Bug 544349

When issuing an fallocate call on xfs which results in insufficient space to complete, XFS will return "28" instead of "ENOSPC" - xfs uses positive errnos internally, and flips them before returning; but in this case it was missed.

It's a trivial fix and easily testable; I know that it's late in RHEL5.5 but we've been touting preallocation as major xfs feature, and this aspect is broken, so would like to get this late-breaking bug fixed in 5.5 if possible.

Thanks,

-Eric

----------

Noticed that through glibc fallocate would return 28 rather than -1
and errno = 28 for ENOSPC. The xfs routines uses XFS_ERROR format
positive return error codes while the syscalls use negative return codes.
Fixup the two cases in xfs_vn_fallocate syscall to convert to negative.

Signed-off-by: Jason Gunthorpe <jgunthorpe>
---
 fs/xfs/linux-2.6/xfs_iops.c |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index cd42ef7..1f3b4b8 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -573,8 +573,8 @@ xfs_vn_fallocate(
 	bf.l_len = len;
 
 	xfs_ilock(ip, XFS_IOLOCK_EXCL);
-	error = xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
-				      0, XFS_ATTR_NOLOCK);
+	error = -xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
+				       0, XFS_ATTR_NOLOCK);
 	if (!error && !(mode & FALLOC_FL_KEEP_SIZE) &&
 	    offset + len > i_size_read(inode))
 		new_size = offset + len;
@@ -585,7 +585,7 @@ xfs_vn_fallocate(
 
 		iattr.ia_valid = ATTR_SIZE;
 		iattr.ia_size = new_size;
-		error = xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
+		error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
 	}
 
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);