summaryrefslogtreecommitdiff
path: root/fs/xfs/xfs_vnodeops.c
diff options
context:
space:
mode:
authorLachlan McIlroy <lachlan@sgi.com>2007-05-08 13:49:46 +1000
committerTim Shimmin <tes@sgi.com>2007-05-08 13:49:46 +1000
commitba87ea699ebd9dd577bf055ebc4a98200e337542 (patch)
tree713b7d32937372fd7c5b8647f14d0e7262fc7075 /fs/xfs/xfs_vnodeops.c
parent2a32963130aec5e157b58ff7dfa3dfa1afdf7ca1 (diff)
downloadlwn-ba87ea699ebd9dd577bf055ebc4a98200e337542.tar.gz
lwn-ba87ea699ebd9dd577bf055ebc4a98200e337542.zip
[XFS] Fix to prevent the notorious 'NULL files' problem after a crash.
The problem that has been addressed is that of synchronising updates of the file size with writes that extend a file. Without the fix the update of a file's size, as a result of a write beyond eof, is independent of when the cached data is flushed to disk. Often the file size update would be written to the filesystem log before the data is flushed to disk. When a system crashes between these two events and the filesystem log is replayed on mount the file's size will be set but since the contents never made it to disk the file is full of holes. If some of the cached data was flushed to disk then it may just be a section of the file at the end that has holes. There are existing fixes to help alleviate this problem, particularly in the case where a file has been truncated, that force cached data to be flushed to disk when the file is closed. If the system crashes while the file(s) are still open then this flushing will never occur. The fix that we have implemented is to introduce a second file size, called the in-memory file size, that represents the current file size as viewed by the user. The existing file size, called the on-disk file size, is the one that get's written to the filesystem log and we only update it when it is safe to do so. When we write to a file beyond eof we only update the in- memory file size in the write operation. Later when the I/O operation, that flushes the cached data to disk completes, an I/O completion routine will update the on-disk file size. The on-disk file size will be updated to the maximum offset of the I/O or to the value of the in-memory file size if the I/O includes eof. SGI-PV: 958522 SGI-Modid: xfs-linux-melb:xfs-kern:28322a Signed-off-by: Lachlan McIlroy <lachlan@sgi.com> Signed-off-by: David Chinner <dgc@sgi.com> Signed-off-by: Tim Shimmin <tes@sgi.com>
Diffstat (limited to 'fs/xfs/xfs_vnodeops.c')
-rw-r--r--fs/xfs/xfs_vnodeops.c40
1 files changed, 20 insertions, 20 deletions
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index 6e49bd362460..e17be3b647be 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -133,7 +133,7 @@ xfs_getattr(
if (!(flags & ATTR_LAZY))
xfs_ilock(ip, XFS_ILOCK_SHARED);
- vap->va_size = ip->i_d.di_size;
+ vap->va_size = XFS_ISIZE(ip);
if (vap->va_mask == XFS_AT_SIZE)
goto all_done;
@@ -496,7 +496,7 @@ xfs_setattr(
if (mask & XFS_AT_SIZE) {
/* Short circuit the truncate case for zero length files */
if ((vap->va_size == 0) &&
- (ip->i_d.di_size == 0) && (ip->i_d.di_nextents == 0)) {
+ (ip->i_size == 0) && (ip->i_d.di_nextents == 0)) {
xfs_iunlock(ip, XFS_ILOCK_EXCL);
lock_flags &= ~XFS_ILOCK_EXCL;
if (mask & XFS_AT_CTIME)
@@ -614,7 +614,7 @@ xfs_setattr(
*/
if (mask & XFS_AT_SIZE) {
code = 0;
- if ((vap->va_size > ip->i_d.di_size) &&
+ if ((vap->va_size > ip->i_size) &&
(flags & ATTR_NOSIZETOK) == 0) {
code = xfs_igrow_start(ip, vap->va_size, credp);
}
@@ -654,10 +654,10 @@ xfs_setattr(
* Truncate file. Must have write permission and not be a directory.
*/
if (mask & XFS_AT_SIZE) {
- if (vap->va_size > ip->i_d.di_size) {
+ if (vap->va_size > ip->i_size) {
xfs_igrow_finish(tp, ip, vap->va_size,
!(flags & ATTR_DMI));
- } else if ((vap->va_size <= ip->i_d.di_size) ||
+ } else if ((vap->va_size <= ip->i_size) ||
((vap->va_size == 0) && ip->i_d.di_nextents)) {
/*
* signal a sync transaction unless
@@ -1221,7 +1221,7 @@ xfs_inactive_free_eofblocks(
* Figure out if there are any blocks beyond the end
* of the file. If not, then there is nothing to do.
*/
- end_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)ip->i_d.di_size));
+ end_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)ip->i_size));
last_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_MAXIOFFSET(mp));
map_len = last_fsb - end_fsb;
if (map_len <= 0)
@@ -1258,7 +1258,7 @@ xfs_inactive_free_eofblocks(
*/
xfs_ilock(ip, XFS_IOLOCK_EXCL);
error = xfs_itruncate_start(ip, XFS_ITRUNC_DEFINITE,
- ip->i_d.di_size);
+ ip->i_size);
if (error) {
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return error;
@@ -1282,7 +1282,7 @@ xfs_inactive_free_eofblocks(
xfs_trans_ihold(tp, ip);
error = xfs_itruncate_finish(&tp, ip,
- ip->i_d.di_size,
+ ip->i_size,
XFS_DATA_FORK,
0);
/*
@@ -1568,7 +1568,7 @@ xfs_release(
if (ip->i_d.di_nlink != 0) {
if ((((ip->i_d.di_mode & S_IFMT) == S_IFREG) &&
- ((ip->i_d.di_size > 0) || (VN_CACHED(vp) > 0 ||
+ ((ip->i_size > 0) || (VN_CACHED(vp) > 0 ||
ip->i_delayed_blks > 0)) &&
(ip->i_df.if_flags & XFS_IFEXTENTS)) &&
(!(ip->i_d.di_flags &
@@ -1629,8 +1629,8 @@ xfs_inactive(
* only one with a reference to the inode.
*/
truncate = ((ip->i_d.di_nlink == 0) &&
- ((ip->i_d.di_size != 0) || (ip->i_d.di_nextents > 0) ||
- (ip->i_delayed_blks > 0)) &&
+ ((ip->i_d.di_size != 0) || (ip->i_size != 0) ||
+ (ip->i_d.di_nextents > 0) || (ip->i_delayed_blks > 0)) &&
((ip->i_d.di_mode & S_IFMT) == S_IFREG));
mp = ip->i_mount;
@@ -1648,7 +1648,7 @@ xfs_inactive(
if (ip->i_d.di_nlink != 0) {
if ((((ip->i_d.di_mode & S_IFMT) == S_IFREG) &&
- ((ip->i_d.di_size > 0) || (VN_CACHED(vp) > 0 ||
+ ((ip->i_size > 0) || (VN_CACHED(vp) > 0 ||
ip->i_delayed_blks > 0)) &&
(ip->i_df.if_flags & XFS_IFEXTENTS) &&
(!(ip->i_d.di_flags &
@@ -4055,14 +4055,14 @@ xfs_alloc_file_space(
allocatesize_fsb = XFS_B_TO_FSB(mp, count);
/* Generate a DMAPI event if needed. */
- if (alloc_type != 0 && offset < ip->i_d.di_size &&
+ if (alloc_type != 0 && offset < ip->i_size &&
(attr_flags&ATTR_DMI) == 0 &&
DM_EVENT_ENABLED(XFS_MTOVFS(mp), ip, DM_EVENT_WRITE)) {
xfs_off_t end_dmi_offset;
end_dmi_offset = offset+len;
- if (end_dmi_offset > ip->i_d.di_size)
- end_dmi_offset = ip->i_d.di_size;
+ if (end_dmi_offset > ip->i_size)
+ end_dmi_offset = ip->i_size;
error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, XFS_ITOV(ip),
offset, end_dmi_offset - offset,
0, NULL);
@@ -4318,11 +4318,11 @@ xfs_free_file_space(
end_dmi_offset = offset + len;
endoffset_fsb = XFS_B_TO_FSBT(mp, end_dmi_offset);
- if (offset < ip->i_d.di_size &&
+ if (offset < ip->i_size &&
(attr_flags & ATTR_DMI) == 0 &&
DM_EVENT_ENABLED(XFS_MTOVFS(mp), ip, DM_EVENT_WRITE)) {
- if (end_dmi_offset > ip->i_d.di_size)
- end_dmi_offset = ip->i_d.di_size;
+ if (end_dmi_offset > ip->i_size)
+ end_dmi_offset = ip->i_size;
error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, vp,
offset, end_dmi_offset - offset,
AT_DELAY_FLAG(attr_flags), NULL);
@@ -4541,7 +4541,7 @@ xfs_change_file_space(
bf->l_start += offset;
break;
case 2: /*SEEK_END*/
- bf->l_start += ip->i_d.di_size;
+ bf->l_start += ip->i_size;
break;
default:
return XFS_ERROR(EINVAL);
@@ -4558,7 +4558,7 @@ xfs_change_file_space(
bf->l_whence = 0;
startoffset = bf->l_start;
- fsize = ip->i_d.di_size;
+ fsize = ip->i_size;
/*
* XFS_IOC_RESVSP and XFS_IOC_UNRESVSP will reserve or unreserve