| Message ID | 1425551805-7314-1-git-send-email-bo.li.liu@oracle.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
On Thu, Mar 5, 2015 at 5:36 AM, Liu Bo <bo.li.liu@oracle.com> wrote: > This problem is uncovered by a test case: > http://patchwork.ozlabs.org/patch/244297. > > Fsync() can report success when it actually doesn't. When we > have several threads running fsync() at the same tiem and in one > fsync() we > get a transaction abortion due to some problems(in the test case it's > disk > failures), and other fsync()s may return successfully which makes > userspace > programs think that data is now safely flushed into disk. > > It's because that after fsyncs() fail btrfs_sync_log() due to disk > failures, > they get to try btrfs_commit_transaction() where it finds that there > is > already a transaction being committed, and they'll just call > wait_for_commit() > and return. Note that we actually check "trans->aborted" in > btrfs_end_transaction, > but it's likely that the error message is still not yet throwed out > and only after > wait_for_commit() we're sure whether the transaction is committed > successfully. > > This add the necessary check and it now passes the test. > > Signed-off-by: Liu Bo <bo.li.liu@oracle.com> > --- > v2: Use a more generic title since it's not only for fsync, but for > others. > > fs/btrfs/transaction.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c > index 7e80f32..bd7ea86 100644 > --- a/fs/btrfs/transaction.c > +++ b/fs/btrfs/transaction.c > @@ -1814,6 +1814,9 @@ int btrfs_commit_transaction(struct > btrfs_trans_handle *trans, > > wait_for_commit(root, cur_trans); > > + if (unlikely(ACCESS_ONCE(cur_trans->aborted))) > + ret = cur_trans->aborted; > + Thanks Liu, but why are we using ACCESS_ONCE here? -chris -- 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 Thu, Mar 05, 2015 at 08:59:57AM -0500, Chris Mason wrote: > > > On Thu, Mar 5, 2015 at 5:36 AM, Liu Bo <bo.li.liu@oracle.com> wrote: > >This problem is uncovered by a test case: > >http://patchwork.ozlabs.org/patch/244297. > > > >Fsync() can report success when it actually doesn't. When we > >have several threads running fsync() at the same tiem and in one > >fsync() we > >get a transaction abortion due to some problems(in the test case > >it's disk > >failures), and other fsync()s may return successfully which makes > >userspace > >programs think that data is now safely flushed into disk. > > > >It's because that after fsyncs() fail btrfs_sync_log() due to disk > >failures, > >they get to try btrfs_commit_transaction() where it finds that > >there is > >already a transaction being committed, and they'll just call > >wait_for_commit() > >and return. Note that we actually check "trans->aborted" in > >btrfs_end_transaction, > >but it's likely that the error message is still not yet throwed > >out and only after > >wait_for_commit() we're sure whether the transaction is committed > >successfully. > > > >This add the necessary check and it now passes the test. > > > >Signed-off-by: Liu Bo <bo.li.liu@oracle.com> > >--- > >v2: Use a more generic title since it's not only for fsync, but > >for others. > > > > fs/btrfs/transaction.c | 3 +++ > > 1 file changed, 3 insertions(+) > > > >diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c > >index 7e80f32..bd7ea86 100644 > >--- a/fs/btrfs/transaction.c > >+++ b/fs/btrfs/transaction.c > >@@ -1814,6 +1814,9 @@ int btrfs_commit_transaction(struct > >btrfs_trans_handle *trans, > > > > wait_for_commit(root, cur_trans); > > > >+ if (unlikely(ACCESS_ONCE(cur_trans->aborted))) > >+ ret = cur_trans->aborted; > >+ > > Thanks Liu, but why are we using ACCESS_ONCE here? It should be not necessary, I just copied it from the first check in btrfs_commit_transaction(), not insisting in using it. Thanks, -liubo -- 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
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 7e80f32..bd7ea86 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1814,6 +1814,9 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, wait_for_commit(root, cur_trans); + if (unlikely(ACCESS_ONCE(cur_trans->aborted))) + ret = cur_trans->aborted; + btrfs_put_transaction(cur_trans); return ret;
This problem is uncovered by a test case: http://patchwork.ozlabs.org/patch/244297. Fsync() can report success when it actually doesn't. When we have several threads running fsync() at the same tiem and in one fsync() we get a transaction abortion due to some problems(in the test case it's disk failures), and other fsync()s may return successfully which makes userspace programs think that data is now safely flushed into disk. It's because that after fsyncs() fail btrfs_sync_log() due to disk failures, they get to try btrfs_commit_transaction() where it finds that there is already a transaction being committed, and they'll just call wait_for_commit() and return. Note that we actually check "trans->aborted" in btrfs_end_transaction, but it's likely that the error message is still not yet throwed out and only after wait_for_commit() we're sure whether the transaction is committed successfully. This add the necessary check and it now passes the test. Signed-off-by: Liu Bo <bo.li.liu@oracle.com> --- v2: Use a more generic title since it's not only for fsync, but for others. fs/btrfs/transaction.c | 3 +++ 1 file changed, 3 insertions(+)