2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* fs/fs-writeback.c
|
|
|
|
*
|
|
|
|
* Copyright (C) 2002, Linus Torvalds.
|
|
|
|
*
|
|
|
|
* Contains all the functions related to writing back and waiting
|
|
|
|
* upon dirty inodes against superblocks, and writing back dirty
|
|
|
|
* pages against inodes. ie: data writeback. Writeout of the
|
|
|
|
* inode itself is not handled here.
|
|
|
|
*
|
2008-10-15 23:01:59 -06:00
|
|
|
* 10Apr2002 Andrew Morton
|
2005-04-16 16:20:36 -06:00
|
|
|
* Split out of fs/inode.c
|
|
|
|
* Additions for address_space-based writeback
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/kernel.h>
|
2011-11-16 21:57:37 -07:00
|
|
|
#include <linux/export.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/spinlock.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>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/fs.h>
|
|
|
|
#include <linux/mm.h>
|
2012-01-07 19:41:55 -07:00
|
|
|
#include <linux/pagemap.h>
|
2009-09-09 01:08:54 -06:00
|
|
|
#include <linux/kthread.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/writeback.h>
|
|
|
|
#include <linux/blkdev.h>
|
|
|
|
#include <linux/backing-dev.h>
|
2010-07-06 21:24:06 -06:00
|
|
|
#include <linux/tracepoint.h>
|
2006-09-30 12:52:18 -06:00
|
|
|
#include "internal.h"
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2012-01-07 19:41:55 -07:00
|
|
|
/*
|
|
|
|
* 4MB minimal write chunk size
|
|
|
|
*/
|
|
|
|
#define MIN_WRITEBACK_PAGES (4096UL >> (PAGE_CACHE_SHIFT - 10))
|
|
|
|
|
2009-09-16 07:18:25 -06:00
|
|
|
/*
|
|
|
|
* Passed into wb_writeback(), essentially a subset of writeback_control
|
|
|
|
*/
|
2010-07-06 00:59:53 -06:00
|
|
|
struct wb_writeback_work {
|
2009-09-16 07:18:25 -06:00
|
|
|
long nr_pages;
|
|
|
|
struct super_block *sb;
|
2011-05-04 19:54:37 -06:00
|
|
|
unsigned long *older_than_this;
|
2009-09-16 07:18:25 -06:00
|
|
|
enum writeback_sync_modes sync_mode;
|
2010-06-06 10:38:15 -06:00
|
|
|
unsigned int tagged_writepages:1;
|
2010-04-01 19:36:30 -06:00
|
|
|
unsigned int for_kupdate:1;
|
|
|
|
unsigned int range_cyclic:1;
|
|
|
|
unsigned int for_background:1;
|
2013-07-02 06:38:35 -06:00
|
|
|
unsigned int for_sync:1; /* sync(2) WB_SYNC_ALL writeback */
|
2011-10-07 21:54:10 -06:00
|
|
|
enum wb_reason reason; /* why was writeback initiated? */
|
2009-09-16 07:18:25 -06:00
|
|
|
|
2009-09-15 12:04:57 -06:00
|
|
|
struct list_head list; /* pending work list */
|
2010-07-06 00:59:53 -06:00
|
|
|
struct completion *done; /* set if the caller waits */
|
2009-09-09 01:08:54 -06:00
|
|
|
};
|
|
|
|
|
2008-04-29 01:58:56 -06:00
|
|
|
/**
|
|
|
|
* writeback_in_progress - determine whether there is writeback in progress
|
|
|
|
* @bdi: the device's backing_dev_info structure.
|
|
|
|
*
|
2009-09-09 01:08:54 -06:00
|
|
|
* Determine whether there is writeback waiting to be handled against a
|
|
|
|
* backing device.
|
2008-04-29 01:58:56 -06:00
|
|
|
*/
|
|
|
|
int writeback_in_progress(struct backing_dev_info *bdi)
|
|
|
|
{
|
2010-08-11 15:17:44 -06:00
|
|
|
return test_bit(BDI_writeback_running, &bdi->state);
|
2008-04-29 01:58:56 -06:00
|
|
|
}
|
ext4: fix potential deadlock in ext4_nonda_switch()
In ext4_nonda_switch(), if the file system is getting full we used to
call writeback_inodes_sb_if_idle(). The problem is that we can be
holding i_mutex already, and this causes a potential deadlock when
writeback_inodes_sb_if_idle() when it tries to take s_umount. (See
lockdep output below).
As it turns out we don't need need to hold s_umount; the fact that we
are in the middle of the write(2) system call will keep the superblock
pinned. Unfortunately writeback_inodes_sb() checks to make sure
s_umount is taken, and the VFS uses a different mechanism for making
sure the file system doesn't get unmounted out from under us. The
simplest way of dealing with this is to just simply grab s_umount
using a trylock, and skip kicking the writeback flusher thread in the
very unlikely case that we can't take a read lock on s_umount without
blocking.
Also, we now check the cirteria for kicking the writeback thread
before we decide to whether to fall back to non-delayed writeback, so
if there are any outstanding delayed allocation writes, we try to get
them resolved as soon as possible.
[ INFO: possible circular locking dependency detected ]
3.6.0-rc1-00042-gce894ca #367 Not tainted
-------------------------------------------------------
dd/8298 is trying to acquire lock:
(&type->s_umount_key#18){++++..}, at: [<c02277d4>] writeback_inodes_sb_if_idle+0x28/0x46
but task is already holding lock:
(&sb->s_type->i_mutex_key#8){+.+...}, at: [<c01ddcce>] generic_file_aio_write+0x5f/0xd3
which lock already depends on the new lock.
2 locks held by dd/8298:
#0: (sb_writers#2){.+.+.+}, at: [<c01ddcc5>] generic_file_aio_write+0x56/0xd3
#1: (&sb->s_type->i_mutex_key#8){+.+...}, at: [<c01ddcce>] generic_file_aio_write+0x5f/0xd3
stack backtrace:
Pid: 8298, comm: dd Not tainted 3.6.0-rc1-00042-gce894ca #367
Call Trace:
[<c015b79c>] ? console_unlock+0x345/0x372
[<c06d62a1>] print_circular_bug+0x190/0x19d
[<c019906c>] __lock_acquire+0x86d/0xb6c
[<c01999db>] ? mark_held_locks+0x5c/0x7b
[<c0199724>] lock_acquire+0x66/0xb9
[<c02277d4>] ? writeback_inodes_sb_if_idle+0x28/0x46
[<c06db935>] down_read+0x28/0x58
[<c02277d4>] ? writeback_inodes_sb_if_idle+0x28/0x46
[<c02277d4>] writeback_inodes_sb_if_idle+0x28/0x46
[<c026f3b2>] ext4_nonda_switch+0xe1/0xf4
[<c0271ece>] ext4_da_write_begin+0x27/0x193
[<c01dcdb0>] generic_file_buffered_write+0xc8/0x1bb
[<c01ddc47>] __generic_file_aio_write+0x1dd/0x205
[<c01ddce7>] generic_file_aio_write+0x78/0xd3
[<c026d336>] ext4_file_write+0x480/0x4a6
[<c0198c1d>] ? __lock_acquire+0x41e/0xb6c
[<c0180944>] ? sched_clock_cpu+0x11a/0x13e
[<c01967e9>] ? trace_hardirqs_off+0xb/0xd
[<c018099f>] ? local_clock+0x37/0x4e
[<c0209f2c>] do_sync_write+0x67/0x9d
[<c0209ec5>] ? wait_on_retry_sync_kiocb+0x44/0x44
[<c020a7b9>] vfs_write+0x7b/0xe6
[<c020a9a6>] sys_write+0x3b/0x64
[<c06dd4bd>] syscall_call+0x7/0xb
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: stable@vger.kernel.org
2012-09-19 20:42:36 -06:00
|
|
|
EXPORT_SYMBOL(writeback_in_progress);
|
2008-04-29 01:58:56 -06:00
|
|
|
|
2010-09-21 03:51:01 -06:00
|
|
|
static inline struct backing_dev_info *inode_to_bdi(struct inode *inode)
|
|
|
|
{
|
|
|
|
struct super_block *sb = inode->i_sb;
|
|
|
|
|
2010-10-04 06:25:33 -06:00
|
|
|
if (strcmp(sb->s_type->name, "bdev") == 0)
|
|
|
|
return inode->i_mapping->backing_dev_info;
|
|
|
|
|
|
|
|
return sb->s_bdi;
|
2010-09-21 03:51:01 -06:00
|
|
|
}
|
|
|
|
|
2010-10-20 18:49:30 -06:00
|
|
|
static inline struct inode *wb_inode(struct list_head *head)
|
|
|
|
{
|
|
|
|
return list_entry(head, struct inode, i_wb_list);
|
|
|
|
}
|
|
|
|
|
2012-01-17 10:18:56 -07:00
|
|
|
/*
|
|
|
|
* Include the creation of the trace points after defining the
|
|
|
|
* wb_writeback_work structure and inline functions so that the definition
|
|
|
|
* remains local to this file.
|
|
|
|
*/
|
|
|
|
#define CREATE_TRACE_POINTS
|
|
|
|
#include <trace/events/writeback.h>
|
|
|
|
|
2011-01-13 16:45:44 -07:00
|
|
|
static void bdi_queue_work(struct backing_dev_info *bdi,
|
|
|
|
struct wb_writeback_work *work)
|
|
|
|
{
|
|
|
|
trace_writeback_queue(bdi, work);
|
|
|
|
|
|
|
|
spin_lock_bh(&bdi->wb_lock);
|
|
|
|
list_add_tail(&work->list, &bdi->work_list);
|
2010-07-25 05:29:22 -06:00
|
|
|
spin_unlock_bh(&bdi->wb_lock);
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
|
|
|
|
mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2010-07-06 00:59:53 -06:00
|
|
|
static void
|
|
|
|
__bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
|
2011-10-07 21:54:10 -06:00
|
|
|
bool range_cyclic, enum wb_reason reason)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2010-07-06 00:59:53 -06:00
|
|
|
struct wb_writeback_work *work;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2009-09-13 12:07:36 -06:00
|
|
|
/*
|
|
|
|
* This is WB_SYNC_NONE writeback, so if allocation fails just
|
|
|
|
* wakeup the thread for old dirty data writeback
|
|
|
|
*/
|
2010-07-06 00:59:53 -06:00
|
|
|
work = kzalloc(sizeof(*work), GFP_ATOMIC);
|
|
|
|
if (!work) {
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
trace_writeback_nowork(bdi);
|
|
|
|
mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);
|
2010-07-06 00:59:53 -06:00
|
|
|
return;
|
2009-09-13 12:07:36 -06:00
|
|
|
}
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-07-06 00:59:53 -06:00
|
|
|
work->sync_mode = WB_SYNC_NONE;
|
|
|
|
work->nr_pages = nr_pages;
|
|
|
|
work->range_cyclic = range_cyclic;
|
2011-10-07 21:54:10 -06:00
|
|
|
work->reason = reason;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-07-06 00:59:53 -06:00
|
|
|
bdi_queue_work(bdi, work);
|
2009-09-16 07:13:54 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* bdi_start_writeback - start writeback
|
|
|
|
* @bdi: the backing device to write from
|
|
|
|
* @nr_pages: the number of pages to write
|
2011-11-23 05:56:45 -07:00
|
|
|
* @reason: reason why some writeback work was initiated
|
2009-09-16 07:13:54 -06:00
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* This does WB_SYNC_NONE opportunistic writeback. The IO is only
|
2011-03-30 19:57:33 -06:00
|
|
|
* started when this function returns, we make no guarantees on
|
2010-06-01 03:08:43 -06:00
|
|
|
* completion. Caller need not hold sb s_umount semaphore.
|
2009-09-16 07:13:54 -06:00
|
|
|
*
|
|
|
|
*/
|
2011-10-07 21:54:10 -06:00
|
|
|
void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
|
|
|
|
enum wb_reason reason)
|
2009-09-16 07:13:54 -06:00
|
|
|
{
|
2011-10-07 21:54:10 -06:00
|
|
|
__bdi_start_writeback(bdi, nr_pages, true, reason);
|
2010-06-08 10:15:15 -06:00
|
|
|
}
|
2009-09-23 06:33:40 -06:00
|
|
|
|
2010-06-08 10:15:15 -06:00
|
|
|
/**
|
|
|
|
* bdi_start_background_writeback - start background writeback
|
|
|
|
* @bdi: the backing device to write from
|
|
|
|
*
|
|
|
|
* Description:
|
2011-01-13 16:45:44 -07:00
|
|
|
* This makes sure WB_SYNC_NONE background writeback happens. When
|
|
|
|
* this function returns, it is only guaranteed that for given BDI
|
|
|
|
* some IO is happening if we are over background dirty threshold.
|
|
|
|
* Caller need not hold sb s_umount semaphore.
|
2010-06-08 10:15:15 -06:00
|
|
|
*/
|
|
|
|
void bdi_start_background_writeback(struct backing_dev_info *bdi)
|
|
|
|
{
|
2011-01-13 16:45:44 -07:00
|
|
|
/*
|
|
|
|
* We just wake up the flusher thread. It will perform background
|
|
|
|
* writeback as soon as there is no other work to do.
|
|
|
|
*/
|
2011-01-13 16:45:46 -07:00
|
|
|
trace_writeback_wake_background(bdi);
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2011-03-22 05:23:41 -06:00
|
|
|
/*
|
|
|
|
* Remove the inode from the writeback list it is on.
|
|
|
|
*/
|
|
|
|
void inode_wb_list_del(struct inode *inode)
|
|
|
|
{
|
2011-04-21 18:19:44 -06:00
|
|
|
struct backing_dev_info *bdi = inode_to_bdi(inode);
|
|
|
|
|
|
|
|
spin_lock(&bdi->wb.list_lock);
|
2011-03-22 05:23:41 -06:00
|
|
|
list_del_init(&inode->i_wb_list);
|
2011-04-21 18:19:44 -06:00
|
|
|
spin_unlock(&bdi->wb.list_lock);
|
2011-03-22 05:23:41 -06:00
|
|
|
}
|
|
|
|
|
2007-10-17 00:30:32 -06:00
|
|
|
/*
|
|
|
|
* Redirty an inode: set its when-it-was dirtied timestamp and move it to the
|
|
|
|
* furthest end of its superblock's dirty-inode list.
|
|
|
|
*
|
|
|
|
* Before stamping the inode's ->dirtied_when, we check to see whether it is
|
2009-09-02 01:19:46 -06:00
|
|
|
* already the most-recently-dirtied inode on the b_dirty list. If that is
|
2007-10-17 00:30:32 -06:00
|
|
|
* the case then the inode must have been redirtied while it was being written
|
|
|
|
* out and we don't reset its dirtied_when.
|
|
|
|
*/
|
2011-04-21 18:19:44 -06:00
|
|
|
static void redirty_tail(struct inode *inode, struct bdi_writeback *wb)
|
2007-10-17 00:30:32 -06:00
|
|
|
{
|
2011-04-21 18:19:44 -06:00
|
|
|
assert_spin_locked(&wb->list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
if (!list_empty(&wb->b_dirty)) {
|
2009-09-02 01:19:46 -06:00
|
|
|
struct inode *tail;
|
2007-10-17 00:30:32 -06:00
|
|
|
|
2010-10-20 18:49:30 -06:00
|
|
|
tail = wb_inode(wb->b_dirty.next);
|
2009-09-02 01:19:46 -06:00
|
|
|
if (time_before(inode->dirtied_when, tail->dirtied_when))
|
2007-10-17 00:30:32 -06:00
|
|
|
inode->dirtied_when = jiffies;
|
|
|
|
}
|
2010-10-20 18:49:30 -06:00
|
|
|
list_move(&inode->i_wb_list, &wb->b_dirty);
|
2007-10-17 00:30:32 -06:00
|
|
|
}
|
|
|
|
|
2007-10-17 00:30:34 -06:00
|
|
|
/*
|
2009-09-02 01:19:46 -06:00
|
|
|
* requeue inode for re-scanning after bdi->b_io list is exhausted.
|
2007-10-17 00:30:34 -06:00
|
|
|
*/
|
2011-04-21 18:19:44 -06:00
|
|
|
static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
|
2007-10-17 00:30:34 -06:00
|
|
|
{
|
2011-04-21 18:19:44 -06:00
|
|
|
assert_spin_locked(&wb->list_lock);
|
2010-10-20 18:49:30 -06:00
|
|
|
list_move(&inode->i_wb_list, &wb->b_more_io);
|
2007-10-17 00:30:34 -06:00
|
|
|
}
|
|
|
|
|
2007-10-17 00:30:44 -06:00
|
|
|
static void inode_sync_complete(struct inode *inode)
|
|
|
|
{
|
2012-05-03 06:47:55 -06:00
|
|
|
inode->i_state &= ~I_SYNC;
|
2012-11-26 17:29:51 -07:00
|
|
|
/* If inode is clean an unused, put it into LRU now... */
|
|
|
|
inode_add_lru(inode);
|
2012-05-03 06:47:55 -06:00
|
|
|
/* Waiters must see I_SYNC cleared before being woken up */
|
2007-10-17 00:30:44 -06:00
|
|
|
smp_mb();
|
|
|
|
wake_up_bit(&inode->i_state, __I_SYNC);
|
|
|
|
}
|
|
|
|
|
2009-04-02 17:56:37 -06:00
|
|
|
static bool inode_dirtied_after(struct inode *inode, unsigned long t)
|
|
|
|
{
|
|
|
|
bool ret = time_after(inode->dirtied_when, t);
|
|
|
|
#ifndef CONFIG_64BIT
|
|
|
|
/*
|
|
|
|
* For inodes being constantly redirtied, dirtied_when can get stuck.
|
|
|
|
* It _appears_ to be in the future, but is actually in distant past.
|
|
|
|
* This test is necessary to prevent such wrapped-around relative times
|
2009-09-23 11:37:09 -06:00
|
|
|
* from permanently stopping the whole bdi writeback.
|
2009-04-02 17:56:37 -06:00
|
|
|
*/
|
|
|
|
ret = ret && time_before_eq(inode->dirtied_when, jiffies);
|
|
|
|
#endif
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2007-10-17 00:30:39 -06:00
|
|
|
/*
|
2012-09-10 18:28:18 -06:00
|
|
|
* Move expired (dirtied before work->older_than_this) dirty inodes from
|
2012-03-09 08:26:22 -07:00
|
|
|
* @delaying_queue to @dispatch_queue.
|
2007-10-17 00:30:39 -06:00
|
|
|
*/
|
2011-04-23 12:27:27 -06:00
|
|
|
static int move_expired_inodes(struct list_head *delaying_queue,
|
2007-10-17 00:30:39 -06:00
|
|
|
struct list_head *dispatch_queue,
|
2011-10-07 21:51:56 -06:00
|
|
|
struct wb_writeback_work *work)
|
2007-10-17 00:30:39 -06:00
|
|
|
{
|
2009-09-24 06:42:33 -06:00
|
|
|
LIST_HEAD(tmp);
|
|
|
|
struct list_head *pos, *node;
|
2009-09-24 07:12:57 -06:00
|
|
|
struct super_block *sb = NULL;
|
2009-09-24 06:42:33 -06:00
|
|
|
struct inode *inode;
|
2009-09-24 07:12:57 -06:00
|
|
|
int do_sb_sort = 0;
|
2011-04-23 12:27:27 -06:00
|
|
|
int moved = 0;
|
2009-09-24 06:42:33 -06:00
|
|
|
|
2007-10-17 00:30:39 -06:00
|
|
|
while (!list_empty(delaying_queue)) {
|
2010-10-20 18:49:30 -06:00
|
|
|
inode = wb_inode(delaying_queue->prev);
|
2011-10-07 21:51:56 -06:00
|
|
|
if (work->older_than_this &&
|
|
|
|
inode_dirtied_after(inode, *work->older_than_this))
|
2007-10-17 00:30:39 -06:00
|
|
|
break;
|
2009-09-24 07:12:57 -06:00
|
|
|
if (sb && sb != inode->i_sb)
|
|
|
|
do_sb_sort = 1;
|
|
|
|
sb = inode->i_sb;
|
2010-10-20 18:49:30 -06:00
|
|
|
list_move(&inode->i_wb_list, &tmp);
|
2011-04-23 12:27:27 -06:00
|
|
|
moved++;
|
2009-09-24 06:42:33 -06:00
|
|
|
}
|
|
|
|
|
2009-09-24 07:12:57 -06:00
|
|
|
/* just one sb in list, splice to dispatch_queue and we're done */
|
|
|
|
if (!do_sb_sort) {
|
|
|
|
list_splice(&tmp, dispatch_queue);
|
2011-04-23 12:27:27 -06:00
|
|
|
goto out;
|
2009-09-24 07:12:57 -06:00
|
|
|
}
|
|
|
|
|
2009-09-24 06:42:33 -06:00
|
|
|
/* Move inodes from one superblock together */
|
|
|
|
while (!list_empty(&tmp)) {
|
2010-10-20 18:49:30 -06:00
|
|
|
sb = wb_inode(tmp.prev)->i_sb;
|
2009-09-24 06:42:33 -06:00
|
|
|
list_for_each_prev_safe(pos, node, &tmp) {
|
2010-10-20 18:49:30 -06:00
|
|
|
inode = wb_inode(pos);
|
2009-09-24 06:42:33 -06:00
|
|
|
if (inode->i_sb == sb)
|
2010-10-20 18:49:30 -06:00
|
|
|
list_move(&inode->i_wb_list, dispatch_queue);
|
2009-09-24 06:42:33 -06:00
|
|
|
}
|
2007-10-17 00:30:39 -06:00
|
|
|
}
|
2011-04-23 12:27:27 -06:00
|
|
|
out:
|
|
|
|
return moved;
|
2007-10-17 00:30:39 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Queue all expired dirty inodes for io, eldest first.
|
2010-08-11 15:17:42 -06:00
|
|
|
* Before
|
|
|
|
* newly dirtied b_dirty b_io b_more_io
|
|
|
|
* =============> gf edc BA
|
|
|
|
* After
|
|
|
|
* newly dirtied b_dirty b_io b_more_io
|
|
|
|
* =============> g fBAedc
|
|
|
|
* |
|
|
|
|
* +--> dequeue for IO
|
2007-10-17 00:30:39 -06:00
|
|
|
*/
|
2011-10-07 21:51:56 -06:00
|
|
|
static void queue_io(struct bdi_writeback *wb, struct wb_writeback_work *work)
|
2009-09-02 01:19:46 -06:00
|
|
|
{
|
2011-04-23 12:27:27 -06:00
|
|
|
int moved;
|
2011-04-21 18:19:44 -06:00
|
|
|
assert_spin_locked(&wb->list_lock);
|
2010-08-11 15:17:42 -06:00
|
|
|
list_splice_init(&wb->b_more_io, &wb->b_io);
|
2011-10-07 21:51:56 -06:00
|
|
|
moved = move_expired_inodes(&wb->b_dirty, &wb->b_io, work);
|
|
|
|
trace_writeback_queue_io(wb, work, moved);
|
2009-09-02 01:19:46 -06:00
|
|
|
}
|
|
|
|
|
2010-03-05 01:21:37 -07:00
|
|
|
static int write_inode(struct inode *inode, struct writeback_control *wbc)
|
2007-10-17 00:30:39 -06:00
|
|
|
{
|
2013-01-11 14:06:37 -07:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (inode->i_sb->s_op->write_inode && !is_bad_inode(inode)) {
|
|
|
|
trace_writeback_write_inode_start(inode, wbc);
|
|
|
|
ret = inode->i_sb->s_op->write_inode(inode, wbc);
|
|
|
|
trace_writeback_write_inode(inode, wbc);
|
|
|
|
return ret;
|
|
|
|
}
|
2009-09-09 01:08:54 -06:00
|
|
|
return 0;
|
2007-10-17 00:30:39 -06:00
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
2012-05-03 06:48:03 -06:00
|
|
|
* Wait for writeback on an inode to complete. Called with i_lock held.
|
|
|
|
* Caller must make sure inode cannot go away when we drop i_lock.
|
2009-06-08 05:35:40 -06:00
|
|
|
*/
|
2012-05-03 06:48:03 -06:00
|
|
|
static void __inode_wait_for_writeback(struct inode *inode)
|
|
|
|
__releases(inode->i_lock)
|
|
|
|
__acquires(inode->i_lock)
|
2009-06-08 05:35:40 -06:00
|
|
|
{
|
|
|
|
DEFINE_WAIT_BIT(wq, &inode->i_state, __I_SYNC);
|
|
|
|
wait_queue_head_t *wqh;
|
|
|
|
|
|
|
|
wqh = bit_waitqueue(&inode->i_state, __I_SYNC);
|
2011-03-22 05:23:36 -06:00
|
|
|
while (inode->i_state & I_SYNC) {
|
|
|
|
spin_unlock(&inode->i_lock);
|
2009-06-08 05:35:40 -06:00
|
|
|
__wait_on_bit(wqh, &wq, inode_wait, TASK_UNINTERRUPTIBLE);
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_lock(&inode->i_lock);
|
2010-05-24 15:32:38 -06:00
|
|
|
}
|
2009-06-08 05:35:40 -06:00
|
|
|
}
|
|
|
|
|
2012-05-03 06:48:03 -06:00
|
|
|
/*
|
|
|
|
* Wait for writeback on an inode to complete. Caller must have inode pinned.
|
|
|
|
*/
|
|
|
|
void inode_wait_for_writeback(struct inode *inode)
|
|
|
|
{
|
|
|
|
spin_lock(&inode->i_lock);
|
|
|
|
__inode_wait_for_writeback(inode);
|
|
|
|
spin_unlock(&inode->i_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Sleep until I_SYNC is cleared. This function must be called with i_lock
|
|
|
|
* held and drops it. It is aimed for callers not holding any inode reference
|
|
|
|
* so once i_lock is dropped, inode can go away.
|
|
|
|
*/
|
|
|
|
static void inode_sleep_on_writeback(struct inode *inode)
|
|
|
|
__releases(inode->i_lock)
|
|
|
|
{
|
|
|
|
DEFINE_WAIT(wait);
|
|
|
|
wait_queue_head_t *wqh = bit_waitqueue(&inode->i_state, __I_SYNC);
|
|
|
|
int sleep;
|
|
|
|
|
|
|
|
prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
|
|
|
|
sleep = inode->i_state & I_SYNC;
|
|
|
|
spin_unlock(&inode->i_lock);
|
|
|
|
if (sleep)
|
|
|
|
schedule();
|
|
|
|
finish_wait(wqh, &wait);
|
|
|
|
}
|
|
|
|
|
2012-05-03 06:47:58 -06:00
|
|
|
/*
|
|
|
|
* Find proper writeback list for the inode depending on its current state and
|
|
|
|
* possibly also change of its state while we were doing writeback. Here we
|
|
|
|
* handle things such as livelock prevention or fairness of writeback among
|
|
|
|
* inodes. This function can be called only by flusher thread - noone else
|
|
|
|
* processes all inodes in writeback lists and requeueing inodes behind flusher
|
|
|
|
* thread's back can have unexpected consequences.
|
|
|
|
*/
|
|
|
|
static void requeue_inode(struct inode *inode, struct bdi_writeback *wb,
|
|
|
|
struct writeback_control *wbc)
|
|
|
|
{
|
|
|
|
if (inode->i_state & I_FREEING)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Sync livelock prevention. Each inode is tagged and synced in one
|
|
|
|
* shot. If still dirty, it will be redirty_tail()'ed below. Update
|
|
|
|
* the dirty time to prevent enqueue and sync it again.
|
|
|
|
*/
|
|
|
|
if ((inode->i_state & I_DIRTY) &&
|
|
|
|
(wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages))
|
|
|
|
inode->dirtied_when = jiffies;
|
|
|
|
|
2012-05-03 06:48:00 -06:00
|
|
|
if (wbc->pages_skipped) {
|
|
|
|
/*
|
|
|
|
* writeback is not making progress due to locked
|
|
|
|
* buffers. Skip this inode for now.
|
|
|
|
*/
|
|
|
|
redirty_tail(inode, wb);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2012-05-03 06:47:58 -06:00
|
|
|
if (mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) {
|
|
|
|
/*
|
|
|
|
* We didn't write back all the pages. nfs_writepages()
|
|
|
|
* sometimes bales out without doing anything.
|
|
|
|
*/
|
|
|
|
if (wbc->nr_to_write <= 0) {
|
|
|
|
/* Slice used up. Queue for next turn. */
|
|
|
|
requeue_io(inode, wb);
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* Writeback blocked by something other than
|
|
|
|
* congestion. Delay the inode for some time to
|
|
|
|
* avoid spinning on the CPU (100% iowait)
|
|
|
|
* retrying writeback of the dirty page/inode
|
|
|
|
* that cannot be performed immediately.
|
|
|
|
*/
|
|
|
|
redirty_tail(inode, wb);
|
|
|
|
}
|
|
|
|
} else if (inode->i_state & I_DIRTY) {
|
|
|
|
/*
|
|
|
|
* Filesystems can dirty the inode during writeback operations,
|
|
|
|
* such as delayed allocation during submission or metadata
|
|
|
|
* updates after data IO completion.
|
|
|
|
*/
|
|
|
|
redirty_tail(inode, wb);
|
|
|
|
} else {
|
|
|
|
/* The inode is clean. Remove from writeback lists. */
|
|
|
|
list_del_init(&inode->i_wb_list);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-06-08 05:35:40 -06:00
|
|
|
/*
|
2012-05-03 06:48:00 -06:00
|
|
|
* Write out an inode and its dirty pages. Do not update the writeback list
|
|
|
|
* linkage. That is left to the caller. The caller is also responsible for
|
|
|
|
* setting I_SYNC flag and calling inode_sync_complete() to clear it.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
static int
|
2012-10-08 17:33:45 -06:00
|
|
|
__writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
struct address_space *mapping = inode->i_mapping;
|
2010-12-01 16:33:37 -07:00
|
|
|
long nr_to_write = wbc->nr_to_write;
|
2009-06-08 05:35:40 -06:00
|
|
|
unsigned dirty;
|
2005-04-16 16:20:36 -06:00
|
|
|
int ret;
|
|
|
|
|
2012-05-03 06:48:00 -06:00
|
|
|
WARN_ON(!(inode->i_state & I_SYNC));
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-01-11 14:06:37 -07:00
|
|
|
trace_writeback_single_inode_start(inode, wbc, nr_to_write);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
ret = do_writepages(mapping, wbc);
|
|
|
|
|
2010-03-05 01:21:21 -07:00
|
|
|
/*
|
|
|
|
* Make sure to wait on the data before writing out the metadata.
|
|
|
|
* This is important for filesystems that modify metadata on data
|
2013-07-02 06:38:35 -06:00
|
|
|
* I/O completion. We don't do it for sync(2) writeback because it has a
|
|
|
|
* separate, external IO completion path and ->sync_fs for guaranteeing
|
|
|
|
* inode metadata is written back correctly.
|
2010-03-05 01:21:21 -07:00
|
|
|
*/
|
2013-07-02 06:38:35 -06:00
|
|
|
if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync) {
|
2010-03-05 01:21:21 -07:00
|
|
|
int err = filemap_fdatawait(mapping);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (ret == 0)
|
|
|
|
ret = err;
|
|
|
|
}
|
|
|
|
|
2010-05-07 03:35:44 -06:00
|
|
|
/*
|
|
|
|
* Some filesystems may redirty the inode during the writeback
|
|
|
|
* due to delalloc, clear dirty metadata flags right before
|
|
|
|
* write_inode()
|
|
|
|
*/
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_lock(&inode->i_lock);
|
2012-05-03 06:47:57 -06:00
|
|
|
/* Clear I_DIRTY_PAGES if we've written out all dirty pages */
|
|
|
|
if (!mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
|
|
|
|
inode->i_state &= ~I_DIRTY_PAGES;
|
2010-05-07 03:35:44 -06:00
|
|
|
dirty = inode->i_state & I_DIRTY;
|
|
|
|
inode->i_state &= ~(I_DIRTY_SYNC | I_DIRTY_DATASYNC);
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_unlock(&inode->i_lock);
|
2010-03-05 01:21:21 -07:00
|
|
|
/* Don't write the inode if only I_DIRTY_PAGES was set */
|
|
|
|
if (dirty & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) {
|
2010-03-05 01:21:37 -07:00
|
|
|
int err = write_inode(inode, wbc);
|
2005-04-16 16:20:36 -06:00
|
|
|
if (ret == 0)
|
|
|
|
ret = err;
|
|
|
|
}
|
2012-05-03 06:48:00 -06:00
|
|
|
trace_writeback_single_inode(inode, wbc, nr_to_write);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Write out an inode's dirty pages. Either the caller has an active reference
|
|
|
|
* on the inode or the inode has I_WILL_FREE set.
|
|
|
|
*
|
|
|
|
* This function is designed to be called for writing back one inode which
|
|
|
|
* we go e.g. from filesystem. Flusher thread uses __writeback_single_inode()
|
|
|
|
* and does more profound writeback list handling in writeback_sb_inodes().
|
|
|
|
*/
|
|
|
|
static int
|
|
|
|
writeback_single_inode(struct inode *inode, struct bdi_writeback *wb,
|
|
|
|
struct writeback_control *wbc)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
spin_lock(&inode->i_lock);
|
|
|
|
if (!atomic_read(&inode->i_count))
|
|
|
|
WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING)));
|
|
|
|
else
|
|
|
|
WARN_ON(inode->i_state & I_WILL_FREE);
|
|
|
|
|
|
|
|
if (inode->i_state & I_SYNC) {
|
|
|
|
if (wbc->sync_mode != WB_SYNC_ALL)
|
|
|
|
goto out;
|
|
|
|
/*
|
2012-05-03 06:48:03 -06:00
|
|
|
* It's a data-integrity sync. We must wait. Since callers hold
|
|
|
|
* inode reference or inode has I_WILL_FREE set, it cannot go
|
|
|
|
* away under us.
|
2012-05-03 06:48:00 -06:00
|
|
|
*/
|
2012-05-03 06:48:03 -06:00
|
|
|
__inode_wait_for_writeback(inode);
|
2012-05-03 06:48:00 -06:00
|
|
|
}
|
|
|
|
WARN_ON(inode->i_state & I_SYNC);
|
|
|
|
/*
|
|
|
|
* Skip inode if it is clean. We don't want to mess with writeback
|
|
|
|
* lists in this function since flusher thread may be doing for example
|
|
|
|
* sync in parallel and if we move the inode, it could get skipped. So
|
|
|
|
* here we make sure inode is on some writeback list and leave it there
|
|
|
|
* unless we have completely cleaned the inode.
|
|
|
|
*/
|
|
|
|
if (!(inode->i_state & I_DIRTY))
|
|
|
|
goto out;
|
|
|
|
inode->i_state |= I_SYNC;
|
|
|
|
spin_unlock(&inode->i_lock);
|
|
|
|
|
2012-10-08 17:33:45 -06:00
|
|
|
ret = __writeback_single_inode(inode, wbc);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-04-21 18:19:44 -06:00
|
|
|
spin_lock(&wb->list_lock);
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_lock(&inode->i_lock);
|
2012-05-03 06:48:00 -06:00
|
|
|
/*
|
|
|
|
* If inode is clean, remove it from writeback lists. Otherwise don't
|
|
|
|
* touch it. See comment above for explanation.
|
|
|
|
*/
|
|
|
|
if (!(inode->i_state & I_DIRTY))
|
|
|
|
list_del_init(&inode->i_wb_list);
|
|
|
|
spin_unlock(&wb->list_lock);
|
2007-10-17 00:30:44 -06:00
|
|
|
inode_sync_complete(inode);
|
2012-05-03 06:48:00 -06:00
|
|
|
out:
|
|
|
|
spin_unlock(&inode->i_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2010-08-29 13:28:09 -06:00
|
|
|
static long writeback_chunk_size(struct backing_dev_info *bdi,
|
|
|
|
struct wb_writeback_work *work)
|
2011-05-04 19:54:37 -06:00
|
|
|
{
|
|
|
|
long pages;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* WB_SYNC_ALL mode does livelock avoidance by syncing dirty
|
|
|
|
* inodes/pages in one big loop. Setting wbc.nr_to_write=LONG_MAX
|
|
|
|
* here avoids calling into writeback_inodes_wb() more than once.
|
|
|
|
*
|
|
|
|
* The intended call sequence for WB_SYNC_ALL writeback is:
|
|
|
|
*
|
|
|
|
* wb_writeback()
|
|
|
|
* writeback_sb_inodes() <== called only once
|
|
|
|
* write_cache_pages() <== called once for each inode
|
|
|
|
* (quickly) tag currently dirty pages
|
|
|
|
* (maybe slowly) sync all tagged pages
|
|
|
|
*/
|
|
|
|
if (work->sync_mode == WB_SYNC_ALL || work->tagged_writepages)
|
|
|
|
pages = LONG_MAX;
|
2010-08-29 13:28:09 -06:00
|
|
|
else {
|
|
|
|
pages = min(bdi->avg_write_bandwidth / 2,
|
|
|
|
global_dirty_limit / DIRTY_SCOPE);
|
|
|
|
pages = min(pages, work->nr_pages);
|
|
|
|
pages = round_down(pages + MIN_WRITEBACK_PAGES,
|
|
|
|
MIN_WRITEBACK_PAGES);
|
|
|
|
}
|
2011-05-04 19:54:37 -06:00
|
|
|
|
|
|
|
return pages;
|
|
|
|
}
|
|
|
|
|
2010-03-11 15:09:47 -07:00
|
|
|
/*
|
|
|
|
* Write a portion of b_io inodes which belong to @sb.
|
2010-06-10 04:07:54 -06:00
|
|
|
*
|
2011-05-04 19:54:37 -06:00
|
|
|
* Return the number of pages and/or inodes written.
|
2010-03-11 15:09:47 -07:00
|
|
|
*/
|
2011-05-04 19:54:37 -06:00
|
|
|
static long writeback_sb_inodes(struct super_block *sb,
|
|
|
|
struct bdi_writeback *wb,
|
|
|
|
struct wb_writeback_work *work)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-05-04 19:54:37 -06:00
|
|
|
struct writeback_control wbc = {
|
|
|
|
.sync_mode = work->sync_mode,
|
|
|
|
.tagged_writepages = work->tagged_writepages,
|
|
|
|
.for_kupdate = work->for_kupdate,
|
|
|
|
.for_background = work->for_background,
|
2013-07-02 06:38:35 -06:00
|
|
|
.for_sync = work->for_sync,
|
2011-05-04 19:54:37 -06:00
|
|
|
.range_cyclic = work->range_cyclic,
|
|
|
|
.range_start = 0,
|
|
|
|
.range_end = LLONG_MAX,
|
|
|
|
};
|
|
|
|
unsigned long start_time = jiffies;
|
|
|
|
long write_chunk;
|
|
|
|
long wrote = 0; /* count both pages and inodes */
|
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
while (!list_empty(&wb->b_io)) {
|
2010-10-20 18:49:30 -06:00
|
|
|
struct inode *inode = wb_inode(wb->b_io.prev);
|
2010-06-10 04:07:54 -06:00
|
|
|
|
|
|
|
if (inode->i_sb != sb) {
|
2011-05-04 19:54:37 -06:00
|
|
|
if (work->sb) {
|
2010-06-10 04:07:54 -06:00
|
|
|
/*
|
|
|
|
* We only want to write back data for this
|
|
|
|
* superblock, move all inodes not belonging
|
|
|
|
* to it back onto the dirty list.
|
|
|
|
*/
|
2011-04-21 18:19:44 -06:00
|
|
|
redirty_tail(inode, wb);
|
2010-06-10 04:07:54 -06:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The inode belongs to a different superblock.
|
|
|
|
* Bounce back to the caller to unpin this and
|
|
|
|
* pin the next superblock.
|
|
|
|
*/
|
2011-05-04 19:54:37 -06:00
|
|
|
break;
|
2010-06-10 04:07:54 -06:00
|
|
|
}
|
|
|
|
|
2010-10-24 11:40:46 -06:00
|
|
|
/*
|
2012-06-08 21:10:55 -06:00
|
|
|
* Don't bother with new inodes or inodes being freed, first
|
|
|
|
* kind does not need periodic writeout yet, and for the latter
|
2010-10-24 11:40:46 -06:00
|
|
|
* kind writeout is handled by the freer.
|
|
|
|
*/
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_lock(&inode->i_lock);
|
2010-10-24 11:40:46 -06:00
|
|
|
if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) {
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-07-12 00:08:50 -06:00
|
|
|
redirty_tail(inode, wb);
|
fs: new inode i_state corruption fix
There was a report of a data corruption
http://lkml.org/lkml/2008/11/14/121. There is a script included to
reproduce the problem.
During testing, I encountered a number of strange things with ext3, so I
tried ext2 to attempt to reduce complexity of the problem. I found that
fsstress would quickly hang in wait_on_inode, waiting for I_LOCK to be
cleared, even though instrumentation showed that unlock_new_inode had
already been called for that inode. This points to memory scribble, or
synchronisation problme.
i_state of I_NEW inodes is not protected by inode_lock because other
processes are not supposed to touch them until I_LOCK (and I_NEW) is
cleared. Adding WARN_ON(inode->i_state & I_NEW) to sites where we modify
i_state revealed that generic_sync_sb_inodes is picking up new inodes from
the inode lists and passing them to __writeback_single_inode without
waiting for I_NEW. Subsequently modifying i_state causes corruption. In
my case it would look like this:
CPU0 CPU1
unlock_new_inode() __sync_single_inode()
reg <- inode->i_state
reg -> reg & ~(I_LOCK|I_NEW) reg <- inode->i_state
reg -> inode->i_state reg -> reg | I_SYNC
reg -> inode->i_state
Non-atomic RMW on CPU1 overwrites CPU0 store and sets I_LOCK|I_NEW again.
Fix for this is rather than wait for I_NEW inodes, just skip over them:
inodes concurrently being created are not subject to data integrity
operations, and should not significantly contribute to dirty memory
either.
After this change, I'm unable to reproduce any of the added warnings or
hangs after ~1hour of running. Previously, the new warnings would start
immediately and hang would happen in under 5 minutes.
I'm also testing on ext3 now, and so far no problems there either. I
don't know whether this fixes the problem reported above, but it fixes a
real problem for me.
Cc: "Jorge Boncompte [DTI2]" <jorge@dti2.net>
Reported-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Cc: Jan Kara <jack@suse.cz>
Cc: <stable@kernel.org>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-12 15:31:38 -06:00
|
|
|
continue;
|
|
|
|
}
|
2012-05-03 06:47:56 -06:00
|
|
|
if ((inode->i_state & I_SYNC) && wbc.sync_mode != WB_SYNC_ALL) {
|
|
|
|
/*
|
|
|
|
* If this inode is locked for writeback and we are not
|
|
|
|
* doing writeback-for-data-integrity, move it to
|
|
|
|
* b_more_io so that writeback can proceed with the
|
|
|
|
* other inodes on s_io.
|
|
|
|
*
|
|
|
|
* We'll have another go at writing back this inode
|
|
|
|
* when we completed a full scan of b_io.
|
|
|
|
*/
|
|
|
|
spin_unlock(&inode->i_lock);
|
|
|
|
requeue_io(inode, wb);
|
|
|
|
trace_writeback_sb_inodes_requeue(inode);
|
|
|
|
continue;
|
|
|
|
}
|
2012-05-03 06:47:59 -06:00
|
|
|
spin_unlock(&wb->list_lock);
|
|
|
|
|
2012-05-03 06:48:00 -06:00
|
|
|
/*
|
|
|
|
* We already requeued the inode if it had I_SYNC set and we
|
|
|
|
* are doing WB_SYNC_NONE writeback. So this catches only the
|
|
|
|
* WB_SYNC_ALL case.
|
|
|
|
*/
|
2012-05-03 06:48:03 -06:00
|
|
|
if (inode->i_state & I_SYNC) {
|
|
|
|
/* Wait for I_SYNC. This function drops i_lock... */
|
|
|
|
inode_sleep_on_writeback(inode);
|
|
|
|
/* Inode may be gone, start again */
|
2012-06-08 09:07:36 -06:00
|
|
|
spin_lock(&wb->list_lock);
|
2012-05-03 06:48:03 -06:00
|
|
|
continue;
|
|
|
|
}
|
2012-05-03 06:48:00 -06:00
|
|
|
inode->i_state |= I_SYNC;
|
|
|
|
spin_unlock(&inode->i_lock);
|
2012-05-03 06:48:03 -06:00
|
|
|
|
2010-08-29 13:28:09 -06:00
|
|
|
write_chunk = writeback_chunk_size(wb->bdi, work);
|
2011-05-04 19:54:37 -06:00
|
|
|
wbc.nr_to_write = write_chunk;
|
|
|
|
wbc.pages_skipped = 0;
|
2011-03-22 05:23:36 -06:00
|
|
|
|
2012-05-03 06:48:03 -06:00
|
|
|
/*
|
|
|
|
* We use I_SYNC to pin the inode in memory. While it is set
|
|
|
|
* evict_inode() will wait so the inode cannot be freed.
|
|
|
|
*/
|
2012-10-08 17:33:45 -06:00
|
|
|
__writeback_single_inode(inode, &wbc);
|
2011-03-22 05:23:36 -06:00
|
|
|
|
2011-05-04 19:54:37 -06:00
|
|
|
work->nr_pages -= write_chunk - wbc.nr_to_write;
|
|
|
|
wrote += write_chunk - wbc.nr_to_write;
|
2012-05-03 06:48:00 -06:00
|
|
|
spin_lock(&wb->list_lock);
|
|
|
|
spin_lock(&inode->i_lock);
|
2011-05-04 19:54:37 -06:00
|
|
|
if (!(inode->i_state & I_DIRTY))
|
|
|
|
wrote++;
|
2012-05-03 06:48:00 -06:00
|
|
|
requeue_inode(inode, wb, &wbc);
|
|
|
|
inode_sync_complete(inode);
|
2011-03-22 05:23:43 -06:00
|
|
|
spin_unlock(&inode->i_lock);
|
2012-05-03 06:48:03 -06:00
|
|
|
cond_resched_lock(&wb->list_lock);
|
2011-05-04 19:54:37 -06:00
|
|
|
/*
|
|
|
|
* bail out to wb_writeback() often enough to check
|
|
|
|
* background threshold and other termination conditions.
|
|
|
|
*/
|
|
|
|
if (wrote) {
|
|
|
|
if (time_is_before_jiffies(start_time + HZ / 10UL))
|
|
|
|
break;
|
|
|
|
if (work->nr_pages <= 0)
|
|
|
|
break;
|
writeback: speed up writeback of big dirty files
After making dirty a 100M file, the normal behavior is to start the
writeback for all data after 30s delays. But sometimes the following
happens instead:
- after 30s: ~4M
- after 5s: ~4M
- after 5s: all remaining 92M
Some analyze shows that the internal io dispatch queues goes like this:
s_io s_more_io
-------------------------
1) 100M,1K 0
2) 1K 96M
3) 0 96M
1) initial state with a 100M file and a 1K file
2) 4M written, nr_to_write <= 0, so write more
3) 1K written, nr_to_write > 0, no more writes(BUG)
nr_to_write > 0 in (3) fools the upper layer to think that data have all
been written out. The big dirty file is actually still sitting in
s_more_io. We cannot simply splice s_more_io back to s_io as soon as s_io
becomes empty, and let the loop in generic_sync_sb_inodes() continue: this
may starve newly expired inodes in s_dirty. It is also not an option to
draw inodes from both s_more_io and s_dirty, an let the loop go on: this
might lead to live locks, and might also starve other superblocks in sync
time(well kupdate may still starve some superblocks, that's another bug).
We have to return when a full scan of s_io completes. So nr_to_write > 0
does not necessarily mean that "all data are written". This patch
introduces a flag writeback_control.more_io to indicate that more io should
be done. With it the big dirty file no longer has to wait for the next
kupdate invokation 5s later.
In sync_sb_inodes() we only set more_io on super_blocks we actually
visited. This avoids the interaction between two pdflush deamons.
Also in __sync_single_inode() we don't blindly keep requeuing the io if the
filesystem cannot progress. Failing to do so may lead to 100% iowait.
Tested-by: Mike Snitzer <snitzer@gmail.com>
Signed-off-by: Fengguang Wu <wfg@mail.ustc.edu.cn>
Cc: Michael Rubin <mrubin@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-04 23:29:36 -07:00
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2011-05-04 19:54:37 -06:00
|
|
|
return wrote;
|
2010-03-11 15:09:47 -07:00
|
|
|
}
|
|
|
|
|
2011-05-04 19:54:37 -06:00
|
|
|
static long __writeback_inodes_wb(struct bdi_writeback *wb,
|
|
|
|
struct wb_writeback_work *work)
|
2010-03-11 15:09:47 -07:00
|
|
|
{
|
2011-05-04 19:54:37 -06:00
|
|
|
unsigned long start_time = jiffies;
|
|
|
|
long wrote = 0;
|
2009-01-06 15:40:25 -07:00
|
|
|
|
2010-03-11 15:09:47 -07:00
|
|
|
while (!list_empty(&wb->b_io)) {
|
2010-10-20 18:49:30 -06:00
|
|
|
struct inode *inode = wb_inode(wb->b_io.prev);
|
2010-03-11 15:09:47 -07:00
|
|
|
struct super_block *sb = inode->i_sb;
|
2009-09-24 07:25:11 -06:00
|
|
|
|
2011-07-07 22:14:41 -06:00
|
|
|
if (!grab_super_passive(sb)) {
|
2011-07-29 22:14:35 -06:00
|
|
|
/*
|
|
|
|
* grab_super_passive() may fail consistently due to
|
|
|
|
* s_umount being grabbed by someone else. Don't use
|
|
|
|
* requeue_io() to avoid busy retrying the inode/sb.
|
|
|
|
*/
|
|
|
|
redirty_tail(inode, wb);
|
2010-06-10 04:07:54 -06:00
|
|
|
continue;
|
2010-03-11 15:09:47 -07:00
|
|
|
}
|
2011-05-04 19:54:37 -06:00
|
|
|
wrote += writeback_sb_inodes(sb, wb, work);
|
2010-06-10 04:07:54 -06:00
|
|
|
drop_super(sb);
|
2010-03-11 15:09:47 -07:00
|
|
|
|
2011-05-04 19:54:37 -06:00
|
|
|
/* refer to the same tests at the end of writeback_sb_inodes */
|
|
|
|
if (wrote) {
|
|
|
|
if (time_is_before_jiffies(start_time + HZ / 10UL))
|
|
|
|
break;
|
|
|
|
if (work->nr_pages <= 0)
|
|
|
|
break;
|
|
|
|
}
|
2010-03-11 15:09:47 -07:00
|
|
|
}
|
2009-09-02 01:19:46 -06:00
|
|
|
/* Leave any unwritten inodes on b_io */
|
2011-05-04 19:54:37 -06:00
|
|
|
return wrote;
|
2009-09-02 01:19:46 -06:00
|
|
|
}
|
|
|
|
|
2011-10-07 21:54:10 -06:00
|
|
|
long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages,
|
|
|
|
enum wb_reason reason)
|
2010-06-10 04:07:54 -06:00
|
|
|
{
|
2011-05-04 19:54:37 -06:00
|
|
|
struct wb_writeback_work work = {
|
|
|
|
.nr_pages = nr_pages,
|
|
|
|
.sync_mode = WB_SYNC_NONE,
|
|
|
|
.range_cyclic = 1,
|
2011-10-07 21:54:10 -06:00
|
|
|
.reason = reason,
|
2011-05-04 19:54:37 -06:00
|
|
|
};
|
2010-06-10 04:07:54 -06:00
|
|
|
|
2011-04-21 18:19:44 -06:00
|
|
|
spin_lock(&wb->list_lock);
|
writeback: refill b_io iff empty
There is no point to carry different refill policies between for_kupdate
and other type of works. Use a consistent "refill b_io iff empty" policy
which can guarantee fairness in an easy to understand way.
A b_io refill will setup a _fixed_ work set with all currently eligible
inodes and start a new round of walk through b_io. The "fixed" work set
means no new inodes will be added to the work set during the walk.
Only when a complete walk over b_io is done, new inodes that are
eligible at the time will be enqueued and the walk be started over.
This procedure provides fairness among the inodes because it guarantees
each inode to be synced once and only once at each round. So all inodes
will be free from starvations.
This change relies on wb_writeback() to keep retrying as long as we made
some progress on cleaning some pages and/or inodes. Without that ability,
the old logic on background works relies on aggressively queuing all
eligible inodes into b_io at every time. But that's not a guarantee.
The below test script completes a slightly faster now:
2.6.39-rc3 2.6.39-rc3-dyn-expire+
------------------------------------------------
all elapsed 256.043 252.367
stddev 24.381 12.530
tar elapsed 30.097 28.808
dd elapsed 13.214 11.782
#!/bin/zsh
cp /c/linux-2.6.38.3.tar.bz2 /dev/shm/
umount /dev/sda7
mkfs.xfs -f /dev/sda7
mount /dev/sda7 /fs
echo 3 > /proc/sys/vm/drop_caches
tic=$(cat /proc/uptime|cut -d' ' -f2)
cd /fs
time tar jxf /dev/shm/linux-2.6.38.3.tar.bz2 &
time dd if=/dev/zero of=/fs/zero bs=1M count=1000 &
wait
sync
tac=$(cat /proc/uptime|cut -d' ' -f2)
echo elapsed: $((tac - tic))
It maintains roughly the same small vs. large file writeout shares, and
offers large files better chances to be written in nice 4M chunks.
Analyzes from Dave Chinner in great details:
Let's say we have lots of inodes with 100 dirty pages being created,
and one large writeback going on. We expire 8 new inodes for every
1024 pages we write back.
With the old code, we do:
b_more_io (large inode) -> b_io (1l)
8 newly expired inodes -> b_io (1l, 8s)
writeback large inode 1024 pages -> b_more_io
b_more_io (large inode) -> b_io (8s, 1l)
8 newly expired inodes -> b_io (8s, 1l, 8s)
writeback 8 small inodes 800 pages
1 large inode 224 pages -> b_more_io
b_more_io (large inode) -> b_io (8s, 1l)
8 newly expired inodes -> b_io (8s, 1l, 8s)
.....
Your new code:
b_more_io (large inode) -> b_io (1l)
8 newly expired inodes -> b_io (1l, 8s)
writeback large inode 1024 pages -> b_more_io
(b_io == 8s)
writeback 8 small inodes 800 pages
b_io empty: (1800 pages written)
b_more_io (large inode) -> b_io (1l)
14 newly expired inodes -> b_io (1l, 14s)
writeback large inode 1024 pages -> b_more_io
(b_io == 14s)
writeback 10 small inodes 1000 pages
1 small inode 24 pages -> b_more_io (1l, 1s(24))
writeback 5 small inodes 500 pages
b_io empty: (2548 pages written)
b_more_io (large inode) -> b_io (1l, 1s(24))
20 newly expired inodes -> b_io (1l, 1s(24), 20s)
......
Rough progression of pages written at b_io refill:
Old code:
total large file % of writeback
1024 224 21.9% (fixed)
New code:
total large file % of writeback
1800 1024 ~55%
2550 1024 ~40%
3050 1024 ~33%
3500 1024 ~29%
3950 1024 ~26%
4250 1024 ~24%
4500 1024 ~22.7%
4700 1024 ~21.7%
4800 1024 ~21.3%
4800 1024 ~21.3%
(pretty much steady state from here)
Ok, so the steady state is reached with a similar percentage of
writeback to the large file as the existing code. Ok, that's good,
but providing some evidence that is doesn't change the shared of
writeback to the large should be in the commit message ;)
The other advantage to this is that we always write 1024 page chunks
to the large file, rather than smaller "whatever remains" chunks.
CC: Jan Kara <jack@suse.cz>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-07-21 20:11:53 -06:00
|
|
|
if (list_empty(&wb->b_io))
|
2011-10-07 21:51:56 -06:00
|
|
|
queue_io(wb, &work);
|
2011-05-04 19:54:37 -06:00
|
|
|
__writeback_inodes_wb(wb, &work);
|
2011-04-21 18:19:44 -06:00
|
|
|
spin_unlock(&wb->list_lock);
|
2010-06-10 04:07:54 -06:00
|
|
|
|
2011-05-04 19:54:37 -06:00
|
|
|
return nr_pages - work.nr_pages;
|
|
|
|
}
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-11-18 13:38:33 -07:00
|
|
|
static bool over_bground_thresh(struct backing_dev_info *bdi)
|
2009-09-09 01:08:54 -06:00
|
|
|
{
|
|
|
|
unsigned long background_thresh, dirty_thresh;
|
|
|
|
|
2010-08-11 15:17:39 -06:00
|
|
|
global_dirty_limits(&background_thresh, &dirty_thresh);
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-11-18 13:38:33 -07:00
|
|
|
if (global_page_state(NR_FILE_DIRTY) +
|
|
|
|
global_page_state(NR_UNSTABLE_NFS) > background_thresh)
|
|
|
|
return true;
|
|
|
|
|
|
|
|
if (bdi_stat(bdi, BDI_RECLAIMABLE) >
|
|
|
|
bdi_dirty_limit(bdi, background_thresh))
|
|
|
|
return true;
|
|
|
|
|
|
|
|
return false;
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
|
2010-08-29 11:22:30 -06:00
|
|
|
/*
|
|
|
|
* Called under wb->list_lock. If there are multiple wb per bdi,
|
|
|
|
* only the flusher working on the first wb should do it.
|
|
|
|
*/
|
|
|
|
static void wb_update_bandwidth(struct bdi_writeback *wb,
|
|
|
|
unsigned long start_time)
|
|
|
|
{
|
2011-10-03 20:46:17 -06:00
|
|
|
__bdi_update_bandwidth(wb->bdi, 0, 0, 0, 0, 0, start_time);
|
2010-08-29 11:22:30 -06:00
|
|
|
}
|
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
/*
|
|
|
|
* Explicit flushing or periodic writeback of "old" data.
|
2009-09-02 01:19:46 -06:00
|
|
|
*
|
2009-09-09 01:08:54 -06:00
|
|
|
* Define "old": the first time one of an inode's pages is dirtied, we mark the
|
|
|
|
* dirtying-time in the inode's address_space. So this periodic writeback code
|
|
|
|
* just walks the superblock inode list, writing back any inodes which are
|
|
|
|
* older than a specific point in time.
|
2009-09-02 01:19:46 -06:00
|
|
|
*
|
2009-09-09 01:08:54 -06:00
|
|
|
* Try to run once per dirty_writeback_interval. But if a writeback event
|
|
|
|
* takes longer than a dirty_writeback_interval interval, then leave a
|
|
|
|
* one-second gap.
|
2009-09-02 01:19:46 -06:00
|
|
|
*
|
2009-09-09 01:08:54 -06:00
|
|
|
* older_than_this takes precedence over nr_to_write. So we'll only write back
|
|
|
|
* all dirty pages if they are all attached to "old" mappings.
|
2009-09-02 01:19:46 -06:00
|
|
|
*/
|
2009-09-16 07:18:25 -06:00
|
|
|
static long wb_writeback(struct bdi_writeback *wb,
|
2010-07-06 00:59:53 -06:00
|
|
|
struct wb_writeback_work *work)
|
2009-09-02 01:19:46 -06:00
|
|
|
{
|
2010-08-29 11:22:30 -06:00
|
|
|
unsigned long wb_start = jiffies;
|
2011-05-04 19:54:37 -06:00
|
|
|
long nr_pages = work->nr_pages;
|
2009-09-09 01:08:54 -06:00
|
|
|
unsigned long oldest_jif;
|
2009-09-16 11:22:48 -06:00
|
|
|
struct inode *inode;
|
2011-05-04 19:54:37 -06:00
|
|
|
long progress;
|
2009-09-02 01:19:46 -06:00
|
|
|
|
2011-04-23 11:26:07 -06:00
|
|
|
oldest_jif = jiffies;
|
2011-05-04 19:54:37 -06:00
|
|
|
work->older_than_this = &oldest_jif;
|
2009-01-06 15:40:25 -07:00
|
|
|
|
2011-04-21 12:06:32 -06:00
|
|
|
spin_lock(&wb->list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
for (;;) {
|
|
|
|
/*
|
2009-09-23 06:33:40 -06:00
|
|
|
* Stop writeback when nr_pages has been consumed
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
2010-07-06 00:59:53 -06:00
|
|
|
if (work->nr_pages <= 0)
|
2009-09-09 01:08:54 -06:00
|
|
|
break;
|
2009-09-02 01:19:46 -06:00
|
|
|
|
2011-01-13 16:45:47 -07:00
|
|
|
/*
|
|
|
|
* Background writeout and kupdate-style writeback may
|
|
|
|
* run forever. Stop them if there is other work to do
|
|
|
|
* so that e.g. sync can proceed. They'll be restarted
|
|
|
|
* after the other works are all done.
|
|
|
|
*/
|
|
|
|
if ((work->for_background || work->for_kupdate) &&
|
|
|
|
!list_empty(&wb->bdi->work_list))
|
|
|
|
break;
|
|
|
|
|
2009-01-06 15:40:25 -07:00
|
|
|
/*
|
2009-09-23 06:33:40 -06:00
|
|
|
* For background writeout, stop when we are below the
|
|
|
|
* background dirty threshold
|
2009-01-06 15:40:25 -07:00
|
|
|
*/
|
2010-11-18 13:38:33 -07:00
|
|
|
if (work->for_background && !over_bground_thresh(wb->bdi))
|
2009-09-09 01:08:54 -06:00
|
|
|
break;
|
2009-01-06 15:40:25 -07:00
|
|
|
|
2011-10-19 03:44:41 -06:00
|
|
|
/*
|
|
|
|
* Kupdate and background works are special and we want to
|
|
|
|
* include all inodes that need writing. Livelock avoidance is
|
|
|
|
* handled by these works yielding to any other work so we are
|
|
|
|
* safe.
|
|
|
|
*/
|
2010-07-21 20:32:30 -06:00
|
|
|
if (work->for_kupdate) {
|
|
|
|
oldest_jif = jiffies -
|
|
|
|
msecs_to_jiffies(dirty_expire_interval * 10);
|
2011-10-19 03:44:41 -06:00
|
|
|
} else if (work->for_background)
|
|
|
|
oldest_jif = jiffies;
|
2010-07-06 21:24:07 -06:00
|
|
|
|
2011-05-04 19:54:37 -06:00
|
|
|
trace_writeback_start(wb->bdi, work);
|
2011-04-21 12:06:32 -06:00
|
|
|
if (list_empty(&wb->b_io))
|
2011-10-07 21:51:56 -06:00
|
|
|
queue_io(wb, work);
|
2010-07-06 00:59:53 -06:00
|
|
|
if (work->sb)
|
2011-05-04 19:54:37 -06:00
|
|
|
progress = writeback_sb_inodes(work->sb, wb, work);
|
2010-06-10 04:07:54 -06:00
|
|
|
else
|
2011-05-04 19:54:37 -06:00
|
|
|
progress = __writeback_inodes_wb(wb, work);
|
|
|
|
trace_writeback_written(wb->bdi, work);
|
2010-07-06 21:24:07 -06:00
|
|
|
|
2010-08-29 11:22:30 -06:00
|
|
|
wb_update_bandwidth(wb, wb_start);
|
2009-09-09 01:08:54 -06:00
|
|
|
|
|
|
|
/*
|
writeback: try more writeback as long as something was written
writeback_inodes_wb()/__writeback_inodes_sb() are not aggressive in that
they only populate possibly a subset of eligible inodes into b_io at
entrance time. When the queued set of inodes are all synced, they just
return, possibly with all queued inode pages written but still
wbc.nr_to_write > 0.
For kupdate and background writeback, there may be more eligible inodes
sitting in b_dirty when the current set of b_io inodes are completed. So
it is necessary to try another round of writeback as long as we made some
progress in this round. When there are no more eligible inodes, no more
inodes will be enqueued in queue_io(), hence nothing could/will be
synced and we may safely bail.
For example, imagine 100 inodes
i0, i1, i2, ..., i90, i91, i99
At queue_io() time, i90-i99 happen to be expired and moved to s_io for
IO. When finished successfully, if their total size is less than
MAX_WRITEBACK_PAGES, nr_to_write will be > 0. Then wb_writeback() will
quit the background work (w/o this patch) while it's still over
background threshold. This will be a fairly normal/frequent case I guess.
Now that we do tagged sync and update inode->dirtied_when after the sync,
this change won't livelock sync(1). I actually tried to write 1 page
per 1ms with this command
write-and-fsync -n10000 -S 1000 -c 4096 /fs/test
and do sync(1) at the same time. The sync completes quickly on ext4,
xfs, btrfs.
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-07-22 10:23:44 -06:00
|
|
|
* Did we write something? Try for more
|
|
|
|
*
|
|
|
|
* Dirty inodes are moved to b_io for writeback in batches.
|
|
|
|
* The completion of the current batch does not necessarily
|
|
|
|
* mean the overall work is done. So we keep looping as long
|
|
|
|
* as made some progress on cleaning pages or inodes.
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
2011-05-04 19:54:37 -06:00
|
|
|
if (progress)
|
2009-09-23 11:32:26 -06:00
|
|
|
continue;
|
|
|
|
/*
|
writeback: try more writeback as long as something was written
writeback_inodes_wb()/__writeback_inodes_sb() are not aggressive in that
they only populate possibly a subset of eligible inodes into b_io at
entrance time. When the queued set of inodes are all synced, they just
return, possibly with all queued inode pages written but still
wbc.nr_to_write > 0.
For kupdate and background writeback, there may be more eligible inodes
sitting in b_dirty when the current set of b_io inodes are completed. So
it is necessary to try another round of writeback as long as we made some
progress in this round. When there are no more eligible inodes, no more
inodes will be enqueued in queue_io(), hence nothing could/will be
synced and we may safely bail.
For example, imagine 100 inodes
i0, i1, i2, ..., i90, i91, i99
At queue_io() time, i90-i99 happen to be expired and moved to s_io for
IO. When finished successfully, if their total size is less than
MAX_WRITEBACK_PAGES, nr_to_write will be > 0. Then wb_writeback() will
quit the background work (w/o this patch) while it's still over
background threshold. This will be a fairly normal/frequent case I guess.
Now that we do tagged sync and update inode->dirtied_when after the sync,
this change won't livelock sync(1). I actually tried to write 1 page
per 1ms with this command
write-and-fsync -n10000 -S 1000 -c 4096 /fs/test
and do sync(1) at the same time. The sync completes quickly on ext4,
xfs, btrfs.
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
2010-07-22 10:23:44 -06:00
|
|
|
* No more inodes for IO, bail
|
2009-09-23 11:32:26 -06:00
|
|
|
*/
|
2010-07-21 22:19:51 -06:00
|
|
|
if (list_empty(&wb->b_more_io))
|
2009-09-09 01:08:54 -06:00
|
|
|
break;
|
2009-09-23 11:32:26 -06:00
|
|
|
/*
|
|
|
|
* Nothing written. Wait for some inode to
|
|
|
|
* become available for writeback. Otherwise
|
|
|
|
* we'll just busyloop.
|
|
|
|
*/
|
|
|
|
if (!list_empty(&wb->b_more_io)) {
|
2011-05-04 19:54:37 -06:00
|
|
|
trace_writeback_wait(wb->bdi, work);
|
2010-10-20 18:49:30 -06:00
|
|
|
inode = wb_inode(wb->b_more_io.prev);
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_lock(&inode->i_lock);
|
2012-05-03 06:47:59 -06:00
|
|
|
spin_unlock(&wb->list_lock);
|
2012-05-03 06:48:03 -06:00
|
|
|
/* This function drops i_lock... */
|
|
|
|
inode_sleep_on_writeback(inode);
|
2012-05-03 06:47:59 -06:00
|
|
|
spin_lock(&wb->list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
}
|
2011-04-21 12:06:32 -06:00
|
|
|
spin_unlock(&wb->list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2011-05-04 19:54:37 -06:00
|
|
|
return nr_pages - work->nr_pages;
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2010-07-06 00:59:53 -06:00
|
|
|
* Return the next wb_writeback_work struct that hasn't been processed yet.
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
2010-07-06 00:59:53 -06:00
|
|
|
static struct wb_writeback_work *
|
2010-08-03 04:51:16 -06:00
|
|
|
get_next_work_item(struct backing_dev_info *bdi)
|
2009-09-09 01:08:54 -06:00
|
|
|
{
|
2010-07-06 00:59:53 -06:00
|
|
|
struct wb_writeback_work *work = NULL;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-07-25 05:29:22 -06:00
|
|
|
spin_lock_bh(&bdi->wb_lock);
|
2010-07-06 00:59:53 -06:00
|
|
|
if (!list_empty(&bdi->work_list)) {
|
|
|
|
work = list_entry(bdi->work_list.next,
|
|
|
|
struct wb_writeback_work, list);
|
|
|
|
list_del_init(&work->list);
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
2010-07-25 05:29:22 -06:00
|
|
|
spin_unlock_bh(&bdi->wb_lock);
|
2010-07-06 00:59:53 -06:00
|
|
|
return work;
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
|
2010-10-30 09:55:52 -06:00
|
|
|
/*
|
|
|
|
* Add in the number of potentially dirty inodes, because each inode
|
|
|
|
* write can dirty pagecache in the underlying blockdev.
|
|
|
|
*/
|
|
|
|
static unsigned long get_nr_dirty_pages(void)
|
|
|
|
{
|
|
|
|
return global_page_state(NR_FILE_DIRTY) +
|
|
|
|
global_page_state(NR_UNSTABLE_NFS) +
|
|
|
|
get_nr_dirty_inodes();
|
|
|
|
}
|
|
|
|
|
2011-01-13 16:45:44 -07:00
|
|
|
static long wb_check_background_flush(struct bdi_writeback *wb)
|
|
|
|
{
|
2010-11-18 13:38:33 -07:00
|
|
|
if (over_bground_thresh(wb->bdi)) {
|
2011-01-13 16:45:44 -07:00
|
|
|
|
|
|
|
struct wb_writeback_work work = {
|
|
|
|
.nr_pages = LONG_MAX,
|
|
|
|
.sync_mode = WB_SYNC_NONE,
|
|
|
|
.for_background = 1,
|
|
|
|
.range_cyclic = 1,
|
2011-10-07 21:54:10 -06:00
|
|
|
.reason = WB_REASON_BACKGROUND,
|
2011-01-13 16:45:44 -07:00
|
|
|
};
|
|
|
|
|
|
|
|
return wb_writeback(wb, &work);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
static long wb_check_old_data_flush(struct bdi_writeback *wb)
|
|
|
|
{
|
|
|
|
unsigned long expired;
|
|
|
|
long nr_pages;
|
|
|
|
|
2010-05-17 04:51:03 -06:00
|
|
|
/*
|
|
|
|
* When set to zero, disable periodic writeback
|
|
|
|
*/
|
|
|
|
if (!dirty_writeback_interval)
|
|
|
|
return 0;
|
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
expired = wb->last_old_flush +
|
|
|
|
msecs_to_jiffies(dirty_writeback_interval * 10);
|
|
|
|
if (time_before(jiffies, expired))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
wb->last_old_flush = jiffies;
|
2010-10-30 09:55:52 -06:00
|
|
|
nr_pages = get_nr_dirty_pages();
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2009-09-16 07:18:25 -06:00
|
|
|
if (nr_pages) {
|
2010-07-06 00:59:53 -06:00
|
|
|
struct wb_writeback_work work = {
|
2009-09-16 07:18:25 -06:00
|
|
|
.nr_pages = nr_pages,
|
|
|
|
.sync_mode = WB_SYNC_NONE,
|
|
|
|
.for_kupdate = 1,
|
|
|
|
.range_cyclic = 1,
|
2011-10-07 21:54:10 -06:00
|
|
|
.reason = WB_REASON_PERIODIC,
|
2009-09-16 07:18:25 -06:00
|
|
|
};
|
|
|
|
|
2010-07-06 00:59:53 -06:00
|
|
|
return wb_writeback(wb, &work);
|
2009-09-16 07:18:25 -06:00
|
|
|
}
|
2009-09-09 01:08:54 -06:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Retrieve work items and do the writeback they describe
|
|
|
|
*/
|
|
|
|
long wb_do_writeback(struct bdi_writeback *wb, int force_wait)
|
|
|
|
{
|
|
|
|
struct backing_dev_info *bdi = wb->bdi;
|
2010-07-06 00:59:53 -06:00
|
|
|
struct wb_writeback_work *work;
|
2009-09-16 07:18:25 -06:00
|
|
|
long wrote = 0;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-08-11 15:17:44 -06:00
|
|
|
set_bit(BDI_writeback_running, &wb->bdi->state);
|
2010-08-03 04:51:16 -06:00
|
|
|
while ((work = get_next_work_item(bdi)) != NULL) {
|
2009-09-09 01:08:54 -06:00
|
|
|
/*
|
|
|
|
* Override sync mode, in case we must wait for completion
|
2010-07-06 00:59:53 -06:00
|
|
|
* because this thread is exiting now.
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
|
|
|
if (force_wait)
|
2010-07-06 00:59:53 -06:00
|
|
|
work->sync_mode = WB_SYNC_ALL;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-07-06 21:24:06 -06:00
|
|
|
trace_writeback_exec(bdi, work);
|
|
|
|
|
2010-07-06 00:59:53 -06:00
|
|
|
wrote += wb_writeback(wb, work);
|
2009-09-09 01:08:54 -06:00
|
|
|
|
|
|
|
/*
|
2010-07-06 00:59:53 -06:00
|
|
|
* Notify the caller of completion if this is a synchronous
|
|
|
|
* work item, otherwise just free it.
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
2010-07-06 00:59:53 -06:00
|
|
|
if (work->done)
|
|
|
|
complete(work->done);
|
|
|
|
else
|
|
|
|
kfree(work);
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check for periodic writeback, kupdated() style
|
|
|
|
*/
|
|
|
|
wrote += wb_check_old_data_flush(wb);
|
2011-01-13 16:45:44 -07:00
|
|
|
wrote += wb_check_background_flush(wb);
|
2010-08-11 15:17:44 -06:00
|
|
|
clear_bit(BDI_writeback_running, &wb->bdi->state);
|
2009-09-09 01:08:54 -06:00
|
|
|
|
|
|
|
return wrote;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Handle writeback of dirty data for the device backed by this bdi. Also
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
* reschedules periodically and does kupdated style flushing.
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
void bdi_writeback_workfn(struct work_struct *work)
|
2009-09-09 01:08:54 -06:00
|
|
|
{
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
struct bdi_writeback *wb = container_of(to_delayed_work(work),
|
|
|
|
struct bdi_writeback, dwork);
|
2010-06-19 15:08:22 -06:00
|
|
|
struct backing_dev_info *bdi = wb->bdi;
|
2009-09-09 01:08:54 -06:00
|
|
|
long pages_written;
|
|
|
|
|
2013-04-30 16:27:24 -06:00
|
|
|
set_worker_desc("flush-%s", dev_name(bdi->dev));
|
2010-10-26 15:22:45 -06:00
|
|
|
current->flags |= PF_SWAPWRITE;
|
2010-07-06 21:24:06 -06:00
|
|
|
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
if (likely(!current_is_workqueue_rescuer() ||
|
|
|
|
list_empty(&bdi->bdi_list))) {
|
2010-07-25 05:29:22 -06:00
|
|
|
/*
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
* The normal path. Keep writing back @bdi until its
|
|
|
|
* work_list is empty. Note that this path is also taken
|
|
|
|
* if @bdi is shutting down even when we're running off the
|
|
|
|
* rescuer as work_list needs to be drained.
|
2010-07-25 05:29:22 -06:00
|
|
|
*/
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
do {
|
|
|
|
pages_written = wb_do_writeback(wb, 0);
|
|
|
|
trace_writeback_pages_written(pages_written);
|
|
|
|
} while (!list_empty(&bdi->work_list));
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* bdi_wq can't get enough workers and we're running off
|
|
|
|
* the emergency worker. Don't hog it. Hopefully, 1024 is
|
|
|
|
* enough for efficient IO.
|
|
|
|
*/
|
|
|
|
pages_written = writeback_inodes_wb(&bdi->wb, 1024,
|
|
|
|
WB_REASON_FORKER_THREAD);
|
2010-07-06 21:24:06 -06:00
|
|
|
trace_writeback_pages_written(pages_written);
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
if (!list_empty(&bdi->work_list) ||
|
|
|
|
(wb_has_dirty_io(wb) && dirty_writeback_interval))
|
|
|
|
queue_delayed_work(bdi_wq, &wb->dwork,
|
|
|
|
msecs_to_jiffies(dirty_writeback_interval * 10));
|
2010-07-06 21:24:06 -06:00
|
|
|
|
writeback: replace custom worker pool implementation with unbound workqueue
Writeback implements its own worker pool - each bdi can be associated
with a worker thread which is created and destroyed dynamically. The
worker thread for the default bdi is always present and serves as the
"forker" thread which forks off worker threads for other bdis.
there's no reason for writeback to implement its own worker pool when
using unbound workqueue instead is much simpler and more efficient.
This patch replaces custom worker pool implementation in writeback
with an unbound workqueue.
The conversion isn't too complicated but the followings are worth
mentioning.
* bdi_writeback->last_active, task and wakeup_timer are removed.
delayed_work ->dwork is added instead. Explicit timer handling is
no longer necessary. Everything works by either queueing / modding
/ flushing / canceling the delayed_work item.
* bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off
bdi_writeback->dwork. On each execution, it processes
bdi->work_list and reschedules itself if there are more things to
do.
The function also handles low-mem condition, which used to be
handled by the forker thread. If the function is running off a
rescuer thread, it only writes out limited number of pages so that
the rescuer can serve other bdis too. This preserves the flusher
creation failure behavior of the forker thread.
* INIT_LIST_HEAD(&bdi->bdi_list) is used to tell
bdi_writeback_workfn() about on-going bdi unregistration so that it
always drains work_list even if it's running off the rescuer. Note
that the original code was broken in this regard. Under memory
pressure, a bdi could finish unregistration with non-empty
work_list.
* The default bdi is no longer special. It now is treated the same as
any other bdi and bdi_cap_flush_forker() is removed.
* BDI_pending is no longer used. Removed.
* Some tracepoints become non-applicable. The following TPs are
removed - writeback_nothread, writeback_wake_thread,
writeback_wake_forker_thread, writeback_thread_start,
writeback_thread_stop.
Everything, including devices coming and going away and rescuer
operation under simulated memory pressure, seems to work fine in my
test setup.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
2013-04-01 20:08:06 -06:00
|
|
|
current->flags &= ~PF_SWAPWRITE;
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2010-06-08 10:15:07 -06:00
|
|
|
* Start writeback of `nr_pages' pages. If `nr_pages' is zero, write back
|
|
|
|
* the whole world.
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
2011-10-07 21:54:10 -06:00
|
|
|
void wakeup_flusher_threads(long nr_pages, enum wb_reason reason)
|
2009-09-09 01:08:54 -06:00
|
|
|
{
|
2010-06-08 10:15:07 -06:00
|
|
|
struct backing_dev_info *bdi;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-07-06 00:59:53 -06:00
|
|
|
if (!nr_pages) {
|
|
|
|
nr_pages = global_page_state(NR_FILE_DIRTY) +
|
2010-06-08 10:15:07 -06:00
|
|
|
global_page_state(NR_UNSTABLE_NFS);
|
|
|
|
}
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2010-06-08 10:15:07 -06:00
|
|
|
rcu_read_lock();
|
2009-09-14 05:12:40 -06:00
|
|
|
list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
|
2009-09-09 01:08:54 -06:00
|
|
|
if (!bdi_has_dirty_io(bdi))
|
|
|
|
continue;
|
2011-10-07 21:54:10 -06:00
|
|
|
__bdi_start_writeback(bdi, nr_pages, false, reason);
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
2009-09-14 05:12:40 -06:00
|
|
|
rcu_read_unlock();
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
static noinline void block_dump___mark_inode_dirty(struct inode *inode)
|
|
|
|
{
|
|
|
|
if (inode->i_ino || strcmp(inode->i_sb->s_id, "bdev")) {
|
|
|
|
struct dentry *dentry;
|
|
|
|
const char *name = "?";
|
|
|
|
|
|
|
|
dentry = d_find_alias(inode);
|
|
|
|
if (dentry) {
|
|
|
|
spin_lock(&dentry->d_lock);
|
|
|
|
name = (const char *) dentry->d_name.name;
|
|
|
|
}
|
|
|
|
printk(KERN_DEBUG
|
|
|
|
"%s(%d): dirtied inode %lu (%s) on %s\n",
|
|
|
|
current->comm, task_pid_nr(current), inode->i_ino,
|
|
|
|
name, inode->i_sb->s_id);
|
|
|
|
if (dentry) {
|
|
|
|
spin_unlock(&dentry->d_lock);
|
|
|
|
dput(dentry);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* __mark_inode_dirty - internal function
|
|
|
|
* @inode: inode to mark
|
|
|
|
* @flags: what kind of dirty (i.e. I_DIRTY_SYNC)
|
|
|
|
* Mark an inode as dirty. Callers should use mark_inode_dirty or
|
|
|
|
* mark_inode_dirty_sync.
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
2009-09-09 01:08:54 -06:00
|
|
|
* Put the inode on the super block's dirty list.
|
|
|
|
*
|
|
|
|
* CAREFUL! We mark it dirty unconditionally, but move it onto the
|
|
|
|
* dirty list only if it is hashed or if it refers to a blockdev.
|
|
|
|
* If it was not hashed, it will never be added to the dirty list
|
|
|
|
* even if it is later hashed, as it will have been marked dirty already.
|
|
|
|
*
|
|
|
|
* In short, make sure you hash any inodes _before_ you start marking
|
|
|
|
* them dirty.
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
2009-09-09 01:08:54 -06:00
|
|
|
* Note that for blockdevs, inode->dirtied_when represents the dirtying time of
|
|
|
|
* the block-special inode (/dev/hda1) itself. And the ->dirtied_when field of
|
|
|
|
* the kernel-internal blockdev inode represents the dirtying time of the
|
|
|
|
* blockdev's pages. This is why for I_DIRTY_PAGES we always use
|
|
|
|
* page->mapping->host, so the page-dirtying time is recorded in the internal
|
|
|
|
* blockdev inode.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2009-09-09 01:08:54 -06:00
|
|
|
void __mark_inode_dirty(struct inode *inode, int flags)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2009-09-09 01:08:54 -06:00
|
|
|
struct super_block *sb = inode->i_sb;
|
2010-07-25 05:29:21 -06:00
|
|
|
struct backing_dev_info *bdi = NULL;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
/*
|
|
|
|
* Don't do this for I_DIRTY_PAGES - that doesn't actually
|
|
|
|
* dirty the inode itself
|
|
|
|
*/
|
|
|
|
if (flags & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) {
|
2013-01-11 14:06:37 -07:00
|
|
|
trace_writeback_dirty_inode_start(inode, flags);
|
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
if (sb->s_op->dirty_inode)
|
2011-05-27 04:53:02 -06:00
|
|
|
sb->s_op->dirty_inode(inode, flags);
|
2013-01-11 14:06:37 -07:00
|
|
|
|
|
|
|
trace_writeback_dirty_inode(inode, flags);
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* make sure that changes are seen by all cpus before we test i_state
|
|
|
|
* -- mikulas
|
|
|
|
*/
|
|
|
|
smp_mb();
|
|
|
|
|
|
|
|
/* avoid the locking if we can */
|
|
|
|
if ((inode->i_state & flags) == flags)
|
|
|
|
return;
|
|
|
|
|
|
|
|
if (unlikely(block_dump))
|
|
|
|
block_dump___mark_inode_dirty(inode);
|
|
|
|
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_lock(&inode->i_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
if ((inode->i_state & flags) != flags) {
|
|
|
|
const int was_dirty = inode->i_state & I_DIRTY;
|
|
|
|
|
|
|
|
inode->i_state |= flags;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the inode is being synced, just update its dirty state.
|
|
|
|
* The unlocker will place the inode on the appropriate
|
|
|
|
* superblock list, based upon its state.
|
|
|
|
*/
|
|
|
|
if (inode->i_state & I_SYNC)
|
2011-03-22 05:23:36 -06:00
|
|
|
goto out_unlock_inode;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Only add valid (hashed) inodes to the superblock's
|
|
|
|
* dirty list. Add blockdev inodes as well.
|
|
|
|
*/
|
|
|
|
if (!S_ISBLK(inode->i_mode)) {
|
2010-10-23 13:19:20 -06:00
|
|
|
if (inode_unhashed(inode))
|
2011-03-22 05:23:36 -06:00
|
|
|
goto out_unlock_inode;
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
2010-06-02 15:38:30 -06:00
|
|
|
if (inode->i_state & I_FREEING)
|
2011-03-22 05:23:36 -06:00
|
|
|
goto out_unlock_inode;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If the inode was already on b_dirty/b_io/b_more_io, don't
|
|
|
|
* reposition it (that would break b_dirty time-ordering).
|
|
|
|
*/
|
|
|
|
if (!was_dirty) {
|
2011-03-22 05:23:41 -06:00
|
|
|
bool wakeup_bdi = false;
|
2010-07-25 05:29:21 -06:00
|
|
|
bdi = inode_to_bdi(inode);
|
|
|
|
|
|
|
|
if (bdi_cap_writeback_dirty(bdi)) {
|
|
|
|
WARN(!test_bit(BDI_registered, &bdi->state),
|
|
|
|
"bdi-%s not registered\n", bdi->name);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If this is the first dirty inode for this
|
|
|
|
* bdi, we have to wake-up the corresponding
|
|
|
|
* bdi thread to make sure background
|
|
|
|
* write-back happens later.
|
|
|
|
*/
|
|
|
|
if (!wb_has_dirty_io(&bdi->wb))
|
|
|
|
wakeup_bdi = true;
|
2009-09-09 01:10:25 -06:00
|
|
|
}
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2011-03-22 05:23:41 -06:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-04-21 18:19:44 -06:00
|
|
|
spin_lock(&bdi->wb.list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
inode->dirtied_when = jiffies;
|
2010-10-20 18:49:30 -06:00
|
|
|
list_move(&inode->i_wb_list, &bdi->wb.b_dirty);
|
2011-04-21 18:19:44 -06:00
|
|
|
spin_unlock(&bdi->wb.list_lock);
|
2011-03-22 05:23:41 -06:00
|
|
|
|
|
|
|
if (wakeup_bdi)
|
|
|
|
bdi_wakeup_thread_delayed(bdi);
|
|
|
|
return;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
}
|
2011-03-22 05:23:36 -06:00
|
|
|
out_unlock_inode:
|
|
|
|
spin_unlock(&inode->i_lock);
|
2010-07-25 05:29:21 -06:00
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(__mark_inode_dirty);
|
|
|
|
|
2009-09-16 07:13:54 -06:00
|
|
|
static void wait_sb_inodes(struct super_block *sb)
|
2009-09-09 01:08:54 -06:00
|
|
|
{
|
|
|
|
struct inode *inode, *old_inode = NULL;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We need to be protected against the filesystem going from
|
|
|
|
* r/o to r/w or vice versa.
|
|
|
|
*/
|
2009-09-16 07:13:54 -06:00
|
|
|
WARN_ON(!rwsem_is_locked(&sb->s_umount));
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2011-03-22 05:23:40 -06:00
|
|
|
spin_lock(&inode_sb_list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Data integrity sync. Must wait for all pages under writeback,
|
|
|
|
* because there may have been pages dirtied before our sync
|
|
|
|
* call, but which had writeout started before we write it out.
|
|
|
|
* In which case, the inode may not be on the dirty list, but
|
|
|
|
* we still have to wait for that writeout.
|
|
|
|
*/
|
2009-09-16 07:13:54 -06:00
|
|
|
list_for_each_entry(inode, &sb->s_inodes, i_sb_list) {
|
2011-03-22 05:23:36 -06:00
|
|
|
struct address_space *mapping = inode->i_mapping;
|
2009-09-09 01:08:54 -06:00
|
|
|
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_lock(&inode->i_lock);
|
|
|
|
if ((inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW)) ||
|
|
|
|
(mapping->nrpages == 0)) {
|
|
|
|
spin_unlock(&inode->i_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
continue;
|
2011-03-22 05:23:36 -06:00
|
|
|
}
|
2009-09-09 01:08:54 -06:00
|
|
|
__iget(inode);
|
2011-03-22 05:23:36 -06:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-03-22 05:23:40 -06:00
|
|
|
spin_unlock(&inode_sb_list_lock);
|
|
|
|
|
2009-09-09 01:08:54 -06:00
|
|
|
/*
|
2011-03-22 05:23:40 -06:00
|
|
|
* We hold a reference to 'inode' so it couldn't have been
|
|
|
|
* removed from s_inodes list while we dropped the
|
|
|
|
* inode_sb_list_lock. We cannot iput the inode now as we can
|
|
|
|
* be holding the last reference and we cannot iput it under
|
|
|
|
* inode_sb_list_lock. So we keep the reference and iput it
|
|
|
|
* later.
|
2009-09-09 01:08:54 -06:00
|
|
|
*/
|
|
|
|
iput(old_inode);
|
|
|
|
old_inode = inode;
|
|
|
|
|
|
|
|
filemap_fdatawait(mapping);
|
|
|
|
|
|
|
|
cond_resched();
|
|
|
|
|
2011-03-22 05:23:40 -06:00
|
|
|
spin_lock(&inode_sb_list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
}
|
2011-03-22 05:23:40 -06:00
|
|
|
spin_unlock(&inode_sb_list_lock);
|
2009-09-09 01:08:54 -06:00
|
|
|
iput(old_inode);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2009-09-02 04:34:32 -06:00
|
|
|
/**
|
2010-10-29 09:16:17 -06:00
|
|
|
* writeback_inodes_sb_nr - writeback dirty inodes from given super_block
|
2009-09-02 04:34:32 -06:00
|
|
|
* @sb: the superblock
|
2010-10-29 09:16:17 -06:00
|
|
|
* @nr: the number of pages to write
|
2011-11-23 05:56:45 -07:00
|
|
|
* @reason: reason why some writeback work initiated
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
2009-09-02 04:34:32 -06:00
|
|
|
* Start writeback on some inodes on this super_block. No guarantees are made
|
|
|
|
* on how many (if any) will be written, and this function does not wait
|
2010-10-29 09:16:17 -06:00
|
|
|
* for IO completion of submitted IO.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2011-10-07 21:54:10 -06:00
|
|
|
void writeback_inodes_sb_nr(struct super_block *sb,
|
|
|
|
unsigned long nr,
|
|
|
|
enum wb_reason reason)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2010-07-06 00:59:53 -06:00
|
|
|
DECLARE_COMPLETION_ONSTACK(done);
|
|
|
|
struct wb_writeback_work work = {
|
2010-06-06 10:38:15 -06:00
|
|
|
.sb = sb,
|
|
|
|
.sync_mode = WB_SYNC_NONE,
|
|
|
|
.tagged_writepages = 1,
|
|
|
|
.done = &done,
|
|
|
|
.nr_pages = nr,
|
2011-10-07 21:54:10 -06:00
|
|
|
.reason = reason,
|
2010-06-08 10:14:43 -06:00
|
|
|
};
|
2009-09-02 04:34:32 -06:00
|
|
|
|
2012-07-03 08:45:27 -06:00
|
|
|
if (sb->s_bdi == &noop_backing_dev_info)
|
|
|
|
return;
|
2010-06-08 10:14:51 -06:00
|
|
|
WARN_ON(!rwsem_is_locked(&sb->s_umount));
|
2010-07-06 00:59:53 -06:00
|
|
|
bdi_queue_work(sb->s_bdi, &work);
|
|
|
|
wait_for_completion(&done);
|
2010-05-17 04:55:07 -06:00
|
|
|
}
|
2010-10-29 09:16:17 -06:00
|
|
|
EXPORT_SYMBOL(writeback_inodes_sb_nr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* writeback_inodes_sb - writeback dirty inodes from given super_block
|
|
|
|
* @sb: the superblock
|
2011-11-23 05:56:45 -07:00
|
|
|
* @reason: reason why some writeback work was initiated
|
2010-10-29 09:16:17 -06:00
|
|
|
*
|
|
|
|
* Start writeback on some inodes on this super_block. No guarantees are made
|
|
|
|
* on how many (if any) will be written, and this function does not wait
|
|
|
|
* for IO completion of submitted IO.
|
|
|
|
*/
|
2011-10-07 21:54:10 -06:00
|
|
|
void writeback_inodes_sb(struct super_block *sb, enum wb_reason reason)
|
2010-10-29 09:16:17 -06:00
|
|
|
{
|
2011-10-07 21:54:10 -06:00
|
|
|
return writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason);
|
2010-10-29 09:16:17 -06:00
|
|
|
}
|
2010-06-01 03:08:43 -06:00
|
|
|
EXPORT_SYMBOL(writeback_inodes_sb);
|
2010-05-17 04:55:07 -06:00
|
|
|
|
2009-12-23 05:57:07 -07:00
|
|
|
/**
|
2013-01-09 22:47:57 -07:00
|
|
|
* try_to_writeback_inodes_sb_nr - try to start writeback if none underway
|
2009-12-23 05:57:07 -07:00
|
|
|
* @sb: the superblock
|
2013-01-09 22:47:57 -07:00
|
|
|
* @nr: the number of pages to write
|
|
|
|
* @reason: the reason of writeback
|
2009-12-23 05:57:07 -07:00
|
|
|
*
|
2013-01-09 22:47:57 -07:00
|
|
|
* Invoke writeback_inodes_sb_nr if no writeback is currently underway.
|
2009-12-23 05:57:07 -07:00
|
|
|
* Returns 1 if writeback was started, 0 if not.
|
|
|
|
*/
|
2013-01-09 22:47:57 -07:00
|
|
|
int try_to_writeback_inodes_sb_nr(struct super_block *sb,
|
|
|
|
unsigned long nr,
|
|
|
|
enum wb_reason reason)
|
2009-12-23 05:57:07 -07:00
|
|
|
{
|
2013-01-09 22:47:57 -07:00
|
|
|
if (writeback_in_progress(sb->s_bdi))
|
2009-12-23 05:57:07 -07:00
|
|
|
return 1;
|
2013-01-09 22:47:57 -07:00
|
|
|
|
|
|
|
if (!down_read_trylock(&sb->s_umount))
|
2009-12-23 05:57:07 -07:00
|
|
|
return 0;
|
2013-01-09 22:47:57 -07:00
|
|
|
|
|
|
|
writeback_inodes_sb_nr(sb, nr, reason);
|
|
|
|
up_read(&sb->s_umount);
|
|
|
|
return 1;
|
2009-12-23 05:57:07 -07:00
|
|
|
}
|
2013-01-09 22:47:57 -07:00
|
|
|
EXPORT_SYMBOL(try_to_writeback_inodes_sb_nr);
|
2009-12-23 05:57:07 -07:00
|
|
|
|
2010-10-29 09:16:17 -06:00
|
|
|
/**
|
2013-01-09 22:47:57 -07:00
|
|
|
* try_to_writeback_inodes_sb - try to start writeback if none underway
|
2010-10-29 09:16:17 -06:00
|
|
|
* @sb: the superblock
|
2011-11-23 05:56:45 -07:00
|
|
|
* @reason: reason why some writeback work was initiated
|
2010-10-29 09:16:17 -06:00
|
|
|
*
|
2013-01-09 22:47:57 -07:00
|
|
|
* Implement by try_to_writeback_inodes_sb_nr()
|
2010-10-29 09:16:17 -06:00
|
|
|
* Returns 1 if writeback was started, 0 if not.
|
|
|
|
*/
|
2013-01-09 22:47:57 -07:00
|
|
|
int try_to_writeback_inodes_sb(struct super_block *sb, enum wb_reason reason)
|
2010-10-29 09:16:17 -06:00
|
|
|
{
|
2013-01-09 22:47:57 -07:00
|
|
|
return try_to_writeback_inodes_sb_nr(sb, get_nr_dirty_pages(), reason);
|
2010-10-29 09:16:17 -06:00
|
|
|
}
|
2013-01-09 22:47:57 -07:00
|
|
|
EXPORT_SYMBOL(try_to_writeback_inodes_sb);
|
2010-10-29 09:16:17 -06:00
|
|
|
|
2009-09-02 04:34:32 -06:00
|
|
|
/**
|
|
|
|
* sync_inodes_sb - sync sb inode pages
|
|
|
|
* @sb: the superblock
|
|
|
|
*
|
|
|
|
* This function writes and waits on any dirty inode belonging to this
|
2011-01-13 16:47:26 -07:00
|
|
|
* super_block.
|
2009-09-02 04:34:32 -06:00
|
|
|
*/
|
2009-09-16 07:13:54 -06:00
|
|
|
void sync_inodes_sb(struct super_block *sb)
|
2009-09-02 04:34:32 -06:00
|
|
|
{
|
2010-07-06 00:59:53 -06:00
|
|
|
DECLARE_COMPLETION_ONSTACK(done);
|
|
|
|
struct wb_writeback_work work = {
|
2010-06-08 10:14:43 -06:00
|
|
|
.sb = sb,
|
|
|
|
.sync_mode = WB_SYNC_ALL,
|
|
|
|
.nr_pages = LONG_MAX,
|
|
|
|
.range_cyclic = 0,
|
2010-07-06 00:59:53 -06:00
|
|
|
.done = &done,
|
2011-10-07 21:54:10 -06:00
|
|
|
.reason = WB_REASON_SYNC,
|
2013-07-02 06:38:35 -06:00
|
|
|
.for_sync = 1,
|
2010-06-08 10:14:43 -06:00
|
|
|
};
|
|
|
|
|
2012-07-03 08:45:27 -06:00
|
|
|
/* Nothing to do? */
|
|
|
|
if (sb->s_bdi == &noop_backing_dev_info)
|
|
|
|
return;
|
2010-06-08 10:14:51 -06:00
|
|
|
WARN_ON(!rwsem_is_locked(&sb->s_umount));
|
|
|
|
|
2010-07-06 00:59:53 -06:00
|
|
|
bdi_queue_work(sb->s_bdi, &work);
|
|
|
|
wait_for_completion(&done);
|
|
|
|
|
2009-09-16 07:13:54 -06:00
|
|
|
wait_sb_inodes(sb);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2009-09-02 04:34:32 -06:00
|
|
|
EXPORT_SYMBOL(sync_inodes_sb);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/**
|
[PATCH] fix nr_unused accounting, and avoid recursing in iput with I_WILL_FREE set
list_move(&inode->i_list, &inode_in_use);
} else {
list_move(&inode->i_list, &inode_unused);
+ inodes_stat.nr_unused++;
}
}
wake_up_inode(inode);
Are you sure the above diff is correct? It was added somewhere between
2.6.5 and 2.6.8. I think it's wrong.
The only way I can imagine the i_count to be zero in the above path, is
that I_WILL_FREE is set. And if I_WILL_FREE is set, then we must not
increase nr_unused. So I believe the above change is buggy and it will
definitely overstate the number of unused inodes and it should be backed
out.
Note that __writeback_single_inode before calling __sync_single_inode, can
drop the spinlock and we can have both the dirty and locked bitflags clear
here:
spin_unlock(&inode_lock);
__wait_on_inode(inode);
iput(inode);
XXXXXXX
spin_lock(&inode_lock);
}
use inode again here
a construct like the above makes zero sense from a reference counting
standpoint.
Either we don't ever use the inode again after the iput, or the
inode_lock should be taken _before_ executing the iput (i.e. a __iput
would be required). Taking the inode_lock after iput means the iget was
useless if we keep using the inode after the iput.
So the only chance the 2.6 was safe to call __writeback_single_inode
with the i_count == 0, is that I_WILL_FREE is set (I_WILL_FREE will
prevent the VM to free the inode in XXXXX).
Potentially calling the above iput with I_WILL_FREE was also wrong
because it would recurse in iput_final (the second mainline bug).
The below (untested) patch fixes the nr_unused accounting, avoids recursing
in iput when I_WILL_FREE is set and makes sure (with the BUG_ON) that we
don't corrupt memory and that all holders that don't set I_WILL_FREE, keeps
a reference on the inode!
Signed-off-by: Andrea Arcangeli <andrea@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-30 16:03:05 -07:00
|
|
|
* write_inode_now - write an inode to disk
|
|
|
|
* @inode: inode to write to disk
|
|
|
|
* @sync: whether the write should be synchronous or not
|
|
|
|
*
|
|
|
|
* This function commits an inode to disk immediately if it is dirty. This is
|
|
|
|
* primarily needed by knfsd.
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
[PATCH] fix nr_unused accounting, and avoid recursing in iput with I_WILL_FREE set
list_move(&inode->i_list, &inode_in_use);
} else {
list_move(&inode->i_list, &inode_unused);
+ inodes_stat.nr_unused++;
}
}
wake_up_inode(inode);
Are you sure the above diff is correct? It was added somewhere between
2.6.5 and 2.6.8. I think it's wrong.
The only way I can imagine the i_count to be zero in the above path, is
that I_WILL_FREE is set. And if I_WILL_FREE is set, then we must not
increase nr_unused. So I believe the above change is buggy and it will
definitely overstate the number of unused inodes and it should be backed
out.
Note that __writeback_single_inode before calling __sync_single_inode, can
drop the spinlock and we can have both the dirty and locked bitflags clear
here:
spin_unlock(&inode_lock);
__wait_on_inode(inode);
iput(inode);
XXXXXXX
spin_lock(&inode_lock);
}
use inode again here
a construct like the above makes zero sense from a reference counting
standpoint.
Either we don't ever use the inode again after the iput, or the
inode_lock should be taken _before_ executing the iput (i.e. a __iput
would be required). Taking the inode_lock after iput means the iget was
useless if we keep using the inode after the iput.
So the only chance the 2.6 was safe to call __writeback_single_inode
with the i_count == 0, is that I_WILL_FREE is set (I_WILL_FREE will
prevent the VM to free the inode in XXXXX).
Potentially calling the above iput with I_WILL_FREE was also wrong
because it would recurse in iput_final (the second mainline bug).
The below (untested) patch fixes the nr_unused accounting, avoids recursing
in iput when I_WILL_FREE is set and makes sure (with the BUG_ON) that we
don't corrupt memory and that all holders that don't set I_WILL_FREE, keeps
a reference on the inode!
Signed-off-by: Andrea Arcangeli <andrea@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-30 16:03:05 -07:00
|
|
|
* The caller must either have a ref on the inode or must have set I_WILL_FREE.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
int write_inode_now(struct inode *inode, int sync)
|
|
|
|
{
|
2011-04-21 18:19:44 -06:00
|
|
|
struct bdi_writeback *wb = &inode_to_bdi(inode)->wb;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct writeback_control wbc = {
|
|
|
|
.nr_to_write = LONG_MAX,
|
2008-02-08 05:20:23 -07:00
|
|
|
.sync_mode = sync ? WB_SYNC_ALL : WB_SYNC_NONE,
|
[PATCH] writeback: fix range handling
When a writeback_control's `start' and `end' fields are used to
indicate a one-byte-range starting at file offset zero, the required
values of .start=0,.end=0 mean that the ->writepages() implementation
has no way of telling that it is being asked to perform a range
request. Because we're currently overloading (start == 0 && end == 0)
to mean "this is not a write-a-range request".
To make all this sane, the patch changes range of writeback_control.
So caller does: If it is calling ->writepages() to write pages, it
sets range (range_start/end or range_cyclic) always.
And if range_cyclic is true, ->writepages() thinks the range is
cyclic, otherwise it just uses range_start and range_end.
This patch does,
- Add LLONG_MAX, LLONG_MIN, ULLONG_MAX to include/linux/kernel.h
-1 is usually ok for range_end (type is long long). But, if someone did,
range_end += val; range_end is "val - 1"
u64val = range_end >> bits; u64val is "~(0ULL)"
or something, they are wrong. So, this adds LLONG_MAX to avoid nasty
things, and uses LLONG_MAX for range_end.
- All callers of ->writepages() sets range_start/end or range_cyclic.
- Fix updates of ->writeback_index. It seems already bit strange.
If it starts at 0 and ended by check of nr_to_write, this last
index may reduce chance to scan end of file. So, this updates
->writeback_index only if range_cyclic is true or whole-file is
scanned.
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Cc: Nathan Scott <nathans@sgi.com>
Cc: Anton Altaparmakov <aia21@cantab.net>
Cc: Steven French <sfrench@us.ibm.com>
Cc: "Vladimir V. Saveliev" <vs@namesys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 03:03:26 -06:00
|
|
|
.range_start = 0,
|
|
|
|
.range_end = LLONG_MAX,
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
if (!mapping_cap_writeback_dirty(inode->i_mapping))
|
2005-11-07 01:59:15 -07:00
|
|
|
wbc.nr_to_write = 0;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
might_sleep();
|
2012-05-03 06:48:00 -06:00
|
|
|
return writeback_single_inode(inode, wb, &wbc);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(write_inode_now);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* sync_inode - write an inode and its pages to disk.
|
|
|
|
* @inode: the inode to sync
|
|
|
|
* @wbc: controls the writeback mode
|
|
|
|
*
|
|
|
|
* sync_inode() will write an inode and its pages to disk. It will also
|
|
|
|
* correctly update the inode on its superblock's dirty inode lists and will
|
|
|
|
* update inode->i_state.
|
|
|
|
*
|
|
|
|
* The caller must have a ref on the inode.
|
|
|
|
*/
|
|
|
|
int sync_inode(struct inode *inode, struct writeback_control *wbc)
|
|
|
|
{
|
2012-05-03 06:48:00 -06:00
|
|
|
return writeback_single_inode(inode, &inode_to_bdi(inode)->wb, wbc);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(sync_inode);
|
2010-10-06 02:48:20 -06:00
|
|
|
|
|
|
|
/**
|
2011-01-13 16:45:48 -07:00
|
|
|
* sync_inode_metadata - write an inode to disk
|
2010-10-06 02:48:20 -06:00
|
|
|
* @inode: the inode to sync
|
|
|
|
* @wait: wait for I/O to complete.
|
|
|
|
*
|
2011-01-13 16:45:48 -07:00
|
|
|
* Write an inode to disk and adjust its dirty state after completion.
|
2010-10-06 02:48:20 -06:00
|
|
|
*
|
|
|
|
* Note: only writes the actual inode, no associated data or other metadata.
|
|
|
|
*/
|
|
|
|
int sync_inode_metadata(struct inode *inode, int wait)
|
|
|
|
{
|
|
|
|
struct writeback_control wbc = {
|
|
|
|
.sync_mode = wait ? WB_SYNC_ALL : WB_SYNC_NONE,
|
|
|
|
.nr_to_write = 0, /* metadata-only */
|
|
|
|
};
|
|
|
|
|
|
|
|
return sync_inode(inode, &wbc);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(sync_inode_metadata);
|