When running several fsx's and other filesystem stress tests, we found
cases where an unmapped buffer was still being sent to submit_bh by the
ext3 dirty data journaling code.
I saw this happen in two ways, both related to another thread doing a
truncate which would unmap the buffer in question.
Either we would get into journal_dirty_data with a bh which was already
unmapped (although journal_dirty_data_fn had checked for this earlier, the
state was not locked at that point), or it would get unmapped in the middle
of journal_dirty_data when we dropped locks to call sync_dirty_buffer.
By re-checking for mapped state after we've acquired the bh state lock, we
should avoid these races. If we find a buffer which is no longer mapped,
we essentially ignore it, because journal_unmap_buffer has already decided
that this buffer can go away.
I've also added tracepoints in these two cases, and made a couple other
tracepoint changes that I found useful in debugging this.
Signed-off-by: Eric Sandeen <esandeen@redhat.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A disk generated some I/O error, after it, I hitted
J_ASSERT(transaction->t_updates > 0) in journal_stop().
It seems to happened on ext3_truncate() path from stack trace. Then,
maybe the following case may trigger J_ASSERT(transaction->t_updates > 0).
ext3_truncate()
-> ext3_free_branches()
-> ext3_journal_test_restart()
-> ext3_journal_restart()
-> journal_restart()
transaction->t_updates--;
/* another process aborted journal */
-> start_this_handle()
returns -EROFS without transaction->t_updates++;
-> ext3_journal_stop()
-> journal_stop()
J_ASSERT(transaction->t_updates > 0)
If journal was aborted in middle of journal_restart(), ext3_truncate()
may trigger J_ASSERT().
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
More white space cleanups in preparation of cloning ext4 from ext3.
Removing spaces that precede a tab.
Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove whitespace from ext3 and jbd, before we clone ext4.
Signed-off-by: Mingming Cao<cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Missed a place where I forgot to convert kfree() to kmem_cache_free() as
part of jbd-manage-its-own-slab changes.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
JBD currently allocates commit and frozen buffers from slabs. With
CONFIG_SLAB_DEBUG, its possible for an allocation to cross the page
boundary causing IO problems.
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=200127
So, instead of allocating these from regular slabs - manage allocation from
its own slabs and disable slab debug for these slabs.
[akpm@osdl.org: cleanups]
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
There are a couple of places where JBD has to check to see whether an unneeded
memory allocation was performed. Usually it _was_ needed, so we end up
calling kfree(NULL). We can micro-optimise that by checking the pointer
before calling kfree().
Thanks to Steven Rostedt <rostedt@goodmis.org> for identifying this.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
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>
The return value of this function is never used, so let's be honest and
declare it as void.
Some places where invalidatepage returned 0, I have inserted comments
suggesting a BUG_ON.
[akpm@osdl.org: JBD BUG fix]
[akpm@osdl.org: rework for git-nfs]
[akpm@osdl.org: don't go BUG in block_invalidate_page()]
Signed-off-by: Neil Brown <neilb@suse.de>
Acked-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The kjournald timer is currently on the kernel thread's stack and the journal
structure points at it. Save a pointer hop by moving the timer into the
journal structure.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Semaphore to mutex conversion.
The conversion was generated via scripts, and the result was validated
automatically via a script as well.
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ben points out that:
When writing files out using O_SYNC, jbd's 1 jiffy delay results in a
significant drop in throughput as the disk sits idle. The patch below
results in a 4-5x performance improvement (from 6.5MB/s to ~24-30MB/s on my
IDE test box) when writing out files using O_SYNC.
So optimise the batching code by omitting it entirely if the process which is
doing a sync write is the same as the one which did the most recent sync
write. If that's true, we're unlikely to get any other processes joining the
transaction.
(Has been in -mm for ages - it took me a long time to get on to performance
testing it)
Numbers, on write-cache-disabled IDE:
/usr/bin/time -p synctest -n 10 -uf -t 1 -p 1 dir-name
Unpatched:
40 seconds
Patched:
35 seconds
Batching disabled:
35 seconds
This is the problematic single-process-doing-fsync case. With multiple
fsyncing processes the numbers are AFACIT unaltered by the patch.
Aside: performance testing and instrumentation shows that the transaction
batching almost doesn't help (testing with synctest -n 1 -uf -t 100 -p 10
dir-name on non-writeback-caching IDE). This is because by the time one
process is running a synchronous commit, a bunch of other processes already
have a transaction handle open, so they're all going to batch into the same
transaction anyway.
The batching seems to offer maybe 5-10% speedup with this workload, but I'm
pretty sure it was more important than that when it was first developed 4-odd
years ago...
Cc: "Stephen C. Tweedie" <sct@redhat.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is the fs/ part of the big kfree cleanup patch.
Remove pointless checks for NULL prior to calling kfree() in fs/.
Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- ->releasepage() annotated (s/int/gfp_t), instances updated
- missing gfp_t in fs/* added
- fixed misannotation from the original sweep caught by bitwise checks:
XFS used __nocast both for gfp_t and for flags used by XFS allocator.
The latter left with unsigned int __nocast; we might want to add a
different type for those but for now let's leave them alone. That,
BTW, is a case when __nocast use had been actively confusing - it had
been used in the same code for two different and similar types, with
no way to catch misuses. Switch of gfp_t to bitwise had caught that
immediately...
One tricky bit is left alone to be dealt with later - mapping->flags is
a mix of gfp_t and error indications. Left alone for now.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use schedule_timeout_{,un}interruptible() instead of
set_current_state()/schedule_timeout() to reduce kernel size. Also use helper
functions to convert between human time units and jiffies rather than constant
HZ division to avoid rounding errors.
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
attached patch should fix the following race:
Proc 1 Proc 2
__flush_batch()
ll_rw_block()
do_get_write_access()
lock_buffer
jh is only waiting for checkpoint
-> b_transaction == NULL ->
do nothing
unlock_buffer
test_set_buffer_locked()
test_clear_buffer_dirty()
__journal_file_buffer()
change the data
submit_bh()
and we have sent wrong data to disk... We now clean the dirty buffer flag
under buffer lock in all cases and hence we know that whenever a buffer is
starting to be journaled we either finish the pending write-out before
attaching a buffer to a transaction or we won't write the buffer until the
transaction is going to be committed.
The test in jbd_unexpected_dirty_buffer() is redundant - remove it.
Furthermore we have to clear the buffer dirty bit under the buffer lock to
prevent races with buffer write-out (and hence prevent returning a buffer with
IO happening).
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This fixes the lots-of-fsx-linux-instances-cause-a-slow-leak bug.
It's been there since 2.6.6, caused by:
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.5/2.6.5-mm4/broken-out/jbd-move-locked-buffers.patch
That patch moves under-writeout ordered-data buffers onto a separate journal
list during commit. It took out the old code which was based on a single
list.
The old code (necessarily) had logic which would restart I/O against buffers
which had been redirtied while they were on the committing transaction's
t_sync_datalist list. The new code only writes buffers once, ignoring
redirtyings by a later transaction, which is good.
But over on the truncate side of things, in journal_unmap_buffer(), we're
treating buffers on the t_locked_list as inviolable things which belong to the
committing transaction, and we just leave them alone during concurrent
truncate-vs-commit.
The net effect is that when truncate tries to invalidate a page whose buffers
are on t_locked_list and have been redirtied, journal_unmap_buffer() just
leaves those buffers alone. truncate will remove the page from its mapping
and we end up with an anonymous clean page with dirty buffers, which is an
illegal state for a page. The JBD commit will not clean those buffers as they
are removed from t_locked_list. The VM (try_to_free_buffers) cannot reclaim
these pages.
The patch teaches journal_unmap_buffer() about buffers which are on the
committing transaction's t_locked_list. These buffers have been written and
I/O has completed. We can take them off the transaction and undirty them
within the context of journal_invalidatepage()->journal_unmap_buffer().
Acked-by: "Stephen C. Tweedie" <sct@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
Eric Sandeen