2005-04-17 05:20:36 +07:00
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*
|
2008-10-16 12:01:59 +07:00
|
|
|
* 10Apr2002 Andrew Morton
|
2005-04-17 05:20:36 +07:00
|
|
|
* Split out of fs/inode.c
|
|
|
|
* Additions for address_space-based writeback
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/kernel.h>
|
2007-09-21 14:19:54 +07:00
|
|
|
#include <linux/module.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#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 15:04:11 +07:00
|
|
|
#include <linux/slab.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/fs.h>
|
|
|
|
#include <linux/mm.h>
|
2009-09-09 14:08:54 +07:00
|
|
|
#include <linux/kthread.h>
|
|
|
|
#include <linux/freezer.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include <linux/writeback.h>
|
|
|
|
#include <linux/blkdev.h>
|
|
|
|
#include <linux/backing-dev.h>
|
|
|
|
#include <linux/buffer_head.h>
|
2010-07-07 10:24:06 +07:00
|
|
|
#include <linux/tracepoint.h>
|
2006-10-01 01:52:18 +07:00
|
|
|
#include "internal.h"
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-09-16 20:18:25 +07:00
|
|
|
/*
|
|
|
|
* Passed into wb_writeback(), essentially a subset of writeback_control
|
|
|
|
*/
|
2010-07-06 13:59:53 +07:00
|
|
|
struct wb_writeback_work {
|
2009-09-16 20:18:25 +07:00
|
|
|
long nr_pages;
|
|
|
|
struct super_block *sb;
|
|
|
|
enum writeback_sync_modes sync_mode;
|
2010-04-02 08:36:30 +07:00
|
|
|
unsigned int for_kupdate:1;
|
|
|
|
unsigned int range_cyclic:1;
|
|
|
|
unsigned int for_background:1;
|
2009-09-16 20:18:25 +07:00
|
|
|
|
2009-09-16 01:04:57 +07:00
|
|
|
struct list_head list; /* pending work list */
|
2010-07-06 13:59:53 +07:00
|
|
|
struct completion *done; /* set if the caller waits */
|
2009-09-09 14:08:54 +07:00
|
|
|
};
|
|
|
|
|
2010-07-07 10:24:06 +07:00
|
|
|
/*
|
|
|
|
* Include the creation of the trace points after defining the
|
|
|
|
* wb_writeback_work structure so that the definition remains local to this
|
|
|
|
* file.
|
|
|
|
*/
|
|
|
|
#define CREATE_TRACE_POINTS
|
|
|
|
#include <trace/events/writeback.h>
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We don't actually have pdflush, but this one is exported though /proc...
|
|
|
|
*/
|
|
|
|
int nr_pdflush_threads;
|
|
|
|
|
2008-04-29 14:58:56 +07:00
|
|
|
/**
|
|
|
|
* writeback_in_progress - determine whether there is writeback in progress
|
|
|
|
* @bdi: the device's backing_dev_info structure.
|
|
|
|
*
|
2009-09-09 14:08:54 +07:00
|
|
|
* Determine whether there is writeback waiting to be handled against a
|
|
|
|
* backing device.
|
2008-04-29 14:58:56 +07:00
|
|
|
*/
|
|
|
|
int writeback_in_progress(struct backing_dev_info *bdi)
|
|
|
|
{
|
2010-08-12 04:17:44 +07:00
|
|
|
return test_bit(BDI_writeback_running, &bdi->state);
|
2008-04-29 14:58:56 +07:00
|
|
|
}
|
|
|
|
|
2010-09-21 16:51:01 +07:00
|
|
|
static inline struct backing_dev_info *inode_to_bdi(struct inode *inode)
|
|
|
|
{
|
|
|
|
struct super_block *sb = inode->i_sb;
|
|
|
|
|
2010-10-04 19:25:33 +07:00
|
|
|
if (strcmp(sb->s_type->name, "bdev") == 0)
|
|
|
|
return inode->i_mapping->backing_dev_info;
|
|
|
|
|
|
|
|
return sb->s_bdi;
|
2010-09-21 16:51:01 +07:00
|
|
|
}
|
|
|
|
|
2010-10-21 07:49:30 +07:00
|
|
|
static inline struct inode *wb_inode(struct list_head *head)
|
|
|
|
{
|
|
|
|
return list_entry(head, struct inode, i_wb_list);
|
|
|
|
}
|
|
|
|
|
2011-01-14 06:45:44 +07:00
|
|
|
/* Wakeup flusher thread or forker thread to fork it. Requires bdi->wb_lock. */
|
|
|
|
static void bdi_wakeup_flusher(struct backing_dev_info *bdi)
|
2009-09-09 14:08:54 +07:00
|
|
|
{
|
writeback: move bdi threads exiting logic to the forker thread
Currently, bdi threads can decide to exit if there were no useful activities
for 5 minutes. However, this causes nasty races: we can easily oops in the
'bdi_queue_work()' if the bdi thread decides to exit while we are waking it up.
And even if we do not oops, but the bdi tread exits immediately after we wake
it up, we'd lose the wake-up event and have an unnecessary delay (up to 5 secs)
in the bdi work processing.
This patch makes the forker thread to be the central place which not only
creates bdi threads, but also kills them if they were inactive long enough.
This better design-wise.
Another reason why this change was done is to prepare for the further changes
which will prevent the bdi threads from waking up every 5 sec and wasting
power. Indeed, when the task does not wake up periodically anymore, it won't be
able to exit either.
This patch also moves the the 'wake_up_bit()' call from the bdi thread to the
forker thread as well. So now the forker thread sets the BDI_pending bit, then
forks the task or kills it, then clears the bit and wakes up the waiting
process.
The only process which may wain on the bit is 'bdi_wb_shutdown()'. This
function was changed as well - now it first removes the bdi from the
'bdi_list', then waits on the 'BDI_pending' bit. Once it wakes up, it is
guaranteed that the forker thread won't race with it, because the bdi is not
visible. Note, the forker thread sets the 'BDI_pending' bit under the
'bdi->wb_lock' which is essential for proper serialization.
And additionally, when we change 'bdi->wb.task', we now take the
'bdi->work_lock', to make sure that we do not lose wake-ups which we otherwise
would when raced with, say, 'bdi_queue_work()'.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2010-07-25 18:29:20 +07:00
|
|
|
if (bdi->wb.task) {
|
|
|
|
wake_up_process(bdi->wb.task);
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* The bdi thread isn't there, wake up the forker thread which
|
|
|
|
* will create and run it.
|
|
|
|
*/
|
2009-09-09 14:08:54 +07:00
|
|
|
wake_up_process(default_backing_dev_info.wb.task);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2011-01-14 06:45:44 +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);
|
|
|
|
if (!bdi->wb.task)
|
|
|
|
trace_writeback_nothread(bdi, work);
|
|
|
|
bdi_wakeup_flusher(bdi);
|
2010-07-25 18:29:22 +07:00
|
|
|
spin_unlock_bh(&bdi->wb_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2010-07-06 13:59:53 +07:00
|
|
|
static void
|
|
|
|
__bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
|
2011-01-14 06:45:44 +07:00
|
|
|
bool range_cyclic)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-07-06 13:59:53 +07:00
|
|
|
struct wb_writeback_work *work;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2009-09-14 01:07:36 +07:00
|
|
|
/*
|
|
|
|
* This is WB_SYNC_NONE writeback, so if allocation fails just
|
|
|
|
* wakeup the thread for old dirty data writeback
|
|
|
|
*/
|
2010-07-06 13:59:53 +07:00
|
|
|
work = kzalloc(sizeof(*work), GFP_ATOMIC);
|
|
|
|
if (!work) {
|
2010-07-07 10:24:06 +07:00
|
|
|
if (bdi->wb.task) {
|
|
|
|
trace_writeback_nowork(bdi);
|
2010-07-06 13:59:53 +07:00
|
|
|
wake_up_process(bdi->wb.task);
|
2010-07-07 10:24:06 +07:00
|
|
|
}
|
2010-07-06 13:59:53 +07:00
|
|
|
return;
|
2009-09-14 01:07:36 +07:00
|
|
|
}
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-07-06 13:59:53 +07:00
|
|
|
work->sync_mode = WB_SYNC_NONE;
|
|
|
|
work->nr_pages = nr_pages;
|
|
|
|
work->range_cyclic = range_cyclic;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-07-06 13:59:53 +07:00
|
|
|
bdi_queue_work(bdi, work);
|
2009-09-16 20:13:54 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* bdi_start_writeback - start writeback
|
|
|
|
* @bdi: the backing device to write from
|
|
|
|
* @nr_pages: the number of pages to write
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* This does WB_SYNC_NONE opportunistic writeback. The IO is only
|
2011-03-31 08:57:33 +07:00
|
|
|
* started when this function returns, we make no guarantees on
|
2010-06-01 16:08:43 +07:00
|
|
|
* completion. Caller need not hold sb s_umount semaphore.
|
2009-09-16 20:13:54 +07:00
|
|
|
*
|
|
|
|
*/
|
2010-06-08 23:15:15 +07:00
|
|
|
void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages)
|
2009-09-16 20:13:54 +07:00
|
|
|
{
|
2011-01-14 06:45:44 +07:00
|
|
|
__bdi_start_writeback(bdi, nr_pages, true);
|
2010-06-08 23:15:15 +07:00
|
|
|
}
|
2009-09-23 19:33:40 +07:00
|
|
|
|
2010-06-08 23:15:15 +07:00
|
|
|
/**
|
|
|
|
* bdi_start_background_writeback - start background writeback
|
|
|
|
* @bdi: the backing device to write from
|
|
|
|
*
|
|
|
|
* Description:
|
2011-01-14 06:45:44 +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 23:15:15 +07:00
|
|
|
*/
|
|
|
|
void bdi_start_background_writeback(struct backing_dev_info *bdi)
|
|
|
|
{
|
2011-01-14 06:45:44 +07:00
|
|
|
/*
|
|
|
|
* We just wake up the flusher thread. It will perform background
|
|
|
|
* writeback as soon as there is no other work to do.
|
|
|
|
*/
|
2011-01-14 06:45:46 +07:00
|
|
|
trace_writeback_wake_background(bdi);
|
2011-01-14 06:45:44 +07:00
|
|
|
spin_lock_bh(&bdi->wb_lock);
|
|
|
|
bdi_wakeup_flusher(bdi);
|
|
|
|
spin_unlock_bh(&bdi->wb_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2011-03-22 18:23:41 +07:00
|
|
|
/*
|
|
|
|
* Remove the inode from the writeback list it is on.
|
|
|
|
*/
|
|
|
|
void inode_wb_list_del(struct inode *inode)
|
|
|
|
{
|
|
|
|
spin_lock(&inode_wb_list_lock);
|
|
|
|
list_del_init(&inode->i_wb_list);
|
|
|
|
spin_unlock(&inode_wb_list_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-10-17 13:30:32 +07:00
|
|
|
/*
|
|
|
|
* 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 14:19:46 +07:00
|
|
|
* already the most-recently-dirtied inode on the b_dirty list. If that is
|
2007-10-17 13:30:32 +07:00
|
|
|
* the case then the inode must have been redirtied while it was being written
|
|
|
|
* out and we don't reset its dirtied_when.
|
|
|
|
*/
|
|
|
|
static void redirty_tail(struct inode *inode)
|
|
|
|
{
|
2009-09-09 14:08:54 +07:00
|
|
|
struct bdi_writeback *wb = &inode_to_bdi(inode)->wb;
|
2007-10-17 13:30:32 +07:00
|
|
|
|
2011-03-22 18:23:41 +07:00
|
|
|
assert_spin_locked(&inode_wb_list_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
if (!list_empty(&wb->b_dirty)) {
|
2009-09-02 14:19:46 +07:00
|
|
|
struct inode *tail;
|
2007-10-17 13:30:32 +07:00
|
|
|
|
2010-10-21 07:49:30 +07:00
|
|
|
tail = wb_inode(wb->b_dirty.next);
|
2009-09-02 14:19:46 +07:00
|
|
|
if (time_before(inode->dirtied_when, tail->dirtied_when))
|
2007-10-17 13:30:32 +07:00
|
|
|
inode->dirtied_when = jiffies;
|
|
|
|
}
|
2010-10-21 07:49:30 +07:00
|
|
|
list_move(&inode->i_wb_list, &wb->b_dirty);
|
2007-10-17 13:30:32 +07:00
|
|
|
}
|
|
|
|
|
2007-10-17 13:30:34 +07:00
|
|
|
/*
|
2009-09-02 14:19:46 +07:00
|
|
|
* requeue inode for re-scanning after bdi->b_io list is exhausted.
|
2007-10-17 13:30:34 +07:00
|
|
|
*/
|
2007-10-17 13:30:38 +07:00
|
|
|
static void requeue_io(struct inode *inode)
|
2007-10-17 13:30:34 +07:00
|
|
|
{
|
2009-09-09 14:08:54 +07:00
|
|
|
struct bdi_writeback *wb = &inode_to_bdi(inode)->wb;
|
|
|
|
|
2011-03-22 18:23:41 +07:00
|
|
|
assert_spin_locked(&inode_wb_list_lock);
|
2010-10-21 07:49:30 +07:00
|
|
|
list_move(&inode->i_wb_list, &wb->b_more_io);
|
2007-10-17 13:30:34 +07:00
|
|
|
}
|
|
|
|
|
2007-10-17 13:30:44 +07:00
|
|
|
static void inode_sync_complete(struct inode *inode)
|
|
|
|
{
|
|
|
|
/*
|
2011-03-22 18:23:41 +07:00
|
|
|
* Prevent speculative execution through
|
|
|
|
* spin_unlock(&inode_wb_list_lock);
|
2007-10-17 13:30:44 +07:00
|
|
|
*/
|
2011-03-22 18:23:41 +07:00
|
|
|
|
2007-10-17 13:30:44 +07:00
|
|
|
smp_mb();
|
|
|
|
wake_up_bit(&inode->i_state, __I_SYNC);
|
|
|
|
}
|
|
|
|
|
2009-04-03 06:56:37 +07:00
|
|
|
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
|
2009-09-24 00:37:09 +07:00
|
|
|
* from permanently stopping the whole bdi writeback.
|
2009-04-03 06:56:37 +07:00
|
|
|
*/
|
|
|
|
ret = ret && time_before_eq(inode->dirtied_when, jiffies);
|
|
|
|
#endif
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2007-10-17 13:30:39 +07:00
|
|
|
/*
|
|
|
|
* Move expired dirty inodes from @delaying_queue to @dispatch_queue.
|
|
|
|
*/
|
|
|
|
static void move_expired_inodes(struct list_head *delaying_queue,
|
|
|
|
struct list_head *dispatch_queue,
|
|
|
|
unsigned long *older_than_this)
|
|
|
|
{
|
2009-09-24 19:42:33 +07:00
|
|
|
LIST_HEAD(tmp);
|
|
|
|
struct list_head *pos, *node;
|
2009-09-24 20:12:57 +07:00
|
|
|
struct super_block *sb = NULL;
|
2009-09-24 19:42:33 +07:00
|
|
|
struct inode *inode;
|
2009-09-24 20:12:57 +07:00
|
|
|
int do_sb_sort = 0;
|
2009-09-24 19:42:33 +07:00
|
|
|
|
2007-10-17 13:30:39 +07:00
|
|
|
while (!list_empty(delaying_queue)) {
|
2010-10-21 07:49:30 +07:00
|
|
|
inode = wb_inode(delaying_queue->prev);
|
2007-10-17 13:30:39 +07:00
|
|
|
if (older_than_this &&
|
2009-04-03 06:56:37 +07:00
|
|
|
inode_dirtied_after(inode, *older_than_this))
|
2007-10-17 13:30:39 +07:00
|
|
|
break;
|
2009-09-24 20:12:57 +07:00
|
|
|
if (sb && sb != inode->i_sb)
|
|
|
|
do_sb_sort = 1;
|
|
|
|
sb = inode->i_sb;
|
2010-10-21 07:49:30 +07:00
|
|
|
list_move(&inode->i_wb_list, &tmp);
|
2009-09-24 19:42:33 +07:00
|
|
|
}
|
|
|
|
|
2009-09-24 20:12:57 +07:00
|
|
|
/* just one sb in list, splice to dispatch_queue and we're done */
|
|
|
|
if (!do_sb_sort) {
|
|
|
|
list_splice(&tmp, dispatch_queue);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2009-09-24 19:42:33 +07:00
|
|
|
/* Move inodes from one superblock together */
|
|
|
|
while (!list_empty(&tmp)) {
|
2010-10-21 07:49:30 +07:00
|
|
|
sb = wb_inode(tmp.prev)->i_sb;
|
2009-09-24 19:42:33 +07:00
|
|
|
list_for_each_prev_safe(pos, node, &tmp) {
|
2010-10-21 07:49:30 +07:00
|
|
|
inode = wb_inode(pos);
|
2009-09-24 19:42:33 +07:00
|
|
|
if (inode->i_sb == sb)
|
2010-10-21 07:49:30 +07:00
|
|
|
list_move(&inode->i_wb_list, dispatch_queue);
|
2009-09-24 19:42:33 +07:00
|
|
|
}
|
2007-10-17 13:30:39 +07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Queue all expired dirty inodes for io, eldest first.
|
2010-08-12 04:17:42 +07:00
|
|
|
* 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 13:30:39 +07:00
|
|
|
*/
|
2009-09-09 14:08:54 +07:00
|
|
|
static void queue_io(struct bdi_writeback *wb, unsigned long *older_than_this)
|
2009-09-02 14:19:46 +07:00
|
|
|
{
|
2011-03-22 18:23:41 +07:00
|
|
|
assert_spin_locked(&inode_wb_list_lock);
|
2010-08-12 04:17:42 +07:00
|
|
|
list_splice_init(&wb->b_more_io, &wb->b_io);
|
2009-09-09 14:08:54 +07:00
|
|
|
move_expired_inodes(&wb->b_dirty, &wb->b_io, older_than_this);
|
2009-09-02 14:19:46 +07:00
|
|
|
}
|
|
|
|
|
2010-03-05 15:21:37 +07:00
|
|
|
static int write_inode(struct inode *inode, struct writeback_control *wbc)
|
2007-10-17 13:30:39 +07:00
|
|
|
{
|
2009-09-09 14:08:54 +07:00
|
|
|
if (inode->i_sb->s_op->write_inode && !is_bad_inode(inode))
|
2010-03-05 15:21:37 +07:00
|
|
|
return inode->i_sb->s_op->write_inode(inode, wbc);
|
2009-09-09 14:08:54 +07:00
|
|
|
return 0;
|
2007-10-17 13:30:39 +07:00
|
|
|
}
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
/*
|
2009-06-08 18:35:40 +07:00
|
|
|
* Wait for writeback on an inode to complete.
|
|
|
|
*/
|
|
|
|
static void inode_wait_for_writeback(struct inode *inode)
|
|
|
|
{
|
|
|
|
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 18:23:36 +07:00
|
|
|
while (inode->i_state & I_SYNC) {
|
|
|
|
spin_unlock(&inode->i_lock);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2009-06-08 18:35:40 +07:00
|
|
|
__wait_on_bit(wqh, &wq, inode_wait, TASK_UNINTERRUPTIBLE);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2010-05-25 04:32:38 +07:00
|
|
|
}
|
2009-06-08 18:35:40 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-22 18:23:43 +07:00
|
|
|
* Write out an inode's dirty pages. Called under inode_wb_list_lock and
|
|
|
|
* inode->i_lock. Either the caller has an active reference on the inode or
|
|
|
|
* the inode has I_WILL_FREE set.
|
2009-06-08 18:35:40 +07:00
|
|
|
*
|
2005-04-17 05:20:36 +07:00
|
|
|
* If `wait' is set, wait on the writeout.
|
|
|
|
*
|
|
|
|
* The whole writeout design is quite complex and fragile. We want to avoid
|
|
|
|
* starvation of particular inodes when others are being redirtied, prevent
|
|
|
|
* livelocks, etc.
|
|
|
|
*/
|
|
|
|
static int
|
2009-06-08 18:35:40 +07:00
|
|
|
writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct address_space *mapping = inode->i_mapping;
|
2009-06-08 18:35:40 +07:00
|
|
|
unsigned dirty;
|
2005-04-17 05:20:36 +07:00
|
|
|
int ret;
|
|
|
|
|
2011-03-22 18:23:43 +07:00
|
|
|
assert_spin_locked(&inode_wb_list_lock);
|
|
|
|
assert_spin_locked(&inode->i_lock);
|
|
|
|
|
2009-06-08 18:35:40 +07:00
|
|
|
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 this inode is locked for writeback and we are not doing
|
2009-09-02 14:19:46 +07:00
|
|
|
* writeback-for-data-integrity, move it to b_more_io so that
|
2009-06-08 18:35:40 +07:00
|
|
|
* writeback can proceed with the other inodes on s_io.
|
|
|
|
*
|
|
|
|
* We'll have another go at writing back this inode when we
|
2009-09-02 14:19:46 +07:00
|
|
|
* completed a full scan of b_io.
|
2009-06-08 18:35:40 +07:00
|
|
|
*/
|
2010-03-05 15:21:37 +07:00
|
|
|
if (wbc->sync_mode != WB_SYNC_ALL) {
|
2009-06-08 18:35:40 +07:00
|
|
|
requeue_io(inode);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* It's a data-integrity sync. We must wait.
|
|
|
|
*/
|
|
|
|
inode_wait_for_writeback(inode);
|
|
|
|
}
|
|
|
|
|
2007-10-17 13:30:44 +07:00
|
|
|
BUG_ON(inode->i_state & I_SYNC);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-05-07 16:35:44 +07:00
|
|
|
/* Set I_SYNC, reset I_DIRTY_PAGES */
|
2007-10-17 13:30:44 +07:00
|
|
|
inode->i_state |= I_SYNC;
|
2010-05-07 16:35:44 +07:00
|
|
|
inode->i_state &= ~I_DIRTY_PAGES;
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
ret = do_writepages(mapping, wbc);
|
|
|
|
|
2010-03-05 15:21:21 +07:00
|
|
|
/*
|
|
|
|
* Make sure to wait on the data before writing out the metadata.
|
|
|
|
* This is important for filesystems that modify metadata on data
|
|
|
|
* I/O completion.
|
|
|
|
*/
|
2010-03-05 15:21:37 +07:00
|
|
|
if (wbc->sync_mode == WB_SYNC_ALL) {
|
2010-03-05 15:21:21 +07:00
|
|
|
int err = filemap_fdatawait(mapping);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (ret == 0)
|
|
|
|
ret = err;
|
|
|
|
}
|
|
|
|
|
2010-05-07 16:35:44 +07:00
|
|
|
/*
|
|
|
|
* Some filesystems may redirty the inode during the writeback
|
|
|
|
* due to delalloc, clear dirty metadata flags right before
|
|
|
|
* write_inode()
|
|
|
|
*/
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2010-05-07 16:35:44 +07:00
|
|
|
dirty = inode->i_state & I_DIRTY;
|
|
|
|
inode->i_state &= ~(I_DIRTY_SYNC | I_DIRTY_DATASYNC);
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
2010-03-05 15:21:21 +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 15:21:37 +07:00
|
|
|
int err = write_inode(inode, wbc);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (ret == 0)
|
|
|
|
ret = err;
|
|
|
|
}
|
|
|
|
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2007-10-17 13:30:44 +07:00
|
|
|
inode->i_state &= ~I_SYNC;
|
2010-06-03 04:38:30 +07:00
|
|
|
if (!(inode->i_state & I_FREEING)) {
|
2010-08-12 04:17:41 +07:00
|
|
|
if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
|
2005-04-17 05:20:36 +07:00
|
|
|
/*
|
|
|
|
* We didn't write back all the pages. nfs_writepages()
|
2010-08-12 04:17:43 +07:00
|
|
|
* sometimes bales out without doing anything.
|
2007-10-17 13:30:35 +07:00
|
|
|
*/
|
2010-08-12 04:17:43 +07:00
|
|
|
inode->i_state |= I_DIRTY_PAGES;
|
|
|
|
if (wbc->nr_to_write <= 0) {
|
2005-04-17 05:20:36 +07:00
|
|
|
/*
|
2010-08-12 04:17:43 +07:00
|
|
|
* slice used up: queue for next turn
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2010-08-12 04:17:43 +07:00
|
|
|
requeue_io(inode);
|
2005-04-17 05:20:36 +07:00
|
|
|
} else {
|
|
|
|
/*
|
2010-08-12 04:17:43 +07:00
|
|
|
* 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.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2007-10-17 13:30:35 +07:00
|
|
|
redirty_tail(inode);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2010-08-12 04:17:41 +07:00
|
|
|
} 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);
|
2005-04-17 05:20:36 +07:00
|
|
|
} else {
|
|
|
|
/*
|
2010-10-23 17:55:17 +07:00
|
|
|
* The inode is clean. At this point we either have
|
|
|
|
* a reference to the inode or it's on it's way out.
|
|
|
|
* No need to add it back to the LRU.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2010-10-21 07:49:30 +07:00
|
|
|
list_del_init(&inode->i_wb_list);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
}
|
2007-10-17 13:30:44 +07:00
|
|
|
inode_sync_complete(inode);
|
2005-04-17 05:20:36 +07:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
/*
|
2010-06-08 23:14:58 +07:00
|
|
|
* For background writeback the caller does not have the sb pinned
|
2009-09-09 14:08:54 +07:00
|
|
|
* before calling writeback. So make sure that we do pin it, so it doesn't
|
|
|
|
* go away while we are writing inodes from it.
|
|
|
|
*/
|
2010-06-08 23:14:58 +07:00
|
|
|
static bool pin_sb_for_writeback(struct super_block *sb)
|
2009-09-09 14:08:54 +07:00
|
|
|
{
|
|
|
|
spin_lock(&sb_lock);
|
2010-06-09 20:31:01 +07:00
|
|
|
if (list_empty(&sb->s_instances)) {
|
|
|
|
spin_unlock(&sb_lock);
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
sb->s_count++;
|
2010-06-09 20:31:01 +07:00
|
|
|
spin_unlock(&sb_lock);
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
if (down_read_trylock(&sb->s_umount)) {
|
2010-06-09 20:31:01 +07:00
|
|
|
if (sb->s_root)
|
2010-06-08 23:14:58 +07:00
|
|
|
return true;
|
2009-09-09 14:08:54 +07:00
|
|
|
up_read(&sb->s_umount);
|
|
|
|
}
|
2010-06-09 20:31:01 +07:00
|
|
|
|
|
|
|
put_super(sb);
|
2010-06-08 23:14:58 +07:00
|
|
|
return false;
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
|
|
|
|
2010-03-12 05:09:47 +07:00
|
|
|
/*
|
|
|
|
* Write a portion of b_io inodes which belong to @sb.
|
2010-06-10 17:07:54 +07:00
|
|
|
*
|
|
|
|
* If @only_this_sb is true, then find and write all such
|
2010-03-12 05:09:47 +07:00
|
|
|
* inodes. Otherwise write only ones which go sequentially
|
|
|
|
* in reverse order.
|
2010-06-10 17:07:54 +07:00
|
|
|
*
|
2010-03-12 05:09:47 +07:00
|
|
|
* Return 1, if the caller writeback routine should be
|
|
|
|
* interrupted. Otherwise return 0.
|
|
|
|
*/
|
2010-06-10 17:07:54 +07:00
|
|
|
static int writeback_sb_inodes(struct super_block *sb, struct bdi_writeback *wb,
|
|
|
|
struct writeback_control *wbc, bool only_this_sb)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2009-09-09 14:08:54 +07:00
|
|
|
while (!list_empty(&wb->b_io)) {
|
2005-04-17 05:20:36 +07:00
|
|
|
long pages_skipped;
|
2010-10-21 07:49:30 +07:00
|
|
|
struct inode *inode = wb_inode(wb->b_io.prev);
|
2010-06-10 17:07:54 +07:00
|
|
|
|
|
|
|
if (inode->i_sb != sb) {
|
|
|
|
if (only_this_sb) {
|
|
|
|
/*
|
|
|
|
* We only want to write back data for this
|
|
|
|
* superblock, move all inodes not belonging
|
|
|
|
* to it back onto the dirty list.
|
|
|
|
*/
|
|
|
|
redirty_tail(inode);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The inode belongs to a different superblock.
|
|
|
|
* Bounce back to the caller to unpin this and
|
|
|
|
* pin the next superblock.
|
|
|
|
*/
|
2010-03-12 05:09:47 +07:00
|
|
|
return 0;
|
2010-06-10 17:07:54 +07:00
|
|
|
}
|
|
|
|
|
2010-10-25 00:40:46 +07:00
|
|
|
/*
|
|
|
|
* Don't bother with new inodes or inodes beeing freed, first
|
|
|
|
* kind does not need peridic writeout yet, and for the latter
|
|
|
|
* kind writeout is handled by the freer.
|
|
|
|
*/
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2010-10-25 00:40:46 +07:00
|
|
|
if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) {
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
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>
2009-03-13 04:31:38 +07:00
|
|
|
requeue_io(inode);
|
|
|
|
continue;
|
|
|
|
}
|
2010-10-25 00:40:46 +07:00
|
|
|
|
2009-04-03 06:56:37 +07:00
|
|
|
/*
|
|
|
|
* Was this inode dirtied after sync_sb_inodes was called?
|
|
|
|
* This keeps sync from extra jobs and livelock.
|
|
|
|
*/
|
2011-03-22 18:23:36 +07:00
|
|
|
if (inode_dirtied_after(inode, wbc->wb_start)) {
|
|
|
|
spin_unlock(&inode->i_lock);
|
2010-03-12 05:09:47 +07:00
|
|
|
return 1;
|
2011-03-22 18:23:36 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
__iget(inode);
|
2011-03-22 18:23:36 +07:00
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
pages_skipped = wbc->pages_skipped;
|
2009-06-08 18:35:40 +07:00
|
|
|
writeback_single_inode(inode, wbc);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (wbc->pages_skipped != pages_skipped) {
|
|
|
|
/*
|
|
|
|
* writeback is not making progress due to locked
|
|
|
|
* buffers. Skip this inode for now.
|
|
|
|
*/
|
writeback: fix time ordering of the per superblock dirty inode lists 3
While writeback is working against a dirty inode it does a check after trying
to write some of the inode's pages:
"did the lower layers skip some of the inode's dirty pages because they were
locked (or under writeback, or whatever)"
If this turns out to be true, we must move the inode back onto s_dirty and
redirty it. The reason for doing this is that fsync() and friends only check
the s_dirty list, and those functions want to know about those pages which
were locked, so they can be waited upon and, if necessary, rewritten.
Problem is, that redirtying was putting the inode onto the tail of s_dirty
without updating its timestamp. This causes a violation of s_dirty ordering.
Fix this by updating inode->dirtied_when when moving the inode onto s_dirty.
But the code is still a bit buggy? If the inode was _already_ dirty then we
don't need to move it at all. Oh well, hopefully it doesn't matter too much,
as that was a redirtying, which was very recent anwyay.
Cc: Mike Waychison <mikew@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-10-17 13:30:34 +07:00
|
|
|
redirty_tail(inode);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2011-03-22 18:23:43 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
iput(inode);
|
2006-03-25 18:07:44 +07:00
|
|
|
cond_resched();
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
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>
2008-02-05 13:29:36 +07:00
|
|
|
if (wbc->nr_to_write <= 0) {
|
|
|
|
wbc->more_io = 1;
|
2010-03-12 05:09:47 +07:00
|
|
|
return 1;
|
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>
2008-02-05 13:29:36 +07:00
|
|
|
}
|
2009-09-09 14:08:54 +07:00
|
|
|
if (!list_empty(&wb->b_more_io))
|
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>
2008-02-05 13:29:36 +07:00
|
|
|
wbc->more_io = 1;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2010-03-12 05:09:47 +07:00
|
|
|
/* b_io is empty */
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2010-06-10 17:07:27 +07:00
|
|
|
void writeback_inodes_wb(struct bdi_writeback *wb,
|
|
|
|
struct writeback_control *wbc)
|
2010-03-12 05:09:47 +07:00
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
|
2010-08-10 07:20:03 +07:00
|
|
|
if (!wbc->wb_start)
|
|
|
|
wbc->wb_start = jiffies; /* livelock avoidance */
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
2010-03-12 05:09:47 +07:00
|
|
|
if (!wbc->for_kupdate || list_empty(&wb->b_io))
|
|
|
|
queue_io(wb, wbc->older_than_this);
|
2009-01-07 05:40:25 +07:00
|
|
|
|
2010-03-12 05:09:47 +07:00
|
|
|
while (!list_empty(&wb->b_io)) {
|
2010-10-21 07:49:30 +07:00
|
|
|
struct inode *inode = wb_inode(wb->b_io.prev);
|
2010-03-12 05:09:47 +07:00
|
|
|
struct super_block *sb = inode->i_sb;
|
2009-09-24 20:25:11 +07:00
|
|
|
|
2010-06-10 17:07:54 +07:00
|
|
|
if (!pin_sb_for_writeback(sb)) {
|
|
|
|
requeue_io(inode);
|
|
|
|
continue;
|
2010-03-12 05:09:47 +07:00
|
|
|
}
|
2010-06-10 17:07:54 +07:00
|
|
|
ret = writeback_sb_inodes(sb, wb, wbc, false);
|
|
|
|
drop_super(sb);
|
2010-03-12 05:09:47 +07:00
|
|
|
|
|
|
|
if (ret)
|
|
|
|
break;
|
|
|
|
}
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2009-09-02 14:19:46 +07:00
|
|
|
/* Leave any unwritten inodes on b_io */
|
|
|
|
}
|
|
|
|
|
2010-06-10 17:07:54 +07:00
|
|
|
static void __writeback_inodes_sb(struct super_block *sb,
|
|
|
|
struct bdi_writeback *wb, struct writeback_control *wbc)
|
|
|
|
{
|
|
|
|
WARN_ON(!rwsem_is_locked(&sb->s_umount));
|
|
|
|
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
2010-06-10 17:07:54 +07:00
|
|
|
if (!wbc->for_kupdate || list_empty(&wb->b_io))
|
|
|
|
queue_io(wb, wbc->older_than_this);
|
|
|
|
writeback_sb_inodes(sb, wb, wbc, true);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2010-06-10 17:07:54 +07:00
|
|
|
}
|
|
|
|
|
2009-09-02 14:19:46 +07:00
|
|
|
/*
|
2009-09-09 14:08:54 +07:00
|
|
|
* The maximum number of pages to writeout in a single bdi flush/kupdate
|
|
|
|
* operation. We do this so we don't hold I_SYNC against an inode for
|
|
|
|
* enormous amounts of time, which would block a userspace task which has
|
|
|
|
* been forced to throttle against that inode. Also, the code reevaluates
|
|
|
|
* the dirty each time it has written this many pages.
|
|
|
|
*/
|
|
|
|
#define MAX_WRITEBACK_PAGES 1024
|
|
|
|
|
|
|
|
static inline bool over_bground_thresh(void)
|
|
|
|
{
|
|
|
|
unsigned long background_thresh, dirty_thresh;
|
|
|
|
|
2010-08-12 04:17:39 +07:00
|
|
|
global_dirty_limits(&background_thresh, &dirty_thresh);
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
return (global_page_state(NR_FILE_DIRTY) +
|
2010-10-27 04:21:45 +07:00
|
|
|
global_page_state(NR_UNSTABLE_NFS) > background_thresh);
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Explicit flushing or periodic writeback of "old" data.
|
2009-09-02 14:19:46 +07:00
|
|
|
*
|
2009-09-09 14:08:54 +07:00
|
|
|
* 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 14:19:46 +07:00
|
|
|
*
|
2009-09-09 14:08:54 +07:00
|
|
|
* 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 14:19:46 +07:00
|
|
|
*
|
2009-09-09 14:08:54 +07:00
|
|
|
* 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 14:19:46 +07:00
|
|
|
*/
|
2009-09-16 20:18:25 +07:00
|
|
|
static long wb_writeback(struct bdi_writeback *wb,
|
2010-07-06 13:59:53 +07:00
|
|
|
struct wb_writeback_work *work)
|
2009-09-02 14:19:46 +07:00
|
|
|
{
|
2009-09-09 14:08:54 +07:00
|
|
|
struct writeback_control wbc = {
|
2010-07-06 13:59:53 +07:00
|
|
|
.sync_mode = work->sync_mode,
|
2009-09-09 14:08:54 +07:00
|
|
|
.older_than_this = NULL,
|
2010-07-06 13:59:53 +07:00
|
|
|
.for_kupdate = work->for_kupdate,
|
|
|
|
.for_background = work->for_background,
|
|
|
|
.range_cyclic = work->range_cyclic,
|
2009-09-09 14:08:54 +07:00
|
|
|
};
|
|
|
|
unsigned long oldest_jif;
|
|
|
|
long wrote = 0;
|
2011-01-14 06:45:48 +07:00
|
|
|
long write_chunk;
|
2009-09-17 00:22:48 +07:00
|
|
|
struct inode *inode;
|
2009-09-02 14:19:46 +07:00
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
if (wbc.for_kupdate) {
|
|
|
|
wbc.older_than_this = &oldest_jif;
|
|
|
|
oldest_jif = jiffies -
|
|
|
|
msecs_to_jiffies(dirty_expire_interval * 10);
|
|
|
|
}
|
2009-09-16 20:18:25 +07:00
|
|
|
if (!wbc.range_cyclic) {
|
|
|
|
wbc.range_start = 0;
|
|
|
|
wbc.range_end = LLONG_MAX;
|
|
|
|
}
|
2009-01-07 05:40:25 +07:00
|
|
|
|
2011-01-14 06:45:48 +07:00
|
|
|
/*
|
|
|
|
* 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_inodes_sb() <== called only once
|
|
|
|
* write_cache_pages() <== called once for each inode
|
|
|
|
* (quickly) tag currently dirty pages
|
|
|
|
* (maybe slowly) sync all tagged pages
|
|
|
|
*/
|
|
|
|
if (wbc.sync_mode == WB_SYNC_NONE)
|
|
|
|
write_chunk = MAX_WRITEBACK_PAGES;
|
|
|
|
else
|
|
|
|
write_chunk = LONG_MAX;
|
|
|
|
|
2010-08-10 07:20:03 +07:00
|
|
|
wbc.wb_start = jiffies; /* livelock avoidance */
|
2009-09-09 14:08:54 +07:00
|
|
|
for (;;) {
|
|
|
|
/*
|
2009-09-23 19:33:40 +07:00
|
|
|
* Stop writeback when nr_pages has been consumed
|
2009-09-09 14:08:54 +07:00
|
|
|
*/
|
2010-07-06 13:59:53 +07:00
|
|
|
if (work->nr_pages <= 0)
|
2009-09-09 14:08:54 +07:00
|
|
|
break;
|
2009-09-02 14:19:46 +07:00
|
|
|
|
2011-01-14 06:45:47 +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;
|
|
|
|
|
2009-01-07 05:40:25 +07:00
|
|
|
/*
|
2009-09-23 19:33:40 +07:00
|
|
|
* For background writeout, stop when we are below the
|
|
|
|
* background dirty threshold
|
2009-01-07 05:40:25 +07:00
|
|
|
*/
|
2010-07-06 13:59:53 +07:00
|
|
|
if (work->for_background && !over_bground_thresh())
|
2009-09-09 14:08:54 +07:00
|
|
|
break;
|
2009-01-07 05:40:25 +07:00
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
wbc.more_io = 0;
|
2011-01-14 06:45:48 +07:00
|
|
|
wbc.nr_to_write = write_chunk;
|
2009-09-09 14:08:54 +07:00
|
|
|
wbc.pages_skipped = 0;
|
2010-07-07 10:24:07 +07:00
|
|
|
|
|
|
|
trace_wbc_writeback_start(&wbc, wb->bdi);
|
2010-07-06 13:59:53 +07:00
|
|
|
if (work->sb)
|
|
|
|
__writeback_inodes_sb(work->sb, wb, &wbc);
|
2010-06-10 17:07:54 +07:00
|
|
|
else
|
|
|
|
writeback_inodes_wb(wb, &wbc);
|
2010-07-07 10:24:07 +07:00
|
|
|
trace_wbc_writeback_written(&wbc, wb->bdi);
|
|
|
|
|
2011-01-14 06:45:48 +07:00
|
|
|
work->nr_pages -= write_chunk - wbc.nr_to_write;
|
|
|
|
wrote += write_chunk - wbc.nr_to_write;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
/*
|
2009-09-24 00:32:26 +07:00
|
|
|
* If we consumed everything, see if we have more
|
2009-09-09 14:08:54 +07:00
|
|
|
*/
|
2009-09-24 00:32:26 +07:00
|
|
|
if (wbc.nr_to_write <= 0)
|
|
|
|
continue;
|
|
|
|
/*
|
|
|
|
* Didn't write everything and we don't have more IO, bail
|
|
|
|
*/
|
|
|
|
if (!wbc.more_io)
|
2009-09-09 14:08:54 +07:00
|
|
|
break;
|
2009-09-24 00:32:26 +07:00
|
|
|
/*
|
|
|
|
* Did we write something? Try for more
|
|
|
|
*/
|
2011-01-14 06:45:48 +07:00
|
|
|
if (wbc.nr_to_write < write_chunk)
|
2009-09-24 00:32:26 +07:00
|
|
|
continue;
|
|
|
|
/*
|
|
|
|
* Nothing written. Wait for some inode to
|
|
|
|
* become available for writeback. Otherwise
|
|
|
|
* we'll just busyloop.
|
|
|
|
*/
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
2009-09-24 00:32:26 +07:00
|
|
|
if (!list_empty(&wb->b_more_io)) {
|
2010-10-21 07:49:30 +07:00
|
|
|
inode = wb_inode(wb->b_more_io.prev);
|
2010-07-07 10:24:07 +07:00
|
|
|
trace_wbc_writeback_wait(&wbc, wb->bdi);
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2009-09-24 00:32:26 +07:00
|
|
|
inode_wait_for_writeback(inode);
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
return wrote;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2010-07-06 13:59:53 +07:00
|
|
|
* Return the next wb_writeback_work struct that hasn't been processed yet.
|
2009-09-09 14:08:54 +07:00
|
|
|
*/
|
2010-07-06 13:59:53 +07:00
|
|
|
static struct wb_writeback_work *
|
2010-08-03 17:51:16 +07:00
|
|
|
get_next_work_item(struct backing_dev_info *bdi)
|
2009-09-09 14:08:54 +07:00
|
|
|
{
|
2010-07-06 13:59:53 +07:00
|
|
|
struct wb_writeback_work *work = NULL;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-07-25 18:29:22 +07:00
|
|
|
spin_lock_bh(&bdi->wb_lock);
|
2010-07-06 13:59:53 +07:00
|
|
|
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 14:08:54 +07:00
|
|
|
}
|
2010-07-25 18:29:22 +07:00
|
|
|
spin_unlock_bh(&bdi->wb_lock);
|
2010-07-06 13:59:53 +07:00
|
|
|
return work;
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
|
|
|
|
2010-10-30 22:55:52 +07:00
|
|
|
/*
|
|
|
|
* 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();
|
|
|
|
}
|
|
|
|
|
2011-01-14 06:45:44 +07:00
|
|
|
static long wb_check_background_flush(struct bdi_writeback *wb)
|
|
|
|
{
|
|
|
|
if (over_bground_thresh()) {
|
|
|
|
|
|
|
|
struct wb_writeback_work work = {
|
|
|
|
.nr_pages = LONG_MAX,
|
|
|
|
.sync_mode = WB_SYNC_NONE,
|
|
|
|
.for_background = 1,
|
|
|
|
.range_cyclic = 1,
|
|
|
|
};
|
|
|
|
|
|
|
|
return wb_writeback(wb, &work);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
static long wb_check_old_data_flush(struct bdi_writeback *wb)
|
|
|
|
{
|
|
|
|
unsigned long expired;
|
|
|
|
long nr_pages;
|
|
|
|
|
2010-05-17 17:51:03 +07:00
|
|
|
/*
|
|
|
|
* When set to zero, disable periodic writeback
|
|
|
|
*/
|
|
|
|
if (!dirty_writeback_interval)
|
|
|
|
return 0;
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
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 22:55:52 +07:00
|
|
|
nr_pages = get_nr_dirty_pages();
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2009-09-16 20:18:25 +07:00
|
|
|
if (nr_pages) {
|
2010-07-06 13:59:53 +07:00
|
|
|
struct wb_writeback_work work = {
|
2009-09-16 20:18:25 +07:00
|
|
|
.nr_pages = nr_pages,
|
|
|
|
.sync_mode = WB_SYNC_NONE,
|
|
|
|
.for_kupdate = 1,
|
|
|
|
.range_cyclic = 1,
|
|
|
|
};
|
|
|
|
|
2010-07-06 13:59:53 +07:00
|
|
|
return wb_writeback(wb, &work);
|
2009-09-16 20:18:25 +07:00
|
|
|
}
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
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 13:59:53 +07:00
|
|
|
struct wb_writeback_work *work;
|
2009-09-16 20:18:25 +07:00
|
|
|
long wrote = 0;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-08-12 04:17:44 +07:00
|
|
|
set_bit(BDI_writeback_running, &wb->bdi->state);
|
2010-08-03 17:51:16 +07:00
|
|
|
while ((work = get_next_work_item(bdi)) != NULL) {
|
2009-09-09 14:08:54 +07:00
|
|
|
/*
|
|
|
|
* Override sync mode, in case we must wait for completion
|
2010-07-06 13:59:53 +07:00
|
|
|
* because this thread is exiting now.
|
2009-09-09 14:08:54 +07:00
|
|
|
*/
|
|
|
|
if (force_wait)
|
2010-07-06 13:59:53 +07:00
|
|
|
work->sync_mode = WB_SYNC_ALL;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-07-07 10:24:06 +07:00
|
|
|
trace_writeback_exec(bdi, work);
|
|
|
|
|
2010-07-06 13:59:53 +07:00
|
|
|
wrote += wb_writeback(wb, work);
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
/*
|
2010-07-06 13:59:53 +07:00
|
|
|
* Notify the caller of completion if this is a synchronous
|
|
|
|
* work item, otherwise just free it.
|
2009-09-09 14:08:54 +07:00
|
|
|
*/
|
2010-07-06 13:59:53 +07:00
|
|
|
if (work->done)
|
|
|
|
complete(work->done);
|
|
|
|
else
|
|
|
|
kfree(work);
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check for periodic writeback, kupdated() style
|
|
|
|
*/
|
|
|
|
wrote += wb_check_old_data_flush(wb);
|
2011-01-14 06:45:44 +07:00
|
|
|
wrote += wb_check_background_flush(wb);
|
2010-08-12 04:17:44 +07:00
|
|
|
clear_bit(BDI_writeback_running, &wb->bdi->state);
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
return wrote;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Handle writeback of dirty data for the device backed by this bdi. Also
|
|
|
|
* wakes up periodically and does kupdated style flushing.
|
|
|
|
*/
|
2010-06-20 04:08:22 +07:00
|
|
|
int bdi_writeback_thread(void *data)
|
2009-09-09 14:08:54 +07:00
|
|
|
{
|
2010-06-20 04:08:22 +07:00
|
|
|
struct bdi_writeback *wb = data;
|
|
|
|
struct backing_dev_info *bdi = wb->bdi;
|
2009-09-09 14:08:54 +07:00
|
|
|
long pages_written;
|
|
|
|
|
2010-10-27 04:22:45 +07:00
|
|
|
current->flags |= PF_SWAPWRITE;
|
2010-06-20 04:08:22 +07:00
|
|
|
set_freezable();
|
2010-07-25 18:29:18 +07:00
|
|
|
wb->last_active = jiffies;
|
2010-06-20 04:08:22 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Our parent may run at a different priority, just set us to normal
|
|
|
|
*/
|
|
|
|
set_user_nice(current, 0);
|
|
|
|
|
2010-07-07 10:24:06 +07:00
|
|
|
trace_writeback_thread_start(bdi);
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
while (!kthread_should_stop()) {
|
2010-07-25 18:29:22 +07:00
|
|
|
/*
|
|
|
|
* Remove own delayed wake-up timer, since we are already awake
|
|
|
|
* and we'll take care of the preriodic write-back.
|
|
|
|
*/
|
|
|
|
del_timer(&wb->wakeup_timer);
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
pages_written = wb_do_writeback(wb, 0);
|
|
|
|
|
2010-07-07 10:24:06 +07:00
|
|
|
trace_writeback_pages_written(pages_written);
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
if (pages_written)
|
2010-07-25 18:29:18 +07:00
|
|
|
wb->last_active = jiffies;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-07-25 18:29:15 +07:00
|
|
|
set_current_state(TASK_INTERRUPTIBLE);
|
2010-08-28 13:52:10 +07:00
|
|
|
if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
|
2010-05-18 19:31:45 +07:00
|
|
|
__set_current_state(TASK_RUNNING);
|
2010-07-25 18:29:15 +07:00
|
|
|
continue;
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
|
|
|
|
2010-07-25 18:29:21 +07:00
|
|
|
if (wb_has_dirty_io(wb) && dirty_writeback_interval)
|
writeback: move bdi threads exiting logic to the forker thread
Currently, bdi threads can decide to exit if there were no useful activities
for 5 minutes. However, this causes nasty races: we can easily oops in the
'bdi_queue_work()' if the bdi thread decides to exit while we are waking it up.
And even if we do not oops, but the bdi tread exits immediately after we wake
it up, we'd lose the wake-up event and have an unnecessary delay (up to 5 secs)
in the bdi work processing.
This patch makes the forker thread to be the central place which not only
creates bdi threads, but also kills them if they were inactive long enough.
This better design-wise.
Another reason why this change was done is to prepare for the further changes
which will prevent the bdi threads from waking up every 5 sec and wasting
power. Indeed, when the task does not wake up periodically anymore, it won't be
able to exit either.
This patch also moves the the 'wake_up_bit()' call from the bdi thread to the
forker thread as well. So now the forker thread sets the BDI_pending bit, then
forks the task or kills it, then clears the bit and wakes up the waiting
process.
The only process which may wain on the bit is 'bdi_wb_shutdown()'. This
function was changed as well - now it first removes the bdi from the
'bdi_list', then waits on the 'BDI_pending' bit. Once it wakes up, it is
guaranteed that the forker thread won't race with it, because the bdi is not
visible. Note, the forker thread sets the 'BDI_pending' bit under the
'bdi->wb_lock' which is essential for proper serialization.
And additionally, when we change 'bdi->wb.task', we now take the
'bdi->work_lock', to make sure that we do not lose wake-ups which we otherwise
would when raced with, say, 'bdi_queue_work()'.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2010-07-25 18:29:20 +07:00
|
|
|
schedule_timeout(msecs_to_jiffies(dirty_writeback_interval * 10));
|
2010-07-25 18:29:21 +07:00
|
|
|
else {
|
|
|
|
/*
|
|
|
|
* We have nothing to do, so can go sleep without any
|
|
|
|
* timeout and save power. When a work is queued or
|
|
|
|
* something is made dirty - we will be woken up.
|
|
|
|
*/
|
2010-07-25 18:29:15 +07:00
|
|
|
schedule();
|
2010-05-18 19:31:45 +07:00
|
|
|
}
|
2010-05-17 17:51:03 +07:00
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
try_to_freeze();
|
|
|
|
}
|
|
|
|
|
writeback: move bdi threads exiting logic to the forker thread
Currently, bdi threads can decide to exit if there were no useful activities
for 5 minutes. However, this causes nasty races: we can easily oops in the
'bdi_queue_work()' if the bdi thread decides to exit while we are waking it up.
And even if we do not oops, but the bdi tread exits immediately after we wake
it up, we'd lose the wake-up event and have an unnecessary delay (up to 5 secs)
in the bdi work processing.
This patch makes the forker thread to be the central place which not only
creates bdi threads, but also kills them if they were inactive long enough.
This better design-wise.
Another reason why this change was done is to prepare for the further changes
which will prevent the bdi threads from waking up every 5 sec and wasting
power. Indeed, when the task does not wake up periodically anymore, it won't be
able to exit either.
This patch also moves the the 'wake_up_bit()' call from the bdi thread to the
forker thread as well. So now the forker thread sets the BDI_pending bit, then
forks the task or kills it, then clears the bit and wakes up the waiting
process.
The only process which may wain on the bit is 'bdi_wb_shutdown()'. This
function was changed as well - now it first removes the bdi from the
'bdi_list', then waits on the 'BDI_pending' bit. Once it wakes up, it is
guaranteed that the forker thread won't race with it, because the bdi is not
visible. Note, the forker thread sets the 'BDI_pending' bit under the
'bdi->wb_lock' which is essential for proper serialization.
And additionally, when we change 'bdi->wb.task', we now take the
'bdi->work_lock', to make sure that we do not lose wake-ups which we otherwise
would when raced with, say, 'bdi_queue_work()'.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2010-07-25 18:29:20 +07:00
|
|
|
/* Flush any work that raced with us exiting */
|
2010-06-20 04:08:22 +07:00
|
|
|
if (!list_empty(&bdi->work_list))
|
|
|
|
wb_do_writeback(wb, 1);
|
2010-07-07 10:24:06 +07:00
|
|
|
|
|
|
|
trace_writeback_thread_stop(bdi);
|
2009-09-09 14:08:54 +07:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2010-06-20 04:08:22 +07:00
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
/*
|
2010-06-08 23:15:07 +07:00
|
|
|
* Start writeback of `nr_pages' pages. If `nr_pages' is zero, write back
|
|
|
|
* the whole world.
|
2009-09-09 14:08:54 +07:00
|
|
|
*/
|
2010-06-08 23:15:07 +07:00
|
|
|
void wakeup_flusher_threads(long nr_pages)
|
2009-09-09 14:08:54 +07:00
|
|
|
{
|
2010-06-08 23:15:07 +07:00
|
|
|
struct backing_dev_info *bdi;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-07-06 13:59:53 +07:00
|
|
|
if (!nr_pages) {
|
|
|
|
nr_pages = global_page_state(NR_FILE_DIRTY) +
|
2010-06-08 23:15:07 +07:00
|
|
|
global_page_state(NR_UNSTABLE_NFS);
|
|
|
|
}
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2010-06-08 23:15:07 +07:00
|
|
|
rcu_read_lock();
|
2009-09-14 18:12:40 +07:00
|
|
|
list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
|
2009-09-09 14:08:54 +07:00
|
|
|
if (!bdi_has_dirty_io(bdi))
|
|
|
|
continue;
|
2011-01-14 06:45:44 +07:00
|
|
|
__bdi_start_writeback(bdi, nr_pages, false);
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
2009-09-14 18:12:40 +07:00
|
|
|
rcu_read_unlock();
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
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.
|
2005-04-17 05:20:36 +07:00
|
|
|
*
|
2009-09-09 14:08:54 +07:00
|
|
|
* 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.
|
2005-04-17 05:20:36 +07:00
|
|
|
*
|
2009-09-09 14:08:54 +07:00
|
|
|
* This function *must* be atomic for the I_DIRTY_PAGES case -
|
|
|
|
* set_page_dirty() is called under spinlock in several places.
|
2005-04-17 05:20:36 +07:00
|
|
|
*
|
2009-09-09 14:08:54 +07:00
|
|
|
* 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.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2009-09-09 14:08:54 +07:00
|
|
|
void __mark_inode_dirty(struct inode *inode, int flags)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2009-09-09 14:08:54 +07:00
|
|
|
struct super_block *sb = inode->i_sb;
|
2010-07-25 18:29:21 +07:00
|
|
|
struct backing_dev_info *bdi = NULL;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
/*
|
|
|
|
* Don't do this for I_DIRTY_PAGES - that doesn't actually
|
|
|
|
* dirty the inode itself
|
|
|
|
*/
|
|
|
|
if (flags & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) {
|
|
|
|
if (sb->s_op->dirty_inode)
|
|
|
|
sb->s_op->dirty_inode(inode);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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 18:23:36 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
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 18:23:36 +07:00
|
|
|
goto out_unlock_inode;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Only add valid (hashed) inodes to the superblock's
|
|
|
|
* dirty list. Add blockdev inodes as well.
|
|
|
|
*/
|
|
|
|
if (!S_ISBLK(inode->i_mode)) {
|
2010-10-24 02:19:20 +07:00
|
|
|
if (inode_unhashed(inode))
|
2011-03-22 18:23:36 +07:00
|
|
|
goto out_unlock_inode;
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
2010-06-03 04:38:30 +07:00
|
|
|
if (inode->i_state & I_FREEING)
|
2011-03-22 18:23:36 +07:00
|
|
|
goto out_unlock_inode;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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 18:23:41 +07:00
|
|
|
bool wakeup_bdi = false;
|
2010-07-25 18:29:21 +07:00
|
|
|
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 14:10:25 +07:00
|
|
|
}
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
|
|
|
spin_lock(&inode_wb_list_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
inode->dirtied_when = jiffies;
|
2010-10-21 07:49:30 +07:00
|
|
|
list_move(&inode->i_wb_list, &bdi->wb.b_dirty);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
|
|
|
|
|
|
|
if (wakeup_bdi)
|
|
|
|
bdi_wakeup_thread_delayed(bdi);
|
|
|
|
return;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
}
|
2011-03-22 18:23:36 +07:00
|
|
|
out_unlock_inode:
|
|
|
|
spin_unlock(&inode->i_lock);
|
2010-07-25 18:29:21 +07:00
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(__mark_inode_dirty);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Write out a superblock's list of dirty inodes. A wait will be performed
|
|
|
|
* upon no inodes, all inodes or the final one, depending upon sync_mode.
|
|
|
|
*
|
|
|
|
* If older_than_this is non-NULL, then only write out inodes which
|
|
|
|
* had their first dirtying at a time earlier than *older_than_this.
|
|
|
|
*
|
|
|
|
* If `bdi' is non-zero then we're being asked to writeback a specific queue.
|
|
|
|
* This function assumes that the blockdev superblock's inodes are backed by
|
|
|
|
* a variety of queues, so all inodes are searched. For other superblocks,
|
|
|
|
* assume that all inodes are backed by the same queue.
|
|
|
|
*
|
|
|
|
* The inodes to be written are parked on bdi->b_io. They are moved back onto
|
|
|
|
* bdi->b_dirty as they are selected for writing. This way, none can be missed
|
|
|
|
* on the writer throttling path, and we get decent balancing between many
|
|
|
|
* throttled threads: we don't want them all piling up on inode_sync_wait.
|
|
|
|
*/
|
2009-09-16 20:13:54 +07:00
|
|
|
static void wait_sb_inodes(struct super_block *sb)
|
2009-09-09 14:08:54 +07:00
|
|
|
{
|
|
|
|
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 20:13:54 +07:00
|
|
|
WARN_ON(!rwsem_is_locked(&sb->s_umount));
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2011-03-22 18:23:40 +07:00
|
|
|
spin_lock(&inode_sb_list_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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 20:13:54 +07:00
|
|
|
list_for_each_entry(inode, &sb->s_inodes, i_sb_list) {
|
2011-03-22 18:23:36 +07:00
|
|
|
struct address_space *mapping = inode->i_mapping;
|
2009-09-09 14:08:54 +07:00
|
|
|
|
2011-03-22 18:23:36 +07:00
|
|
|
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 14:08:54 +07:00
|
|
|
continue;
|
2011-03-22 18:23:36 +07:00
|
|
|
}
|
2009-09-09 14:08:54 +07:00
|
|
|
__iget(inode);
|
2011-03-22 18:23:36 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-03-22 18:23:40 +07:00
|
|
|
spin_unlock(&inode_sb_list_lock);
|
|
|
|
|
2009-09-09 14:08:54 +07:00
|
|
|
/*
|
2011-03-22 18:23:40 +07:00
|
|
|
* 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 14:08:54 +07:00
|
|
|
*/
|
|
|
|
iput(old_inode);
|
|
|
|
old_inode = inode;
|
|
|
|
|
|
|
|
filemap_fdatawait(mapping);
|
|
|
|
|
|
|
|
cond_resched();
|
|
|
|
|
2011-03-22 18:23:40 +07:00
|
|
|
spin_lock(&inode_sb_list_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
}
|
2011-03-22 18:23:40 +07:00
|
|
|
spin_unlock(&inode_sb_list_lock);
|
2009-09-09 14:08:54 +07:00
|
|
|
iput(old_inode);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2009-09-02 17:34:32 +07:00
|
|
|
/**
|
2010-10-29 22:16:17 +07:00
|
|
|
* writeback_inodes_sb_nr - writeback dirty inodes from given super_block
|
2009-09-02 17:34:32 +07:00
|
|
|
* @sb: the superblock
|
2010-10-29 22:16:17 +07:00
|
|
|
* @nr: the number of pages to write
|
2005-04-17 05:20:36 +07:00
|
|
|
*
|
2009-09-02 17:34:32 +07:00
|
|
|
* 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 22:16:17 +07:00
|
|
|
* for IO completion of submitted IO.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2010-10-29 22:16:17 +07:00
|
|
|
void writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-07-06 13:59:53 +07:00
|
|
|
DECLARE_COMPLETION_ONSTACK(done);
|
|
|
|
struct wb_writeback_work work = {
|
2010-06-08 23:14:43 +07:00
|
|
|
.sb = sb,
|
|
|
|
.sync_mode = WB_SYNC_NONE,
|
2010-07-06 13:59:53 +07:00
|
|
|
.done = &done,
|
2010-10-29 22:16:17 +07:00
|
|
|
.nr_pages = nr,
|
2010-06-08 23:14:43 +07:00
|
|
|
};
|
2009-09-02 17:34:32 +07:00
|
|
|
|
2010-06-08 23:14:51 +07:00
|
|
|
WARN_ON(!rwsem_is_locked(&sb->s_umount));
|
2010-07-06 13:59:53 +07:00
|
|
|
bdi_queue_work(sb->s_bdi, &work);
|
|
|
|
wait_for_completion(&done);
|
2010-05-17 17:55:07 +07:00
|
|
|
}
|
2010-10-29 22:16:17 +07:00
|
|
|
EXPORT_SYMBOL(writeback_inodes_sb_nr);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* writeback_inodes_sb - writeback dirty inodes from given super_block
|
|
|
|
* @sb: the superblock
|
|
|
|
*
|
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
void writeback_inodes_sb(struct super_block *sb)
|
|
|
|
{
|
2010-10-30 23:05:48 +07:00
|
|
|
return writeback_inodes_sb_nr(sb, get_nr_dirty_pages());
|
2010-10-29 22:16:17 +07:00
|
|
|
}
|
2010-06-01 16:08:43 +07:00
|
|
|
EXPORT_SYMBOL(writeback_inodes_sb);
|
2010-05-17 17:55:07 +07:00
|
|
|
|
2009-12-23 19:57:07 +07:00
|
|
|
/**
|
|
|
|
* writeback_inodes_sb_if_idle - start writeback if none underway
|
|
|
|
* @sb: the superblock
|
|
|
|
*
|
|
|
|
* Invoke writeback_inodes_sb if no writeback is currently underway.
|
|
|
|
* Returns 1 if writeback was started, 0 if not.
|
|
|
|
*/
|
|
|
|
int writeback_inodes_sb_if_idle(struct super_block *sb)
|
|
|
|
{
|
|
|
|
if (!writeback_in_progress(sb->s_bdi)) {
|
2010-06-08 23:14:51 +07:00
|
|
|
down_read(&sb->s_umount);
|
2009-12-23 19:57:07 +07:00
|
|
|
writeback_inodes_sb(sb);
|
2010-06-08 23:14:51 +07:00
|
|
|
up_read(&sb->s_umount);
|
2009-12-23 19:57:07 +07:00
|
|
|
return 1;
|
|
|
|
} else
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(writeback_inodes_sb_if_idle);
|
|
|
|
|
2010-10-29 22:16:17 +07:00
|
|
|
/**
|
|
|
|
* writeback_inodes_sb_if_idle - start writeback if none underway
|
|
|
|
* @sb: the superblock
|
|
|
|
* @nr: the number of pages to write
|
|
|
|
*
|
|
|
|
* Invoke writeback_inodes_sb if no writeback is currently underway.
|
|
|
|
* Returns 1 if writeback was started, 0 if not.
|
|
|
|
*/
|
|
|
|
int writeback_inodes_sb_nr_if_idle(struct super_block *sb,
|
|
|
|
unsigned long nr)
|
|
|
|
{
|
|
|
|
if (!writeback_in_progress(sb->s_bdi)) {
|
|
|
|
down_read(&sb->s_umount);
|
|
|
|
writeback_inodes_sb_nr(sb, nr);
|
|
|
|
up_read(&sb->s_umount);
|
|
|
|
return 1;
|
|
|
|
} else
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(writeback_inodes_sb_nr_if_idle);
|
|
|
|
|
2009-09-02 17:34:32 +07:00
|
|
|
/**
|
|
|
|
* sync_inodes_sb - sync sb inode pages
|
|
|
|
* @sb: the superblock
|
|
|
|
*
|
|
|
|
* This function writes and waits on any dirty inode belonging to this
|
2011-01-14 06:47:26 +07:00
|
|
|
* super_block.
|
2009-09-02 17:34:32 +07:00
|
|
|
*/
|
2009-09-16 20:13:54 +07:00
|
|
|
void sync_inodes_sb(struct super_block *sb)
|
2009-09-02 17:34:32 +07:00
|
|
|
{
|
2010-07-06 13:59:53 +07:00
|
|
|
DECLARE_COMPLETION_ONSTACK(done);
|
|
|
|
struct wb_writeback_work work = {
|
2010-06-08 23:14:43 +07:00
|
|
|
.sb = sb,
|
|
|
|
.sync_mode = WB_SYNC_ALL,
|
|
|
|
.nr_pages = LONG_MAX,
|
|
|
|
.range_cyclic = 0,
|
2010-07-06 13:59:53 +07:00
|
|
|
.done = &done,
|
2010-06-08 23:14:43 +07:00
|
|
|
};
|
|
|
|
|
2010-06-08 23:14:51 +07:00
|
|
|
WARN_ON(!rwsem_is_locked(&sb->s_umount));
|
|
|
|
|
2010-07-06 13:59:53 +07:00
|
|
|
bdi_queue_work(sb->s_bdi, &work);
|
|
|
|
wait_for_completion(&done);
|
|
|
|
|
2009-09-16 20:13:54 +07:00
|
|
|
wait_sb_inodes(sb);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2009-09-02 17:34:32 +07:00
|
|
|
EXPORT_SYMBOL(sync_inodes_sb);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
/**
|
[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>
2005-10-31 06:03:05 +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.
|
2005-04-17 05:20:36 +07:00
|
|
|
*
|
[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>
2005-10-31 06:03:05 +07:00
|
|
|
* The caller must either have a ref on the inode or must have set I_WILL_FREE.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
|
|
|
int write_inode_now(struct inode *inode, int sync)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
struct writeback_control wbc = {
|
|
|
|
.nr_to_write = LONG_MAX,
|
2008-02-08 19:20:23 +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 16:03:26 +07:00
|
|
|
.range_start = 0,
|
|
|
|
.range_end = LLONG_MAX,
|
2005-04-17 05:20:36 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
if (!mapping_cap_writeback_dirty(inode->i_mapping))
|
2005-11-07 15:59:15 +07:00
|
|
|
wbc.nr_to_write = 0;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
might_sleep();
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
2011-03-22 18:23:43 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2009-06-08 18:35:40 +07:00
|
|
|
ret = writeback_single_inode(inode, &wbc);
|
2011-03-22 18:23:43 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (sync)
|
2007-10-17 13:30:44 +07:00
|
|
|
inode_sync_wait(inode);
|
2005-04-17 05:20:36 +07:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
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)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_lock(&inode_wb_list_lock);
|
2011-03-22 18:23:43 +07:00
|
|
|
spin_lock(&inode->i_lock);
|
2009-06-08 18:35:40 +07:00
|
|
|
ret = writeback_single_inode(inode, wbc);
|
2011-03-22 18:23:43 +07:00
|
|
|
spin_unlock(&inode->i_lock);
|
2011-03-22 18:23:41 +07:00
|
|
|
spin_unlock(&inode_wb_list_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(sync_inode);
|
2010-10-06 15:48:20 +07:00
|
|
|
|
|
|
|
/**
|
2011-01-14 06:45:48 +07:00
|
|
|
* sync_inode_metadata - write an inode to disk
|
2010-10-06 15:48:20 +07:00
|
|
|
* @inode: the inode to sync
|
|
|
|
* @wait: wait for I/O to complete.
|
|
|
|
*
|
2011-01-14 06:45:48 +07:00
|
|
|
* Write an inode to disk and adjust its dirty state after completion.
|
2010-10-06 15:48:20 +07:00
|
|
|
*
|
|
|
|
* 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);
|