Whamcloud - gitweb
phil [Thu, 27 Nov 2003 06:31:34 +0000 (06:31 +0000)]
Slightly less obvious, but still very innocent, parts of b_eq:
- code or includes in the main build, but contained in #if __KERNEL__
- userspace-only portals code
- very few and minor other changes, such as renaming a function
shaver [Thu, 27 Nov 2003 01:05:52 +0000 (01:05 +0000)]
b=2303: Unregister replies when skipping ping-requests from disconnected
imports, to avoid tripping assertions when we free the requests.
r=robert
eeb [Wed, 26 Nov 2003 21:42:28 +0000 (21:42 +0000)]
* Fixed socknal forwarding race in the new version of
ksocknal_process_receive().
phil [Wed, 26 Nov 2003 18:52:12 +0000 (18:52 +0000)]
backing out eeb's socknal checkin from yesterday:
11/25/2003 12:22 eeb lustre/portals/knals/socknal/socknal.c 1.1.2.13
11/25/2003 12:22 eeb lustre/portals/knals/socknal/socknal_cb.c 1.1.2.16
LLNL found a race condition in overnight testing, causes LBUG.
rread [Wed, 26 Nov 2003 01:26:46 +0000 (01:26 +0000)]
b=2296
r=shaver
- Build ldlm into ptlrpc.
- Remove all the ptlrpc_ldlm_* hooks
This almost certainly breaks building when $(SRDIR) != $(OBJDIR).
rread [Tue, 25 Nov 2003 20:29:39 +0000 (20:29 +0000)]
- unregister the elan nal
eeb [Tue, 25 Nov 2003 17:22:29 +0000 (17:22 +0000)]
* Always hold kss_lock while setting/inspecting ksnc_{rx,tx}_ready.
* Removed kss_lock ops from ksocknal_process_{transmit,receive}().
ksocknal_scheduler() does this locking now.
* Included the --single_socket lconf flag.
ericm [Tue, 25 Nov 2003 14:43:38 +0000 (14:43 +0000)]
liblustre: as andreas pointed out, open() need to convert O_XXX flags
to FMODE_XXX which is internally used by lustre. remove the original
hacking code.
tianying [Tue, 25 Nov 2003 12:51:54 +0000 (12:51 +0000)]
1. move llog_origin_handle_cancel to llog_server.c
2. destroy useless plain logs during both llog_cleanup and llog_setup
3. change some CERRORs to CWARNs while just print out status information
jacob [Tue, 25 Nov 2003 07:12:15 +0000 (07:12 +0000)]
b=1941
Improvements from jerrifer's patch, with some changes.
- use lmc --batch for performance goodness
- remove 'read_conf' as it is confusing and redundant
- misc cleanups, improvements
rread [Mon, 24 Nov 2003 23:57:02 +0000 (23:57 +0000)]
cleanup clients correctly in insanity.sh
set the mds upcall, so the mds can recovery from ost failures
alex [Mon, 24 Nov 2003 22:13:36 +0000 (22:13 +0000)]
- port onto virgin 2.4.22
niu [Mon, 24 Nov 2003 08:59:21 +0000 (08:59 +0000)]
Move device type specific llog ioctl codes from generic obdclass to mds/obdfilter/mdc.
phil [Mon, 24 Nov 2003 04:45:54 +0000 (04:45 +0000)]
b=1883
- Adds llite_close.c; Zach's originally, somewhat heavily hacked
- Adds an MDS_DONE_WRITING rpc and handler
- Adds client-side code for storing the MDS FID and epoch in the
obdo's inline field in BRW_WRITE requests
phil [Mon, 24 Nov 2003 00:24:53 +0000 (00:24 +0000)]
b=2192
ll_inode_getattr (-> ll_lsm_getattr) now takes an obdo, which it fills with the
attributes. The caller is responsible, if it wishes, for updating the fields
in the inode, and the one previous caller now contains that extra code. We'll
see whether additional abstraction is required during the KMS work.
This is a prerequisite for the first 30% of the KMS that we decided to do as a
fix for this bug.
ericm [Sat, 22 Nov 2003 07:09:16 +0000 (07:09 +0000)]
liblustre: forgot to commit last night: workaround O_WRONLY (b2282)
wangdi [Fri, 21 Nov 2003 18:20:14 +0000 (18:20 +0000)]
add special device inode lustre handler
alex [Fri, 21 Nov 2003 15:06:24 +0000 (15:06 +0000)]
- patch to fix 64-bit pointer arithmetic bug in the ext3 extended
attributes code.
alex [Fri, 21 Nov 2003 14:42:56 +0000 (14:42 +0000)]
- rename the patch
alex [Fri, 21 Nov 2003 14:36:01 +0000 (14:36 +0000)]
- patches against 2.4.19-pre1 and series
rread [Fri, 21 Nov 2003 09:10:57 +0000 (09:10 +0000)]
move client_df to test-framework.
rread [Fri, 21 Nov 2003 08:42:40 +0000 (08:42 +0000)]
- integrate support for FAILURE_MODE into test-framework
- add configs for mdev
rread [Fri, 21 Nov 2003 03:27:14 +0000 (03:27 +0000)]
- cleanup old active files
- cleanup insanity test_0
- add checks to make sure the nodes are up during setup
rread [Fri, 21 Nov 2003 01:04:56 +0000 (01:04 +0000)]
add usage() to test_framework, and use getopts cause getopt is broken
insanity fixes for hardware
shaver [Thu, 20 Nov 2003 21:41:14 +0000 (21:41 +0000)]
Small test and upcall fixes; we fail over during a write now.
adilger [Thu, 20 Nov 2003 12:23:22 +0000 (12:23 +0000)]
Fix bulk RPC checksum code and page poisoning for debugging bug 2254.
b=1703,b=2254
rread [Thu, 20 Nov 2003 07:43:00 +0000 (07:43 +0000)]
- Integrate insanity.sh with test-framework
- Centralize the config data to files in tests/cfg/
- Different config file can be used by setting CONFIG=file or
using command line option: --config <file>
- FAILURE_MODE option for insanity.sh:
FAILURE_MODE=SOFT - shutdown services with lconf --force
FAILURE_MODE=HARD - power down the service nodes using $POWER_DOWN
and $POWER_UP to reboot.
This feature should be integrated with the other tests.
- replay-single.sh, replay-dual.sh, and conf-sanity.sh all pass on hardware
- replay-ost-dual.sh and insanity.sh don't pass yet
wangchao [Thu, 20 Nov 2003 06:06:55 +0000 (06:06 +0000)]
bug: 1944
reviewer: Andreas
Add "lfs osts" and "lfs check <mds|osts|servers>" to lfs.
wangdi [Thu, 20 Nov 2003 03:25:17 +0000 (03:25 +0000)]
add lbuild scripts, please read lbuild --help
rread [Thu, 20 Nov 2003 00:50:56 +0000 (00:50 +0000)]
- during disconnect, downgrade import level early for non-replayable
obds. This prevents ptlrpc from automatically reconnecting on a
network failure.
adilger [Wed, 19 Nov 2003 21:40:51 +0000 (21:40 +0000)]
Fix local-large-inode.sh script so it is actually using large inodes (my bad),
although in my defence acc-sm.sh would have broken until people updated their
kernel, which has mostly happened now.
Make the other configs create an MDS large enough to hold inodes for bonnie++
with all-external EA data.
adilger [Wed, 19 Nov 2003 21:36:22 +0000 (21:36 +0000)]
Return GFP_NOFS fix to socknal that was lost in autoconn code shuffle.
b=1933
shaver [Wed, 19 Nov 2003 18:53:50 +0000 (18:53 +0000)]
MDS-side OSCs always ping their OSTs for prompt failure detection.
ericm [Wed, 19 Nov 2003 11:38:36 +0000 (11:38 +0000)]
liblustre: release it lock a bit later, abvoid reference to null lock,
should be the reason of LASSERT in simul test.
wangdi [Wed, 19 Nov 2003 11:10:09 +0000 (11:10 +0000)]
add conf-sanity.sh
ericm [Wed, 19 Nov 2003 09:57:20 +0000 (09:57 +0000)]
liblustre: minor fix for replay_single test, even 60 to too much for vmware.
rread [Tue, 18 Nov 2003 20:36:11 +0000 (20:36 +0000)]
- Remove last references to OBD_CLASS_UUID. Now the obd->obd_uuid is
used as the client uuid for self and lctl exports.
zab [Tue, 18 Nov 2003 18:43:55 +0000 (18:43 +0000)]
- fix the osc histogram spinlock inits that my uml build missed.
take a shot.
shaver [Tue, 18 Nov 2003 18:30:56 +0000 (18:30 +0000)]
b=2247: detect misbehaving clients that reuse UUIDs _and_ import handles (such
as libtest before the srand fix) and error, rather than LBUGing.
rread [Tue, 18 Nov 2003 09:52:39 +0000 (09:52 +0000)]
insanity.sh - this is test17 turned into a regular test, if you could
call this test regular. Still a work in progress, but does at least
setup/cleanup.
ericm [Tue, 18 Nov 2003 08:58:34 +0000 (08:58 +0000)]
liblustre: fix test_lock_cancel for zeroconf.
alex [Mon, 17 Nov 2003 23:28:28 +0000 (23:28 +0000)]
- now devices can use memory allocated via uaf
adilger [Mon, 17 Nov 2003 19:26:28 +0000 (19:26 +0000)]
Allow current b_devel to work with older (read LLNL) kernels until they are
updated. Bumped LUSTRE_KERNEL_VERSION to 31 so we know which kernels have
the required patches, but don't make it a hard requirement.
b=1933
alex [Mon, 17 Nov 2003 17:32:56 +0000 (17:32 +0000)]
- lustre patch of b1933
NOTE: don't forget to apply gfp_memalloc patche first
alex [Mon, 17 Nov 2003 17:32:24 +0000 (17:32 +0000)]
- kernel part of b1933
niu [Mon, 17 Nov 2003 06:42:02 +0000 (06:42 +0000)]
b: 2194
r: Peter
Add lctl interface for llog.
adilger [Sun, 16 Nov 2003 23:18:42 +0000 (23:18 +0000)]
Sigh, patch I had submitted (which I grabbed) was not what was added to tree
(minor changes from SCT on the way).
adilger [Sun, 16 Nov 2003 23:01:14 +0000 (23:01 +0000)]
Slight context diff for journal callback patch from vanilla, add to suse series.
adilger [Sun, 16 Nov 2003 22:55:54 +0000 (22:55 +0000)]
Add journal callback patch for SuSE 2.4.19 ia64 kernel (in vanilla 2.4.20).
adilger [Sun, 16 Nov 2003 22:08:54 +0000 (22:08 +0000)]
New patch files for SuSE 2.4.19 ia64 kernel as used by SDSC. Mostly cosmetic
except vfs_intent still used the old-style path_init()/path_walk(), and it
already had EA patches applied.
adilger [Sun, 16 Nov 2003 21:59:19 +0000 (21:59 +0000)]
New series for SuSE-2.4.19 ia64 kernel for SDSC machines.
adilger [Sun, 16 Nov 2003 21:43:24 +0000 (21:43 +0000)]
Remove gratuitous newline hunk, which causes a conflict on SuSE-2.4.19.
eeb [Sun, 16 Nov 2003 17:45:55 +0000 (17:45 +0000)]
* Plugged a spurious timeout hole
alex [Sun, 16 Nov 2003 16:40:30 +0000 (16:40 +0000)]
- lightly-tested, undocumented debuggin tool to catch use-after-free bugs
NOTE: documentation will be available soon
rread [Sat, 15 Nov 2003 01:17:52 +0000 (01:17 +0000)]
b=2250
create lctl commands set_lustre_upcall and set_timeout, so they
can be saved as part of the 0conf log.
rread [Sat, 15 Nov 2003 01:05:26 +0000 (01:05 +0000)]
on hardware, the script runs too fast for these multiops to die, cause
the /mnt/lustre to be busy. So, have test 32 cause recovery on the client and wait
a bit before finishing.
rread [Fri, 14 Nov 2003 23:55:42 +0000 (23:55 +0000)]
- Don't wake up the pinger, if it's not enabled
adilger [Fri, 14 Nov 2003 23:09:31 +0000 (23:09 +0000)]
Make sure lwt_event_t packs into a page evenly on ia64.
adilger [Fri, 14 Nov 2003 23:03:25 +0000 (23:03 +0000)]
Some more wire constant checks.
b=1321
rread [Fri, 14 Nov 2003 22:32:24 +0000 (22:32 +0000)]
More test cleanups.
rread [Fri, 14 Nov 2003 22:04:23 +0000 (22:04 +0000)]
Wake up the pinger after adding an import, to reset the timer.
adilger [Fri, 14 Nov 2003 21:49:34 +0000 (21:49 +0000)]
Re-add lgd_len in case it is needed at some point (conflict resolution error).
b=1821
rread [Fri, 14 Nov 2003 20:52:11 +0000 (20:52 +0000)]
- remove reference to lgd_len, which has disappeared
adilger [Fri, 14 Nov 2003 20:12:30 +0000 (20:12 +0000)]
Interoperability for different PAGE_SIZE/wordsize machines. Tested on ia64
and i386 separately, and with ia64 client + i386 MDS/OST.
Mostly aligning structs to have 64-bit fields aligned on 64-bit boundaries.
Remove some VFS constants and replace them with Lustre constants on the wire.
Since the MDS doesn't really open files itself, we don't need to convert from
the wire constants back to local flags at all.
Frobbing of niobufs on the targets to split them into PAGE_SIZE chunks (this
may be a problem on large PAGE_SIZE servers with small PAGE_SIZE clients,
not sure yet).
I have tested and this appears to be compatible with old filesystems.
At worst we should only need another --write_conf on the MDS.
b=686, b=1821, b=1343, b=2042
adilger [Fri, 14 Nov 2003 19:27:01 +0000 (19:27 +0000)]
Use PTL_MD_MAX_PAGES instead of PTL_MD_MAX_IOV where needed, so we send the
right number of pages over the wire for ia64/i386 interoperability.
b=1821
ericm [Fri, 14 Nov 2003 15:05:48 +0000 (15:05 +0000)]
liblustre: update lrun script.
ericm [Fri, 14 Nov 2003 11:13:26 +0000 (11:13 +0000)]
liblustre:
add more test items into replay_single.
ericm [Fri, 14 Nov 2003 09:37:33 +0000 (09:37 +0000)]
liblustre: add more test in replay_single, and some other adjustment.
phil [Fri, 14 Nov 2003 09:24:31 +0000 (09:24 +0000)]
I touched one thing in lustre_mds.h, and I was dismayed to see the
entire source tree rebuild!
It took just a few minutes to remove a number of #includes which
violated all manner of abstraction boundary. ericm assures me that I
have not broken anything major in liblustre.
Also, I moved a good chunk of inline functions into llite_internal.h
(which is where I secretly believe that most of lustre_lite.h will end up),
and renamed ll_ino2fid (which no longer takes an inode) to mdc_pack_fid.
rread [Fri, 14 Nov 2003 01:12:31 +0000 (01:12 +0000)]
- fix del_route
alex [Thu, 13 Nov 2003 20:49:31 +0000 (20:49 +0000)]
- update this patch to apply against vanilla-2.4.22
zab [Thu, 13 Nov 2003 20:21:48 +0000 (20:21 +0000)]
Silence most of the warnings in the x86_64 build, and kill some Real Bugs in
the process. (sanity checked in an x86 uml build, as well)
- include linux/init.h from some modules that use __init and company
- x86_64 has 64 bit longs but uses 'long long' for u64 in the kernel. we
add its own specific section to LP*64 definitions. Sadly, there are still
a huge number of warnings from this gcc when we try to LP*64 print uint64_t
obdo arguments.
- use min_t in some defines to avoid the duplicate const prefacing that comes
with nesting min()
- LPX64 printing -1UL doesn't work so well
- i_nlink can be any of unsigned short, int, or long. when printing cast
to unsigned long.
- kern_addr_valid takes a 'unsigned long' argument and shifts it.. sigh.
- don't #define EXPORT_SYMTAB in the source, that's the job of the kernel
makefiles
- x86_64/suse has THREAD_SIZE defined in userspace
- s/long flags/unsigned long flags/. Again.
- va_list ap = NULL; is not legal, as far as I know. it doesn't seem
to actually do anything in this code path either.
- x86_64 passes CONFIG_X86 but had thread->rsp as the stack pointer. this
code is going away eventually anyway.
- print a pointer with %p, not %x
- some portals callback pointer definitions used size_t while some nal's
functions used ptl_size_t in the function defintions. I switched the
functions over to avoid pushing changes on all the nals.
alex [Thu, 13 Nov 2003 15:30:09 +0000 (15:30 +0000)]
- ide_end_request() was used wrong way: fixed
NOTE: tested on the all related series
alex [Thu, 13 Nov 2003 15:06:15 +0000 (15:06 +0000)]
- kmem_cache_validate patch has been removed from all the series
NOTE: all the supported series still build
ericm [Thu, 13 Nov 2003 13:01:48 +0000 (13:01 +0000)]
liblustre: zconf adjustment for recovery_small and replay_single.
ericm [Thu, 13 Nov 2003 06:56:20 +0000 (06:56 +0000)]
liblustre:
- adjust the way of zeroconf
- bring lltest/recovery_small to zeroconf
tianying [Thu, 13 Nov 2003 05:57:06 +0000 (05:57 +0000)]
b:2215 - OSTs fetch unlink llog records from MDS post replay
1. add lop_connect and lop_precleanup to llog_operations
2. rename llog_obd_ctxt to llog_ctxt; llog_commit_data to llog_canceld_ctxt
3. split out llog functions in llog_client.c and llog_server.c and remove llogd.c
4. add one test-59 to sanity.sh to verify cancellation of llog records async
5. fix calling of mds_cleanup_orphans, add test-34 to replay-single.sh
6. fix some codes about llog
rread [Thu, 13 Nov 2003 01:27:36 +0000 (01:27 +0000)]
- return replay-ost-single to runable state
now just need to pass
- commonize the zconf mount/umount
adilger [Thu, 13 Nov 2003 01:00:08 +0000 (01:00 +0000)]
Fix ia64 compile warnings. Please folks, use LPU64 instead of %llu.
rread [Wed, 12 Nov 2003 23:17:22 +0000 (23:17 +0000)]
- enable KMOD
(CONFIG_MODE_TT=y is added by make oldconfig)
rread [Wed, 12 Nov 2003 23:14:01 +0000 (23:14 +0000)]
sample modules.conf for zeroconf
adilger [Wed, 12 Nov 2003 23:11:46 +0000 (23:11 +0000)]
Remove kmem_cache_validate_hp from hp-pnnl-2.4.20 series.
rread [Wed, 12 Nov 2003 21:16:28 +0000 (21:16 +0000)]
b=2244
r=shaver
Pass force flag to ldlm_cleanup.
rread [Wed, 12 Nov 2003 21:11:24 +0000 (21:11 +0000)]
- fix set_route
- add kptlrouter to list of modules
adilger [Wed, 12 Nov 2003 19:49:42 +0000 (19:49 +0000)]
Add process pid into console log messages.
ericm [Wed, 12 Nov 2003 14:15:15 +0000 (14:15 +0000)]
liblustre: support zeroconf.
but current existed tests still use dumpfile, switch to zconf later.
wangchao [Wed, 12 Nov 2003 07:56:04 +0000 (07:56 +0000)]
b=2134
r=shaver
add a test case of unix domain socket test to sanity.sh (test case 54)
ericm [Wed, 12 Nov 2003 07:31:46 +0000 (07:31 +0000)]
liblustre: add missing lib for shared so creation.
adilger [Wed, 12 Nov 2003 06:27:19 +0000 (06:27 +0000)]
Don't have a separate (inconsistent) MTU define for portals routers.
Print out portals router messages with nicely formatted NIDs.
rread [Wed, 12 Nov 2003 06:18:48 +0000 (06:18 +0000)]
- add /sbin/mount.lustre to rpm
rread [Wed, 12 Nov 2003 06:18:10 +0000 (06:18 +0000)]
cleanup replay-single.sh
rread [Wed, 12 Nov 2003 02:41:21 +0000 (02:41 +0000)]
b=2243
fix router config breakage. Now that the router config commands are
NAL_CMDs, need to add a gw_nal to pcfg.
rread [Wed, 12 Nov 2003 00:54:34 +0000 (00:54 +0000)]
create ibnal/Makefile
shaver [Wed, 12 Nov 2003 00:17:01 +0000 (00:17 +0000)]
Quiet warnings.
rread [Tue, 11 Nov 2003 23:13:14 +0000 (23:13 +0000)]
Add new "lustre" fs type which supports only zeroconf mounts.
- old zeroconf client code removed from lustre-lite, and the lconf
--zeroconf option deleted
- common code factored out for ll/lustre fill_super and put_super
- lconf still uses lustre_lite (but not for long)
llmount will be used by mount for lustre filesystems if copied to
/sbin/mount.lustre:
mount -t lustre mds_host:/mds_servicee/profile /mnt/lustre
Multiple mounts of the same filesystem are supported.
Remove unused mds and filter nspath code.
eeb [Tue, 11 Nov 2003 16:42:34 +0000 (16:42 +0000)]
* fixed kportal_router_cmd() to return -EINVAL on unrecognised command
wangdi [Tue, 11 Nov 2003 10:31:39 +0000 (10:31 +0000)]
update netconsole-2.4.20-rh.patch
ericm [Tue, 11 Nov 2003 05:18:13 +0000 (05:18 +0000)]
liblustre:
- support llog facilities
- remove unused wait_for_completion/completion pairs
zab [Tue, 11 Nov 2003 01:09:42 +0000 (01:09 +0000)]
- we don't have an arch/x86_64 for kmem_cache_validate(), but the code
doesn't use it anymore, so we just won't include it in the series.
alex [Tue, 11 Nov 2003 00:35:13 +0000 (00:35 +0000)]
- async deletion for suse kernels