* [PATCH] fix extent length in xfs_io bmap
@ 2007-01-16 5:37 Utako Kusaka
2007-01-22 7:14 ` Barry Naujok
0 siblings, 1 reply; 2+ messages in thread
From: Utako Kusaka @ 2007-01-16 5:37 UTC (permalink / raw)
To: xfs
Hi,
In bmap command in xfs_io, there is a difference in the length of the extent group
between "bmap" and "bmap -n nn".
It occurs if the file size > max extent size and the extents are allocated contiguously.
This patch fixes it.
Test fs:
# xfs_info /dev/sx8/14p1
meta-data=/dev/sx8/14p1 isize=256 agcount=16, agsize=7631000 blks
= sectsz=512 attr=0
data = bsize=4096 blocks=122096000, imaxpct=25
= sunit=0 swidth=0 blks, unwritten=1
naming =version 2 bsize=4096
log =internal bsize=4096 blocks=32768, version=1
= sectsz=512 sunit=0 blks
realtime =none extsz=4096 blocks=0, rtextents=0
Example:
u.bmx[0-5] = [startoff,startblock,blockcount,extentflag]
0:[ 0, 1048588,2097088,0]
1:[2097088, 3145676,2097088,0]
2:[4194176, 5242764,2097088,0]
3:[6291264, 7339852, 291136,0]
4:[6582400, 8388616,2097088,0]
5:[8679488,10485704,1806272,0]
# xfs_io file2
xfs_io> bmap -v
file2:
EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL
0: [0..52659199]: 8388704..61047903 0 (8388704..61047903) 52659200 ...*
1: [52659200..83886079]: 61048064..92274943 1 (64..31226943) 31226880 ...**
xfs_io> bmap -v -n 1
file2:
EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL
0: [0..16776703]: 8388704..25165407 0 (8388704..25165407) 16776704 ...*
xfs_io> bmap -v -n 2
file2:
EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL
0: [0..52659199]: 8388704..61047903 0 (8388704..61047903) 52659200
1: [52659200..69435903]: 61048064..77824767 1 (64..16776767) 16776704 ...**
Signed-off-by: Utako Kusaka <utako@tnes.nec.co.jp>
---
--- xfsprogs-2.8.18-orgn/io/bmap.c 2006-12-13 13:57:22.000000000 +0900
+++ xfsprogs-2.8.18/io/bmap.c 2006-12-21 11:07:09.573116475 +0900
@@ -78,6 +78,7 @@ bmap_f(
int bmv_iflags = 0; /* flags for XFS_IOC_GETBMAPX */
int i = 0;
int c;
+ int egcnt;
while ((c = getopt(argc, argv, "adln:pv")) != EOF) {
switch (c) {
@@ -136,7 +137,7 @@ bmap_f(
}
}
- map_size = nflag ? nflag+1 : 32; /* initial guess - 256 */
+ map_size = nflag ? nflag+2 : 32; /* initial guess - 256 */
map = malloc(map_size*sizeof(*map));
if (map == NULL) {
fprintf(stderr, _("%s: malloc of %d bytes failed.\n"),
@@ -232,9 +233,10 @@ bmap_f(
return 0;
}
}
+ egcnt = nflag ? min(nflag, map->bmv_entries) : map->bmv_entries;
printf("%s:\n", file->name);
if (!vflag) {
- for (i = 0; i < map->bmv_entries; i++) {
+ for (i = 0; i < egcnt; i++) {
printf("\t%d: [%lld..%lld]: ", i,
(long long) map[i + 1].bmv_offset,
(long long)(map[i + 1].bmv_offset +
@@ -288,7 +290,7 @@ bmap_f(
* Go through the extents and figure out the width
* needed for all columns.
*/
- for (i = 0; i < map->bmv_entries; i++) {
+ for (i = 0; i < egcnt; i++) {
snprintf(rbuf, sizeof(rbuf), "[%lld..%lld]:",
(long long) map[i + 1].bmv_offset,
(long long)(map[i + 1].bmv_offset +
@@ -325,7 +327,7 @@ bmap_f(
aoff_w, _("AG-OFFSET"),
tot_w, _("TOTAL"),
flg ? _(" FLAGS") : "");
- for (i = 0; i < map->bmv_entries; i++) {
+ for (i = 0; i < egcnt; i++) {
flg = FLG_NULL;
if (map[i + 1].bmv_oflags & BMV_OF_PREALLOC) {
flg |= FLG_PRE;
^ permalink raw reply [flat|nested] 2+ messages in thread* RE: [PATCH] fix extent length in xfs_io bmap
2007-01-16 5:37 [PATCH] fix extent length in xfs_io bmap Utako Kusaka
@ 2007-01-22 7:14 ` Barry Naujok
0 siblings, 0 replies; 2+ messages in thread
From: Barry Naujok @ 2007-01-22 7:14 UTC (permalink / raw)
To: 'Utako Kusaka', xfs
Hi Utako,
I'll push this patch in for the next xfsprogs update.
Thanks,
Barry.
> -----Original Message-----
> From: xfs-bounce@oss.sgi.com [mailto:xfs-bounce@oss.sgi.com]
> On Behalf Of Utako Kusaka
> Sent: Tuesday, 16 January 2007 4:37 PM
> To: xfs@oss.sgi.com
> Subject: [PATCH] fix extent length in xfs_io bmap
>
> Hi,
>
> In bmap command in xfs_io, there is a difference in the
> length of the extent group
> between "bmap" and "bmap -n nn".
> It occurs if the file size > max extent size and the extents
> are allocated contiguously.
> This patch fixes it.
>
> Test fs:
> # xfs_info /dev/sx8/14p1
> meta-data=/dev/sx8/14p1 isize=256 agcount=16,
> agsize=7631000 blks
> = sectsz=512 attr=0
> data = bsize=4096
> blocks=122096000, imaxpct=25
> = sunit=0 swidth=0 blks,
> unwritten=1
> naming =version 2 bsize=4096
> log =internal bsize=4096 blocks=32768, version=1
> = sectsz=512 sunit=0 blks
> realtime =none extsz=4096 blocks=0, rtextents=0
>
>
> Example:
> u.bmx[0-5] = [startoff,startblock,blockcount,extentflag]
> 0:[ 0, 1048588,2097088,0]
> 1:[2097088, 3145676,2097088,0]
> 2:[4194176, 5242764,2097088,0]
> 3:[6291264, 7339852, 291136,0]
> 4:[6582400, 8388616,2097088,0]
> 5:[8679488,10485704,1806272,0]
>
> # xfs_io file2
> xfs_io> bmap -v
> file2:
> EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET
> TOTAL
> 0: [0..52659199]: 8388704..61047903 0
> (8388704..61047903) 52659200 ...*
> 1: [52659200..83886079]: 61048064..92274943 1
> (64..31226943) 31226880 ...**
> xfs_io> bmap -v -n 1
> file2:
> EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET
> TOTAL
> 0: [0..16776703]: 8388704..25165407 0
> (8388704..25165407) 16776704 ...*
> xfs_io> bmap -v -n 2
> file2:
> EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET
> TOTAL
> 0: [0..52659199]: 8388704..61047903 0
> (8388704..61047903) 52659200
> 1: [52659200..69435903]: 61048064..77824767 1
> (64..16776767) 16776704 ...**
>
>
> Signed-off-by: Utako Kusaka <utako@tnes.nec.co.jp>
> ---
>
> --- xfsprogs-2.8.18-orgn/io/bmap.c 2006-12-13
> 13:57:22.000000000 +0900
> +++ xfsprogs-2.8.18/io/bmap.c 2006-12-21 11:07:09.573116475 +0900
> @@ -78,6 +78,7 @@ bmap_f(
> int bmv_iflags = 0; /* flags for
> XFS_IOC_GETBMAPX */
> int i = 0;
> int c;
> + int egcnt;
>
> while ((c = getopt(argc, argv, "adln:pv")) != EOF) {
> switch (c) {
> @@ -136,7 +137,7 @@ bmap_f(
> }
> }
>
> - map_size = nflag ? nflag+1 : 32; /* initial
> guess - 256 */
> + map_size = nflag ? nflag+2 : 32; /* initial
> guess - 256 */
> map = malloc(map_size*sizeof(*map));
> if (map == NULL) {
> fprintf(stderr, _("%s: malloc of %d bytes failed.\n"),
> @@ -232,9 +233,10 @@ bmap_f(
> return 0;
> }
> }
> + egcnt = nflag ? min(nflag, map->bmv_entries) : map->bmv_entries;
> printf("%s:\n", file->name);
> if (!vflag) {
> - for (i = 0; i < map->bmv_entries; i++) {
> + for (i = 0; i < egcnt; i++) {
> printf("\t%d: [%lld..%lld]: ", i,
> (long long) map[i + 1].bmv_offset,
> (long long)(map[i + 1].bmv_offset +
> @@ -288,7 +290,7 @@ bmap_f(
> * Go through the extents and figure out the width
> * needed for all columns.
> */
> - for (i = 0; i < map->bmv_entries; i++) {
> + for (i = 0; i < egcnt; i++) {
> snprintf(rbuf, sizeof(rbuf), "[%lld..%lld]:",
> (long long) map[i + 1].bmv_offset,
> (long long)(map[i + 1].bmv_offset +
> @@ -325,7 +327,7 @@ bmap_f(
> aoff_w, _("AG-OFFSET"),
> tot_w, _("TOTAL"),
> flg ? _(" FLAGS") : "");
> - for (i = 0; i < map->bmv_entries; i++) {
> + for (i = 0; i < egcnt; i++) {
> flg = FLG_NULL;
> if (map[i + 1].bmv_oflags & BMV_OF_PREALLOC) {
> flg |= FLG_PRE;
>
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2007-01-22 7:09 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-01-16 5:37 [PATCH] fix extent length in xfs_io bmap Utako Kusaka
2007-01-22 7:14 ` Barry Naujok
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox