Whamcloud - gitweb
fs/lustre-release.git
13 years agofile hostlist.c was initially added on branch b1_6.
yangsheng [Fri, 21 Nov 2008 15:27:36 +0000 (15:27 +0000)]
file hostlist.c was initially added on branch b1_6.

13 years agofile hash.h was initially added on branch b1_6.
yangsheng [Fri, 21 Nov 2008 15:27:35 +0000 (15:27 +0000)]
file hash.h was initially added on branch b1_6.

13 years agofile hash.c was initially added on branch b1_6.
yangsheng [Fri, 21 Nov 2008 15:27:34 +0000 (15:27 +0000)]
file hash.c was initially added on branch b1_6.

13 years agofile lshowmount.c was initially added on branch b1_6.
yangsheng [Fri, 21 Nov 2008 15:27:33 +0000 (15:27 +0000)]
file lshowmount.c was initially added on branch b1_6.

13 years agofile lshowmount.8 was initially added on branch b1_6.
yangsheng [Fri, 21 Nov 2008 15:27:31 +0000 (15:27 +0000)]
file lshowmount.8 was initially added on branch b1_6.

13 years agoBranch HEAD
vs [Thu, 20 Nov 2008 23:24:31 +0000 (23:24 +0000)]
Branch HEAD
b=17132
i=adilger

Use raid5/6 rhel5 improvements

13 years agofix build on sles9.
shadow [Thu, 20 Nov 2008 21:21:23 +0000 (21:21 +0000)]
fix build on sles9.

Branch HEAD

13 years agofix 'make rpm'
shadow [Thu, 20 Nov 2008 15:19:26 +0000 (15:19 +0000)]
fix 'make rpm'

13 years agofix typo.
shadow [Thu, 20 Nov 2008 12:51:23 +0000 (12:51 +0000)]
fix typo.

Branch HEAD

13 years ago*** empty log message ***
shadow [Thu, 20 Nov 2008 09:25:26 +0000 (09:25 +0000)]
*** empty log message ***

13 years agofirst part of 2.6.26 support (lnet/libcfs part)
shadow [Thu, 20 Nov 2008 09:24:35 +0000 (09:24 +0000)]
first part of 2.6.26 support (lnet/libcfs part)

Branch HEAD
b=14250
i=maxim
i=deen

13 years agoBranch HEAD
liangzhen [Thu, 20 Nov 2008 06:54:16 +0000 (06:54 +0000)]
Branch HEAD

o2iblnd asserts when ib_poll_cq() returns an error

b=17533
i=maxim
i=Christopher Morrone

13 years agoBranch HEAD
fanyong [Thu, 20 Nov 2008 05:51:00 +0000 (05:51 +0000)]
Branch HEAD
b=16947
i=h.huang
i=yury.umanets

Hold lli_lock when access lli_sai to prevent NULL pointer.

13 years agonot send LOV EA under replay, we can't know about they size at this
shadow [Thu, 20 Nov 2008 05:07:25 +0000 (05:07 +0000)]
not send LOV EA under replay, we can't know about they size at this
time. Don't allow client connect to mds before any ost connected,
for avoid problems with LOV EA size and returning EIO to client.

Branch HEAD
b=16080
i=umka
i=tappro

13 years agoBranch HEAD
fanyong [Thu, 20 Nov 2008 02:57:49 +0000 (02:57 +0000)]
Branch HEAD
b=13058
i=alexey.lyashkov
i=tianzy

64 bit quota support check when autoconf.

13 years agoBranch HEAD
anserper [Thu, 20 Nov 2008 01:19:08 +0000 (01:19 +0000)]
Branch HEAD
b=17371
i=Johann Lombardi
testcase for 17371

13 years agoBranch HEAD
vitaly [Wed, 19 Nov 2008 21:28:19 +0000 (21:28 +0000)]
Branch HEAD
b=16129

the test number fix

13 years agoBranch b1_8_gate
vitaly [Wed, 19 Nov 2008 21:14:46 +0000 (21:14 +0000)]
Branch b1_8_gate
b=16129
i=adilger
i=green

- a high priority request list is added into service;
- once a lock is canceled, all the IO requests, including coming
ones, under this lock, are moved into this list;
- PING is also added into this list;
- once a lock cancel timeout occurs, the timeout is prolonged
if there is an IO rpc under this lock;
- another request list is added into the export, used to speed up
the rpc-lock matching.

13 years agoUpdated for 2.0 build 07
stinson1947 [Wed, 19 Nov 2008 20:03:19 +0000 (20:03 +0000)]
Updated for 2.0 build 07

13 years agob=17740
pravins [Wed, 19 Nov 2008 19:54:00 +0000 (19:54 +0000)]
b=17740
allocate separate dentry for Iterator context.

13 years agob=17740
pravins [Wed, 19 Nov 2008 19:29:49 +0000 (19:29 +0000)]
b=17740
pass NULL to bypass capa while destroying orphan lov object.

13 years agob=17653
grev [Wed, 19 Nov 2008 19:20:13 +0000 (19:20 +0000)]
b=17653
i=Adilger
test_21c fix: restore config

13 years agob=17477
grev [Wed, 19 Nov 2008 19:04:56 +0000 (19:04 +0000)]
b=17477
i=Yury.Umanets
run acc-sm:formatall() only if forced

13 years agob=17477
grev [Wed, 19 Nov 2008 17:56:07 +0000 (17:56 +0000)]
b=17477
i=Adilger
force replay-dual to check and mount MOUNT2

13 years agob=17686
yury [Wed, 19 Nov 2008 14:18:37 +0000 (14:18 +0000)]
b=17686
r=shadow,panda

- fixes race in ptlrpcd caused busy obd and inability to cleanup;
- cleanups and debugs in llcd code.

13 years agoRemoving unwanted patches
deshmukh [Wed, 19 Nov 2008 13:22:35 +0000 (13:22 +0000)]
Removing unwanted patches

13 years agoReverting the changes
deshmukh [Wed, 19 Nov 2008 12:59:17 +0000 (12:59 +0000)]
Reverting the changes

13 years agoLand b_head_interop_disk on HEAD (20081119_1314)
deshmukh [Wed, 19 Nov 2008 08:55:54 +0000 (08:55 +0000)]
Land b_head_interop_disk  on HEAD (20081119_1314)
b=11826
i=nikita
i=adilger

13 years agoExported some ldiskfs functions which are required for iop
deshmukh [Wed, 19 Nov 2008 06:47:42 +0000 (06:47 +0000)]
Exported some ldiskfs functions which are required for iop
i=nikita
i=adilger
b=11826

13 years agoBranch HEAD
huanghua [Wed, 19 Nov 2008 02:29:18 +0000 (02:29 +0000)]
Branch HEAD
b=17083
i=johann

add warning message while waiting for exports to be destroyed.

13 years agob=15699
nathan [Wed, 19 Nov 2008 01:20:00 +0000 (01:20 +0000)]
b=15699
libcfs proc debug change requires this

13 years agob=17037
yury [Tue, 18 Nov 2008 10:25:14 +0000 (10:25 +0000)]
b=17037
- fixes merge issue in HEAD which caused deadlock on ctxt->loc_sem

13 years agoBranch HEAD
anserper [Tue, 18 Nov 2008 04:08:21 +0000 (04:08 +0000)]
Branch HEAD
b=12433
i=Oleg Drokin
i=Yury Umanets

fix imp_inval message

13 years agoBranch HEAD
tianzy [Mon, 17 Nov 2008 07:09:51 +0000 (07:09 +0000)]
Branch HEAD
fix a compile error in sles9 and sles10
i=johann
i=nathan

13 years agoBranch HEAD
tianzy [Mon, 17 Nov 2008 07:06:45 +0000 (07:06 +0000)]
Branch HEAD
decay qos ost/oss penalties if MDS is not creating objects
i=nathan
i=johann

13 years agoBranch HEAD
liangzhen [Mon, 17 Nov 2008 03:10:28 +0000 (03:10 +0000)]
Branch HEAD

remove LASSERT on message type from network

b=17372
i=isaac
i=Maxim

13 years agoLand b_head_quota onto HEAD (20081116_0105)
fanyong [Sat, 15 Nov 2008 18:39:35 +0000 (18:39 +0000)]
Land b_head_quota onto HEAD (20081116_0105)
b=13058
i=johann
i=yury.umanets

13 years agoBranch b_head_quota
fanyong [Sat, 15 Nov 2008 15:33:22 +0000 (15:33 +0000)]
Branch b_head_quota
b=13058
i=nikita.danilov
i=johann
i=robert.read

Debug patch for quota on HEAD.

13 years agobranch: HEAD
ericm [Fri, 14 Nov 2008 19:54:21 +0000 (19:54 +0000)]
branch: HEAD
reserve some space in mdt_body and mdt_rec_xxx structures and add
MDS_SETATTR64_REC to allow possible expansion of uid/gid to 64 bits
in the future.
b=17601
r=adilger
r=huanghua
r=fanyong

13 years agob=16488
grev [Fri, 14 Nov 2008 19:20:44 +0000 (19:20 +0000)]
b=16488
i=Oleg.Drokin
RACER accs-m test suit

13 years agob=15699
nathan [Fri, 14 Nov 2008 17:24:56 +0000 (17:24 +0000)]
b=15699
i=adilger
i=manoj
move string-based mask setting into a standalone fn; will also be used by changelogs

13 years agob=17122
grev [Fri, 14 Nov 2008 10:28:02 +0000 (10:28 +0000)]
b=17122
i=Adilger
skip sanity test_100 for NETTYPE != tcp

13 years agocl_lock_free(): remove assertion that does not work in liblustre.
nikita [Thu, 13 Nov 2008 13:44:41 +0000 (13:44 +0000)]
cl_lock_free(): remove assertion that does not work in liblustre.

13 years agob=17479
yury [Thu, 13 Nov 2008 10:15:44 +0000 (10:15 +0000)]
b=17479
r=adilger,behlendorf1

- avoid div/mod in lustre_hash code

13 years agob=16074
green [Thu, 13 Nov 2008 00:42:02 +0000 (00:42 +0000)]
b=16074
r=adilger, Lee Ward

Do not get extra reference on source inode since we already hold the reference
from previous name.

13 years agovvp: always write page from the beginning: sptlrpc depends on kiov_offset being 0.
nikita [Wed, 12 Nov 2008 21:00:37 +0000 (21:00 +0000)]
vvp: always write page from the beginning: sptlrpc depends on kiov_offset being 0.

13 years agolov_object: simplify and document lov_object life-time rules.
nikita [Wed, 12 Nov 2008 20:59:18 +0000 (20:59 +0000)]
lov_object: simplify and document lov_object life-time rules.

13 years agocl_page_list_own(): if page was truncated concurrently, remove it from the queue...
nikita [Wed, 12 Nov 2008 20:58:06 +0000 (20:58 +0000)]
cl_page_list_own(): if page was truncated concurrently, remove it from the queue at once.

13 years agob=17634
grev [Wed, 12 Nov 2008 20:52:41 +0000 (20:52 +0000)]
b=17634
i=Yury.Umanets
insanity cleanup (remove dup fn)

13 years agob=17310
yury [Wed, 12 Nov 2008 19:32:36 +0000 (19:32 +0000)]
b=17310
r=shadow,vitaly

- check rq_phase correctly in ptlrpc_expired_set() and couple of other places.

13 years ago- trivial doxygen fixes
yury [Wed, 12 Nov 2008 17:33:02 +0000 (17:33 +0000)]
- trivial doxygen fixes

13 years ago b=17037
yury [Wed, 12 Nov 2008 17:31:10 +0000 (17:31 +0000)]
 b=17037
 r=tappro,wangdi

  - fixes ost cleanup issue due to missed llcd_put() in the case ost does not receive disconnect from mds;

  - do not sleep on hanging llcd. Instead assert on it _after_ stopping recov_thread's ptlrpcd which should kill any remeining llcds;

  - fixes and cleanups, comments.

13 years agob=16488
grev [Wed, 12 Nov 2008 16:07:34 +0000 (16:07 +0000)]
b=16488
i=Oleg.Drokin
new runracer script

13 years agofile runracer was initially added on branch b1_8_gate.
grev [Wed, 12 Nov 2008 16:06:06 +0000 (16:06 +0000)]
file runracer was initially added on branch b1_8_gate.

13 years agoBranch HEAD
yangsheng [Wed, 12 Nov 2008 14:36:24 +0000 (14:36 +0000)]
Branch HEAD
b=17379

As adilger asked, remove the "out_release:" entirely.

13 years agob=17555
grev [Wed, 12 Nov 2008 11:19:59 +0000 (11:19 +0000)]
b=17555
i=Adilger
use current config instead of reformat fs to have single ost

13 years agoosc_ldlm_glimpse_ast(): keep lock mutex over glimpse processing, for uniformity with...
nikita [Wed, 12 Nov 2008 11:16:25 +0000 (11:16 +0000)]
osc_ldlm_glimpse_ast(): keep lock mutex over glimpse processing, for uniformity with other ASTs.

13 years agocl_lock_request(): add detailed logging for 17665.
nikita [Wed, 12 Nov 2008 11:15:46 +0000 (11:15 +0000)]
cl_lock_request(): add detailed logging for 17665.

13 years agob=16918
alex [Wed, 12 Nov 2008 08:09:08 +0000 (08:09 +0000)]
b=16918
i=adilger

 - fix test checking ROC cache invalidation

13 years agocl_object_glimpse(): do not use loop control variable after the loop has terminated.
nikita [Tue, 11 Nov 2008 11:48:35 +0000 (11:48 +0000)]
cl_object_glimpse(): do not use loop control variable after the loop has terminated.

13 years agopeacify lockdep during mount.
nikita [Tue, 11 Nov 2008 11:47:41 +0000 (11:47 +0000)]
peacify lockdep during mount.

13 years agoBranch HEAD
liangzhen [Tue, 11 Nov 2008 08:47:59 +0000 (08:47 +0000)]
Branch HEAD

Assign msg_ev for GET before lnet_prep_send for REPLY

i=isaac

13 years agoBranch HEAD
tianzy [Tue, 11 Nov 2008 07:54:34 +0000 (07:54 +0000)]
Branch HEAD
Remove usage of trace_call_on_all_cpus()
b=15878
i=adilger
i=robert.read

13 years agoBranch HEAD
liangzhen [Tue, 11 Nov 2008 07:41:59 +0000 (07:41 +0000)]
Branch HEAD

Allow user to specify small value for o2iblnd::concurrent_sends

b=15983
i=isaac

13 years agobranch HEAD
zhanghc [Tue, 11 Nov 2008 07:12:49 +0000 (07:12 +0000)]
branch HEAD

b=17176

fixed a bug in 14774 patch -- compare peer's nid instead of self's nid
in ptlrpc_connection during select failover MDS/OST nodes

i=deen

13 years agob=16461
pravins [Tue, 11 Nov 2008 06:51:43 +0000 (06:51 +0000)]
b=16461
i=umka
i=h.huang

patch to add lr_padding to lu_range struct. this field will be used by
compact fld work.

13 years agoBranch HEAD
yangsheng [Tue, 11 Nov 2008 06:06:02 +0000 (06:06 +0000)]
Branch HEAD
b=17458
i=shadow, bobijam

Update kernel to SLES10 SP2 2.6.16.60-0.31.

13 years agobranch HEAD
zhanghc [Tue, 11 Nov 2008 03:47:05 +0000 (03:47 +0000)]
branch HEAD

b=17495

move the check of recovering state of the OST in osc_precreate
out of "if (oscc->oscc_last_id < oscc->oscc_next_id)" condition
so create operation don't use recovering OST

i=adilger
i=nathan.rutman

13 years agoremove forgotten debugging code that was causing 17643.
nikita [Mon, 10 Nov 2008 18:27:54 +0000 (18:27 +0000)]
remove forgotten debugging code that was causing 17643.

13 years agoUpdated for the next bi-weekly b2.0 build.
stinson1947 [Mon, 10 Nov 2008 17:03:30 +0000 (17:03 +0000)]
Updated for the next bi-weekly b2.0 build.

13 years agob=17617 (in fact this is part of 17310 for HEAD)
yury [Mon, 10 Nov 2008 14:07:22 +0000 (14:07 +0000)]
b=17617 (in fact this is part of 17310 for HEAD)

- fixes manual_umount_client() in conf_sanity.sh to handle --force correctly as it is used by number of tests (this part should be missed in tests porting);

- add assert before decref mgc refcounter;
- fixed possible leaked niduuid in case of error;
- kill @rc in server_put_super() as it is not used;
- use OBD_FAIL_CHECK_ORSET in ptlrpc_check_set() to avoid assert;
- enable test_45 as it is working now.

13 years ago Branch HEAD
liuy [Mon, 10 Nov 2008 12:53:52 +0000 (12:53 +0000)]
 Branch HEAD
 add b_hd_params_tree branch

13 years agob=17235
grev [Mon, 10 Nov 2008 09:49:42 +0000 (09:49 +0000)]
b=17235
i=Alexander.Zarochentsev
cos new test_33a

13 years agob=17310
yury [Sat, 8 Nov 2008 13:31:43 +0000 (13:31 +0000)]
b=17310
r=shadow,johann

- make sure that no new inflight rpcs may come after ptlrpcd_deactivate_import() for both
synchronous and asynchronous sending. To do so we make sure that imp_inflight++ is done only when
permission is granted by ptlrpc_import_delay_req() which makes decision should req be sent,
deferred or killed as import is not in the state to send it in observable future. For async
sending, rpc is only counted inflight when its added to sending or delaying list instead of just
adding it to set for processing.

This fixes assert in ptlrpc_invalidate_import() and as number of other issues;

- synchronize imp_inflight and the presence on sending or delaying list for ptlrpc_queue_wait()
case. So that, now it is guaranteed that if im_inflight != 0 we may always find hanging rpc either
in sending or in delaying list;

- make sure that in ptlrcp_queue_wait() we remove rpc from sending or delaying list and dec
inflight only after ptlrpc_unregister_reply() is done. This way we make sure that accounting is
correct. Rpc can't be returned to the pool or counted finished until lnet lets us go with finished
reply unlink;

- check for inflight and rq_list in pinger;

- comments, cleanups;

13 years agorecovery-small.sh: test_55: fix obvious typos.
nikita [Sat, 8 Nov 2008 13:18:30 +0000 (13:18 +0000)]
recovery-small.sh: test_55: fix obvious typos.

13 years agoland clio.
nikita [Fri, 7 Nov 2008 23:54:43 +0000 (23:54 +0000)]
land clio.
b=14166

13 years agob=17511
yury [Fri, 7 Nov 2008 22:37:57 +0000 (22:37 +0000)]
b=17511
r=johann,adilger

  - removes deadlock possibility by disabling rehash in hash_del() operations and moving hash_add()
    out of spin_locks when calling. Hash table has own mechanisms for protecting its structures and it
    also has hash_add_unique() method for using in concurrent run contexts;

    - fixed missed lh_put() in hash_add_unique() which led to extra refs in some cases (extra ref to
    export) and inability to cleanup;

    - fixed __lustre_hash_set_theta() which set @max theta into ->lh_min_theta;

    - in lustre_hash_rehash_size() disable rehash also for the case when new and old hash sizes equal
    in corner cases (max_size or min_size). Before this fix it could be possible to do needless
    rehashes when size is actually did not change but we do this expensive operation;

    - disable rehash in hash_add_unique() if no actual add happened since entry with the same key is
    already found in the table;

    - some cleanups in hash table code;

13 years agob=17477
grev [Fri, 7 Nov 2008 20:35:25 +0000 (20:35 +0000)]
b=17477
i=Adilger
replace cleanup_and_setup_lustre fn by check_and_setup_lustre fn

13 years agoBranch HEAD
zam [Fri, 7 Nov 2008 17:57:56 +0000 (17:57 +0000)]
Branch HEAD
b=17235
i=adigler@sun.com

add -r 'deletepath' option to createmany.c; createmany.c code cleanup.

13 years agob=17477
grev [Fri, 7 Nov 2008 17:15:13 +0000 (17:15 +0000)]
b=17477
i=Adilger
check config if lustre is mounted before acc-sm run

13 years agob=14384
grev [Fri, 7 Nov 2008 16:34:28 +0000 (16:34 +0000)]
b=14384
i=Brian
assert_DIR cleanup

13 years agob=17445
yury [Fri, 7 Nov 2008 15:14:29 +0000 (15:14 +0000)]
b=17445
r=tappro,johann

- proper locking in rq pool freeing time

13 years ago- mark 45 HEAD_EXCEPT as it fails in HEAD (failure is not related to 17310 where...
yury [Fri, 7 Nov 2008 14:45:59 +0000 (14:45 +0000)]
- mark 45 HEAD_EXCEPT as it fails in HEAD (failure is not related to 17310 where this test was added) and new bug for it is filed - 17617.

13 years agoBranch HEAD
liangzhen [Fri, 7 Nov 2008 14:41:11 +0000 (14:41 +0000)]
Branch HEAD

mode of module_param is not correct

13 years agoBranch HEAD
liangzhen [Fri, 7 Nov 2008 14:25:53 +0000 (14:25 +0000)]
Branch HEAD

Support zero-copy receive of Chelsio driver

b=15093
i=Isaac
i=Maxim

13 years agob=17310
yury [Fri, 7 Nov 2008 14:01:03 +0000 (14:01 +0000)]
b=17310
r=johann,shadow

- fixes in ptlrcp to handle possible long reply unlink for HEAD.

13 years agoBranch HEAD
yangsheng [Fri, 7 Nov 2008 09:12:28 +0000 (09:12 +0000)]
Branch HEAD
b=17357

13 years agoBranch HEAD
yangsheng [Fri, 7 Nov 2008 04:39:29 +0000 (04:39 +0000)]
Branch HEAD
b=17357

Back the changes.

13 years agoBranch HEAD
bobijam [Fri, 7 Nov 2008 03:16:55 +0000 (03:16 +0000)]
Branch HEAD
b=16578
o=adilger

A faster way to get long string.

13 years agob=17235
grev [Thu, 6 Nov 2008 20:12:42 +0000 (20:12 +0000)]
b=17235
i=Alexander.Zarochentsev
i=Adilger
COS replay-dual test_21b

13 years agoBranch HEAD
zam [Thu, 6 Nov 2008 18:11:14 +0000 (18:11 +0000)]
Branch HEAD
b=15393
i=alex.zhuravlev@sun.com
i=tappro@sun.com

Commit on sharing. Eliminate inter-client dependencies between
uncommitted transactions by doing transaction commits.
Thereby clients may recovery independently.

13 years agob=16551
grev [Thu, 6 Nov 2008 11:47:15 +0000 (11:47 +0000)]
b=16551
i=Adilger
correct remote_[mds|ost] fn to work correctly on configuration with
several MDS/OSS nodes

13 years agoBranch HEAD
yangsheng [Thu, 6 Nov 2008 05:21:48 +0000 (05:21 +0000)]
Branch HEAD
b=17357
i=johann, yangsheng

Fixed the endian issues for HEAD.
Author: shadow

13 years agoBranch HEAD
huanghua [Thu, 6 Nov 2008 03:25:10 +0000 (03:25 +0000)]
Branch HEAD
b=17083
i=yury.umanets
i=nikita.danilov

Before destroying the whole mdt stack, just wait a little bit for the
obd_zombie_impexp_thread to destroy all the exports.

13 years agob=17505
zhanghc [Thu, 6 Nov 2008 02:24:44 +0000 (02:24 +0000)]
b=17505

removing mfd from 'closing_list' to fix ASSERTION(list_empty(&mfd->mfd_list))
in mdt_mfd_free in mdt_open.c

i=robert.read

13 years agoBranch HEAD
anserper [Thu, 6 Nov 2008 00:38:30 +0000 (00:38 +0000)]
Branch HEAD
b=17371
i=Johann Lombardi
i=Oleg Drokin

fix a possible race between umount  and requeue thread

13 years agob=16438
kalpak [Wed, 5 Nov 2008 10:03:55 +0000 (10:03 +0000)]
b=16438
  i=adilger
  i=girish

  Mounting a filesystem with extents feature will fail on big-endian systems since ext3-based ldiskfs is not supported on big-endian systems. This can be over-riden with "bigendian_extents" mount option.

13 years agoBranch HEAD
tianzy [Wed, 5 Nov 2008 09:20:02 +0000 (09:20 +0000)]
Branch HEAD
fix panic on debug_daemon
b=15878
i=shadow
i=zhenyu.xu

13 years agob=16438
kalpak [Wed, 5 Nov 2008 09:13:07 +0000 (09:13 +0000)]
b=16438
i=adilger
i=girish

Mounting a filesystem with extents feature will fail on big-endian systems since ext3-based ldiskfs is not supported on big-endian systems. This can be over-riden with "bigendian_extents" mount option.

13 years agob=16918
alex [Wed, 5 Nov 2008 07:31:54 +0000 (07:31 +0000)]
b=16918

 - do cache invalidation once read is complete