Whamcloud - gitweb
fs/lustre-release.git
14 years agob=21968 conf-sanity changes for config mgs and mds are not combined
Elena Gryaznova [Wed, 24 Feb 2010 15:53:14 +0000 (18:53 +0300)]
b=21968 conf-sanity changes for config mgs and mds are not combined

i=Nathan.Rutman

includes t-f:combined_mgs_mds ()

14 years agob=21097 quote the passing on of it's own arguments through $@
Jian Yu [Tue, 23 Feb 2010 07:14:00 +0000 (15:14 +0800)]
b=21097 quote the passing on of it's own arguments through $@

Fix do_nodes() in acc-sm test-framework to quote the passing on
of it's own arguments through $@ to do_node().

i=jian.yu
i=edg

14 years agob=21869 improve filter_fid2dentry() error message
Landen [Mon, 22 Feb 2010 03:19:38 +0000 (11:19 +0800)]
b=21869 improve filter_fid2dentry() error message

i=landen
i=girish

14 years agob=20433 decrease the usage of memory on clients.
Landen [Mon, 22 Feb 2010 03:08:46 +0000 (11:08 +0800)]
b=20433 decrease the usage of memory on clients.

1. On clients, recycle dentries and inodes unused.
2. Delete the code related to ll_deathrow(att 6215 in bug 1443). It
   is useless now.

i=adilger
i=oleg

14 years agob=21137 ext4 extent allocation is slower than in ext3
dzogin [Fri, 19 Feb 2010 16:19:17 +0000 (11:19 -0500)]
b=21137 ext4 extent allocation is slower than in ext3

 Increase the default value of MB_DEFAULT_ORDER2_REQS to 8, enlarge ext4
 preallocation table for 2048 4K blocks extents creation.

 i=alex.zhuravlev
 i=andrew.perepechko

14 years agoUpdated with Oracle copyright. v1_8_2_50
Terry Rutledge [Fri, 19 Feb 2010 17:12:09 +0000 (10:12 -0700)]
Updated with Oracle copyright.
2010-01-29  Sun Microsystems, Inc.
        * version 3.1.1

-------------------------------------------------------------------------------

14 years agob=22074 incorrect triggering of synchronous IO
Brian Behlendorf [Wed, 17 Feb 2010 17:49:23 +0000 (12:49 -0500)]
b=22074 incorrect triggering of synchronous IO

The OSC can mistakenly fall back to synchronous IO when the max_dirty_mb
limit is reached and no write requests have yet been issued.  This can
occur when the dirty pages are spread over many files all of which are
below the optimal request size.

o=behlendorf1@llnl.gov (Brian Behlendorf)
i=adilger
i=tom.wang

14 years agob=20383 fix errant m4 "dnl" usage
Brian J. Murrell [Wed, 17 Feb 2010 16:51:46 +0000 (11:51 -0500)]
b=20383 fix errant m4 "dnl" usage

Some dnl() usage seems to have been causing some errors in the resulting
configure script.

o=tappro
i=brian
i=adilger

14 years agob=21913 conf-sanity test_35b FAIL due to client reconnection has not been triggered
Dmitry Zogin [Fri, 12 Feb 2010 22:27:32 +0000 (17:27 -0500)]
b=21913 conf-sanity test_35b FAIL due to client reconnection has not been triggered

 Improve the robustness of conf-sanity test_35b.

 i=johann

14 years agoRevert "b=20989 fix race in nfs export code"
Johann Lombardi [Wed, 17 Feb 2010 11:49:30 +0000 (12:49 +0100)]
Revert "b=20989 fix race in nfs export code"

This reverts commit 09f7487326e23dfed472bfbf752b5ee58c225855.

Conflicts:

lustre/ChangeLog

14 years agob=21829 fix broken llobdstat and add a counter parameter
Brian J. Murrell [Fri, 12 Feb 2010 20:52:12 +0000 (15:52 -0500)]
b=21829 fix broken llobdstat and add a counter parameter

Need to make sure we limit the search for OBD stats files to the obdfilter
subdirectory of "/proc/fs/lustre".
Add a counter argument to limit the number of items returned when using the
interval parameter.
Fix lots of whitespace atrocities as well as better format some of the code.

i=cliff
i=wangyb

14 years agob=13520 PTLRPC_PAUSE_REQ checking should ignore PING.
Eric Mei [Fri, 12 Feb 2010 18:48:08 +0000 (11:48 -0700)]
b=13520 PTLRPC_PAUSE_REQ checking should ignore PING.

r=nathan
r=wangdi

14 years agob=20355 Add $(PTHREAD_LIBS) to lctl and lfs build
Cliff White [Wed, 3 Feb 2010 19:41:54 +0000 (11:41 -0800)]
b=20355 Add $(PTHREAD_LIBS) to lctl and lfs build

i=johann

$(PTHREAD_LIBS) is needed to compile lctl and lfs for BG/P

14 years agob=20918 t-f max recovery time estimation
Elena Gryaznova [Thu, 11 Feb 2010 20:51:33 +0000 (23:51 +0300)]
b=20918 t-f max recovery time estimation

i=Nathan.Rutman
i=Brian.Murrell

inform user about the estimated maximum recovery time value
to help him to set the server FAILOVER period properly

14 years agob=21932 disable some tests on NFSCLIENT
Elena Gryaznova [Thu, 11 Feb 2010 20:29:40 +0000 (23:29 +0300)]
b=21932 disable some tests on NFSCLIENT

i=Oleg.Drokin

disable simul,  write_disjoint, write_append_truncate for NFSCLIENT mode;
setstripe on lustre client (nfsserver) instead of nfs client for ior test

14 years agob=21919 Optimize quota_ctl operations by sending requests in parallel
Andrew Perepechko [Thu, 11 Feb 2010 19:48:35 +0000 (22:48 +0300)]
b=21919 Optimize quota_ctl operations by sending requests in parallel

i=Johann Lombardi
i=ZhiYong Tian

Based on a patch from Joseph Herring (LLNL).
Send MDS->OST quota_ctl requests in parallel, do not resend.
Compiled from two attachments in the ticket.

14 years agob=18030 dealock fix
Alexander.Zarochentev [Thu, 11 Feb 2010 17:28:08 +0000 (20:28 +0300)]
b=18030 dealock fix

start the transaction earlier in llog_lvfs_destroy to get transaction start
and inode mutex lock nested properly.

i=adilger
i=alex.zhuravlev

14 years agob=21264 workaround dd bus error
Alexander.Zarochentev [Thu, 11 Feb 2010 15:59:25 +0000 (18:59 +0300)]
b=21264 workaround dd bus error

A buggy coreutils/gettext combination workaround.  Suppressing dd xfer
statistic makes dd do not call gnu gettext library and avoid crashing.

i=robert.read

14 years agob=15057 fix file ownerships in lustre-modules RPM
Brian J. Murrell [Wed, 10 Feb 2010 18:47:45 +0000 (13:47 -0500)]
b=15057 fix file ownerships in lustre-modules RPM

The files in the lustre-modules RPM were not being set with a correct
owner and were therefore just using what was on the filesystem.

i=yangsheng
i=wangyb

14 years agob=21665 a small fix for "lfs osts"
Andrew Perepechko [Wed, 10 Feb 2010 16:34:14 +0000 (19:34 +0300)]
b=21665 a small fix for "lfs osts"

a=Joseph Herring
i=Andreas Dilger
i=Andrew Perepechko

Actually, we don't want to traverse the directory tree,
so return a positive value from sem_init to terminate
the traversal before it starts.

14 years agob=21882 handle SLV==1 on client side
Vitaly Fertman [Wed, 10 Feb 2010 16:12:16 +0000 (19:12 +0300)]
b=21882 handle SLV==1 on client side

Initialize ldlm pool SLV to 0 on client side to handle SLV==1 obtained from server correctly

i=rreed
i=adilger

14 years agob=21882 lru resize SLV can get stuck
Vitaly Fertman [Wed, 10 Feb 2010 16:11:32 +0000 (19:11 +0300)]
b=21882 lru resize SLV can get stuck

calculate SLV with a greater precision to not lose small changes due to interger math truncation;
round up SLV only if the amount of granted locks less than the limit to not get stuck with this SLV

i-adilger
i=green

14 years agob=21666 prevent use of OFED source dir instead of headers
Brian J. Murrell [Mon, 8 Feb 2010 19:41:49 +0000 (14:41 -0500)]
b=21666 prevent use of OFED source dir instead of headers

Try to determine if the user is pointing configure at the OFED source
directory intead of the devel/headers directory and error out of
configure if so and display an informative warning.

i=panda
i=whitebear

14 years agob=19553 Ignore broken cancel_dirty_page() in OFED 1.4.1
Brian J. Murrell [Mon, 8 Feb 2010 19:34:16 +0000 (14:34 -0500)]
b=19553 Ignore broken cancel_dirty_page() in OFED 1.4.1

OFED 1.4.1 had a broken implementation of cancel_dirty_page for SLES10.
This patch detects that and ignores the function if found.

i=johann
i=yangsheng

14 years agob=19336 Get rid of the EXTRA_VERSION_DELIMETER shenanigans
Brian J. Murrell [Mon, 8 Feb 2010 17:54:41 +0000 (12:54 -0500)]
b=19336 Get rid of the EXTRA_VERSION_DELIMETER shenanigans

We used to carry around a bunch of baggage in order to specify what kind
of delimeter to put between the version and "extra version".  The truth of
the matter is that this should always be "-".

This patch includes some support for a build system developer to force
an uncached rebuild of all products.

i=wangyb
i=yangsheng

14 years agob=21882 lru-resize comment cleanups i=adilger
Vitaly Fertman [Thu, 4 Feb 2010 22:39:32 +0000 (01:39 +0300)]
b=21882 lru-resize comment cleanups i=adilger

14 years agob=20718 sanity-quota test_14b: improve error message
Elena Gryaznova [Wed, 3 Feb 2010 19:25:32 +0000 (22:25 +0300)]
b=20718 sanity-quota test_14b: improve error message

i=Andrew.Perepechko

make check_runas_id_ret () more verbose

14 years agob=21961 add changelog entry
Johann Lombardi [Thu, 4 Feb 2010 22:24:44 +0000 (23:24 +0100)]
b=21961 add changelog entry

14 years agob=21961 (17914) ignore trailing -mdc when determining index number
Nathan Rutman [Thu, 4 Feb 2010 21:58:08 +0000 (13:58 -0800)]
b=21961 (17914) ignore trailing -mdc when determining index number

a=jinshan.xiong
i=nathan
i=h.huang

14 years agob=21990 add parallel-scale EXCEPT list
Elena Gryaznova [Thu, 4 Feb 2010 16:30:52 +0000 (19:30 +0300)]
b=21990 add parallel-scale EXCEPT list

i=Minh.Diep

14 years agob=21966 avoid divide-by-zero in lprocfs_rd_import()
Andreas Dilger [Thu, 4 Feb 2010 12:37:29 +0000 (13:37 +0100)]
b=21966 avoid divide-by-zero in lprocfs_rd_import()
i=johann

14 years agob=16909 simplify MDT/OST service start message
Johann Lombardi [Thu, 4 Feb 2010 08:15:55 +0000 (09:15 +0100)]
b=16909 simplify MDT/OST service start message

i=nathan
i=adilger

14 years agob=21686 add changelog entry
Johann Lombardi [Thu, 4 Feb 2010 08:12:44 +0000 (09:12 +0100)]
b=21686 add changelog entry

14 years agob=16909 Simplify MDT/OST service start message
Landen [Thu, 4 Feb 2010 07:57:49 +0000 (15:57 +0800)]
b=16909 Simplify MDT/OST service start message

i=nathan
i=adilger

14 years agob=18489 test_116, test_118k cleanup
Elena Gryaznova [Wed, 3 Feb 2010 21:58:09 +0000 (00:58 +0300)]
b=18489 test_116, test_118k cleanup

i=Andrew.Perepechko (panda)
i=Andreas.Dilger

14 years agob=21953 use separate failover counter for each facet
Elena Gryaznova [Wed, 3 Feb 2010 21:37:08 +0000 (00:37 +0300)]
b=21953 use separate failover counter for each facet

i=Mikhail.Pershin (tappro)

14 years agoadd 1.8.3 section in the lnet changelog
Johann Lombardi [Wed, 3 Feb 2010 18:46:33 +0000 (19:46 +0100)]
add 1.8.3 section in the lnet changelog

14 years agob=21147 call build_lqs only from generic_quota_on
Andrew Perepechko [Wed, 3 Feb 2010 17:13:59 +0000 (20:13 +0300)]
b=21147 call build_lqs only from generic_quota_on

i=Johann Lombardi
i=ZhiYong Tian

14 years agob=21259 "lfs check" is only allowed for root.
Dmitry Zogin [Wed, 3 Feb 2010 16:54:11 +0000 (11:54 -0500)]
b=21259 "lfs check" is only allowed for root.

 Code cleanup around obd_class_*() functions and sanity test for non-root lfs check

 i=adilger
 i=andrew.perepechko

14 years agob=21632 Kernel update to OEL5.4 2.6.18-164.11.1.0.1.el5.
yangsheng [Tue, 2 Feb 2010 15:15:37 +0000 (23:15 +0800)]
b=21632 Kernel update to OEL5.4 2.6.18-164.11.1.0.1.el5.

i=johann

14 years agobump version to 1.8.2.50
Johann Lombardi [Wed, 3 Feb 2010 18:35:51 +0000 (19:35 +0100)]
bump version to 1.8.2.50

14 years agoadd changelog section for 1.8.3
Johann Lombardi [Wed, 3 Feb 2010 18:34:07 +0000 (19:34 +0100)]
add changelog section for 1.8.3

14 years agoRevert debug patch from b=21364
Johann Lombardi [Tue, 2 Feb 2010 10:49:29 +0000 (11:49 +0100)]
Revert debug patch from b=21364

This reverts commit 818de83d3200ae48dae7096500ba0118b8f95976.
I inadvertently committed my debug patch.

14 years agob=20970 need add an additional barrier for write_disjoint
Landen [Fri, 29 Jan 2010 07:13:42 +0000 (15:13 +0800)]
b=20970 need add an additional barrier for write_disjoint

i=rread
i=grev

14 years agob=21948 skip parallel grouplock test for NFSCLIENT mode
Elena Gryaznova [Thu, 28 Jan 2010 15:06:18 +0000 (18:06 +0300)]
b=21948 skip parallel grouplock test for NFSCLIENT mode

i=Johann

14 years agob=21900 ost-pools test_25: FAIL: /mnt/lustre/d0.ost-pools/d25/file1 not allocated...
Dmitry Zogin [Wed, 27 Jan 2010 18:04:27 +0000 (13:04 -0500)]
b=21900 ost-pools test_25: FAIL: /mnt/lustre/d0.ost-pools/d25/file1 not allocated from OSTs 0.

 Modify ost-pools test_25 to wait for MDS-OST connection to re-establish.

 i=johann

14 years agodelete test_12 in sanity-quota.sh
Landen [Wed, 27 Jan 2010 07:48:03 +0000 (15:48 +0800)]
delete test_12 in sanity-quota.sh

14 years agob=21686 fail the request if its obd_device stopping
root [Sat, 30 Jan 2010 07:02:05 +0000 (15:02 +0800)]
b=21686 fail the request if its obd_device stopping

in ldlm_handle_enqueue, the request should be failed
if its obd_device had been marked as "fail"(obd_fail=1),
which will be set during umount.

i=johann@sun.com
i=oleg.drokin@sun.com

14 years agoMerge branch 'b1_8' of git@git.lustre.org:prime/lustre into b1_8
Johann Lombardi [Tue, 2 Feb 2010 10:09:15 +0000 (11:09 +0100)]
Merge branch 'b1_8' of git@git.lustre.org:prime/lustre into b1_8

14 years agoForgot to update these when I updated the lustre/ChangeLog. Added date
Terry Rutledge [Sat, 30 Jan 2010 20:57:52 +0000 (13:57 -0700)]
Forgot to update these when I updated the lustre/ChangeLog.  Added date
and another supported version of OFED.

14 years agob=21364 debug patch
Johann Lombardi [Fri, 29 Jan 2010 00:53:20 +0000 (01:53 +0100)]
b=21364 debug patch

14 years agob=21815 lustre_hash_rehash_key() should use lh_read_unlock()
Johann Lombardi [Wed, 27 Jan 2010 14:39:49 +0000 (15:39 +0100)]
b=21815 lustre_hash_rehash_key() should use lh_read_unlock()

lh_read_lock() is no-op if rehash is disabled, so we should
use lh_read_unlock() in this function.
This should not have any consequence, but better to fix it.

14 years agob=21815 move assertion under write lock
Johann Lombardi [Wed, 27 Jan 2010 14:27:07 +0000 (15:27 +0100)]
b=21815 move assertion under write lock

14 years agob=21815 print more debug info in lustre_hash_exit when assertion fails
Johann Lombardi [Wed, 27 Jan 2010 12:49:37 +0000 (13:49 +0100)]
b=21815 print more debug info in lustre_hash_exit when assertion fails

14 years agob=19405 do not flag a request as rq_replay for non replayable imports
Vladimir V. Saveliev [Tue, 26 Jan 2010 16:12:40 +0000 (17:12 +0100)]
b=19405  do not flag a request as rq_replay for non replayable imports

i=ericm
i=robert

14 years agob=21906 LBUG doesn't print stack trace on sles9 because show_stack not exported
Johann Lombardi [Tue, 26 Jan 2010 16:04:06 +0000 (17:04 +0100)]
b=21906 LBUG doesn't print stack trace on sles9 because show_stack not exported

14 years agoRevert "b=21097 fix md5sum error in metadata-updates.sh" 1.8.2 v1_8_2_RC4
Johann Lombardi [Sat, 23 Jan 2010 00:36:25 +0000 (01:36 +0100)]
Revert "b=21097 fix md5sum error in metadata-updates.sh"

This reverts commit 89b5d6f0e40b35bcc93d6830568e823d67e8f364.

14 years agob=17682 fix time unit in message
Johann Lombardi [Fri, 22 Jan 2010 23:03:01 +0000 (00:03 +0100)]
b=17682 fix time unit in message

14 years agob=21448 send recovery rpc ASAP
Johann Lombardi [Fri, 22 Jan 2010 21:16:02 +0000 (22:16 +0100)]
b=21448 send recovery rpc ASAP

i=robert.read
i=tappro

14 years agob=21406 fix deadlock between kjournald2 and ost_io thread
Johann Lombardi [Fri, 22 Jan 2010 21:03:26 +0000 (22:03 +0100)]
b=21406 fix deadlock between kjournald2 and ost_io thread

i=adilger
i=girish

Calling clear_page_dirty_for_io() is no longer needed since
we are granted that no dirty pages can be left in the page
cache by partial truncate. The problem is that
clear_page_dirty_for_io() can temporarilly mark the page
as dirty in the radix tree, which can cause deadlock
between jbd commit and bulk write handling.

14 years agob=17569 add force_over_16tb for rhel5/ext4
Johann Lombardi [Fri, 22 Jan 2010 20:49:33 +0000 (21:49 +0100)]
b=17569 add force_over_16tb for rhel5/ext4

16TB is the next limit.

14 years agob=17569 remove force_over_8tb for rhel5/ext4 since it is now tested
Johann Lombardi [Fri, 22 Jan 2010 15:26:27 +0000 (16:26 +0100)]
b=17569 remove force_over_8tb for rhel5/ext4 since it is now tested

14 years agob=21686 revert attach 25564 bug 19557
Johann Lombardi [Fri, 22 Jan 2010 15:00:30 +0000 (16:00 +0100)]
b=21686 revert attach 25564 bug 19557

14 years agob=21632 Update RHEL5.4 kernel to 2.6.16-164.11.1.el5.
yangsheng [Fri, 22 Jan 2010 12:53:02 +0000 (20:53 +0800)]
b=21632 Update RHEL5.4 kernel to 2.6.16-164.11.1.el5.

14 years agob=21147 fix unnecessary semaphore release in generic_quota_on
Andrew Perepechko [Thu, 21 Jan 2010 12:08:28 +0000 (15:08 +0300)]
b=21147 fix unnecessary semaphore release in generic_quota_on

i=Johann Lombardi

14 years agoadd missing changelog entries
Johann Lombardi [Thu, 21 Jan 2010 10:43:42 +0000 (11:43 +0100)]
add missing changelog entries

14 years agob=21595 jbd2/rhel5: don't call jbd callback with spinlock hold
Rahul Deshmukh [Thu, 21 Jan 2010 07:42:38 +0000 (13:12 +0530)]
b=21595 jbd2/rhel5: don't call jbd callback with spinlock hold

since the callback is allowed to sleep (e.g. take semaphore), we should
not hold any spinlocks when involing it.
jbd2/sles11 is fixed already.

i=johann
i=girish

14 years agob=21828 drop number of active requests when queued for recovery
Dmitry Zogin [Wed, 20 Jan 2010 23:54:49 +0000 (18:54 -0500)]
b=21828 drop number of active requests when queued for recovery

Now that we take a reference on the original request instead of
making a copy of it for recovery.  We need to drop the number of
active requests or the queued requests will prevent all request
processing when they exceed (srv->srv_threads_running - 1).

 i=nathan.rutman
 i=tappro

14 years agob=21826 refuse to invalidate operational quota files when they are in use
Andrew Perepechko [Wed, 20 Jan 2010 20:06:38 +0000 (23:06 +0300)]
b=21826 refuse to invalidate operational quota files when they are in use

an attempt to invalidate operational quota files on the quota master is not actually permitted by VFS (returning -EPERM), but we should not depend on that and should return the error earlier.

i=Johann Lombardi
i=ZhiYong Tian

14 years agob=19742 fix llite fiemap interfaces v1_8_2_RC3
Rahul Deshmukh [Mon, 18 Jan 2010 21:49:40 +0000 (22:49 +0100)]
b=19742 fix llite fiemap interfaces

i=johann

llite can get fiemap requests through ioctl or directly
through the ->fiemap vfs inode's operation (newer kernel).
Unfortunately, both interfaces take different arguments,
so the purpose of this patch is to fix this.

14 years agob=19742 fix fiemap patches for rhel5
Rahul Deshmukh [Mon, 18 Jan 2010 21:48:01 +0000 (22:48 +0100)]
b=19742 fix fiemap patches for rhel5

i=girish
i=andreas

14 years agob=21370 sanity 27x: double the qos_maxage timeout
James Simmons [Mon, 18 Jan 2010 17:36:07 +0000 (18:36 +0100)]
b=21370 sanity 27x: double the qos_maxage timeout
i=adilger

14 years agob=19742 add missing fiemap patches to rhel5 series
Johann Lombardi [Mon, 18 Jan 2010 15:50:39 +0000 (16:50 +0100)]
b=19742 add missing fiemap patches to rhel5 series

i=adilger
i=girish

14 years agob=21846 define lqs_key for quota lqs
Johann Lombardi [Mon, 18 Jan 2010 15:01:59 +0000 (16:01 +0100)]
b=21846 define lqs_key for quota lqs

i=adilger
i=landen

14 years agoupdate supported kernels in changelog and which_patch
Johann Lombardi [Mon, 18 Jan 2010 14:53:27 +0000 (15:53 +0100)]
update supported kernels in changelog and which_patch

14 years agob=20758 Update kernel to 2.6.16.60-0.42.8.
yangsheng [Mon, 18 Jan 2010 14:35:18 +0000 (22:35 +0800)]
b=20758 Update kernel to 2.6.16.60-0.42.8.

i=johann
i=landen

14 years agob=18690 disable rehash for quota v1_8_2_RC2
Johann Lombardi [Mon, 18 Jan 2010 10:10:33 +0000 (11:10 +0100)]
b=18690 disable rehash for quota

quota can uses a key of 0 for root and the rehash code
assert on this. Disable rehashing for quota lqs for now.

14 years agoUpdated with correct version string for 1.8.2 RC1. v1_8_2 v1_8_2_RC1
Terry Rutledge [Sat, 16 Jan 2010 18:26:35 +0000 (11:26 -0700)]
Updated with correct version string for 1.8.2 RC1.

14 years agoUpdated for 1.8.2 RC1. Added release date.
Terry Rutledge [Sat, 16 Jan 2010 18:25:01 +0000 (11:25 -0700)]
Updated for 1.8.2 RC1.  Added release date.

14 years agob=21564 add changelog entry
Girish Shilamkar [Sat, 16 Jan 2010 17:34:06 +0000 (18:34 +0100)]
b=21564 add changelog entry

14 years agob=21564 Print mmp_check_interval in kmmpd, fix sles10 & rhel5/ext3 too
Girish Shilamkar [Sat, 16 Jan 2010 16:16:57 +0000 (21:46 +0530)]
b=21564 Print mmp_check_interval in kmmpd, fix sles10 & rhel5/ext3 too

The patch not only prints mmp_check_interval but also makes it possible
to abort mount operation in case it takes too long.

i=adilger

14 years agob=21574 more pinger fixes
Johann Lombardi [Sat, 16 Jan 2010 14:39:22 +0000 (15:39 +0100)]
b=21574 more pinger fixes

i=oleg
i=andrew

- ptlrpc_update_next_ping(): don't postpone next ping when "soon"
  is set and a ping request is already scheduled before the new
  deadline.
- It is usually fine to extend the deadline for the next ping
  since we are granted that the pinger will wake up before
  this new deadline and update his timer.
  However, the purpose of ptlrpc_pinger_commit_expected() is to
  schedule ping earlier. To support this, i've changed
  ptlrpc_update_next_ping() to wake up the pinger if the new
  ping deadline is before the pinger is supposed to wake up.

14 years agob=21574 PING_INTERVAL_SHORT should not postpone the next ping
Johann Lombardi [Sat, 16 Jan 2010 14:27:17 +0000 (15:27 +0100)]
b=21574 PING_INTERVAL_SHORT should not postpone the next ping

i=oleg
i=andrew

Most our tests run with obd_timeout=20s, so PING_INTERVAL=5s, while
PING_INTERVAL_SHORT=7s. ptlrpc_pinger_commit_expected() was actually
not intended to delay pings.
Although we would prefer to schedule the next ping after a
bit more than 5s (jbd commit time), using 5s instead of 7s
is not a big deal since we will have to only wait for 5
additional seconds in the worst case.

14 years agob=21574 define ptlrpc_pinger_commit_expected for liblustre
Johann Lombardi [Sat, 16 Jan 2010 14:21:28 +0000 (15:21 +0100)]
b=21574 define ptlrpc_pinger_commit_expected for liblustre

i=oleg
i=andrew

14 years agob=21574 schedule ping asap instead of delaying it
Johann Lombardi [Sat, 16 Jan 2010 14:19:03 +0000 (15:19 +0100)]
b=21574 schedule ping asap instead of delaying it

i=oleg
i=andrew

The intent was to schedule a ping as soon as possible to know
sooner rather than later that the transno has been committed.
This is used by the async journal feature to unpin pages
in memory sooner.

14 years agob=20928 skip sanity 202 on ib since unaligned dio not supported by o2iblnd
Johann Lombardi [Sat, 16 Jan 2010 14:16:03 +0000 (15:16 +0100)]
b=20928 skip sanity 202 on ib since unaligned dio not supported by o2iblnd

14 years agob=21564 Print mmp_check_interval in kmmpd
Girish Shilamkar [Sat, 16 Jan 2010 08:32:41 +0000 (14:02 +0530)]
b=21564 Print mmp_check_interval in kmmpd

The patch not only prints mmp_check_interval but also makes it possible
to abort mount operation in case it takes too long.

i=adilger

14 years agoadd missing changelog entries
Johann Lombardi [Sat, 16 Jan 2010 01:04:38 +0000 (02:04 +0100)]
add missing changelog entries

14 years agob=18399 add missing patch to sles11 series
Johann Lombardi [Sat, 16 Jan 2010 00:21:12 +0000 (01:21 +0100)]
b=18399 add missing patch to sles11 series

14 years agob=21097 fix md5sum error in metadata-updates.sh
Ed Giesen [Sat, 16 Jan 2010 00:15:25 +0000 (01:15 +0100)]
b=21097 fix md5sum error in metadata-updates.sh

14 years agobump version
Johann Lombardi [Sat, 16 Jan 2010 00:05:23 +0000 (01:05 +0100)]
bump version

14 years agofix another build issue
Johann Lombardi [Fri, 15 Jan 2010 23:36:37 +0000 (00:36 +0100)]
fix another build issue

14 years agofix build issue
Johann Lombardi [Fri, 15 Jan 2010 16:37:42 +0000 (17:37 +0100)]
fix build issue

14 years agob=21565 filter_last_id() NULL dereference
Dmitry Zogin [Fri, 15 Jan 2010 14:35:48 +0000 (09:35 -0500)]
b=21565 filter_last_id() NULL dereference

 lprocfs_filter_rd_last_id() should check for the fully setup obd device,
 before proceeding further.

 i=johann
 i=andrew.perepechko

14 years agob=11680 don't call LBUG if reading force_lbug
Johann Lombardi [Fri, 15 Jan 2010 11:03:58 +0000 (12:03 +0100)]
b=11680 don't call LBUG if reading force_lbug

Should not happen because the permission is 0200,
but better to check.

14 years agob=18690 enable rehash for hash tables that intended to use it
Johann Lombardi [Fri, 15 Jan 2010 10:32:27 +0000 (11:32 +0100)]
b=18690 enable rehash for hash tables that intended to use it

14 years agob=11680 Add /proc/sys/lnet/force_lbug
Christopher J. Morrone [Fri, 15 Jan 2010 00:27:17 +0000 (01:27 +0100)]
b=11680 Add /proc/sys/lnet/force_lbug

This patch adds a proc entry called force_lbug.

14 years agob=19720 use min_t() to force comparison to unsigned
Brian J. Murrell [Thu, 14 Jan 2010 21:15:02 +0000 (16:15 -0500)]
b=19720 use min_t() to force comparison to unsigned

In older kernels num_online_cpus() is an int, and in newer
kernels it is an unsigned so force the comparison to unsigned
so that it's portable to both new and old kernels.

i=panda
i=whitebear

14 years agob=21411 Improvement for AT.
yangsheng [Thu, 14 Jan 2010 13:55:18 +0000 (21:55 +0800)]
b=21411 Improvement for AT.

i=nathan
i=tappro

14 years agofix tiny nit from previous commit
Johann Lombardi [Thu, 14 Jan 2010 14:20:02 +0000 (15:20 +0100)]
fix tiny nit from previous commit