Whamcloud - gitweb
fs/lustre-release.git
16 years agob=6427
alex [Wed, 1 Jun 2005 10:26:37 +0000 (10:26 +0000)]
b=6427

 - pass OBD_MD_FLID for raw lookup

16 years agoBranch HEAD
adilger [Wed, 1 Jun 2005 07:42:11 +0000 (07:42 +0000)]
Branch HEAD
Build fix for HEAD.
r=oleg (original fix)

16 years ago- one more fix about jumping to wrong label in mds_create_objects()
yury [Wed, 1 Jun 2005 07:12:10 +0000 (07:12 +0000)]
- one more fix about jumping to wrong label in mds_create_objects()

16 years ago- fixed jump to wrong label in mds_create_objects() in error path.
yury [Wed, 1 Jun 2005 07:06:45 +0000 (07:06 +0000)]
- fixed jump to wrong label in mds_create_objects() in error path.

16 years ago- minor fixes in smfs to support raw lookups
alex [Tue, 31 May 2005 21:39:59 +0000 (21:39 +0000)]
- minor fixes in smfs to support raw lookups

16 years agoBranch: b_cray-1_4_2
green [Tue, 31 May 2005 20:25:50 +0000 (20:25 +0000)]
Branch: b_cray-1_4_2
If liblustre was built, add liblustre.{a,so} to the lustre package

16 years agoadd larry's changes that allow using mpirun or yod for running IOR.
nic [Tue, 31 May 2005 20:04:25 +0000 (20:04 +0000)]
add larry's changes that allow using mpirun or yod for running IOR.

16 years agob=5843
alex [Tue, 31 May 2005 19:52:48 +0000 (19:52 +0000)]
b=5843

 - don't postpone orphan unlinks because

16 years agoquiet some verbose gss messages.
ericm [Tue, 31 May 2005 18:28:51 +0000 (18:28 +0000)]
quiet some verbose gss messages.

16 years agoBranch b1_4
adilger [Tue, 31 May 2005 18:15:47 +0000 (18:15 +0000)]
Branch b1_4
On x86_64 the kernel declares __u64 as long long, but userspace uses long.
Fix LP*64 printf format macros to handle this difference to quiet warnings.

16 years agoBrach HEAD
wangdi [Tue, 31 May 2005 14:52:12 +0000 (14:52 +0000)]
Brach HEAD
r:alex
add lookup-raw patches for raw lookup acceptance tests

16 years ago- don't try to show nid for non-connected exports
alex [Tue, 31 May 2005 12:41:17 +0000 (12:41 +0000)]
- don't try to show nid for non-connected exports

16 years ago- add debug message
tappro [Tue, 31 May 2005 11:31:54 +0000 (11:31 +0000)]
- add debug message

16 years agob=6379
alex [Tue, 31 May 2005 11:20:31 +0000 (11:20 +0000)]
b=6379

 - even a simpler fix: just skip removal if orphan is already unlinked
   semaphore on pending dir should be enough

16 years agob=6414,6366
tappro [Tue, 31 May 2005 11:02:50 +0000 (11:02 +0000)]
b=6414,6366
r=alex

it seems that this ASSERT is not needed here really. Such case is allowed

16 years ago* Fix for 6357: handle RapidArray memory registration failures.
eeb [Tue, 31 May 2005 10:24:25 +0000 (10:24 +0000)]
*   Fix for 6357: handle RapidArray memory registration failures.

16 years ago- show nid for stale clients
alex [Tue, 31 May 2005 10:10:25 +0000 (10:10 +0000)]
- show nid for stale clients

16 years ago- export show_task()
alex [Tue, 31 May 2005 07:58:02 +0000 (07:58 +0000)]
- export show_task()

16 years agob=6379
alex [Tue, 31 May 2005 07:16:22 +0000 (07:16 +0000)]
b=6379

 - check for open count only in mds_cleanup_orphans()
 - drop orphan flag to avoid simultaneous unlinking from PENDING/

16 years agob=6403
alex [Tue, 31 May 2005 06:41:53 +0000 (06:41 +0000)]
b=6403

 - revert wrong fix back

16 years agoHEAD
adilger [Mon, 30 May 2005 20:25:33 +0000 (20:25 +0000)]
HEAD
Fix error messages.

16 years ago- minor debug message fix
alex [Mon, 30 May 2005 13:50:05 +0000 (13:50 +0000)]
- minor debug message fix

16 years agob=6403
alex [Mon, 30 May 2005 11:30:07 +0000 (11:30 +0000)]
b=6403

 - remove ugly debug bits
 - dump FL_REPLAY reqs once after connect to avoid flooding

16 years ago- more compilation fixes after discussing them with Andreas
yury [Mon, 30 May 2005 07:42:57 +0000 (07:42 +0000)]
- more compilation fixes after discussing them with Andreas

16 years agob=6413
alex [Sun, 29 May 2005 22:43:10 +0000 (22:43 +0000)]
b=6413

 - wake recovery thread up on any {REQ|LOCK}_REPLAY_DONE

16 years agob=6379
alex [Sun, 29 May 2005 21:03:52 +0000 (21:03 +0000)]
b=6379

 - drop orphan flag to prevent a race between mds_cleanup_orphans() and
   mds_mfd_close() when the both try to unlink the same orphan

16 years ago- many gcc4 compilation fixes (warnings)
yury [Sun, 29 May 2005 16:17:17 +0000 (16:17 +0000)]
- many gcc4 compilation fixes (warnings)

16 years agob=6403
alex [Sun, 29 May 2005 09:34:30 +0000 (09:34 +0000)]
b=6403

 - ll_dentry_iput() clears all l_ast_data and this makes impossible to
   close a file upon openlock cancel

16 years ago- minor debug messages fixes
alex [Sat, 28 May 2005 11:16:28 +0000 (11:16 +0000)]
- minor debug messages fixes

16 years agoBranch b1_4
adilger [Fri, 27 May 2005 22:45:28 +0000 (22:45 +0000)]
Branch b1_4
Re-enable rate-limiting of CDEBUG messages.  The lack of which is causing
grief on large clusters during recovery.
b=4311

16 years agofile llog_reader.c was initially added on branch b_cray_newconfig.
lincent [Fri, 27 May 2005 21:31:04 +0000 (21:31 +0000)]
file llog_reader.c was initially added on branch b_cray_newconfig.

16 years agob=6403
alex [Fri, 27 May 2005 21:02:00 +0000 (21:02 +0000)]
b=6403

 - additional check because a req with FL_REPLAY can be w/o repmsg

16 years ago- landing MOD (MDS Originated destroy) plus few ldlm and MDS fixes.
yury [Fri, 27 May 2005 18:19:41 +0000 (18:19 +0000)]
- landing MOD (MDS Originated destroy) plus few ldlm and MDS fixes.

16 years agob=6403
alex [Fri, 27 May 2005 13:38:44 +0000 (13:38 +0000)]
b=6403

 - additional debug mess(age?) to learn 6K depth open req queue on cnbuild

16 years agob=6366
tappro [Fri, 27 May 2005 12:51:54 +0000 (12:51 +0000)]
b=6366
r=alex
- delete_inode() for backfs inode should be called synchronous with smfs. To do this smfs_clear_inode_info() is called from smfs_delete_inode().

16 years agoBranch HEAD
wangdi [Fri, 27 May 2005 09:50:41 +0000 (09:50 +0000)]
Branch HEAD
add types.h in Makefile.am for some tests build

16 years agob=6403
alex [Thu, 26 May 2005 22:55:27 +0000 (22:55 +0000)]
b=6403

 - the situation looks valid with opencache locks, but it floods consoles
   over. make it silent for a while

16 years agoupdate to krb5 script: check krytab for root.
ericm [Thu, 26 May 2005 19:19:56 +0000 (19:19 +0000)]
update to krb5 script: check krytab for root.

16 years agob=6394
ericm [Thu, 26 May 2005 19:14:59 +0000 (19:14 +0000)]
b=6394
* don't touch base rpcpipe chain.
* invalidate svc gss cache item in case of init error.

16 years ago- minor log format fix
alex [Thu, 26 May 2005 19:02:36 +0000 (19:02 +0000)]
- minor log format fix

16 years agoquite some unnecessary messages and a compiler warning.
ericm [Thu, 26 May 2005 17:13:15 +0000 (17:13 +0000)]
quite some unnecessary messages and a compiler warning.

16 years agob=6398
alex [Thu, 26 May 2005 17:05:54 +0000 (17:05 +0000)]
b=6398

 - additional assert

16 years ago- don't log open replays
alex [Thu, 26 May 2005 16:59:58 +0000 (16:59 +0000)]
- don't log open replays

16 years agob=6398
alex [Thu, 26 May 2005 11:08:39 +0000 (11:08 +0000)]
b=6398

 - target_start_recovery_thread() starts a thread and must wait for
   completion because that thread sets recovering flag. but routine
   kernel_thread() returns pid > 0 in successful case, while the
   target_start_recover_thread() expects 0. so, first routine
   target_start_recovery_thread() returns immediately and chain
   mds_postsetup() -> mds_dt_connect() is called with obd_recovering==0.
   so, mds_postrecov_common() is called and it calls mds_cleanup_orphans()

16 years agob=6368
alex [Wed, 25 May 2005 21:35:00 +0000 (21:35 +0000)]
b=6368

 - use time_after_eq() instead of simple comparing

16 years ago* add a bit in ptlrpc_request to indicate ptlrpcs error, gss recovery
ericm [Wed, 25 May 2005 18:08:29 +0000 (18:08 +0000)]
* add a bit in ptlrpc_request to indicate ptlrpcs error, gss recovery
  fixes, need new nfsutils patches version 8.
* more precise error notify between kernel and lgssd.
* minor gss api param fix. and various minor gss fixes.

16 years agofile upcall_cache.c was initially added on branch b1_4_bug3389.
adilger [Wed, 25 May 2005 15:33:20 +0000 (15:33 +0000)]
file upcall_cache.c was initially added on branch b1_4_bug3389.

16 years agoBranch: HEAD
wangdi [Wed, 25 May 2005 07:54:27 +0000 (07:54 +0000)]
Branch: HEAD
remove unsupported target on HEAD

16 years agoBranch: HEAD
wangdi [Wed, 25 May 2005 07:52:32 +0000 (07:52 +0000)]
Branch: HEAD
remove SMFS options in autoMakefile.am because we should build smfs by default
on HEAD

16 years agoBranch HEAD
wangdi [Tue, 24 May 2005 10:11:43 +0000 (10:11 +0000)]
Branch HEAD
1) remove unsupported target in lustre-core.m4 according to boc's suggest
2) add some args on 2.6-fc3 target

16 years ago- enable tcp keepalive back
alex [Tue, 24 May 2005 10:01:06 +0000 (10:01 +0000)]
- enable tcp keepalive back

16 years ago- disable tcp keepalive for debugging purpose
alex [Tue, 24 May 2005 09:54:48 +0000 (09:54 +0000)]
- disable tcp keepalive for debugging purpose

16 years agorevert unapproved change (lustre-build HEAD is shared by all branches; the proper...
jacob [Tue, 24 May 2005 09:52:53 +0000 (09:52 +0000)]
revert unapproved change (lustre-build HEAD is shared by all branches; the proper place for this change is in HEAD's .target.in files)

16 years agoBrach: HEAD
wangdi [Tue, 24 May 2005 09:08:07 +0000 (09:08 +0000)]
Brach: HEAD
add kernel-2.6.10-2.6-fc3-i686-smp.config for lbuild on HEAD

16 years agoBranch HEAD
wangdi [Tue, 24 May 2005 08:56:07 +0000 (08:56 +0000)]
Branch HEAD
change LINUX26 from 0 to 1 for HEAD in lbuild

16 years agoBranch: HEAD
wangdi [Tue, 24 May 2005 07:47:16 +0000 (07:47 +0000)]
Branch: HEAD
some fix for fc3 lbuild on HEAD

16 years ago- grrr, they are unsigned ...
alex [Mon, 23 May 2005 15:38:39 +0000 (15:38 +0000)]
- grrr, they are unsigned ...

16 years ago- changes for KML setting and variables in accordance with smfs plugin API
tappro [Mon, 23 May 2005 11:29:34 +0000 (11:29 +0000)]
- changes for KML setting and variables in accordance with smfs plugin API

16 years ago- KML fixes and updates due to SMFS plugin API.
tappro [Mon, 23 May 2005 11:25:57 +0000 (11:25 +0000)]
- KML fixes and updates due to SMFS plugin API.

16 years ago- brown paper bug (negative value is good)
alex [Mon, 23 May 2005 10:27:32 +0000 (10:27 +0000)]
- brown paper bug (negative value is good)

16 years ago- take HZ into account
alex [Sun, 22 May 2005 12:10:37 +0000 (12:10 +0000)]
- take HZ into account

16 years ago- fixed CDEBUG message
yury [Sun, 22 May 2005 09:44:39 +0000 (09:44 +0000)]
- fixed CDEBUG message

16 years ago- on cnbuild ping time is set to some strange value, the patch is intended
alex [Sat, 21 May 2005 19:00:19 +0000 (19:00 +0000)]
- on cnbuild ping time is set to some strange value, the patch is intended
  to catch and fix the problem

16 years ago- handling out possible signals to @current while waiting for mount.
yury [Sat, 21 May 2005 13:54:26 +0000 (13:54 +0000)]
- handling out possible signals to @current while waiting for mount.
- checking for possible dentry mounting even if wait function returned failure - this helps to avpid post mounts.
- many changes in sanity-gns.sh

16 years agoBranch HEAD
wangdi [Sat, 21 May 2005 12:51:11 +0000 (12:51 +0000)]
Branch HEAD
In smfs_helper, replace list_for_each_entry_safe with list_for_each_entry,
because we may unplugin the plg list

16 years ago- fixes in sanity-gns.sh
yury [Sat, 21 May 2005 11:14:10 +0000 (11:14 +0000)]
- fixes in sanity-gns.sh

16 years ago- bug fixed: wrong lock type was passed to ldlm_lock_decref()
alex [Sat, 21 May 2005 08:07:24 +0000 (08:07 +0000)]
- bug fixed: wrong lock type was passed to ldlm_lock_decref()

16 years agor=jacob
phil [Fri, 20 May 2005 23:03:16 +0000 (23:03 +0000)]
r=jacob

check out libsysio HEAD for all lustre builds

16 years agob=3359
phil [Fri, 20 May 2005 22:55:11 +0000 (22:55 +0000)]
b=3359
r=lee@sandia.gov

libsysio fixes which always apply the umask before calling liblustre.
With these changes, liblustre can stop worrying about the umask entirely.

16 years agoAs long as we have lustre hack flags, let's at least compile with -O0,
phil [Fri, 20 May 2005 22:53:54 +0000 (22:53 +0000)]
As long as we have lustre hack flags, let's at least compile with -O0,
which makes debugging at least theoretically possible.

16 years agoadd -DLUSTRE_UTILS=1 to the utils build
phil [Fri, 20 May 2005 22:44:08 +0000 (22:44 +0000)]
add -DLUSTRE_UTILS=1 to the utils build

16 years ago- Fixes a catstrophic bug introduced in the merge, wherein liblustre OBD_ALLOC
phil [Fri, 20 May 2005 22:35:35 +0000 (22:35 +0000)]
- Fixes a catstrophic bug introduced in the merge, wherein liblustre OBD_ALLOC
  wasn't zeroing memory
- eliminate CDEBUG helper dupes; everything references CDEBUG now
- I wanted to keep obd_memory, portals_debug, et al in liblustre, but I can't
  have those bits compiled into userspace utils.  Hence the LUSTRE_UTILS
  preprocessor definition to distinguish them.

16 years agomountain need Qlogic 2300 driver as well.
brian [Fri, 20 May 2005 19:03:42 +0000 (19:03 +0000)]
mountain need Qlogic 2300 driver as well.

16 years agoQuadrics' elan config variables from mountain.
brian [Fri, 20 May 2005 16:43:32 +0000 (16:43 +0000)]
Quadrics' elan config variables from mountain.

16 years agoBranch: HEAD
wangdi [Fri, 20 May 2005 15:52:00 +0000 (15:52 +0000)]
Branch: HEAD
some fixes about write back cache.
1)cleanup smfs_init_inode_info
2)miss spin_unlock in class_find_client_obd
3)remove PLG_ALL flags of OBJECTS and LOGS dir on MDS
4)some minor fix in cmobd.sh for WB cache

16 years agob=6360
alex [Fri, 20 May 2005 11:35:45 +0000 (11:35 +0000)]
b=6360

 - couple asserts and debug info

16 years agoadd some real userspace definitions for CDEBUG et al
phil [Thu, 19 May 2005 23:22:02 +0000 (23:22 +0000)]
add some real userspace definitions for CDEBUG et al

16 years agob=6316
alex [Thu, 19 May 2005 21:18:22 +0000 (21:18 +0000)]
b=6316

 - all clients report their minimal transno to be replayed
 - server waits for all clients to connect in order to collect
   minimal transno's from clients and replay all requests
   in right order

16 years agob=6358
alex [Thu, 19 May 2005 21:14:53 +0000 (21:14 +0000)]
b=6358

 - simplest mkdir placement to test the idea on mountain

16 years agob=6336
jacob [Thu, 19 May 2005 20:35:56 +0000 (20:35 +0000)]
b=6336

only disable qswnal when building against qsnet inside the kernel.

16 years agob=6336
jacob [Thu, 19 May 2005 20:33:32 +0000 (20:33 +0000)]
b=6336

disable qswnal if neither CONFIG_QSNET nor CONFIG_QSNET_MODULES is
defined (ie on ppc64).

16 years agob=6265
jacob [Thu, 19 May 2005 19:58:38 +0000 (19:58 +0000)]
b=6265

remove .pyc files when uninstalling

16 years agoBranch: HEAD
wangdi [Thu, 19 May 2005 16:45:33 +0000 (16:45 +0000)]
Branch: HEAD
1.port new lustre_lcfg structure from b1_4 to HEAD
2.some fix on test scripts for WB cache

16 years ago* changed router lookups to always return the local destination
eeb [Thu, 19 May 2005 14:04:32 +0000 (14:04 +0000)]
*   changed router lookups to always return the local destination
    socknal + qswnal changes to match
    simplified "will this be forwarded" predicate in ptl_send()

*   fixed bug in multi-instance socknal to pass the correct NI to the tx
    teardown function to allow for the case where the tx hasn't even had a conn
    assigned yet.

*   fixed routerstat stats formatting

*   removed osc::check_elan_limit() and replaced with max_unfragmented_pages()
    which actually works.  NB it depend on the assumption that file and page
    offsets are aligned!

*   fixed lconf to stay away from portals, except to 'modprobe' for it first in
    'load_module()', and to rmmod all the NALs + portals when ptlrpc gets
    unloaded.

16 years ago- wrong command lctl dk /r/tmp/debug in test 55
alex [Thu, 19 May 2005 11:35:16 +0000 (11:35 +0000)]
- wrong command lctl dk /r/tmp/debug in test 55

16 years agob=6332
ericm [Wed, 18 May 2005 23:26:15 +0000 (23:26 +0000)]
b=6332
create dir across MDS's, subdir need default ACL of its parent to determine
its own default ACL and access ACL. API obd_create() is added 2 params
to hold acl data.

16 years ago- minor fixes in the script
alex [Wed, 18 May 2005 21:29:12 +0000 (21:29 +0000)]
- minor fixes in the script

16 years ago- ptlrpc_ping_interpret is needless:
alex [Wed, 18 May 2005 20:44:06 +0000 (20:44 +0000)]
- ptlrpc_ping_interpret is needless:
  general timeout infrastructure is enough for cascading timeouts

16 years ago- ptlrpc_request_handle_notconn() shouldn't reconnect if recovery on
alex [Wed, 18 May 2005 20:41:05 +0000 (20:41 +0000)]
- ptlrpc_request_handle_notconn() shouldn't reconnect if recovery on
  import is disabled adminitratively

16 years agob=6344
tappro [Wed, 18 May 2005 11:52:27 +0000 (11:52 +0000)]
b=6344
r=alex

- fix bug with flag DCACHE_CROSS_REF that can be unsetted sometimes.

16 years agoBranch: HEAD
wangdi [Wed, 18 May 2005 09:02:56 +0000 (09:02 +0000)]
Branch: HEAD
some fix about x86_64 lkcd patch

16 years agoBranch HEAD
wangdi [Tue, 17 May 2005 10:26:35 +0000 (10:26 +0000)]
Branch HEAD
add some config options for CGG

16 years agob=6274
tappro [Tue, 17 May 2005 09:02:45 +0000 (09:02 +0000)]
b=6274
r=alex
-fix nasty bug with using function in stack

16 years agob=5881, 6274
tappro [Mon, 16 May 2005 18:33:01 +0000 (18:33 +0000)]
b=5881, 6274
r=alex

-fix bug with setting i_ino (thanks alex)
-fix issue with non-freed dentry after error
-cleanup smfs_do_lookup() code

16 years agoSmall fix for OS X uninstall script
liangzhen [Mon, 16 May 2005 18:33:00 +0000 (18:33 +0000)]
Small fix for OS X uninstall script

16 years agoFiles needed by OS X package
liangzhen [Mon, 16 May 2005 18:07:50 +0000 (18:07 +0000)]
Files needed by OS X package

16 years ago- many changes in sanity-gns.sh. Enabled last test (3b).
yury [Mon, 16 May 2005 16:22:41 +0000 (16:22 +0000)]
- many changes in sanity-gns.sh. Enabled last test (3b).

16 years ago- fixed umounting all GNS entrie son lustre umount (added umount_lustre patch)
yury [Mon, 16 May 2005 08:04:51 +0000 (08:04 +0000)]
- fixed umounting all GNS entrie son lustre umount (added umount_lustre patch)
- some typos fixed.

16 years ago- fixes and improvements in GNS:
yury [Mon, 16 May 2005 06:48:21 +0000 (06:48 +0000)]
- fixes and improvements in GNS:
  - making it work in some stat cases (stat target is behind mount object)
  - making it work in sys_mount() control path (is mount point is behind mount object)

  - to avoid deadlock is mount case is_subdir() is used for detecting is current dentry is
    parent of dentry being in progress of mounting.

  - do not mount on revalidate if there is not valid @nd applied.
  - using nd->flags if possible for both lookup and revalidate cases.
  - added debug info printing in mounting cases.
  - totaly removed -ERESTARTSYS approach stuff.

  - fixed bug with many waiters on ->ll_gns_mount_finished - only first of them will find GNS
    in IDLE state and will start mounting and the rest will assert on wrong GNS state (MOUNTING)

  - reinit ->ll_gns_mount_finished after all waiters are awakaned, this fixes the bug when
    after first using this completion it does not functions, because complete_all() has
    slighly diff. semanthics then complete() and namely does not just decrease ->done but
    sets it to MAX_INT/2.

  - fully rewritten sanity-gns.sh to make it more flexible and easy to add new sanity tests.

16 years agoBranch: HEAD
wangdi [Sun, 15 May 2005 17:07:11 +0000 (17:07 +0000)]
Branch: HEAD
Should return 0, if the return value of mds_postrecov_common > 0