Whamcloud - gitweb
fs/lustre-release.git
21 years ago- change version of HEAD to 0.5
rread [Fri, 12 Jul 2002 08:53:35 +0000 (08:53 +0000)]
- change version of HEAD to 0.5

21 years ago- fix typos
rread [Fri, 12 Jul 2002 08:53:03 +0000 (08:53 +0000)]
- fix typos

21 years ago- add module aliases to /modules.conf
rread [Fri, 12 Jul 2002 07:01:47 +0000 (07:01 +0000)]
- add module aliases to /modules.conf

21 years ago- fix rpm build
rread [Fri, 12 Jul 2002 07:01:01 +0000 (07:01 +0000)]
- fix rpm build

21 years ago- fixes from santa fe
rread [Fri, 12 Jul 2002 06:59:55 +0000 (06:59 +0000)]
- fixes from santa fe

21 years ago- run config in production
rread [Fri, 12 Jul 2002 03:02:27 +0000 (03:02 +0000)]
- run config in production

21 years agoRemove hard-coded mountpoint from runtests script.
adilger [Thu, 11 Jul 2002 22:55:00 +0000 (22:55 +0000)]
Remove hard-coded mountpoint from runtests script.

21 years agoFix statfs for new ptlrpc interface in head.
adilger [Thu, 11 Jul 2002 22:53:34 +0000 (22:53 +0000)]
Fix statfs for new ptlrpc interface in head.

21 years agoRemove sgid #warning. This is already done in the filesystem and the MDS.
adilger [Thu, 11 Jul 2002 22:52:24 +0000 (22:52 +0000)]
Remove sgid #warning.  This is already done in the filesystem and the MDS.
It does not really make sense to do it in the MDC.

21 years agoI broke llcleanup earlier. This fixes it.
pschwan [Thu, 11 Jul 2002 22:43:00 +0000 (22:43 +0000)]
I broke llcleanup earlier.  This fixes it.

21 years ago- The server side of the DLM wasn't always handling the invalid handle case
pschwan [Thu, 11 Jul 2002 15:37:29 +0000 (15:37 +0000)]
- The server side of the DLM wasn't always handling the invalid handle case
very gracefully.  Since there are races that make invalid handles not hugely
uncommon, that's fixed now.
- Even on the client side, we could get cancelled between ldlm_lock_decref() and
ldlm_lock_cancel(), thus passing an invalid handle into cancel().  This is
handled now.
- Fixed another leaked request.
- Fixed an IT_SETATTR bug, even though we don't currently exercise that code
path.
- Fixed common.sh for the multiple mount cleanup case
- llcleanup.sh was unloading things in the wrong order (MDS is pinned by LDLM
and needs to be unloaded later in the process).  Fixed.

21 years agomdc_connect needs to use a level of LUSTRE_CONN_NEW.
shaver [Wed, 10 Jul 2002 20:31:23 +0000 (20:31 +0000)]
mdc_connect needs to use a level of LUSTRE_CONN_NEW.
(I broke mounting on the trunk before, this fixes it.)

21 years agoDefault to an rq_level of LUSTRE_CONN_FULL.
shaver [Mon, 8 Jul 2002 19:23:02 +0000 (19:23 +0000)]
Default to an rq_level of LUSTRE_CONN_FULL.
Use l_killable_pending in place of explicit lists of tests.

21 years ago- updates for rpm
rread [Sat, 6 Jul 2002 23:00:46 +0000 (23:00 +0000)]
- updates for rpm

21 years agoChange TCP port number.
adilger [Fri, 5 Jul 2002 23:26:23 +0000 (23:26 +0000)]
Change TCP port number.

21 years agoChange TCP acceptor port number to be 2432, an unused Coda port. You need
adilger [Fri, 5 Jul 2002 23:21:40 +0000 (23:21 +0000)]
Change TCP acceptor port number to be 2432, an unused Coda port.  You need
to update all of your <myname>.cfg files (or XML if you use it) to use the
new port number if you want to use tcpdump, and to maintain consistency
across scripts.

21 years agoMove the on-wire stucts/definitions into lustre_idl.h, for the CVS head.
uid34591 [Fri, 5 Jul 2002 22:50:42 +0000 (22:50 +0000)]
Move the on-wire stucts/definitions into lustre_idl.h, for the CVS head.

21 years agoFix "tags" target if portals is a symlink
uid34591 [Fri, 5 Jul 2002 22:39:03 +0000 (22:39 +0000)]
Fix "tags" target if portals is a symlink

21 years agoInclude headers explicitly that were previously included from lustre_net.h.
uid34591 [Fri, 5 Jul 2002 21:53:27 +0000 (21:53 +0000)]
Include headers explicitly that were previously included from lustre_net.h.

21 years agoFix osc Makefile to link ll_pack.c (module osc.o should not depend on ost.o).
uid34591 [Fri, 5 Jul 2002 21:23:03 +0000 (21:23 +0000)]
Fix osc Makefile to link ll_pack.c (module osc.o should not depend on ost.o).
Remove old cruft.

21 years ago- Minor connection change to pass uuid's in.
uid23919 [Fri, 5 Jul 2002 21:20:11 +0000 (21:20 +0000)]
- Minor connection change to pass uuid's in.

21 years agoRemoved the deprecated forced-localhost DLM bits; they're confusing the
pschwan [Fri, 5 Jul 2002 20:55:57 +0000 (20:55 +0000)]
Removed the deprecated forced-localhost DLM bits; they're confusing the
portals routing code and making me sad.

21 years ago- use UUIDs for mounting
braam [Fri, 5 Jul 2002 20:53:05 +0000 (20:53 +0000)]
- use UUIDs for mounting

21 years ago- change PACKAGE to lustre
rread [Fri, 5 Jul 2002 20:05:27 +0000 (20:05 +0000)]
- change PACKAGE to lustre
- update spec

21 years ago- fix documentation updates. The master document is now called
braam [Fri, 5 Jul 2002 19:23:33 +0000 (19:23 +0000)]
- fix documentation updates.  The master document is now called
  lustre.pdf.    Still need the CVS Tag or HEAD-version if not
  present....
- Minor edits to references etc to get compilation going.

21 years agoItems of note:
pschwan [Fri, 5 Jul 2002 18:56:59 +0000 (18:56 +0000)]
Items of note:
* Fixes rename
* Allows for symlinks larger than LL_INLINESZ (of course, intent-based symlink
  code isn't written yet)

Items of lesser note, although their importance is not diminished by the mere
inclusion in this seemingly unimportant list:
- We were ignoring reint errors in ldlm_intent_policy.  No more!
- ll_dir_readpage error "handling" led directly to a panic.  Hopefully that's
  fixed now.

21 years ago- Make now correctly rebuilds
braam [Fri, 5 Jul 2002 18:48:01 +0000 (18:48 +0000)]
- Make now correctly rebuilds
- The version was given an "author tag" to make it visible

21 years agoPrevent C-c and C-z from locking us up, and make most of our waits
shaver [Fri, 5 Jul 2002 18:03:59 +0000 (18:03 +0000)]
Prevent C-c and C-z from locking us up, and make most of our waits
uninterruptible.  We'll move to a more robust system shortly, but
this will make for a usable testing environment in the interim.

21 years agoChange module data to use "info@clusterfs.com" email per Peter.
adilger [Fri, 5 Jul 2002 10:38:40 +0000 (10:38 +0000)]
Change module data to use "info@clusterfs.com" email per Peter.

21 years agoAdd statfs fixups to head.
adilger [Fri, 5 Jul 2002 10:17:21 +0000 (10:17 +0000)]
Add statfs fixups to head.

21 years agoAdd closing quote so that obdecho compiles.
adilger [Fri, 5 Jul 2002 08:02:12 +0000 (08:02 +0000)]
Add closing quote so that obdecho compiles.

21 years agoAdd statfs support for proper blocks count to head.
adilger [Fri, 5 Jul 2002 07:41:44 +0000 (07:41 +0000)]
Add statfs support for proper blocks count to head.
Includes fix for type == 0 (truncate) problem.
Removes ost_get_info call - it is not used and causes confusion.

21 years agoPass 64-bit object numbers wherever possible, instead of truncating to 32-bit.
adilger [Fri, 5 Jul 2002 07:37:46 +0000 (07:37 +0000)]
Pass 64-bit object numbers wherever possible, instead of truncating to 32-bit.

21 years agotidy runfailure-ost a touch
shaver [Fri, 5 Jul 2002 01:35:04 +0000 (01:35 +0000)]
tidy runfailure-ost a touch

21 years agothink i got it this time
rread [Fri, 5 Jul 2002 01:17:41 +0000 (01:17 +0000)]
think i got it this time

21 years ago-again
rread [Fri, 5 Jul 2002 01:15:37 +0000 (01:15 +0000)]
-again

21 years ago- RCS version experimet
rread [Fri, 5 Jul 2002 01:15:10 +0000 (01:15 +0000)]
- RCS version experimet

21 years ago- fixes an extra intent_release in open_namei
braam [Thu, 4 Jul 2002 23:26:32 +0000 (23:26 +0000)]
- fixes an extra intent_release in open_namei

21 years ago- install setup scripts and config
rread [Thu, 4 Jul 2002 22:26:55 +0000 (22:26 +0000)]
- install setup scripts and config

21 years ago- install to /usr/sbin
rread [Thu, 4 Jul 2002 21:34:05 +0000 (21:34 +0000)]
- install to /usr/sbin

21 years ago - updated to new dtd
rread [Thu, 4 Jul 2002 21:23:19 +0000 (21:23 +0000)]
 - updated to new dtd
 - network now in <node>
 - new service_id added to profile for network
 - this will probably not work with obdctl anymore, new tools on the way

21 years ago- Quiet the FIXMEs from CERRORs to CDEBUGs, because they were reducing LLNL
pschwan [Thu, 4 Jul 2002 19:58:38 +0000 (19:58 +0000)]
- Quiet the FIXMEs from CERRORs to CDEBUGs, because they were reducing LLNL
consoles to a crawl
- more DLM debug infrastructure, mostly preventative.
- check for the forgotten ptlrpc_abort() case in reply_in_callback
- cleanup the intent-test and runfailure-mds

21 years agoDump the lock before LBUG()ing in ldlm_lock_destroy, for easier debugging of
shaver [Thu, 4 Jul 2002 14:45:38 +0000 (14:45 +0000)]
Dump the lock before LBUG()ing in ldlm_lock_destroy, for easier debugging of
mismatched refcount issues.

21 years ago- add copyright, license, and attribution
rread [Thu, 4 Jul 2002 01:54:01 +0000 (01:54 +0000)]
- add copyright, license, and attribution

21 years ago- fix lookup bug
rread [Thu, 4 Jul 2002 01:37:09 +0000 (01:37 +0000)]
- fix lookup bug

21 years agoinitial version
rread [Thu, 4 Jul 2002 01:00:50 +0000 (01:00 +0000)]
initial version

21 years ago- Added lctl.c which is basically ptlctl/obdctl/debugctl all
brian [Thu, 4 Jul 2002 00:12:31 +0000 (00:12 +0000)]
- Added lctl.c which is basically ptlctl/obdctl/debugctl all
  frankensteined together.  It should be possible to configure
  all of portals/lustre through it.
- network.c contains the portals bits
  device.c contains the device bits
  debug.c contains the debug bits
- Some functionality has been consolidated when it was redundant,
  there's still more of that to do!
- The thread handling that was in obdctl is broken for a little bit.
- All the original *ctl tools are uneffected.
- Several commands have been slightly renamed do to name collisions,
  for instance the obdctl list command is now device_list, so as
  not to conflict with debug_list or route_list.  It'll be pretty
  obvious.

21 years ago- Added lctl.c which is basically ptlctl/obdctl/debugctl all
behlendo [Thu, 4 Jul 2002 00:12:30 +0000 (00:12 +0000)]
- Added lctl.c which is basically ptlctl/obdctl/debugctl all
  frankensteined together.  It should be possible to configure
  all of portals/lustre through it.
- network.c contains the portals bits
  device.c contains the device bits
  debug.c contains the debug bits
- Some functionality has been consolidated when it was redundant,
  there's still more of that to do!
- The thread handling that was in obdctl is broken for a little bit.
- All the original *ctl tools are uneffected.
- Several commands have been slightly renamed do to name collisions,
  for instance the obdctl list command is now device_list, so as
  not to conflict with debug_list or route_list.  It'll be pretty
  obvious.

21 years ago- first additions for rename.
braam [Wed, 3 Jul 2002 23:19:32 +0000 (23:19 +0000)]
- first additions for rename.
- NOTE: it appears the intent regression is broken at present (freeing
  a ref'd lock)

21 years agoFix ordering of module unload in head.
adilger [Wed, 3 Jul 2002 20:46:02 +0000 (20:46 +0000)]
Fix ordering of module unload in head.

21 years agoclean up old patches
braam [Wed, 3 Jul 2002 18:44:15 +0000 (18:44 +0000)]
clean up old patches

21 years agopatch for Jim's 2.4.18-chaos5
braam [Wed, 3 Jul 2002 18:41:36 +0000 (18:41 +0000)]
patch for Jim's 2.4.18-chaos5

21 years agoAdd note for 0.4.2
braam [Wed, 3 Jul 2002 18:39:04 +0000 (18:39 +0000)]
Add note for 0.4.2

21 years ago- update the patch with yesterday's fix
braam [Wed, 3 Jul 2002 18:29:20 +0000 (18:29 +0000)]
- update the patch with yesterday's fix

21 years agoMinor fixes to get setup to work again. Fix some brainos in obdclass to
braam [Tue, 2 Jul 2002 23:17:36 +0000 (23:17 +0000)]
Minor fixes to get setup to work again. Fix some brainos in obdclass to
provide robustness.

21 years ago- initial version
rread [Tue, 2 Jul 2002 20:27:25 +0000 (20:27 +0000)]
- initial version

21 years ago- add missing unlocks.
braam [Tue, 2 Jul 2002 08:41:34 +0000 (08:41 +0000)]
- add missing unlocks.

21 years ago - create device
rread [Tue, 2 Jul 2002 02:41:17 +0000 (02:41 +0000)]
 - create device
 - make note of need to modify /etc/modules.conf

21 years ago- experiment: attempt to add the cvs version to formatted documentation.
rread [Tue, 2 Jul 2002 01:36:41 +0000 (01:36 +0000)]
- experiment: attempt to add the cvs version to formatted documentation.
Added a rule to create 'lustre.lyx' which is a copy of master.lyx,
modified to include the cvs version.

Normally, "make lustre.pdf" should produce a pdf with the proper version on
the cover page, but on my system the whole Date tag is not showing up. Not sure why.

21 years ago- remove unneeded parameter to obd_create/destroy
rread [Tue, 2 Jul 2002 01:32:42 +0000 (01:32 +0000)]
- remove unneeded parameter to obd_create/destroy

21 years ago-Minor fixes to ensure the DTD is valid. net-local.xml passes validation
behlendo [Mon, 1 Jul 2002 23:45:32 +0000 (23:45 +0000)]
-Minor fixes to ensure the DTD is valid.  net-local.xml passes validation
 with the provided validation tool.

 >xmllint --valid --noout net-local.xml

21 years ago- other half of the buffer fix.
braam [Mon, 1 Jul 2002 23:24:26 +0000 (23:24 +0000)]
- other half of the buffer fix.

21 years ago- fix for variable size buffer handling.
braam [Mon, 1 Jul 2002 22:55:42 +0000 (22:55 +0000)]
- fix for variable size buffer handling.

21 years ago- more LDLM refcount locking infrastructure
pschwan [Mon, 1 Jul 2002 21:23:34 +0000 (21:23 +0000)]
- more LDLM refcount locking infrastructure
- fixed the error handling paths of unlink and rmdir, so that unlinking a
non-existant file returns an error
- fixed the DLM bug that caused the perl script to hang (there's a new bug now
  that it triggers under heavier load that causes it to crash)
- fixed a second crashing bug that occurred if a blocked lock gets completed
  and then cancelled all before the server-side enqueue finishes (amazing but
  true)
- osc_open() now returns the actual rc instead of 0
- fixed a race condition in ptlrpc/service.c related to thread shutdown (nice
  catch, Mike)

21 years agoFixed some path-slash badness in common.sh, and added embryonic runfailure-ost.
shaver [Mon, 1 Jul 2002 19:34:20 +0000 (19:34 +0000)]
Fixed some path-slash badness in common.sh, and added embryonic runfailure-ost.

21 years agoA Perl script to hammer concurrently at multiple mount points
pschwan [Mon, 1 Jul 2002 15:26:52 +0000 (15:26 +0000)]
A Perl script to hammer concurrently at multiple mount points

21 years ago- much of the striping configuration managment and setup builds up
braam [Mon, 1 Jul 2002 07:30:40 +0000 (07:30 +0000)]
- much of the striping configuration managment and setup builds up
  this patch
- the mds has a new lovconfig command to tell it the UUID's and
  default striping pattern of the targets it needs to use (these are
  the UUID's of the OSC's typically).
- To make this scalable I changed some of the memory management in the
  class ioctl handling
- The LOV device has a trivial attach method and setup only tells it
  what MDC to use to get its information, by giving it the MDC-UUID.
- As discussed before, the MDS really provides the persistent storage
  for the LOV, what little it needs.  So during the obd_connect call
  for the object storage target (which is made from read_super) the
  storage target learns how it is striped and then connects to all the
  targets.
- We are in need of better configuration scripts for this stuff and
  tomorrow we will push the XML configurations a little further.
- Updated the documentation
- Began to cleanup the /proc/lustre/ stuff -- have some neat ideas
  about that and SNMP now.

21 years ago- updated LDLM_DEBUG to give more refcount info
pschwan [Mon, 1 Jul 2002 06:35:51 +0000 (06:35 +0000)]
- updated LDLM_DEBUG to give more refcount info
- made fixme a macro, so that it shows us where it's called from
- fixed a DLM deadlock (unbalanced l_lock)
- fixed the refcount bug in ldlm_lock_decref
- fixed the refcount bug in ldlm_cli_enqueue in the failed/aborted case
- the lock slab cleans up now!
- fixed the ``connection foo has refcount -61'' bug
- found, but have not yet fixed, a subtle ctrl-c-during-aborted-ldlm-enqueue
bug that can be triggered if you abort the hanging Perl test at _just_ the right
time.
- fixed request leaks in osc_connect and mdc_connect
- we create an import in osc_connect but never use or free it -- I didn't
remove this code, assuming it was going to be used soon?

21 years ago- same for mdc
braam [Sun, 30 Jun 2002 20:34:25 +0000 (20:34 +0000)]
- same for mdc

21 years agofix a blunder in osc_connect that frees the connection before
braam [Sun, 30 Jun 2002 20:33:59 +0000 (20:33 +0000)]
fix a blunder in osc_connect that frees the connection before
returning success.

21 years agoFix a few compiler warnings.
braam [Sun, 30 Jun 2002 08:03:40 +0000 (08:03 +0000)]
Fix a few compiler warnings.

21 years ago- except for fixing the segfault in the documentation build, this is
braam [Sun, 30 Jun 2002 07:39:40 +0000 (07:39 +0000)]
- except for fixing the segfault in the documentation build, this is
  merely a cleanup checkin.  Functions in the obdclass system were
  given sensible names, old unused headers were removed, and stuff
  that isn't used widely was sometimes move from generic headers to
  specific subsystems.

21 years ago- Fixes memory and refcount links in the DLM
pschwan [Sun, 30 Jun 2002 03:22:06 +0000 (03:22 +0000)]
- Fixes memory and refcount links in the DLM
- Fixes a long-standing request leak
- The lock slab is not yet cleaned up correctly in some cases, but I'm on it

21 years ago- small issues
braam [Sun, 30 Jun 2002 03:19:02 +0000 (03:19 +0000)]
- small issues

- magic and versions into Lustre messages
- obdo cleanups, remove unused fields.
- small updates to object protocol documentation.

21 years agoFile I/O fix: move the lock name space to the filter itself.
braam [Sat, 29 Jun 2002 13:10:22 +0000 (13:10 +0000)]
File I/O fix: move the lock name space to the filter itself.

21 years agoMinor bug fixes to get export handles used by the DLM. I believe the
braam [Sat, 29 Jun 2002 12:53:02 +0000 (12:53 +0000)]
Minor bug fixes to get export handles used by the DLM.  I believe the
head is pretty close to where it was before in functionality, but the
cleanups can now continue further.

21 years ago- fix make dist
rread [Sat, 29 Jun 2002 00:41:48 +0000 (00:41 +0000)]
- fix make dist
- bring spec file uptodate
- add rpms rule to main Makefile.am

21 years ago- need to add include for PAGE_SIZE
rread [Sat, 29 Jun 2002 00:40:47 +0000 (00:40 +0000)]
- need to add include for PAGE_SIZE

21 years ago- shared MDS/OST connection routines
braam [Fri, 28 Jun 2002 20:05:03 +0000 (20:05 +0000)]
- shared MDS/OST connection routines

21 years ago- The same changes made to the OST are now working in the MDS
braam [Fri, 28 Jun 2002 20:04:21 +0000 (20:04 +0000)]
- The same changes made to the OST are now working in the MDS
- Significant simplification is starting to emerge now, as expected.
- Next is the locking subsystem.

- Added new terms to the glossary.

21 years ago- Refined the DTD futher: 'id' changed to 'name'; tree structure rearranged
brian [Fri, 28 Jun 2002 19:49:00 +0000 (19:49 +0000)]
- Refined the DTD futher: 'id' changed to 'name'; tree structure rearranged
  and brought more inline with the documentation; other general ongoing
  cleanup.
- DTD changes reflected in net-local.xml.
- obdctl properly uses of UUIDs during config, on going work here.

21 years ago- Refined the DTD futher: 'id' changed to 'name'; tree structure rearranged
behlendo [Fri, 28 Jun 2002 19:48:59 +0000 (19:48 +0000)]
- Refined the DTD futher: 'id' changed to 'name'; tree structure rearranged
  and brought more inline with the documentation; other general ongoing
  cleanup.
- DTD changes reflected in net-local.xml.
- obdctl properly uses of UUIDs during config, on going work here.

21 years agoAdd a new patch for the new LLNL kernel. This contains many serious changes
braam [Fri, 28 Jun 2002 06:54:02 +0000 (06:54 +0000)]
Add a new patch for the new LLNL kernel.  This contains many serious changes
from the previous one.

21 years ago- more fixes to export handling. Probably tomorrow they can be
braam [Fri, 28 Jun 2002 01:53:38 +0000 (01:53 +0000)]
- more fixes to export handling.  Probably tomorrow they can be
  completed.

21 years ago- More changes in the connection handle stuff. We are back to where
braam [Thu, 27 Jun 2002 23:15:06 +0000 (23:15 +0000)]
- More changes in the connection handle stuff.  We are back to where
  obdecho works over the net.

21 years agoDon't update your trees yet..... Brian and I are trying to get our
braam [Thu, 27 Jun 2002 21:09:07 +0000 (21:09 +0000)]
Don't update your trees yet..... Brian and I are trying to get our
work in sync.
- This is a fairly pervasive change to get away from "device
  numbers".
- Introduce connection handles everywhere (treat these as
  lustre_handles).
- On targets map UUID's to handles upon connect and use those handles
  in the future.

21 years agoAnother slightly less rough cut at XML based configuration. There
brian [Wed, 26 Jun 2002 22:09:29 +0000 (22:09 +0000)]
Another slightly less rough cut at XML based configuration.  There
is still a large reliance on the existing scripts for loopback
support, portals config, and module loading.  I'll try and resolve
some of these things after some consultation about what the best
way to do that is.

- Several minor changes were made to the DTD, we're still not
  validating against it but it should be consistent with the example.
- Simple code added for ldlm, mds, obd, ost, osc, and mdc configuration.
- Other misc support and fixes.

A local filesystem should be mountable with the net-local.xml config:

~/tests/llxmlsetup.sh net-local.xml

21 years agoAnother slightly less rough cut at XML based configuration. There
behlendo [Wed, 26 Jun 2002 22:09:28 +0000 (22:09 +0000)]
Another slightly less rough cut at XML based configuration.  There
is still a large reliance on the existing scripts for loopback
support, portals config, and module loading.  I'll try and resolve
some of these things after some consultation about what the best
way to do that is.

- Several minor changes were made to the DTD, we're still not
  validating against it but it should be consistent with the example.
- Simple code added for ldlm, mds, obd, ost, osc, and mdc configuration.
- Other misc support and fixes.

A local filesystem should be mountable with the net-local.xml config:

~/tests/llxmlsetup.sh net-local.xml

21 years agoStart maintaining the release changelog file.
braam [Wed, 26 Jun 2002 00:44:59 +0000 (00:44 +0000)]
Start maintaining the release changelog file.

21 years ago- ldlm fixes to get it to pass intent-test.sh, although it doesn't quite clean
pschwan [Tue, 25 Jun 2002 22:09:20 +0000 (22:09 +0000)]
- ldlm fixes to get it to pass intent-test.sh, although it doesn't quite clean
up correctly afterwards yet.

21 years agoMerge trivial changes from branch to head.
adilger [Tue, 25 Jun 2002 07:06:17 +0000 (07:06 +0000)]
Merge trivial changes from branch to head.
The only real change is the filter multi-page BRW failure cleanup code.

21 years ago- Cleanup of the lock callback infrastructure - not quite functional
braam [Tue, 25 Jun 2002 04:49:57 +0000 (04:49 +0000)]
- Cleanup of the lock callback infrastructure - not quite functional
  yet.

- API improvements.
- Documentation update.

21 years agoRemove obdext2.cfg as we no longer have obdext2
adilger [Mon, 24 Jun 2002 20:11:56 +0000 (20:11 +0000)]
Remove obdext2.cfg as we no longer have obdext2

21 years ago- fixes to let a client do correct lock cancellation during namespace
braam [Mon, 24 Jun 2002 18:08:00 +0000 (18:08 +0000)]
- fixes to let a client do correct lock cancellation during namespace
  free.

21 years agostatus: on ./mcreate I believe the client lock on the file and the mds
braam [Mon, 24 Jun 2002 17:22:36 +0000 (17:22 +0000)]
status: on ./mcreate I believe the client lock on the file and the mds
lock on the directory are both correctly enqueued.

- updated documentation regarding lock / resource refcounting
- removed unnecessary resource_getref before policy (thanks phil)
- local locks to addref, server locks don't
- variety of small fixes
- print lock/type names
- add one reference on lock for creation, put it away once for destroy
  [otherwise 0 ref locks can "live" and that causes the resources grief]
- small fix to change_resource

21 years agoFix o_valid and o_size for truncate through osc/ost/filterobd (on HEAD).
adilger [Mon, 24 Jun 2002 16:57:47 +0000 (16:57 +0000)]
Fix o_valid and o_size for truncate through osc/ost/filterobd (on HEAD).
This should also remove the need to explicitly set mtime/ctime in
ll_truncate(), but that has not yet been tested and causes no real harm.

21 years ago- first fixes to get things working again: current state:
braam [Mon, 24 Jun 2002 07:43:13 +0000 (07:43 +0000)]
- first fixes to get things working again: current state:
  ldlm_cli_enqueue returns a lock, but it misses one refcount.

21 years agolast compile fixes: WARNING nothing works probably.
braam [Mon, 24 Jun 2002 06:32:05 +0000 (06:32 +0000)]
last compile fixes: WARNING nothing works probably.

21 years ago- more fixes
braam [Mon, 24 Jun 2002 06:25:23 +0000 (06:25 +0000)]
- more fixes

21 years ago- mds compile fixes
braam [Mon, 24 Jun 2002 06:24:19 +0000 (06:24 +0000)]
- mds compile fixes