Whamcloud - gitweb
adilger [Fri, 3 Oct 2003 17:41:34 +0000 (17:41 +0000)]
Fix POSIX creat.13, fstat.1, open.18, stat.3 mtime/atime on new files.
b=2020
adilger [Fri, 3 Oct 2003 16:14:31 +0000 (16:14 +0000)]
Print hex objids with leading 0x in lfind/lfs find.
eeb [Fri, 3 Oct 2003 10:18:44 +0000 (10:18 +0000)]
* Added echo_destroy_export()
* Made lctl::disconnect ignore inappropriate NALs silently
rread [Thu, 2 Oct 2003 19:52:53 +0000 (19:52 +0000)]
merge devel to lcfg
eeb [Thu, 2 Oct 2003 15:00:13 +0000 (15:00 +0000)]
* new lgmnal replaces gmnal
rread [Wed, 1 Oct 2003 23:51:15 +0000 (23:51 +0000)]
* Add a MOUNTOPT config record in the config log. This is saved as a
profile, which ll_fill_super uses to determine which MDC and LOV
devices to connect to.
* Lconf creates a cleanup log, too.,
* replay-single.sh updated to use current 0-conf scheme. It isn't
pretty yet, but getting there.
rread [Wed, 1 Oct 2003 06:50:50 +0000 (06:50 +0000)]
* adding functions to send llog requests of ptlrpc (untested)
* fixed IOC_OBD_PARSE to parse a recorded config log
alex [Tue, 30 Sep 2003 15:55:14 +0000 (15:55 +0000)]
- UML patch against 2.6.0-test5
wangdi [Tue, 30 Sep 2003 14:47:30 +0000 (14:47 +0000)]
add netconole netdump and kgdb support in 2.6
eeb [Tue, 30 Sep 2003 09:26:09 +0000 (09:26 +0000)]
* Merged HEAD diffs
youfeng [Tue, 30 Sep 2003 06:08:07 +0000 (06:08 +0000)]
b 1585 Support immutable attributes to prevent accidental file deletion from the archives
rread [Mon, 29 Sep 2003 23:42:32 +0000 (23:42 +0000)]
* recording config records _appears_ to work
rread [Mon, 29 Sep 2003 23:26:50 +0000 (23:26 +0000)]
Config interface now uses the new struct lustre_cfg for all config
commands. Config commands are sent by lctl using the
OBD_IOC_PROCESS_CFG ioctl, and they can be recorded with
OBD_IOC_RECORD (untested). The configuration process is now stateless,
and each command refers to the device it operates on by name, if
needed. The lctl interface has changed slightly:
* newdev is not needed. (It continues to exist for compatibility, but
is a NOOP.) The attach command now selects the new device.
* cfg_device <name> is used to select the device that is being
configured. This takes the place of the "device; probe" for config
commands only.
The current config ops are:
LCFG_ATTACH
LCFG_DETACH
LCFG_SETUP
LCFG_CLEANUP
LCFG_LOV_SET_CONFIG
LCFG_ADD_UUID
LCFG_DEL_UUID
The other ioctls are unchanged. The portals config commands have not
been incorporated yet.
adilger [Mon, 29 Sep 2003 21:26:48 +0000 (21:26 +0000)]
Fix ext3 htree directory corruption in this combined htree patch.
b=1516
alex [Mon, 29 Sep 2003 16:09:33 +0000 (16:09 +0000)]
- minor changes in order to get 2.6-compatibility
NOTE: kdev_t and dev_t issue need to be investigated in depth
alex [Sun, 28 Sep 2003 10:52:18 +0000 (10:52 +0000)]
- ext3-wantedi patch against 2.6.0-test5
shaver [Sun, 28 Sep 2003 05:25:08 +0000 (05:25 +0000)]
- wake up the processing thread when a client completes, because it may make
queued + completed = max, at which point we should proceed.
- fix operand inversion in procfs stats calculation.
shaver [Sat, 27 Sep 2003 18:58:41 +0000 (18:58 +0000)]
1742: fix calculation of remaining clients vs. queue length
wangdi [Sat, 27 Sep 2003 14:43:14 +0000 (14:43 +0000)]
update lustre_kernel.spec.in
youfeng [Fri, 26 Sep 2003 03:59:10 +0000 (03:59 +0000)]
b=1615
a greppable magic string to aid LLNL in extracting the lustre messages from
syslog. "Lustre:" for less important messages, and "LustreError" for more
important messages.
alex [Thu, 25 Sep 2003 21:21:48 +0000 (21:21 +0000)]
- rh-2.4.22 series for latest RH kernel
- some patches were ported
NOTE: it builds, it runs lustre and simple tests (cp, ls, stat) work
zab [Thu, 25 Sep 2003 15:48:16 +0000 (15:48 +0000)]
- I've given this script to enough people now that it seems reasonable
to keep it in cvs
adilger [Thu, 25 Sep 2003 08:57:57 +0000 (08:57 +0000)]
Add mrename program to avoid GNU "mv" changing args and doing lstat.
adilger [Thu, 25 Sep 2003 08:55:40 +0000 (08:55 +0000)]
Add mkdir support to multiop.
phil [Thu, 25 Sep 2003 04:54:45 +0000 (04:54 +0000)]
re-committing on b_devel:
b=1997
r=alex
My previous fix for truncate/write lock inversion was close, but badly
flawed. I failed to remember that truncate will internally restart
the transaction -- start a new one, for all intents and purposes -- so
the ordering was backwards. "i_sem before transaction" is the
cardinal rule.
I tried to avoid that because I didn't want to hold the i_sem across
the entire disk I/O in filter_commitrw_write. After some discussion
with Alex, however, we decided that the i_sem need only be held for
the block allocation, and that the BKL suffices for updating i_size.
So the ordering in the write path is now: take i_sem, start transaction,
call filter_direct_io, f_d_io does block alloc, drops i_sem, does I/O.
shaver [Wed, 24 Sep 2003 21:12:08 +0000 (21:12 +0000)]
Part of b=1742:
Allow replay to skip missing transnos if all connected clients are either
complete or have a request in the queue. (The original reply for transno N
might not make it to the wire before the server crashes, but N+1 might have had
a shorter path out. The reply-acks ensure that we can't miss a dependency this
way.)
shaver [Wed, 24 Sep 2003 20:50:15 +0000 (20:50 +0000)]
Make the OST and MDS devices smaller, to minutely speed up the test runs.
Add support for debug-daemon logging and ONLY=cleanup (helpful, I've found,
in debugging).
Start OST before, and stop after, the MDS.
zab [Tue, 23 Sep 2003 22:15:21 +0000 (22:15 +0000)]
- back out the mdc close deferral changes (bug 1883) until mds size
management is ready and really needs it and both the txtbsy (bug 2003) and
mdc semaphore regressions (bug 2008) are taken care of.
alex [Tue, 23 Sep 2003 15:31:24 +0000 (15:31 +0000)]
- assert condition have been changed a bit
alex [Tue, 23 Sep 2003 10:30:42 +0000 (10:30 +0000)]
- this patch is aggregate one for ext-2.4-patch-[1234].patch
NOTE: the patch is being used by pdirops series only, but I'd like to use
it as replacement in all 2.4 kernels
alex [Tue, 23 Sep 2003 10:27:10 +0000 (10:27 +0000)]
- recent bugfixes have been propagated to pdirops series
jacob [Tue, 23 Sep 2003 06:13:47 +0000 (06:13 +0000)]
- lustre-kernel.spec.in: .spec for building a lustre-enabled kernel
- linux-2.4.20-rh-i686-smp.conf: config file for 2.4.20-20.9
- lmake: small wrapper script to build and install linux+lustre
- rh-2.4.target: configuration for lmake for rh-2.4.20 series
- lwizard: script which generates a lustre config file
shaver [Mon, 22 Sep 2003 12:00:53 +0000 (12:00 +0000)]
b=1995: simple recovery-progress/state reporting, in lieu of the CERROR spew.
r=phil, more to come post-LLP.
niu [Mon, 22 Sep 2003 02:33:04 +0000 (02:33 +0000)]
b 1944
fix the bug about parse argument when lfs run in interactive mode.
braam [Sat, 20 Sep 2003 21:04:55 +0000 (21:04 +0000)]
- add code to set_dev_rdonly to flush all transactions up to the calling point
- add a replay test which replays a re-open of a file, where the re-opening is responsible for creating objects.
- add replay code to handle that.
braam [Sat, 20 Sep 2003 17:45:45 +0000 (17:45 +0000)]
- land b_log on b_devel
shaver [Sat, 20 Sep 2003 16:51:06 +0000 (16:51 +0000)]
b=1896:
Two fixes, both required for this bug:
- destroy the import when the export is freed, not when it's disconnected
(which might be due to recovery, and might be racing with ASTs to access
the same import data). (Robert's patch plus a comment fix and an
EXPORT_SYMBOL.)
- take a reference in ptlrpc_fail_export, because obd_disconnect always nets
us -2 (all other callers have another reference, either from a request or
a local-connect artifact), and we only "own" one -- the one in the hash
table, which represents the network connection.
Also: macroize class_export_put and class_export_get for a major simplification
of export-refcount debugging.
shaver [Sat, 20 Sep 2003 14:17:44 +0000 (14:17 +0000)]
b=1958: call lib_recv when simulating failure in lib_parse, just as we do
for all other error-returning cases. Prevents crash when using
"lctl> fail _all_" with the socknal. (Patch from Phil.)
rread [Fri, 19 Sep 2003 21:28:47 +0000 (21:28 +0000)]
* use records that have a tail for the simple record tests.
* change verify_handle to scan the bitmap to deterime the correct
last_idx and active rec count.
* test 4 now uses verify_handle, and fails the last_idx check after
writing one rec.
* run-llog.sh now tells lctl to ignore_errors, so it can cleanup
after a failed test.
niu [Fri, 19 Sep 2003 08:47:49 +0000 (08:47 +0000)]
b: 1944
r: braam
Create user tool lfs. port lstripe & lfind to lfs.
shaver [Thu, 18 Sep 2003 19:26:39 +0000 (19:26 +0000)]
Fix warning.
braam [Thu, 18 Sep 2003 18:12:35 +0000 (18:12 +0000)]
- Niu's changes are back in.
- Changed utime.c to use a non-null time
shaver [Thu, 18 Sep 2003 18:05:44 +0000 (18:05 +0000)]
Fix warnings in ll_sparseness_write/verify.
braam [Thu, 18 Sep 2003 17:33:15 +0000 (17:33 +0000)]
- niu's inode changes broke sanity.sh 36a.
niu [Thu, 18 Sep 2003 07:55:57 +0000 (07:55 +0000)]
b: 1929
r: braam
Improve the code organization in llite.
braam [Thu, 18 Sep 2003 04:23:39 +0000 (04:23 +0000)]
- fix obdo_create cleanup issue in loggin
- make test_llog part of sanity
jerrifer [Thu, 18 Sep 2003 02:36:52 +0000 (02:36 +0000)]
test sparse pwrite
braam [Thu, 18 Sep 2003 02:11:21 +0000 (02:11 +0000)]
- a variety of log fixes to the catalog system
- can now create catalog entries and secondary logs
- some obdo is not being cleaned up.
zab [Wed, 17 Sep 2003 22:06:43 +0000 (22:06 +0000)]
b=1883
r=braam
Wait for dirty data to be handled (writeback, truncate, etc) before issueing
the mdc close. This is done by tracking opens for write and dirty data and
deferring mdc close at file_release if there aren't other opens for write and
there is dirty data. We perform these deferred closes when the dirty data is
taken care of or when another open for write comes along.
Also, a hot sanity test to verify the behaviour.
alex [Wed, 17 Sep 2003 17:38:57 +0000 (17:38 +0000)]
- port of fast ea patch onto 2.4.20-vanilla
alex [Wed, 17 Sep 2003 15:53:16 +0000 (15:53 +0000)]
- ext3_update_inode() zeros inode body on first calling
alex [Wed, 17 Sep 2003 14:55:31 +0000 (14:55 +0000)]
- this version of fast ea patch clears magic in inode in
ext3_xattr_drop_inode() and allows ext3 with large inode to be mounted
eeb [Wed, 17 Sep 2003 11:31:16 +0000 (11:31 +0000)]
* LWT
- single 'where'
- interrupts disabled for whole event
- removed per-cpu get_cycles() offset calculation
- changed event formatting
* socknal
- stop using write_space for tx timeout; consider connection alive
while it can buffer more stuff; potentially the whole socket
buffer must drain within the tx timeout.
- changed peer death notification
- nicer daemon names
- restored RH - specific affinity FTTB
tianying [Wed, 17 Sep 2003 06:12:11 +0000 (06:12 +0000)]
Add mds_cleanup_orphans function by tian ying.
alex [Tue, 16 Sep 2003 21:24:36 +0000 (21:24 +0000)]
- minor cleanups in order to get chaos-2.4.18-pdirops series working
radhika [Tue, 16 Sep 2003 19:12:05 +0000 (19:12 +0000)]
Backed out the changes made by Zhao JianHui (Jerrifer) for b=1222.
phil [Tue, 16 Sep 2003 18:23:24 +0000 (18:23 +0000)]
Followup cleanup commit to the last one: move the setting of the
completion ast to ldlm_lock_create, where the blocking AST is set.
ericm [Tue, 16 Sep 2003 09:53:47 +0000 (09:53 +0000)]
[liblustre]: if pnode revalidate success, intent should be attached onto
the inode. it's the temporary way since libsysio don't recongnize intent
at the moment.
ericm [Tue, 16 Sep 2003 09:32:09 +0000 (09:32 +0000)]
[liblustre]: fix inode->i_ref leak. if pnode revalidation failed we need
remove the stale inode right away to prevent others access it again which
cause i_ref leak. it's an libsysio limitation, not sure if it's the best
way.
jerrifer [Tue, 16 Sep 2003 07:49:24 +0000 (07:49 +0000)]
for sanity.sh test_44a
jerrifer [Tue, 16 Sep 2003 07:48:15 +0000 (07:48 +0000)]
for snaity.sh test_44a
phil [Tue, 16 Sep 2003 03:16:23 +0000 (03:16 +0000)]
The easiest and least controversial part of the patch to eliminate the
gymnastics associated with setting l_completion_ast on the client.
We used to set l_completion_ast when the lock is created, so that a
completion AST which arrives before the RPC reply will still be
handled correctly. Later, we would NULL it out so that lock_enqueue's
call to grant_lock didn't needlessly execute again. This was
race-prone and stupid.
Now ldlm_grant_lock takes a flag, so we can suppress needless calls
without this mess.
eeb [Mon, 15 Sep 2003 21:55:28 +0000 (21:55 +0000)]
* pruned LWT output to account for wrapped cpus
* calculate get_cycles() offsets per-cpu
alex [Mon, 15 Sep 2003 16:29:17 +0000 (16:29 +0000)]
- port of extents onto 2.4.20-vanilla
NOTE: tested by dbench 4/8 in UML only
eeb [Mon, 15 Sep 2003 14:02:03 +0000 (14:02 +0000)]
* Added LWT facility
alex [Mon, 15 Sep 2003 06:32:05 +0000 (06:32 +0000)]
- vanilla-2.4.20 series did't build for i386 arch w/o this patch
alex [Mon, 15 Sep 2003 06:27:10 +0000 (06:27 +0000)]
- raw lookup support in vfs and ext3
NOTE: patch could be applied for all supported 2.4 kernels
braam [Mon, 15 Sep 2003 05:04:04 +0000 (05:04 +0000)]
- lstripe code for b_devel
- error handling is not complete yet (sanity 27f fails)
- lfind interaction (27g 34's) untested
ericm [Sun, 14 Sep 2003 15:15:30 +0000 (15:15 +0000)]
[liblustre]: pnode revalidate.
alex [Sun, 14 Sep 2003 13:38:59 +0000 (13:38 +0000)]
- fix against bug 1935 (OOPS in removing current working directory)
wangdi [Sun, 14 Sep 2003 11:46:47 +0000 (11:46 +0000)]
fix bugs for building rpm
wangdi [Sun, 14 Sep 2003 11:36:23 +0000 (11:36 +0000)]
update lustre_kernel.spec.in
ericm [Sat, 13 Sep 2003 16:48:21 +0000 (16:48 +0000)]
[liblustre]: get most of liblustre functions work again, except:
- pnode revalidation
- file i/o
alex [Sat, 13 Sep 2003 14:12:03 +0000 (14:12 +0000)]
- bug 1931 (link_path_walk bug) fix propogated into those kernels
alex [Sat, 13 Sep 2003 11:47:31 +0000 (11:47 +0000)]
- fix agaist bug #1322 'clients looping in revalidate after recovery failure'
alex [Fri, 12 Sep 2003 19:54:37 +0000 (19:54 +0000)]
- needed .pc file added
phil [Fri, 12 Sep 2003 19:17:22 +0000 (19:17 +0000)]
- replace some hard-coded filter bits with a cpp constant
- create 3 groups instead of 2, as a start to fixing echo_create
ericm [Fri, 12 Sep 2003 16:56:02 +0000 (16:56 +0000)]
[liblustre]: open(O_CREAT)/close()
ericm [Fri, 12 Sep 2003 14:58:03 +0000 (14:58 +0000)]
[liblustre]: stat(), intent related stuff, not complete.
wangdi [Fri, 12 Sep 2003 12:01:05 +0000 (12:01 +0000)]
add spec files building lustre modules + kernel in one rpm
tianying [Fri, 12 Sep 2003 01:04:27 +0000 (01:04 +0000)]
Using generation in mdc_readpage function committed by tian ying.
zab [Thu, 11 Sep 2003 19:26:54 +0000 (19:26 +0000)]
- add an LASSERTF() variant so we can get some printf style data along
with our assert.
phil [Thu, 11 Sep 2003 18:28:03 +0000 (18:28 +0000)]
b=1903
Add /proc/sys/portals/console which defaults to 1; if you set it to 0,
it will suppress all console output except LASSERTs
phil [Thu, 11 Sep 2003 18:25:34 +0000 (18:25 +0000)]
fix annoying printf warnings in mkdirdeep
alex [Thu, 11 Sep 2003 15:58:42 +0000 (15:58 +0000)]
- i_filterdata field in struct inode
niu [Thu, 11 Sep 2003 07:55:56 +0000 (07:55 +0000)]
* commit in.
niu [Thu, 11 Sep 2003 07:17:40 +0000 (07:17 +0000)]
* first commit in.
adilger [Wed, 10 Sep 2003 22:38:27 +0000 (22:38 +0000)]
Add .pc files for dynamic-locks and vfs-pdirops for 2.4.20-rh kernel.
adilger [Wed, 10 Sep 2003 21:59:22 +0000 (21:59 +0000)]
Fix usage, error messages.
alex [Wed, 10 Sep 2003 19:51:07 +0000 (19:51 +0000)]
- fast EA (stored in large inode body) support against 2.4.18-chaos
alex [Wed, 10 Sep 2003 19:46:28 +0000 (19:46 +0000)]
- large inode support for 2.4.18-chaos
NOTE: in fact, all we need is already there, but only check
alex [Wed, 10 Sep 2003 06:06:40 +0000 (06:06 +0000)]
- uml for 2.4.18-chaos and series file
adilger [Tue, 9 Sep 2003 11:19:18 +0000 (11:19 +0000)]
Fix for missing open flags for vfs_intent-2.4.20-rh kernel patch.
This change was also needed for the vfs_intent-2.4.20-hp patch. b=1877
Also add bug 1344 revalidate loop fixes to some of the vfs_intent patches.
The diff is fairly large, so that the files in the 2.4.20-rh patch are
in the same order as in all of the other vfs_intent patches to make
comparison and maintenance easier. Some other minor changes are made to
keep the patches consistent.
There is still a difference between vfs_intent-2.4.20-vanilla and -rh
in follow_down, but I don't know which one is the "keeper".
adilger [Tue, 9 Sep 2003 04:58:14 +0000 (04:58 +0000)]
Add 2.4.21 jbd fix patches to rh-2.4.20 and chaos-2.4.20 kernel series now
that I had a chance to test them against the rh-2.4.20 kernel.
b=1871
phil [Tue, 9 Sep 2003 04:49:16 +0000 (04:49 +0000)]
land b_llog on b_devel:
reorganization and API update of the logging infrastructure. minimal
changes to the other code.
phil [Tue, 9 Sep 2003 03:54:45 +0000 (03:54 +0000)]
merge b_llpmd into b_devel. the major highlights:
- new I/O backend
- new client page cache and llite/lov/osc plumbing
- pre-creation of OST objects
- most of the OBD protocol now revolves around exports, not obd_devices
phil [Tue, 9 Sep 2003 00:39:06 +0000 (00:39 +0000)]
- clear 'rc' if rc == LLOG_EMPTY in llog_init_handle, so we don't free
the header and return an error that our caller doesn't expect
- don't call llog_lvfs_pad if left == reclen, only if 0 < left < reclen
- fix the uuid magic in llog_test
- call llog_init_handle in the right places, and don't forget to close
the log if one of the 1a checks fail
phil [Mon, 8 Sep 2003 23:02:25 +0000 (23:02 +0000)]
- If the first user command isn't ENQUEUE, postsetup would be run too
late, which causes the 'sanity' 1a hang that coop sees. Fixed by
running postsetup if any of the non-setup RPCs arrives.
- Remove overwriting of lgh_last_idx in llog_init_handle
- init the handle from the test code (not quite right yet)
- #if 0 out the filter catalog functions until they build
zab [Mon, 8 Sep 2003 22:41:02 +0000 (22:41 +0000)]
at least get 2.6 building, more testing happens right after gdb stops 2.6 uml
from booting. 2.4 still builds and works.
- 2.4/2.6 compat macro cleaning
- add a ll wrapper for dev_t so mknod can take int in 2.4 and dev_t in 2.6
- 2.6 has to call clear_page_dirty to keep the page dirty counts accurate,
we introduce a helper for 2.4
- tee hee, put lli_vfs_inode back. its important, and stuff.
- add a removepage patch for 2.6
- move readpage into rw.c as both ports now use it
- llite_internal spring cleaning
- update 2.6 writepage to use the Writeback bits
- minor filter_io_26 cleanups, more to be done
- move osc_lock_contains into the right obd_ops initializer
phil [Mon, 8 Sep 2003 22:18:38 +0000 (22:18 +0000)]
- new abstraction in the llog_* function wrappers
- fix double-allocation of handle headers
- added llog_read_header()
- more llog tests