linux-sg2042/fs/jbd/commit.c

980 lines
29 KiB
C
Raw Normal View History

/*
* linux/fs/jbd/commit.c
*
* Written by Stephen C. Tweedie <sct@redhat.com>, 1998
*
* Copyright 1998 Red Hat corp --- All Rights Reserved
*
* This file is part of the Linux kernel and is made available under
* the terms of the GNU General Public License, version 2, or at your
* option, any later version, incorporated herein by reference.
*
* Journal commit routines for the generic filesystem journaling code;
* part of the ext2fs journaling system.
*/
#include <linux/time.h>
#include <linux/fs.h>
#include <linux/jbd.h>
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <trace/events/jbd.h>
/*
* Default IO end handler for temporary BJ_IO buffer_heads.
*/
static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate)
{
BUFFER_TRACE(bh, "");
if (uptodate)
set_buffer_uptodate(bh);
else
clear_buffer_uptodate(bh);
unlock_buffer(bh);
}
/*
* When an ext3-ordered file is truncated, it is possible that many pages are
jbd: positively dispose the unmapped data buffers in journal_commit_transaction() After ext3-ordered files are truncated, there is a possibility that the pages which cannot be estimated still remain. Remaining pages can be released when the system has really few memory. So, it is not memory leakage. But the resource management software etc. may not work correctly. It is possible that journal_unmap_buffer() cannot release the buffers, and the pages to which they belong because they are attached to a commiting transaction and journal_unmap_buffer() cannot release them. To release such the buffers and the pages later, journal_unmap_buffer() leaves it to journal_commit_transaction(). (journal_unmap_buffer() puts the mark 'BH_Freed' to the buffers so that journal_commit_transaction() can identify whether they can be released or not.) In the journalled mode and the writeback mode, jbd does with only metadata buffers. But in the ordered mode, jbd does with metadata buffers and also data buffers. Actually, journal_commit_transaction() releases only the metadata buffers of which release is demanded by journal_unmap_buffer(), and also releases the pages to which they belong if possible. As a result, the data buffers of which release is demanded by journal_unmap_buffer() remain after a transaction commits. And also the pages to which they belong remain. Such the remained pages don't have mapping any longer. Due to this fact, there is a possibility that the pages which cannot be estimated remain. The metadata buffers marked 'BH_Freed' and the pages to which they belong can be released at 'JBD: commit phase 7'. Therefore, by applying the same code into 'JBD: commit phase 2' (where the data buffers are done with), journal_commit_transaction() can also release the data buffers marked 'BH_Freed' and the pages to which they belong. As a result, all the buffers marked 'BH_Freed' can be released, and also all the pages to which these buffers belong can be released at journal_commit_transaction(). So, the page which cannot be estimated is lost. <<Excerpt of code at 'JBD: commit phase 7'>> > spin_lock(&journal->j_list_lock); > while (commit_transaction->t_forget) { > transaction_t *cp_transaction; > struct buffer_head *bh; > > jh = commit_transaction->t_forget; >... > if (buffer_freed(bh)) { > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_freed(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_jbddirty(bh); > } > > if (buffer_jbddirty(bh)) { > JBUFFER_TRACE(jh, "add to new checkpointing trans"); > __journal_insert_checkpoint(jh, commit_transaction); > JBUFFER_TRACE(jh, "refile for checkpoint writeback"); > __journal_refile_buffer(jh); > jbd_unlock_bh_state(bh); > } else { > J_ASSERT_BH(bh, !buffer_dirty(bh)); > ... > JBUFFER_TRACE(jh, "refile or unfile freed buffer"); > __journal_refile_buffer(jh); > if (!jh->b_transaction) { > jbd_unlock_bh_state(bh); > /* needs a brelse */ > journal_remove_journal_head(bh); > release_buffer_page(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > } else > } **************************************************************** * Apply the code of "^^^^^^" lines into 'JBD: commit phase 2' * **************************************************************** At journal_commit_transaction() code, there is one extra message in the series of jbd debug messages. ("JBD: commit phase 2") This patch fixes it, too. Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 16:46:29 +08:00
* not successfully freed, because they are attached to a committing transaction.
* After the transaction commits, these pages are left on the LRU, with no
* ->mapping, and with attached buffers. These pages are trivially reclaimable
* by the VM, but their apparent absence upsets the VM accounting, and it makes
* the numbers in /proc/meminfo look odd.
*
* So here, we have a buffer which has just come off the forget list. Look to
* see if we can strip all buffers from the backing page.
*
jbd: positively dispose the unmapped data buffers in journal_commit_transaction() After ext3-ordered files are truncated, there is a possibility that the pages which cannot be estimated still remain. Remaining pages can be released when the system has really few memory. So, it is not memory leakage. But the resource management software etc. may not work correctly. It is possible that journal_unmap_buffer() cannot release the buffers, and the pages to which they belong because they are attached to a commiting transaction and journal_unmap_buffer() cannot release them. To release such the buffers and the pages later, journal_unmap_buffer() leaves it to journal_commit_transaction(). (journal_unmap_buffer() puts the mark 'BH_Freed' to the buffers so that journal_commit_transaction() can identify whether they can be released or not.) In the journalled mode and the writeback mode, jbd does with only metadata buffers. But in the ordered mode, jbd does with metadata buffers and also data buffers. Actually, journal_commit_transaction() releases only the metadata buffers of which release is demanded by journal_unmap_buffer(), and also releases the pages to which they belong if possible. As a result, the data buffers of which release is demanded by journal_unmap_buffer() remain after a transaction commits. And also the pages to which they belong remain. Such the remained pages don't have mapping any longer. Due to this fact, there is a possibility that the pages which cannot be estimated remain. The metadata buffers marked 'BH_Freed' and the pages to which they belong can be released at 'JBD: commit phase 7'. Therefore, by applying the same code into 'JBD: commit phase 2' (where the data buffers are done with), journal_commit_transaction() can also release the data buffers marked 'BH_Freed' and the pages to which they belong. As a result, all the buffers marked 'BH_Freed' can be released, and also all the pages to which these buffers belong can be released at journal_commit_transaction(). So, the page which cannot be estimated is lost. <<Excerpt of code at 'JBD: commit phase 7'>> > spin_lock(&journal->j_list_lock); > while (commit_transaction->t_forget) { > transaction_t *cp_transaction; > struct buffer_head *bh; > > jh = commit_transaction->t_forget; >... > if (buffer_freed(bh)) { > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_freed(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_jbddirty(bh); > } > > if (buffer_jbddirty(bh)) { > JBUFFER_TRACE(jh, "add to new checkpointing trans"); > __journal_insert_checkpoint(jh, commit_transaction); > JBUFFER_TRACE(jh, "refile for checkpoint writeback"); > __journal_refile_buffer(jh); > jbd_unlock_bh_state(bh); > } else { > J_ASSERT_BH(bh, !buffer_dirty(bh)); > ... > JBUFFER_TRACE(jh, "refile or unfile freed buffer"); > __journal_refile_buffer(jh); > if (!jh->b_transaction) { > jbd_unlock_bh_state(bh); > /* needs a brelse */ > journal_remove_journal_head(bh); > release_buffer_page(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > } else > } **************************************************************** * Apply the code of "^^^^^^" lines into 'JBD: commit phase 2' * **************************************************************** At journal_commit_transaction() code, there is one extra message in the series of jbd debug messages. ("JBD: commit phase 2") This patch fixes it, too. Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 16:46:29 +08:00
* Called under journal->j_list_lock. The caller provided us with a ref
* against the buffer, and we drop that here.
*/
static void release_buffer_page(struct buffer_head *bh)
{
struct page *page;
if (buffer_dirty(bh))
goto nope;
if (atomic_read(&bh->b_count) != 1)
goto nope;
page = bh->b_page;
if (!page)
goto nope;
if (page->mapping)
goto nope;
/* OK, it's a truncated page */
if (!trylock_page(page))
goto nope;
page_cache_get(page);
__brelse(bh);
try_to_free_buffers(page);
unlock_page(page);
page_cache_release(page);
return;
nope:
__brelse(bh);
}
jbd: positively dispose the unmapped data buffers in journal_commit_transaction() After ext3-ordered files are truncated, there is a possibility that the pages which cannot be estimated still remain. Remaining pages can be released when the system has really few memory. So, it is not memory leakage. But the resource management software etc. may not work correctly. It is possible that journal_unmap_buffer() cannot release the buffers, and the pages to which they belong because they are attached to a commiting transaction and journal_unmap_buffer() cannot release them. To release such the buffers and the pages later, journal_unmap_buffer() leaves it to journal_commit_transaction(). (journal_unmap_buffer() puts the mark 'BH_Freed' to the buffers so that journal_commit_transaction() can identify whether they can be released or not.) In the journalled mode and the writeback mode, jbd does with only metadata buffers. But in the ordered mode, jbd does with metadata buffers and also data buffers. Actually, journal_commit_transaction() releases only the metadata buffers of which release is demanded by journal_unmap_buffer(), and also releases the pages to which they belong if possible. As a result, the data buffers of which release is demanded by journal_unmap_buffer() remain after a transaction commits. And also the pages to which they belong remain. Such the remained pages don't have mapping any longer. Due to this fact, there is a possibility that the pages which cannot be estimated remain. The metadata buffers marked 'BH_Freed' and the pages to which they belong can be released at 'JBD: commit phase 7'. Therefore, by applying the same code into 'JBD: commit phase 2' (where the data buffers are done with), journal_commit_transaction() can also release the data buffers marked 'BH_Freed' and the pages to which they belong. As a result, all the buffers marked 'BH_Freed' can be released, and also all the pages to which these buffers belong can be released at journal_commit_transaction(). So, the page which cannot be estimated is lost. <<Excerpt of code at 'JBD: commit phase 7'>> > spin_lock(&journal->j_list_lock); > while (commit_transaction->t_forget) { > transaction_t *cp_transaction; > struct buffer_head *bh; > > jh = commit_transaction->t_forget; >... > if (buffer_freed(bh)) { > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_freed(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_jbddirty(bh); > } > > if (buffer_jbddirty(bh)) { > JBUFFER_TRACE(jh, "add to new checkpointing trans"); > __journal_insert_checkpoint(jh, commit_transaction); > JBUFFER_TRACE(jh, "refile for checkpoint writeback"); > __journal_refile_buffer(jh); > jbd_unlock_bh_state(bh); > } else { > J_ASSERT_BH(bh, !buffer_dirty(bh)); > ... > JBUFFER_TRACE(jh, "refile or unfile freed buffer"); > __journal_refile_buffer(jh); > if (!jh->b_transaction) { > jbd_unlock_bh_state(bh); > /* needs a brelse */ > journal_remove_journal_head(bh); > release_buffer_page(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > } else > } **************************************************************** * Apply the code of "^^^^^^" lines into 'JBD: commit phase 2' * **************************************************************** At journal_commit_transaction() code, there is one extra message in the series of jbd debug messages. ("JBD: commit phase 2") This patch fixes it, too. Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 16:46:29 +08:00
/*
* Decrement reference counter for data buffer. If it has been marked
* 'BH_Freed', release it and the page to which it belongs if possible.
*/
static void release_data_buffer(struct buffer_head *bh)
{
if (buffer_freed(bh)) {
clear_buffer_freed(bh);
release_buffer_page(bh);
} else
put_bh(bh);
}
/*
* Try to acquire jbd_lock_bh_state() against the buffer, when j_list_lock is
* held. For ranking reasons we must trylock. If we lose, schedule away and
* return 0. j_list_lock is dropped in this case.
*/
static int inverted_lock(journal_t *journal, struct buffer_head *bh)
{
if (!jbd_trylock_bh_state(bh)) {
spin_unlock(&journal->j_list_lock);
schedule();
return 0;
}
return 1;
}
/* Done it all: now write the commit record. We should have
* cleaned up our previous buffers by now, so if we are in abort
* mode we can now just skip the rest of the journal write
* entirely.
*
* Returns 1 if the journal needs to be aborted or 0 on success
*/
static int journal_write_commit_record(journal_t *journal,
transaction_t *commit_transaction)
{
struct journal_head *descriptor;
struct buffer_head *bh;
journal_header_t *header;
int ret;
if (is_journal_aborted(journal))
return 0;
descriptor = journal_get_descriptor_buffer(journal);
if (!descriptor)
return 1;
bh = jh2bh(descriptor);
header = (journal_header_t *)(bh->b_data);
header->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER);
header->h_blocktype = cpu_to_be32(JFS_COMMIT_BLOCK);
header->h_sequence = cpu_to_be32(commit_transaction->t_tid);
JBUFFER_TRACE(descriptor, "write commit block");
set_buffer_dirty(bh);
if (journal->j_flags & JFS_BARRIER)
ret = __sync_dirty_buffer(bh, WRITE_SYNC | WRITE_FLUSH_FUA);
else
ret = sync_dirty_buffer(bh);
put_bh(bh); /* One for getblk() */
journal_put_journal_head(descriptor);
return (ret == -EIO);
}
static void journal_do_submit_data(struct buffer_head **wbuf, int bufs,
int write_op)
{
int i;
for (i = 0; i < bufs; i++) {
wbuf[i]->b_end_io = end_buffer_write_sync;
/* We use-up our safety reference in submit_bh() */
submit_bh(write_op, wbuf[i]);
}
}
/*
* Submit all the data buffers to disk
*/
static int journal_submit_data_buffers(journal_t *journal,
transaction_t *commit_transaction,
int write_op)
{
struct journal_head *jh;
struct buffer_head *bh;
int locked;
int bufs = 0;
struct buffer_head **wbuf = journal->j_wbuf;
int err = 0;
/*
* Whenever we unlock the journal and sleep, things can get added
* onto ->t_sync_datalist, so we have to keep looping back to
* write_out_data until we *know* that the list is empty.
*
* Cleanup any flushed data buffers from the data list. Even in
* abort mode, we want to flush this out as soon as possible.
*/
write_out_data:
cond_resched();
spin_lock(&journal->j_list_lock);
while (commit_transaction->t_sync_datalist) {
jh = commit_transaction->t_sync_datalist;
bh = jh2bh(jh);
locked = 0;
/* Get reference just to make sure buffer does not disappear
* when we are forced to drop various locks */
get_bh(bh);
/* If the buffer is dirty, we need to submit IO and hence
* we need the buffer lock. We try to lock the buffer without
* blocking. If we fail, we need to drop j_list_lock and do
* blocking lock_buffer().
*/
if (buffer_dirty(bh)) {
if (!trylock_buffer(bh)) {
BUFFER_TRACE(bh, "needs blocking lock");
spin_unlock(&journal->j_list_lock);
trace_jbd_do_submit_data(journal,
commit_transaction);
/* Write out all data to prevent deadlocks */
journal_do_submit_data(wbuf, bufs, write_op);
bufs = 0;
lock_buffer(bh);
spin_lock(&journal->j_list_lock);
}
locked = 1;
}
/* We have to get bh_state lock. Again out of order, sigh. */
if (!inverted_lock(journal, bh)) {
jbd_lock_bh_state(bh);
spin_lock(&journal->j_list_lock);
}
/* Someone already cleaned up the buffer? */
if (!buffer_jbd(bh) || bh2jh(bh) != jh
|| jh->b_transaction != commit_transaction
|| jh->b_jlist != BJ_SyncData) {
jbd_unlock_bh_state(bh);
if (locked)
unlock_buffer(bh);
BUFFER_TRACE(bh, "already cleaned up");
jbd: positively dispose the unmapped data buffers in journal_commit_transaction() After ext3-ordered files are truncated, there is a possibility that the pages which cannot be estimated still remain. Remaining pages can be released when the system has really few memory. So, it is not memory leakage. But the resource management software etc. may not work correctly. It is possible that journal_unmap_buffer() cannot release the buffers, and the pages to which they belong because they are attached to a commiting transaction and journal_unmap_buffer() cannot release them. To release such the buffers and the pages later, journal_unmap_buffer() leaves it to journal_commit_transaction(). (journal_unmap_buffer() puts the mark 'BH_Freed' to the buffers so that journal_commit_transaction() can identify whether they can be released or not.) In the journalled mode and the writeback mode, jbd does with only metadata buffers. But in the ordered mode, jbd does with metadata buffers and also data buffers. Actually, journal_commit_transaction() releases only the metadata buffers of which release is demanded by journal_unmap_buffer(), and also releases the pages to which they belong if possible. As a result, the data buffers of which release is demanded by journal_unmap_buffer() remain after a transaction commits. And also the pages to which they belong remain. Such the remained pages don't have mapping any longer. Due to this fact, there is a possibility that the pages which cannot be estimated remain. The metadata buffers marked 'BH_Freed' and the pages to which they belong can be released at 'JBD: commit phase 7'. Therefore, by applying the same code into 'JBD: commit phase 2' (where the data buffers are done with), journal_commit_transaction() can also release the data buffers marked 'BH_Freed' and the pages to which they belong. As a result, all the buffers marked 'BH_Freed' can be released, and also all the pages to which these buffers belong can be released at journal_commit_transaction(). So, the page which cannot be estimated is lost. <<Excerpt of code at 'JBD: commit phase 7'>> > spin_lock(&journal->j_list_lock); > while (commit_transaction->t_forget) { > transaction_t *cp_transaction; > struct buffer_head *bh; > > jh = commit_transaction->t_forget; >... > if (buffer_freed(bh)) { > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_freed(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_jbddirty(bh); > } > > if (buffer_jbddirty(bh)) { > JBUFFER_TRACE(jh, "add to new checkpointing trans"); > __journal_insert_checkpoint(jh, commit_transaction); > JBUFFER_TRACE(jh, "refile for checkpoint writeback"); > __journal_refile_buffer(jh); > jbd_unlock_bh_state(bh); > } else { > J_ASSERT_BH(bh, !buffer_dirty(bh)); > ... > JBUFFER_TRACE(jh, "refile or unfile freed buffer"); > __journal_refile_buffer(jh); > if (!jh->b_transaction) { > jbd_unlock_bh_state(bh); > /* needs a brelse */ > journal_remove_journal_head(bh); > release_buffer_page(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > } else > } **************************************************************** * Apply the code of "^^^^^^" lines into 'JBD: commit phase 2' * **************************************************************** At journal_commit_transaction() code, there is one extra message in the series of jbd debug messages. ("JBD: commit phase 2") This patch fixes it, too. Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 16:46:29 +08:00
release_data_buffer(bh);
continue;
}
if (locked && test_clear_buffer_dirty(bh)) {
BUFFER_TRACE(bh, "needs writeout, adding to array");
wbuf[bufs++] = bh;
__journal_file_buffer(jh, commit_transaction,
BJ_Locked);
jbd_unlock_bh_state(bh);
if (bufs == journal->j_wbufsize) {
spin_unlock(&journal->j_list_lock);
trace_jbd_do_submit_data(journal,
commit_transaction);
journal_do_submit_data(wbuf, bufs, write_op);
bufs = 0;
goto write_out_data;
}
} else if (!locked && buffer_locked(bh)) {
__journal_file_buffer(jh, commit_transaction,
BJ_Locked);
jbd_unlock_bh_state(bh);
put_bh(bh);
} else {
BUFFER_TRACE(bh, "writeout complete: unfile");
if (unlikely(!buffer_uptodate(bh)))
err = -EIO;
__journal_unfile_buffer(jh);
jbd_unlock_bh_state(bh);
if (locked)
unlock_buffer(bh);
jbd: positively dispose the unmapped data buffers in journal_commit_transaction() After ext3-ordered files are truncated, there is a possibility that the pages which cannot be estimated still remain. Remaining pages can be released when the system has really few memory. So, it is not memory leakage. But the resource management software etc. may not work correctly. It is possible that journal_unmap_buffer() cannot release the buffers, and the pages to which they belong because they are attached to a commiting transaction and journal_unmap_buffer() cannot release them. To release such the buffers and the pages later, journal_unmap_buffer() leaves it to journal_commit_transaction(). (journal_unmap_buffer() puts the mark 'BH_Freed' to the buffers so that journal_commit_transaction() can identify whether they can be released or not.) In the journalled mode and the writeback mode, jbd does with only metadata buffers. But in the ordered mode, jbd does with metadata buffers and also data buffers. Actually, journal_commit_transaction() releases only the metadata buffers of which release is demanded by journal_unmap_buffer(), and also releases the pages to which they belong if possible. As a result, the data buffers of which release is demanded by journal_unmap_buffer() remain after a transaction commits. And also the pages to which they belong remain. Such the remained pages don't have mapping any longer. Due to this fact, there is a possibility that the pages which cannot be estimated remain. The metadata buffers marked 'BH_Freed' and the pages to which they belong can be released at 'JBD: commit phase 7'. Therefore, by applying the same code into 'JBD: commit phase 2' (where the data buffers are done with), journal_commit_transaction() can also release the data buffers marked 'BH_Freed' and the pages to which they belong. As a result, all the buffers marked 'BH_Freed' can be released, and also all the pages to which these buffers belong can be released at journal_commit_transaction(). So, the page which cannot be estimated is lost. <<Excerpt of code at 'JBD: commit phase 7'>> > spin_lock(&journal->j_list_lock); > while (commit_transaction->t_forget) { > transaction_t *cp_transaction; > struct buffer_head *bh; > > jh = commit_transaction->t_forget; >... > if (buffer_freed(bh)) { > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_freed(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_jbddirty(bh); > } > > if (buffer_jbddirty(bh)) { > JBUFFER_TRACE(jh, "add to new checkpointing trans"); > __journal_insert_checkpoint(jh, commit_transaction); > JBUFFER_TRACE(jh, "refile for checkpoint writeback"); > __journal_refile_buffer(jh); > jbd_unlock_bh_state(bh); > } else { > J_ASSERT_BH(bh, !buffer_dirty(bh)); > ... > JBUFFER_TRACE(jh, "refile or unfile freed buffer"); > __journal_refile_buffer(jh); > if (!jh->b_transaction) { > jbd_unlock_bh_state(bh); > /* needs a brelse */ > journal_remove_journal_head(bh); > release_buffer_page(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > } else > } **************************************************************** * Apply the code of "^^^^^^" lines into 'JBD: commit phase 2' * **************************************************************** At journal_commit_transaction() code, there is one extra message in the series of jbd debug messages. ("JBD: commit phase 2") This patch fixes it, too. Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 16:46:29 +08:00
release_data_buffer(bh);
}
if (need_resched() || spin_needbreak(&journal->j_list_lock)) {
spin_unlock(&journal->j_list_lock);
goto write_out_data;
}
}
spin_unlock(&journal->j_list_lock);
trace_jbd_do_submit_data(journal, commit_transaction);
journal_do_submit_data(wbuf, bufs, write_op);
return err;
}
/*
* journal_commit_transaction
*
* The primary function for committing a transaction to the log. This
* function is called by the journal thread to begin a complete commit.
*/
void journal_commit_transaction(journal_t *journal)
{
transaction_t *commit_transaction;
struct journal_head *jh, *new_jh, *descriptor;
struct buffer_head **wbuf = journal->j_wbuf;
int bufs;
int flags;
int err;
unsigned int blocknr;
jbd: improve fsync batching There is a flaw with the way jbd handles fsync batching. If we fsync() a file and we were not the last person to run fsync() on this fs then we automatically sleep for 1 jiffie in order to wait for new writers to join into the transaction before forcing the commit. The problem with this is that with really fast storage (ie a Clariion) the time it takes to commit a transaction to disk is way faster than 1 jiffie in most cases, so sleeping means waiting longer with nothing to do than if we just committed the transaction and kept going. Ric Wheeler noticed this when using fs_mark with more than 1 thread, the throughput would plummet as he added more threads. This patch attempts to fix this problem by recording the average time in nanoseconds that it takes to commit a transaction to disk, and what time we started the transaction. If we run an fsync() and we have been running for less time than it takes to commit the transaction to disk, we sleep for the delta amount of time and then commit to disk. We acheive sub-jiffie sleeping using schedule_hrtimeout. This means that the wait time is auto-tuned to the speed of the underlying disk, instead of having this static timeout. I weighted the average according to somebody's comments (Andreas Dilger I think) in order to help normalize random outliers where we take way longer or way less time to commit than the average. I also have a min() check in there to make sure we don't sleep longer than a jiffie in case our storage is super slow, this was requested by Andrew. I unfortunately do not have access to a Clariion, so I had to use a ramdisk to represent a super fast array. I tested with a SATA drive with barrier=1 to make sure there was no regression with local disks, I tested with a 4 way multipathed Apple Xserve RAID array and of course the ramdisk. I ran the following command fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 -t $i where $i was 2, 4, 8, 16 and 32. I mkfs'ed the fs each time. Here are my results type threads with patch without patch sata 2 24.6 26.3 sata 4 49.2 48.1 sata 8 70.1 67.0 sata 16 104.0 94.1 sata 32 153.6 142.7 xserve 2 246.4 222.0 xserve 4 480.0 440.8 xserve 8 829.5 730.8 xserve 16 1172.7 1026.9 xserve 32 1816.3 1650.5 ramdisk 2 2538.3 1745.6 ramdisk 4 2942.3 661.9 ramdisk 8 2882.5 999.8 ramdisk 16 2738.7 1801.9 ramdisk 32 2541.9 2394.0 Signed-off-by: Josef Bacik <jbacik@redhat.com> Cc: Andreas Dilger <adilger@sun.com> Cc: Arjan van de Ven <arjan@infradead.org> Cc: Ric Wheeler <rwheeler@redhat.com> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-08 10:07:24 +08:00
ktime_t start_time;
u64 commit_time;
char *tagp = NULL;
journal_header_t *header;
journal_block_tag_t *tag = NULL;
int space_left = 0;
int first_tag = 0;
int tag_flag;
int i;
struct blk_plug plug;
int write_op = WRITE;
/*
* First job: lock down the current transaction and wait for
* all outstanding updates to complete.
*/
/* Do we need to erase the effects of a prior journal_flush? */
if (journal->j_flags & JFS_FLUSHED) {
jbd_debug(3, "super block updated\n");
mutex_lock(&journal->j_checkpoint_mutex);
/*
* We hold j_checkpoint_mutex so tail cannot change under us.
* We don't need any special data guarantees for writing sb
* since journal is empty and it is ok for write to be
* flushed only with transaction commit.
*/
journal_update_sb_log_tail(journal, journal->j_tail_sequence,
journal->j_tail, WRITE_SYNC);
mutex_unlock(&journal->j_checkpoint_mutex);
} else {
jbd_debug(3, "superblock not updated\n");
}
J_ASSERT(journal->j_running_transaction != NULL);
J_ASSERT(journal->j_committing_transaction == NULL);
commit_transaction = journal->j_running_transaction;
J_ASSERT(commit_transaction->t_state == T_RUNNING);
trace_jbd_start_commit(journal, commit_transaction);
jbd_debug(1, "JBD: starting commit of transaction %d\n",
commit_transaction->t_tid);
spin_lock(&journal->j_state_lock);
commit_transaction->t_state = T_LOCKED;
trace_jbd_commit_locking(journal, commit_transaction);
spin_lock(&commit_transaction->t_handle_lock);
while (commit_transaction->t_updates) {
DEFINE_WAIT(wait);
prepare_to_wait(&journal->j_wait_updates, &wait,
TASK_UNINTERRUPTIBLE);
if (commit_transaction->t_updates) {
spin_unlock(&commit_transaction->t_handle_lock);
spin_unlock(&journal->j_state_lock);
schedule();
spin_lock(&journal->j_state_lock);
spin_lock(&commit_transaction->t_handle_lock);
}
finish_wait(&journal->j_wait_updates, &wait);
}
spin_unlock(&commit_transaction->t_handle_lock);
J_ASSERT (commit_transaction->t_outstanding_credits <=
journal->j_max_transaction_buffers);
/*
* First thing we are allowed to do is to discard any remaining
* BJ_Reserved buffers. Note, it is _not_ permissible to assume
* that there are no such buffers: if a large filesystem
* operation like a truncate needs to split itself over multiple
* transactions, then it may try to do a journal_restart() while
* there are still BJ_Reserved buffers outstanding. These must
* be released cleanly from the current transaction.
*
* In this case, the filesystem must still reserve write access
* again before modifying the buffer in the new transaction, but
* we do not require it to remember exactly which old buffers it
* has reserved. This is consistent with the existing behaviour
* that multiple journal_get_write_access() calls to the same
* buffer are perfectly permissible.
*/
while (commit_transaction->t_reserved_list) {
jh = commit_transaction->t_reserved_list;
JBUFFER_TRACE(jh, "reserved, unused: refile");
/*
* A journal_get_undo_access()+journal_release_buffer() may
* leave undo-committed data.
*/
if (jh->b_committed_data) {
struct buffer_head *bh = jh2bh(jh);
jbd_lock_bh_state(bh);
jbd_free(jh->b_committed_data, bh->b_size);
jh->b_committed_data = NULL;
jbd_unlock_bh_state(bh);
}
journal_refile_buffer(journal, jh);
}
/*
* Now try to drop any written-back buffers from the journal's
* checkpoint lists. We do this *before* commit because it potentially
* frees some memory
*/
spin_lock(&journal->j_list_lock);
__journal_clean_checkpoint_list(journal);
spin_unlock(&journal->j_list_lock);
jbd_debug (3, "JBD: commit phase 1\n");
/*
* Clear revoked flag to reflect there is no revoked buffers
* in the next transaction which is going to be started.
*/
journal_clear_buffer_revoked_flags(journal);
/*
* Switch to a new revoke table.
*/
journal_switch_revoke_table(journal);
trace_jbd_commit_flushing(journal, commit_transaction);
commit_transaction->t_state = T_FLUSH;
journal->j_committing_transaction = commit_transaction;
journal->j_running_transaction = NULL;
jbd: improve fsync batching There is a flaw with the way jbd handles fsync batching. If we fsync() a file and we were not the last person to run fsync() on this fs then we automatically sleep for 1 jiffie in order to wait for new writers to join into the transaction before forcing the commit. The problem with this is that with really fast storage (ie a Clariion) the time it takes to commit a transaction to disk is way faster than 1 jiffie in most cases, so sleeping means waiting longer with nothing to do than if we just committed the transaction and kept going. Ric Wheeler noticed this when using fs_mark with more than 1 thread, the throughput would plummet as he added more threads. This patch attempts to fix this problem by recording the average time in nanoseconds that it takes to commit a transaction to disk, and what time we started the transaction. If we run an fsync() and we have been running for less time than it takes to commit the transaction to disk, we sleep for the delta amount of time and then commit to disk. We acheive sub-jiffie sleeping using schedule_hrtimeout. This means that the wait time is auto-tuned to the speed of the underlying disk, instead of having this static timeout. I weighted the average according to somebody's comments (Andreas Dilger I think) in order to help normalize random outliers where we take way longer or way less time to commit than the average. I also have a min() check in there to make sure we don't sleep longer than a jiffie in case our storage is super slow, this was requested by Andrew. I unfortunately do not have access to a Clariion, so I had to use a ramdisk to represent a super fast array. I tested with a SATA drive with barrier=1 to make sure there was no regression with local disks, I tested with a 4 way multipathed Apple Xserve RAID array and of course the ramdisk. I ran the following command fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 -t $i where $i was 2, 4, 8, 16 and 32. I mkfs'ed the fs each time. Here are my results type threads with patch without patch sata 2 24.6 26.3 sata 4 49.2 48.1 sata 8 70.1 67.0 sata 16 104.0 94.1 sata 32 153.6 142.7 xserve 2 246.4 222.0 xserve 4 480.0 440.8 xserve 8 829.5 730.8 xserve 16 1172.7 1026.9 xserve 32 1816.3 1650.5 ramdisk 2 2538.3 1745.6 ramdisk 4 2942.3 661.9 ramdisk 8 2882.5 999.8 ramdisk 16 2738.7 1801.9 ramdisk 32 2541.9 2394.0 Signed-off-by: Josef Bacik <jbacik@redhat.com> Cc: Andreas Dilger <adilger@sun.com> Cc: Arjan van de Ven <arjan@infradead.org> Cc: Ric Wheeler <rwheeler@redhat.com> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-08 10:07:24 +08:00
start_time = ktime_get();
commit_transaction->t_log_start = journal->j_head;
wake_up(&journal->j_wait_transaction_locked);
spin_unlock(&journal->j_state_lock);
jbd_debug (3, "JBD: commit phase 2\n");
if (tid_geq(journal->j_commit_waited, commit_transaction->t_tid))
write_op = WRITE_SYNC;
/*
* Now start flushing things to disk, in the order they appear
* on the transaction lists. Data blocks go first.
*/
blk_start_plug(&plug);
err = journal_submit_data_buffers(journal, commit_transaction,
write_op);
blk_finish_plug(&plug);
/*
* Wait for all previously submitted IO to complete.
*/
spin_lock(&journal->j_list_lock);
while (commit_transaction->t_locked_list) {
struct buffer_head *bh;
jh = commit_transaction->t_locked_list->b_tprev;
bh = jh2bh(jh);
get_bh(bh);
if (buffer_locked(bh)) {
spin_unlock(&journal->j_list_lock);
wait_on_buffer(bh);
spin_lock(&journal->j_list_lock);
}
if (unlikely(!buffer_uptodate(bh))) {
if (!trylock_page(bh->b_page)) {
spin_unlock(&journal->j_list_lock);
lock_page(bh->b_page);
spin_lock(&journal->j_list_lock);
}
if (bh->b_page->mapping)
set_bit(AS_EIO, &bh->b_page->mapping->flags);
unlock_page(bh->b_page);
SetPageError(bh->b_page);
err = -EIO;
}
if (!inverted_lock(journal, bh)) {
put_bh(bh);
spin_lock(&journal->j_list_lock);
continue;
}
if (buffer_jbd(bh) && bh2jh(bh) == jh &&
jh->b_transaction == commit_transaction &&
jbd: Fix oops in journal_remove_journal_head() journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 journal_commit_transaction() ... processing t_forget list __journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); journal_try_to_free_buffers() journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() journal_put_journal_head(jh) journal_remove_journal_head(bh); journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(), [__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara <jack@suse.cz>
2011-06-25 05:11:59 +08:00
jh->b_jlist == BJ_Locked)
__journal_unfile_buffer(jh);
jbd: Fix oops in journal_remove_journal_head() journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 journal_commit_transaction() ... processing t_forget list __journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); journal_try_to_free_buffers() journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() journal_put_journal_head(jh) journal_remove_journal_head(bh); journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(), [__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara <jack@suse.cz>
2011-06-25 05:11:59 +08:00
jbd_unlock_bh_state(bh);
jbd: positively dispose the unmapped data buffers in journal_commit_transaction() After ext3-ordered files are truncated, there is a possibility that the pages which cannot be estimated still remain. Remaining pages can be released when the system has really few memory. So, it is not memory leakage. But the resource management software etc. may not work correctly. It is possible that journal_unmap_buffer() cannot release the buffers, and the pages to which they belong because they are attached to a commiting transaction and journal_unmap_buffer() cannot release them. To release such the buffers and the pages later, journal_unmap_buffer() leaves it to journal_commit_transaction(). (journal_unmap_buffer() puts the mark 'BH_Freed' to the buffers so that journal_commit_transaction() can identify whether they can be released or not.) In the journalled mode and the writeback mode, jbd does with only metadata buffers. But in the ordered mode, jbd does with metadata buffers and also data buffers. Actually, journal_commit_transaction() releases only the metadata buffers of which release is demanded by journal_unmap_buffer(), and also releases the pages to which they belong if possible. As a result, the data buffers of which release is demanded by journal_unmap_buffer() remain after a transaction commits. And also the pages to which they belong remain. Such the remained pages don't have mapping any longer. Due to this fact, there is a possibility that the pages which cannot be estimated remain. The metadata buffers marked 'BH_Freed' and the pages to which they belong can be released at 'JBD: commit phase 7'. Therefore, by applying the same code into 'JBD: commit phase 2' (where the data buffers are done with), journal_commit_transaction() can also release the data buffers marked 'BH_Freed' and the pages to which they belong. As a result, all the buffers marked 'BH_Freed' can be released, and also all the pages to which these buffers belong can be released at journal_commit_transaction(). So, the page which cannot be estimated is lost. <<Excerpt of code at 'JBD: commit phase 7'>> > spin_lock(&journal->j_list_lock); > while (commit_transaction->t_forget) { > transaction_t *cp_transaction; > struct buffer_head *bh; > > jh = commit_transaction->t_forget; >... > if (buffer_freed(bh)) { > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_freed(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > clear_buffer_jbddirty(bh); > } > > if (buffer_jbddirty(bh)) { > JBUFFER_TRACE(jh, "add to new checkpointing trans"); > __journal_insert_checkpoint(jh, commit_transaction); > JBUFFER_TRACE(jh, "refile for checkpoint writeback"); > __journal_refile_buffer(jh); > jbd_unlock_bh_state(bh); > } else { > J_ASSERT_BH(bh, !buffer_dirty(bh)); > ... > JBUFFER_TRACE(jh, "refile or unfile freed buffer"); > __journal_refile_buffer(jh); > if (!jh->b_transaction) { > jbd_unlock_bh_state(bh); > /* needs a brelse */ > journal_remove_journal_head(bh); > release_buffer_page(bh); > ^^^^^^^^^^^^^^^^^^^^^^^^ > } else > } **************************************************************** * Apply the code of "^^^^^^" lines into 'JBD: commit phase 2' * **************************************************************** At journal_commit_transaction() code, there is one extra message in the series of jbd debug messages. ("JBD: commit phase 2") This patch fixes it, too. Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 16:46:29 +08:00
release_data_buffer(bh);
cond_resched_lock(&journal->j_list_lock);
}
spin_unlock(&journal->j_list_lock);
if (err) {
char b[BDEVNAME_SIZE];
printk(KERN_WARNING
"JBD: Detected IO errors while flushing file data "
"on %s\n", bdevname(journal->j_fs_dev, b));
if (journal->j_flags & JFS_ABORT_ON_SYNCDATA_ERR)
journal_abort(journal, err);
err = 0;
}
blk_start_plug(&plug);
journal_write_revoke_records(journal, commit_transaction, write_op);
/*
* If we found any dirty or locked buffers, then we should have
* looped back up to the write_out_data label. If there weren't
* any then journal_clean_data_list should have wiped the list
* clean by now, so check that it is in fact empty.
*/
J_ASSERT (commit_transaction->t_sync_datalist == NULL);
jbd_debug (3, "JBD: commit phase 3\n");
/*
* Way to go: we have now written out all of the data for a
* transaction! Now comes the tricky part: we need to write out
* metadata. Loop over the transaction's entire buffer list:
*/
spin_lock(&journal->j_state_lock);
commit_transaction->t_state = T_COMMIT;
spin_unlock(&journal->j_state_lock);
trace_jbd_commit_logging(journal, commit_transaction);
jbd: fix possible journal overflow issues There are several cases where the running transaction can get buffers added to its BJ_Metadata list which it never dirtied, which makes its t_nr_buffers counter end up larger than its t_outstanding_credits counter. This will cause issues when starting new transactions as while we are logging buffers we decrement t_outstanding_buffers, so when t_outstanding_buffers goes negative, we will report that we need less space in the journal than we actually need, so transactions will be started even though there may not be enough room for them. In the worst case scenario (which admittedly is almost impossible to reproduce) this will result in the journal running out of space. The fix is to only refile buffers from the committing transaction to the running transactions BJ_Modified list when b_modified is set on that journal, which is the only way to be sure if the running transaction has modified that buffer. This patch also fixes an accounting error in journal_forget, it is possible that we can call journal_forget on a buffer without having modified it, only gotten write access to it, so instead of freeing a credit, we only do so if the buffer was modified. The assert will help catch if this problem occurs. Without these two patches I could hit this assert within minutes of running postmark, with them this issue no longer arises. Thank you, Signed-off-by: Josef Bacik <jbacik@redhat.com> Cc: <linux-ext4@vger.kernel.org> Acked-by: Jan Kara <jack@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-28 17:16:12 +08:00
J_ASSERT(commit_transaction->t_nr_buffers <=
commit_transaction->t_outstanding_credits);
descriptor = NULL;
bufs = 0;
while (commit_transaction->t_buffers) {
/* Find the next buffer to be journaled... */
jh = commit_transaction->t_buffers;
/* If we're in abort mode, we just un-journal the buffer and
release it. */
if (is_journal_aborted(journal)) {
clear_buffer_jbddirty(jh2bh(jh));
JBUFFER_TRACE(jh, "journal is aborting: refile");
journal_refile_buffer(journal, jh);
/* If that was the last one, we need to clean up
* any descriptor buffers which may have been
* already allocated, even if we are now
* aborting. */
if (!commit_transaction->t_buffers)
goto start_journal_io;
continue;
}
/* Make sure we have a descriptor block in which to
record the metadata buffer. */
if (!descriptor) {
struct buffer_head *bh;
J_ASSERT (bufs == 0);
jbd_debug(4, "JBD: get descriptor\n");
descriptor = journal_get_descriptor_buffer(journal);
if (!descriptor) {
journal_abort(journal, -EIO);
continue;
}
bh = jh2bh(descriptor);
jbd_debug(4, "JBD: got buffer %llu (%p)\n",
(unsigned long long)bh->b_blocknr, bh->b_data);
header = (journal_header_t *)&bh->b_data[0];
header->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER);
header->h_blocktype = cpu_to_be32(JFS_DESCRIPTOR_BLOCK);
header->h_sequence = cpu_to_be32(commit_transaction->t_tid);
tagp = &bh->b_data[sizeof(journal_header_t)];
space_left = bh->b_size - sizeof(journal_header_t);
first_tag = 1;
set_buffer_jwrite(bh);
set_buffer_dirty(bh);
wbuf[bufs++] = bh;
/* Record it so that we can wait for IO
completion later */
BUFFER_TRACE(bh, "ph3: file as descriptor");
journal_file_buffer(descriptor, commit_transaction,
BJ_LogCtl);
}
/* Where is the buffer to be written? */
err = journal_next_log_block(journal, &blocknr);
/* If the block mapping failed, just abandon the buffer
and repeat this loop: we'll fall into the
refile-on-abort condition above. */
if (err) {
journal_abort(journal, err);
continue;
}
/*
* start_this_handle() uses t_outstanding_credits to determine
* the free space in the log, but this counter is changed
* by journal_next_log_block() also.
*/
commit_transaction->t_outstanding_credits--;
/* Bump b_count to prevent truncate from stumbling over
the shadowed buffer! @@@ This can go if we ever get
rid of the BJ_IO/BJ_Shadow pairing of buffers. */
get_bh(jh2bh(jh));
/* Make a temporary IO buffer with which to write it out
(this will requeue both the metadata buffer and the
temporary IO buffer). new_bh goes on BJ_IO*/
set_buffer_jwrite(jh2bh(jh));
/*
* akpm: journal_write_metadata_buffer() sets
* new_bh->b_transaction to commit_transaction.
* We need to clean this up before we release new_bh
* (which is of type BJ_IO)
*/
JBUFFER_TRACE(jh, "ph3: write metadata");
flags = journal_write_metadata_buffer(commit_transaction,
jh, &new_jh, blocknr);
set_buffer_jwrite(jh2bh(new_jh));
wbuf[bufs++] = jh2bh(new_jh);
/* Record the new block's tag in the current descriptor
buffer */
tag_flag = 0;
if (flags & 1)
tag_flag |= JFS_FLAG_ESCAPE;
if (!first_tag)
tag_flag |= JFS_FLAG_SAME_UUID;
tag = (journal_block_tag_t *) tagp;
tag->t_blocknr = cpu_to_be32(jh2bh(jh)->b_blocknr);
tag->t_flags = cpu_to_be32(tag_flag);
tagp += sizeof(journal_block_tag_t);
space_left -= sizeof(journal_block_tag_t);
if (first_tag) {
memcpy (tagp, journal->j_uuid, 16);
tagp += 16;
space_left -= 16;
first_tag = 0;
}
/* If there's no more to do, or if the descriptor is full,
let the IO rip! */
if (bufs == journal->j_wbufsize ||
commit_transaction->t_buffers == NULL ||
space_left < sizeof(journal_block_tag_t) + 16) {
jbd_debug(4, "JBD: Submit %d IOs\n", bufs);
/* Write an end-of-descriptor marker before
submitting the IOs. "tag" still points to
the last tag we set up. */
tag->t_flags |= cpu_to_be32(JFS_FLAG_LAST_TAG);
start_journal_io:
for (i = 0; i < bufs; i++) {
struct buffer_head *bh = wbuf[i];
lock_buffer(bh);
clear_buffer_dirty(bh);
set_buffer_uptodate(bh);
bh->b_end_io = journal_end_buffer_io_sync;
submit_bh(write_op, bh);
}
cond_resched();
/* Force a new descriptor to be generated next
time round the loop. */
descriptor = NULL;
bufs = 0;
}
}
blk_finish_plug(&plug);
/* Lo and behold: we have just managed to send a transaction to
the log. Before we can commit it, wait for the IO so far to
complete. Control buffers being written are on the
transaction's t_log_list queue, and metadata buffers are on
the t_iobuf_list queue.
Wait for the buffers in reverse order. That way we are
less likely to be woken up until all IOs have completed, and
so we incur less scheduling load.
*/
jbd_debug(3, "JBD: commit phase 4\n");
/*
* akpm: these are BJ_IO, and j_list_lock is not needed.
* See __journal_try_to_free_buffer.
*/
wait_for_iobuf:
while (commit_transaction->t_iobuf_list != NULL) {
struct buffer_head *bh;
jh = commit_transaction->t_iobuf_list->b_tprev;
bh = jh2bh(jh);
if (buffer_locked(bh)) {
wait_on_buffer(bh);
goto wait_for_iobuf;
}
if (cond_resched())
goto wait_for_iobuf;
if (unlikely(!buffer_uptodate(bh)))
err = -EIO;
clear_buffer_jwrite(bh);
JBUFFER_TRACE(jh, "ph4: unfile after journal write");
journal_unfile_buffer(journal, jh);
/*
* ->t_iobuf_list should contain only dummy buffer_heads
* which were created by journal_write_metadata_buffer().
*/
BUFFER_TRACE(bh, "dumping temporary bh");
journal_put_journal_head(jh);
__brelse(bh);
J_ASSERT_BH(bh, atomic_read(&bh->b_count) == 0);
free_buffer_head(bh);
/* We also have to unlock and free the corresponding
shadowed buffer */
jh = commit_transaction->t_shadow_list->b_tprev;
bh = jh2bh(jh);
clear_buffer_jwrite(bh);
J_ASSERT_BH(bh, buffer_jbddirty(bh));
/* The metadata is now released for reuse, but we need
to remember it against this transaction so that when
we finally commit, we can do any checkpointing
required. */
JBUFFER_TRACE(jh, "file as BJ_Forget");
journal_file_buffer(jh, commit_transaction, BJ_Forget);
/*
* Wake up any transactions which were waiting for this
* IO to complete. The barrier must be here so that changes
* by journal_file_buffer() take effect before wake_up_bit()
* does the waitqueue check.
*/
smp_mb();
wake_up_bit(&bh->b_state, BH_Unshadow);
JBUFFER_TRACE(jh, "brelse shadowed buffer");
__brelse(bh);
}
J_ASSERT (commit_transaction->t_shadow_list == NULL);
jbd_debug(3, "JBD: commit phase 5\n");
/* Here we wait for the revoke record and descriptor record buffers */
wait_for_ctlbuf:
while (commit_transaction->t_log_list != NULL) {
struct buffer_head *bh;
jh = commit_transaction->t_log_list->b_tprev;
bh = jh2bh(jh);
if (buffer_locked(bh)) {
wait_on_buffer(bh);
goto wait_for_ctlbuf;
}
if (cond_resched())
goto wait_for_ctlbuf;
if (unlikely(!buffer_uptodate(bh)))
err = -EIO;
BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile");
clear_buffer_jwrite(bh);
journal_unfile_buffer(journal, jh);
journal_put_journal_head(jh);
__brelse(bh); /* One for getblk */
/* AKPM: bforget here */
}
if (err)
journal_abort(journal, err);
jbd_debug(3, "JBD: commit phase 6\n");
/* All metadata is written, now write commit record and do cleanup */
spin_lock(&journal->j_state_lock);
J_ASSERT(commit_transaction->t_state == T_COMMIT);
commit_transaction->t_state = T_COMMIT_RECORD;
spin_unlock(&journal->j_state_lock);
if (journal_write_commit_record(journal, commit_transaction))
err = -EIO;
if (err)
journal_abort(journal, err);
/* End of a transaction! Finally, we can do checkpoint
processing: any buffers committed as a result of this
transaction can be removed from any checkpoint list it was on
before. */
jbd_debug(3, "JBD: commit phase 7\n");
J_ASSERT(commit_transaction->t_sync_datalist == NULL);
J_ASSERT(commit_transaction->t_buffers == NULL);
J_ASSERT(commit_transaction->t_checkpoint_list == NULL);
J_ASSERT(commit_transaction->t_iobuf_list == NULL);
J_ASSERT(commit_transaction->t_shadow_list == NULL);
J_ASSERT(commit_transaction->t_log_list == NULL);
restart_loop:
/*
* As there are other places (journal_unmap_buffer()) adding buffers
* to this list we have to be careful and hold the j_list_lock.
*/
spin_lock(&journal->j_list_lock);
while (commit_transaction->t_forget) {
transaction_t *cp_transaction;
struct buffer_head *bh;
jbd: Fix oops in journal_remove_journal_head() journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 journal_commit_transaction() ... processing t_forget list __journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); journal_try_to_free_buffers() journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() journal_put_journal_head(jh) journal_remove_journal_head(bh); journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(), [__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara <jack@suse.cz>
2011-06-25 05:11:59 +08:00
int try_to_free = 0;
jh = commit_transaction->t_forget;
spin_unlock(&journal->j_list_lock);
bh = jh2bh(jh);
jbd: Fix oops in journal_remove_journal_head() journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 journal_commit_transaction() ... processing t_forget list __journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); journal_try_to_free_buffers() journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() journal_put_journal_head(jh) journal_remove_journal_head(bh); journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(), [__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara <jack@suse.cz>
2011-06-25 05:11:59 +08:00
/*
* Get a reference so that bh cannot be freed before we are
* done with it.
*/
get_bh(bh);
jbd_lock_bh_state(bh);
J_ASSERT_JH(jh, jh->b_transaction == commit_transaction ||
jh->b_transaction == journal->j_running_transaction);
/*
* If there is undo-protected committed data against
* this buffer, then we can remove it now. If it is a
* buffer needing such protection, the old frozen_data
* field now points to a committed version of the
* buffer, so rotate that field to the new committed
* data.
*
* Otherwise, we can just throw away the frozen data now.
*/
if (jh->b_committed_data) {
jbd_free(jh->b_committed_data, bh->b_size);
jh->b_committed_data = NULL;
if (jh->b_frozen_data) {
jh->b_committed_data = jh->b_frozen_data;
jh->b_frozen_data = NULL;
}
} else if (jh->b_frozen_data) {
jbd_free(jh->b_frozen_data, bh->b_size);
jh->b_frozen_data = NULL;
}
spin_lock(&journal->j_list_lock);
cp_transaction = jh->b_cp_transaction;
if (cp_transaction) {
JBUFFER_TRACE(jh, "remove from old cp transaction");
__journal_remove_checkpoint(jh);
}
/* Only re-checkpoint the buffer_head if it is marked
* dirty. If the buffer was added to the BJ_Forget list
* by journal_forget, it may no longer be dirty and
* there's no point in keeping a checkpoint record for
* it. */
/* A buffer which has been freed while still being
* journaled by a previous transaction may end up still
* being dirty here, but we want to avoid writing back
* that buffer in the future after the "add to orphan"
* operation been committed, That's not only a performance
* gain, it also stops aliasing problems if the buffer is
* left behind for writeback and gets reallocated for another
* use in a different page. */
if (buffer_freed(bh) && !jh->b_next_transaction) {
clear_buffer_freed(bh);
clear_buffer_jbddirty(bh);
}
if (buffer_jbddirty(bh)) {
JBUFFER_TRACE(jh, "add to new checkpointing trans");
__journal_insert_checkpoint(jh, commit_transaction);
if (is_journal_aborted(journal))
clear_buffer_jbddirty(bh);
} else {
J_ASSERT_BH(bh, !buffer_dirty(bh));
jbd: Fix oops in journal_remove_journal_head() journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 journal_commit_transaction() ... processing t_forget list __journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); journal_try_to_free_buffers() journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() journal_put_journal_head(jh) journal_remove_journal_head(bh); journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(), [__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara <jack@suse.cz>
2011-06-25 05:11:59 +08:00
/*
* The buffer on BJ_Forget list and not jbddirty means
[PATCH] jbd: fix BUG in journal_commit_transaction() Fix possible assertion failure in journal_commit_transaction() on jh->b_next_transaction == NULL (when we are processing BJ_Forget list and buffer is not jbddirty). !jbddirty buffers can be placed on BJ_Forget list for example by journal_forget() or by __dispose_buffer() - generally such buffer means that it has been freed by this transaction. Freed buffers should not be reallocated until the transaction has committed (that's why we have the assertion there) but they *can* be reallocated when the transaction has already been committed to disk and we are just processing the BJ_Forget list (as soon as we remove b_committed_data from the bitmap bh, ext3 will be able to reallocate buffers freed by the committing transaction). So we have to also count with the case that the buffer has been reallocated and b_next_transaction has been already set. And one more subtle point: it can happen that we manage to reallocate the buffer and also mark it jbddirty. Then we also add the freed buffer to the checkpoint list of the committing trasaction. But that should do no harm. Non-jbddirty buffers should be filed to BJ_Reserved and not BJ_Metadata list. It can actually happen that we refile such buffers during the commit phase when we reallocate in the running transaction blocks deleted in committing transaction (and that can happen if the committing transaction already wrote all the data and is just cleaning up BJ_Forget list). Signed-off-by: Jan Kara <jack@suse.cz> Acked-by: "Stephen C. Tweedie" <sct@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 17:05:25 +08:00
* it has been freed by this transaction and hence it
* could not have been reallocated until this
* transaction has committed. *BUT* it could be
* reallocated once we have written all the data to
* disk and before we process the buffer on BJ_Forget
jbd: Fix oops in journal_remove_journal_head() journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 journal_commit_transaction() ... processing t_forget list __journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); journal_try_to_free_buffers() journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() journal_put_journal_head(jh) journal_remove_journal_head(bh); journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(), [__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara <jack@suse.cz>
2011-06-25 05:11:59 +08:00
* list.
*/
if (!jh->b_next_transaction)
try_to_free = 1;
}
jbd: Fix oops in journal_remove_journal_head() journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 journal_commit_transaction() ... processing t_forget list __journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); journal_try_to_free_buffers() journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() journal_put_journal_head(jh) journal_remove_journal_head(bh); journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]journal_refile_buffer(), [__]journal_unfile_buffer(), and __journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara <jack@suse.cz>
2011-06-25 05:11:59 +08:00
JBUFFER_TRACE(jh, "refile or unfile freed buffer");
__journal_refile_buffer(jh);
jbd_unlock_bh_state(bh);
if (try_to_free)
release_buffer_page(bh);
else
__brelse(bh);
cond_resched_lock(&journal->j_list_lock);
}
spin_unlock(&journal->j_list_lock);
/*
* This is a bit sleazy. We use j_list_lock to protect transition
* of a transaction into T_FINISHED state and calling
* __journal_drop_transaction(). Otherwise we could race with
* other checkpointing code processing the transaction...
*/
spin_lock(&journal->j_state_lock);
spin_lock(&journal->j_list_lock);
/*
* Now recheck if some buffers did not get attached to the transaction
* while the lock was dropped...
*/
if (commit_transaction->t_forget) {
spin_unlock(&journal->j_list_lock);
spin_unlock(&journal->j_state_lock);
goto restart_loop;
}
/* Done with this transaction! */
jbd_debug(3, "JBD: commit phase 8\n");
J_ASSERT(commit_transaction->t_state == T_COMMIT_RECORD);
commit_transaction->t_state = T_FINISHED;
J_ASSERT(commit_transaction == journal->j_committing_transaction);
journal->j_commit_sequence = commit_transaction->t_tid;
journal->j_committing_transaction = NULL;
jbd: improve fsync batching There is a flaw with the way jbd handles fsync batching. If we fsync() a file and we were not the last person to run fsync() on this fs then we automatically sleep for 1 jiffie in order to wait for new writers to join into the transaction before forcing the commit. The problem with this is that with really fast storage (ie a Clariion) the time it takes to commit a transaction to disk is way faster than 1 jiffie in most cases, so sleeping means waiting longer with nothing to do than if we just committed the transaction and kept going. Ric Wheeler noticed this when using fs_mark with more than 1 thread, the throughput would plummet as he added more threads. This patch attempts to fix this problem by recording the average time in nanoseconds that it takes to commit a transaction to disk, and what time we started the transaction. If we run an fsync() and we have been running for less time than it takes to commit the transaction to disk, we sleep for the delta amount of time and then commit to disk. We acheive sub-jiffie sleeping using schedule_hrtimeout. This means that the wait time is auto-tuned to the speed of the underlying disk, instead of having this static timeout. I weighted the average according to somebody's comments (Andreas Dilger I think) in order to help normalize random outliers where we take way longer or way less time to commit than the average. I also have a min() check in there to make sure we don't sleep longer than a jiffie in case our storage is super slow, this was requested by Andrew. I unfortunately do not have access to a Clariion, so I had to use a ramdisk to represent a super fast array. I tested with a SATA drive with barrier=1 to make sure there was no regression with local disks, I tested with a 4 way multipathed Apple Xserve RAID array and of course the ramdisk. I ran the following command fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 -t $i where $i was 2, 4, 8, 16 and 32. I mkfs'ed the fs each time. Here are my results type threads with patch without patch sata 2 24.6 26.3 sata 4 49.2 48.1 sata 8 70.1 67.0 sata 16 104.0 94.1 sata 32 153.6 142.7 xserve 2 246.4 222.0 xserve 4 480.0 440.8 xserve 8 829.5 730.8 xserve 16 1172.7 1026.9 xserve 32 1816.3 1650.5 ramdisk 2 2538.3 1745.6 ramdisk 4 2942.3 661.9 ramdisk 8 2882.5 999.8 ramdisk 16 2738.7 1801.9 ramdisk 32 2541.9 2394.0 Signed-off-by: Josef Bacik <jbacik@redhat.com> Cc: Andreas Dilger <adilger@sun.com> Cc: Arjan van de Ven <arjan@infradead.org> Cc: Ric Wheeler <rwheeler@redhat.com> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-08 10:07:24 +08:00
commit_time = ktime_to_ns(ktime_sub(ktime_get(), start_time));
/*
* weight the commit time higher than the average time so we don't
* react too strongly to vast changes in commit time
*/
if (likely(journal->j_average_commit_time))
journal->j_average_commit_time = (commit_time*3 +
journal->j_average_commit_time) / 4;
else
journal->j_average_commit_time = commit_time;
spin_unlock(&journal->j_state_lock);
if (commit_transaction->t_checkpoint_list == NULL &&
commit_transaction->t_checkpoint_io_list == NULL) {
__journal_drop_transaction(journal, commit_transaction);
} else {
if (journal->j_checkpoint_transactions == NULL) {
journal->j_checkpoint_transactions = commit_transaction;
commit_transaction->t_cpnext = commit_transaction;
commit_transaction->t_cpprev = commit_transaction;
} else {
commit_transaction->t_cpnext =
journal->j_checkpoint_transactions;
commit_transaction->t_cpprev =
commit_transaction->t_cpnext->t_cpprev;
commit_transaction->t_cpnext->t_cpprev =
commit_transaction;
commit_transaction->t_cpprev->t_cpnext =
commit_transaction;
}
}
spin_unlock(&journal->j_list_lock);
trace_jbd_end_commit(journal, commit_transaction);
jbd_debug(1, "JBD: commit %d complete, head %d\n",
journal->j_commit_sequence, journal->j_tail_sequence);
wake_up(&journal->j_wait_done_commit);
}