linux/fs/ext4/page-io.c

495 lines
13 KiB
C
Raw Normal View History

/*
* linux/fs/ext4/page-io.c
*
* This contains the new page_io functions for ext4
*
* Written by Theodore Ts'o, 2010.
*/
#include <linux/fs.h>
#include <linux/time.h>
#include <linux/jbd2.h>
#include <linux/highuid.h>
#include <linux/pagemap.h>
#include <linux/quotaops.h>
#include <linux/string.h>
#include <linux/buffer_head.h>
#include <linux/writeback.h>
#include <linux/pagevec.h>
#include <linux/mpage.h>
#include <linux/namei.h>
#include <linux/aio.h>
#include <linux/uio.h>
#include <linux/bio.h>
#include <linux/workqueue.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/mm.h>
#include <linux/ratelimit.h>
#include "ext4_jbd2.h"
#include "xattr.h"
#include "acl.h"
static struct kmem_cache *io_end_cachep;
int __init ext4_init_pageio(void)
{
io_end_cachep = KMEM_CACHE(ext4_io_end, SLAB_RECLAIM_ACCOUNT);
if (io_end_cachep == NULL)
return -ENOMEM;
return 0;
}
void ext4_exit_pageio(void)
{
kmem_cache_destroy(io_end_cachep);
}
/*
* Print an buffer I/O error compatible with the fs/buffer.c. This
* provides compatibility with dmesg scrapers that look for a specific
* buffer I/O error message. We really need a unified error reporting
* structure to userspace ala Digital Unix's uerf system, but it's
* probably not going to happen in my lifetime, due to LKML politics...
*/
static void buffer_io_error(struct buffer_head *bh)
{
char b[BDEVNAME_SIZE];
printk_ratelimited(KERN_ERR "Buffer I/O error on device %s, logical block %llu\n",
bdevname(bh->b_bdev, b),
(unsigned long long)bh->b_blocknr);
}
static void ext4_finish_bio(struct bio *bio)
{
int i;
int error = !test_bit(BIO_UPTODATE, &bio->bi_flags);
struct bio_vec *bvec;
bio_for_each_segment_all(bvec, bio, i) {
struct page *page = bvec->bv_page;
struct buffer_head *bh, *head;
unsigned bio_start = bvec->bv_offset;
unsigned bio_end = bio_start + bvec->bv_len;
unsigned under_io = 0;
unsigned long flags;
if (!page)
continue;
if (error) {
SetPageError(page);
set_bit(AS_EIO, &page->mapping->flags);
}
bh = head = page_buffers(page);
/*
* We check all buffers in the page under BH_Uptodate_Lock
* to avoid races with other end io clearing async_write flags
*/
local_irq_save(flags);
bit_spin_lock(BH_Uptodate_Lock, &head->b_state);
do {
if (bh_offset(bh) < bio_start ||
bh_offset(bh) + bh->b_size > bio_end) {
if (buffer_async_write(bh))
under_io++;
continue;
}
clear_buffer_async_write(bh);
if (error)
buffer_io_error(bh);
} while ((bh = bh->b_this_page) != head);
bit_spin_unlock(BH_Uptodate_Lock, &head->b_state);
local_irq_restore(flags);
if (!under_io)
end_page_writeback(page);
}
}
static void ext4_release_io_end(ext4_io_end_t *io_end)
{
struct bio *bio, *next_bio;
BUG_ON(!list_empty(&io_end->list));
BUG_ON(io_end->flag & EXT4_IO_END_UNWRITTEN);
WARN_ON(io_end->handle);
if (atomic_dec_and_test(&EXT4_I(io_end->inode)->i_ioend_count))
wake_up_all(ext4_ioend_wq(io_end->inode));
for (bio = io_end->bio; bio; bio = next_bio) {
next_bio = bio->bi_private;
ext4_finish_bio(bio);
bio_put(bio);
}
kmem_cache_free(io_end_cachep, io_end);
}
static void ext4_clear_io_unwritten_flag(ext4_io_end_t *io_end)
{
struct inode *inode = io_end->inode;
io_end->flag &= ~EXT4_IO_END_UNWRITTEN;
/* Wake up anyone waiting on unwritten extent conversion */
if (atomic_dec_and_test(&EXT4_I(inode)->i_unwritten))
wake_up_all(ext4_ioend_wq(inode));
}
/*
* Check a range of space and convert unwritten extents to written. Note that
* we are protected from truncate touching same part of extent tree by the
* fact that truncate code waits for all DIO to finish (thus exclusion from
* direct IO is achieved) and also waits for PageWriteback bits. Thus we
* cannot get to ext4_ext_truncate() before all IOs overlapping that range are
* completed (happens from ext4_free_ioend()).
*/
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
static int ext4_end_io(ext4_io_end_t *io)
{
struct inode *inode = io->inode;
loff_t offset = io->offset;
ssize_t size = io->size;
handle_t *handle = io->handle;
int ret = 0;
ext4_debug("ext4_end_io_nolock: io 0x%p from inode %lu,list->next 0x%p,"
"list->prev 0x%p\n",
io, inode->i_ino, io->list.next, io->list.prev);
io->handle = NULL; /* Following call will use up the handle */
ret = ext4_convert_unwritten_extents(handle, inode, offset, size);
if (ret < 0) {
ext4_msg(inode->i_sb, KERN_EMERG,
"failed to convert unwritten extents to written "
"extents -- potential data loss! "
"(inode %lu, offset %llu, size %zd, error %d)",
inode->i_ino, offset, size, ret);
}
ext4_clear_io_unwritten_flag(io);
ext4_release_io_end(io);
return ret;
}
static void dump_completed_IO(struct inode *inode, struct list_head *head)
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
{
#ifdef EXT4FS_DEBUG
struct list_head *cur, *before, *after;
ext4_io_end_t *io, *io0, *io1;
if (list_empty(head))
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
return;
ext4_debug("Dump inode %lu completed io list\n", inode->i_ino);
list_for_each_entry(io, head, list) {
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
cur = &io->list;
before = cur->prev;
io0 = container_of(before, ext4_io_end_t, list);
after = cur->next;
io1 = container_of(after, ext4_io_end_t, list);
ext4_debug("io 0x%p from inode %lu,prev 0x%p,next 0x%p\n",
io, inode->i_ino, io0, io1);
}
#endif
}
/* Add the io_end to per-inode completed end_io list. */
static void ext4_add_complete_io(ext4_io_end_t *io_end)
{
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
struct ext4_inode_info *ei = EXT4_I(io_end->inode);
struct ext4_sb_info *sbi = EXT4_SB(io_end->inode->i_sb);
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
struct workqueue_struct *wq;
unsigned long flags;
/* Only reserved conversions from writeback should enter here */
WARN_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN));
WARN_ON(!io_end->handle && sbi->s_journal);
spin_lock_irqsave(&ei->i_completed_io_lock, flags);
wq = sbi->rsv_conversion_wq;
if (list_empty(&ei->i_rsv_conversion_list))
queue_work(wq, &ei->i_rsv_conversion_work);
list_add_tail(&io_end->list, &ei->i_rsv_conversion_list);
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
}
static int ext4_do_flush_completed_IO(struct inode *inode,
struct list_head *head)
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
{
ext4_io_end_t *io;
struct list_head unwritten;
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
unsigned long flags;
struct ext4_inode_info *ei = EXT4_I(inode);
int err, ret = 0;
spin_lock_irqsave(&ei->i_completed_io_lock, flags);
dump_completed_IO(inode, head);
list_replace_init(head, &unwritten);
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
while (!list_empty(&unwritten)) {
io = list_entry(unwritten.next, ext4_io_end_t, list);
BUG_ON(!(io->flag & EXT4_IO_END_UNWRITTEN));
list_del_init(&io->list);
err = ext4_end_io(io);
if (unlikely(!ret && err))
ret = err;
}
return ret;
}
/*
* work on completed IO, to convert unwritten extents to extents
ext4: completed_io locking cleanup Current unwritten extent conversion state-machine is very fuzzy. - For unknown reason it performs conversion under i_mutex. What for? My diagnosis: We already protect extent tree with i_data_sem, truncate and punch_hole should wait for DIO, so the only data we have to protect is end_io->flags modification, but only flush_completed_IO and end_io_work modified this flags and we can serialize them via i_completed_io_lock. Currently all these games with mutex_trylock result in the following deadlock truncate: kworker: ext4_setattr ext4_end_io_work mutex_lock(i_mutex) inode_dio_wait(inode) ->BLOCK DEADLOCK<- mutex_trylock() inode_dio_done() #TEST_CASE1_BEGIN MNT=/mnt_scrach unlink $MNT/file fallocate -l $((1024*1024*1024)) $MNT/file aio-stress -I 100000 -O -s 100m -n -t 1 -c 10 -o 2 -o 3 $MNT/file sleep 2 truncate -s 0 $MNT/file #TEST_CASE1_END Or use 286's xfstests https://github.com/dmonakhov/xfstests/blob/devel/286 This patch makes state machine simple and clean: (1) xxx_end_io schedule final extent conversion simply by calling ext4_add_complete_io(), which append it to ei->i_completed_io_list NOTE1: because of (2A) work should be queued only if ->i_completed_io_list was empty, otherwise the work is scheduled already. (2) ext4_flush_completed_IO is responsible for handling all pending end_io from ei->i_completed_io_list Flushing sequence consists of following stages: A) LOCKED: Atomically drain completed_io_list to local_list B) Perform extents conversion C) LOCKED: move converted io's to to_free list for final deletion This logic depends on context which we was called from. D) Final end_io context destruction NOTE1: i_mutex is no longer required because end_io->flags modification is protected by ei->ext4_complete_io_lock Full list of changes: - Move all completion end_io related routines to page-io.c in order to improve logic locality - Move open coded logic from various xx_end_xx routines to ext4_add_complete_io() - remove EXT4_IO_END_FSYNC - Improve SMP scalability by removing useless i_mutex which does not protect io->flags anymore. - Reduce lock contention on i_completed_io_lock by optimizing list walk. - Rename ext4_end_io_nolock to end4_end_io and make it static - Check flush completion status to ext4_ext_punch_hole(). Because it is not good idea to punch blocks from corrupted inode. Changes since V3 (in request to Jan's comments): Fall back to active flush_completed_IO() approach in order to prevent performance issues with nolocked DIO reads. Changes since V2: Fix use-after-free caused by race truncate vs end_io_work Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-29 04:14:55 +00:00
*/
void ext4_end_io_rsv_work(struct work_struct *work)
{
struct ext4_inode_info *ei = container_of(work, struct ext4_inode_info,
i_rsv_conversion_work);
ext4_do_flush_completed_IO(&ei->vfs_inode, &ei->i_rsv_conversion_list);
}
ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags)
{
ext4_io_end_t *io = kmem_cache_zalloc(io_end_cachep, flags);
if (io) {
ext4: handle writeback of inodes which are being freed The following BUG can occur when an inode which is getting freed when it still has dirty pages outstanding, and it gets deleted (in this because it was the target of a rename). In ordered mode, we need to make sure the data pages are written just in case we crash before the rename (or unlink) is committed. If the inode is being freed then when we try to igrab the inode, we end up tripping the BUG_ON at fs/ext4/page-io.c:146. To solve this problem, we need to keep track of the number of io callbacks which are pending, and avoid destroying the inode until they have all been completed. That way we don't have to bump the inode count to keep the inode from being destroyed; an approach which doesn't work because the count could have already been dropped down to zero before the inode writeback has started (at which point we're not allowed to bump the count back up to 1, since it's already started getting freed). Thanks to Dave Chinner for suggesting this approach, which is also used by XFS. kernel BUG at /scratch_space/linux-2.6/fs/ext4/page-io.c:146! Call Trace: [<ffffffff811075b1>] ext4_bio_write_page+0x172/0x307 [<ffffffff811033a7>] mpage_da_submit_io+0x2f9/0x37b [<ffffffff811068d7>] mpage_da_map_and_submit+0x2cc/0x2e2 [<ffffffff811069b3>] mpage_add_bh_to_extent+0xc6/0xd5 [<ffffffff81106c66>] write_cache_pages_da+0x2a4/0x3ac [<ffffffff81107044>] ext4_da_writepages+0x2d6/0x44d [<ffffffff81087910>] do_writepages+0x1c/0x25 [<ffffffff810810a4>] __filemap_fdatawrite_range+0x4b/0x4d [<ffffffff810815f5>] filemap_fdatawrite_range+0xe/0x10 [<ffffffff81122a2e>] jbd2_journal_begin_ordered_truncate+0x7b/0xa2 [<ffffffff8110615d>] ext4_evict_inode+0x57/0x24c [<ffffffff810c14a3>] evict+0x22/0x92 [<ffffffff810c1a3d>] iput+0x212/0x249 [<ffffffff810bdf16>] dentry_iput+0xa1/0xb9 [<ffffffff810bdf6b>] d_kill+0x3d/0x5d [<ffffffff810be613>] dput+0x13a/0x147 [<ffffffff810b990d>] sys_renameat+0x1b5/0x258 [<ffffffff81145f71>] ? _atomic_dec_and_lock+0x2d/0x4c [<ffffffff810b2950>] ? cp_new_stat+0xde/0xea [<ffffffff810b29c1>] ? sys_newlstat+0x2d/0x38 [<ffffffff810b99c6>] sys_rename+0x16/0x18 [<ffffffff81002a2b>] system_call_fastpath+0x16/0x1b Reported-by: Nick Bowler <nbowler@elliptictech.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Tested-by: Nick Bowler <nbowler@elliptictech.com>
2010-11-08 18:43:33 +00:00
atomic_inc(&EXT4_I(inode)->i_ioend_count);
io->inode = inode;
INIT_LIST_HEAD(&io->list);
atomic_set(&io->count, 1);
}
return io;
}
void ext4_put_io_end_defer(ext4_io_end_t *io_end)
{
if (atomic_dec_and_test(&io_end->count)) {
if (!(io_end->flag & EXT4_IO_END_UNWRITTEN) || !io_end->size) {
ext4_release_io_end(io_end);
return;
}
ext4_add_complete_io(io_end);
}
}
int ext4_put_io_end(ext4_io_end_t *io_end)
{
int err = 0;
if (atomic_dec_and_test(&io_end->count)) {
if (io_end->flag & EXT4_IO_END_UNWRITTEN) {
err = ext4_convert_unwritten_extents(io_end->handle,
io_end->inode, io_end->offset,
io_end->size);
io_end->handle = NULL;
ext4_clear_io_unwritten_flag(io_end);
}
ext4_release_io_end(io_end);
}
return err;
}
ext4_io_end_t *ext4_get_io_end(ext4_io_end_t *io_end)
{
atomic_inc(&io_end->count);
return io_end;
}
/* BIO completion function for page writeback */
static void ext4_end_bio(struct bio *bio, int error)
{
ext4_io_end_t *io_end = bio->bi_private;
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-11 22:44:27 +00:00
sector_t bi_sector = bio->bi_iter.bi_sector;
BUG_ON(!io_end);
bio->bi_end_io = NULL;
if (test_bit(BIO_UPTODATE, &bio->bi_flags))
error = 0;
ext4: handle writeback of inodes which are being freed The following BUG can occur when an inode which is getting freed when it still has dirty pages outstanding, and it gets deleted (in this because it was the target of a rename). In ordered mode, we need to make sure the data pages are written just in case we crash before the rename (or unlink) is committed. If the inode is being freed then when we try to igrab the inode, we end up tripping the BUG_ON at fs/ext4/page-io.c:146. To solve this problem, we need to keep track of the number of io callbacks which are pending, and avoid destroying the inode until they have all been completed. That way we don't have to bump the inode count to keep the inode from being destroyed; an approach which doesn't work because the count could have already been dropped down to zero before the inode writeback has started (at which point we're not allowed to bump the count back up to 1, since it's already started getting freed). Thanks to Dave Chinner for suggesting this approach, which is also used by XFS. kernel BUG at /scratch_space/linux-2.6/fs/ext4/page-io.c:146! Call Trace: [<ffffffff811075b1>] ext4_bio_write_page+0x172/0x307 [<ffffffff811033a7>] mpage_da_submit_io+0x2f9/0x37b [<ffffffff811068d7>] mpage_da_map_and_submit+0x2cc/0x2e2 [<ffffffff811069b3>] mpage_add_bh_to_extent+0xc6/0xd5 [<ffffffff81106c66>] write_cache_pages_da+0x2a4/0x3ac [<ffffffff81107044>] ext4_da_writepages+0x2d6/0x44d [<ffffffff81087910>] do_writepages+0x1c/0x25 [<ffffffff810810a4>] __filemap_fdatawrite_range+0x4b/0x4d [<ffffffff810815f5>] filemap_fdatawrite_range+0xe/0x10 [<ffffffff81122a2e>] jbd2_journal_begin_ordered_truncate+0x7b/0xa2 [<ffffffff8110615d>] ext4_evict_inode+0x57/0x24c [<ffffffff810c14a3>] evict+0x22/0x92 [<ffffffff810c1a3d>] iput+0x212/0x249 [<ffffffff810bdf16>] dentry_iput+0xa1/0xb9 [<ffffffff810bdf6b>] d_kill+0x3d/0x5d [<ffffffff810be613>] dput+0x13a/0x147 [<ffffffff810b990d>] sys_renameat+0x1b5/0x258 [<ffffffff81145f71>] ? _atomic_dec_and_lock+0x2d/0x4c [<ffffffff810b2950>] ? cp_new_stat+0xde/0xea [<ffffffff810b29c1>] ? sys_newlstat+0x2d/0x38 [<ffffffff810b99c6>] sys_rename+0x16/0x18 [<ffffffff81002a2b>] system_call_fastpath+0x16/0x1b Reported-by: Nick Bowler <nbowler@elliptictech.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Tested-by: Nick Bowler <nbowler@elliptictech.com>
2010-11-08 18:43:33 +00:00
if (error) {
struct inode *inode = io_end->inode;
ext4: handle writeback of inodes which are being freed The following BUG can occur when an inode which is getting freed when it still has dirty pages outstanding, and it gets deleted (in this because it was the target of a rename). In ordered mode, we need to make sure the data pages are written just in case we crash before the rename (or unlink) is committed. If the inode is being freed then when we try to igrab the inode, we end up tripping the BUG_ON at fs/ext4/page-io.c:146. To solve this problem, we need to keep track of the number of io callbacks which are pending, and avoid destroying the inode until they have all been completed. That way we don't have to bump the inode count to keep the inode from being destroyed; an approach which doesn't work because the count could have already been dropped down to zero before the inode writeback has started (at which point we're not allowed to bump the count back up to 1, since it's already started getting freed). Thanks to Dave Chinner for suggesting this approach, which is also used by XFS. kernel BUG at /scratch_space/linux-2.6/fs/ext4/page-io.c:146! Call Trace: [<ffffffff811075b1>] ext4_bio_write_page+0x172/0x307 [<ffffffff811033a7>] mpage_da_submit_io+0x2f9/0x37b [<ffffffff811068d7>] mpage_da_map_and_submit+0x2cc/0x2e2 [<ffffffff811069b3>] mpage_add_bh_to_extent+0xc6/0xd5 [<ffffffff81106c66>] write_cache_pages_da+0x2a4/0x3ac [<ffffffff81107044>] ext4_da_writepages+0x2d6/0x44d [<ffffffff81087910>] do_writepages+0x1c/0x25 [<ffffffff810810a4>] __filemap_fdatawrite_range+0x4b/0x4d [<ffffffff810815f5>] filemap_fdatawrite_range+0xe/0x10 [<ffffffff81122a2e>] jbd2_journal_begin_ordered_truncate+0x7b/0xa2 [<ffffffff8110615d>] ext4_evict_inode+0x57/0x24c [<ffffffff810c14a3>] evict+0x22/0x92 [<ffffffff810c1a3d>] iput+0x212/0x249 [<ffffffff810bdf16>] dentry_iput+0xa1/0xb9 [<ffffffff810bdf6b>] d_kill+0x3d/0x5d [<ffffffff810be613>] dput+0x13a/0x147 [<ffffffff810b990d>] sys_renameat+0x1b5/0x258 [<ffffffff81145f71>] ? _atomic_dec_and_lock+0x2d/0x4c [<ffffffff810b2950>] ? cp_new_stat+0xde/0xea [<ffffffff810b29c1>] ? sys_newlstat+0x2d/0x38 [<ffffffff810b99c6>] sys_rename+0x16/0x18 [<ffffffff81002a2b>] system_call_fastpath+0x16/0x1b Reported-by: Nick Bowler <nbowler@elliptictech.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Tested-by: Nick Bowler <nbowler@elliptictech.com>
2010-11-08 18:43:33 +00:00
ext4_warning(inode->i_sb, "I/O error writing to inode %lu "
"(offset %llu size %ld starting block %llu)",
inode->i_ino,
(unsigned long long) io_end->offset,
(long) io_end->size,
(unsigned long long)
2011-02-07 17:46:14 +00:00
bi_sector >> (inode->i_blkbits - 9));
ext4: handle writeback of inodes which are being freed The following BUG can occur when an inode which is getting freed when it still has dirty pages outstanding, and it gets deleted (in this because it was the target of a rename). In ordered mode, we need to make sure the data pages are written just in case we crash before the rename (or unlink) is committed. If the inode is being freed then when we try to igrab the inode, we end up tripping the BUG_ON at fs/ext4/page-io.c:146. To solve this problem, we need to keep track of the number of io callbacks which are pending, and avoid destroying the inode until they have all been completed. That way we don't have to bump the inode count to keep the inode from being destroyed; an approach which doesn't work because the count could have already been dropped down to zero before the inode writeback has started (at which point we're not allowed to bump the count back up to 1, since it's already started getting freed). Thanks to Dave Chinner for suggesting this approach, which is also used by XFS. kernel BUG at /scratch_space/linux-2.6/fs/ext4/page-io.c:146! Call Trace: [<ffffffff811075b1>] ext4_bio_write_page+0x172/0x307 [<ffffffff811033a7>] mpage_da_submit_io+0x2f9/0x37b [<ffffffff811068d7>] mpage_da_map_and_submit+0x2cc/0x2e2 [<ffffffff811069b3>] mpage_add_bh_to_extent+0xc6/0xd5 [<ffffffff81106c66>] write_cache_pages_da+0x2a4/0x3ac [<ffffffff81107044>] ext4_da_writepages+0x2d6/0x44d [<ffffffff81087910>] do_writepages+0x1c/0x25 [<ffffffff810810a4>] __filemap_fdatawrite_range+0x4b/0x4d [<ffffffff810815f5>] filemap_fdatawrite_range+0xe/0x10 [<ffffffff81122a2e>] jbd2_journal_begin_ordered_truncate+0x7b/0xa2 [<ffffffff8110615d>] ext4_evict_inode+0x57/0x24c [<ffffffff810c14a3>] evict+0x22/0x92 [<ffffffff810c1a3d>] iput+0x212/0x249 [<ffffffff810bdf16>] dentry_iput+0xa1/0xb9 [<ffffffff810bdf6b>] d_kill+0x3d/0x5d [<ffffffff810be613>] dput+0x13a/0x147 [<ffffffff810b990d>] sys_renameat+0x1b5/0x258 [<ffffffff81145f71>] ? _atomic_dec_and_lock+0x2d/0x4c [<ffffffff810b2950>] ? cp_new_stat+0xde/0xea [<ffffffff810b29c1>] ? sys_newlstat+0x2d/0x38 [<ffffffff810b99c6>] sys_rename+0x16/0x18 [<ffffffff81002a2b>] system_call_fastpath+0x16/0x1b Reported-by: Nick Bowler <nbowler@elliptictech.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Tested-by: Nick Bowler <nbowler@elliptictech.com>
2010-11-08 18:43:33 +00:00
}
if (io_end->flag & EXT4_IO_END_UNWRITTEN) {
/*
* Link bio into list hanging from io_end. We have to do it
* atomically as bio completions can be racing against each
* other.
*/
bio->bi_private = xchg(&io_end->bio, bio);
ext4_put_io_end_defer(io_end);
} else {
/*
* Drop io_end reference early. Inode can get freed once
* we finish the bio.
*/
ext4_put_io_end_defer(io_end);
ext4_finish_bio(bio);
bio_put(bio);
}
}
void ext4_io_submit(struct ext4_io_submit *io)
{
struct bio *bio = io->io_bio;
if (bio) {
bio_get(io->io_bio);
submit_bio(io->io_op, io->io_bio);
BUG_ON(bio_flagged(io->io_bio, BIO_EOPNOTSUPP));
bio_put(io->io_bio);
}
io->io_bio = NULL;
}
void ext4_io_submit_init(struct ext4_io_submit *io,
struct writeback_control *wbc)
{
io->io_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE);
io->io_bio = NULL;
io->io_end = NULL;
}
static int io_submit_init_bio(struct ext4_io_submit *io,
struct buffer_head *bh)
{
int nvecs = bio_get_nr_vecs(bh->b_bdev);
struct bio *bio;
bio = bio_alloc(GFP_NOIO, min(nvecs, BIO_MAX_PAGES));
if (!bio)
return -ENOMEM;
block: Abstract out bvec iterator Immutable biovecs are going to require an explicit iterator. To implement immutable bvecs, a later patch is going to add a bi_bvec_done member to this struct; for now, this patch effectively just renames things. Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: "Ed L. Cashin" <ecashin@coraid.com> Cc: Nick Piggin <npiggin@kernel.dk> Cc: Lars Ellenberg <drbd-dev@lists.linbit.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Yehuda Sadeh <yehuda@inktank.com> Cc: Sage Weil <sage@inktank.com> Cc: Alex Elder <elder@inktank.com> Cc: ceph-devel@vger.kernel.org Cc: Joshua Morris <josh.h.morris@us.ibm.com> Cc: Philip Kelleher <pjk1939@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Cc: Neil Brown <neilb@suse.de> Cc: Alasdair Kergon <agk@redhat.com> Cc: Mike Snitzer <snitzer@redhat.com> Cc: dm-devel@redhat.com Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Cc: Boaz Harrosh <bharrosh@panasas.com> Cc: Benny Halevy <bhalevy@tonian.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Chris Mason <chris.mason@fusionio.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Jaegeuk Kim <jaegeuk.kim@samsung.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@kernel.org> Cc: Joern Engel <joern@logfs.org> Cc: Prasad Joshi <prasadjoshi.linux@gmail.com> Cc: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Ben Myers <bpm@sgi.com> Cc: xfs@oss.sgi.com Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Len Brown <len.brown@intel.com> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Herton Ronaldo Krzesinski <herton.krzesinski@canonical.com> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Guo Chao <yan@linux.vnet.ibm.com> Cc: Tejun Heo <tj@kernel.org> Cc: Asai Thambi S P <asamymuthupa@micron.com> Cc: Selvan Mani <smani@micron.com> Cc: Sam Bradshaw <sbradshaw@micron.com> Cc: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Cc: "Roger Pau Monné" <roger.pau@citrix.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Jiang Liu <jiang.liu@huawei.com> Cc: Nitin Gupta <ngupta@vflare.org> Cc: Jerome Marchand <jmarchand@redhat.com> Cc: Joe Perches <joe@perches.com> Cc: Peng Tao <tao.peng@emc.com> Cc: Andy Adamson <andros@netapp.com> Cc: fanchaoting <fanchaoting@cn.fujitsu.com> Cc: Jie Liu <jeff.liu@oracle.com> Cc: Sunil Mushran <sunil.mushran@gmail.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Namjae Jeon <namjae.jeon@samsung.com> Cc: Pankaj Kumar <pankaj.km@samsung.com> Cc: Dan Magenheimer <dan.magenheimer@oracle.com> Cc: Mel Gorman <mgorman@suse.de>6
2013-10-11 22:44:27 +00:00
bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
bio->bi_bdev = bh->b_bdev;
bio->bi_end_io = ext4_end_bio;
bio->bi_private = ext4_get_io_end(io->io_end);
io->io_bio = bio;
io->io_next_block = bh->b_blocknr;
return 0;
}
static int io_submit_add_bh(struct ext4_io_submit *io,
struct inode *inode,
struct buffer_head *bh)
{
int ret;
if (io->io_bio && bh->b_blocknr != io->io_next_block) {
submit_and_retry:
ext4_io_submit(io);
}
if (io->io_bio == NULL) {
ret = io_submit_init_bio(io, bh);
if (ret)
return ret;
}
ret = bio_add_page(io->io_bio, bh->b_page, bh->b_size, bh_offset(bh));
if (ret != bh->b_size)
goto submit_and_retry;
io->io_next_block++;
return 0;
}
int ext4_bio_write_page(struct ext4_io_submit *io,
struct page *page,
int len,
struct writeback_control *wbc)
{
struct inode *inode = page->mapping->host;
unsigned block_start, blocksize;
struct buffer_head *bh, *head;
int ret = 0;
int nr_submitted = 0;
blocksize = 1 << inode->i_blkbits;
2011-02-07 17:46:14 +00:00
BUG_ON(!PageLocked(page));
BUG_ON(PageWriteback(page));
set_page_writeback(page);
ClearPageError(page);
/*
* In the first loop we prepare and mark buffers to submit. We have to
* mark all buffers in the page before submitting so that
* end_page_writeback() cannot be called from ext4_bio_end_io() when IO
* on the first buffer finishes and we are still working on submitting
* the second buffer.
*/
bh = head = page_buffers(page);
do {
block_start = bh_offset(bh);
if (block_start >= len) {
/*
* Comments copied from block_write_full_page_endio:
*
* The page straddles i_size. It must be zeroed out on
* each and every writepage invocation because it may
* be mmapped. "A file is mapped in multiples of the
* page size. For a file that is not a multiple of
* the page size, the remaining memory is zeroed when
* mapped, and writes to that region are not written
* out to the file."
*/
zero_user_segment(page, block_start,
block_start + blocksize);
clear_buffer_dirty(bh);
set_buffer_uptodate(bh);
continue;
}
if (!buffer_dirty(bh) || buffer_delay(bh) ||
!buffer_mapped(bh) || buffer_unwritten(bh)) {
/* A hole? We can safely clear the dirty bit */
if (!buffer_mapped(bh))
clear_buffer_dirty(bh);
if (io->io_bio)
ext4_io_submit(io);
continue;
}
if (buffer_new(bh)) {
clear_buffer_new(bh);
unmap_underlying_metadata(bh->b_bdev, bh->b_blocknr);
}
set_buffer_async_write(bh);
} while ((bh = bh->b_this_page) != head);
/* Now submit buffers to write */
bh = head = page_buffers(page);
do {
if (!buffer_async_write(bh))
continue;
ret = io_submit_add_bh(io, inode, bh);
if (ret) {
/*
* We only get here on ENOMEM. Not much else
* we can do but mark the page as dirty, and
* better luck next time.
*/
redirty_page_for_writepage(wbc, page);
break;
}
nr_submitted++;
clear_buffer_dirty(bh);
} while ((bh = bh->b_this_page) != head);
/* Error stopped previous loop? Clean up buffers... */
if (ret) {
do {
clear_buffer_async_write(bh);
bh = bh->b_this_page;
} while (bh != head);
}
unlock_page(page);
/* Nothing submitted - we have to end page writeback */
if (!nr_submitted)
end_page_writeback(page);
return ret;
}