xfs: Fix error return for fallocate() on XFS
authorJason Gunthorpe <jgunthorpe@obsidianresearch.com>
Tue, 24 Nov 2009 21:52:53 +0000 (21:52 +0000)
committerAlex Elder <aelder@sgi.com>
Fri, 11 Dec 2009 21:11:23 +0000 (15:11 -0600)
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@obsidianresearch.com>
Reviewed-by: Eric Sandeen <sandeen@sandeen.net>
Signed-off-by: Alex Elder <aelder@sgi.com>
fs/xfs/linux-2.6/xfs_iops.c

index cd42ef78f6b54e705ad604e3ccfc4eb1a3df3743..1f3b4b8f7dd45195b71e41f6ea4df2a6a0b110c9 100644 (file)
@@ -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);