Whamcloud - gitweb
jacob [Tue, 3 May 2005 22:16:07 +0000 (22:16 +0000)]
b=3031
- this is what i should have done originally
nathan [Tue, 3 May 2005 22:13:12 +0000 (22:13 +0000)]
Branch b1_4
b=6223
Fix this related one too.
nathan [Tue, 3 May 2005 21:54:49 +0000 (21:54 +0000)]
b=6223
This should fix the crash (need to use vfree, not kfree). But it won't
allow lfs getstripe to work for 420 OSTs -- OBD_MAX_IOCTL_BUFFER of 8k allows a
max of around 180 OBDs.
jacob [Tue, 3 May 2005 21:46:41 +0000 (21:46 +0000)]
b=3031
- remove .flags files on make clean
jacob [Tue, 3 May 2005 21:25:15 +0000 (21:25 +0000)]
b=3418
- add dependencies on libs
green [Tue, 3 May 2005 21:02:00 +0000 (21:02 +0000)]
vanilla-2.4.29 support.
adilger [Fri, 29 Apr 2005 22:28:50 +0000 (22:28 +0000)]
Branch: b1_4
Add a similar fix to the mds_connect() to handle NULL connect data, in
case we ever use echo_client against the MDS.
Print out the request for bad connect requests so we can get the source NID.
This was happening at NCSA for a long time and no way to find which client
was trying to conenct.
nic [Fri, 29 Apr 2005 22:00:13 +0000 (22:00 +0000)]
r=phil
only set data elements with a non-NULL data. this fixes echo_client connects
adilger [Fri, 29 Apr 2005 20:24:51 +0000 (20:24 +0000)]
Branch: b1_4
- don't reference lr_lvb_data until after we hold lr_lvb_sem
b=6170
adilger [Thu, 28 Apr 2005 23:25:40 +0000 (23:25 +0000)]
Branch: b1_4
Fix compile warnings on 64-bit.
nathan [Thu, 28 Apr 2005 22:02:49 +0000 (22:02 +0000)]
Branch b1_4
b=6151
utilities all want /dev/obd
adilger [Thu, 28 Apr 2005 20:00:21 +0000 (20:00 +0000)]
Branch: b1_4
Placeholders for constants already defined in b_cray.
adilger [Thu, 28 Apr 2005 19:50:10 +0000 (19:50 +0000)]
Branch: b1_4
On Altix PAGE_SIZE is #defined as getpagesize() in userspace and CPP
can't deal with that. Instead, assume we are limited by the RPC size
and not the number of pages (which is currently true for any arch)
and don't use PAGE_SIZE in any CPP conditionals.
adilger [Thu, 28 Apr 2005 18:36:04 +0000 (18:36 +0000)]
Branch: b1_4
Fix quota test valid compile warnings.
adilger [Thu, 28 Apr 2005 00:22:59 +0000 (00:22 +0000)]
Branch: b1_4
Free lsm if there was an error allocating lsr.
adilger [Thu, 28 Apr 2005 00:15:20 +0000 (00:15 +0000)]
Branch: b1_4
Define the quota constants for
adilger [Thu, 28 Apr 2005 00:03:30 +0000 (00:03 +0000)]
Branch: b1_4
Don't leave lock on resource list in ldlm_handle_enqueue() error path, as
that will LBUG when lock is destroyed. Regression test for same.
b=6149, b=6184
r=phil (original patch)
adilger [Wed, 27 Apr 2005 21:19:05 +0000 (21:19 +0000)]
Branch: b1_4
Restore lov.sh testing to 5 OSTs.
b=4554
green [Wed, 27 Apr 2005 20:01:28 +0000 (20:01 +0000)]
Changelog update for commits on 6203, 5007
adilger [Wed, 27 Apr 2005 19:04:13 +0000 (19:04 +0000)]
Branch: b1_4
- reserve enough journal credits in fsfilt_start_log for setattr
b=4554
adilger [Wed, 27 Apr 2005 18:11:06 +0000 (18:11 +0000)]
Branch: b1_4
Improve lconf performance for large configurations by adding a cache
for lookups instead of searching the DB for each entry.
This sped up a 420-OST config --write-conf from 10h to 6 minutes...
b=6204
r=fergal (original patch)
green [Wed, 27 Apr 2005 17:35:37 +0000 (17:35 +0000)]
b=6203,5654
r=adilger
Release i_sem and i_alloc_sem prior to getting DLM lock in setattr_raw to
avoid lock inversion issues.
green [Wed, 27 Apr 2005 08:22:11 +0000 (08:22 +0000)]
b=5907
r=adilger
Do not allow to remove mountpoints on lustre
jacob [Wed, 27 Apr 2005 04:27:28 +0000 (04:27 +0000)]
bump version
jacob [Tue, 26 Apr 2005 23:53:29 +0000 (23:53 +0000)]
b=5782
- added --disable-server and --disable-client configure options (5782)
jacob [Tue, 26 Apr 2005 23:47:20 +0000 (23:47 +0000)]
be more quiet
jacob [Tue, 26 Apr 2005 22:03:51 +0000 (22:03 +0000)]
b=6205
r=adilger
- move the lustre kernel modules to their own package
- lbuild now does 'make rpms' to build lustre modules, more or less
- pass --with-linux to lbuild to build only lustre rpms against an
existing kernel source tree
- drop the lustre-ldap rpm
green [Tue, 26 Apr 2005 20:49:30 +0000 (20:49 +0000)]
b=5907
r=adilger
Do not unhash mountpoint dentries and silence error message
adilger [Tue, 26 Apr 2005 17:38:08 +0000 (17:38 +0000)]
Branch: b1_4
Quiet compile warning about unused lustre_delete_dquot().
green [Mon, 25 Apr 2005 16:30:47 +0000 (16:30 +0000)]
Update changelog with 6196 bugfix info
green [Mon, 25 Apr 2005 08:40:36 +0000 (08:40 +0000)]
b=6196
r=adilger
For inodes with no ost objects call vmtruncate from inode_setattr() to update
inode->i_size, becasue size is only stored on MDS.
adilger [Sat, 23 Apr 2005 08:37:55 +0000 (08:37 +0000)]
Branch: b1_4
Put test58 files into a subdirectory.
adilger [Sat, 23 Apr 2005 08:08:06 +0000 (08:08 +0000)]
Branch: b1_4
Reduce number of files created because small test setups don't have so many
files and it doesn't really improve the test. Also, checkstat will fail
if there are too many files on the command line.
adilger [Sat, 23 Apr 2005 00:15:10 +0000 (00:15 +0000)]
Branch: b1_4
Proper fix for hokey error message quieting hack, per Nathan's feedback.
adilger [Fri, 22 Apr 2005 19:53:06 +0000 (19:53 +0000)]
Branch: b1_4
Fix thinko.
adilger [Fri, 22 Apr 2005 17:54:56 +0000 (17:54 +0000)]
Branch: b1_4
Print file/directory for failed ioctl.
adilger [Fri, 22 Apr 2005 17:54:23 +0000 (17:54 +0000)]
Branch: b1_4
Return a different error code than fail_abort so we know where this test failed.
adilger [Fri, 22 Apr 2005 17:46:54 +0000 (17:46 +0000)]
Branch: b1_4
Quiet noisy llog/filter_setattr messages caused by quota llog replay.
adilger [Fri, 22 Apr 2005 17:39:58 +0000 (17:39 +0000)]
Branch: b1_4
Improved error messages for debugging of recent problems.
Use a constant for the minimum acceptable LOV stripe size.
Return limit of stripe_size * num_stripes for default_striped files on 32-bit.
adilger [Fri, 22 Apr 2005 17:12:46 +0000 (17:12 +0000)]
Branch: b1_4
Initialize the blocksize for stat() for non-regular files.
b=6062
adilger [Fri, 22 Apr 2005 16:39:56 +0000 (16:39 +0000)]
Branch: b1_4
Create larger MDS by default in order to avoid OOS on the MDS because of
quota setattr records.
b=5893
nathan [Fri, 22 Apr 2005 01:14:26 +0000 (01:14 +0000)]
Branch b1_4
b=5921
r=adilger
Land b1_4_5921 ping evictor
jacob [Tue, 19 Apr 2005 23:55:59 +0000 (23:55 +0000)]
add support for sles 8:
- new target for sles 8 kernel
- post(un) scripts for kernel rpms
- fixup quota build on sles 8 and sles 9
- update patches for a more recent sles 8 kernel
phil [Tue, 19 Apr 2005 19:17:19 +0000 (19:17 +0000)]
two uml patches I forgot to commit
phil [Fri, 15 Apr 2005 23:05:56 +0000 (23:05 +0000)]
Add 2.6-fc3.series, based on 2.6-rhel4.series. It includes a (new) UML
patch, and a modified iopen-misc and nfs-cifs-intent patch. There's also
a tiny ext3-patch-fuzz-fixup, which allows me to use the rhel4 ldiskfs series.
adilger [Thu, 14 Apr 2005 10:50:29 +0000 (10:50 +0000)]
Branch: b1_4
Handle the new LOV config when running with a config in LDAP.
b=6064
r=fergal (original patch)
adilger [Thu, 14 Apr 2005 05:47:12 +0000 (05:47 +0000)]
Branch: b1_4
Update ChangeLog with a few missing entries.
adilger [Wed, 13 Apr 2005 21:14:58 +0000 (21:14 +0000)]
Branch: b1_4
If there is an IO error the bio completion routine will be called multiple
times. We don't want to free the dreq until the last completion.
b=6046
r=alex (original patch)
adilger [Wed, 13 Apr 2005 17:01:19 +0000 (17:01 +0000)]
Branch: b1_4
Update build version - we already sent out 1.4.1.10...
adilger [Wed, 13 Apr 2005 02:31:31 +0000 (02:31 +0000)]
Branch: b1_4
Don't hold client locks being cancelled on a temporary worklist from l_lru
when doing "echo clear > /proc/fs/lustre/ldlm/namespaces/*/lru_size".
Instead we use l_pending_chain, which is only used on the server.
b=5666
r=phil
adilger [Wed, 13 Apr 2005 00:24:10 +0000 (00:24 +0000)]
Branch: b1_4
b=6133
adilger [Tue, 12 Apr 2005 09:43:52 +0000 (09:43 +0000)]
Branch: b1_4
Have local, lov, uml, mount2fs.sh configs all generate the config.xml
file based on their own filename instead of hard coding it. This makes
it easier to copy the config to another name, edit it, and then run
"NAME=foo sh llmount.sh" and similar tests.
Allow mount2fs.sh config to have some parameters overridden, and add
"--group fs{1,2}" to the config to allo testing startup/shutdown of
each filesystem independently.
adilger [Tue, 12 Apr 2005 09:36:16 +0000 (09:36 +0000)]
Branch: b1_4
Grab some extra debugging for bug 5195 in case it is a real bug and
not just write flush race because we can't stop kupdated.
b=5195
adilger [Tue, 12 Apr 2005 09:31:49 +0000 (09:31 +0000)]
Branch: b1_4
Fix up offset printing for fsx.
adilger [Tue, 12 Apr 2005 09:20:34 +0000 (09:20 +0000)]
Branch: b1_4
Don't override START and CLEAN, in case it is desirable to run a full
unmount-after-every-test sanity.sh (which is no longer the default).
adilger [Tue, 12 Apr 2005 09:19:27 +0000 (09:19 +0000)]
Branch: b1_4
Fix somewhat broken "lustre status" output, which previously always output
"running" or "recovery" then "stopped". Now it checks for servers and
clients in recovery and also if modules are loaded (which can be a sign
of problems). It will check status regardless of whether there is a
config.xml file as not all sites are configured that way.
adilger [Tue, 12 Apr 2005 09:14:03 +0000 (09:14 +0000)]
Branch: b1_4
Log MDS disconnection to obdfilter kernel log.
Don't set request portals until after request alloc is verified.
b=5515
adilger [Tue, 12 Apr 2005 09:10:50 +0000 (09:10 +0000)]
Branch: b1_4
Fix copy+paste bug from "upcall removal" landing. b=3262 r=nathan, green
Add debugging for missing reconnect RPC. b=5933
adilger [Tue, 12 Apr 2005 08:59:55 +0000 (08:59 +0000)]
Branch: b1_4
Minor corrections.
adilger [Tue, 12 Apr 2005 08:59:24 +0000 (08:59 +0000)]
Branch: b1_4
Try to unload ksocknal/kqswnal/kptlrouter for environments that auto-load
these modules because of modules.conf (e.g. mdev).
green [Tue, 12 Apr 2005 08:30:00 +0000 (08:30 +0000)]
b=6141
Sigh, more DQ_* conversion to DQ_*_B and bitops.
nathan [Mon, 11 Apr 2005 17:25:19 +0000 (17:25 +0000)]
Branch b1_4
b=4834
bdev, not kdev on 2.6
adilger [Mon, 11 Apr 2005 16:58:16 +0000 (16:58 +0000)]
Branch: b1_4
list which modules are still loaded.
b=6133
nathan [Sat, 9 Apr 2005 16:52:18 +0000 (16:52 +0000)]
Branch b1_4
b=4834
r=adilger
llog_finish before mntput
nathan [Fri, 8 Apr 2005 18:18:52 +0000 (18:18 +0000)]
Branch b1_4
b=4834
unused in series files
nathan [Fri, 8 Apr 2005 18:16:35 +0000 (18:16 +0000)]
Branch b1_4
b=4834
r=adilger
(from adilger's b_cray fix)
Change the dev_read_only patch to do set the device r/w only at
block device shutdown, and not just when Lustre does a mntput
(which may still be racy with threads accessing the superblock
before the are stopped).
nathan [Fri, 8 Apr 2005 17:01:38 +0000 (17:01 +0000)]
Branch b1_4
Forceful test cleanup (from b_cray)
On tests that evict clients, make sure clients are reconnected at the end
of the test.
adilger [Fri, 8 Apr 2005 11:35:47 +0000 (11:35 +0000)]
Branch: b1_4
Don't call bdevname() with dcache_lock held. Instead print MDC name.
b=5259
adilger [Fri, 8 Apr 2005 11:01:41 +0000 (11:01 +0000)]
Branch: b1_4
b=4982
green [Fri, 8 Apr 2005 08:04:52 +0000 (08:04 +0000)]
b=4982
r=adilger
Stick block_invalidatepage back in, as Andreas says trusting in jbd's
invalidate page logic can be dangerous and needs more investigations.
cvs2svn [Fri, 8 Apr 2005 07:46:18 +0000 (07:46 +0000)]
This commit was manufactured by cvs2svn to create branch 'b1_4'.
wangdi [Fri, 8 Apr 2005 07:46:17 +0000 (07:46 +0000)]
Branch: HEAD
1)minor fix and cleanup dev_rd_only kernel patch on HEAD
2)add sec support on kernel config
3)some cleanup
green [Fri, 8 Apr 2005 07:23:09 +0000 (07:23 +0000)]
r=adilger
Small optimisation. Do not drop partially truncated page from cache on client,
so that when later somebody decides to write there, we won't need to fetch it
from OST again (synchronously).
green [Fri, 8 Apr 2005 07:21:54 +0000 (07:21 +0000)]
b=4982
r=adilger
Flush pagecache on ost before doing directio writes, or else we might end up
with stale data being written to disk later.
phil [Fri, 8 Apr 2005 04:24:51 +0000 (04:24 +0000)]
It is so unbelievably embarrassing when I run llobdstat during the training,
it prints Mb/s, and everyone laughs at how CFS can't tell the difference
between megabits and megabytes.
nathan [Thu, 7 Apr 2005 19:30:29 +0000 (19:30 +0000)]
Branch b1_4
b=6080
try to isolate test 24's fallout
alex [Thu, 7 Apr 2005 16:40:42 +0000 (16:40 +0000)]
- disable CDEBUG() ratelimit for a while
wangdi [Thu, 7 Apr 2005 15:41:12 +0000 (15:41 +0000)]
Branch: HEAD
add signal bug fix to uml patch
yury [Thu, 7 Apr 2005 10:54:01 +0000 (10:54 +0000)]
- cleanups in sanity-gns.sh
yury [Thu, 7 Apr 2005 10:49:37 +0000 (10:49 +0000)]
- removed hardcoded checking for ".mntinfo" on MDS to see if we are trying to open GNS mount object. By now only suid bit on directory is checked. This should not break anything, as it is not used for any purpose.
- cleanups, moving not needed to be widely known llite functions to llite_internal.h
- added test_4a to sanity-gns.sh. It checks how GNS mounting works with 20 concurrent threads aiming to mount the same object.
wangdi [Thu, 7 Apr 2005 10:19:37 +0000 (10:19 +0000)]
Branch: HEAD
minor fix in lkcd patch
wangdi [Thu, 7 Apr 2005 09:13:45 +0000 (09:13 +0000)]
Branch: HEAD
some patches fix for lustre in FC3
adilger [Wed, 6 Apr 2005 23:43:55 +0000 (23:43 +0000)]
Branch: b1_4
- update gmnal to use PTL_MTU, fix module refcounting, for HP
b=5786
r=mdoyle (original patch)
adilger [Wed, 6 Apr 2005 22:25:12 +0000 (22:25 +0000)]
Branch: b1_4
Quiet compile warnings on ia64.
adilger [Wed, 6 Apr 2005 21:56:14 +0000 (21:56 +0000)]
Branch: b1_4
Conversely, loff_t is always %llu.
adilger [Wed, 6 Apr 2005 21:52:29 +0000 (21:52 +0000)]
Branch: b1_4
Use LPU64 instead of %llu to print __u64 values, as ia64 uses "long"
for 64-bit numbers.
adilger [Wed, 6 Apr 2005 18:51:23 +0000 (18:51 +0000)]
Branch: b1_4
Revert "last little bit of cleanup" for lli_size_sem locking. Getting this
lock (a semaphore which can sleep) in ll_nopage and/or ll_ap_refresh_count
causes the client to lock up (symptoms include kupdated stuck, no free mem,
client unresponsive to network = elan timeouts).
Still need to figure proper locking solution here. A spinlock is good for
all places except for ll_setattr_raw->vmtruncate path.
b=5654 (added under), b=6057, b=6059, b=6077 (possible fix)
wangdi [Wed, 6 Apr 2005 16:44:03 +0000 (16:44 +0000)]
Branch HEAD
update dev_read_only patch from b1_4, thanks for nathan's reminding
wangdi [Wed, 6 Apr 2005 16:25:49 +0000 (16:25 +0000)]
Branch: HEAD
minor fix in Makefile for uml patch
wangdi [Wed, 6 Apr 2005 16:18:43 +0000 (16:18 +0000)]
Branch: HEAD
add uml and vmware kernel config for FC3
wangdi [Wed, 6 Apr 2005 16:10:05 +0000 (16:10 +0000)]
Branch HEAD
add lustre FC3 series
wangdi [Wed, 6 Apr 2005 16:05:44 +0000 (16:05 +0000)]
Branch: HEAD
add FC3 kernel patches
nic [Wed, 6 Apr 2005 15:42:55 +0000 (15:42 +0000)]
add the $TESTNAME to all errors to make auto-vetting easier
adilger [Wed, 6 Apr 2005 12:08:39 +0000 (12:08 +0000)]
Branch: b1_4
Add support for mountfsoptions and clientoptions to LDAP.
b=5873
r=fergal (original patch)
adilger [Wed, 6 Apr 2005 11:57:42 +0000 (11:57 +0000)]
Branch: b_cray
Allow utils to be built with cray portals.
b=5918
r=bogl (original change)
eeb [Wed, 6 Apr 2005 11:40:59 +0000 (11:40 +0000)]
* Updated gmnalnid.c with patch from HP
eeb [Wed, 6 Apr 2005 10:49:52 +0000 (10:49 +0000)]
* changed vibnal compilation checks to allow gcc < 3.0.0 and
allow HIGHMEM4G on x86 (bug 5852)
nathan [Wed, 6 Apr 2005 01:55:27 +0000 (01:55 +0000)]
Branch b1_4
b=5949
The end of mds_connect was a mess. In an error case, it dropped too many
export refs. I think this might be the cause of 5949, 6030, 6058
Same commit, different rev. I jumped the gun on undoing this.
nathan [Wed, 6 Apr 2005 01:43:19 +0000 (01:43 +0000)]
Branch b1_4
b=5949
quickly undo broken stuff before anyone realizes it broke.
nathan [Wed, 6 Apr 2005 00:13:30 +0000 (00:13 +0000)]
Branch b1_4
b=5949
The end of mds_connect was a mess. In an error case, it dropped too many
export refs. I think this might be the cause of 5949, 6030, 6058