Whamcloud - gitweb
fs/lustre-release.git
13 years agob=23985 Fix upstream jbd2 code issue. 1.8.5 v1_8_5_RC4
yangsheng [Mon, 15 Nov 2010 14:43:46 +0000 (22:43 +0800)]
b=23985 Fix upstream jbd2 code issue.

The jbd2_stats_proc_init() was placed on wrong location in
  jbd2_journal_init_dev().  This may cause /proc/fs/jdb2/<dev>/*
  cannot be created when using external journal device.

i=adilger
i=johann

13 years agoChangelog update
Andrew Perepechko [Mon, 15 Nov 2010 12:43:06 +0000 (15:43 +0300)]
Changelog update

13 years agob=24051 sanity test_76 fix
Andrew Perepechko [Mon, 15 Nov 2010 12:29:06 +0000 (15:29 +0300)]
b=24051 sanity test_76 fix

Due to per-cpu slab caches, "active" stat from /proc/slabinfo
is not always accurate.

i=ZhiYong Tian

13 years agob=23766 atime isn't properly updated on the MDS
Andrew Perepechko [Mon, 15 Nov 2010 12:21:11 +0000 (15:21 +0300)]
b=23766 atime isn't properly updated on the MDS

Based on the patch from Robin Humble (ANU)

i=Andreas Dilger
i=Vitaly Fertman

13 years agob=21587 don't LBUG if transno has changed during replay v1_8_5_RC3
Andrew Perepechko [Wed, 27 Oct 2010 04:54:18 +0000 (12:54 +0800)]
b=21587 don't LBUG if transno has changed during replay

i=johann
i=tappro

13 years agob=23988 Changes sd-iostats patch for 2.6.32 kernel.
yangsheng [Fri, 22 Oct 2010 09:43:38 +0000 (17:43 +0800)]
b=23988 Changes sd-iostats patch for 2.6.32 kernel.

Also statistic with REQ_TYPE_BLOCK_PC command.

13 years agoRevert "b=23988 Changes sd-iostats patch for 2.6.32 kernel."
Johann Lombardi [Thu, 21 Oct 2010 12:27:41 +0000 (14:27 +0200)]
Revert "b=23988 Changes sd-iostats patch for 2.6.32 kernel."

This reverts commit 50a915fdbb340f04a850778870d027c4c3303aab.

13 years agob=23965 Add fiemap patch for ldiskfs base on ext4(sles11 sp1).
yangsheng [Thu, 21 Oct 2010 12:06:52 +0000 (20:06 +0800)]
b=23965 Add fiemap patch for ldiskfs base on ext4(sles11 sp1).

i=girish
i=johann

13 years agob=23988 Changes sd-iostats patch for 2.6.32 kernel.
yangsheng [Thu, 21 Oct 2010 12:00:24 +0000 (20:00 +0800)]
b=23988 Changes sd-iostats patch for 2.6.32 kernel.

13 years agob=24003 Shows stripe_count: 65535 but should be '-1'.
yangsheng [Thu, 21 Oct 2010 10:42:00 +0000 (18:42 +0800)]
b=24003 Shows stripe_count: 65535 but should be '-1'.

i=johann

13 years agob=21587 don't clean up echo_client is refcount is not 0
Johann Lombardi [Wed, 20 Oct 2010 17:46:23 +0000 (19:46 +0200)]
b=21587 don't clean up echo_client is refcount is not 0

i=adilger

13 years agob=23966 Revert "b=21137 sles11 with 1.8 is slower than 1.6 sles10 for O_DIRECT single...
Johann Lombardi [Wed, 20 Oct 2010 09:37:53 +0000 (11:37 +0200)]
b=23966 Revert "b=21137 sles11 with 1.8 is slower than 1.6 sles10 for O_DIRECT single file IOR writes"

This reverts commit 1242fac546a8c252dee5715bb218133795211028.
This introduces issues with SATA drivers. See bug 23966.

13 years agob=23721 download OFED from OFA v1_8_5_RC2
Brian J. Murrell [Thu, 14 Oct 2010 18:34:15 +0000 (20:34 +0200)]
b=23721 download OFED from OFA

Download the OFED distribution directly from OFA to ameliorate
re-distribution complications.

i=yangsheng
i=wangyb

13 years agofix nit in lbuild-sles10
Johann Lombardi [Thu, 14 Oct 2010 16:42:27 +0000 (18:42 +0200)]
fix nit in lbuild-sles10

13 years agoUpdated for Lustre 1.8.5. v1_8_5_RC1
Terry Rutledge [Thu, 14 Oct 2010 00:11:32 +0000 (17:11 -0700)]
Updated for Lustre 1.8.5.

13 years agofix changelog
Johann Lombardi [Wed, 13 Oct 2010 22:41:11 +0000 (00:41 +0200)]
fix changelog

13 years agoset version to 1.8.5.0 for RC1
Johann Lombardi [Wed, 13 Oct 2010 22:35:22 +0000 (00:35 +0200)]
set version to 1.8.5.0 for RC1

13 years agob=22514 Update to RHEL5.5 & OEL5.5 latest kernel.
yangsheng [Wed, 13 Oct 2010 14:22:24 +0000 (22:22 +0800)]
b=22514 Update to RHEL5.5 & OEL5.5 latest kernel.

13 years agob=20744 add changelog entry and set REBUILD_OFED_DEVEL_RPM for patched kernel only
Johann Lombardi [Wed, 13 Oct 2010 21:53:05 +0000 (23:53 +0200)]
b=20744 add changelog entry and set REBUILD_OFED_DEVEL_RPM for patched kernel only

13 years agob=23911 enable REBUILD_OFED_DEVEL_RPM by default
Johann Lombardi [Wed, 13 Oct 2010 09:34:26 +0000 (11:34 +0200)]
b=23911 enable REBUILD_OFED_DEVEL_RPM by default

13 years agob=20744 set arch correctly for ofed-devel
Brian J. Murrell [Tue, 12 Oct 2010 19:56:32 +0000 (15:56 -0400)]
b=20744 set arch correctly for ofed-devel

Oh boy, more fallout from our not building i586 kernels for sles.
If we are building lustre servers and NOT rebuilding the Suse ofed
RPM, then we need to look for the i586 ofed-devel package.

i=yangsheng

13 years agob=20744 add changelog entry
Johann Lombardi [Tue, 12 Oct 2010 15:55:46 +0000 (17:55 +0200)]
b=20744 add changelog entry

13 years agob=20744 Update SLES10 SP3 kernel 2.6.16.60-0.69.1.
yangsheng [Tue, 12 Oct 2010 15:50:36 +0000 (23:50 +0800)]
b=20744 Update SLES10 SP3 kernel 2.6.16.60-0.69.1.

i=johann
i=brian

13 years agob=21137 sles11 with 1.8 is slower than 1.6 sles10 for O_DIRECT single file IOR writes
Dmitry Zogin [Tue, 12 Oct 2010 14:38:03 +0000 (10:38 -0400)]
b=21137 sles11 with 1.8 is slower than 1.6 sles10 for O_DIRECT single file IOR writes

Fix __bio_add_page() in order to prevent IO split on some drivers.

 i=andreas.dilger
 i=sheng.yang

13 years agob=20533 set the default max_sectors to the raid5/6 stripe size
Andrew Perepechko [Tue, 12 Oct 2010 12:33:43 +0000 (16:33 +0400)]
b=20533 set the default max_sectors to the raid5/6 stripe size

a=Kevin Van Maren
i=Jinshan Xiong
i=Atul Vidwansa

13 years agob=23701 reduce stack pressure by uninlining some ptlrpc and mds routines
Andrew Perepechko [Tue, 12 Oct 2010 08:45:35 +0000 (12:45 +0400)]
b=23701 reduce stack pressure by uninlining some ptlrpc and mds routines

i=Oleg Drokin
i=Alexander Zarochentsev
i=Johann Lombardi

13 years agob=23542 recovery-small: test_27 is not time bounded
Elena Gryaznova [Mon, 11 Oct 2010 14:17:59 +0000 (18:17 +0400)]
b=23542 recovery-small: test_27 is not time bounded

i=Brian.Murrell

13 years agob=23537 do_rpc_nodes does not propagate NAME to remote nodes causing rpc.sh to fail...
Elena Gryaznova [Mon, 11 Oct 2010 16:35:37 +0000 (20:35 +0400)]
b=23537 do_rpc_nodes does not propagate NAME to remote nodes causing rpc.sh to fail abruptly

o=Chris.Horn <hornc@cray.com>
i=grev

13 years agobump version to 1.8.4.55
Johann Lombardi [Mon, 11 Oct 2010 13:11:23 +0000 (15:11 +0200)]
bump version to 1.8.4.55

13 years agob=23781 fix another potential obdo leak
Johann Lombardi [Mon, 11 Oct 2010 12:12:00 +0000 (14:12 +0200)]
b=23781 fix another potential obdo leak

i=hongchao

13 years agob=20744 remedial fix for inkernel ofed handling
Brian Murell [Mon, 11 Oct 2010 12:02:16 +0000 (14:02 +0200)]
b=20744 remedial fix for inkernel ofed handling

i=yangsheng

It seems that the patch to add inkernel handling for SLES10's OFED introduced a
bug for other inkernel uses.  This fixes that bug.

13 years agobump version to 1.8.4.54 v1_8_4_54
Johann Lombardi [Fri, 8 Oct 2010 16:16:37 +0000 (18:16 +0200)]
bump version to 1.8.4.54

13 years agob=23588 wait for new OST addition propagation to clients
Vladimir Saveliev [Fri, 8 Oct 2010 13:39:06 +0000 (17:39 +0400)]
b=23588 wait for new OST addition propagation to clients

i=nathan.rutman

13 years agob=20744 hacks for Novell bug 639581
Brian J. Murrell [Fri, 8 Oct 2010 14:17:41 +0000 (10:17 -0400)]
b=20744 hacks for Novell bug 639581

13 years agob=20744 hacks for Novell bug 578796
Brian J. Murrell [Fri, 8 Oct 2010 14:17:40 +0000 (10:17 -0400)]
b=20744 hacks for Novell bug 578796

These hacks are needed to deal with Novell bug 578796.

13 years agob=20744 use Novell supplied ofed-devel
Brian J. Murrell [Fri, 8 Oct 2010 14:17:39 +0000 (10:17 -0400)]
b=20744 use Novell supplied ofed-devel

Several changes to support using ofed-devel supplied in SLES10 SP3
(currently as PTFs).
Fix a bug with passing arguments to rpm2cpio that should not be passed.
Add a hook for distro specific module building after the kernel[-devel]
build and unpack.
Fix resolve_arch to be able to resolve either the "canonical" arch for
i?86, or the arch that the patchless kernel is expected to be.
Bail out of download_ofed earlier if using "inkernel".
Fix a bug where an rpm2cpio command was not being error checked.
Fix a bug where we were not processing and making available files
in all of the RPMS/i?86 subdirs.

i=mjmac
i=wangyb

13 years agob=23685 don't define unused variables
Brian J. Murrell [Wed, 6 Oct 2010 21:13:37 +0000 (17:13 -0400)]
b=23685 don't define unused variables

When CONFIG_SYSCTL is not defined in the kernel config, a couple of
variables are left unused, yet still defined.

i=Dmitry Zogin
i=Andrew Perepechko

13 years agob=23804 LC_WALK_SPACE_HAS_DATA_SEM on servers only
Brian J. Murrell [Wed, 6 Oct 2010 21:13:36 +0000 (17:13 -0400)]
b=23804 LC_WALK_SPACE_HAS_DATA_SEM on servers only

The LC_WALK_SPACE_HAS_DATA_SEM check is only needed on servers
so guard it for server configures only.
i=Sheng Yang
i=Andreas Dilger

13 years agob=23721 disable rdac and mptlinux by default
Andrew Perepechko [Fri, 8 Oct 2010 12:38:13 +0000 (16:38 +0400)]
b=23721 disable rdac and mptlinux by default

Disable the (downloading and) building of the MPTLINUX and RDAC
packages by default and provide a commandline switch to enable them.

i=yangsheng
i=wangyb

13 years agob=21610 Update to sles11 sp1 latest kernel 2.6.32.19-0.2.
yangsheng [Fri, 8 Oct 2010 09:43:15 +0000 (17:43 +0800)]
b=21610 Update to sles11 sp1 latest kernel 2.6.32.19-0.2.

i=adilger
i=girish

13 years agoAdd missing changelog entries
Johann Lombardi [Fri, 8 Oct 2010 09:19:48 +0000 (11:19 +0200)]
Add missing changelog entries

13 years agoRevert "b=21760 Application hung in direct I/O"
Johann Lombardi [Thu, 7 Oct 2010 22:25:32 +0000 (00:25 +0200)]
Revert "b=21760 Application hung in direct I/O"

This reverts commit 91d48dd54d07c3901da34227037a32412243e08a.

With this patch, ptlrpcd sleeps until bulk buffers are unlinked by
the network stack. Meanwhile, no requests/replies are processed
by ptlrpcd which can cause various kind of problems (e.g. eviction).

13 years agob=23645 remove another i_mutex warning with 2.6.32
Johann Lombardi [Thu, 7 Oct 2010 20:05:25 +0000 (22:05 +0200)]
b=23645 remove another i_mutex warning with 2.6.32

hold inode mutex before calling lookup_on_len() when creating
objects on MDS.

13 years agob=23596 account direct i/o inflight separately from non-direct i/o
Andrew Perepechko [Thu, 7 Oct 2010 14:27:39 +0000 (18:27 +0400)]
b=23596 account direct i/o inflight separately from non-direct i/o

Account direct i/o inflight rpcs separately from non-direct i/o so
that direct i/o, which is limited by max_rpcs_in_flight, should not
block non-direct i/o, which is not limited by max_rpcs_in_flight.

i=Oleg Drokin
i=Alexander Zarochentsev
i=Johann Lombardi (author of the original patch)

13 years agob=23878 add debug info
Johann Lombardi [Thu, 7 Oct 2010 08:06:12 +0000 (10:06 +0200)]
b=23878 add debug info

13 years agoAdd support for OFED 1.5.2
Johann Lombardi [Thu, 7 Oct 2010 07:55:36 +0000 (09:55 +0200)]
Add support for OFED 1.5.2

13 years agob=22168 write_disjoint: retry if write receives EINTR.
Eric Mei [Wed, 6 Oct 2010 15:16:23 +0000 (09:16 -0600)]
b=22168 write_disjoint: retry if write receives EINTR.

o=robert.read
r=eric.mei

13 years agob=23216 a fix for a possible memory leak in echo_client_kbrw
Andrew Perepechko [Wed, 6 Oct 2010 15:17:45 +0000 (19:17 +0400)]
b=23216 a fix for a possible memory leak in echo_client_kbrw

b=Oleg Drokin
b=Hong-Chao Zhang

13 years agoMerge branch 'b1_8' of git.lustre.org:prime/lustre into b1_8
Andrew Perepechko [Wed, 6 Oct 2010 15:14:25 +0000 (19:14 +0400)]
Merge branch 'b1_8' of git.lustre.org:prime/lustre into b1_8

13 years agob=23827 fix per-NID reporting on outstanding writes
Andrew Perepechko [Wed, 6 Oct 2010 15:12:52 +0000 (19:12 +0400)]
b=23827  fix per-NID reporting on outstanding writes

a=Kevin Van Maren
i=Johann Lombardi
i=Andreas Dilger

13 years agob=23780 hold write reference on i_data_sem for walk_space
Johann Lombardi [Wed, 6 Oct 2010 13:51:13 +0000 (15:51 +0200)]
b=23780 hold write reference on i_data_sem for walk_space

i=girish

Restore the old behavior and live with the fiemap locking issue for now.

13 years agob=23701 a build fix
Andrew Perepechko [Tue, 5 Oct 2010 17:55:38 +0000 (21:55 +0400)]
b=23701 a build fix

13 years agob=23710 reduce stack pressure by uninlining mds_log_lost_precreated
Andrew Perepechko [Tue, 5 Oct 2010 16:56:14 +0000 (20:56 +0400)]
b=23710 reduce stack pressure by uninlining mds_log_lost_precreated

i=Johann Lombardi

13 years agob=22546 sanity-quota test_18c: @@@@@@ FAIL: expect 104857600, got 102760448. Verifyin...
Elena Gryaznova [Tue, 5 Oct 2010 16:46:33 +0000 (20:46 +0400)]
b=22546 sanity-quota test_18c: @@@@@@ FAIL: expect 104857600, got 102760448. Verifying file failed!

o=James.Simmons <simmonsja@ornl.gov>
i=Brian.Murrell
i=grev
i=landen

13 years agob=23701 reduce stack pressure in ptlrpc_server_handle_request
Andrew Perepechko [Tue, 5 Oct 2010 16:43:18 +0000 (20:43 +0400)]
b=23701 reduce stack pressure in ptlrpc_server_handle_request

i=Oleg Drokin
i=Alexander Zarochentsev
i=Johann Lombardi

13 years agoUpdated for 1.8.5 build 03. v1_8_4_53
Terry Rutledge [Fri, 1 Oct 2010 14:53:48 +0000 (07:53 -0700)]
Updated for 1.8.5 build 03.

13 years agob=22107 a fix for the warning message in filter_destroy_internal
Andrew Perepechko [Thu, 30 Sep 2010 12:45:54 +0000 (16:45 +0400)]
b=22107 a fix for the warning message in filter_destroy_internal

filter_destroy_internal is entered with inode references from
lookup and from the resource used by the extent lock, so the
check is changed to i_count != 2.

i=Vitaly Fertman
i=Johann Lombardi

13 years agob=22766 cascading_rw: take lmm_stripe_count returned by ioctl(LL_IOC_LOV_SETSTRIPE)
Vladimir Saveliev [Wed, 29 Sep 2010 21:21:37 +0000 (01:21 +0400)]
b=22766 cascading_rw: take lmm_stripe_count returned by ioctl(LL_IOC_LOV_SETSTRIPE)

+cleanup in ll_dir_ioctl(): do not save return code of copy_to_user and copy_from_user

i=adilger

13 years agob=22770 Removed the LASSERT since (conn == NULL) is a legitimate case.
yangsheng [Mon, 20 Sep 2010 14:32:01 +0000 (22:32 +0800)]
b=22770 Removed the LASSERT since (conn == NULL) is a legitimate case.

i=adigler
i=panda

13 years agobump version to 1.8.4.52
Johann Lombardi [Tue, 28 Sep 2010 14:01:37 +0000 (16:01 +0200)]
bump version to 1.8.4.52

13 years agob=23781 fix obdo leak issue
hongchao.zhang [Sun, 12 Sep 2010 19:52:59 +0000 (03:52 +0800)]
b=23781 fix obdo leak issue

in ll_setattr_raw, if ll_extent_lock call fails, the newly allocated
obdo has not been freed before return.

i=johann

13 years agob=23706 sles11sp1: jbd2 stat patch is missing
Dmitry Zogin [Fri, 24 Sep 2010 19:05:47 +0000 (15:05 -0400)]
b=23706 sles11sp1: jbd2 stat patch is missing

 Temp patch to disable obdfilter-survey.sh 1b/2b tests,
 if jbd stats are not available.

 i=grev

13 years agoexplicitly set KVERS; allow all automake packages
Brian J. Murrell [Thu, 23 Sep 2010 15:23:34 +0000 (11:23 -0400)]
explicitly set KVERS; allow all automake packages

Explicitly set KVERS to what it'd default value would be if it's
not set by the caller.
Allow any of the compatible automake* packages in debian/control.

13 years agob=22117 Patch to limit mmp interval.
Girish Shilamkar [Thu, 2 Sep 2010 18:03:34 +0000 (23:33 +0530)]
b=22117 Patch to limit mmp interval.

i=adilger
i=kalpak

13 years agomove .debs into /debs
Brian J. Murrell [Sun, 19 Sep 2010 04:27:13 +0000 (00:27 -0400)]
move .debs into /debs

Move the built .debs from .. into a /debs subdir of the build tree.
We in fact could almost make debian/rules do this using dh_builddeb's
--destdir argument, however dpkg-genchanges wants to find them in ..
and there is no way to pass dpkg-genchange's -u argument to it from
dpkg-buildpackage.  So thus, our hack.

13 years agob=22934 conf_sanity writeconf method is never tested
Elena Gryaznova [Thu, 16 Sep 2010 18:40:22 +0000 (22:40 +0400)]
b=22934 conf_sanity writeconf method is never tested

i=Andrew.Perepechko

use the correct mds dev

13 years agob=17682 Improve debug capability for production systems
Dmitry Zogin [Thu, 16 Sep 2010 12:16:41 +0000 (08:16 -0400)]
b=17682 Improve debug capability for production systems

 Improve performance of D_RPCTRACE debugging.

 i=andreas.dilger
 i=andrew.perepechko

13 years agob=22378 Correct MDS client stats
Dmitry Zogin [Thu, 16 Sep 2010 12:13:08 +0000 (08:13 -0400)]
b=22378 Correct MDS client stats

 Interoperability fix for test_133[abc]  sanity.sh tests

 o=grev
 i=dmitry.zoguine

13 years agoIgnore generated ll_decode_filter_fid file
Andreas Dilger [Wed, 1 Sep 2010 09:07:53 +0000 (03:07 -0600)]
Ignore generated ll_decode_filter_fid file

13 years agob=23271 handle e2fsck and lfsck return values
Vladimir Saveliev [Tue, 14 Sep 2010 11:45:05 +0000 (15:45 +0400)]
b=23271 handle e2fsck and lfsck return values

to avoid test-framework.sh' trap

i=grev
i=jian.yu

13 years agob=20101 lfs ost enhancements
Andrew Perepechko [Thu, 16 Sep 2010 11:12:40 +0000 (15:12 +0400)]
b=20101 lfs ost enhancements

a=James Simmons <simmonsja@ornl.gov>
i=Andreas Dilger
i=Brian Murrell

13 years agob=23200 1.8<->2.0 interop: sanityN test_12: @@@@@@ FAIL: test_12 failed with 1
Rahul Deshmukh [Thu, 16 Sep 2010 08:08:09 +0000 (13:38 +0530)]
b=23200 1.8<->2.0 interop: sanityN test_12: @@@@@@ FAIL: test_12 failed with 1

b=23200 1.8<->2.0 interop: sanityN test_12: @@@@@@ FAIL: test_12 failed with 1

This patch fixed the 1.8<->2.0 interop: sanityN test_12

i=rread

13 years agob=23216 a fix for a possible memory leak in filter_server_init_data
Andrew Perepechko [Wed, 15 Sep 2010 21:11:56 +0000 (01:11 +0400)]
b=23216 a fix for a possible memory leak in filter_server_init_data

i=Alexander Zarochentsev
i=Dmitry Zogin

13 years agob=21137 sles11 with 1.8 is slower than 1.6 sles10 for O_DIRECT single file IOR writes
Dmitry Zogin [Wed, 15 Sep 2010 19:41:32 +0000 (15:41 -0400)]
b=21137 sles11 with 1.8 is slower than 1.6 sles10 for O_DIRECT single file IOR writes

 Extra debug logging in filter_do_bio()

 i=andreas.dilger
 i=andrew.perepechko

13 years agob=22820 too many default ACLs break directory access on new directories
Andrew Perepechko [Wed, 15 Sep 2010 17:28:02 +0000 (21:28 +0400)]
b=22820 too many default ACLs break directory access on new directories

The attached patch was a result of the discussion with Andreas Dilger on lustre
discuss starting here: http://lists.lustre.org/pipermail/lustre-discuss/2010-May/013177.html

a=Frederik Ferner <frederik.ferner@diamond.ac.uk>
i=Johann Lombardi
i=Andreas Dilger

13 years agob=23179 MDS node unresponsive
Isaac Huang [Mon, 13 Sep 2010 16:43:48 +0000 (10:43 -0600)]
b=23179 MDS node unresponsive

Better hash distribution, doubled hash size for the lnet cookie hash.

i=zhen.liang
i=andreas.dilger

13 years agob=23683 Bring upstream patch for ldiskfs.
yangsheng [Sat, 11 Sep 2010 18:54:23 +0000 (02:54 +0800)]
b=23683 Bring upstream patch for ldiskfs.

13 years agob=23701 remove a variable from stack, allocate from slab
Andrew Perepechko [Sat, 11 Sep 2010 17:45:35 +0000 (21:45 +0400)]
b=23701 remove a variable from stack, allocate from slab

Through the assembly of mds_open from the crash dump, it seems,
although the iattr var declaration is put into a compound statement,
the stack for the variable is reserved from the beginning to the
end of mds_open processing. According to crash, the size of iattr is 80 bytes.

i=Oleg Drokin
i=Alexander Zarochentsev

13 years agob=23701 remove a variable from stack, allocate from slab
Andrew Perepechko [Sat, 11 Sep 2010 17:43:59 +0000 (21:43 +0400)]
b=23701 remove a variable from stack, allocate from slab

i=Oleg Drokin
i=Alexander Zarochentsev

13 years agobz=23673 add "updates" dir to lustre-modules location.
Vitaly Fertman [Fri, 10 Sep 2010 21:48:40 +0000 (01:48 +0400)]
bz=23673 add "updates" dir to lustre-modules location.

o=grev
i=vitaly

13 years agob=22378 Correct MDS client stats
Dmitry Zogin [Fri, 10 Sep 2010 18:45:35 +0000 (14:45 -0400)]
b=22378 Correct MDS client stats

 Fixing sanity.sh test_133c

 i=grev
 i=andrew.perepechko

13 years agob=23669 change pios parameters according to space available
Elena Gryaznova [Thu, 9 Sep 2010 19:54:49 +0000 (23:54 +0400)]
b=23669 change pios parameters according to space available

i=Andrew.Perepechko

13 years agob=23174 Lustre inode size is not coherent across nodes.
Dmitry Zogin [Tue, 7 Sep 2010 15:36:01 +0000 (11:36 -0400)]
b=23174 Lustre inode size is not coherent across nodes.

 Update lvbo from disk when AST fails with EINVAL. Lvbo is updated on EINVAL
 error in ldlm_handle_ast_error(). The updates in filter_intent_policy() and
 ldlm_cb_interpret() have been removed as redundant.

 i=vitaly.fertman
 i=andrew.perepechko

13 years agopackage liblustre tests for debian
Brian J. Murrell [Wed, 1 Sep 2010 00:03:12 +0000 (20:03 -0400)]
package liblustre tests for debian

The liblustre tests were not being packaged into lustre-tests.

13 years agomake debs refinement
Brian J. Murrell [Wed, 1 Sep 2010 00:03:11 +0000 (20:03 -0400)]
make debs refinement

When looking for the most recent tag, look for only tags matching v*.
That should hopefully weed out any non-versioning tags.

Don't convert debian/changelog commits to patches since the changelog
has to be in place right when dpkg-buildpackage starts.
This does mean that one should NEVER commit to another file in the same
commit as a debian/changelog commit.

Make sure debian/patche{s,d} are clean before starting.

13 years agoRevert "b=22117 Patch to limit mmp interval."
Andrew Perepechko [Mon, 6 Sep 2010 15:34:04 +0000 (19:34 +0400)]
Revert "b=22117 Patch to limit mmp interval."

This reverts commit 0791b46099339efcf608ff99822be01766f90656.

13 years agob=22589 llite_lib.c:234: undefined-reference to `__stack_chk_fail_local'
Brian J. Murrell [Tue, 31 Aug 2010 23:49:00 +0000 (19:49 -0400)]
b=22589 llite_lib.c:234: undefined-reference to `__stack_chk_fail_local'

If liblustre.so is linked with -nostdlib it produces errors such as:
./llite_lib.o: In function `liblustre_process_log':
lustre/liblustre/llite_lib.c:234: undefined reference to `__stack_chk_fail_local'
on platforms where the stack-smashing protector is used (i.e. Ubuntu).

i=johann

13 years agob=22117 Patch to limit mmp interval.
Girish Shilamkar [Thu, 2 Sep 2010 18:03:34 +0000 (23:33 +0530)]
b=22117 Patch to limit mmp interval.

i=adilger
i=kalpak

13 years agob=23363 obdfilter-survey: no echo client and ost_name not set
Elena Gryaznova [Fri, 3 Sep 2010 14:04:18 +0000 (18:04 +0400)]
b=23363 obdfilter-survey: no echo client and ost_name not set

i=Andrew.Perepechko

libecho fix: use NETTYPE instead of tcp only

13 years agob=17748 fix sanityn test 33 to support multiple OSS nodes
Jian Yu [Fri, 3 Sep 2010 14:23:28 +0000 (22:23 +0800)]
b=17748 fix sanityn test 33 to support multiple OSS nodes

i=vitaly.fertman
i=grev

13 years agob=23702 removed space for compiler.
yangsheng [Sat, 4 Sep 2010 00:44:15 +0000 (08:44 +0800)]
b=23702 removed space for compiler.

i=panda

13 years agob=21563 Metadata performance has degraded for some operations between 1.6.5 and 1.8.1
Dmitry Zogin [Tue, 31 Aug 2010 19:54:08 +0000 (15:54 -0400)]
b=21563 Metadata performance has degraded for some operations between 1.6.5 and 1.8.1

 Change mdsrate tool to handle time with double resolution and use MPI_Wtime().
 Also stat operations measurements do not require --iter argument anymore, and do
 not use readdir() internally.

 i=johann
 i=iandrew.perepechko
 i=robert.read

13 years agob=22978 better conformation to coding standards
Brian J. Murrell [Tue, 31 Aug 2010 19:07:47 +0000 (15:07 -0400)]
b=22978 better conformation to coding standards

Per adilger:
- the preferred Lustre coding style is to not use assignments inside
  conditionals
- Lustre coding style recommends against using negation on non-boolean
  values
- while fixing this code, we may as well print the flag in octal, which
is how it is normally defined.

i=whitebear
i=panda

13 years agob=23216 a fix for echo_client_async_page memory allocation failure handling
Andrew Perepechko [Tue, 31 Aug 2010 13:18:28 +0000 (17:18 +0400)]
b=23216 a fix for echo_client_async_page memory allocation failure handling

i=Dmitry Zogin
i=ZhiYong Tian

13 years agob=23014 SGPDDSURVEY typo fix
Elena Gryaznova [Mon, 30 Aug 2010 12:13:25 +0000 (16:13 +0400)]
b=23014 SGPDDSURVEY typo fix

i=Andrew.Perepechko

set default SGPDDSURVEY value properly

13 years agob=21128 ldlm_cancel_lru fixes for stack overflow v4 and handle PF_MEMALLOC
pravin [Mon, 30 Aug 2010 09:27:47 +0000 (14:57 +0530)]
b=21128 ldlm_cancel_lru fixes for stack overflow v4 and handle PF_MEMALLOC

i=oleg
i=rahul
        avoid stack overflow by delegating work to separate thread

13 years agobump version to 1.8.4.51 v1_8_4_51
Johann Lombardi [Fri, 27 Aug 2010 22:00:36 +0000 (00:00 +0200)]
bump version to 1.8.4.51

13 years agob=23503 Oops at __percpu_counter_add+0x1b
Dmitry Zogin [Fri, 13 Aug 2010 16:52:12 +0000 (12:52 -0400)]
b=23503 Oops at __percpu_counter_add+0x1b

 Use bdi_init()/bdi_destroy() to proper initialize backing_dev_info
 structure regardless of HAVE_NEW_BACKING_DEV_INFO.

 i=sheng.yang
 i=johann.lombardi

13 years agob=22588 fix compile issue with gcc 4.4
Robert Read [Fri, 27 Aug 2010 21:36:31 +0000 (23:36 +0200)]
b=22588 fix compile issue with gcc 4.4

i=brian
i=andrew

add may_alias attribute to struct lov_user_md_v1

13 years agob=23645 remove warning printed at mount time with 2.6.32
Johann Lombardi [Fri, 27 Aug 2010 21:32:11 +0000 (23:32 +0200)]
b=23645 remove warning printed at mount time with 2.6.32

Make sure inode mutex on the CWD is taken when calling lookup_on_len().