Directories are only updated transactionally, which means fsync only
needs to flush the log the inode is currently dirty, but not bother
with checking for dirty data, non-transactional updates, and most
importanly doesn't have to flush disk caches except as part of a
transaction commit.
While the first two optimizations can't easily be measured, the
latter actually makes a difference when doing lots of fsync that do
not actually have to commit the inode, e.g. because an earlier fsync
already pushed the log far enough.
The new xfs_dir_fsync is identical to xfs_nfs_commit_metadata except
for the prototype, but I'm not sure creating a common helper for the
two is worth it given how simple the functions are.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
There is no reason to keep a reference to the inode even if we unlock
it during transaction commit because we never drop a reference between
the ijoin and commit. Also use this fact to merge xfs_trans_ijoin_ref
back into xfs_trans_ijoin - the third argument decides if an unlock
is needed now.
I'm actually starting to wonder if allowing inodes to be unlocked
at transaction commit really is worth the effort. The only real
benefit is that they can be unlocked earlier when commiting a
synchronous transactions, but that could be solved by doing the
log force manually after the unlock, too.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
Only read the LSN we need to push to with the ilock held, and then release
it before we do the log force to improve concurrency.
This also removes the only direct caller of _xfs_trans_commit, thus
allowing it to be merged into the plain xfs_trans_commit again.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
xfs_bmapi() currently handles both extent map reading and
allocation. As a result, the code is littered with "if (wr)"
branches to conditionally do allocation operations if required.
This makes the code much harder to follow and causes significant
indent issues with the code.
Given that read mapping is much simpler than allocation, we can
split out read mapping from xfs_bmapi() and reuse the logic that
we have already factored out do do all the hard work of handling the
extent map manipulations. The results in a much simpler function for
the common extent read operations, and will allow the allocation
code to be simplified in another commit.
Once xfs_bmapi_read() is implemented, convert all the callers of
xfs_bmapi() that are only reading extents to use the new function.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
Currently a buffered reader or writer can add pages to the pagecache
while we are waiting for the iolock in xfs_file_dio_aio_write. Prevent
this by re-checking mapping->nrpages after we got the iolock, and if
nessecary upgrade the lock to exclusive mode. To simplify this a bit
only take the ilock inside of xfs_file_aio_write_checks.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
We now have an i_dio_count filed and surrounding infrastructure to wait
for direct I/O completion instead of i_icount, and we have never needed
to iocount waits for buffered I/O given that we only set the page uptodate
after finishing all required work. Thus remove i_iocount, and replace
the actually needed waits with calls to inode_dio_wait.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
For append write workloads, extending the file requires a certain
amount of exclusive locking to be done up front to ensure sanity in
things like ensuring that we've zeroed any allocated regions
between the old EOF and the start of the new IO.
For single threads, this typically isn't a problem, and for large
IOs we don't serialise enough for it to be a problem for two
threads on really fast block devices. However for smaller IO and
larger thread counts we have a problem.
Take 4 concurrent sequential, single block sized and aligned IOs.
After the first IO is submitted but before it completes, we end up
with this state:
IO 1 IO 2 IO 3 IO 4
+-------+-------+-------+-------+
^ ^
| |
| |
| |
| \- ip->i_new_size
\- ip->i_size
And the IO is done without exclusive locking because offset <=
ip->i_size. When we submit IO 2, we see offset > ip->i_size, and
grab the IO lock exclusive, because there is a chance we need to do
EOF zeroing. However, there is already an IO in progress that avoids
the need for IO zeroing because offset <= ip->i_new_size. hence we
could avoid holding the IO lock exlcusive for this. Hence after
submission of the second IO, we'd end up this state:
IO 1 IO 2 IO 3 IO 4
+-------+-------+-------+-------+
^ ^
| |
| |
| |
| \- ip->i_new_size
\- ip->i_size
There is no need to grab the i_mutex of the IO lock in exclusive
mode if we don't need to invalidate the page cache. Taking these
locks on every direct IO effective serialises them as taking the IO
lock in exclusive mode has to wait for all shared holders to drop
the lock. That only happens when IO is complete, so effective it
prevents dispatch of concurrent direct IO writes to the same inode.
And so you can see that for the third concurrent IO, we'd avoid
exclusive locking for the same reason we avoided the exclusive lock
for the second IO.
Fixing this is a bit more complex than that, because we need to hold
a write-submission local value of ip->i_new_size to that clearing
the value is only done if no other thread has updated it before our
IO completes.....
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
There is no need to grab the i_mutex of the IO lock in exclusive
mode if we don't need to invalidate the page cache. Taking these
locks on every direct IO effective serialises them as taking the IO
lock in exclusive mode has to wait for all shared holders to drop
the lock. That only happens when IO is complete, so effective it
prevents dispatch of concurrent direct IO reads to the same inode.
Fix this by taking the IO lock shared to check the page cache state,
and only then drop it and take the IO lock exclusively if there is
work to be done. Hence for the normal direct IO case, no exclusive
locking will occur.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Tested-by: Joern Engel <joern@logfs.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
Use the move from Linux 2.6 to Linux 3.x as an excuse to kill the
annoying subdirectories in the XFS source code. Besides the large
amount of file rename the only changes are to the Makefile, a few
files including headers with the subdirectory prefix, and the binary
sysctl compat code that includes a header under fs/xfs/ from
kernel/.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>
Christoph Hellwig