Btrfs devel - Jun 2011 - [PATCH] Btrfs: set no_trans_join after trying to expand the transaction

We can lockup if we try to allow new writers join the transaction and we have
flushoncommit set or have a pending snapshot.  This is because we set
no_trans_join and then loop around and try to wait for ordered extents again.
The problem is the ordered endio stuff needs to join the transaction, which it
can''t do because no_trans_join is set.  So instead wait until after
this loop to
set no_trans_join and then make sure to wait for num_writers == 1 in case
anybody got started in between us exiting the loop and setting no_trans_join.
This could easily be reproduced by mounting -o flushoncommit and running xfstest
13.  It cannot be reproduced with this patch.  Thanks,

Reported-by: Jim Schutt <jaschut@sandia.gov>
Signed-off-by: Josef Bacik <josef@redhat.com>
---
 fs/btrfs/transaction.c |   14 +++++++++++---
 1 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 8754997..d0f1c07 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -1239,12 +1239,20 @@ int btrfs_commit_transaction(struct btrfs_trans_handle
*trans,
 			schedule_timeout(1);
 
 		finish_wait(&cur_trans->writer_wait, &wait);
-		spin_lock(&root->fs_info->trans_lock);
-		root->fs_info->trans_no_join = 1;
-		spin_unlock(&root->fs_info->trans_lock);
 	} while (atomic_read(&cur_trans->num_writers) > 1 ||
 		 (should_grow && cur_trans->num_joined != joined));
 
+	/*
+	 * Ok now we need to make sure to block out any other joins while we
+	 * commit the transaction.  We could have started a join before setting
+	 * no_join so make sure to wait for num_writers to == 1 again.
+	 */
+	spin_lock(&root->fs_info->trans_lock);
+	root->fs_info->trans_no_join = 1;
+	spin_unlock(&root->fs_info->trans_lock);
+	wait_event(cur_trans->writer_wait,
+		   atomic_read(&cur_trans->num_writers) == 1);
+
 	ret = create_pending_snapshots(trans, root->fs_info);
 	BUG_ON(ret);
 
-- 
1.7.5.2

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Josef Bacik wrote:
> We can lockup if we try to allow new writers join the transaction and we
have
> flushoncommit set or have a pending snapshot.  This is because we set
> no_trans_join and then loop around and try to wait for ordered extents
again.
> The problem is the ordered endio stuff needs to join the transaction, which
it
> can''t do because no_trans_join is set.  So instead wait until
after this loop to
> set no_trans_join and then make sure to wait for num_writers == 1 in case
> anybody got started in between us exiting the loop and setting
no_trans_join.
> This could easily be reproduced by mounting -o flushoncommit and running
xfstest
> 13.  It cannot be reproduced with this patch.  Thanks,
FWIW, this version of the patch works fine for me
as well.

-- Jim
> 
> Reported-by: Jim Schutt <jaschut@sandia.gov>
> Signed-off-by: Josef Bacik <josef@redhat.com>
> ---
>  fs/btrfs/transaction.c |   14 +++++++++++---
>  1 files changed, 11 insertions(+), 3 deletions(-)
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On 06/15/2011 10:36 AM, Jim Schutt wrote:
> 
> Josef Bacik wrote:
>> We can lockup if we try to allow new writers join the transaction and
>> we have
>> flushoncommit set or have a pending snapshot.  This is because we set
>> no_trans_join and then loop around and try to wait for ordered extents
>> again.
>> The problem is the ordered endio stuff needs to join the transaction,
>> which it
>> can''t do because no_trans_join is set.  So instead wait until
after
>> this loop to
>> set no_trans_join and then make sure to wait for num_writers == 1 in
case
>> anybody got started in between us exiting the loop and setting
>> no_trans_join.
>> This could easily be reproduced by mounting -o flushoncommit and
>> running xfstest
>> 13.  It cannot be reproduced with this patch.  Thanks,
> 
> FWIW, this version of the patch works fine for me
> as well.
> 
Great, thanks for testing.

Josef
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html