Whamcloud - gitweb
zab [Tue, 2 Mar 2004 01:34:30 +0000 (01:34 +0000)]
- the start of moving things from portals.o to libcfs.o. the 2.4 kernel
build still works though others probably don't.
eeb [Fri, 20 Feb 2004 20:15:59 +0000 (20:15 +0000)]
* Brought portals error and handle declarations into line with the
portals spec.
cvs2svn [Sun, 27 Jul 2003 02:40:00 +0000 (02:40 +0000)]
This commit was manufactured by cvs2svn to create branch
'unlabeled-1.1.4.2.6'.
eeb [Sun, 27 Jul 2003 02:39:59 +0000 (02:39 +0000)]
* Merged b_devel updates
braam [Sat, 26 Jul 2003 23:14:37 +0000 (23:14 +0000)]
- do not send unused uid and gid fields in mdc/mds update packets
- correctly handle subdirectories of sgid directories (owner set to
parent id - test 6g added for this)
- fix for 1607 (test 19c)
- remove unused kernel configuration files
eeb [Sat, 26 Jul 2003 16:59:48 +0000 (16:59 +0000)]
* Guarded #define EXPORT_SYMTAB
* Bit the bullet and #defined PTL_MTU
NB currently defined in <portals/lib-types.h>; maybe should move...
phil [Sat, 26 Jul 2003 02:28:13 +0000 (02:28 +0000)]
merge recent fixes from b_devel into b_filterio
shaver [Fri, 25 Jul 2003 23:04:34 +0000 (23:04 +0000)]
b=1600: my version of Andreas' patch to use write_record to ensure that the
MDS server data is written at creation time
b=1597: cope with entering recovery/replay without a preceding timeout
We pass this initial single-node/single-mount replay test, yay.
braam [Fri, 25 Jul 2003 22:55:09 +0000 (22:55 +0000)]
- add page private field
braam [Fri, 25 Jul 2003 22:42:16 +0000 (22:42 +0000)]
- add the forgotten error export to chaos-2.4.18 series
phil [Fri, 25 Jul 2003 21:54:56 +0000 (21:54 +0000)]
merge b_devel into b_filterio
braam [Fri, 25 Jul 2003 21:28:46 +0000 (21:28 +0000)]
- backport a few name changes in the patches to 2.4.18
- include some new I/O functions.
phil [Fri, 25 Jul 2003 21:18:15 +0000 (21:18 +0000)]
land b_unify on b_devel
- patches for 2.6
- linus-approved APIs
- lots of little reorg and fixes for 2.4/2.6 compatibility
- better code to manage MDC lock cancellation and dentry memory pressure
phil [Fri, 25 Jul 2003 18:55:06 +0000 (18:55 +0000)]
merge b_devel into b_unify, in preparation for landing
braam [Thu, 24 Jul 2003 22:03:44 +0000 (22:03 +0000)]
- add __d_drop for 2.4
braam [Thu, 24 Jul 2003 05:45:28 +0000 (05:45 +0000)]
- some compatibility macros.
- shift the open flag setting to namei.c as per Trond's & Linus
patches
- this code gets fairly far in sanity.
braam [Thu, 24 Jul 2003 02:21:19 +0000 (02:21 +0000)]
- Linus style patches for 2.4. 2.4 passes sanity again.
braam [Wed, 23 Jul 2003 04:31:42 +0000 (04:31 +0000)]
- THIS BREAKS b_unify
2.6.0 patches now fixed to Linus specs. 2.6.0 compiles, untested.
beginning of similar changes for 2.4 changes
braam [Wed, 23 Jul 2003 04:31:22 +0000 (04:31 +0000)]
- THIS BREAKS b_unify
2.6.0 patches now fixed to Linus specs. 2.6.0 compiles, untested.
beginning of similar changes for 2.4 changes
phil [Tue, 22 Jul 2003 14:23:27 +0000 (14:23 +0000)]
merge Andreas's last commit into b_unify
phil [Tue, 22 Jul 2003 14:05:20 +0000 (14:05 +0000)]
partial merge of b_unify into b_devel (20030722_0202)
phil [Tue, 22 Jul 2003 13:50:44 +0000 (13:50 +0000)]
partial merge of b_unify into b_devel (20030722_0202)
phil [Tue, 22 Jul 2003 03:37:05 +0000 (03:37 +0000)]
merge b_devel into b_unify
phil [Tue, 22 Jul 2003 03:35:45 +0000 (03:35 +0000)]
merge b_devel into b_unify
phil [Sat, 19 Jul 2003 18:41:50 +0000 (18:41 +0000)]
Implement llog_cleanup_commmit_master, so that the orphan code stops
leaking threads (and a little memory)
mmex [Fri, 18 Jul 2003 01:28:35 +0000 (01:28 +0000)]
- update rh-2.4.20 configuration to compile correctly
adilger [Thu, 17 Jul 2003 21:07:20 +0000 (21:07 +0000)]
Set mtime and ctime on old inode if truncating by thread.
Also, mark the inodes dirty _after_ we modify them (done in orphan_add/del),
in case of crash before the real truncate is called.
I thought this might be the cause of bug 1530 (deleted files come back to
life), but that would not be the case even if there is a problem in the
delete thread as the name unlink is done before the delete thread gets
involved.
zab [Thu, 17 Jul 2003 17:35:20 +0000 (17:35 +0000)]
- bring recent b_unify changes into b_filterio
(UNIFY_FILTERIO_UPDATE_CHILD_20030717)
eeb [Thu, 17 Jul 2003 13:03:07 +0000 (13:03 +0000)]
* merged current b_devel
adilger [Thu, 17 Jul 2003 07:57:33 +0000 (07:57 +0000)]
Landing statfs cache code.
- preliminary for prealloc (bug 974)
- preliminary for QOS (bug 855)
- speeds up parallel df
- speeds up lprocfs
Lprocfs underwent a bunch of cleanup (mostly from Intel, partly from me)
to remove lots of macros and strangeness w.r.t. per-target statfs values.
Still open for discussion is how old the cached statfs values can be.
I arbitrarily picked 1 second for client statfs, lprocfs, and target
statfs caches. We may want to change that, or allow it to be tunable.
Note that this code is different than that in bug 855, in that the
max_age parameter is an absolute age in jiffies, whereas the patch
in bug 855 used a relative time in seconds. The thing I'm trying
to avoid is if we say "I want a result up to 1 second old", but we
get delayed during an RPC and we don't know what "1 second old" is
related to.
mmex [Thu, 17 Jul 2003 00:50:53 +0000 (00:50 +0000)]
- updated vanilla series to include uml 6 patch
zab [Tue, 15 Jul 2003 19:36:50 +0000 (19:36 +0000)]
- bring recent b_unify fixes into b_filterio
(UNIFY_FILTERIO_UPDATE_CHILD_20030715)
shaver [Tue, 15 Jul 2003 08:53:19 +0000 (08:53 +0000)]
Land b_unify on b_devel.
Highlights include:
- intent rework from Peter and Phil
- unification of VFS intent strategy for 2.4 and 2.5/2.6
- select 2.5 kernel patches from b_ad
- all the whitespace fixes you can imagine.
Complete diff for sanity-checking at http://off.net/~shaver/unify-landing.diff.
braam [Tue, 15 Jul 2003 00:01:15 +0000 (00:01 +0000)]
- remaining b_unify ---> b_ad merge
braam [Mon, 14 Jul 2003 23:42:19 +0000 (23:42 +0000)]
- merge b_unify into b_ad. This makes b_ad almost ready to be landed
on b_unify and we may just do that next.
shaver [Sun, 13 Jul 2003 00:32:15 +0000 (00:32 +0000)]
merge b_intent into b_unify, now that we know it passes
alex_tomas [Sat, 12 Jul 2003 10:03:52 +0000 (10:03 +0000)]
- once again (it seems I have to wake up ...)
alex_tomas [Sat, 12 Jul 2003 09:51:49 +0000 (09:51 +0000)]
- inode-protection-from-pdflush.patch: an inode may have I_SKIP_PDFLUSH
flag which protects it from to be processed by pdflush's writeback
mechanism. this allows us to work locking problem with lustre_fwrite()
around
- MDS sets I_SKIP_PDFLUSH flag on 'last record' file
- filter_prep() sets I_SKIP_PDFLUSH flag
- inode-protection-from-pdflush.patch has been included in kgdb-2.5.73
series
alex_tomas [Sat, 12 Jul 2003 09:36:55 +0000 (09:36 +0000)]
- ext3-noread-inode.patch added
braam [Sat, 12 Jul 2003 08:28:58 +0000 (08:28 +0000)]
- cleanup in super and llite_lib to make this closer to b_unify.
mmex [Fri, 11 Jul 2003 01:40:56 +0000 (01:40 +0000)]
- add adilger's asyncdel patchs, the 2.4.20 versions.
alex_tomas [Thu, 10 Jul 2003 20:43:09 +0000 (20:43 +0000)]
- almost all routines from super.c and super25.c extracted into llite_lib.c
pschwan [Thu, 10 Jul 2003 18:42:21 +0000 (18:42 +0000)]
this is a large portion of the outstanding changes on b_ad, almost all
of which look trivial and readily acceptable to b_devel. landing
first on b_unify to fix a few bugs and test more heavily
shaver [Thu, 10 Jul 2003 01:01:21 +0000 (01:01 +0000)]
b=1397: Add --pwfile to lactive, specifying a file from which it will read the
LDAP password.
braam [Wed, 9 Jul 2003 18:39:52 +0000 (18:39 +0000)]
- another merge of b_devel into b_ad
shaver [Wed, 9 Jul 2003 00:35:41 +0000 (00:35 +0000)]
lconf support for client-side loading of management client as required.
Bump config version to account for new elements, just to be safe.
braam [Tue, 8 Jul 2003 22:41:20 +0000 (22:41 +0000)]
- we could call this "merely a merge from b_devel". Notes:
- lprocfs is not managable
- this oopses on mount, will be fixed soon.
pschwan [Sat, 5 Jul 2003 22:47:43 +0000 (22:47 +0000)]
land b_merge on b_devel. Includes b_mount and b_orphan, which includes:
- orphans are moved into the PENDING directory for possible recovery
- replayed opens now open by fid for orphan/rename safety (1042)
- last close of an orphan inode generates a transno (683)
- avoid CERROR in normal ll_setattr_raw() error case (1500)
Added code which is currently disabled:
- dentry pinning from b_mount for cwd and mountpoints (1020)
- logging code for orphans (and many other things)
adilger [Fri, 4 Jul 2003 18:29:34 +0000 (18:29 +0000)]
Quiet a number of ia64 compile errors (not all of which are ia64 specific).
Strangely, it seems that (int) / (size_t) is promoted to some non-int type
on ia64, I don't know why, since the result could never become larger than int.
pschwan [Fri, 4 Jul 2003 08:56:58 +0000 (08:56 +0000)]
merge b_orphan into b_merge, with orphan code disabled
pschwan [Thu, 3 Jul 2003 19:55:01 +0000 (19:55 +0000)]
Merge b_devel into b_orphan (
20030703 update tag)
shaver [Wed, 2 Jul 2003 18:11:02 +0000 (18:11 +0000)]
b=1492: use a single per-mount/client UUID for all server connections.
rread [Wed, 2 Jul 2003 17:25:42 +0000 (17:25 +0000)]
Add the same module directory code that lconf uses to find the Lustre python modules.
rread [Wed, 2 Jul 2003 16:26:42 +0000 (16:26 +0000)]
Use the correct path for the XSL and top.ldif when installed from the rpm.
rread [Tue, 1 Jul 2003 07:35:30 +0000 (07:35 +0000)]
b=1424
r=shaver,others
- cleanup import level and generation
- fixes several bugs:
b=1420 ost and client both timeout, client never reconnects
b=1440 very small change to call the lov set_active in recovery
b=1448 another small change to send DISCONNECT at a normal level
braam [Mon, 30 Jun 2003 20:52:38 +0000 (20:52 +0000)]
- move things forward to 2.5.73, include lkcd patches
shaver [Mon, 30 Jun 2003 20:11:30 +0000 (20:11 +0000)]
b=957: add pinger for recoverable clients, in order to promptly detect server
failure and join recovery. Currently DISABLED (#ifdef ENABLE_PINGER)
pending some discussion about deployment.
Also: use the --config (or file.xml) instead of 'none' in the mount line, so
that mounted filesystems have a little more context.
pschwan [Thu, 26 Jun 2003 17:49:34 +0000 (17:49 +0000)]
- Many unused variable warning fixes (several of which were not really a
problem, I think)
- list_splice_init() isn't defined in 2.4.18
- FSTYPE change for lov.sh
adilger [Tue, 24 Jun 2003 23:34:23 +0000 (23:34 +0000)]
Fix for bug 1466, delete thread locking up machine during orphan recovery.
rread [Wed, 18 Jun 2003 21:06:29 +0000 (21:06 +0000)]
b=1397
- lconf creates a readonly connection to LDAP
adilger [Wed, 18 Jun 2003 18:10:07 +0000 (18:10 +0000)]
Fix patches that were eaten by patch scripts.
adilger [Wed, 18 Jun 2003 09:07:20 +0000 (09:07 +0000)]
Fix delete thread so that it doesn't sleep uninterruptibly and raise load avg.
Do not start delete thread if asyncdel mount option is not given.
rread [Mon, 16 Jun 2003 23:01:18 +0000 (23:01 +0000)]
- fix --select
adilger [Mon, 16 Jun 2003 07:04:59 +0000 (07:04 +0000)]
Remove excessive use of macros in the new lprocfs stats counters. This was
the source of one bug, and wasn't necessary. Also a general cleanup.
Removed duplicate "read/write" counter, as this data will already be contained
in the "read/write_bytes" counter as the number of events.
Found out that lprocfs RPC counters do not work under UML, because
"get_cycles()" always returns 0.
thantry [Sat, 14 Jun 2003 01:32:16 +0000 (01:32 +0000)]
Convert spaces into tabs for patch to apply cleanly.
eeb [Fri, 13 Jun 2003 13:22:08 +0000 (13:22 +0000)]
* Merged b_devel changes into b_eq (again). b_eq and b_devel have same
acceptance-small symptoms viz....
1/ Need to sleep before lconf -d on llite client to ensure all
connection refcounts are 0 (llmountcleanup.sh on b_eq has a 10 second
sleep).
2/ When running dbench 50, the OST complains about long-running
obdfilter ops. Eventually the MDS or OST will crash/hang.
rread [Fri, 13 Jun 2003 00:49:17 +0000 (00:49 +0000)]
Config support for the namespace mount option, Peter will fill in the
kernel details.
lmc --add mds ... --nspath /mnt/point
lmc --add ost ... --nspath /mnt/point
I added these options to local.sh, not sure if they need to be in
there permanently.
eeb [Thu, 12 Jun 2003 17:34:24 +0000 (17:34 +0000)]
* Merged in b_devel again
cvs2svn [Wed, 11 Jun 2003 18:04:57 +0000 (18:04 +0000)]
This commit was manufactured by cvs2svn to create branch
'unlabeled-1.1.4.2.6'.
eeb [Wed, 11 Jun 2003 18:04:56 +0000 (18:04 +0000)]
* Merged latest b_devel
* **** NOT TESTED ****
eeb [Wed, 11 Jun 2003 13:51:17 +0000 (13:51 +0000)]
* Single event queue per network interface.
* Pre-allocated server requests, individually posted for receives
Servers now configured by XXX_NREQ (# requests) and XXX_MAXREQSIZE
(size of each replies buffer).
* Asynchronous server-side reply completion/notification. struct
ptlrpc_reply_state objects splits off state reply so request can be
reposted immediately.
* exp_outstanding_reply now a list of replies
* bulk descriptors sized so bd_iov can contain all the fragments.
ptlrpc_prep_bulk_page() is now returns void (i.e. no errors; may fail
an LASSERT if the bulk desc was sized incorrectly).
* NOT tested saved locks yet; bringing up to date with b_devel first...
adilger [Tue, 10 Jun 2003 23:00:53 +0000 (23:00 +0000)]
Add some comments to the iopen patches to make intended behaviour clear.
pschwan [Tue, 10 Jun 2003 10:48:27 +0000 (10:48 +0000)]
b_devel -> b_orphan merge for
20030610
rread [Sat, 7 Jun 2003 00:07:44 +0000 (00:07 +0000)]
beginning of multinet in lconf
- start the acceptor on gateways
- use mynid for all network types
- many routing code cleanups
- experimental support for cluster id
rread [Thu, 5 Jun 2003 04:12:30 +0000 (04:12 +0000)]
merge from b_devel -> b_ad
DEVEL_AD_PARENT_20030526 -> DEVEL_AD_PARENT_20030605
Large, squirrelly merge. It mounts, and acceptance_small.sh finishes
the test and then fails in cleanup, just like b_devel.
adilger [Wed, 4 Jun 2003 21:34:43 +0000 (21:34 +0000)]
Merge from b_devel -> b_orphan
DEVEL_ORPHAN_UPDATE_PARENT_20030528 -> DEVEL_ORPHAN_UPDATE_PARENT_20030604
This is a non-functional commit, with osc/osc_request.c not compiling because
it is missing the "llog_commit_master" data struct, and does not have any
commit threads started yet. I'm beginning to think the right thing to do
here is to create a single global llog_commit_master struct for the OSC, and
start a single thread at module setup time, and clean it all up at module
unload time, but I could be convinced otherwise.
wangdi [Wed, 4 Jun 2003 01:59:10 +0000 (01:59 +0000)]
fix the a flaw in iopen-2.4.18.patch
adilger [Tue, 3 Jun 2003 23:27:02 +0000 (23:27 +0000)]
Remove bogus extra arg from delete thread warning message.
adilger [Tue, 3 Jun 2003 22:52:07 +0000 (22:52 +0000)]
Include the EXT3_MOUNT_ASYNCDEL addition from ext3-delete-thread into the
context of the iopen patch (as it is in the 2.4.20 version), otherwise the
iopen patch fails to apply.
adilger [Tue, 3 Jun 2003 07:38:09 +0000 (07:38 +0000)]
Add THREAD_NAME compat macro, to remove duplication of UML thread-name stuff.
Make list_move() available if not defined (based on presence of
list_for_each_entry(), which is also defined in 2.5).
wangdi [Mon, 2 Jun 2003 09:07:28 +0000 (09:07 +0000)]
add jdike-2.5.69-uml.config
wangdi [Sat, 31 May 2003 09:53:36 +0000 (09:53 +0000)]
add check_get_page in i386 arch
adilger [Wed, 28 May 2003 18:28:35 +0000 (18:28 +0000)]
More from b_devel->b_orphan merge (I guess I didn't commit anything last night).
DEVEL_ORPHAN_UPDATE_PARENT_20030522->DEVEL_ORPHAN_UPDATE_PARENT_20030528
adilger [Wed, 28 May 2003 17:32:11 +0000 (17:32 +0000)]
More from b_devel->b_orphan merge.
DEVEL_ORPHAN_UPDATE_PARENT_20030522->DEVEL_ORPHAN_UPDATE_PARENT_20030528
adilger [Wed, 28 May 2003 17:19:20 +0000 (17:19 +0000)]
More from b_devel->b_orphan merge.
DEVEL_ORPHAN_UPDATE_PARENT_20030522->DEVEL_ORPHAN_UPDATE_PARENT_20030528
adilger [Wed, 28 May 2003 08:50:45 +0000 (08:50 +0000)]
Merge b_devel->b_orphan:
DEVEL_ORPHAN_UPDATE_PARENT_20030522->DEVEL_ORPHAN_UPDATE_PARENT_20030528
meijia [Wed, 28 May 2003 06:58:55 +0000 (06:58 +0000)]
[TCPNAL]: make tcpnal_recv() could handle multi-iovectors.
meijia [Wed, 28 May 2003 03:05:09 +0000 (03:05 +0000)]
[TCPNAL]: make tcpnal_send() would send large buffers array.
adilger [Tue, 27 May 2003 20:59:13 +0000 (20:59 +0000)]
Minor change to noread patch (was missing fix from not-yet-working 2.5 version).
zab [Tue, 27 May 2003 01:25:16 +0000 (01:25 +0000)]
get b_ad compiling after the b_devel merge.
- move the possibly broken waitfor_one_page to one place
- o_mtime is seconds, that's what ll_ts2u64 returns. s/%Lu/LPU64/
- get rid of client_tgt2uuid2obd copies, export it from obdclass
- move client_obd_{setup,cleanup} into obdclass, export, kill dupes
- remove empty ptlrpc_lib
- get rid of random sanobd dupe in lov
- hoist ll_alloc_brw out of a kernel version ifdef
- no, seriously, unsigned long flags.
pschwan [Tue, 27 May 2003 00:43:11 +0000 (00:43 +0000)]
- Commit wangdi's configure macro for detecting 2.4.20-rh (with backported 2.5
structure changes)
- move the signal compat macros to a new portals_compat25.h
- use them everywhere for 2.5.x and 2.4.20-rh
zab [Mon, 26 May 2003 22:50:20 +0000 (22:50 +0000)]
- bring recent b_devel changes into b_ad, only minor corrections needed. the
only significant of which was making sure changes in super.c made it to
b_ad's llite_lib.c
wangdi [Mon, 26 May 2003 11:04:05 +0000 (11:04 +0000)]
fix portals/archdep.m4
pschwan [Mon, 26 May 2003 08:18:18 +0000 (08:18 +0000)]
b=1173
Remove all files from lib/, put the functions into various other
modules as appropriate, and remove the bits to create the symlinks.
Tested in UML and lightly on dev. Builds, inserts, mounts.
adilger [Sun, 25 May 2003 18:17:54 +0000 (18:17 +0000)]
Mostly working full cookie create-send-cancel cycle for MDS unlinks.
MDS now establishes a connection (LOV only, getting UUID from LOVDESC pending
bug 1280 resolution).
We cannot clean up as a result of this (MDS->LOV connection). Might be
possible to fix by changing cleanup order so we do llite/mdc/mds cleanup
before lov/osc/ost cleanup.
meijia [Sat, 24 May 2003 08:49:12 +0000 (08:49 +0000)]
fix for build portals in user space
eeb [Fri, 23 May 2003 22:04:03 +0000 (22:04 +0000)]
* Added portals + lustre wire constant and struct size/offset assertions
* Added portals + lustre wirecheck programs to generate these assertions
on the reference platform.
meijia [Fri, 23 May 2003 10:01:12 +0000 (10:01 +0000)]
merge b_devel to b_eq since there are problems with old b_devel
DEVEL_EQ_UPDATE_PARENT_20030523 -> DEVEL_EQ_UPDATE_CHILD_20030523
adilger [Fri, 23 May 2003 07:03:13 +0000 (07:03 +0000)]
Merge b_devel -> b_orphan.
DEVEL_ORPHAN_UPDATE_PARENT_20030522 -> DEVEL_ORPHAN_UPDATE_PARENT_20030522
meijia [Wed, 21 May 2003 15:43:09 +0000 (15:43 +0000)]
merge b_devel to b_eq:
DEVEL_EQ_UPDATE_PARENT_20030521 -> DEVEL_EQ_UPDATE_CHILD_20030521