Whamcloud - gitweb
fs/lustre-release.git
22 years agoBug#1380 migrating xproc to Lprocfs
thantry [Wed, 11 Jun 2003 21:57:36 +0000 (21:57 +0000)]
Bug#1380 migrating xproc to Lprocfs
Bug#1377 use per-cpu lprocfs counters
obdstat will be replaced by llobdstat.pl

22 years agob_devel -> b_orphan merge for 20030610
pschwan [Tue, 10 Jun 2003 10:48:02 +0000 (10:48 +0000)]
b_devel -> b_orphan merge for 20030610

22 years agob=1228
pschwan [Fri, 6 Jun 2003 06:16:31 +0000 (06:16 +0000)]
b=1228
Add generation numbers and inode pointers to the "VFS Op" debug statements

22 years agoadd ext3-delete_thread-2.4.18.patch
wangdi [Tue, 3 Jun 2003 01:40:23 +0000 (01:40 +0000)]
add ext3-delete_thread-2.4.18.patch

22 years agoThis wholesale change breaks down llvisualize into multiple modules
thantry [Tue, 3 Jun 2003 01:20:30 +0000 (01:20 +0000)]
This wholesale change breaks down llvisualize into multiple modules
to allow easy extensibility and to reduce memory footprint when
running with large logs.

llascii.pm is the module used for timing analysis, html and graph generation
(--vfssummary option). No new features have been added, but lot of
redundant code has been removed.

llgtk.pm is the visual component for the gtktree. This has been completely
rewritten, in part due to bugs #1271 and #717. The --serverrpc switch has
(for stitching server logs into client view) been fixed. The --uml option
will allow loading of logs which are time warped at multiple places in the log.

llcommon.pm contains code that is common to both the analysis mode (--vfssummary option)
and the visual mode.

llconsts.pm contains the constants, and any global variables used.

22 years agoadd tcp-zero-copy.patch
wangdi [Thu, 29 May 2003 06:30:18 +0000 (06:30 +0000)]
add tcp-zero-copy.patch

22 years agoadd delete thread patch in chaos
wangdi [Thu, 29 May 2003 02:43:14 +0000 (02:43 +0000)]
add delete thread patch in  chaos

22 years agoExtra debugging for a situation that may no longer be possible: if we
pschwan [Wed, 28 May 2003 20:38:21 +0000 (20:38 +0000)]
Extra debugging for a situation that may no longer be possible: if we
enter ll_alloc_brw with an i_blksize of < PAGE_CACHE_SIZE, we would
allocate enough space for zero pages.  Then LASSERT soon thereafter.

The fix that Andreas made to ll_file_open might prevent this, but just
in case, if we ever get into here with llwp->max == 0, print an error
and force it to be 1.

22 years agoFix for LASSERT failure in ll_inode_getattr.
adilger [Tue, 27 May 2003 23:06:30 +0000 (23:06 +0000)]
Fix for LASSERT failure in ll_inode_getattr.
Also removes get_info("brw_size") and instead uses o_blksize from the OSC
  to set the "optimum" block size for IO (ala st_blksize from fstat).  We
  do not yet pass this from the OST to the client, but we should as soon
  as BA OSTs do.
This also removes a problem where we returned bad values for st_blksize at
  times (4096 when the inode was created, 65536 * num_osts later, instead
  of 65536 * num_stripes).

22 years agoRemove lib/simple.c functions from ldlm/ldlm_lib.c into obdclass/simple.c
adilger [Tue, 27 May 2003 21:40:58 +0000 (21:40 +0000)]
Remove lib/simple.c functions from ldlm/ldlm_lib.c into obdclass/simple.c
Remove duplicate declarations of client_obd_{setup,cleanup}() in ldlm/ldlm_lib.c
Remove duplicate declarations of client_sanobd_setup()/path2dev.

22 years agorename exports_hp.pc to exports_2.4.20.pc
wangdi [Tue, 27 May 2003 11:33:31 +0000 (11:33 +0000)]
rename exports_hp.pc to exports_2.4.20.pc

22 years agomv exports_hp.patch to exports_2.4.20.patch
wangdi [Tue, 27 May 2003 11:29:36 +0000 (11:29 +0000)]
mv exports_hp.patch to exports_2.4.20.patch

22 years agorename dev_read_only_hp to dev_read_only_2.4.20
wangdi [Tue, 27 May 2003 11:22:21 +0000 (11:22 +0000)]
rename dev_read_only_hp to dev_read_only_2.4.20

22 years agorename dev_read_only_hp.patch dev_read_only_2.4.20.patch
wangdi [Tue, 27 May 2003 11:15:17 +0000 (11:15 +0000)]
rename dev_read_only_hp.patch dev_read_only_2.4.20.patch

22 years ago- remove unused "granted_callback" and "cp_data" fields from the
pschwan [Thu, 22 May 2003 18:50:49 +0000 (18:50 +0000)]
- remove unused "granted_callback" and "cp_data" fields from the
  ldlm_lock structure and APIs
- fix warning in llite/iod.c

22 years agomerge b_ad (cleaned up a bit) into b_devel; entirely 2.5 compat fixes
pschwan [Mon, 19 May 2003 06:29:47 +0000 (06:29 +0000)]
merge b_ad (cleaned up a bit) into b_devel; entirely 2.5 compat fixes

22 years agothe chaos kernel can use the iopen-2.4.18.patch
pschwan [Mon, 19 May 2003 05:19:55 +0000 (05:19 +0000)]
the chaos kernel can use the iopen-2.4.18.patch

22 years agoland b_proto on b_devel:
pschwan [Mon, 19 May 2003 03:14:49 +0000 (03:14 +0000)]
land b_proto on b_devel:
 - major rework of the rpc and bulk APIs, lots of new assertions and cleanups
 - many rpc and bulk bugfixes, including bugs 937, 1057, 1072, 1162
 - perform brw and getattr operations to multiple OSTs in parallel
 - all bulks are tied to exactly one request; multiple requests can be
   tied to a single request_set

22 years agolanding b_open onto b_devel:
pschwan [Sun, 18 May 2003 16:39:21 +0000 (16:39 +0000)]
landing b_open onto b_devel:
 - kernel patches for bug 1203
 - fixes directory open()s
 - re-enables metadata caching on the MDS
 - fixes bug in ext3_noread kernel patch
 - fixes some symbol export problems in other kernels
 - fixes compilation problems when running on Red Hat 2.4.20 kernels

If you update past this merge, you will require kernel version 18

22 years agoThe regex for hostname now contains the fully qualified hostname.
thantry [Mon, 12 May 2003 17:32:47 +0000 (17:32 +0000)]
The regex for hostname now contains the fully qualified hostname.
So, changing the parser to accomodate the change

22 years agoHowever did I miss this ?
thantry [Fri, 9 May 2003 00:08:35 +0000 (00:08 +0000)]
However did I miss this ?

22 years agoMissed commits from b_devel merge.
adilger [Mon, 5 May 2003 23:07:51 +0000 (23:07 +0000)]
Missed commits from b_devel merge.

22 years agopartial merge of b_open into b_devel:
pschwan [Mon, 5 May 2003 04:30:23 +0000 (04:30 +0000)]
partial merge of b_open into b_devel:
 - Lustre kernel patch version 16
 - extN no longer built; ext3 patched as part of the kernel patch
 - related changes to lconf/lmc, to load and mkfs ext3 instead of extN now

22 years agoThis file keeps getting lost it seems.
adilger [Fri, 2 May 2003 23:24:13 +0000 (23:24 +0000)]
This file keeps getting lost it seems.

22 years agoObvious issue with parser in the absence of the start and
thantry [Tue, 29 Apr 2003 03:56:41 +0000 (03:56 +0000)]
Obvious issue with parser in the absence of the start and
the end line switch. Fixed the same.

22 years agob=1178
pschwan [Sun, 27 Apr 2003 21:52:52 +0000 (21:52 +0000)]
b=1178
- Preparations for removing unused "addr" from lustre_handle
- remove ioc_addr from kernel/userspace communications

22 years agoAdding the capability for sectional views into visualize.
thantry [Thu, 24 Apr 2003 23:40:20 +0000 (23:40 +0000)]
Adding the capability for sectional views into visualize.
This adds the switches --s <start-line> and --e <end-line> to enable
looking at events of interests in really large logs. Applies for both
analysis and visualization

22 years agoUpdated llparser for bug #1167. In addition, it also skips all processes
thantry [Thu, 24 Apr 2003 00:58:10 +0000 (00:58 +0000)]
Updated llparser for bug #1167. In addition, it also skips all processes
that do not perform RPCs summarizing them to stderr

22 years agoDon't do nbytes calculation when not using VFSTRACE debugging.
adilger [Wed, 23 Apr 2003 09:02:49 +0000 (09:02 +0000)]
Don't do nbytes calculation when not using VFSTRACE debugging.

22 years agoFixed a logical error which was causing the first RPC to be skipped
thantry [Wed, 23 Apr 2003 01:16:29 +0000 (01:16 +0000)]
Fixed a logical error which was causing the first RPC to be skipped
when creating HTML links. A more user-friendly output format

22 years ago- fix a lost fix for exec.c
braam [Tue, 22 Apr 2003 18:25:11 +0000 (18:25 +0000)]
- fix a lost fix for exec.c

22 years ago- make a few debugging cases less chatty
zab [Fri, 18 Apr 2003 18:27:08 +0000 (18:27 +0000)]
- make a few debugging cases less chatty
- get a PR lock to ->eof in ll_file_read to purge extending writes and
  make them visible to our post-enqueue getattr

22 years ago* Incorporates Zach's feedback on bug #783 to change the CDEBUG
thantry [Thu, 17 Apr 2003 02:08:49 +0000 (02:08 +0000)]
* Incorporates Zach's feedback on bug #783 to change the CDEBUG
  to represent actual bytes written (for partial pages)
* Updated llparser.pm to display html links only when all RPC lines
  for a transaction are present

22 years ago* Updated visualize that can deal with wrapped logs, and very large logs
thantry [Wed, 16 Apr 2003 21:37:24 +0000 (21:37 +0000)]
* Updated visualize that can deal with wrapped logs, and very large logs
  (~50M).
* Broke out the visual part of visualize into a separate module
* Updated the concurrency view

22 years agoFix inode->i_blksize (for fstat st_blksize and hopefully improved cp
adilger [Fri, 11 Apr 2003 04:55:39 +0000 (04:55 +0000)]
Fix inode->i_blksize (for fstat st_blksize and hopefully improved cp
performance) so that it is the maximum vector write size (num_stripes *
max_iov_size) as told by obd_get_info("brw_size").

Also change the obd_get_info() API to fix some minor abuses (passing part
of the key in via val, returning data via the return code).  Also remove
any pretense of obd_get_info() allocating memory for the caller, following
normal portals caller-allocates behaviour.  Make inputs and returns be
fixed size, so they can be network-encoded across different arches.

22 years agoUpdated version of llvisualize. Some issues remain with time-wrapped
thantry [Thu, 10 Apr 2003 21:54:17 +0000 (21:54 +0000)]
Updated version of llvisualize. Some issues remain with time-wrapped
logs. Tested for the --vfssummary and --htm <htmfilename> option
To invoke
   ./llvisualize --vfssummary --htm htm1 <cdebug-log-files>
Will result in an ASCII text output that provides Marker/VFS/RPC timing
summary. It will also produce an easily navigatable main HTML file "htm1.html"
and graph files (one for each log file) that provide a visual representation
of concurrency in the cluster.

22 years agoBug fix for Bug #783. Reviewed by Phil.
thantry [Thu, 10 Apr 2003 17:35:55 +0000 (17:35 +0000)]
Bug fix for Bug #783. Reviewed by Phil.
Needed for llvisualize functionality.

22 years agoThis gives a good first shot at fixing the patch management issues:
braam [Wed, 9 Apr 2003 10:25:15 +0000 (10:25 +0000)]
This gives a good first shot at fixing the patch management issues:

 - we followed the instructions that are being posted on the wiki
   and wich were discussed this morning

 - patches are now fairly easy to create and update

22 years agooops, this has been sitting here for a few days now.
zab [Tue, 8 Apr 2003 20:52:28 +0000 (20:52 +0000)]
oops, this has been sitting here for a few days now.
- add some page cache stats to ll_sb_info and export them with seq_file so that
  a shell spinning during an io run can give us a nice picture of what the
  client is doing.

22 years agoIt's just _my_ UML, I guess. sigh. I'll update or something, and try
pschwan [Fri, 4 Apr 2003 16:55:46 +0000 (16:55 +0000)]
It's just _my_ UML, I guess.  sigh.  I'll update or something, and try
again.

22 years ago- squelch apparently-harmless "STRANGE intent release" message
pschwan [Fri, 4 Apr 2003 04:36:58 +0000 (04:36 +0000)]
- squelch apparently-harmless "STRANGE intent release" message

- Zach's recent commit started using GPLONLY symbols, which my UML
can't resolve without a dorky preprocessor hack.  Oh well.

- Print the remote NID in timeout and "cancel out of sync" messages,
so they're useful

22 years agob=957
pschwan [Wed, 2 Apr 2003 03:56:06 +0000 (03:56 +0000)]
b=957
- Infrastructure for the pinger thread; needs multi-rpc and some refinements
- Add last_request_time to obd_export

- Fixed export leak in lov_brw success case
- test22 rears its head again: tar gets pissy if it can't access $PWD,
  which is common for homedirs at LLNL.  cd to /tmp before running tar.

Merged many of the changes from b_proto:
- multi-rpc infrastructure, but not the key bits
- removed some bitfield flags in favour of individual struct members
- merged obd_brw_write and obd_brw_read into a single obd_brw_internal
- move the Elan-related page wrangling from the IOD into the OSC

22 years agofix up file size and cache mechanisms to account for the disconnect between
zab [Wed, 2 Apr 2003 00:03:16 +0000 (00:03 +0000)]
fix up file size and cache mechanisms to account for the disconnect between
acquiring a file region extent and getting 0..N callbacks for obd locks that
the acquiry generated.

- truncate_inode_pages(,0) could race with dirtying and lose data.  when a lock
  is canceled we now purge/invalidate the (striped) pages that were covered by
  the (lov->obd) lock
- remove the getattr flag business and instead do getattrs around every lock
  acquiry.  this brings us closer to how things will behave when we return
  file size with locks, but it will slow things down until then.
- maintain an rbtree of extents which record the page offsets that are dirty
  in the page cache.  for now it helps us decide if we should trust getattr.
  in the future it will help purging and the write space reservation.
- always round extents to page size in osc_enqueue
- while I'm in here, get rid of the ISBLK stuff in _file_write and let
  generic_file_write set ppos to i_size under the i_sem

this has passed tests under a large lov with /tmp backing, but do keep an eye
out for data/size inconsistency problems.

22 years agoAdd Zach's patch to export truncate_complete_page(); still needs an
pschwan [Mon, 24 Mar 2003 17:41:09 +0000 (17:41 +0000)]
Add Zach's patch to export truncate_complete_page(); still needs an
hp-pnnl and vanilla equivalent

22 years agoadding error handling in runas.c in case the malloc() fails
pengzhao [Tue, 18 Mar 2003 05:22:39 +0000 (05:22 +0000)]
adding error handling  in runas.c in case the malloc() fails

22 years agoThis commit was manufactured by cvs2svn to create branch 'unlabeled-1.2.2'.
cvs2svn [Mon, 17 Mar 2003 20:48:25 +0000 (20:48 +0000)]
This commit was manufactured by cvs2svn to create branch 'unlabeled-1.2.2'.

22 years agoSeems I missed the cutoff for the old b_devel branch, and this was lost.
adilger [Mon, 17 Mar 2003 20:48:24 +0000 (20:48 +0000)]
Seems I missed the cutoff for the old b_devel branch, and this was lost.

22 years agoQuiet a couple of ia64 compile warnings.
adilger [Mon, 17 Mar 2003 20:23:42 +0000 (20:23 +0000)]
Quiet a couple of ia64 compile warnings.

22 years agofile vanilla-2.5.63 was initially added on branch b_devel.
amrutjoshi [Sat, 15 Mar 2003 16:20:05 +0000 (16:20 +0000)]
file vanilla-2.5.63 was initially added on branch b_devel.

22 years agofile lustre-2.5.63.pc was initially added on branch b_devel.
amrutjoshi [Sat, 15 Mar 2003 16:20:04 +0000 (16:20 +0000)]
file lustre-2.5.63.pc was initially added on branch b_devel.

22 years agofile lustre-2.5.63.patch was initially added on branch b_devel.
amrutjoshi [Sat, 15 Mar 2003 16:20:03 +0000 (16:20 +0000)]
file lustre-2.5.63.patch was initially added on branch b_devel.

22 years agofile extN.patch-2.5.63 was initially added on branch b_devel.
amrutjoshi [Sat, 15 Mar 2003 16:20:02 +0000 (16:20 +0000)]
file extN.patch-2.5.63 was initially added on branch b_devel.

22 years agofile mkdirdeep.c was initially added on branch b_devel.
thantry [Sat, 15 Mar 2003 02:36:04 +0000 (02:36 +0000)]
file mkdirdeep.c was initially added on branch b_devel.

22 years agoBugzilla 895
thantry [Sat, 15 Mar 2003 02:35:56 +0000 (02:35 +0000)]
Bugzilla 895

22 years agofile extN-delete_thread.diff was initially added on branch b_devel.
adilger [Sat, 15 Mar 2003 01:32:48 +0000 (01:32 +0000)]
file extN-delete_thread.diff was initially added on branch b_devel.

22 years agofile ext3-orphan_lock.diff was initially added on branch b_devel.
adilger [Sat, 15 Mar 2003 01:32:47 +0000 (01:32 +0000)]
file ext3-orphan_lock.diff was initially added on branch b_devel.

22 years agofile ext3-truncate_blocks.diff was initially added on branch b_devel.
adilger [Fri, 14 Mar 2003 21:12:05 +0000 (21:12 +0000)]
file ext3-truncate_blocks.diff was initially added on branch b_devel.

22 years ago- remove the lli_open_count that was accidentally re-introduced with b_io
zab [Thu, 13 Mar 2003 22:20:12 +0000 (22:20 +0000)]
- remove the lli_open_count that was accidentally re-introduced with b_io
- silence an unused variable warning in runas

22 years agorunas.c is modified to use execvp() instead of system(), suggested by Zach
pengzhao [Thu, 13 Mar 2003 04:56:19 +0000 (04:56 +0000)]
runas.c is modified to use execvp() instead of system(), suggested by Zach

22 years agofile checkstack.pl was initially added on branch b_devel.
adilger [Wed, 12 Mar 2003 00:58:10 +0000 (00:58 +0000)]
file checkstack.pl was initially added on branch b_devel.

22 years agoThis commit was manufactured by cvs2svn to create branch 'unlabeled-1.2.2'.
cvs2svn [Tue, 11 Mar 2003 23:37:57 +0000 (23:37 +0000)]
This commit was manufactured by cvs2svn to create branch 'unlabeled-1.2.2'.

22 years agomerge b_devel into HEAD. Includes: 0.6.0
pschwan [Tue, 11 Mar 2003 23:37:27 +0000 (23:37 +0000)]
merge b_devel into HEAD.  Includes:

- client-side I/O cache
- O_DIRECT fixes for IA64
- liblustre improvements
- various small bug fixes

22 years agoRemove obsolete rh-8.0 series file.
adilger [Tue, 11 Mar 2003 21:07:11 +0000 (21:07 +0000)]
Remove obsolete rh-8.0 series file.

22 years agofile runas.c was initially added on branch branch_bug828.
pengzhao [Tue, 11 Mar 2003 07:17:05 +0000 (07:17 +0000)]
file runas.c was initially added on branch branch_bug828.

22 years agoBug 828 is fixed by Peng Zhao, arrpoved by Andreas.
pengzhao [Tue, 11 Mar 2003 07:17:03 +0000 (07:17 +0000)]
Bug 828 is fixed by Peng Zhao, arrpoved by Andreas.
runas.c allows the root to "runas" another user to do things.

22 years agofile ext3-2.5-noread.diff was initially added on branch b_devel.
adilger [Sat, 8 Mar 2003 15:59:27 +0000 (15:59 +0000)]
file ext3-2.5-noread.diff was initially added on branch b_devel.

22 years agofile ptlrpc_module.c was initially added on branch b_recovery.
shaver [Fri, 7 Mar 2003 20:16:03 +0000 (20:16 +0000)]
file ptlrpc_module.c was initially added on branch b_recovery.

22 years agofile ptlrpc_internal.h was initially added on branch b_recovery.
shaver [Fri, 7 Mar 2003 20:16:02 +0000 (20:16 +0000)]
file ptlrpc_internal.h was initially added on branch b_recovery.

22 years agoCommenting out liblustre from make targets until Makefile.am in liblustre is fixed
amrutjoshi [Wed, 5 Mar 2003 06:47:27 +0000 (06:47 +0000)]
Commenting out liblustre from make targets until Makefile.am in liblustre is fixed

22 years agoRemove extraneous whitespace.
adilger [Tue, 4 Mar 2003 17:30:44 +0000 (17:30 +0000)]
Remove extraneous whitespace.

22 years agob=941: zab's fix for ia64 compiles. untested, but it does compile
rread [Tue, 4 Mar 2003 00:38:12 +0000 (00:38 +0000)]
b=941: zab's fix for ia64 compiles. untested, but it does compile

22 years agofile iod-stock-24-exports_hp.patch was initially added on branch b_devel.
rread [Tue, 4 Mar 2003 00:38:08 +0000 (00:38 +0000)]
file iod-stock-24-exports_hp.patch was initially added on branch b_devel.

22 years agoland b_io on b_devel (b_io will be deleted as soon as it's confirmed
pschwan [Sun, 2 Mar 2003 05:58:14 +0000 (05:58 +0000)]
land b_io on b_devel (b_io will be deleted as soon as it's confirmed
that the merge happend without incident)

22 years agoland 0.5.20.3 b_devel onto HEAD (b_devel will remain) 0.5.20.3
pschwan [Sun, 2 Mar 2003 05:23:24 +0000 (05:23 +0000)]
land 0.5.20.3 b_devel onto HEAD (b_devel will remain)

22 years ago- Flush every line so it works better with pdsh
behlendo [Fri, 28 Feb 2003 03:31:35 +0000 (03:31 +0000)]
- Flush every line so it works better with pdsh

22 years agofile ext3-largefile.diff was initially added on branch b_devel.
adilger [Thu, 27 Feb 2003 23:44:48 +0000 (23:44 +0000)]
file ext3-largefile.diff was initially added on branch b_devel.

22 years agoUpdate HP VFS patch for v13.
adilger [Thu, 27 Feb 2003 08:04:45 +0000 (08:04 +0000)]
Update HP VFS patch for v13.

22 years agob=912
pschwan [Wed, 26 Feb 2003 23:10:14 +0000 (23:10 +0000)]
b=912
- Lustre patch version 13:
  - merge the iod-rmap-exports patch from b_io
  - add and export __d_rehash, which optionally avoids taking dcache_lock
- fix bug 912 race by calling d_rehash without dropping the dcache_lock
- Fix earlier botched patch for 913

22 years agofile recovery-cleanup.sh was initially added on branch b_devel.
shaver [Tue, 25 Feb 2003 16:40:47 +0000 (16:40 +0000)]
file recovery-cleanup.sh was initially added on branch b_devel.

22 years agoChanged to use absolte paths instead of relative paths for symbols in case
adilger [Tue, 25 Feb 2003 01:02:17 +0000 (01:02 +0000)]
Changed to use absolte paths instead of relative paths for symbols in case
that made a difference (it didn't).

22 years agofile mlink.c was initially added on branch b_devel.
shaver [Mon, 24 Feb 2003 21:48:17 +0000 (21:48 +0000)]
file mlink.c was initially added on branch b_devel.

22 years agoDropped-request and -reply tests for link and unlink.
shaver [Mon, 24 Feb 2003 21:45:58 +0000 (21:45 +0000)]
Dropped-request and -reply tests for link and unlink.

22 years agoDropped-request and dropped-reply tests for rename.
shaver [Mon, 24 Feb 2003 19:37:04 +0000 (19:37 +0000)]
Dropped-request and dropped-reply tests for rename.

22 years ago- Quell (spurious, but I don't blame gcc for getting it wrong) "uninitialized
shaver [Mon, 24 Feb 2003 19:35:18 +0000 (19:35 +0000)]
- Quell (spurious, but I don't blame gcc for getting it wrong) "uninitialized
  use of dchild" warning.

- Reconstruct rename replies (easy, until I go back and make them all work in
  the failed-over case).

22 years ago- macroize and share check-for-resent-req code.
shaver [Mon, 24 Feb 2003 17:05:30 +0000 (17:05 +0000)]
- macroize and share check-for-resent-req code.
- defensive assertion in mds_close_mfd to help catch dentry mismanagement
- transnos for getattr_name and open, always.
- don't free the repmsg too early when we're dropping replies for testing
- ldlm_intent_policy knows how to fix up lock handles for reconstructed requests,
  and not give the lock away to the client twice
- tchmod exits with errno on failure.

22 years agoUpdate lcrash map-making script.
adilger [Sat, 22 Feb 2003 22:23:44 +0000 (22:23 +0000)]
Update lcrash map-making script.

22 years agofile .cvsignore was initially added on branch b_devel.
braam [Sat, 22 Feb 2003 03:15:55 +0000 (03:15 +0000)]
file .cvsignore was initially added on branch b_devel.

22 years agofile compile.sh was initially added on branch b_devel.
adilger [Fri, 21 Feb 2003 11:44:54 +0000 (11:44 +0000)]
file compile.sh was initially added on branch b_devel.

22 years agofile Makefile.am was initially added on branch b_devel.
braam [Fri, 21 Feb 2003 11:05:36 +0000 (11:05 +0000)]
file Makefile.am was initially added on branch b_devel.

22 years agofile libtest.c was initially added on branch b_devel.
braam [Fri, 21 Feb 2003 11:05:35 +0000 (11:05 +0000)]
file libtest.c was initially added on branch b_devel.

22 years agofix uninitialized variable warning
pschwan [Thu, 20 Feb 2003 22:16:51 +0000 (22:16 +0000)]
fix uninitialized variable warning

22 years agob=892
pschwan [Thu, 20 Feb 2003 22:07:11 +0000 (22:07 +0000)]
b=892
osc_close used to walk through the request list, trying to find the
corresponding osc_open.  This is slow, racy and doomed to failure, so we
need to instead save a point to the request in osc_open.

Instead of passing a handle up to whoever called us, pass 32 bytes of
opaque data.  These get passed back to us at various points, through
the obdo's inline field.

22 years agoFixed to make version_tag.pl work outside srcdir.
amrutjoshi [Thu, 20 Feb 2003 11:51:12 +0000 (11:51 +0000)]
Fixed to make version_tag.pl work outside srcdir.

22 years agoYou might not believe it, but the open-reply tests pass now.
shaver [Thu, 20 Feb 2003 09:02:07 +0000 (09:02 +0000)]
You might not believe it, but the open-reply tests pass now.

22 years agoReconstruct replies for open. Lightly tested (successful open case).
shaver [Thu, 20 Feb 2003 08:59:38 +0000 (08:59 +0000)]
Reconstruct replies for open.  Lightly tested (successful open case).

22 years ago- Store open-request XID in the MFD, so that we can find it during reply
shaver [Thu, 20 Feb 2003 07:36:15 +0000 (07:36 +0000)]
- Store open-request XID in the MFD, so that we can find it during reply
  reconstruction.
- Untested reply reconstruction for GETATTR_NAME (easy) and OPEN (not easy,
  ahem).
- Don't leak dentry refs when reconstructing replies for reint_setattr or
  reint_create.
- Don't double-dput dchild if dentry_open fails.  (Not yet in test suite, bug
  894 filed to trick someone into adding it.)

22 years agodon't LBUG if we can't lookup the parent in mds_open and mds_reint_create; it's
pschwan [Thu, 20 Feb 2003 07:09:40 +0000 (07:09 +0000)]
don't LBUG if we can't lookup the parent in mds_open and mds_reint_create; it's
entirely possible that we're racing with rmdir()

22 years agofile lustre_peer.c was initially added on branch b_malt.
rread [Thu, 20 Feb 2003 06:52:52 +0000 (06:52 +0000)]
file lustre_peer.c was initially added on branch b_malt.

22 years agofile statone.c was initially added on branch b_devel.
shaver [Wed, 19 Feb 2003 22:01:20 +0000 (22:01 +0000)]
file statone.c was initially added on branch b_devel.

22 years agoAdd tests for reply reconstruction and request retransmission for
shaver [Wed, 19 Feb 2003 20:28:42 +0000 (20:28 +0000)]
Add tests for reply reconstruction and request retransmission for
reint_setattr.