatch_requests+0xa4/0x2a0 [c000001fffe979c0] [c0000000005acac0] __blk_mq_run_hw_queue+0x100/0x2c0 [c000001fffe97a00] [c0000000005ad478] __blk_mq_delay_run_hw_queue+0x118/0x130 [c000001fffe97a40] [c0000000005ad61c] blk_mq_start_hw_queues+0x6c/0xa0 [c000001fffe97a80] [c000000000797aac] scsi_kick_queue+0x2c/0x60 [c000001fffe97aa0] [c000000000797cf0] scsi_run_queue+0x210/0x360 [c000001fffe97b10] [c00000000079b888] scsi_run_host_queues+0x48/0x80 [c000001fffe97b40] [c0000000007b6090] ipr_ioa_bringdown_done+0x70/0x1e0 [c000001fffe97bc0] [c0000000007bc860] ipr_reset_ioa_job+0x80/0xf0 [c000001fffe97bf0] [c0000000007b4d50] ipr_reset_timer_done+0xd0/0x100 [c000001fffe97c30] [c0000000001937bc] call_timer_fn+0xdc/0x4b0 [c000001fffe97cf0] [c000000000193d08] expire_timers+0x178/0x330 [c000001fffe97d60] [c0000000001940c8] run_timer_softirq+0xb8/0x120 [c000001fffe97de0] [c000000000b726a8] __do_softirq+0x168/0x6d8 [c000001fffe97ef0] [c0000000000df2c8] irq_exit+0x108/0x150 [c000001fffe97f10] [c000000000017bf4] __do_irq+0x2a4/0x4a0 [c000001fffe97f90] [c00000000002da50] call_do_irq+0x14/0x24 [c0000007fad93aa0] [c000000000017e8c] do_IRQ+0x9c/0x140 [c0000007fad93af0] [c000000000008b98] hardware_interrupt_common+0x138/0x140 Reported-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-01 09:21:30 -06:00
if (!ioa_cfg->hrrq[IPR_INIT_HRRQ].removing_ioa) {
ioa_cfg->scsi_unblock = 0;
ioa_cfg->scsi_blocked = 1;
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
scsi_block_requests(ioa_cfg->host);
scsi: ipr: Fix scsi-mq lockdep issue Fixes the following lockdep warning that can occur when scsi-mq is enabled with ipr due to ipr calling scsi_unblock_requests from irq context. The fix is to move the call to scsi_unblock_requests to ipr's existing workqueue. stack backtrace: CPU: 28 PID: 0 Comm: swapper/28 Not tainted 4.13.0-rc2-gcc6x-gf74c89b #1 Call Trace: [c000001fffe97550] [c000000000b50818] dump_stack+0xe8/0x160 (unreliable) [c000001fffe97590] [c0000000001586d0] print_usage_bug+0x2d0/0x390 [c000001fffe97640] [c000000000158f34] mark_lock+0x7a4/0x8e0 [c000001fffe976f0] [c00000000015a000] __lock_acquire+0x6a0/0x1a70 [c000001fffe97860] [c00000000015befc] lock_acquire+0xec/0x2e0 [c000001fffe97930] [c000000000b71514] _raw_spin_lock+0x44/0x70 [c000001fffe97960] [c0000000005b60f4] blk_mq_sched_dispatch_requests+0xa4/0x2a0 [c000001fffe979c0] [c0000000005acac0] __blk_mq_run_hw_queue+0x100/0x2c0 [c000001fffe97a00] [c0000000005ad478] __blk_mq_delay_run_hw_queue+0x118/0x130 [c000001fffe97a40] [c0000000005ad61c] blk_mq_start_hw_queues+0x6c/0xa0 [c000001fffe97a80] [c000000000797aac] scsi_kick_queue+0x2c/0x60 [c000001fffe97aa0] [c000000000797cf0] scsi_run_queue+0x210/0x360 [c000001fffe97b10] [c00000000079b888] scsi_run_host_queues+0x48/0x80 [c000001fffe97b40] [c0000000007b6090] ipr_ioa_bringdown_done+0x70/0x1e0 [c000001fffe97bc0] [c0000000007bc860] ipr_reset_ioa_job+0x80/0xf0 [c000001fffe97bf0] [c0000000007b4d50] ipr_reset_timer_done+0xd0/0x100 [c000001fffe97c30] [c0000000001937bc] call_timer_fn+0xdc/0x4b0 [c000001fffe97cf0] [c000000000193d08] expire_timers+0x178/0x330 [c000001fffe97d60] [c0000000001940c8] run_timer_softirq+0xb8/0x120 [c000001fffe97de0] [c000000000b726a8] __do_softirq+0x168/0x6d8 [c000001fffe97ef0] [c0000000000df2c8] irq_exit+0x108/0x150 [c000001fffe97f10] [c000000000017bf4] __do_irq+0x2a4/0x4a0 [c000001fffe97f90] [c00000000002da50] call_do_irq+0x14/0x24 [c0000007fad93aa0] [c000000000017e8c] do_IRQ+0x9c/0x140 [c0000007fad93af0] [c000000000008b98] hardware_interrupt_common+0x138/0x140 Reported-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-01 09:21:30 -06:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_cmd = ipr_get_free_ipr_cmnd(ioa_cfg);
ioa_cfg->reset_cmd = ipr_cmd;
ipr_cmd->job_step = job_step;
ipr_cmd->u.shutdown_type = shutdown_type;
ipr_reset_ioa_job(ipr_cmd);
}
/**
* ipr_initiate_ioa_reset - Initiate an adapter reset
* @ioa_cfg: ioa config struct
* @shutdown_type: shutdown type
*
* Description: This function will initiate the reset of the given adapter.
* If the caller needs to wait on the completion of the reset,
* the caller must sleep on the reset_wait_q.
*
* Return value:
* none
**/
static void ipr_initiate_ioa_reset(struct ipr_ioa_cfg *ioa_cfg,
enum ipr_shutdown_type shutdown_type)
{
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
int i;
if (ioa_cfg->hrrq[IPR_INIT_HRRQ].ioa_is_dead)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
[SCSI] ipr: Stop reading adapter dump prematurely When the ipr driver decides to dump the adapter, it changes the sdt_state to GET_DUMP, then prepares the adapter so that the dump can be read. However, if the ipr worker thread wakes up for some reason before the driver has put the adapter in a state where it can succesfully dump the adapter, the driver will start dumping the adapter too early, which can potentially trigger a BUG check in the pci config blocking API. Fix this by adding a new sdt_state to differentiate between the ipr driver wanting to dump the adapter in the near future and wanting to dump the adapter now. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-09-21 07:51:11 -06:00
if (ioa_cfg->in_reset_reload) {
if (ioa_cfg->sdt_state == GET_DUMP)
ioa_cfg->sdt_state = WAIT_FOR_DUMP;
else if (ioa_cfg->sdt_state == READ_DUMP)
ioa_cfg->sdt_state = ABORT_DUMP;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioa_cfg->reset_retries++ >= IPR_NUM_RESET_RELOAD_RETRIES) {
dev_err(&ioa_cfg->pdev->dev,
"IOA taken offline - error recovery failed\n");
ioa_cfg->reset_retries = 0;
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].ioa_is_dead = 1;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
wmb();
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioa_cfg->in_ioa_bringdown) {
ioa_cfg->reset_cmd = NULL;
ioa_cfg->in_reset_reload = 0;
ipr_fail_all_ops(ioa_cfg);
wake_up_all(&ioa_cfg->reset_wait_q);
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
if (!ioa_cfg->hrrq[IPR_INIT_HRRQ].removing_ioa) {
scsi: ipr: Fix scsi-mq lockdep issue Fixes the following lockdep warning that can occur when scsi-mq is enabled with ipr due to ipr calling scsi_unblock_requests from irq context. The fix is to move the call to scsi_unblock_requests to ipr's existing workqueue. stack backtrace: CPU: 28 PID: 0 Comm: swapper/28 Not tainted 4.13.0-rc2-gcc6x-gf74c89b #1 Call Trace: [c000001fffe97550] [c000000000b50818] dump_stack+0xe8/0x160 (unreliable) [c000001fffe97590] [c0000000001586d0] print_usage_bug+0x2d0/0x390 [c000001fffe97640] [c000000000158f34] mark_lock+0x7a4/0x8e0 [c000001fffe976f0] [c00000000015a000] __lock_acquire+0x6a0/0x1a70 [c000001fffe97860] [c00000000015befc] lock_acquire+0xec/0x2e0 [c000001fffe97930] [c000000000b71514] _raw_spin_lock+0x44/0x70 [c000001fffe97960] [c0000000005b60f4] blk_mq_sched_dispatch_requests+0xa4/0x2a0 [c000001fffe979c0] [c0000000005acac0] __blk_mq_run_hw_queue+0x100/0x2c0 [c000001fffe97a00] [c0000000005ad478] __blk_mq_delay_run_hw_queue+0x118/0x130 [c000001fffe97a40] [c0000000005ad61c] blk_mq_start_hw_queues+0x6c/0xa0 [c000001fffe97a80] [c000000000797aac] scsi_kick_queue+0x2c/0x60 [c000001fffe97aa0] [c000000000797cf0] scsi_run_queue+0x210/0x360 [c000001fffe97b10] [c00000000079b888] scsi_run_host_queues+0x48/0x80 [c000001fffe97b40] [c0000000007b6090] ipr_ioa_bringdown_done+0x70/0x1e0 [c000001fffe97bc0] [c0000000007bc860] ipr_reset_ioa_job+0x80/0xf0 [c000001fffe97bf0] [c0000000007b4d50] ipr_reset_timer_done+0xd0/0x100 [c000001fffe97c30] [c0000000001937bc] call_timer_fn+0xdc/0x4b0 [c000001fffe97cf0] [c000000000193d08] expire_timers+0x178/0x330 [c000001fffe97d60] [c0000000001940c8] run_timer_softirq+0xb8/0x120 [c000001fffe97de0] [c000000000b726a8] __do_softirq+0x168/0x6d8 [c000001fffe97ef0] [c0000000000df2c8] irq_exit+0x108/0x150 [c000001fffe97f10] [c000000000017bf4] __do_irq+0x2a4/0x4a0 [c000001fffe97f90] [c00000000002da50] call_do_irq+0x14/0x24 [c0000007fad93aa0] [c000000000017e8c] do_IRQ+0x9c/0x140 [c0000007fad93af0] [c000000000008b98] hardware_interrupt_common+0x138/0x140 Reported-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-08-01 09:21:30 -06:00
ioa_cfg->scsi_unblock = 1;
schedule_work(&ioa_cfg->work_q);
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return;
} else {
ioa_cfg->in_ioa_bringdown = 1;
shutdown_type = IPR_SHUTDOWN_NONE;
}
}
_ipr_initiate_ioa_reset(ioa_cfg, ipr_reset_shutdown_ioa,
shutdown_type);
}
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
/**
* ipr_reset_freeze - Hold off all I/O activity
* @ipr_cmd: ipr command struct
*
* Description: If the PCI slot is frozen, hold off all I/O
* activity; then, as soon as the slot is available again,
* initiate an adapter reset.
*/
static int ipr_reset_freeze(struct ipr_cmnd *ipr_cmd)
{
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg;
int i;
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
/* Disallow new interrupts, avoid loop */
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].allow_interrupts = 0;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
wmb();
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_pending_q);
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
ipr_cmd->done = ipr_reset_ioa_job;
return IPR_RC_JOB_RETURN;
}
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
/**
* ipr_pci_mmio_enabled - Called when MMIO has been re-enabled
* @pdev: PCI device struct
*
* Description: This routine is called to tell us that the MMIO
* access to the IOA has been restored
*/
static pci_ers_result_t ipr_pci_mmio_enabled(struct pci_dev *pdev)
{
unsigned long flags = 0;
struct ipr_ioa_cfg *ioa_cfg = pci_get_drvdata(pdev);
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
if (!ioa_cfg->probe_done)
pci_save_state(pdev);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
return PCI_ERS_RESULT_NEED_RESET;
}
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
/**
* ipr_pci_frozen - Called when slot has experienced a PCI bus error.
* @pdev: PCI device struct
*
* Description: This routine is called to tell us that the PCI bus
* is down. Can't do anything here, except put the device driver
* into a holding pattern, waiting for the PCI bus to come back.
*/
static void ipr_pci_frozen(struct pci_dev *pdev)
{
unsigned long flags = 0;
struct ipr_ioa_cfg *ioa_cfg = pci_get_drvdata(pdev);
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
if (ioa_cfg->probe_done)
_ipr_initiate_ioa_reset(ioa_cfg, ipr_reset_freeze, IPR_SHUTDOWN_NONE);
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
}
/**
* ipr_pci_slot_reset - Called when PCI slot has been reset.
* @pdev: PCI device struct
*
* Description: This routine is called by the pci error recovery
* code after the PCI slot has been reset, just before we
* should resume normal operations.
*/
static pci_ers_result_t ipr_pci_slot_reset(struct pci_dev *pdev)
{
unsigned long flags = 0;
struct ipr_ioa_cfg *ioa_cfg = pci_get_drvdata(pdev);
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
if (ioa_cfg->probe_done) {
if (ioa_cfg->needs_warm_reset)
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
else
_ipr_initiate_ioa_reset(ioa_cfg, ipr_reset_restore_cfg_space,
IPR_SHUTDOWN_NONE);
} else
wake_up_all(&ioa_cfg->eeh_wait_q);
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
return PCI_ERS_RESULT_RECOVERED;
}
/**
* ipr_pci_perm_failure - Called when PCI slot is dead for good.
* @pdev: PCI device struct
*
* Description: This routine is called when the PCI bus has
* permanently failed.
*/
static void ipr_pci_perm_failure(struct pci_dev *pdev)
{
unsigned long flags = 0;
struct ipr_ioa_cfg *ioa_cfg = pci_get_drvdata(pdev);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
int i;
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
if (ioa_cfg->probe_done) {
if (ioa_cfg->sdt_state == WAIT_FOR_DUMP)
ioa_cfg->sdt_state = ABORT_DUMP;
ioa_cfg->reset_retries = IPR_NUM_RESET_RELOAD_RETRIES - 1;
ioa_cfg->in_ioa_bringdown = 1;
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].allow_cmds = 0;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
wmb();
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
} else
wake_up_all(&ioa_cfg->eeh_wait_q);
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
}
/**
* ipr_pci_error_detected - Called when a PCI error is detected.
* @pdev: PCI device struct
* @state: PCI channel state
*
* Description: Called when a PCI error is detected.
*
* Return value:
* PCI_ERS_RESULT_NEED_RESET or PCI_ERS_RESULT_DISCONNECT
*/
static pci_ers_result_t ipr_pci_error_detected(struct pci_dev *pdev,
pci_channel_state_t state)
{
switch (state) {
case pci_channel_io_frozen:
ipr_pci_frozen(pdev);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
return PCI_ERS_RESULT_CAN_RECOVER;
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
case pci_channel_io_perm_failure:
ipr_pci_perm_failure(pdev);
return PCI_ERS_RESULT_DISCONNECT;
break;
default:
break;
}
return PCI_ERS_RESULT_NEED_RESET;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_probe_ioa_part2 - Initializes IOAs found in ipr_probe_ioa(..)
* @ioa_cfg: ioa cfg struct
*
scripts/spelling.txt: add "intialization" pattern and fix typo instances Fix typos and add the following to the scripts/spelling.txt: intialization||initialization The "inintialization" in drivers/acpi/spcr.c is a different pattern but I fixed it as well in this commit. Link: http://lkml.kernel.org/r/1481573103-11329-16-git-send-email-yamada.masahiro@socionext.com Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-02-27 15:29:20 -07:00
* Description: This is the second phase of adapter initialization
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* This function takes care of initilizing the adapter to the point
* where it can accept new commands.
* Return value:
drivers/scsi/: Spelling fixes Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Acked-by: James Smart <james.smart@emulex.com> Acked-by: Darrick J. Wong <djwong@us.ibm.com> Acked-by: David Somayajulu <david.somayajulu@qlogic.com> Acked-by: Mark Salyzyn <mark_salyzyn@adaptec.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
2008-02-03 08:28:22 -07:00
* 0 on success / -EIO on failure
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static int ipr_probe_ioa_part2(struct ipr_ioa_cfg *ioa_cfg)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
int rc = 0;
unsigned long host_lock_flags = 0;
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, host_lock_flags);
dev_dbg(&ioa_cfg->pdev->dev, "ioa_cfg adx: 0x%p\n", ioa_cfg);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ioa_cfg->probe_done = 1;
[SCSI] ipr: Driver initialization fix for kexec/kdump When kexec booting a kernel when the previous kernel did not call ipr's shutdown method, the ipr adapter does not get properly initialized, which can result in the ipr adapter completing commands issued by the previous kernel. Fix ipr to detect this scenario by reading the adapter's interrupt mask register and the microprocessor interrupt register. If the interrupt mask register indicates that interrupts are enabled or the reset alert bit is set when the card is probed, this means the card is in an unknown state and we hard reset the card. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-17 08:35:12 -07:00
if (ioa_cfg->needs_hard_reset) {
ioa_cfg->needs_hard_reset = 0;
ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_NONE);
} else
_ipr_initiate_ioa_reset(ioa_cfg, ipr_reset_enable_ioa,
IPR_SHUTDOWN_NONE);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, host_lock_flags);
LEAVE;
return rc;
}
/**
* ipr_free_cmd_blks - Frees command blocks allocated for an adapter
* @ioa_cfg: ioa config struct
*
* Return value:
* none
**/
static void ipr_free_cmd_blks(struct ipr_ioa_cfg *ioa_cfg)
{
int i;
ipr: Fix possible error path oops during initialization Fixes a possible oops during adapter initialization in some memory allocation failure error paths scenarios. Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:55 -06:00
if (ioa_cfg->ipr_cmnd_list) {
for (i = 0; i < IPR_NUM_CMD_BLKS; i++) {
if (ioa_cfg->ipr_cmnd_list[i])
dma_pool_free(ioa_cfg->ipr_cmd_pool,
ioa_cfg->ipr_cmnd_list[i],
ioa_cfg->ipr_cmnd_list_dma[i]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: Fix possible error path oops during initialization Fixes a possible oops during adapter initialization in some memory allocation failure error paths scenarios. Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:55 -06:00
ioa_cfg->ipr_cmnd_list[i] = NULL;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
if (ioa_cfg->ipr_cmd_pool)
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_pool_destroy(ioa_cfg->ipr_cmd_pool);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Increase max concurrent oustanding commands Increase the total number of max concurrent outstanding commands for the most recent family of adapters in order to improve overall adapter performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:10 -06:00
kfree(ioa_cfg->ipr_cmnd_list);
kfree(ioa_cfg->ipr_cmnd_list_dma);
ioa_cfg->ipr_cmnd_list = NULL;
ioa_cfg->ipr_cmnd_list_dma = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->ipr_cmd_pool = NULL;
}
/**
* ipr_free_mem - Frees memory allocated for an adapter
* @ioa_cfg: ioa cfg struct
*
* Return value:
* nothing
**/
static void ipr_free_mem(struct ipr_ioa_cfg *ioa_cfg)
{
int i;
kfree(ioa_cfg->res_entries);
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&ioa_cfg->pdev->dev, sizeof(struct ipr_misc_cbs),
ioa_cfg->vpd_cbs, ioa_cfg->vpd_cbs_dma);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_free_cmd_blks(ioa_cfg);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++)
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&ioa_cfg->pdev->dev,
sizeof(u32) * ioa_cfg->hrrq[i].size,
ioa_cfg->hrrq[i].host_rrq,
ioa_cfg->hrrq[i].host_rrq_dma);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&ioa_cfg->pdev->dev, ioa_cfg->cfg_table_size,
ioa_cfg->u.cfg_table, ioa_cfg->cfg_table_dma);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
for (i = 0; i < IPR_MAX_HCAMS; i++) {
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&ioa_cfg->pdev->dev,
sizeof(struct ipr_hostrcb),
ioa_cfg->hostrcb[i],
ioa_cfg->hostrcb_dma[i]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ipr_free_dump(ioa_cfg);
kfree(ioa_cfg->trace);
}
/**
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
* ipr_free_irqs - Free all allocated IRQs for the adapter.
* @ioa_cfg: ipr cfg struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
* This function frees all allocated IRQs for the
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* specified adapter.
*
* Return value:
* none
**/
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
static void ipr_free_irqs(struct ipr_ioa_cfg *ioa_cfg)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct pci_dev *pdev = ioa_cfg->pdev;
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
int i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
for (i = 0; i < ioa_cfg->nvectors; i++)
free_irq(pci_irq_vector(pdev, i), &ioa_cfg->hrrq[i]);
pci_free_irq_vectors(pdev);
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
/**
* ipr_free_all_resources - Free all allocated resources for an adapter.
* @ipr_cmd: ipr command struct
*
* This function frees all allocated resources for the
* specified adapter.
*
* Return value:
* none
**/
static void ipr_free_all_resources(struct ipr_ioa_cfg *ioa_cfg)
{
struct pci_dev *pdev = ioa_cfg->pdev;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
ENTER;
ipr_free_irqs(ioa_cfg);
if (ioa_cfg->reset_work_q)
destroy_workqueue(ioa_cfg->reset_work_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
iounmap(ioa_cfg->hdw_dma_regs);
pci_release_regions(pdev);
ipr_free_mem(ioa_cfg);
scsi_host_put(ioa_cfg->host);
pci_disable_device(pdev);
LEAVE;
}
/**
* ipr_alloc_cmd_blks - Allocate command blocks for an adapter
* @ioa_cfg: ioa config struct
*
* Return value:
* 0 on success / -ENOMEM on allocation failure
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static int ipr_alloc_cmd_blks(struct ipr_ioa_cfg *ioa_cfg)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_cmnd *ipr_cmd;
struct ipr_ioarcb *ioarcb;
dma_addr_t dma_addr;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
int i, entries_each_hrrq, hrrq_id = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
ioa_cfg->ipr_cmd_pool = dma_pool_create(IPR_NAME, &ioa_cfg->pdev->dev,
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
sizeof(struct ipr_cmnd), 512, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ioa_cfg->ipr_cmd_pool)
return -ENOMEM;
[SCSI] ipr: Increase max concurrent oustanding commands Increase the total number of max concurrent outstanding commands for the most recent family of adapters in order to improve overall adapter performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:10 -06:00
ioa_cfg->ipr_cmnd_list = kcalloc(IPR_NUM_CMD_BLKS, sizeof(struct ipr_cmnd *), GFP_KERNEL);
ioa_cfg->ipr_cmnd_list_dma = kcalloc(IPR_NUM_CMD_BLKS, sizeof(dma_addr_t), GFP_KERNEL);
if (!ioa_cfg->ipr_cmnd_list || !ioa_cfg->ipr_cmnd_list_dma) {
ipr_free_cmd_blks(ioa_cfg);
return -ENOMEM;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
if (ioa_cfg->hrrq_num > 1) {
if (i == 0) {
entries_each_hrrq = IPR_NUM_INTERNAL_CMD_BLKS;
ioa_cfg->hrrq[i].min_cmd_id = 0;
scsi: ipr: fix incorrect indentation of assignment statement Remove one extraneous level of indentation on an assignment statement. Signed-off-by: Colin Ian King <colin.king@canonical.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-12-01 06:33:27 -07:00
ioa_cfg->hrrq[i].max_cmd_id =
(entries_each_hrrq - 1);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
} else {
entries_each_hrrq =
IPR_NUM_BASE_CMD_BLKS/
(ioa_cfg->hrrq_num - 1);
ioa_cfg->hrrq[i].min_cmd_id =
IPR_NUM_INTERNAL_CMD_BLKS +
(i - 1) * entries_each_hrrq;
ioa_cfg->hrrq[i].max_cmd_id =
(IPR_NUM_INTERNAL_CMD_BLKS +
i * entries_each_hrrq - 1);
}
} else {
entries_each_hrrq = IPR_NUM_CMD_BLKS;
ioa_cfg->hrrq[i].min_cmd_id = 0;
ioa_cfg->hrrq[i].max_cmd_id = (entries_each_hrrq - 1);
}
ioa_cfg->hrrq[i].size = entries_each_hrrq;
}
BUG_ON(ioa_cfg->hrrq_num == 0);
i = IPR_NUM_CMD_BLKS -
ioa_cfg->hrrq[ioa_cfg->hrrq_num - 1].max_cmd_id - 1;
if (i > 0) {
ioa_cfg->hrrq[ioa_cfg->hrrq_num - 1].size += i;
ioa_cfg->hrrq[ioa_cfg->hrrq_num - 1].max_cmd_id += i;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
for (i = 0; i < IPR_NUM_CMD_BLKS; i++) {
scsi: ipr: Use dma_pool_zalloc() Use dma_pool_zalloc() instead of dma_pool_alloc + memset Signed-off-by: Souptick Joarder <jrdr.linux@gmail.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-03-08 06:11:57 -07:00
ipr_cmd = dma_pool_zalloc(ioa_cfg->ipr_cmd_pool,
GFP_KERNEL, &dma_addr);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ipr_cmd) {
ipr_free_cmd_blks(ioa_cfg);
return -ENOMEM;
}
ioa_cfg->ipr_cmnd_list[i] = ipr_cmd;
ioa_cfg->ipr_cmnd_list_dma[i] = dma_addr;
ioarcb = &ipr_cmd->ioarcb;
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
ipr_cmd->dma_addr = dma_addr;
if (ioa_cfg->sis64)
ioarcb->a.ioarcb_host_pci_addr64 = cpu_to_be64(dma_addr);
else
ioarcb->a.ioarcb_host_pci_addr = cpu_to_be32(dma_addr);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->host_response_handle = cpu_to_be32(i << 2);
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ioa_cfg->sis64) {
ioarcb->u.sis64_addr_data.data_ioadl_addr =
cpu_to_be64(dma_addr + offsetof(struct ipr_cmnd, i.ioadl64));
ioarcb->u.sis64_addr_data.ioasa_host_pci_addr =
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
cpu_to_be64(dma_addr + offsetof(struct ipr_cmnd, s.ioasa64));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
} else {
ioarcb->write_ioadl_addr =
cpu_to_be32(dma_addr + offsetof(struct ipr_cmnd, i.ioadl));
ioarcb->read_ioadl_addr = ioarcb->write_ioadl_addr;
ioarcb->ioasa_host_pci_addr =
[SCSI] ipr: include the resource path in the IOA status area structure The IOA status area now includes the new resource path field for 64 bit adapters. This patch changes the driver to fix the ioasa structure and to use the correct structure definition based on the type of adatper. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-05-10 10:13:27 -06:00
cpu_to_be32(dma_addr + offsetof(struct ipr_cmnd, s.ioasa));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioarcb->ioasa_len = cpu_to_be16(sizeof(struct ipr_ioasa));
ipr_cmd->cmd_index = i;
ipr_cmd->ioa_cfg = ioa_cfg;
ipr_cmd->sense_buffer_dma = dma_addr +
offsetof(struct ipr_cmnd, sense_buffer);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_cmd->ioarcb.cmd_pkt.hrrq_id = hrrq_id;
ipr_cmd->hrrq = &ioa_cfg->hrrq[hrrq_id];
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
if (i >= ioa_cfg->hrrq[hrrq_id].max_cmd_id)
hrrq_id++;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
return 0;
}
/**
* ipr_alloc_mem - Allocate memory for an adapter
* @ioa_cfg: ioa config struct
*
* Return value:
* 0 on success / non-zero for error
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static int ipr_alloc_mem(struct ipr_ioa_cfg *ioa_cfg)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct pci_dev *pdev = ioa_cfg->pdev;
int i, rc = -ENOMEM;
ENTER;
treewide: kzalloc() -> kcalloc() The kzalloc() function has a 2-factor argument form, kcalloc(). This patch replaces cases of: kzalloc(a * b, gfp) with: kcalloc(a * b, gfp) as well as handling cases of: kzalloc(a * b * c, gfp) with: kzalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kzalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kzalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kzalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kzalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kzalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(char) * COUNT + COUNT , ...) | kzalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kzalloc + kcalloc ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kzalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kzalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kzalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kzalloc(C1 * C2 * C3, ...) | kzalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kzalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kzalloc(sizeof(THING) * C2, ...) | kzalloc(sizeof(TYPE) * C2, ...) | kzalloc(C1 * C2 * C3, ...) | kzalloc(C1 * C2, ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - (E1) * E2 + E1, E2 , ...) | - kzalloc + kcalloc ( - (E1) * (E2) + E1, E2 , ...) | - kzalloc + kcalloc ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 15:03:40 -06:00
ioa_cfg->res_entries = kcalloc(ioa_cfg->max_devs_supported,
sizeof(struct ipr_resource_entry),
GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ioa_cfg->res_entries)
goto out;
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
for (i = 0; i < ioa_cfg->max_devs_supported; i++) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_add_tail(&ioa_cfg->res_entries[i].queue, &ioa_cfg->free_res_q);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ioa_cfg->res_entries[i].ioa_cfg = ioa_cfg;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
ioa_cfg->vpd_cbs = dma_alloc_coherent(&pdev->dev,
sizeof(struct ipr_misc_cbs),
&ioa_cfg->vpd_cbs_dma,
GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ioa_cfg->vpd_cbs)
goto out_free_res_entries;
if (ipr_alloc_cmd_blks(ioa_cfg))
goto out_free_vpd_cbs;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
ioa_cfg->hrrq[i].host_rrq = dma_alloc_coherent(&pdev->dev,
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
sizeof(u32) * ioa_cfg->hrrq[i].size,
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
&ioa_cfg->hrrq[i].host_rrq_dma,
GFP_KERNEL);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
if (!ioa_cfg->hrrq[i].host_rrq) {
while (--i > 0)
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&pdev->dev,
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
sizeof(u32) * ioa_cfg->hrrq[i].size,
ioa_cfg->hrrq[i].host_rrq,
ioa_cfg->hrrq[i].host_rrq_dma);
goto out_ipr_free_cmd_blocks;
}
ioa_cfg->hrrq[i].ioa_cfg = ioa_cfg;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
ioa_cfg->u.cfg_table = dma_alloc_coherent(&pdev->dev,
ioa_cfg->cfg_table_size,
&ioa_cfg->cfg_table_dma,
GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (!ioa_cfg->u.cfg_table)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
goto out_free_host_rrq;
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
for (i = 0; i < IPR_MAX_HCAMS; i++) {
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
ioa_cfg->hostrcb[i] = dma_alloc_coherent(&pdev->dev,
sizeof(struct ipr_hostrcb),
&ioa_cfg->hostrcb_dma[i],
GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ioa_cfg->hostrcb[i])
goto out_free_hostrcb_dma;
ioa_cfg->hostrcb[i]->hostrcb_dma =
ioa_cfg->hostrcb_dma[i] + offsetof(struct ipr_hostrcb, hcam);
[SCSI] ipr: Add support for logging SAS fabric errors Adds support for logging SAS fabric errors logged by the ipr firmware. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:35 -07:00
ioa_cfg->hostrcb[i]->ioa_cfg = ioa_cfg;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_add_tail(&ioa_cfg->hostrcb[i]->queue, &ioa_cfg->hostrcb_free_q);
}
treewide: kzalloc() -> kcalloc() The kzalloc() function has a 2-factor argument form, kcalloc(). This patch replaces cases of: kzalloc(a * b, gfp) with: kcalloc(a * b, gfp) as well as handling cases of: kzalloc(a * b * c, gfp) with: kzalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kzalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kzalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kzalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kzalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kzalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(char) * COUNT + COUNT , ...) | kzalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kzalloc + kcalloc ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kzalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kzalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kzalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kzalloc(C1 * C2 * C3, ...) | kzalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kzalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kzalloc(sizeof(THING) * C2, ...) | kzalloc(sizeof(TYPE) * C2, ...) | kzalloc(C1 * C2 * C3, ...) | kzalloc(C1 * C2, ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - (E1) * E2 + E1, E2 , ...) | - kzalloc + kcalloc ( - (E1) * (E2) + E1, E2 , ...) | - kzalloc + kcalloc ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 15:03:40 -06:00
ioa_cfg->trace = kcalloc(IPR_NUM_TRACE_ENTRIES,
sizeof(struct ipr_trace_entry),
GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ioa_cfg->trace)
goto out_free_hostrcb_dma;
rc = 0;
out:
LEAVE;
return rc;
out_free_hostrcb_dma:
while (i-- > 0) {
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&pdev->dev, sizeof(struct ipr_hostrcb),
ioa_cfg->hostrcb[i],
ioa_cfg->hostrcb_dma[i]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&pdev->dev, ioa_cfg->cfg_table_size,
ioa_cfg->u.cfg_table, ioa_cfg->cfg_table_dma);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
out_free_host_rrq:
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&pdev->dev,
sizeof(u32) * ioa_cfg->hrrq[i].size,
ioa_cfg->hrrq[i].host_rrq,
ioa_cfg->hrrq[i].host_rrq_dma);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
out_ipr_free_cmd_blocks:
ipr_free_cmd_blks(ioa_cfg);
out_free_vpd_cbs:
ipr: convert to generic DMA API Even though the ipr driver is only used on PCI, convert it to use the generic DMA API. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:08 -06:00
dma_free_coherent(&pdev->dev, sizeof(struct ipr_misc_cbs),
ioa_cfg->vpd_cbs, ioa_cfg->vpd_cbs_dma);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
out_free_res_entries:
kfree(ioa_cfg->res_entries);
goto out;
}
/**
* ipr_initialize_bus_attr - Initialize SCSI bus attributes to default values
* @ioa_cfg: ioa config struct
*
* Return value:
* none
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static void ipr_initialize_bus_attr(struct ipr_ioa_cfg *ioa_cfg)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
int i;
for (i = 0; i < IPR_MAX_NUM_BUSES; i++) {
ioa_cfg->bus_attr[i].bus = i;
ioa_cfg->bus_attr[i].qas_enabled = 0;
ioa_cfg->bus_attr[i].bus_width = IPR_DEFAULT_BUS_WIDTH;
if (ipr_max_speed < ARRAY_SIZE(ipr_max_bus_speeds))
ioa_cfg->bus_attr[i].max_xfer_rate = ipr_max_bus_speeds[ipr_max_speed];
else
ioa_cfg->bus_attr[i].max_xfer_rate = IPR_U160_SCSI_RATE;
}
}
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
/**
* ipr_init_regs - Initialize IOA registers
* @ioa_cfg: ioa config struct
*
* Return value:
* none
**/
static void ipr_init_regs(struct ipr_ioa_cfg *ioa_cfg)
{
const struct ipr_interrupt_offsets *p;
struct ipr_interrupts *t;
void __iomem *base;
p = &ioa_cfg->chip_cfg->regs;
t = &ioa_cfg->regs;
base = ioa_cfg->hdw_dma_regs;
t->set_interrupt_mask_reg = base + p->set_interrupt_mask_reg;
t->clr_interrupt_mask_reg = base + p->clr_interrupt_mask_reg;
t->clr_interrupt_mask_reg32 = base + p->clr_interrupt_mask_reg32;
t->sense_interrupt_mask_reg = base + p->sense_interrupt_mask_reg;
t->sense_interrupt_mask_reg32 = base + p->sense_interrupt_mask_reg32;
t->clr_interrupt_reg = base + p->clr_interrupt_reg;
t->clr_interrupt_reg32 = base + p->clr_interrupt_reg32;
t->sense_interrupt_reg = base + p->sense_interrupt_reg;
t->sense_interrupt_reg32 = base + p->sense_interrupt_reg32;
t->ioarrin_reg = base + p->ioarrin_reg;
t->sense_uproc_interrupt_reg = base + p->sense_uproc_interrupt_reg;
t->sense_uproc_interrupt_reg32 = base + p->sense_uproc_interrupt_reg32;
t->set_uproc_interrupt_reg = base + p->set_uproc_interrupt_reg;
t->set_uproc_interrupt_reg32 = base + p->set_uproc_interrupt_reg32;
t->clr_uproc_interrupt_reg = base + p->clr_uproc_interrupt_reg;
t->clr_uproc_interrupt_reg32 = base + p->clr_uproc_interrupt_reg32;
if (ioa_cfg->sis64) {
t->init_feedback_reg = base + p->init_feedback_reg;
t->dump_addr_reg = base + p->dump_addr_reg;
t->dump_data_reg = base + p->dump_data_reg;
t->endian_swap_reg = base + p->endian_swap_reg;
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_init_ioa_cfg - Initialize IOA config struct
* @ioa_cfg: ioa config struct
* @host: scsi host struct
* @pdev: PCI dev struct
*
* Return value:
* none
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static void ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg,
struct Scsi_Host *host, struct pci_dev *pdev)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
int i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->host = host;
ioa_cfg->pdev = pdev;
ioa_cfg->log_level = ipr_log_level;
[SCSI] ipr: Runtime reset Some IPR RAID adapter will automatically create single device RAID arrays for all attached devices when the card is initialized. Setting the RUNTIME_RESET doorbell bit will prevent this from occurring, since we only want this behavior the first time the card is initialized and not each time the card happens to get reset. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:01:54 -07:00
ioa_cfg->doorbell = IPR_DOORBELL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
sprintf(ioa_cfg->eye_catcher, IPR_EYECATCHER);
sprintf(ioa_cfg->trace_start, IPR_TRACE_START_LABEL);
sprintf(ioa_cfg->cfg_table_start, IPR_CFG_TBL_START);
sprintf(ioa_cfg->resource_table_label, IPR_RES_TABLE_LABEL);
sprintf(ioa_cfg->ipr_hcam_label, IPR_HCAM_LABEL);
sprintf(ioa_cfg->ipr_cmd_label, IPR_CMD_LABEL);
INIT_LIST_HEAD(&ioa_cfg->hostrcb_free_q);
INIT_LIST_HEAD(&ioa_cfg->hostrcb_pending_q);
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
INIT_LIST_HEAD(&ioa_cfg->hostrcb_report_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
INIT_LIST_HEAD(&ioa_cfg->free_res_q);
INIT_LIST_HEAD(&ioa_cfg->used_res_q);
WorkStruct: make allyesconfig Fix up for make allyesconfig. Signed-Off-By: David Howells <dhowells@redhat.com>
2006-11-22 07:57:56 -07:00
INIT_WORK(&ioa_cfg->work_q, ipr_worker_thread);
scsi: ipr: System hung while dlpar adding primary ipr adapter back While dlpar adding primary ipr adapter back, driver goes through adapter initialization then schedule ipr_worker_thread to start te disk scan by dropping the host lock, calling scsi_add_device. Then get the adapter reset request again, so driver does scsi_block_requests, this will cause the scsi_add_device get hung until we unblock. But we can't run ipr_worker_thread to do the unblock because its stuck in scsi_add_device. This patch fixes the issue. [mkp: typo and whitespace fixes] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-09-20 18:32:12 -06:00
INIT_WORK(&ioa_cfg->scsi_add_work_q, ipr_add_remove_thread);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
init_waitqueue_head(&ioa_cfg->reset_wait_q);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
init_waitqueue_head(&ioa_cfg->msi_wait_q);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
init_waitqueue_head(&ioa_cfg->eeh_wait_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ioa_cfg->sdt_state = INACTIVE;
ipr_initialize_bus_attr(ioa_cfg);
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
ioa_cfg->max_devs_supported = ipr_max_devs;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
if (ioa_cfg->sis64) {
host->max_id = IPR_MAX_SIS64_TARGETS_PER_BUS;
host->max_lun = IPR_MAX_SIS64_LUNS_PER_TARGET;
if (ipr_max_devs > IPR_MAX_SIS64_DEVS)
ioa_cfg->max_devs_supported = IPR_MAX_SIS64_DEVS;
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ioa_cfg->cfg_table_size = (sizeof(struct ipr_config_table_hdr64)
+ ((sizeof(struct ipr_config_table_entry64)
* ioa_cfg->max_devs_supported)));
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
} else {
host->max_id = IPR_MAX_NUM_TARGETS_PER_BUS;
host->max_lun = IPR_MAX_NUM_LUNS_PER_TARGET;
if (ipr_max_devs > IPR_MAX_PHYSICAL_DEVS)
ioa_cfg->max_devs_supported = IPR_MAX_PHYSICAL_DEVS;
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ioa_cfg->cfg_table_size = (sizeof(struct ipr_config_table_hdr)
+ ((sizeof(struct ipr_config_table_entry)
* ioa_cfg->max_devs_supported)));
[SCSI] ipr: update the configuration table code for the next generation chip This patch changes the configuration table structures and related code such that both 32 bit and 64 bit based adapters can work with the driver. This patch also implements the code to generate the virtual bus/id/lun values for devices connected to the new adapters. It also implements support for the new device resource path. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:59 -07:00
}
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ipr: add support for async scanning to speed up boot Switch device scanning logic in the ipr driver to use the async scan API. This speeds up boot times, particularly on large systems. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong<wenxiong@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-12-02 11:47:37 -07:00
host->max_channel = IPR_VSET_BUS;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
host->unique_id = host->host_no;
host->max_cmd_len = IPR_MAX_CDB_LEN;
[SCSI] ipr: Increase max concurrent oustanding commands Increase the total number of max concurrent outstanding commands for the most recent family of adapters in order to improve overall adapter performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:10 -06:00
host->can_queue = ioa_cfg->max_cmds;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
pci_set_drvdata(pdev, ioa_cfg);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
for (i = 0; i < ARRAY_SIZE(ioa_cfg->hrrq); i++) {
INIT_LIST_HEAD(&ioa_cfg->hrrq[i].hrrq_free_q);
INIT_LIST_HEAD(&ioa_cfg->hrrq[i].hrrq_pending_q);
spin_lock_init(&ioa_cfg->hrrq[i]._lock);
if (i == 0)
ioa_cfg->hrrq[i].lock = ioa_cfg->host->host_lock;
else
ioa_cfg->hrrq[i].lock = &ioa_cfg->hrrq[i]._lock;
[SCSI] ipr: add hardware assisted smart dump functionality This patch adds the hardware assisted smart dump functionality for the next generation IOA PCI interface chip. Signea-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:14 -07:00
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
* ipr_get_chip_info - Find adapter chip information
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @dev_id: PCI device id struct
*
* Return value:
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
* ptr to chip information on success / NULL on failure
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static const struct ipr_chip_t *
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
ipr_get_chip_info(const struct pci_device_id *dev_id)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
int i;
for (i = 0; i < ARRAY_SIZE(ipr_chip); i++)
if (ipr_chip[i].vendor == dev_id->vendor &&
ipr_chip[i].device == dev_id->device)
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
return &ipr_chip[i];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return NULL;
}
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
/**
* ipr_wait_for_pci_err_recovery - Wait for any PCI error recovery to complete
* during probe time
* @ioa_cfg: ioa config struct
*
* Return value:
* None
**/
static void ipr_wait_for_pci_err_recovery(struct ipr_ioa_cfg *ioa_cfg)
{
struct pci_dev *pdev = ioa_cfg->pdev;
if (pci_channel_offline(pdev)) {
wait_event_timeout(ioa_cfg->eeh_wait_q,
!pci_channel_offline(pdev),
IPR_PCI_ERROR_RECOVERY_TIMEOUT);
pci_restore_state(pdev);
}
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
static void name_msi_vectors(struct ipr_ioa_cfg *ioa_cfg)
{
int vec_idx, n = sizeof(ioa_cfg->vectors_info[0].desc) - 1;
for (vec_idx = 0; vec_idx < ioa_cfg->nvectors; vec_idx++) {
snprintf(ioa_cfg->vectors_info[vec_idx].desc, n,
"host%d-%d", ioa_cfg->host->host_no, vec_idx);
ioa_cfg->vectors_info[vec_idx].
desc[strlen(ioa_cfg->vectors_info[vec_idx].desc)] = 0;
}
}
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
static int ipr_request_other_msi_irqs(struct ipr_ioa_cfg *ioa_cfg,
struct pci_dev *pdev)
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
{
int i, rc;
for (i = 1; i < ioa_cfg->nvectors; i++) {
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
rc = request_irq(pci_irq_vector(pdev, i),
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ipr_isr_mhrrq,
0,
ioa_cfg->vectors_info[i].desc,
&ioa_cfg->hrrq[i]);
if (rc) {
while (--i >= 0)
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
free_irq(pci_irq_vector(pdev, i),
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
&ioa_cfg->hrrq[i]);
return rc;
}
}
return 0;
}
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
/**
* ipr_test_intr - Handle the interrupt generated in ipr_test_msi().
* @pdev: PCI device struct
*
* Description: Simply set the msi_received flag to 1 indicating that
* Message Signaled Interrupts are supported.
*
* Return value:
* 0 on success / non-zero on failure
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static irqreturn_t ipr_test_intr(int irq, void *devp)
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
{
struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)devp;
unsigned long lock_flags = 0;
irqreturn_t rc = IRQ_HANDLED;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
dev_info(&ioa_cfg->pdev->dev, "Received IRQ : %d\n", irq);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ioa_cfg->msi_received = 1;
wake_up(&ioa_cfg->msi_wait_q);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
return rc;
}
/**
* ipr_test_msi - Test for Message Signaled Interrupt (MSI) support.
* @pdev: PCI device struct
*
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
* Description: This routine sets up and initiates a test interrupt to determine
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
* if the interrupt is received via the ipr_test_intr() service routine.
* If the tests fails, the driver will fall back to LSI.
*
* Return value:
* 0 on success / non-zero on failure
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static int ipr_test_msi(struct ipr_ioa_cfg *ioa_cfg, struct pci_dev *pdev)
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
{
int rc;
volatile u32 int_reg;
unsigned long lock_flags = 0;
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
int irq = pci_irq_vector(pdev, 0);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
init_waitqueue_head(&ioa_cfg->msi_wait_q);
ioa_cfg->msi_received = 0;
ipr_mask_and_clear_interrupts(ioa_cfg, ~IPR_PCII_IOA_TRANS_TO_OPER);
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
writel(IPR_PCII_IO_DEBUG_ACKNOWLEDGE, ioa_cfg->regs.clr_interrupt_mask_reg32);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
rc = request_irq(irq, ipr_test_intr, 0, IPR_NAME, ioa_cfg);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
if (rc) {
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
dev_err(&pdev->dev, "Can not assign irq %d\n", irq);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
return rc;
} else if (ipr_debug)
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
dev_info(&pdev->dev, "IRQ assigned: %d\n", irq);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
writel(IPR_PCII_IO_DEBUG_ACKNOWLEDGE, ioa_cfg->regs.sense_interrupt_reg32);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
int_reg = readl(ioa_cfg->regs.sense_interrupt_reg);
wait_event_timeout(ioa_cfg->msi_wait_q, ioa_cfg->msi_received, HZ);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
ipr_mask_and_clear_interrupts(ioa_cfg, ~IPR_PCII_IOA_TRANS_TO_OPER);
if (!ioa_cfg->msi_received) {
/* MSI test failed */
dev_info(&pdev->dev, "MSI test failed. Falling back to LSI.\n");
rc = -EOPNOTSUPP;
} else if (ipr_debug)
dev_info(&pdev->dev, "MSI test succeeded.\n");
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
free_irq(irq, ioa_cfg);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
LEAVE;
return rc;
}
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
/* ipr_probe_ioa - Allocates memory and does first stage of initialization
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
* @pdev: PCI device struct
* @dev_id: PCI device id struct
*
* Return value:
* 0 on success / non-zero on failure
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static int ipr_probe_ioa(struct pci_dev *pdev,
const struct pci_device_id *dev_id)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioa_cfg *ioa_cfg;
struct Scsi_Host *host;
unsigned long ipr_regs_pci;
void __iomem *ipr_regs;
[SCSI] Signedness issue in drivers/scsi/ipr.c gcc 4.1 with some extra warnings show the following: drivers/scsi/ipr.c:6361: warning: comparison of unsigned expression < 0 is always false drivers/scsi/ipr.c:6385: warning: comparison of unsigned expression < 0 is always false drivers/scsi/ipr.c:6415: warning: comparison of unsigned expression < 0 is always false The problem is that rc is of the type u32, which can never be smaller than zero, therefore all three error handling checks get useless. This patch changes it to a normal int, because all usages / all functions it get used with expect an int. Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de> Acked-by: Brian King <brking@us.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 17:59:07 -06:00
int rc = PCIBIOS_SUCCESSFUL;
[SCSI] ipr: Better handle adapter boot time errors If an ipr adapter encounters an adapter error requiring an adapter reset to recover from prior to driver load time, the error will be ignored and recovery will not happen until the initial timeout occurs waiting for the firmware to come ready, which means a five minute timeout. Fix is to read the interrupt register before clearing any of the interrupts at probe time. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-02 09:44:11 -06:00
volatile u32 mask, uproc, interrupts;
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
unsigned long lock_flags, driver_lock_flags;
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
unsigned int irq_flag;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
dev_info(&pdev->dev, "Found IOA with IRQ: %d\n", pdev->irq);
host = scsi_host_alloc(&driver_template, sizeof(*ioa_cfg));
if (!host) {
dev_err(&pdev->dev, "call to scsi_host_alloc failed!\n");
rc = -ENOMEM;
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ioa_cfg = (struct ipr_ioa_cfg *)host->hostdata;
memset(ioa_cfg, 0, sizeof(struct ipr_ioa_cfg));
[SCSI] libsas, ipr: cleanup ata_host flags initialization via ata_host_init libsas and ipr pass flags to ata_host_init that are meant for the port. ata_host flags: ATA_HOST_SIMPLEX = (1 << 0), /* Host is simplex, one DMA channel per host only */ ATA_HOST_STARTED = (1 << 1), /* Host started */ ATA_HOST_PARALLEL_SCAN = (1 << 2), /* Ports on this host can be scanned in parallel */ ATA_HOST_IGNORE_ATA = (1 << 3), /* Ignore ATA devices on this host. */ flags passed by libsas: ATA_FLAG_SATA = (1 << 1), ATA_FLAG_PIO_DMA = (1 << 7), /* PIO cmds via DMA */ ATA_FLAG_NCQ = (1 << 10), /* host supports NCQ */ The only one that aliases is ATA_HOST_STARTED which is a 'don't care' in the libsas and ipr cases since ata_hosts from these sources are not registered with libata. Reported-by: Hannes Reinecke <hare@suse.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Acked-by: Brian King <brking@us.ibm.com> Acked-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-07-09 22:06:08 -06:00
ata_host_init(&ioa_cfg->ata_host, &pdev->dev, &ipr_sata_ops);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
ioa_cfg->ipr_chip = ipr_get_chip_info(dev_id);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
if (!ioa_cfg->ipr_chip) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
dev_err(&pdev->dev, "Unknown adapter chipset 0x%04X 0x%04X\n",
dev_id->vendor, dev_id->device);
goto out_scsi_host_put;
}
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
/* set SIS 32 or SIS 64 */
ioa_cfg->sis64 = ioa_cfg->ipr_chip->sis_type == IPR_SIS64 ? 1 : 0;
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
ioa_cfg->chip_cfg = ioa_cfg->ipr_chip->cfg;
[SCSI] ipr: Remove unnecessary interrupt clearing on new adapters The latest ipr hardware no longer requires the driver to issue any MMIOs to clear the interrupt so remove this to optimize performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:08 -06:00
ioa_cfg->clear_isr = ioa_cfg->chip_cfg->clear_isr;
[SCSI] ipr: Increase max concurrent oustanding commands Increase the total number of max concurrent outstanding commands for the most recent family of adapters in order to improve overall adapter performance. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-03-14 20:20:10 -06:00
ioa_cfg->max_cmds = ioa_cfg->chip_cfg->max_cmds;
ipr: differentiate pci-x and pci-e based adapters MSI has only been tested on and known to work with PCI-E based adapters. This patch adds a field to struct ipr_chip_t to indicate which type of interrupt to use based on what is known about the chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-17 10:55:35 -06:00
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
if (ipr_transop_timeout)
ioa_cfg->transop_timeout = ipr_transop_timeout;
else if (dev_id->driver_data & IPR_USE_LONG_TRANSOP_TIMEOUT)
ioa_cfg->transop_timeout = IPR_LONG_OPERATIONAL_TIMEOUT;
else
ioa_cfg->transop_timeout = IPR_OPERATIONAL_TIMEOUT;
PCI: Change all drivers to use pci_device->revision Instead of all drivers reading pci config space to get the revision ID, they can now use the pci_device->revision member. This exposes some issues where drivers where reading a word or a dword for the revision number, and adding useless error-handling around the read. Some drivers even just read it for no purpose of all. In devices where the revision ID is being copied over and used in what appears to be the equivalent of hotpath, I have left the copy code and the cached copy as not to influence the driver's performance. Compile tested with make all{yes,mod}config on x86_64 and i386. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Acked-by: Dave Jones <davej@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2007-06-08 16:46:36 -06:00
ioa_cfg->revid = pdev->revision;
[SCSI] ipr: Use PCI-E reset API for new ipr adapter Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-07 16:09:05 -06:00
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ipr_init_ioa_cfg(ioa_cfg, host, pdev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_regs_pci = pci_resource_start(pdev, 0);
rc = pci_request_regions(pdev, IPR_NAME);
if (rc < 0) {
dev_err(&pdev->dev,
"Couldn't register memory range of registers\n");
goto out_scsi_host_put;
}
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
rc = pci_enable_device(pdev);
if (rc || pci_channel_offline(pdev)) {
if (pci_channel_offline(pdev)) {
ipr_wait_for_pci_err_recovery(ioa_cfg);
rc = pci_enable_device(pdev);
}
if (rc) {
dev_err(&pdev->dev, "Cannot enable adapter\n");
ipr_wait_for_pci_err_recovery(ioa_cfg);
goto out_release_regions;
}
}
[SCSI] advansys, arcmsr, ipr, nsp32, qla1280, stex: use pci_ioremap_bar() Use the newly introduced pci_ioremap_bar() function in drivers/scsi. pci_ioremap_bar() just takes a pci device and a bar number, with the goal of making it really hard to get wrong, while also having a central place to stick sanity checks. Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Matthew Wilcox <willy@linux.intel.com> Cc: Brian King <brking@us.ibm.com> Cc: Ed Lin <ed.lin@promise.com> Cc: Nick Cheng <nick.cheng@areca.com.tw> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-09-28 17:18:02 -06:00
ipr_regs = pci_ioremap_bar(pdev, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (!ipr_regs) {
dev_err(&pdev->dev,
"Couldn't map memory range of registers\n");
rc = -ENOMEM;
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
goto out_disable;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
ioa_cfg->hdw_dma_regs = ipr_regs;
ioa_cfg->hdw_dma_regs_pci = ipr_regs_pci;
ioa_cfg->ioa_mailbox = ioa_cfg->chip_cfg->mailbox + ipr_regs;
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ipr_init_regs(ioa_cfg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (ioa_cfg->sis64) {
ipr: set coherent DMA mask Use dma_set_mask_and_coherent() to set both the DMA and coherent DMA mask. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:09 -06:00
rc = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
if (rc < 0) {
ipr: set coherent DMA mask Use dma_set_mask_and_coherent() to set both the DMA and coherent DMA mask. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:09 -06:00
dev_dbg(&pdev->dev, "Failed to set 64 bit DMA mask\n");
rc = dma_set_mask_and_coherent(&pdev->dev,
DMA_BIT_MASK(32));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
}
} else
ipr: set coherent DMA mask Use dma_set_mask_and_coherent() to set both the DMA and coherent DMA mask. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:09 -06:00
rc = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
[SCSI] ipr: add support for new adapter command structures for the next generation chip Change the adapter command structures such that both 32 bit and 64 bit based adapters can work with the driver. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:23:36 -07:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (rc < 0) {
ipr: set coherent DMA mask Use dma_set_mask_and_coherent() to set both the DMA and coherent DMA mask. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2014-10-30 16:27:09 -06:00
dev_err(&pdev->dev, "Failed to set DMA mask\n");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
goto cleanup_nomem;
}
rc = pci_write_config_byte(pdev, PCI_CACHE_LINE_SIZE,
ioa_cfg->chip_cfg->cache_line_size);
if (rc != PCIBIOS_SUCCESSFUL) {
dev_err(&pdev->dev, "Write of cache line size failed\n");
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ipr_wait_for_pci_err_recovery(ioa_cfg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
rc = -EIO;
goto cleanup_nomem;
}
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
/* Issue MMIO read to ensure card is not in EEH */
interrupts = readl(ioa_cfg->regs.sense_interrupt_reg);
ipr_wait_for_pci_err_recovery(ioa_cfg);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
if (ipr_number_of_msix > IPR_MAX_MSIX_VECTORS) {
dev_err(&pdev->dev, "The max number of MSIX is %d\n",
IPR_MAX_MSIX_VECTORS);
ipr_number_of_msix = IPR_MAX_MSIX_VECTORS;
}
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
irq_flag = PCI_IRQ_LEGACY;
if (ioa_cfg->ipr_chip->has_msi)
irq_flag |= PCI_IRQ_MSI | PCI_IRQ_MSIX;
rc = pci_alloc_irq_vectors(pdev, 1, ipr_number_of_msix, irq_flag);
if (rc < 0) {
ipr_wait_for_pci_err_recovery(ioa_cfg);
goto cleanup_nomem;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
ioa_cfg->nvectors = rc;
if (!pdev->msi_enabled && !pdev->msix_enabled)
ioa_cfg->clear_isr = 1;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
pci_set_master(pdev);
if (pci_channel_offline(pdev)) {
ipr_wait_for_pci_err_recovery(ioa_cfg);
pci_set_master(pdev);
if (pci_channel_offline(pdev)) {
rc = -EIO;
goto out_msi_disable;
}
}
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
if (pdev->msi_enabled || pdev->msix_enabled) {
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
rc = ipr_test_msi(ioa_cfg, pdev);
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
switch (rc) {
case 0:
dev_info(&pdev->dev,
"Request for %d MSI%ss succeeded.", ioa_cfg->nvectors,
pdev->msix_enabled ? "-X" : "");
break;
case -EOPNOTSUPP:
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ipr_wait_for_pci_err_recovery(ioa_cfg);
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
pci_free_irq_vectors(pdev);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ioa_cfg->nvectors = 1;
scsi: ipr: Fix runaway IRQs when falling back from MSI to LSI LSIs must be ack'ed with an MMIO otherwise they remain asserted forever. This is controlled by the "clear_isr" flag. While we set that flag properly when deciding initially whether to use LSIs or MSIs, we fail to set it if we first chose MSIs, the test fails, then fallback to LSIs. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-11-30 14:28:55 -07:00
ioa_cfg->clear_isr = 1;
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
break;
default:
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
goto out_msi_disable;
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
}
}
ioa_cfg->hrrq_num = min3(ioa_cfg->nvectors,
(unsigned int)num_online_cpus(),
(unsigned int)IPR_MAX_HRRQ_NUM);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if ((rc = ipr_save_pcix_cmd_reg(ioa_cfg)))
[SCSI] ipr: reorder error handling code to include iounmap The out_msi_disable label should be before cleanup_nomem to additionally benefit from the call to iounmap. Subsequent gotos are adjusted to go to out_msi_disable instead of cleanup_nomem, which now follows it. This is safe because pci_disable_msi does nothing if pci_enable_msi was not called. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @r@ expression e1,e2; statement S; @@ e1 = pci_ioremap_bar(...); ... when != e1 = e2 when != iounmap(e1) when any ( if (<+...e1...+>) S | if(...) { ... return 0; } | if (...) { ... when != iounmap(e1) when != if (...) { ... iounmap(e1) ... } * return ...; } else S ) // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-07-11 15:08:25 -06:00
goto out_msi_disable;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if ((rc = ipr_set_pcix_cmd_reg(ioa_cfg)))
[SCSI] ipr: reorder error handling code to include iounmap The out_msi_disable label should be before cleanup_nomem to additionally benefit from the call to iounmap. Subsequent gotos are adjusted to go to out_msi_disable instead of cleanup_nomem, which now follows it. This is safe because pci_disable_msi does nothing if pci_enable_msi was not called. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @r@ expression e1,e2; statement S; @@ e1 = pci_ioremap_bar(...); ... when != e1 = e2 when != iounmap(e1) when any ( if (<+...e1...+>) S | if(...) { ... return 0; } | if (...) { ... when != iounmap(e1) when != if (...) { ... iounmap(e1) ... } * return ...; } else S ) // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-07-11 15:08:25 -06:00
goto out_msi_disable;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
rc = ipr_alloc_mem(ioa_cfg);
if (rc < 0) {
dev_err(&pdev->dev,
"Couldn't allocate enough memory for device driver!\n");
[SCSI] ipr: reorder error handling code to include iounmap The out_msi_disable label should be before cleanup_nomem to additionally benefit from the call to iounmap. Subsequent gotos are adjusted to go to out_msi_disable instead of cleanup_nomem, which now follows it. This is safe because pci_disable_msi does nothing if pci_enable_msi was not called. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @r@ expression e1,e2; statement S; @@ e1 = pci_ioremap_bar(...); ... when != e1 = e2 when != iounmap(e1) when any ( if (<+...e1...+>) S | if(...) { ... return 0; } | if (...) { ... when != iounmap(e1) when != if (...) { ... iounmap(e1) ... } * return ...; } else S ) // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-07-11 15:08:25 -06:00
goto out_msi_disable;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
/* Save away PCI config space for use following IOA reset */
rc = pci_save_state(pdev);
if (rc != PCIBIOS_SUCCESSFUL) {
dev_err(&pdev->dev, "Failed to save PCI config space\n");
rc = -EIO;
goto cleanup_nolog;
}
[SCSI] ipr: Driver initialization fix for kexec/kdump When kexec booting a kernel when the previous kernel did not call ipr's shutdown method, the ipr adapter does not get properly initialized, which can result in the ipr adapter completing commands issued by the previous kernel. Fix ipr to detect this scenario by reading the adapter's interrupt mask register and the microprocessor interrupt register. If the interrupt mask register indicates that interrupts are enabled or the reset alert bit is set when the card is probed, this means the card is in an unknown state and we hard reset the card. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-17 08:35:12 -07:00
/*
* If HRRQ updated interrupt is not masked, or reset alert is set,
* the card is in an unknown state and needs a hard reset
*/
[SCSI] ipr: add support for multiple stages of initialization This patch adds support for using the new IOA initialization feedback register. It also enables 64 bit support in the ipr_ioafp_identify_hrrq and ipr_mask_and_clear_interrupts routines. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:26 -07:00
mask = readl(ioa_cfg->regs.sense_interrupt_mask_reg32);
interrupts = readl(ioa_cfg->regs.sense_interrupt_reg32);
uproc = readl(ioa_cfg->regs.sense_uproc_interrupt_reg32);
[SCSI] ipr: Driver initialization fix for kexec/kdump When kexec booting a kernel when the previous kernel did not call ipr's shutdown method, the ipr adapter does not get properly initialized, which can result in the ipr adapter completing commands issued by the previous kernel. Fix ipr to detect this scenario by reading the adapter's interrupt mask register and the microprocessor interrupt register. If the interrupt mask register indicates that interrupts are enabled or the reset alert bit is set when the card is probed, this means the card is in an unknown state and we hard reset the card. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-17 08:35:12 -07:00
if ((mask & IPR_PCII_HRRQ_UPDATED) == 0 || (uproc & IPR_UPROCI_RESET_ALERT))
ioa_cfg->needs_hard_reset = 1;
[SCSI] ipr: Always initiate hard reset in kdump kernel During kdump testing I noticed timeouts when initialising each IPR adapter. While the driver has logic to detect an adapter in an indeterminate state, it wasn't triggering and each adapter went through a 5 minute timeout before finally going operational. Some analysis showed the needs_hard_reset flag wasn't getting set. We can check the reset_devices kernel parameter which is set by kdump and force a full reset. This fixes the problem. Signed-off-by: Anton Blanchard <anton@samba.org> Cc: <stable@kernel.org> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-08-01 03:43:45 -06:00
if ((interrupts & IPR_PCII_ERROR_INTERRUPTS) || reset_devices)
[SCSI] ipr: Better handle adapter boot time errors If an ipr adapter encounters an adapter error requiring an adapter reset to recover from prior to driver load time, the error will be ignored and recovery will not happen until the initial timeout occurs waiting for the firmware to come ready, which means a five minute timeout. Fix is to read the interrupt register before clearing any of the interrupts at probe time. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-02 09:44:11 -06:00
ioa_cfg->needs_hard_reset = 1;
if (interrupts & IPR_PCII_IOA_UNIT_CHECKED)
ioa_cfg->ioa_unit_checked = 1;
[SCSI] ipr: Driver initialization fix for kexec/kdump When kexec booting a kernel when the previous kernel did not call ipr's shutdown method, the ipr adapter does not get properly initialized, which can result in the ipr adapter completing commands issued by the previous kernel. Fix ipr to detect this scenario by reading the adapter's interrupt mask register and the microprocessor interrupt register. If the interrupt mask register indicates that interrupts are enabled or the reset alert bit is set when the card is probed, this means the card is in an unknown state and we hard reset the card. Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-17 08:35:12 -07:00
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_mask_and_clear_interrupts(ioa_cfg, ~IPR_PCII_IOA_TRANS_TO_OPER);
[SCSI] ipr: Reduce lock contention This patch reduces lock contention while implementing distributed completion processing. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:51 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
if (pdev->msi_enabled || pdev->msix_enabled) {
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
name_msi_vectors(ioa_cfg);
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
rc = request_irq(pci_irq_vector(pdev, 0), ipr_isr, 0,
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
ioa_cfg->vectors_info[0].desc,
&ioa_cfg->hrrq[0]);
if (!rc)
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
rc = ipr_request_other_msi_irqs(ioa_cfg, pdev);
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
} else {
rc = request_irq(pdev->irq, ipr_isr,
IRQF_SHARED,
IPR_NAME, &ioa_cfg->hrrq[0]);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (rc) {
dev_err(&pdev->dev, "Couldn't register IRQ %d! rc=%d\n",
pdev->irq, rc);
goto cleanup_nolog;
}
[SCSI] ipr: Use PCI-E reset API for new ipr adapter Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-07 16:09:05 -06:00
if ((dev_id->driver_data & IPR_USE_PCI_WARM_RESET) ||
(dev_id->device == PCI_DEVICE_ID_IBM_OBSIDIAN_E && !ioa_cfg->revid)) {
ioa_cfg->needs_warm_reset = 1;
ioa_cfg->reset = ipr_reset_slot_reset;
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
ioa_cfg->reset_work_q = alloc_ordered_workqueue("ipr_reset_%d",
WQ_MEM_RECLAIM, host->host_no);
if (!ioa_cfg->reset_work_q) {
dev_err(&pdev->dev, "Couldn't register reset workqueue\n");
ipr: Fix error return code in ipr_probe_ioa() Fix to return error code -ENOMEM from the workqueue alloc error handling case instead of 0, as done elsewhere in this function. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-07-29 10:00:45 -06:00
rc = -ENOMEM;
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
goto out_free_irq;
}
[SCSI] ipr: Use PCI-E reset API for new ipr adapter Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-07 16:09:05 -06:00
} else
ioa_cfg->reset = ipr_reset_start_bist;
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
spin_lock_irqsave(&ipr_driver_lock, driver_lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_add_tail(&ioa_cfg->queue, &ipr_ioa_head);
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
spin_unlock_irqrestore(&ipr_driver_lock, driver_lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
LEAVE;
out:
return rc;
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
out_free_irq:
ipr_free_irqs(ioa_cfg);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
cleanup_nolog:
ipr_free_mem(ioa_cfg);
ipr: add test for MSI interrupt support The return value from pci_enable_msi() can not always be trusted. This patch adds code to generate an interrupt after MSI has been enabled and tests whether or not we can receive and process it. If the tests fails, then fall back to LSI. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2009-06-16 16:13:28 -06:00
out_msi_disable:
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
ipr_wait_for_pci_err_recovery(ioa_cfg);
scsi: ipr: Use pci_irq_allocate_vectors Switch the ipr driver to use pci_alloc_irq_vectors. We need to two calls to pci_alloc_irq_vectors as ipr only supports multiple MSI-X vectors, but not multiple MSI vectors. Otherwise this cleans up a lot of cruft and allows to use a common request_irq loop for irq types, which happens to only iterate over a single line in the non MSI-X case. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-09-11 07:31:24 -06:00
pci_free_irq_vectors(pdev);
[SCSI] ipr: reorder error handling code to include iounmap The out_msi_disable label should be before cleanup_nomem to additionally benefit from the call to iounmap. Subsequent gotos are adjusted to go to out_msi_disable instead of cleanup_nomem, which now follows it. This is safe because pci_disable_msi does nothing if pci_enable_msi was not called. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @r@ expression e1,e2; statement S; @@ e1 = pci_ioremap_bar(...); ... when != e1 = e2 when != iounmap(e1) when any ( if (<+...e1...+>) S | if(...) { ... return 0; } | if (...) { ... when != iounmap(e1) when != if (...) { ... iounmap(e1) ... } * return ...; } else S ) // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-07-11 15:08:25 -06:00
cleanup_nomem:
iounmap(ipr_regs);
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
out_disable:
pci_disable_device(pdev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
out_release_regions:
pci_release_regions(pdev);
out_scsi_host_put:
scsi_host_put(host);
goto out;
}
/**
* ipr_initiate_ioa_bringdown - Bring down an adapter
* @ioa_cfg: ioa config struct
* @shutdown_type: shutdown type
*
* Description: This function will initiate bringing down the adapter.
* This consists of issuing an IOA shutdown to the adapter
* to flush the cache, and running BIST.
* If the caller needs to wait on the completion of the reset,
* the caller must sleep on the reset_wait_q.
*
* Return value:
* none
**/
static void ipr_initiate_ioa_bringdown(struct ipr_ioa_cfg *ioa_cfg,
enum ipr_shutdown_type shutdown_type)
{
ENTER;
if (ioa_cfg->sdt_state == WAIT_FOR_DUMP)
ioa_cfg->sdt_state = ABORT_DUMP;
ioa_cfg->reset_retries = 0;
ioa_cfg->in_ioa_bringdown = 1;
ipr_initiate_ioa_reset(ioa_cfg, shutdown_type);
LEAVE;
}
/**
* __ipr_remove - Remove a single adapter
* @pdev: pci device struct
*
* Adapter hot plug remove entry point.
*
* Return value:
* none
**/
static void __ipr_remove(struct pci_dev *pdev)
{
unsigned long host_lock_flags = 0;
struct ipr_ioa_cfg *ioa_cfg = pci_get_drvdata(pdev);
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
int i;
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
unsigned long driver_lock_flags;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ENTER;
spin_lock_irqsave(ioa_cfg->host->host_lock, host_lock_flags);
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
while (ioa_cfg->in_reset_reload) {
[SCSI] ipr: Prevent overlapped adapter resets This patch fixes some scenarios where an ipr adapter could get reset overlapped, which could cause very long timeouts to occur, or PCI bus errors. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:06 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, host_lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, host_lock_flags);
}
[SCSI] ipr: Fix oops while resetting an ipr adapter When resetting an ipr adapter, we use scsi_block_requests to block any new commands from scsi core, and then unblock after the reset. When hotplug removing an adapter, we shut it down and go through this same code, but we've seen issues with scsi_unblock_requests running after the adapter's memory has been freed. There is really no need to block/unblock when the adapter is being removed, so this patch skips the block/unblock and will immediately fail any commands that happen to make it to queuecommand while the adapter is being shutdown. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-30 22:45:08 -07:00
for (i = 0; i < ioa_cfg->hrrq_num; i++) {
spin_lock(&ioa_cfg->hrrq[i]._lock);
ioa_cfg->hrrq[i].removing_ioa = 1;
spin_unlock(&ioa_cfg->hrrq[i]._lock);
}
wmb();
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
ipr_initiate_ioa_bringdown(ioa_cfg, IPR_SHUTDOWN_NORMAL);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, host_lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
workqueue: deprecate flush[_delayed]_work_sync() flush[_delayed]_work_sync() are now spurious. Mark them deprecated and convert all users to flush[_delayed]_work(). If you're cc'd and wondering what's going on: Now all workqueues are non-reentrant and the regular flushes guarantee that the work item is not pending or running on any CPU on return, so there's no reason to use the sync flushes at all and they're going away. This patch doesn't make any functional difference. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Russell King <linux@arm.linux.org.uk> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Ian Campbell <ian.campbell@citrix.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Mattia Dongili <malattia@linux.it> Cc: Kent Yoder <key@linux.vnet.ibm.com> Cc: David Airlie <airlied@linux.ie> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Karsten Keil <isdn@linux-pingi.de> Cc: Bryan Wu <bryan.wu@canonical.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mauro Carvalho Chehab <mchehab@infradead.org> Cc: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: David Woodhouse <dwmw2@infradead.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: linux-wireless@vger.kernel.org Cc: Anton Vorontsov <cbou@mail.ru> Cc: Sangbeom Kim <sbkim73@samsung.com> Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Petr Vandrovec <petr@vandrovec.name> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Avi Kivity <avi@redhat.com>
2012-08-20 15:51:24 -06:00
flush_work(&ioa_cfg->work_q);
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
if (ioa_cfg->reset_work_q)
flush_workqueue(ioa_cfg->reset_work_q);
[SCSI] ipr: dlpar failed when adding an adapter back Reinitialize resource queue prior to freeing resource entries to ensure they are not referenced. This fixes an issue with target_destoy accessing memory after it was freed. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-03-14 12:52:24 -06:00
INIT_LIST_HEAD(&ioa_cfg->used_res_q);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_lock_irqsave(ioa_cfg->host->host_lock, host_lock_flags);
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
spin_lock_irqsave(&ipr_driver_lock, driver_lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
list_del(&ioa_cfg->queue);
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
spin_unlock_irqrestore(&ipr_driver_lock, driver_lock_flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
if (ioa_cfg->sdt_state == ABORT_DUMP)
ioa_cfg->sdt_state = WAIT_FOR_DUMP;
spin_unlock_irqrestore(ioa_cfg->host->host_lock, host_lock_flags);
ipr_free_all_resources(ioa_cfg);
LEAVE;
}
/**
* ipr_remove - IOA hot plug remove entry point
* @pdev: pci device struct
*
* Adapter hot plug remove entry point.
*
* Return value:
* none
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static void ipr_remove(struct pci_dev *pdev)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioa_cfg *ioa_cfg = pci_get_drvdata(pdev);
ENTER;
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
ipr_remove_trace_file(&ioa_cfg->host->shost_dev.kobj,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
&ipr_trace_attr);
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
ipr_remove_dump_file(&ioa_cfg->host->shost_dev.kobj,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
&ipr_dump_attr);
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
sysfs_remove_bin_file(&ioa_cfg->host->shost_dev.kobj,
&ipr_ioa_async_err_log);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
scsi_remove_host(ioa_cfg->host);
__ipr_remove(pdev);
LEAVE;
}
/**
* ipr_probe - Adapter hot plug add entry point
*
* Return value:
* 0 on success / non-zero on failure
**/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static int ipr_probe(struct pci_dev *pdev, const struct pci_device_id *dev_id)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
struct ipr_ioa_cfg *ioa_cfg;
ipr: Wait to do async scan until scsi host is initialized When performing an async scan, make sure the kthread doing scanning doesn't start before the scsi host is fully initialized. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-07-15 13:48:03 -06:00
unsigned long flags;
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
int rc, i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
rc = ipr_probe_ioa(pdev, dev_id);
if (rc)
return rc;
ioa_cfg = pci_get_drvdata(pdev);
rc = ipr_probe_ioa_part2(ioa_cfg);
if (rc) {
__ipr_remove(pdev);
return rc;
}
rc = scsi_add_host(ioa_cfg->host, &pdev->dev);
if (rc) {
__ipr_remove(pdev);
return rc;
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
rc = ipr_create_trace_file(&ioa_cfg->host->shost_dev.kobj,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
&ipr_trace_attr);
if (rc) {
scsi_remove_host(ioa_cfg->host);
__ipr_remove(pdev);
return rc;
}
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
rc = sysfs_create_bin_file(&ioa_cfg->host->shost_dev.kobj,
&ipr_ioa_async_err_log);
if (rc) {
ipr_remove_dump_file(&ioa_cfg->host->shost_dev.kobj,
&ipr_dump_attr);
ipr_remove_trace_file(&ioa_cfg->host->shost_dev.kobj,
&ipr_trace_attr);
scsi_remove_host(ioa_cfg->host);
__ipr_remove(pdev);
return rc;
}
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
rc = ipr_create_dump_file(&ioa_cfg->host->shost_dev.kobj,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
&ipr_dump_attr);
if (rc) {
scsi: ipr: Add asynchronous error notification This patch implements functions for pushing HCAM (host controlled asynchronous messages) error buffers to userspace through sysfs attributes. Reads to the "async_err_log" attribute will result in a single HCAM buffer being copied to userspace; one can process the next HCAM buffer by writing any string to the same attribute. A new list was added to the ioa_cfg structure to store the HCAM buffers for later reporting. We also send a KOBJ_CHANGE event whenever a new HCAM buffer is made available to userspace. Signed-off-by: Heitor Ricardo Alves de Siqueira <halves@linux.vnet.ibm.com> Signed-off-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-08-24 11:56:51 -06:00
sysfs_remove_bin_file(&ioa_cfg->host->shost_dev.kobj,
&ipr_ioa_async_err_log);
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-21 16:13:36 -07:00
ipr_remove_trace_file(&ioa_cfg->host->shost_dev.kobj,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
&ipr_trace_attr);
scsi_remove_host(ioa_cfg->host);
__ipr_remove(pdev);
return rc;
}
ipr: Fix sync scsi scan Commit b195d5e2bffd ("ipr: Wait to do async scan until scsi host is initialized") fixed async scan for ipr, but broke sync scan for ipr. This fixes sync scan back up. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reported-and-tested-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-08 16:53:12 -06:00
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
ioa_cfg->scan_enabled = 1;
schedule_work(&ioa_cfg->work_q);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
ioa_cfg->iopoll_weight = ioa_cfg->chip_cfg->iopoll_weight;
block: remove old blk_iopoll_enabled variable This was a debugging measure to toggle enabled/disabled when testing. But for real production setups, it's not safe to toggle this setting without either reloading drivers of quiescing IO first. Neither of which the toggle enforces. Additionally, it makes drivers deal with the conditional state. Remove it completely. It's up to the driver whether iopoll is enabled or not. Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-13 09:38:42 -06:00
if (ioa_cfg->iopoll_weight && ioa_cfg->sis64 && ioa_cfg->nvectors > 1) {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
for (i = 1; i < ioa_cfg->hrrq_num; i++) {
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
irq_poll_init(&ioa_cfg->hrrq[i].iopoll,
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
ioa_cfg->iopoll_weight, ipr_iopoll);
}
}
ipr: Fix sync scsi scan Commit b195d5e2bffd ("ipr: Wait to do async scan until scsi host is initialized") fixed async scan for ipr, but broke sync scan for ipr. This fixes sync scan back up. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reported-and-tested-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-08-08 16:53:12 -06:00
scsi_scan_host(ioa_cfg->host);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
return 0;
}
/**
* ipr_shutdown - Shutdown handler.
[PATCH] PCI: make drivers use the pci shutdown callback instead of the driver core callback. Now we can change the pci core to always set this pointer, as pci drivers should use it, not the driver core callback. Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-23 18:35:56 -06:00
* @pdev: pci device struct
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
*
* This function is invoked upon system shutdown/reboot. It will issue
* an adapter shutdown to the adapter to flush the write cache.
*
* Return value:
* none
**/
[PATCH] PCI: make drivers use the pci shutdown callback instead of the driver core callback. Now we can change the pci core to always set this pointer, as pci drivers should use it, not the driver core callback. Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-23 18:35:56 -06:00
static void ipr_shutdown(struct pci_dev *pdev)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{
[PATCH] PCI: make drivers use the pci shutdown callback instead of the driver core callback. Now we can change the pci core to always set this pointer, as pci drivers should use it, not the driver core callback. Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-23 18:35:56 -06:00
struct ipr_ioa_cfg *ioa_cfg = pci_get_drvdata(pdev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
unsigned long lock_flags = 0;
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
enum ipr_shutdown_type shutdown_type = IPR_SHUTDOWN_NORMAL;
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
int i;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
block: remove old blk_iopoll_enabled variable This was a debugging measure to toggle enabled/disabled when testing. But for real production setups, it's not safe to toggle this setting without either reloading drivers of quiescing IO first. Neither of which the toggle enforces. Additionally, it makes drivers deal with the conditional state. Remove it completely. It's up to the driver whether iopoll is enabled or not. Signed-off-by: Jens Axboe <axboe@fb.com>
2014-03-13 09:38:42 -06:00
if (ioa_cfg->iopoll_weight && ioa_cfg->sis64 && ioa_cfg->nvectors > 1) {
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
ioa_cfg->iopoll_weight = 0;
for (i = 1; i < ioa_cfg->hrrq_num; i++)
irq_poll: make blk-iopoll available outside the block layer The new name is irq_poll as iopoll is already taken. Better suggestions welcome. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Bart Van Assche <bart.vanassche@sandisk.com>
2015-11-10 06:56:14 -07:00
irq_poll_disable(&ioa_cfg->hrrq[i].iopoll);
[SCSI] ipr: Implement block iopoll This patch implements blk iopoll in ipr driver for performance improvement. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:52 -07:00
}
ipr: fix small coding style issues Fixing small trivial coding style issues. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-08-22 15:14:14 -06:00
while (ioa_cfg->in_reset_reload) {
[SCSI] ipr: Prevent overlapped adapter resets This patch fixes some scenarios where an ipr adapter could get reset overlapped, which could cause very long timeouts to occur, or PCI bus errors. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:06 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
}
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
if (ipr_fast_reboot && system_state == SYSTEM_RESTART && ioa_cfg->sis64)
shutdown_type = IPR_SHUTDOWN_QUIESCE;
ipr_initiate_ioa_bringdown(ioa_cfg, shutdown_type);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
if (ipr_fast_reboot && system_state == SYSTEM_RESTART && ioa_cfg->sis64) {
ipr: Reset in task context The pci_set_pcie_reset_state has changed semantics to not be callable from interrupt context, so change ipr's usage of the API to comply with this change by ensuring this occurs from a workqueue. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:52 -06:00
ipr_free_irqs(ioa_cfg);
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
pci_disable_device(ioa_cfg->pdev);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
static struct pci_device_id ipr_pci_table[] = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_MYLEX, PCI_DEVICE_ID_IBM_GEMSTONE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_5702, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_MYLEX, PCI_DEVICE_ID_IBM_GEMSTONE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_5703, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_MYLEX, PCI_DEVICE_ID_IBM_GEMSTONE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_573D, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_MYLEX, PCI_DEVICE_ID_IBM_GEMSTONE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_573E, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CITRINE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_571B, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CITRINE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_572E, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CITRINE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_571A, 0, 0, 0 },
[SCSI] ipr: New PCI Ids Adds support for some new ipr adapters Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:42 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CITRINE,
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_575B, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT },
[SCSI] ipr: New PCI Ids Adds support for some new ipr adapters Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:42 -07:00
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_OBSIDIAN,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_572A, 0, 0, 0 },
[SCSI] ipr: New PCI Ids Adds support for some new ipr adapters Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:42 -07:00
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_OBSIDIAN,
[SCSI] ipr: Increase adapter operational timeout Increases the adapter operational timeout for some adapters that support dual controller configurations, since they may take longer to come ready. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:13 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_572B, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT },
[SCSI] ipr: PCI IDs for new SAS adapters Adds PCI IDs for some new ipr SAS adapters. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:10 -07:00
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_OBSIDIAN,
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_575C, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT },
[SCSI] ipr: New PCI Ids Adds support for some new ipr adapters Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:42 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_572A, 0, 0, 0 },
[SCSI] ipr: New PCI Ids Adds support for some new ipr adapters Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:42 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN,
[SCSI] ipr: Increase adapter operational timeout Increases the adapter operational timeout for some adapters that support dual controller configurations, since they may take longer to come ready. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:13 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_572B, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT},
[SCSI] ipr: PCI IDs for new SAS adapters Adds PCI IDs for some new ipr SAS adapters. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:10 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN,
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_575C, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT },
[SCSI] ipr: Add new PCI-E IDs to device table Adds support for some new PCI-E ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:53 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN_E,
[SCSI] ipr: Increase adapter operational timeout Increases the adapter operational timeout for some adapters that support dual controller configurations, since they may take longer to come ready. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-04-26 15:00:13 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_574E, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT },
[SCSI] ipr: Add new PCI-E IDs to device table Adds support for some new PCI-E ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:53 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN_E,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B3, 0, 0, 0 },
[SCSI] ipr: add support for new Obsidian-E embedded adapter This patch allows the driver to recognize a new Obsidian-E based adapter that uses a new subsystem ID. This patch also fixes a few tab/space problems. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-06-24 14:34:14 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN_E,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57CC, 0, 0, 0 },
[SCSI] ipr: PCI IDs for new SAS adapters Adds PCI IDs for some new ipr SAS adapters. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:10 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN_E,
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B7, 0, 0,
[SCSI] ipr: Use PCI-E reset API for new ipr adapter Use a newly added PCI API to issue a PCI Fundamental reset (warm reset) to a new ipr PCI-E adapter. Typically, the ipr adapter uses the start BIST bit in config space to reset an adapter. Issuing start BIST on this particular adapter results in the PCI-E logic on the card losing sync, which causes PCI-E errors, making the card unusable. The only reset mechanism that exists on this hardware that does not have this problem is PCI Fundamental reset (warm reset). Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-07 16:09:05 -06:00
IPR_USE_LONG_TRANSOP_TIMEOUT | IPR_USE_PCI_WARM_RESET },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_SNIPE,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_2780, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_SCAMP,
[SCSI] ipr: Remove usage of pci driver data Since ipr handles dynamic ids, it must handle driver_data not being set, so remove the current usage of driver_data so it can be used for other things in future patches. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-23 10:25:23 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_571E, 0, 0, 0 },
[SCSI] ipr: New PCI Ids Adds support for some new ipr adapters Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 16:02:42 -07:00
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_SCAMP,
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_571F, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT },
[SCSI] ipr: PCI IDs for new SAS adapters Adds PCI IDs for some new ipr SAS adapters. Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-11-21 09:28:10 -07:00
{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_SCAMP,
[SCSI] ipr: Make adapter operational timeout be per adapter type Some ipr adapters may take longer than others to come operational. This patch makes this timeout different for different adapters, while still preserving the module parameter which can be used to globally override the default. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-03-29 11:42:40 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_572F, 0, 0,
IPR_USE_LONG_TRANSOP_TIMEOUT },
[SCSI] ipr: adds PCI ID definitions for new adapters This patch adds the PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. New entries have been added to the ipr_pci_table[] array for the adapters and to the ipr_chip[] array for the new versions of the chip. Older entries have been removed for cards that did not ship. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:38 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B5, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_574D, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B2, 0, 0, 0 },
[SCSI] ipr: Add sereral new CCIN definitions for new adapters support Add the appropriate definitions and table entries for new adapter support. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:47 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57C0, 0, 0, 0 },
[SCSI] ipr: add definitions for additional adapter Add the appropriate definition and table entry for an additional adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2011-10-27 12:58:21 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57C3, 0, 0, 0 },
[SCSI] ipr: add definitions for a new adapter There was an addition to the hardware roadmap that includes a new adapter. This patch adds the new definitions for the adapter. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-10-18 21:24:34 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROC_FPGA_E2,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57C4, 0, 0, 0 },
[SCSI] ipr: update PCI ID definitions for new adapters This patch updates some PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-02-23 12:54:55 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
[SCSI] ipr: adds PCI ID definitions for new adapters This patch adds the PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. New entries have been added to the ipr_pci_table[] array for the adapters and to the ipr_chip[] array for the new versions of the chip. Older entries have been removed for cards that did not ship. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:38 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B4, 0, 0, 0 },
[SCSI] ipr: update PCI ID definitions for new adapters This patch updates some PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-02-23 12:54:55 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
[SCSI] ipr: adds PCI ID definitions for new adapters This patch adds the PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. New entries have been added to the ipr_pci_table[] array for the adapters and to the ipr_chip[] array for the new versions of the chip. Older entries have been removed for cards that did not ship. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:38 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57B1, 0, 0, 0 },
[SCSI] ipr: update PCI ID definitions for new adapters This patch updates some PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-02-23 12:54:55 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
[SCSI] ipr: adds PCI ID definitions for new adapters This patch adds the PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. New entries have been added to the ipr_pci_table[] array for the adapters and to the ipr_chip[] array for the new versions of the chip. Older entries have been removed for cards that did not ship. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:38 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57C6, 0, 0, 0 },
[SCSI] ipr: update PCI ID definitions for new adapters This patch updates some PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-02-23 12:54:55 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57C8, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
[SCSI] ipr: adds PCI ID definitions for new adapters This patch adds the PCI ID definitions for new adapters based on the next generation 64 bit IOA PCI interface chip. New entries have been added to the ipr_pci_table[] array for the adapters and to the ipr_chip[] array for the new versions of the chip. Older entries have been removed for cards that did not ship. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:38 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57CE, 0, 0, 0 },
[SCSI] ipr: Add sereral new CCIN definitions for new adapters support Add the appropriate definitions and table entries for new adapter support. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:47 -07:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57D5, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57D6, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57D7, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57D8, 0, 0, 0 },
[SCSI] ipr: Add sereral new CCIN definitions for new adapters support Add the appropriate definitions and table entries for new adapter support. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-07-10 09:46:27 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57D9, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
[SCSI] ipr: Add new CCIN definition for new hardware support Add the appropriate definition and table entry for new hardware support. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:40 -07:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57DA, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
[SCSI] ipr: Add sereral new CCIN definitions for new adapters support Add the appropriate definitions and table entries for new adapter support. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-07-10 09:46:27 -06:00
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57EB, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57EC, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57ED, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57EE, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57EF, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_57F0, 0, 0, 0 },
[SCSI] ipr: Add new CCIN definition for Grand Canyon support Add the appropriate definition and table entry for new hardware support. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-03-12 15:08:52 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_2CCA, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_2CD2, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CROCODILE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_2CCD, 0, 0, 0 },
ipr: Add new CCIN for new adapters support Add the appropriate definitions and table entries for new adapters support. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Gabriel Krisman Bertazi <krisman@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2016-07-12 15:02:07 -06:00
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_RATTLESNAKE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_580A, 0, 0, 0 },
{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_RATTLESNAKE,
PCI_VENDOR_ID_IBM, IPR_SUBS_DEV_ID_580B, 0, 0, 0 },
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
{ }
};
MODULE_DEVICE_TABLE(pci, ipr_pci_table);
scsi: make pci error handlers const Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2012-09-07 10:33:16 -06:00
static const struct pci_error_handlers ipr_err_handler = {
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
.error_detected = ipr_pci_error_detected,
[SCSI] ipr: Handle early EEH If, when the ipr driver loads, the adapter is in an EEH error state, it will currently oops and not be able to recover, as it attempts to access memory that has not yet been allocated. We've seen this occur in some kexec scenarios. The following patch fixes the oops and also allows the driver to recover from these probe time EEH errors. [jejb: checkpatch fix] Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2014-01-21 11:16:41 -07:00
.mmio_enabled = ipr_pci_mmio_enabled,
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
.slot_reset = ipr_pci_slot_reset,
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
static struct pci_driver ipr_driver = {
.name = IPR_NAME,
.id_table = ipr_pci_table,
.probe = ipr_probe,
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-21 14:08:55 -07:00
.remove = ipr_remove,
[PATCH] PCI: make drivers use the pci shutdown callback instead of the driver core callback. Now we can change the pci core to always set this pointer, as pci drivers should use it, not the driver core callback. Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-23 18:35:56 -06:00
.shutdown = ipr_shutdown,
[SCSI] PCI Error Recovery: IPR SCSI device driver Various PCI bus errors can be signaled by newer PCI controllers. This patch adds the PCI error recovery callbacks to the IPR SCSI device driver. The patch has been tested, and appears to work well. Signed-off-by: Linas Vepstas <linas@linas.org> Signed-off-by: Brian King <brking@us.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-02-03 15:52:42 -07:00
.err_handler = &ipr_err_handler,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
};
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
/**
* ipr_halt_done - Shutdown prepare completion
*
* Return value:
* none
**/
static void ipr_halt_done(struct ipr_cmnd *ipr_cmd)
{
[SCSI] ipr: Add support for MSI-X and distributed completion The new generation IBM SAS Controllers will support MSI-X interrupts and Distributed Completion Processing features. This patch add these support in ipr device driver. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-01-11 16:43:50 -07:00
list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
}
/**
* ipr_halt - Issue shutdown prepare to all adapters
*
* Return value:
* NOTIFY_OK on success / NOTIFY_DONE on failure
**/
static int ipr_halt(struct notifier_block *nb, ulong event, void *buf)
{
struct ipr_cmnd *ipr_cmd;
struct ipr_ioa_cfg *ioa_cfg;
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
unsigned long flags = 0, driver_lock_flags;
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
if (event != SYS_RESTART && event != SYS_HALT && event != SYS_POWER_OFF)
return NOTIFY_DONE;
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
spin_lock_irqsave(&ipr_driver_lock, driver_lock_flags);
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
list_for_each_entry(ioa_cfg, &ipr_ioa_head, queue) {
spin_lock_irqsave(ioa_cfg->host->host_lock, flags);
ipr: Reboot speed improvements Currently when performing a reboot with an ipr adapter, the adapter gets shutdown completely, flushing all write cache, as well as performing a full hardware reset of the card during the shutdown phase of the old kernel. This ensures the adapter is in a fully quiesced state across the reboot. There are scenarios, however, such as when performing kexec, where this full adapter shutdown is not required and not desired, since it can make the reboot process take noticeably longer. This patch adds a module parameter to allow for skipping the full shutdown during reboot. Rather than performing a full adapter shutdown and reset, we simply cancel any outstanding error buffers, place the adapter into a state where it has no memory of any DMA addresses from the old kernel, then disable the device. This significantly speeds up kexec boot, particularly in configurations with multiple ipr adapters. Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Reviewed-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Daniel Kreling <kreling@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
2015-03-26 10:23:50 -06:00
if (!ioa_cfg->hrrq[IPR_INIT_HRRQ].allow_cmds ||
(ipr_fast_reboot && event == SYS_RESTART && ioa_cfg->sis64)) {
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
continue;
}
ipr_cmd = ipr_get_free_ipr_cmnd(ioa_cfg);
ipr_cmd->ioarcb.res_handle = cpu_to_be32(IPR_IOA_RES_HANDLE);
ipr_cmd->ioarcb.cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
ipr_cmd->ioarcb.cmd_pkt.cdb[0] = IPR_IOA_SHUTDOWN;
ipr_cmd->ioarcb.cmd_pkt.cdb[1] = IPR_SHUTDOWN_PREPARE_FOR_NORMAL;
ipr_do_req(ipr_cmd, ipr_halt_done, ipr_timeout, IPR_DEVICE_RESET_TIMEOUT);
spin_unlock_irqrestore(ioa_cfg->host->host_lock, flags);
}
[SCSI] ipr: possible irq lock inversion dependency detected When enable lockdep, seeing "possible irq lock inversion dependency detected" error. This patch fixes the issue. Signed-off-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Acked-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-05-24 08:59:13 -06:00
spin_unlock_irqrestore(&ipr_driver_lock, driver_lock_flags);
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
return NOTIFY_OK;
}
static struct notifier_block ipr_notifier = {
ipr_halt, NULL, 0
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
/**
* ipr_init - Module entry point
*
* Return value:
* 0 on success / negative value on failure
**/
static int __init ipr_init(void)
{
ipr_info("IBM Power RAID SCSI Device Driver version: %s %s\n",
IPR_DRIVER_VERSION, IPR_DRIVER_DATE);
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
register_reboot_notifier(&ipr_notifier);
[SCSI] pci_module_init conversion in scsi subsystem Converts pci_module_init() to pci_register_driver() in the scsi subsys on 23 drivers which only return the value of pci_module_init(). Signed-off-by: Henrik Kretzschmar <henne@nachtwindheim.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-09-25 17:58:58 -06:00
return pci_register_driver(&ipr_driver);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
}
/**
* ipr_exit - Module unload
*
* Module unload entry point.
*
* Return value:
* none
**/
static void __exit ipr_exit(void)
{
[SCSI] ipr: implement shutdown changes and remove obsolete write cache parameter This patch adds a reboot notifier that will issue a shutdown prepare command to all adapters. This helps to prevent a problem where the primary adapter can get shut down before the secondary adapter and cause the secondary adapter to fail over and log and error. This patch also removes the "enable_cache" paramater as it is obsolete. Write cache for an adapter is now controlled from the iprconfig utility. Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2010-02-19 14:24:21 -07:00
unregister_reboot_notifier(&ipr_notifier);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 16:20:36 -06:00
pci_unregister_driver(&ipr_driver);
}
module_init(ipr_init);
module_exit(ipr_exit);