[PATCH v2] xfs: Fix overallocation in xfs_buf_allocate_memory()

[PATCH v2] xfs: Fix overallocation in xfs_buf_allocate_memory()

[Jan Kara]

Commit de1cbee which removed b_file_offset in favor of b_bn introduced a bug
causing xfs_buf_allocate_memory() to overestimate the number of necessary
pages. The problem is that xfs_buf_alloc() sets b_bn to -1 and thus effectively
every buffer is straddling a page boundary which causes
xfs_buf_allocate_memory() to allocate two pages and use vmalloc() for access
which is unnecessary.

Dave says xfs_buf_alloc() doesn't need to set b_bn to -1 anymore since the
buffer is inserted into the cache only after being fully initialized now.
So just make xfs_buf_alloc() fill in proper block number from the beginning.

CC: David Chinner <dchinner@redhat.com>
Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/xfs/xfs_buf.c |   16 ++--------------
 1 files changed, 2 insertions(+), 14 deletions(-)

diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 172d3cc..a4beb42 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -201,14 +201,7 @@ xfs_buf_alloc(
 	bp->b_length = numblks;
 	bp->b_io_length = numblks;
 	bp->b_flags = flags;
-
-	/*
-	 * We do not set the block number here in the buffer because we have not
-	 * finished initialising the buffer. We insert the buffer into the cache
-	 * in this state, so this ensures that we are unable to do IO on a
-	 * buffer that hasn't been fully initialised.
-	 */
-	bp->b_bn = XFS_BUF_DADDR_NULL;
+	bp->b_bn = blkno;
 	atomic_set(&bp->b_pin_count, 0);
 	init_waitqueue_head(&bp->b_waiters);
 
@@ -567,11 +560,6 @@ xfs_buf_get(
 	if (bp != new_bp)
 		xfs_buf_free(new_bp);
 
-	/*
-	 * Now we have a workable buffer, fill in the block number so
-	 * that we can do IO on it.
-	 */
-	bp->b_bn = blkno;
 	bp->b_io_length = bp->b_length;
 
 found:
@@ -772,7 +760,7 @@ xfs_buf_get_uncached(
 	int			error, i;
 	xfs_buf_t		*bp;
 
-	bp = xfs_buf_alloc(target, 0, numblks, 0);
+	bp = xfs_buf_alloc(target, XFS_BUF_DADDR_NULL, numblks, 0);
 	if (unlikely(bp == NULL))
 		goto fail;
 
-- 
1.7.1

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH v2] xfs: Fix overallocation in xfs_buf_allocate_memory()

[Christoph Hellwig]

On Wed, Jun 06, 2012 at 12:32:26AM +0200, Jan Kara wrote:
> Commit de1cbee which removed b_file_offset in favor of b_bn introduced a bug
> causing xfs_buf_allocate_memory() to overestimate the number of necessary
> pages. The problem is that xfs_buf_alloc() sets b_bn to -1 and thus effectively
> every buffer is straddling a page boundary which causes
> xfs_buf_allocate_memory() to allocate two pages and use vmalloc() for access
> which is unnecessary.
> 
> Dave says xfs_buf_alloc() doesn't need to set b_bn to -1 anymore since the
> buffer is inserted into the cache only after being fully initialized now.
> So just make xfs_buf_alloc() fill in proper block number from the beginning.

Looks good.  The b_io_length initialization in xfs_buf_get can also be
removed, but as that's a cosmetic change it can way for 3.6.

Reviewed-by: Christoph Hellwig <hch@lst.de>

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH v2] xfs: Fix overallocation in xfs_buf_allocate_memory()

[Dave Chinner]

On Wed, Jun 06, 2012 at 12:32:26AM +0200, Jan Kara wrote:
> Commit de1cbee which removed b_file_offset in favor of b_bn introduced a bug
> causing xfs_buf_allocate_memory() to overestimate the number of necessary
> pages. The problem is that xfs_buf_alloc() sets b_bn to -1 and thus effectively
> every buffer is straddling a page boundary which causes
> xfs_buf_allocate_memory() to allocate two pages and use vmalloc() for access
> which is unnecessary.
> 
> Dave says xfs_buf_alloc() doesn't need to set b_bn to -1 anymore since the
> buffer is inserted into the cache only after being fully initialized now.
> So just make xfs_buf_alloc() fill in proper block number from the beginning.
> 
> CC: David Chinner <dchinner@redhat.com>
> Signed-off-by: Jan Kara <jack@suse.cz>

Hmmm, the

Reviewed-by: Dave Chinner <dchinner@redhat.com>

I sent yesterday doesn't appear to have made it to the list. Oh
well, try again...

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs