2019-05-29 17:57:49 -06:00
|
|
|
/* SPDX-License-Identifier: GPL-2.0-only */
|
dmaengine: driver for the iop32x, iop33x, and iop13xx raid engines
The Intel(R) IOP series of i/o processors integrate an Xscale core with
raid acceleration engines. The capabilities per platform are:
iop219:
(2) copy engines
iop321:
(2) copy engines
(1) xor and block fill engine
iop33x:
(2) copy and crc32c engines
(1) xor, xor zero sum, pq, pq zero sum, and block fill engine
iop34x (iop13xx):
(2) copy, crc32c, xor, xor zero sum, and block fill engines
(1) copy, crc32c, xor, xor zero sum, pq, pq zero sum, and block fill engine
The driver supports the features of the async_tx api:
* asynchronous notification of operation completion
* implicit (interupt triggered) handling of inter-channel transaction
dependencies
The driver adapts to the platform it is running by two methods.
1/ #include <asm/arch/adma.h> which defines the hardware specific
iop_chan_* and iop_desc_* routines as a series of static inline
functions
2/ The private platform data attached to the platform_device defines the
capabilities of the channels
20070626: Callbacks are run in a tasklet. Given the recent discussion on
LKML about killing tasklets in favor of workqueues I did a quick conversion
of the driver. Raid5 resync performance dropped from 50MB/s to 30MB/s, so
the tasklet implementation remains until a generic softirq interface is
available.
Changelog:
* fixed a slot allocation bug in do_iop13xx_adma_xor that caused too few
slots to be requested eventually leading to data corruption
* enabled the slot allocation routine to attempt to free slots before
returning -ENOMEM
* switched the cleanup routine to solely use the software chain and the
status register to determine if a descriptor is complete. This is
necessary to support other IOP engines that do not have status writeback
capability
* make the driver iop generic
* modified the allocation routines to understand allocating a group of
slots for a single operation
* added a null xor initialization operation for the xor only channel on
iop3xx
* support xor operations on buffers larger than the hardware maximum
* split the do_* routines into separate prep, src/dest set, submit stages
* added async_tx support (dependent operations initiation at cleanup time)
* simplified group handling
* added interrupt support (callbacks via tasklets)
* brought the pending depth inline with ioat (i.e. 4 descriptors)
* drop dma mapping methods, suggested by Chris Leech
* don't use inline in C files, Adrian Bunk
* remove static tasklet declarations
* make iop_adma_alloc_slots easier to read and remove chances for a
corrupted descriptor chain
* fix locking bug in iop_adma_alloc_chan_resources, Benjamin Herrenschmidt
* convert capabilities over to dma_cap_mask_t
* fixup sparse warnings
* add descriptor flush before iop_chan_enable
* checkpatch.pl fixes
* gpl v2 only correction
* move set_src, set_dest, submit to async_tx methods
* move group_list and phys to async_tx
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2007-01-02 13:52:26 -07:00
|
|
|
/*
|
|
|
|
|
* Copyright © 2006, Intel Corporation.
|
|
|
|
|
*/
|
|
|
|
|
#ifndef IOP_ADMA_H
|
|
|
|
|
#define IOP_ADMA_H
|
|
|
|
|
#include <linux/types.h>
|
|
|
|
|
#include <linux/dmaengine.h>
|
|
|
|
|
#include <linux/interrupt.h>
|
|
|
|
|
|
|
|
|
|
#define IOP_ADMA_SLOT_SIZE 32
|
|
|
|
|
#define IOP_ADMA_THRESHOLD 4
|
2008-11-11 13:12:33 -07:00
|
|
|
#ifdef DEBUG
|
|
|
|
|
#define IOP_PARANOIA 1
|
|
|
|
|
#else
|
|
|
|
|
#define IOP_PARANOIA 0
|
|
|
|
|
#endif
|
|
|
|
|
#define iop_paranoia(x) BUG_ON(IOP_PARANOIA && (x))
|
dmaengine: driver for the iop32x, iop33x, and iop13xx raid engines
The Intel(R) IOP series of i/o processors integrate an Xscale core with
raid acceleration engines. The capabilities per platform are:
iop219:
(2) copy engines
iop321:
(2) copy engines
(1) xor and block fill engine
iop33x:
(2) copy and crc32c engines
(1) xor, xor zero sum, pq, pq zero sum, and block fill engine
iop34x (iop13xx):
(2) copy, crc32c, xor, xor zero sum, and block fill engines
(1) copy, crc32c, xor, xor zero sum, pq, pq zero sum, and block fill engine
The driver supports the features of the async_tx api:
* asynchronous notification of operation completion
* implicit (interupt triggered) handling of inter-channel transaction
dependencies
The driver adapts to the platform it is running by two methods.
1/ #include <asm/arch/adma.h> which defines the hardware specific
iop_chan_* and iop_desc_* routines as a series of static inline
functions
2/ The private platform data attached to the platform_device defines the
capabilities of the channels
20070626: Callbacks are run in a tasklet. Given the recent discussion on
LKML about killing tasklets in favor of workqueues I did a quick conversion
of the driver. Raid5 resync performance dropped from 50MB/s to 30MB/s, so
the tasklet implementation remains until a generic softirq interface is
available.
Changelog:
* fixed a slot allocation bug in do_iop13xx_adma_xor that caused too few
slots to be requested eventually leading to data corruption
* enabled the slot allocation routine to attempt to free slots before
returning -ENOMEM
* switched the cleanup routine to solely use the software chain and the
status register to determine if a descriptor is complete. This is
necessary to support other IOP engines that do not have status writeback
capability
* make the driver iop generic
* modified the allocation routines to understand allocating a group of
slots for a single operation
* added a null xor initialization operation for the xor only channel on
iop3xx
* support xor operations on buffers larger than the hardware maximum
* split the do_* routines into separate prep, src/dest set, submit stages
* added async_tx support (dependent operations initiation at cleanup time)
* simplified group handling
* added interrupt support (callbacks via tasklets)
* brought the pending depth inline with ioat (i.e. 4 descriptors)
* drop dma mapping methods, suggested by Chris Leech
* don't use inline in C files, Adrian Bunk
* remove static tasklet declarations
* make iop_adma_alloc_slots easier to read and remove chances for a
corrupted descriptor chain
* fix locking bug in iop_adma_alloc_chan_resources, Benjamin Herrenschmidt
* convert capabilities over to dma_cap_mask_t
* fixup sparse warnings
* add descriptor flush before iop_chan_enable
* checkpatch.pl fixes
* gpl v2 only correction
* move set_src, set_dest, submit to async_tx methods
* move group_list and phys to async_tx
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2007-01-02 13:52:26 -07:00
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* struct iop_adma_device - internal representation of an ADMA device
|
|
|
|
|
* @pdev: Platform device
|
|
|
|
|
* @id: HW ADMA Device selector
|
|
|
|
|
* @dma_desc_pool: base of DMA descriptor region (DMA address)
|
|
|
|
|
* @dma_desc_pool_virt: base of DMA descriptor region (CPU address)
|
|
|
|
|
* @common: embedded struct dma_device
|
|
|
|
|
*/
|
|
|
|
|
struct iop_adma_device {
|
|
|
|
|
struct platform_device *pdev;
|
|
|
|
|
int id;
|
|
|
|
|
dma_addr_t dma_desc_pool;
|
|
|
|
|
void *dma_desc_pool_virt;
|
|
|
|
|
struct dma_device common;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* struct iop_adma_chan - internal representation of an ADMA device
|
|
|
|
|
* @pending: allows batching of hardware operations
|
|
|
|
|
* @lock: serializes enqueue/dequeue operations to the slot pool
|
|
|
|
|
* @mmr_base: memory mapped register base
|
|
|
|
|
* @chain: device chain view of the descriptors
|
|
|
|
|
* @device: parent device
|
|
|
|
|
* @common: common dmaengine channel object members
|
|
|
|
|
* @last_used: place holder for allocation to continue from where it left off
|
|
|
|
|
* @all_slots: complete domain of slots usable by the channel
|
|
|
|
|
* @slots_allocated: records the actual size of the descriptor slot pool
|
|
|
|
|
* @irq_tasklet: bottom half where iop_adma_slot_cleanup runs
|
|
|
|
|
*/
|
|
|
|
|
struct iop_adma_chan {
|
|
|
|
|
int pending;
|
|
|
|
|
spinlock_t lock; /* protects the descriptor slot pool */
|
|
|
|
|
void __iomem *mmr_base;
|
|
|
|
|
struct list_head chain;
|
|
|
|
|
struct iop_adma_device *device;
|
|
|
|
|
struct dma_chan common;
|
|
|
|
|
struct iop_adma_desc_slot *last_used;
|
|
|
|
|
struct list_head all_slots;
|
|
|
|
|
int slots_allocated;
|
|
|
|
|
struct tasklet_struct irq_tasklet;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
* struct iop_adma_desc_slot - IOP-ADMA software descriptor
|
|
|
|
|
* @slot_node: node on the iop_adma_chan.all_slots list
|
|
|
|
|
* @chain_node: node on the op_adma_chan.chain list
|
|
|
|
|
* @hw_desc: virtual address of the hardware descriptor chain
|
|
|
|
|
* @phys: hardware address of the hardware descriptor chain
|
|
|
|
|
* @group_head: first operation in a transaction
|
|
|
|
|
* @slot_cnt: total slots used in an transaction (group of operations)
|
|
|
|
|
* @slots_per_op: number of slots per operation
|
|
|
|
|
* @idx: pool index
|
2009-09-08 18:53:02 -06:00
|
|
|
* @tx_list: list of descriptors that are associated with one operation
|
dmaengine: driver for the iop32x, iop33x, and iop13xx raid engines
The Intel(R) IOP series of i/o processors integrate an Xscale core with
raid acceleration engines. The capabilities per platform are:
iop219:
(2) copy engines
iop321:
(2) copy engines
(1) xor and block fill engine
iop33x:
(2) copy and crc32c engines
(1) xor, xor zero sum, pq, pq zero sum, and block fill engine
iop34x (iop13xx):
(2) copy, crc32c, xor, xor zero sum, and block fill engines
(1) copy, crc32c, xor, xor zero sum, pq, pq zero sum, and block fill engine
The driver supports the features of the async_tx api:
* asynchronous notification of operation completion
* implicit (interupt triggered) handling of inter-channel transaction
dependencies
The driver adapts to the platform it is running by two methods.
1/ #include <asm/arch/adma.h> which defines the hardware specific
iop_chan_* and iop_desc_* routines as a series of static inline
functions
2/ The private platform data attached to the platform_device defines the
capabilities of the channels
20070626: Callbacks are run in a tasklet. Given the recent discussion on
LKML about killing tasklets in favor of workqueues I did a quick conversion
of the driver. Raid5 resync performance dropped from 50MB/s to 30MB/s, so
the tasklet implementation remains until a generic softirq interface is
available.
Changelog:
* fixed a slot allocation bug in do_iop13xx_adma_xor that caused too few
slots to be requested eventually leading to data corruption
* enabled the slot allocation routine to attempt to free slots before
returning -ENOMEM
* switched the cleanup routine to solely use the software chain and the
status register to determine if a descriptor is complete. This is
necessary to support other IOP engines that do not have status writeback
capability
* make the driver iop generic
* modified the allocation routines to understand allocating a group of
slots for a single operation
* added a null xor initialization operation for the xor only channel on
iop3xx
* support xor operations on buffers larger than the hardware maximum
* split the do_* routines into separate prep, src/dest set, submit stages
* added async_tx support (dependent operations initiation at cleanup time)
* simplified group handling
* added interrupt support (callbacks via tasklets)
* brought the pending depth inline with ioat (i.e. 4 descriptors)
* drop dma mapping methods, suggested by Chris Leech
* don't use inline in C files, Adrian Bunk
* remove static tasklet declarations
* make iop_adma_alloc_slots easier to read and remove chances for a
corrupted descriptor chain
* fix locking bug in iop_adma_alloc_chan_resources, Benjamin Herrenschmidt
* convert capabilities over to dma_cap_mask_t
* fixup sparse warnings
* add descriptor flush before iop_chan_enable
* checkpatch.pl fixes
* gpl v2 only correction
* move set_src, set_dest, submit to async_tx methods
* move group_list and phys to async_tx
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2007-01-02 13:52:26 -07:00
|
|
|
* @async_tx: support for the async_tx api
|
|
|
|
|
* @group_list: list of slots that make up a multi-descriptor transaction
|
|
|
|
|
* for example transfer lengths larger than the supported hw max
|
|
|
|
|
* @xor_check_result: result of zero sum
|
|
|
|
|
* @crc32_result: result crc calculation
|
|
|
|
|
*/
|
|
|
|
|
struct iop_adma_desc_slot {
|
|
|
|
|
struct list_head slot_node;
|
|
|
|
|
struct list_head chain_node;
|
|
|
|
|
void *hw_desc;
|
|
|
|
|
struct iop_adma_desc_slot *group_head;
|
|
|
|
|
u16 slot_cnt;
|
|
|
|
|
u16 slots_per_op;
|
|
|
|
|
u16 idx;
|
2009-09-08 18:53:02 -06:00
|
|
|
struct list_head tx_list;
|
dmaengine: driver for the iop32x, iop33x, and iop13xx raid engines
The Intel(R) IOP series of i/o processors integrate an Xscale core with
raid acceleration engines. The capabilities per platform are:
iop219:
(2) copy engines
iop321:
(2) copy engines
(1) xor and block fill engine
iop33x:
(2) copy and crc32c engines
(1) xor, xor zero sum, pq, pq zero sum, and block fill engine
iop34x (iop13xx):
(2) copy, crc32c, xor, xor zero sum, and block fill engines
(1) copy, crc32c, xor, xor zero sum, pq, pq zero sum, and block fill engine
The driver supports the features of the async_tx api:
* asynchronous notification of operation completion
* implicit (interupt triggered) handling of inter-channel transaction
dependencies
The driver adapts to the platform it is running by two methods.
1/ #include <asm/arch/adma.h> which defines the hardware specific
iop_chan_* and iop_desc_* routines as a series of static inline
functions
2/ The private platform data attached to the platform_device defines the
capabilities of the channels
20070626: Callbacks are run in a tasklet. Given the recent discussion on
LKML about killing tasklets in favor of workqueues I did a quick conversion
of the driver. Raid5 resync performance dropped from 50MB/s to 30MB/s, so
the tasklet implementation remains until a generic softirq interface is
available.
Changelog:
* fixed a slot allocation bug in do_iop13xx_adma_xor that caused too few
slots to be requested eventually leading to data corruption
* enabled the slot allocation routine to attempt to free slots before
returning -ENOMEM
* switched the cleanup routine to solely use the software chain and the
status register to determine if a descriptor is complete. This is
necessary to support other IOP engines that do not have status writeback
capability
* make the driver iop generic
* modified the allocation routines to understand allocating a group of
slots for a single operation
* added a null xor initialization operation for the xor only channel on
iop3xx
* support xor operations on buffers larger than the hardware maximum
* split the do_* routines into separate prep, src/dest set, submit stages
* added async_tx support (dependent operations initiation at cleanup time)
* simplified group handling
* added interrupt support (callbacks via tasklets)
* brought the pending depth inline with ioat (i.e. 4 descriptors)
* drop dma mapping methods, suggested by Chris Leech
* don't use inline in C files, Adrian Bunk
* remove static tasklet declarations
* make iop_adma_alloc_slots easier to read and remove chances for a
corrupted descriptor chain
* fix locking bug in iop_adma_alloc_chan_resources, Benjamin Herrenschmidt
* convert capabilities over to dma_cap_mask_t
* fixup sparse warnings
* add descriptor flush before iop_chan_enable
* checkpatch.pl fixes
* gpl v2 only correction
* move set_src, set_dest, submit to async_tx methods
* move group_list and phys to async_tx
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2007-01-02 13:52:26 -07:00
|
|
|
struct dma_async_tx_descriptor async_tx;
|
|
|
|
|
union {
|
|
|
|
|
u32 *xor_check_result;
|
|
|
|
|
u32 *crc32_result;
|
2009-08-28 15:32:04 -06:00
|
|
|
u32 *pq_check_result;
|
dmaengine: driver for the iop32x, iop33x, and iop13xx raid engines
The Intel(R) IOP series of i/o processors integrate an Xscale core with
raid acceleration engines. The capabilities per platform are:
iop219:
(2) copy engines
iop321:
(2) copy engines
(1) xor and block fill engine
iop33x:
(2) copy and crc32c engines
(1) xor, xor zero sum, pq, pq zero sum, and block fill engine
iop34x (iop13xx):
(2) copy, crc32c, xor, xor zero sum, and block fill engines
(1) copy, crc32c, xor, xor zero sum, pq, pq zero sum, and block fill engine
The driver supports the features of the async_tx api:
* asynchronous notification of operation completion
* implicit (interupt triggered) handling of inter-channel transaction
dependencies
The driver adapts to the platform it is running by two methods.
1/ #include <asm/arch/adma.h> which defines the hardware specific
iop_chan_* and iop_desc_* routines as a series of static inline
functions
2/ The private platform data attached to the platform_device defines the
capabilities of the channels
20070626: Callbacks are run in a tasklet. Given the recent discussion on
LKML about killing tasklets in favor of workqueues I did a quick conversion
of the driver. Raid5 resync performance dropped from 50MB/s to 30MB/s, so
the tasklet implementation remains until a generic softirq interface is
available.
Changelog:
* fixed a slot allocation bug in do_iop13xx_adma_xor that caused too few
slots to be requested eventually leading to data corruption
* enabled the slot allocation routine to attempt to free slots before
returning -ENOMEM
* switched the cleanup routine to solely use the software chain and the
status register to determine if a descriptor is complete. This is
necessary to support other IOP engines that do not have status writeback
capability
* make the driver iop generic
* modified the allocation routines to understand allocating a group of
slots for a single operation
* added a null xor initialization operation for the xor only channel on
iop3xx
* support xor operations on buffers larger than the hardware maximum
* split the do_* routines into separate prep, src/dest set, submit stages
* added async_tx support (dependent operations initiation at cleanup time)
* simplified group handling
* added interrupt support (callbacks via tasklets)
* brought the pending depth inline with ioat (i.e. 4 descriptors)
* drop dma mapping methods, suggested by Chris Leech
* don't use inline in C files, Adrian Bunk
* remove static tasklet declarations
* make iop_adma_alloc_slots easier to read and remove chances for a
corrupted descriptor chain
* fix locking bug in iop_adma_alloc_chan_resources, Benjamin Herrenschmidt
* convert capabilities over to dma_cap_mask_t
* fixup sparse warnings
* add descriptor flush before iop_chan_enable
* checkpatch.pl fixes
* gpl v2 only correction
* move set_src, set_dest, submit to async_tx methods
* move group_list and phys to async_tx
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2007-01-02 13:52:26 -07:00
|
|
|
};
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
struct iop_adma_platform_data {
|
|
|
|
|
int hw_id;
|
|
|
|
|
dma_cap_mask_t cap_mask;
|
|
|
|
|
size_t pool_size;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
#define to_iop_sw_desc(addr_hw_desc) \
|
|
|
|
|
container_of(addr_hw_desc, struct iop_adma_desc_slot, hw_desc)
|
|
|
|
|
#define iop_hw_desc_slot_idx(hw_desc, idx) \
|
|
|
|
|
( (void *) (((unsigned long) hw_desc) + ((idx) << 5)) )
|
|
|
|
|
#endif
|