On Thu 01-08-13 15:01:07, Paul Gortmaker wrote:
> Backport of jbd2 commit 0ef54180e0187117062939202b96faf04c8673bc
> ("jbd2: drop checkpoint mutex when waiting in __jbd2_log_wait_for_space()")
>
> Quoting that commit:
> --------------
> While trying to debug an an issue under extreme I/O loading
> on preempt-rt kernels, the following backtrace was observed
> via SysRQ output:
>
> rm D ffff8802203afbc0 4600 4878 4748 0x00000000
> ffff8802217bfb78 0000000000000082 ffff88021fc2bb80 ffff88021fc2bb80
> ffff88021fc2bb80 ffff8802217bffd8 ffff8802217bffd8 ffff8802217bffd8
> ffff88021f1d4c80 ffff88021fc2bb80 ffff8802217bfb88 ffff88022437b000
> Call Trace:
> [<ffffffff8172dc34>] schedule+0x24/0x70
> [<ffffffff81225b5d>] jbd2_log_wait_commit+0xbd/0x140
> [<ffffffff81060390>] ? __init_waitqueue_head+0x50/0x50
> [<ffffffff81223635>] jbd2_log_do_checkpoint+0xf5/0x520
> [<ffffffff81223b09>] __jbd2_log_wait_for_space+0xa9/0x1f0
> [<ffffffff8121dc40>] start_this_handle.isra.10+0x2e0/0x530
> [<ffffffff81060390>] ? __init_waitqueue_head+0x50/0x50
> [<ffffffff8121e0a3>] jbd2__journal_start+0xc3/0x110
> [<ffffffff811de7ce>] ? ext4_rmdir+0x6e/0x230
> [<ffffffff8121e0fe>] jbd2_journal_start+0xe/0x10
> [<ffffffff811f308b>] ext4_journal_start_sb+0x5b/0x160
> [<ffffffff811de7ce>] ext4_rmdir+0x6e/0x230
> [<ffffffff811435c5>] vfs_rmdir+0xd5/0x140
> [<ffffffff8114370f>] do_rmdir+0xdf/0x120
> [<ffffffff8105c6b4>] ? task_work_run+0x44/0x80
> [<ffffffff81002889>] ? do_notify_resume+0x89/0x100
> [<ffffffff817361ae>] ? int_signal+0x12/0x17
> [<ffffffff81145d85>] sys_unlinkat+0x25/0x40
> [<ffffffff81735f22>] system_call_fastpath+0x16/0x1b
>
> What is interesting here, is that we call log_wait_commit, from
> within wait_for_space, but we are still holding the checkpoint_mutex
> as it surrounds mostly the whole of wait_for_space. And then, as we
> are waiting, journal_commit_transaction can run, and if the JBD2_FLUSHED
> bit is set, then we will also try to take the same checkpoint_mutex.
>
> It seems that we need to drop the checkpoint_mutex while sitting in
> jbd2_log_wait_commit, if we want to guarantee that progress can be made
> by jbd2_journal_commit_transaction(). There does not seem to be
> anything preempt-rt specific about this, other then perhaps increasing
> the odds of it happening.
> --------------
As much as this is a sensible optimization, I don't agree with the
analysis that the system can deadlock when JBD2_FLUSHED is set. We call
jbd2_log_wait_commit() only if there is already committing transaction and
that gets set *after* JBD2_FLUSHED is handled. So it would be good if you
went back and actually checked where exactly did we hang in
jbd2_journal_commit_transaction()...
Honza
>
> Signed-off-by: Paul Gortmaker <paul.gortmaker@xxxxxxxxxxxxx>
> ---
> fs/jbd/checkpoint.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
> index 08c0304..8e8e0ee 100644
> --- a/fs/jbd/checkpoint.c
> +++ b/fs/jbd/checkpoint.c
> @@ -160,7 +160,15 @@ void __log_wait_for_space(journal_t *journal)
> /* We were able to recover space; yay! */
> ;
> } else if (tid) {
> + /*
> + * jbd_journal_commit_transaction() may want
> + * to take the checkpoint_mutex if JBD_FLUSHED
> + * is set. So we need to temporarily drop it.
> + */
> + mutex_unlock(&journal->j_checkpoint_mutex);
> log_wait_commit(journal, tid);
> + spin_lock(&journal->j_state_lock);
> + continue;
> } else {
> printk(KERN_ERR "%s: needed %d blocks and "
> "only had %d space available\n",
> --
> 1.8.1.2
>
--
Jan Kara <jack@xxxxxxx>
SUSE Labs, CR
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
