Whamcloud - gitweb
alex [Sat, 13 Sep 2003 18:31:41 +0000 (18:31 +0000)]
- ext3_writepage_trans_blocks() accounts that page may need several blocks.
thus we need not account it twice
alex [Sat, 13 Sep 2003 12:17:39 +0000 (12:17 +0000)]
- fix agaist bug #1322 'clients looping in revalidate after recovery failure'
girishc [Sat, 13 Sep 2003 11:28:36 +0000 (11:28 +0000)]
b=1825
fix minor bug during rename
alex [Fri, 12 Sep 2003 20:51:07 +0000 (20:51 +0000)]
- i_filterdata added into struct inode
phil [Fri, 12 Sep 2003 19:13:46 +0000 (19:13 +0000)]
b=1807
- backport oscc_create rewrite from b_llp_hp
- send osc_creates to a new OST thread, so that time-sensitive
creations will not have to compete with billions of read/write requests
girishc [Fri, 12 Sep 2003 13:25:11 +0000 (13:25 +0000)]
ll_setattr not supported in b_llp_hp
shaver [Thu, 11 Sep 2003 22:04:20 +0000 (22:04 +0000)]
b=1792: Fix iopen_connect_dentry to properly rehash the __iopen__ dentry when
coverting it into a real, "connected" dentry.
girishc [Thu, 11 Sep 2003 17:05:34 +0000 (17:05 +0000)]
Bringing NFS export code uptodate with b_llp_hp
ericm [Tue, 9 Sep 2003 14:37:10 +0000 (14:37 +0000)]
merge b_devel -> b_eq:
20030909
only kernel pass sanity, liblustre still broken
adilger [Tue, 9 Sep 2003 11:19:11 +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".
phil [Tue, 9 Sep 2003 04:49:12 +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 03:55:05 +0000 (03:55 +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 01:51:07 +0000 (01:51 +0000)]
- export llog_cat_put
- disable the get/put_catalog bits, until they're a little more cooked
phil [Tue, 9 Sep 2003 00:39:05 +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
braam [Mon, 8 Sep 2003 23:37:44 +0000 (23:37 +0000)]
- make catalog api's available again, now based on the other layers.
phil [Mon, 8 Sep 2003 23:02:24 +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
phil [Mon, 8 Sep 2003 22:18:37 +0000 (22:18 +0000)]
- new abstraction in the llog_* function wrappers
- fix double-allocation of handle headers
- added llog_read_header()
- more llog tests
braam [Mon, 8 Sep 2003 22:07:47 +0000 (22:07 +0000)]
- initialize handles
braam [Mon, 8 Sep 2003 18:31:11 +0000 (18:31 +0000)]
- unify llog_open and llog_create
phil [Mon, 8 Sep 2003 17:08:32 +0000 (17:08 +0000)]
- set the mds's and ost's obd_logops
- initial oa to null in llog_lvfs_create, avoids crashing
- in llog test 1a, don't try to close if the open failed
- 'lctl modules' hook for llog_test
phil [Mon, 8 Sep 2003 07:51:02 +0000 (07:51 +0000)]
- steal from the reserved bits instead of the bitmap bits, another
piece of sage Andreas advice
- add an llh_flags field to the header
- rename confusing lgh_index to lgh_last_idx
- if the log is marked ZAP_WHEN_EMPTY, the log is full, and all of the
records are cancelled, call llog_destroy from llog_cancel_rec
- extend llog_open to also support reopen-by-name
- add llog_vfs_read_blob and llog_vfs_read_header
- removed a bunch of unnecessary EXPORT_SYMBOLs
- add llog_lvfs_open()
phil [Mon, 8 Sep 2003 06:25:28 +0000 (06:25 +0000)]
- pad llog_* body structures correctly (thanks Andreas)
- countless build fixes
braam [Mon, 8 Sep 2003 05:35:19 +0000 (05:35 +0000)]
- few compile fixes
braam [Mon, 8 Sep 2003 05:14:19 +0000 (05:14 +0000)]
- next steps towards a working llogging api
phil [Mon, 8 Sep 2003 05:10:41 +0000 (05:10 +0000)]
- remove obd_run_ctxt from obd->u.mds and obd->u.filter; put it in the
obd_device
- add lvfs_callback_ops, and the first one, the l_fid2dentry
- add callback ops to obd_run_ctxt
- change the 100 uses of mds->mds_ctxt and filter->fo_ctxt
- add a handful of llog tests
braam [Mon, 8 Sep 2003 04:07:01 +0000 (04:07 +0000)]
- fixes to the api's
phil [Mon, 8 Sep 2003 03:12:41 +0000 (03:12 +0000)]
Fix a few build errors
braam [Mon, 8 Sep 2003 02:39:13 +0000 (02:39 +0000)]
- the same llog reorg again, now on the right branch...
adilger [Sun, 7 Sep 2003 08:46:05 +0000 (08:46 +0000)]
Fixups to checkstack target (get diffs in right order, show top 30 abusers).
Don't show more functions with less than 200 bytes stack on ia64.
adilger [Sun, 7 Sep 2003 05:24:07 +0000 (05:24 +0000)]
Update checkstack program, to output more useful "stack_usage function name".
Add a "make checkstack" target which shows the current "worst offenders" list
of stack users, or the difference in this list since the last time that target
was run.
There is also a "make checkstack-clean" tarket which resets the "saved" stack
values so that the stack usage diff doesn't show old items.
phil [Fri, 5 Sep 2003 18:23:40 +0000 (18:23 +0000)]
merge b_devel into b_llpmd
adilger [Fri, 5 Sep 2003 00:16:36 +0000 (00:16 +0000)]
Add jbd-* fixes to some of the series files where I know they apply:
- vanilla-2.4.20 and hp-pnnl-2.4.20 need all three
- chaos-2.4.18 doesn't need the flushtime patch (it is in their tree already)
I'm not sure whether chaos-2.4.20 and rh-2.4.20 need them or not (patches
were originally from RH kernels, so they may already be in rh-2.4.20).
ericm [Wed, 3 Sep 2003 16:21:34 +0000 (16:21 +0000)]
revert the last fix for cygwin, which make ogdb-uml always null.
wangdi [Tue, 2 Sep 2003 13:57:04 +0000 (13:57 +0000)]
add kgdb over udp patch
ericm [Thu, 28 Aug 2003 15:25:38 +0000 (15:25 +0000)]
[portals]: in handler of SIGALRM, we must reenable SIGALRM before longjmp,
because it's automatically disabled when entering handler.
ericm [Thu, 28 Aug 2003 09:09:13 +0000 (09:09 +0000)]
[tcpnal]: use passed-in port number to start listening socket.
ericm [Tue, 26 Aug 2003 03:28:43 +0000 (03:28 +0000)]
[portals]:
PtlEQWait_timeout(): if left_over is 0, don't call alarm with it,
otherwise we'll not got timeout notification.
phil [Sat, 23 Aug 2003 22:21:18 +0000 (22:21 +0000)]
merge b_devel into b_llpio, including v24 kernel patch
cvs2svn [Fri, 22 Aug 2003 21:40:01 +0000 (21:40 +0000)]
This commit was manufactured by cvs2svn to create branch 'unlabeled-1.1.4'.
phil [Fri, 22 Aug 2003 21:39:57 +0000 (21:39 +0000)]
fix socknal build on vanilla kernels by adding socket exports
bumped the kernel patch version to 23, but really only vanilla-2.4.20 changed
ericm [Fri, 22 Aug 2003 15:02:33 +0000 (15:02 +0000)]
merge b_devel -> b_eq:
20030822
wangdi [Fri, 22 Aug 2003 09:05:16 +0000 (09:05 +0000)]
file Makefile.am was initially added on branch b_devel.
rread [Fri, 22 Aug 2003 02:19:00 +0000 (02:19 +0000)]
b=1803
r=shaver
New import state machine, as documented on the lustre wiki in
ImportStates.
A new function, ptlrpc_connect_import, performs all import connects
and moves the import from the DISCON state to either FULL, EVICTED,
REPLAY, or RECOVER, depending on the situation. Unlike the levels, the
states are now exact, and the request->rq_send_state much match the
import state to be sent.
Passes recovery/01, replay-small, and replay-dual.
rread [Fri, 22 Aug 2003 02:18:55 +0000 (02:18 +0000)]
file import.c was initially added on branch b_devel.
ericm [Thu, 21 Aug 2003 10:37:15 +0000 (10:37 +0000)]
LiuKai's another Cygwin fix for data alignment, add #pragma
ericm [Thu, 21 Aug 2003 04:54:47 +0000 (04:54 +0000)]
[tcpnal]: service socket listening on port 988
ericm [Thu, 21 Aug 2003 04:16:54 +0000 (04:16 +0000)]
[tcpnal]: merge LiuKai's cygwin patch
ericm [Thu, 21 Aug 2003 02:53:50 +0000 (02:53 +0000)]
[tcpnal]: enlarge iov array in tcpnal_send().
ericm [Wed, 20 Aug 2003 11:23:33 +0000 (11:23 +0000)]
[tcpnal]:
- don't consider pid when doing connection search, it's conform to the
ksocknal behavior.
ericm [Wed, 20 Aug 2003 11:09:01 +0000 (11:09 +0000)]
[tcpnal]:
- protect tcpnal internal data structure
ericm [Wed, 20 Aug 2003 08:46:12 +0000 (08:46 +0000)]
[tcpnal]:
- add nal_cb locking, fix define of state_lock in user space.
ericm [Wed, 20 Aug 2003 08:22:54 +0000 (08:22 +0000)]
[tcpnal}:
temporarily work around for nal thread endless waiting problem.
girishc [Wed, 20 Aug 2003 07:43:23 +0000 (07:43 +0000)]
file nfs_export_kernel-2.4.20.pc was initially added on branch b_nfsdevel.
ericm [Wed, 20 Aug 2003 07:22:07 +0000 (07:22 +0000)]
[tcpnal]: rewrite the data out path.
Get rid of pipe as a bridge between user threads and nal threads, which
is not so efficient and don't work well in Cygwin. Now user thread might
directly call into portals lib via lib_dispatch.
2 issues remain:
- sometimes during startup low level nal threads could run into endless
waiting.
- need more sycronous facilities to prevent race.
behlendo [Tue, 19 Aug 2003 19:13:24 +0000 (19:13 +0000)]
file LLNL_Changelog was initially added on branch b_llnl_stable.
phil [Tue, 19 Aug 2003 17:38:30 +0000 (17:38 +0000)]
merge b_multinet into HEAD
phil [Tue, 19 Aug 2003 17:10:03 +0000 (17:10 +0000)]
b=1505
Disable some console messages from failed statfs(), reconnect
ericm [Tue, 19 Aug 2003 12:08:53 +0000 (12:08 +0000)]
[liblustre]: merge LiuKai's cygwin patch
- add portals/include/cygwin-ioctl.h
- don't compile accepter.c in liblustre
- add missing le-cpu convert macro
- use windows API directly at some places
ericm [Tue, 19 Aug 2003 12:08:49 +0000 (12:08 +0000)]
file cygwin-ioctl.h was initially added on branch b_eq.
alex [Mon, 18 Aug 2003 09:40:33 +0000 (09:40 +0000)]
file uml_2.6.0_test3 was initially added on branch b_llpmd.
alex [Mon, 18 Aug 2003 09:40:19 +0000 (09:40 +0000)]
file uml-patch-2.6.0-test3-1.pc was initially added on branch b_llpmd.
alex [Mon, 18 Aug 2003 09:40:10 +0000 (09:40 +0000)]
file uml-patch-2.6.0-test3-1.patch was initially added on branch b_llpmd.
braam [Sat, 16 Aug 2003 05:08:17 +0000 (05:08 +0000)]
file osc_create.c was initially added on branch b_llpmd.
adilger [Fri, 15 Aug 2003 20:28:09 +0000 (20:28 +0000)]
Fix for truncate/write inversion.
b=1639
r=phil
girishc [Thu, 14 Aug 2003 17:20:39 +0000 (17:20 +0000)]
file nfs_export_kernel-2.4.20.patch was initially added on branch b_nfsdevel.
mdoyle [Thu, 14 Aug 2003 14:07:51 +0000 (14:07 +0000)]
file lgmnalnid.c was initially added on branch b_myrinet.
ericm [Thu, 14 Aug 2003 07:04:43 +0000 (07:04 +0000)]
[tcpnal]:
tcpnal_recv() need call lib_finalize() even the message size
is 0, e.g. ACK. hit by TianYing.
phil [Thu, 14 Aug 2003 05:01:54 +0000 (05:01 +0000)]
merge three weeks of b_devel fixes into b_filterio
rread [Wed, 13 Aug 2003 20:16:22 +0000 (20:16 +0000)]
b=1513
r=shaver
* add conn_cnt to import, export, and lustre_msg. The conn_cnt is
increased by the client whenever it connects or reconnects. The
server ignores failed BRWs with an old conn_cnt.
Also, when a bulk is resent, the xid is changed, so the previous one
will definitely fail.
phil [Wed, 13 Aug 2003 19:02:31 +0000 (19:02 +0000)]
Fix size validation with getattr intents. b=1768
phil [Wed, 13 Aug 2003 18:19:25 +0000 (18:19 +0000)]
b=1592
Read past EOF would clear res->dentry in preprw; we would oops trying
to dput it in commitrw. Fixed.
adilger [Wed, 13 Aug 2003 17:00:00 +0000 (17:00 +0000)]
file iosanity.sh was initially added on branch b_devel.
ericm [Wed, 13 Aug 2003 03:53:45 +0000 (03:53 +0000)]
[tcpnal]: workaround: serialize access to procbridge pipe. TianYing's work
exposed the race condition here.
phil [Tue, 12 Aug 2003 23:24:08 +0000 (23:24 +0000)]
b=1642
r=zab
- Land a cleanup of the preprw_read/commitrw_read path from b_filterio.
The old dentry cleanup in error cases could not be overseen.
- Backport a copy of Zach's fix for bug 1741, believed the be the
cause of the intermittent partial-page corruption
adilger [Tue, 12 Aug 2003 23:18:08 +0000 (23:18 +0000)]
file write_append_truncate.c was initially added on branch b_devel.
adilger [Tue, 12 Aug 2003 16:59:45 +0000 (16:59 +0000)]
file sleeptest.c was initially added on branch b_devel.
adilger [Tue, 12 Aug 2003 16:18:37 +0000 (16:18 +0000)]
Exit early from mds_open() if we get an error.
b=1749
r=phil
adilger [Tue, 12 Aug 2003 16:13:58 +0000 (16:13 +0000)]
Fix import levels when a reconnect happens without a previous timeout.
b=1597
r=shaver
adilger [Tue, 12 Aug 2003 06:39:26 +0000 (06:39 +0000)]
Return an error from lov_create() if all OSCs are inactive.
b=1751
r=phil,jacob
adilger [Tue, 12 Aug 2003 06:26:29 +0000 (06:26 +0000)]
Don't LBUG if we get bad stripe data back from the MDS (normally a bug, but
not one that we want to crash on).
jacob [Fri, 8 Aug 2003 20:10:29 +0000 (20:10 +0000)]
- Fix build with rh 2.4.20 kernel series
r=phil
b=1663
rread [Fri, 8 Aug 2003 10:04:36 +0000 (10:04 +0000)]
file keepopen.py was initially added on branch b_devel.
adilger [Fri, 8 Aug 2003 06:43:09 +0000 (06:43 +0000)]
Land fix for bug 1558 on HEAD. r=phil,shaver
adilger [Fri, 8 Aug 2003 05:56:59 +0000 (05:56 +0000)]
Land fix for bug 1558 on HEAD. r=phil,shaver
shaver [Thu, 7 Aug 2003 17:10:52 +0000 (17:10 +0000)]
fix some lprocfs rot, b=1732, r=phik
ericm [Thu, 7 Aug 2003 10:27:56 +0000 (10:27 +0000)]
file liblustre_sanity_uml.sh was initially added on branch b_eq.
behlendo [Tue, 5 Aug 2003 18:43:21 +0000 (18:43 +0000)]
file LLNL_ChangeLog was initially added on branch b_llnl_devel.
phil [Tue, 5 Aug 2003 15:58:40 +0000 (15:58 +0000)]
invoking section 3 of the GNU LGPL, to instead apply the terms of the GPL
mdoyle [Tue, 5 Aug 2003 10:11:24 +0000 (10:11 +0000)]
Fixed license statment
phil [Sun, 3 Aug 2003 23:57:51 +0000 (23:57 +0000)]
b=1598
Final version of the read performance patch, tested on ALC.
phil [Sun, 3 Aug 2003 21:32:48 +0000 (21:32 +0000)]
merge b_llnl_devel into b_multinet in preparation for landing
phil [Sun, 3 Aug 2003 21:06:34 +0000 (21:06 +0000)]
* Landing fix on HEAD, originally committed to b_devel
b=707
r=shaver
Makes the LRU size a /proc tunable, "lru_size". If you shrink the LRU
size below the number of existing unused locks, they're cancelled
immediately. Echo "clear" to cancel all locks without changing the value.
phil [Sun, 3 Aug 2003 21:04:49 +0000 (21:04 +0000)]
* Landing fix on HEAD, originally committed to b_devel
r=rread
move spin_unlock outside the condition, noticed by eric.
phil [Sun, 3 Aug 2003 21:04:09 +0000 (21:04 +0000)]
* Landing fix on HEAD, originally committed to b_devel
b=1625
r=adilger
Avoid dereferencing 'lock' after we LDLM_LOCK_PUT, by saving a copy of
the namespace pointer and using that to unlock.
Alternatively, we could drop the NS lock and then do the PUT, but this
is an equally low-impact fix.
phil [Sun, 3 Aug 2003 20:55:11 +0000 (20:55 +0000)]
Oops, I think I had an unsaved buffer; backout accidental commit.
phil [Sun, 3 Aug 2003 19:44:44 +0000 (19:44 +0000)]
b=1598
r=shaver
Fix the 3x read performance regression caused by clobbering the
client's i_blksize number, which sets an upper bound on the amount of
readahead done by a single ll_readpage. The practical upshot was that
all reads after the first one would do 4k to the disk and 4k over the
wire.
phil [Sun, 3 Aug 2003 16:39:45 +0000 (16:39 +0000)]
file stat_fs.h was initially added on branch b_devel.
ericm [Sun, 3 Aug 2003 07:06:35 +0000 (07:06 +0000)]
file rbtree.c was initially added on branch b_eq.
ericm [Sun, 3 Aug 2003 07:06:34 +0000 (07:06 +0000)]
file rbtree.h was initially added on branch b_eq.
phil [Sat, 2 Aug 2003 19:13:50 +0000 (19:13 +0000)]
b=1644
r=eeb
This patch fixes the LBUG by clearing the bits that indicate that the
OST/echo_server is sending valid cache grant information.
Li Zongliang tested this patch but was unable to check it in.