Whamcloud - gitweb
tianzy [Tue, 5 May 2009 11:03:10 +0000 (11:03 +0000)]
Branch b1_8
use lqs to judge which uid/gid are set quota limitaton and
which uid/gid are not, and treat them differently.
b=18574
i=johann
i=panda
tianzy [Tue, 5 May 2009 10:57:38 +0000 (10:57 +0000)]
Branch b1_8
part I
1. mainly fix the lqs_key of lustre_qunit_size.
2. clean quota_compute_lqs().
3. only supply quota_search_lqs() to outside and make quota_create_lqs() internal.
b=18616
i=johann
i=panda
liuy [Tue, 5 May 2009 08:31:52 +0000 (08:31 +0000)]
Branch b1_8
some improvements:
-use fd->hints structure instead of MPI Info routines
-change hints name as romio requires
-check lustre/lustre_user.h header file in configure instead of giving
the lustre structs/constants
-add new MPE logging
-revert the hint settings in ad_lustre_open.c
-remove the two confusing hints "contig_data" and "samesize"
-fix avail_cb_node calculation in ad_lustre_aggregate.c
bobijam [Tue, 5 May 2009 02:12:07 +0000 (02:12 +0000)]
Branch b1_8
b=17402
i=vitaly.fertman
i=eric.mei
Clear up nid_stat's nid_exp_ref_count usage.
brian [Mon, 4 May 2009 21:45:44 +0000 (21:45 +0000)]
b=17671
i=zhen.liang
i=sheng.yang
Enable building of OFED 1.4.1 RC4 on b1_8 for testing in anticipation of
a 1.8.1 release with OFED 1.4.1 GA.
This includes RHEL5, which disables it's current "inkernel" specification
and reverts it's kernel configs accordingly.
brian [Mon, 4 May 2009 21:30:59 +0000 (21:30 +0000)]
b=17671
i=zhen.liang
i=sheng.yang
Enable Lustre to be built with OFED 1.4.1.
This is just the glue needed for the build with OFED 1.4.1 to actually work.
OFED 1.4.1 won't actually get built with a Lustre branch until a target file
specifies:
OFED_VERSION=1.4.1-rc4
Note that OFED 1.4.1 only supports RHEL4, RHEL5 and SLES10 targets. SLES9
users are out of luck (but have been for some time with OFED).
anserper [Mon, 4 May 2009 20:03:34 +0000 (20:03 +0000)]
b=19319
i=Alexander Zarochentsev
don't pass garbage in mode with O_CREAT
brian [Mon, 4 May 2009 18:50:25 +0000 (18:50 +0000)]
b=16312
i=wangyb
i=jack
- fixes vanilla build: typo
- enhanced incomplete cache detection
- put ext3 source into kernel-devel for rhel5 to get rid of ugly lbuild hack
- clean up the RPM smp type specification as well as .config selection
- we can actually get rid of the rhel5 -smp .configs in the lustre tree now
- find the kernel rpm by looking at rpm attributes instead of just filenames
- fix a bug where some old code was setting global variables it is no longer
able to set
- add "lnxmin" to the sles10 target file to deal with the breakage that bug
19336 will aim to fix properly
grev [Mon, 4 May 2009 12:34:16 +0000 (12:34 +0000)]
b=19143
i=Robert.Read
t-f pri-facet sec-facet different devs names fix
girish [Fri, 1 May 2009 15:23:05 +0000 (15:23 +0000)]
b=19058
b=19184
i=adilger,girish
o=Bernd Schubert
o=Brian Behlendorf
ext3_remount() should return correct error code
add locking for creation of jbd_4k slab cache. Only needed for RHEL5 kernel
shadow [Fri, 1 May 2009 11:40:35 +0000 (11:40 +0000)]
set notransno before readonly.
Branch b1_8
b=19085
i=rread
i=adilger
nathan [Thu, 30 Apr 2009 22:37:16 +0000 (22:37 +0000)]
b=19293
i=rread
i=ericm
move AT tunables to proc/sys/lustre, add conf_param path
yangsheng [Wed, 29 Apr 2009 09:28:25 +0000 (09:28 +0000)]
Branch b1_8
b=19024
i=shadow, bobijam
Update to RHEL5.3 2.6.18-128.1.6.el5 kernel.
yangsheng [Wed, 29 Apr 2009 07:44:13 +0000 (07:44 +0000)]
Branch b1_8
b=19212
i=shadow, bobijam
Update to SLES10 2.6.16.60-0.37 kernel.
yangsheng [Wed, 29 Apr 2009 06:31:24 +0000 (06:31 +0000)]
Branch b1_8
b=19276
i=johann, shadow, nathan
Fixed the swab issue for PPC platform.
bobijam [Wed, 29 Apr 2009 02:21:54 +0000 (02:21 +0000)]
Branch b1_8
b=18073
i=johann
i=adilger
i=sheng.yang
use "\\" as escape character to prevent '.' from being converted to '/' in
lctl get/set_param.
tianzy [Wed, 29 Apr 2009 00:58:53 +0000 (00:58 +0000)]
Branch b1_8
add lcounter_read_positive() so that it always returns positive integer
b=17098
i=shadow
i=jay
wangyb [Tue, 28 Apr 2009 08:15:20 +0000 (08:15 +0000)]
branch b1_8
b=18015
i=zhenyu.xu
i=sheng.yang
check if synchronize_rcu is available, otherwise use synchronize_kernel
fanyong [Sat, 25 Apr 2009 10:57:01 +0000 (10:57 +0000)]
Branch b1_8
b=19171
i=tianzy
i=robert.read
Hold obd reference count when quota recovery.
grev [Sat, 25 Apr 2009 09:36:24 +0000 (09:36 +0000)]
b=19262
i=Robert.Read
include functions.sh properly
fanyong [Sat, 25 Apr 2009 06:47:44 +0000 (06:47 +0000)]
Branch b1_8
b=15691
i=eric.mei
i=peng.ye
Another diagnostic patch to collect more log on both client and OSS.
adilger [Sat, 25 Apr 2009 01:12:15 +0000 (01:12 +0000)]
Branch b1_8
Quiet compiler warning.
b=15981
o=girish.shilamkar
i=adilger
i=robert.read
brian [Fri, 24 Apr 2009 18:20:41 +0000 (18:20 +0000)]
b=16312
i=robert.read
i=yibin.wang
Make lbuild build kernels based on the vendor's own SRPM rather than all of
the gyrations we used to go through to build the kernel tarballs and so on.
This should give people a kernel (and associated packages) that looks much
more like the kernel (and packages) they are used to getting from their
vendor.
This has resulted in a much easier to understand lbuild and once we finally
fully drop support for rhel4 and sles9 we can get rid of the
lustre-kernel-2.4.spec.in, build/lbuild.old_school and build/lmake files.
Anyone know knows anything about the build system will understand what kind
of simplicity in understanding of how the build works that will yield.
bobijam [Fri, 24 Apr 2009 02:18:16 +0000 (02:18 +0000)]
Branch b1_8
b=18683
i=joahnn
i=sheng.yang
If no options are given, tunefs.lustre/mkfs.lustre just print info.
adilger [Thu, 23 Apr 2009 08:13:02 +0000 (08:13 +0000)]
Branch b1_8
Quiet useless debug log spamming message.
b=19114
adilger [Thu, 23 Apr 2009 07:59:21 +0000 (07:59 +0000)]
Branch b1_8
Fix error message to match actual filenames.
Remove trailing whitespace.
b=19114
shadow [Thu, 23 Apr 2009 06:12:23 +0000 (06:12 +0000)]
correctly skip time estimate if in recovery
Branch b1_8
b=19223
i=johann
i=nathan
fanyong [Thu, 23 Apr 2009 04:48:08 +0000 (04:48 +0000)]
Branch b1_8
b=19149
i=tianzy
i=panda
Make sure quota has been released to master in test_6.
stinson1947 [Wed, 22 Apr 2009 19:06:00 +0000 (19:06 +0000)]
Updated for the next release. Changes provided by Johann.
shadow [Wed, 22 Apr 2009 18:07:32 +0000 (18:07 +0000)]
Allocate inodes for llog in last inode group for avoid broke recovery.
Branch b1_8
b=18192
i=adilger
i=shadow
i=girish
tappro [Wed, 22 Apr 2009 17:58:24 +0000 (17:58 +0000)]
- set close_req properly in mdc_close()
b:3633
i:vitaly,adilger
anserper [Wed, 22 Apr 2009 17:00:10 +0000 (17:00 +0000)]
b=19115
i=Johann Lombardi
just check for the oops
tappro [Wed, 22 Apr 2009 09:52:00 +0000 (09:52 +0000)]
- update versions for partially committed open request
b:19224
i:adilger,rread
grev [Tue, 21 Apr 2009 18:56:39 +0000 (18:56 +0000)]
b=19116
i=Robert.Read
convert "-" to X to create client variable name
grev [Tue, 21 Apr 2009 18:12:17 +0000 (18:12 +0000)]
b=19175
i=Robert.Read
check_catastrophe fix
tianzy [Tue, 21 Apr 2009 09:40:11 +0000 (09:40 +0000)]
Branch b1_8
1. delete the lc_watchdog_touch_ms()
2. changing to seconds instead of ms
3. delete lcw_time in lc_watchdog
b=18582
i=nathan
i=johann
grev [Mon, 20 Apr 2009 18:45:25 +0000 (18:45 +0000)]
b=19140
i=Alexander.Zarochentsev
PTLDEBUG/SUBSYSTEM t-f set_param fix:
accept the string value
anserper [Mon, 20 Apr 2009 17:44:13 +0000 (17:44 +0000)]
b=18695
o=Johann Lombardi
i=Andreas Dilger
i=Alexey Lyashkov
i=Andrew Perepechko
don't pack i_size on close when it's not needed
grev [Fri, 17 Apr 2009 11:58:16 +0000 (11:58 +0000)]
b=19107
i=Maxim.Patlasov
acc-sm liblustre skip list
tappro [Fri, 17 Apr 2009 05:24:59 +0000 (05:24 +0000)]
- fix typo in 18734 commit
tappro [Thu, 16 Apr 2009 18:08:40 +0000 (18:08 +0000)]
- take reference on export for commit callback
- move mfd clenanup to mds_disconnect() as mds_export_destroy() may be delayed
and busy inodes may remain during umount
b:18734
i:adilger,shadow
zhanghc [Thu, 16 Apr 2009 15:15:58 +0000 (15:15 +0000)]
b=17485
don't reuse orphan objects in OST during
establishing connection in MDS
i=shadow
i=bobijam
zam [Thu, 16 Apr 2009 08:00:59 +0000 (08:00 +0000)]
Branch b1_8
b=18988
i=Alexey.Lyashkov
new sanity test for for O_APPEND.
yangsheng [Wed, 15 Apr 2009 10:32:54 +0000 (10:32 +0000)]
Branch b1_8
b=18033
Fixed the typo.
zhanghc [Wed, 15 Apr 2009 08:45:59 +0000 (08:45 +0000)]
b=18733
add protection (spin_lock(&lock->l_lock) before
using lock->l_resource in "lock_res_and_lock"
i=johann
i=green
nathan [Tue, 14 Apr 2009 21:50:57 +0000 (21:50 +0000)]
b=19047
i=adilger
i=nrutman
add llog_reader man page
shadow [Tue, 14 Apr 2009 18:43:54 +0000 (18:43 +0000)]
revert fix
b=16839
yangsheng [Tue, 14 Apr 2009 15:42:04 +0000 (15:42 +0000)]
Branch b1_8
b=18399
i=adilger, johann
Patch kernel to avoid deadlock in prune_icache().
Authro: ZhaoHongChao
yangsheng [Tue, 14 Apr 2009 12:54:50 +0000 (12:54 +0000)]
file prune-icache-use-trylock-sles10.patch was added on branch b1_8 on 2009-04-14 15:42:03 +0000
yangsheng [Tue, 14 Apr 2009 12:54:48 +0000 (12:54 +0000)]
file prune-icache-use-trylock-rhel5.patch was added on branch b1_8 on 2009-04-14 15:42:03 +0000
anserper [Tue, 14 Apr 2009 00:04:32 +0000 (00:04 +0000)]
b=18728
i=Alex Zhuravlev
i=Alexander Zarochentsev
decrease response time for write-waiting threads
anserper [Mon, 13 Apr 2009 23:56:01 +0000 (23:56 +0000)]
b=16890
i=ZhiYong Tian
i=Johann Lombardi
admin qf memset reduction, backport from the kernel patch (5.93 KB, patch)
anserper [Mon, 13 Apr 2009 23:51:24 +0000 (23:51 +0000)]
b=18951
i=Nathan Rutman
i=Vladimir Saveliev
fix for a specific race in multiop
anserper [Mon, 13 Apr 2009 23:37:22 +0000 (23:37 +0000)]
b=19063
i=Elena Gryaznova
quota_save_version must wait until switch to the new version has happened
anserper [Mon, 13 Apr 2009 23:00:14 +0000 (23:00 +0000)]
b=19115
i=Andreas Dilger
i=Alexander Zarochentsev
avoid oops coming from a wrong userspace flag passed to open(2); test 33b is temporarily disabled
shadow [Mon, 13 Apr 2009 19:00:58 +0000 (19:00 +0000)]
add lazystatfs mount option to allow statfs(2) to skip down OSTs
Branch b1_8
b=17974
i=green
i=vitaly
grev [Sat, 11 Apr 2009 09:34:00 +0000 (09:34 +0000)]
b=18779
i=Vladimir.Saveliev
i=Nathan.Rutman
test_34c fix
grev [Fri, 10 Apr 2009 20:51:41 +0000 (20:51 +0000)]
b=18271
i=Nathan.Rutman
writeconf fn fix
grev [Fri, 10 Apr 2009 15:00:25 +0000 (15:00 +0000)]
b=19139
i=Jian.Yu
typo END_RUN_FILE name fix
grev [Fri, 10 Apr 2009 14:46:49 +0000 (14:46 +0000)]
b=19129
i=Alexander.Zarochentsev
zconf_mount_clients remote command fix
yujian [Fri, 10 Apr 2009 09:09:53 +0000 (09:09 +0000)]
Branch b1_8
b=18501
i=nathan
Document the limitation for lustre_config in the upgrading case and
specify "--nomgs" option to a non-co-located MDT.
bobijam [Fri, 10 Apr 2009 02:50:44 +0000 (02:50 +0000)]
Branch b1_8
b=18918
i=nathan.rutman
fix service name on (un)mount for MGS.b=18918
bobijam [Fri, 10 Apr 2009 01:46:15 +0000 (01:46 +0000)]
Branch b1_8
b=18638
i=jinshan.xiong (jay)
i=andrew.perepachko (panda)
ldlm_lock_put() use RCU to call ldlm_lock_free, so need call synchronize_rcu()
to wait a grace period elapsed, so that ldlm_lock_free() get a chance to be
called.
adilger [Thu, 9 Apr 2009 00:05:16 +0000 (00:05 +0000)]
Branch b1_8
Reinstate upper limit on LDLM threads.
b=18688
i=robert.read
zhanghc [Wed, 8 Apr 2009 13:08:26 +0000 (13:08 +0000)]
b=17491
eliminate "wrpped times" stuff in original patch
i=nathan.rutman
i=robert.read
shadow [Wed, 8 Apr 2009 11:02:23 +0000 (11:02 +0000)]
kill one configure test.
Branch b1_8
b=14250
i=shadow
i=adilger
shadow [Wed, 8 Apr 2009 06:25:36 +0000 (06:25 +0000)]
Don't allow reconnect to already connected import.
Branch b1_8
b=16839
i=green
i=bzzz
bobijam [Wed, 8 Apr 2009 02:50:28 +0000 (02:50 +0000)]
Branch b1_8
b=17378
i=robert.read
add test case: sanity.sh test_17h().
bobijam [Wed, 8 Apr 2009 02:18:02 +0000 (02:18 +0000)]
Branch b1_8
b=18688
o=adilger
i=nathan.rutman
i=johann
Description: Allow tuning service thread via /proc
Details : For each service a new
/proc/fs/lustre/{service}/*/thread_{min,max,started} entry is
created that can be used to set min/max thread counts, and get the
current number of running threads.
anserper [Tue, 7 Apr 2009 22:22:09 +0000 (22:22 +0000)]
b=18619
i=Vitaly Fertman
i=Oleg Drokin
make oap_async_flags access atomic
grev [Tue, 7 Apr 2009 18:04:59 +0000 (18:04 +0000)]
b=18080
i=Alexander.Zarochentsev
add sanityN test_35 to A_E list
anserper [Mon, 6 Apr 2009 19:46:36 +0000 (19:46 +0000)]
b=18695
o=Johann Lombardi
i=Andreas Dilger
the regression test
nathan [Mon, 6 Apr 2009 19:36:32 +0000 (19:36 +0000)]
b=18798
i=rread
i=adliger
Track import connection state changes in a new osc/mdc proc file
Add overview-type data to the osc/mdc import proc file.
yangsheng [Mon, 6 Apr 2009 17:25:47 +0000 (17:25 +0000)]
Branch b1_8
b=17701
i=shadow, adilger
Cleanup and some fixes for lov_user_md_v{1,3} swab.
shadow [Mon, 6 Apr 2009 15:33:17 +0000 (15:33 +0000)]
fix warning.
grev [Sat, 4 Apr 2009 02:05:17 +0000 (02:05 +0000)]
b=18914
i=Brian
check_client_load: try to connect several times to avoid ENFILE
grev [Sat, 4 Apr 2009 01:05:54 +0000 (01:05 +0000)]
b=17973
i=Brian
mpi_run: process mpi output to get an error
grev [Sat, 4 Apr 2009 00:11:39 +0000 (00:11 +0000)]
b=19010
i=Alexander.Zarochentsev (Zam)
t-f sanity check if env is sane
johann [Fri, 3 Apr 2009 23:42:35 +0000 (23:42 +0000)]
Branch b1_8
b=18695
i=adilger
i=shadow
don't update i_size on MDS_CLOSE for directories.
johann [Fri, 3 Apr 2009 23:38:58 +0000 (23:38 +0000)]
Branch b1_8
b=18695
i=adilger
Add LASSERTF in fsfilt_ext3_setattr() when truncating a directory.
grev [Fri, 3 Apr 2009 19:34:06 +0000 (19:34 +0000)]
b=17973
i=Brian
run_* scripts fix: kill PGID properly
grev [Fri, 3 Apr 2009 17:35:11 +0000 (17:35 +0000)]
b=18404
o=Adilger
i=grev
i=Johann
write-append-truncate cleanup
zhanghc [Fri, 3 Apr 2009 09:21:04 +0000 (09:21 +0000)]
b=18364
1, fix a problem in osc_brw_async, which use the same
obdo when there is too much data to be written into
OST and trigger a LASSERT(!(oa->o_valid & bits))
in function "osc_announce_cached"
2, initialize the local variable "obdo" (struct obdo)
in different functions, for it may cause some
strange problem in system
i=adilger@sun.com
i=johann@sun.com
adilger [Thu, 2 Apr 2009 06:28:33 +0000 (06:28 +0000)]
Branch b1_8
Fix unused variable warning due to code restructuring.
b=17710
adilger [Thu, 2 Apr 2009 05:10:45 +0000 (05:10 +0000)]
Branch b1_8
Ignore generated files.
adilger [Thu, 2 Apr 2009 04:38:19 +0000 (04:38 +0000)]
Branch b1_8
Fix line wrap to 80 columns.
adilger [Thu, 2 Apr 2009 04:35:12 +0000 (04:35 +0000)]
Branch b1_8
Update comment.
adilger [Thu, 2 Apr 2009 04:29:17 +0000 (04:29 +0000)]
Branch b1_8
Quiet compiler warning for comparison between signed and unsigned types.
adilger [Thu, 2 Apr 2009 04:15:00 +0000 (04:15 +0000)]
Branch b1_8
Quiet unused "path" variable warning.
anserper [Wed, 1 Apr 2009 21:42:35 +0000 (21:42 +0000)]
b=18317
i=Johann Lombardi
i=ZhiYong Tian
wait until all lqs have been freed on cleanup
grev [Wed, 1 Apr 2009 16:46:52 +0000 (16:46 +0000)]
b=18696
i=Tappro
RECOVERY_RANDOM_SCALE typo fix
grev [Wed, 1 Apr 2009 16:13:31 +0000 (16:13 +0000)]
b=15266
i=Brian
new mdsrate-lookup-10dirs test (cmd3-6)
correct spelling
grev [Wed, 1 Apr 2009 16:06:51 +0000 (16:06 +0000)]
b=15266
i=Brian
new mdsrate-lookup-10dirs test (cmd3-6)
bobijam [Wed, 1 Apr 2009 02:35:27 +0000 (02:35 +0000)]
Branch b1_8
b=18895
i=tappro
i=alexey.lyashkov (shadow)
when mds_setup() fails the error handle path needs clean possibly built
nid_stat which builds nid_proc on obd's obd_proc_exports_entry.
anserper [Tue, 31 Mar 2009 20:58:50 +0000 (20:58 +0000)]
b=18972
i=Elena
fix a typo in test 24i of sanity.sh
shadow [Tue, 31 Mar 2009 07:22:05 +0000 (07:22 +0000)]
don't leak obd reference on failed llog setup
Branch b1_8
b=18896
i=tappro
i=rread
tappro [Tue, 31 Mar 2009 06:21:37 +0000 (06:21 +0000)]
- avoid assertion if second open|create found orphan again
b:18927
i:rread,johann
shadow [Tue, 31 Mar 2009 06:00:17 +0000 (06:00 +0000)]
allow kill process which stick in waiting statahead result.
Branch b1_8
b=18902
i=fanyong
i=tappro
fanyong [Tue, 31 Mar 2009 04:40:49 +0000 (04:40 +0000)]
Branch b1_8
b=18903
i=eric.mei
i=w.li
typo fix
wangdi [Mon, 30 Mar 2009 20:53:36 +0000 (20:53 +0000)]
Branch: b1_8
b=18794
remove event once there are no obd registered under that event.
i=Ericm,Huanghua