Whamcloud - gitweb
fs/lustre-release.git
20 years ago2.4.21-suse-171 patches and series
phil [Sat, 17 Apr 2004 05:05:06 +0000 (05:05 +0000)]
2.4.21-suse-171 patches and series

20 years agoLanding HEAD changes for 2.6 onto b1_2 - kernel patches only.
adilger [Fri, 16 Apr 2004 17:38:58 +0000 (17:38 +0000)]
Landing HEAD changes for 2.6 onto b1_2 - kernel patches only.
This gets the diff under half a million lines ;-).

20 years agoFix braino.
adilger [Fri, 16 Apr 2004 16:55:11 +0000 (16:55 +0000)]
Fix braino.
b=3159
r=zab

20 years agoUpdate build version.
adilger [Fri, 16 Apr 2004 01:32:44 +0000 (01:32 +0000)]
Update build version.

20 years agoPrint out lock handle that causes assertion failure.
adilger [Fri, 16 Apr 2004 01:02:14 +0000 (01:02 +0000)]
Print out lock handle that causes assertion failure.
b=3159

20 years agoThis commit was manufactured by cvs2svn to create branch 'b1_2'.
cvs2svn [Wed, 14 Apr 2004 20:39:46 +0000 (20:39 +0000)]
This commit was manufactured by cvs2svn to create branch 'b1_2'.

20 years agoLand b_smallfix onto HEAD (20040414_1359)
adilger [Wed, 14 Apr 2004 20:39:45 +0000 (20:39 +0000)]
Land b_smallfix onto HEAD (20040414_1359)
b=3116, b=3064, b=3107, b=3122, b=1519, 2.6 kernel fixes, group lock fixes

20 years agoFix the RPC lprocfs stats. The lprocfs entries need to be created before
adilger [Wed, 14 Apr 2004 20:36:54 +0000 (20:36 +0000)]
Fix the RPC lprocfs stats.  The lprocfs entries need to be created before
we call ptlrpc_init_svc() or it will silently not create any "stats" entry.
b=1519

20 years agoAdd a common error handling function for AST errors. This fixes a case where
adilger [Wed, 14 Apr 2004 20:29:30 +0000 (20:29 +0000)]
Add a common error handling function for AST errors.  This fixes a case where
we don't drop locks on failed completion ASTs and end up evicting the client.
b=3145

20 years agofix compiler warning
nic [Wed, 14 Apr 2004 19:01:38 +0000 (19:01 +0000)]
fix compiler warning

20 years agoDon't dereference negative dentry.
adilger [Wed, 14 Apr 2004 16:33:17 +0000 (16:33 +0000)]
Don't dereference negative dentry.
b=3106

20 years ago- tagging RC_CURRENT
buffalo-pull [Tue, 13 Apr 2004 23:26:40 +0000 (23:26 +0000)]
- tagging RC_CURRENT

20 years agoUpdate build version.
adilger [Tue, 13 Apr 2004 19:08:29 +0000 (19:08 +0000)]
Update build version.

20 years agoAdd missing part of bug 2706 fix.
adilger [Tue, 13 Apr 2004 18:38:14 +0000 (18:38 +0000)]
Add missing part of bug 2706 fix.
b=2706
r=phil

20 years agoOnly create lprocfs entries after a device is set up (instead of at attach
adilger [Tue, 13 Apr 2004 18:34:28 +0000 (18:34 +0000)]
Only create lprocfs entries after a device is set up (instead of at attach
time) so that they don't try to use initialized data.
b=1519

20 years ago- tagging RC_CURRENT
buffalo-pull [Tue, 13 Apr 2004 04:04:01 +0000 (04:04 +0000)]
- tagging RC_CURRENT

20 years ago* Landed b_cray_portals_merge.
eeb [Tue, 13 Apr 2004 01:58:28 +0000 (01:58 +0000)]
*   Landed b_cray_portals_merge.

    Superbug 2871:
    Sub-bugs fixed 2869 2870 3034 3036 3037 3035 3038 3135

20 years agoland b_inodebits
nic [Mon, 12 Apr 2004 21:45:14 +0000 (21:45 +0000)]
land b_inodebits

20 years agoFix resend race during recovery (don't set rq_timeout = 0).
adilger [Mon, 12 Apr 2004 21:29:37 +0000 (21:29 +0000)]
Fix resend race during recovery (don't set rq_timeout = 0).
b=2950
r=robert

20 years agoDon't wait forever in sleep_on() if two threads call portals_debug_dumplog()
adilger [Mon, 12 Apr 2004 19:54:43 +0000 (19:54 +0000)]
Don't wait forever in sleep_on() if two threads call portals_debug_dumplog()
at the same time.
b=3107
r=zab

20 years agoDisable ll_readpage() log dumping (for debugging bug 2805), it caused bug 3122.
adilger [Thu, 8 Apr 2004 19:46:57 +0000 (19:46 +0000)]
Disable ll_readpage() log dumping (for debugging bug 2805), it caused bug 3122.
b=3122

20 years agoDon't dereference an ERR_PTR() in the mds_reint_link() error handling path.
adilger [Thu, 8 Apr 2004 19:12:01 +0000 (19:12 +0000)]
Don't dereference an ERR_PTR() in the mds_reint_link() error handling path.
b=3107
r=phil

20 years agoPoint people at the autolustre rpms, too
phil [Thu, 8 Apr 2004 18:44:12 +0000 (18:44 +0000)]
Point people at the autolustre rpms, too

20 years agoadd entry for 3118
nic [Thu, 8 Apr 2004 17:19:51 +0000 (17:19 +0000)]
add entry for 3118

20 years agob=3118
nic [Thu, 8 Apr 2004 16:55:55 +0000 (16:55 +0000)]
b=3118
r=adilger

fix badly initialized nameidata. Bug found and fix reported by 'johann lombardi <johann.lombardi@bull.net>'

20 years agoUpdate build tag.
adilger [Wed, 7 Apr 2004 21:53:25 +0000 (21:53 +0000)]
Update build tag.

20 years agoFix patch breakage.
adilger [Wed, 7 Apr 2004 21:53:00 +0000 (21:53 +0000)]
Fix patch breakage.

20 years agoFix patch breakage.
adilger [Wed, 7 Apr 2004 21:52:05 +0000 (21:52 +0000)]
Fix patch breakage.

20 years agoFix duplicate merge of this hunk.
adilger [Wed, 7 Apr 2004 21:46:19 +0000 (21:46 +0000)]
Fix duplicate merge of this hunk.

20 years agoland b_smallfix 20040407_1414:
jacob [Wed, 7 Apr 2004 19:12:05 +0000 (19:12 +0000)]
land b_smallfix 20040407_1414:

b=3057, 2947, 2925, 3050, 3041, 2849, 1662

20 years ago- put the exit(1)s back in after MPI_Abort() in the hopes that MPI
zab [Wed, 7 Apr 2004 17:31:23 +0000 (17:31 +0000)]
- put the exit(1)s back in after MPI_Abort() in the hopes that MPI
  implementations which don't exit in MPI_Abort() will honor the exit(1)s.

20 years ago* changed ptlrpc_unregister_reply(), ptlrpc_unregister_bulk() and
eeb [Tue, 6 Apr 2004 22:45:41 +0000 (22:45 +0000)]
*     changed ptlrpc_unregister_reply(), ptlrpc_unregister_bulk() and
      ptlrpc_abort_bulk() to always l_wait_event() after doing
      PtlMDUnlink() on the relevant MD.  Previous code assumed that when
      PtlMDUnlink() returned PTL_MD_INVALID, the relevant callback handler
      must have run already, which isn't true for liblustre.  The
      l_wait_event() will be a NOOP for kernel lustre (condition already
      true) but gives the liblustre event handler a chance to call all
      the callbacks.

20 years agoIn ll_file_read() use (*ppos + count - 1) for end of read instead of the value
adilger [Tue, 6 Apr 2004 22:29:15 +0000 (22:29 +0000)]
In ll_file_read() use (*ppos + count - 1) for end of read instead of the value
in policy->l_extent.end.  This makes the intent of the comparison more clear,
and avoids any problems if the policy data is ever changed (e.g. to return the
granted extent or similar).

Useful debugging for lock extent problems.
b=2925

20 years agoMinor correctness fixes for EA compat support:
adilger [Tue, 6 Apr 2004 22:24:54 +0000 (22:24 +0000)]
Minor correctness fixes for EA compat support:
- print the right size in the error message
- work correctly some day in the future when we have v2 EAs
b=3059

20 years agoUse the highest extent to determine the kms value, not the last one.
adilger [Tue, 6 Apr 2004 19:10:44 +0000 (19:10 +0000)]
Use the highest extent to determine the kms value, not the last one.
Try not to grant overlapping locks for the same client, to speed lock cancel.
b=2925
r=phil

20 years agob=3054
nic [Mon, 5 Apr 2004 19:43:35 +0000 (19:43 +0000)]
b=3054
r=zab

add case for 2.6 and /proc/sys/vm/dirty_writeback_centisecs where 2.4 uses kupdated

20 years ago* Fix for 2895
eeb [Mon, 5 Apr 2004 19:26:16 +0000 (19:26 +0000)]
*   Fix for 2895

20 years agoUpdate version tag.
adilger [Mon, 5 Apr 2004 18:36:21 +0000 (18:36 +0000)]
Update version tag.

20 years agoDebugging for mdc_close oops.
adilger [Mon, 5 Apr 2004 18:19:03 +0000 (18:19 +0000)]
Debugging for mdc_close oops.
b=2410
r=robert

20 years agoDon't update to the kms value for a lock that is just in the process of
adilger [Mon, 5 Apr 2004 18:14:48 +0000 (18:14 +0000)]
Don't update to the kms value for a lock that is just in the process of
being cancelled.
b=3057
r=phil

20 years agob=3106
phil [Sun, 4 Apr 2004 20:36:38 +0000 (20:36 +0000)]
b=3106
Don't deref NULL inode if we make it into filter_lvbo_init for (in
this case) the wrong object group

20 years agoNULL the filter arrays as they are freed to avoid potential problems on cleanup.
adilger [Sat, 3 Apr 2004 19:28:33 +0000 (19:28 +0000)]
NULL the filter arrays as they are freed to avoid potential problems on cleanup.
r=phil

20 years agoInitialize lock handle so that we don't fail ASSERT on error cleanup.
adilger [Sat, 3 Apr 2004 19:27:21 +0000 (19:27 +0000)]
Initialize lock handle so that we don't fail ASSERT on error cleanup.
b=3057
r=phil

20 years agoQuiet compiler warning.
adilger [Sat, 3 Apr 2004 04:40:53 +0000 (04:40 +0000)]
Quiet compiler warning.
b=1557

20 years agoDon't dereference uninitialized variable.
adilger [Sat, 3 Apr 2004 04:39:18 +0000 (04:39 +0000)]
Don't dereference uninitialized variable.
b=2543

20 years ago- tagging RC_CURRENT
buffalo-pull [Sat, 3 Apr 2004 04:23:14 +0000 (04:23 +0000)]
- tagging RC_CURRENT

20 years agocreate linux/configs if not present
nic [Fri, 2 Apr 2004 21:21:53 +0000 (21:21 +0000)]
create linux/configs if not present

20 years ago- tagging RC_CURRENT
buffalo-pull [Fri, 2 Apr 2004 02:11:34 +0000 (02:11 +0000)]
- tagging RC_CURRENT

20 years agoChange default stripe size based on ALC/ADEV testing.
adilger [Thu, 1 Apr 2004 19:50:19 +0000 (19:50 +0000)]
Change default stripe size based on ALC/ADEV testing.

20 years agoMinor changes to sync b1_2 with HEAD that shouldn't affect anything.
adilger [Thu, 1 Apr 2004 12:04:48 +0000 (12:04 +0000)]
Minor changes to sync b1_2 with HEAD that shouldn't affect anything.

20 years agoMinor changes to sync b1_2 with HEAD that shouldn't affect anything.
adilger [Thu, 1 Apr 2004 12:03:07 +0000 (12:03 +0000)]
Minor changes to sync b1_2 with HEAD that shouldn't affect anything.

20 years agoDisable tests for which fixes are not currently in b1_2.
adilger [Thu, 1 Apr 2004 11:41:45 +0000 (11:41 +0000)]
Disable tests for which fixes are not currently in b1_2.

20 years agoFix sanity test.
adilger [Thu, 1 Apr 2004 06:42:31 +0000 (06:42 +0000)]
Fix sanity test.
b=1414

20 years agob=3041
jacob [Thu, 1 Apr 2004 04:09:23 +0000 (04:09 +0000)]
b=3041

re-add things CVS mistakenly removed.

20 years agob=3104
jacob [Thu, 1 Apr 2004 04:06:36 +0000 (04:06 +0000)]
b=3104

Do not cvs add/rm on one branch, then cvs up -r to another branch, and
expect CVS to cope.

20 years agob=3041
jacob [Thu, 1 Apr 2004 03:34:59 +0000 (03:34 +0000)]
b=3041
b=3058

Clean up the autoconf and automake bits:

 * now liblustre and kernel modules can be built in the same tree
use --disable-modules, --disable-liblustre to disable those
bits
 * 2.6 modules are now built as .ko, and lconf can deal with them
 * fixes for builddir != srcdir

20 years ago- tagging RC_CURRENT
buffalo-pull [Thu, 1 Apr 2004 03:25:01 +0000 (03:25 +0000)]
- tagging RC_CURRENT

20 years agoDon't allow setstripe on a file/directory without write permission for same.
adilger [Thu, 1 Apr 2004 00:52:40 +0000 (00:52 +0000)]
Don't allow setstripe on a file/directory without write permission for same.
Regression tests for setstripe permission check.
b=1414

20 years agoFix mismerge from HEAD.
adilger [Thu, 1 Apr 2004 00:48:48 +0000 (00:48 +0000)]
Fix mismerge from HEAD.

20 years agob=3082
nic [Thu, 1 Apr 2004 00:31:26 +0000 (00:31 +0000)]
b=3082
r=andreas

sys_fchmod was missing setattr_raw magic, add chmod_common for both sys_fchmod and sys_chmod
adjust locking around chown_common() setattr_raw RPC, as i_sem need to be held over setattr

20 years agoMerge from HEAD to b1_2 select fixes for 1.2.2
adilger [Wed, 31 Mar 2004 22:36:42 +0000 (22:36 +0000)]
Merge from HEAD to b1_2 select fixes for 1.2.2
b=2922, b=2948, b=1414

20 years agob=2948
adilger [Wed, 31 Mar 2004 22:26:54 +0000 (22:26 +0000)]
b=2948

20 years agoMerge fixes from b_smallfix to b1_2 for 1.2.2
adilger [Wed, 31 Mar 2004 20:37:02 +0000 (20:37 +0000)]
Merge fixes from b_smallfix to b1_2 for 1.2.2
b=2947, b=2543, b=1557, b=2009

20 years agoKeep tests in sync with HEAD.
adilger [Wed, 31 Mar 2004 19:38:09 +0000 (19:38 +0000)]
Keep tests in sync with HEAD.

20 years agoBump kernel patch version to 34, to match HEAD.
adilger [Wed, 31 Mar 2004 19:30:21 +0000 (19:30 +0000)]
Bump kernel patch version to 34, to match HEAD.
b=2517, b=2399 are the changes for this kernel version

20 years agoLand b_smallfix onto HEAD (20040330_2312)
adilger [Wed, 31 Mar 2004 18:18:04 +0000 (18:18 +0000)]
Land b_smallfix onto HEAD (20040330_2312)
b=2947, b=2543, b=1557, b=2009

20 years agoFix test errors for 6 & 7 (wasn't waiting for async delete to complete).
adilger [Wed, 31 Mar 2004 04:24:00 +0000 (04:24 +0000)]
Fix test errors for 6 & 7 (wasn't waiting for async delete to complete).
b=2921

20 years agoRemove ancient files.
adilger [Wed, 31 Mar 2004 04:11:48 +0000 (04:11 +0000)]
Remove ancient files.

20 years ago- tagging RC_CURRENT
buffalo-pull [Wed, 31 Mar 2004 01:17:54 +0000 (01:17 +0000)]
- tagging RC_CURRENT

20 years agoDon't copy lvb into possibly NULL reply on ldlm_handle_enqueue() error
adilger [Wed, 31 Mar 2004 01:02:01 +0000 (01:02 +0000)]
Don't copy lvb into possibly NULL reply on ldlm_handle_enqueue() error
client won't use it anyways.  Regression test for same.
Fix secondary bug hit during regression tests in mds_mfd_close() cleanup.
b=2983

20 years agob=1414
adilger [Tue, 30 Mar 2004 21:28:34 +0000 (21:28 +0000)]
b=1414

20 years ago- tagging RC_CURRENT
buffalo-pull [Tue, 30 Mar 2004 03:26:44 +0000 (03:26 +0000)]
- tagging RC_CURRENT

20 years agochange function declarations to correctly reflect the MPI_Abort exiting, remove
nic [Mon, 29 Mar 2004 16:57:21 +0000 (16:57 +0000)]
change function declarations to correctly reflect the MPI_Abort exiting, remove
exits. this is what should have been done in the first place...yeah.

20 years agoliblustre:
ericm [Mon, 29 Mar 2004 06:02:28 +0000 (06:02 +0000)]
liblustre:
 - support VPATH build (still some unclean stuff in liblustre build)

20 years ago- tagging RC_CURRENT
buffalo-pull [Sat, 27 Mar 2004 03:06:18 +0000 (03:06 +0000)]
- tagging RC_CURRENT

20 years agofix compiler warnings
nic [Fri, 26 Mar 2004 16:35:50 +0000 (16:35 +0000)]
fix compiler warnings

20 years ago to get Lustre built with 2.6.3+mm4, LTIME_S(dentry->d_inode->i_size) is not correct
hitao [Fri, 26 Mar 2004 08:33:13 +0000 (08:33 +0000)]
 to get Lustre built with 2.6.3+mm4, LTIME_S(dentry->d_inode->i_size) is not correct

20 years ago- tagging RC_CURRENT
buffalo-pull [Fri, 26 Mar 2004 00:39:39 +0000 (00:39 +0000)]
- tagging RC_CURRENT

20 years agoDon't free resource lvb if there is a lookup error (causes later assertion).
adilger [Thu, 25 Mar 2004 20:16:58 +0000 (20:16 +0000)]
Don't free resource lvb if there is a lookup error (causes later assertion).
Don't dereference dentry after dputting it.
Remove use of obdo in filter_lvb, since we don't actually need it.
b=2922

20 years agoloff_t is a long long in the kernel, liblustre should use the same to
adilger [Thu, 25 Mar 2004 20:08:53 +0000 (20:08 +0000)]
loff_t is a long long in the kernel, liblustre should use the same to
avoid compiler warnings on 64-bit (this #define is for windows port only).

20 years agoMove test 66[ab] from sanity.sh to recovery-small.sh because we need to do
adilger [Thu, 25 Mar 2004 18:42:47 +0000 (18:42 +0000)]
Move test 66[ab] from sanity.sh to recovery-small.sh because we need to do
ops on the OST, and that isn't possible with sanity.sh if it is remote.
Fix test replay-ost-single test 7 for bug 2123 also.
b=2983, b=2123

20 years ago- tagging RC_CURRENT
buffalo-pull [Thu, 25 Mar 2004 04:33:11 +0000 (04:33 +0000)]
- tagging RC_CURRENT

20 years agob=2985
zab [Thu, 25 Mar 2004 01:33:16 +0000 (01:33 +0000)]
b=2985
r=zab

use MPI_Abort() in the exiting error paths of write_* so that other processes
are pulled out of MPI_Barrier().  I added a -f <filename> argument to
write_disjoint while I was veryfying this.

20 years agoFix for bug 2921 (test failure only). We had coincidentally created a file
adilger [Wed, 24 Mar 2004 09:21:03 +0000 (09:21 +0000)]
Fix for bug 2921 (test failure only).  We had coincidentally created a file
in test 4 with the same size, and if we didn't wait for its deletion then
it could so happen that the "kbytesfree before" would equal "kbytesfree after"
since that file was freed while we wrote the new one.
b=2921

20 years agoDon't copy lvb into reply message on error, since that message might not have
adilger [Wed, 24 Mar 2004 05:40:47 +0000 (05:40 +0000)]
Don't copy lvb into reply message on error, since that message might not have
allocated enough space for the lvb and the client will ignore it anyways.
Also fixes a NULL request deref in mds_mfd_close() when cleaning up opens
on the MDS.  The initial sanity test (66b) triggered a new bug 2986, but
another test with multiop tests this properly.
b=2983

20 years ago- tagging RC_CURRENT
buffalo-pull [Wed, 24 Mar 2004 03:08:43 +0000 (03:08 +0000)]
- tagging RC_CURRENT

20 years agoDon't copy lvb into reply message on error, since that message might not have
adilger [Wed, 24 Mar 2004 02:50:22 +0000 (02:50 +0000)]
Don't copy lvb into reply message on error, since that message might not have
allocated enough space for the lvb and the client will ignore it anyways.
Also fixes a NULL request deref in mds_mfd_close() when cleaning up opens
on the MDS.  The initial sanity test (66b) triggered a new bug 2986, but
another test with multiop tests this properly.
b=2983

20 years agoHandle SF IOR better by working harder to avoid granting conflicting locks.
adilger [Wed, 24 Mar 2004 00:36:19 +0000 (00:36 +0000)]
Handle SF IOR better by working harder to avoid granting conflicting locks.
From b1_2.
b=2919

20 years agoHandle deleted working directory (from b1_2) for 2.4.18-chaos kernel.
adilger [Wed, 24 Mar 2004 00:31:48 +0000 (00:31 +0000)]
Handle deleted working directory (from b1_2) for 2.4.18-chaos kernel.
b=2399

20 years agoDon't crash in expired_lock_main() with racing client eviction/lock completion.
adilger [Wed, 24 Mar 2004 00:29:15 +0000 (00:29 +0000)]
Don't crash in expired_lock_main() with racing client eviction/lock completion.
b=2867

20 years agoFiled bug 2991 for this issue.
adilger [Tue, 23 Mar 2004 21:06:37 +0000 (21:06 +0000)]
Filed bug 2991 for this issue.
b=2991

20 years agob=2973
rread [Tue, 23 Mar 2004 21:02:07 +0000 (21:02 +0000)]
b=2973

- add a new test (18a) to use lctl --deactivate to confirm page cache is
  cleared after invalidation.
- rename 18 to 18b, and increase wait to ensure the cache is checked after
  recovery

20 years agoDisable this check, because it fails on LLNL kernels until the xattr patches
adilger [Tue, 23 Mar 2004 20:16:45 +0000 (20:16 +0000)]
Disable this check, because it fails on LLNL kernels until the xattr patches
are updated.

20 years agoEnable EXT3_FS_XATTR in defconfig for new kernel configs to make it easier
adilger [Tue, 23 Mar 2004 19:45:35 +0000 (19:45 +0000)]
Enable EXT3_FS_XATTR in defconfig for new kernel configs to make it easier
for users to get it right.

20 years agoIgnore generated files.
adilger [Tue, 23 Mar 2004 19:35:17 +0000 (19:35 +0000)]
Ignore generated files.

20 years agoDon't enable D_MALLOC by default, slipped in during debugging.
adilger [Tue, 23 Mar 2004 19:32:58 +0000 (19:32 +0000)]
Don't enable D_MALLOC by default, slipped in during debugging.

20 years agoDon't have D_MALLOC on by default. This slipped in during debugging.
adilger [Tue, 23 Mar 2004 17:20:13 +0000 (17:20 +0000)]
Don't have D_MALLOC on by default.  This slipped in during debugging.

20 years agoallow lconf to work with python1.5.
nic [Tue, 23 Mar 2004 16:02:02 +0000 (16:02 +0000)]
allow lconf to work with python1.5.
do not support http urls until python2, as that is the first occurance of
    urllib2, where the error handling for http is finally sane

20 years agoDon't print an error message for a normal race.
adilger [Tue, 23 Mar 2004 07:16:25 +0000 (07:16 +0000)]
Don't print an error message for a normal race.
b=2526

20 years ago *b: 2948
tianying [Tue, 23 Mar 2004 03:46:11 +0000 (03:46 +0000)]
   *b: 2948
   *r: Robert
     the pointer of oti->oti_logcookies should be increased when an osc is inactive.