Whamcloud - gitweb
adilger [Thu, 17 Mar 2005 23:22:39 +0000 (23:22 +0000)]
Branch: HEAD
Build fix for RHEL4.
b=5733
r=green (original patch)
eeb [Thu, 17 Mar 2005 19:16:17 +0000 (19:16 +0000)]
* fix for 5890: different compilers produce different wire struct offsets
This gives vibnal its own wirecheck utility and adds compile-time
assertions to vibnal to ensure all platforms agree.
eeb [Thu, 17 Mar 2005 18:54:57 +0000 (18:54 +0000)]
* vibnal fix + earlier arp failure reporting for bug 5852
yury [Thu, 17 Mar 2005 08:50:46 +0000 (08:50 +0000)]
- added check for existing TAG file before catting it
zhaoqiang [Thu, 17 Mar 2005 03:11:10 +0000 (03:11 +0000)]
*** empty log message ***
yury [Wed, 16 Mar 2005 14:53:14 +0000 (14:53 +0000)]
- find_assigned_loop() should check all loop devices and do not return after first not accessible.
- looking for free loop device is moved to stand alone function find_free_loop()
- int init_loop() do not check if passed size > 8000 before mkfs() as this is checked in mkfs()
- do not stop looking for free loop device after first not accessible device.
yury [Wed, 16 Mar 2005 11:49:32 +0000 (11:49 +0000)]
- fix in test_16 in conf sanity
eeb [Wed, 16 Mar 2005 11:04:11 +0000 (11:04 +0000)]
* fix for 5876 to change vibnal connection request/response message
marshalling into priv_data to avoid ia64 alignment traps.
tappro [Wed, 16 Mar 2005 10:57:38 +0000 (10:57 +0000)]
b=5881
r=wangdi
fixes in accordance with SMFS changes
tappro [Wed, 16 Mar 2005 10:55:48 +0000 (10:55 +0000)]
b=5881
r=wangdi
smfs fixes to make it works in 2.6.7 kernel and in conformity with smfs hld/dld documents.
yury [Wed, 16 Mar 2005 08:05:07 +0000 (08:05 +0000)]
- fix in mds_cancel_cookie_cb() with freeing lsm.
niu [Wed, 16 Mar 2005 07:37:57 +0000 (07:37 +0000)]
add debug mask for quota
alex [Tue, 15 Mar 2005 14:20:10 +0000 (14:20 +0000)]
b=5874
additional debug info
yury [Tue, 15 Mar 2005 07:41:35 +0000 (07:41 +0000)]
- returned back old schema of mds_verify_child()
yury [Tue, 15 Mar 2005 07:25:03 +0000 (07:25 +0000)]
- fixed wrongly inverted condition in lmv_create_obj()
alex [Mon, 14 Mar 2005 22:20:17 +0000 (22:20 +0000)]
b=5765
r=adilger
choose proper obd to match lock that protects directory content
yury [Mon, 14 Mar 2005 19:16:37 +0000 (19:16 +0000)]
- memory leak fixes from b_fid.
yury [Sun, 13 Mar 2005 11:30:13 +0000 (11:30 +0000)]
- more checks and verbosity in test_16 of conf-sanity.sh to see what happens when it does not pass.
yury [Sat, 12 Mar 2005 15:55:51 +0000 (15:55 +0000)]
- mds_verify_child() is converted to use fid component of lustre_id instead of pair ino+gen for checking if inode has changed while we were waiting for lock.
alex [Fri, 11 Mar 2005 22:17:22 +0000 (22:17 +0000)]
b=5863
- assertion was wrong
yury [Fri, 11 Mar 2005 18:24:01 +0000 (18:24 +0000)]
- fixed using not initialized ctxt with pop_ctxt() in mds_lock_and_check_slave()
eeb [Fri, 11 Mar 2005 16:50:44 +0000 (16:50 +0000)]
* fixed bug tracefile.c: tage_alloc() as encountered in bug 5857
* Asserted tage->page != NULL in tage_free; how could it ever be otherwise?
* Moved what seemed like an obviously incorrect assertion in trace_get_tage()
that tcd_pages was not empty into the "debug daemon buffer overflowed"
branch.
Someone should review this change; I'm flagging it in bug 5857
eeb [Fri, 11 Mar 2005 11:10:46 +0000 (11:10 +0000)]
* fixed 5000: socknal autoconnect race (outgoing v. incoming)
* minor socknal cleanups
wangdi [Fri, 11 Mar 2005 09:37:41 +0000 (09:37 +0000)]
exp might be NULL
B: HEAD
yury [Fri, 11 Mar 2005 09:17:21 +0000 (09:17 +0000)]
- update from HEAD
alex [Thu, 10 Mar 2005 22:13:14 +0000 (22:13 +0000)]
b=5209
__journal_refile_buffer() shouldn't file buffer
on metadata list if the buffer is clean
eeb [Thu, 10 Mar 2005 16:25:36 +0000 (16:25 +0000)]
* fix for 5809: vibnal tx_sending race
nikita [Wed, 9 Mar 2005 11:39:00 +0000 (11:39 +0000)]
turn debugging on by default for now
wangdi [Wed, 9 Mar 2005 08:01:06 +0000 (08:01 +0000)]
Branch: HEAD
Log:
we should lock the ino not fid in mds_lock_new_child, for protect ino being reused after unlink.
b:5843
r:alex
alex [Tue, 8 Mar 2005 09:01:51 +0000 (09:01 +0000)]
- fix against credits leak in journal_release_buffer()
accepted in mainline (2.6.11-pre)
liuduo [Tue, 8 Mar 2005 03:43:35 +0000 (03:43 +0000)]
uptodate with HEAD, resovle lots of conflictions
adilger [Mon, 7 Mar 2005 22:17:51 +0000 (22:17 +0000)]
Branch: b1_2
Revert the RPM requirement for /boot/System.map-%{kversion}. Not all kernels
install this file. Instead, use it if it exists, but fall back to old
behaviour if it doesn't (though we depmod against the right kernel at least).
eeb [Mon, 7 Mar 2005 12:24:23 +0000 (12:24 +0000)]
* further lctl debug_file fixes
- truncate output file
- close streams
eeb [Mon, 7 Mar 2005 11:46:21 +0000 (11:46 +0000)]
* vibnal fixes
5807 - incorrect outstanding SEND accounting
changed refcounting inlines to macros to improve logging
cleanup unused per-work-item debug on send
* lctl debug_file wasn't passing the mode argument to open(O_CREAT)
fixed with restrictive (owner only) permissions
* 5803 ranal initialisation fix
yury [Sun, 6 Mar 2005 18:58:33 +0000 (18:58 +0000)]
- different fixes in MDS
adilger [Sun, 6 Mar 2005 07:29:22 +0000 (07:29 +0000)]
Branch: b1_4
Print current portals allocation when we hit enomem in socknal for debugging.
b=5601
adilger [Sat, 5 Mar 2005 19:35:06 +0000 (19:35 +0000)]
Merge b1_4_bgl from b1_4 (20050304_1757)
- don't LASSERT in ll_release on NULL lld with NFS export (4655, 5760)
- hold NS lock when calling handle_ast_error->del_waiting_lock (5746)
- fix setattr mtime regression from lovcleanup merge (4829, 5669)
- workaround for 2.6 crash in ll_unhash_aliases (5687, 5210)
- small ext3 extents cleanups and fixes (5733)
- improved mballoc code, several small races and bugs fixed (5733, 5638)
- kernel version 43 - fix remove_suid bugs in both 2.4 and 2.6 (5695)
- avoid needless client->OST connect, fix handle mismatch (5317)
- fix DLM error path that led to out-of-sync client, long delays (5779)
- support common vfs-enforced mount options (nodev,nosuid,noexec) (5637)
- fix several locking issues related to i_size (5492,5624,5654,5672)
- mdc_enqueue does not need max_mds_easize request buffer on send (5707)
- fix lustre/lustrefs init scripts for SuSE (patch from Scali, 5702)
- don't hold the pinger_sem in ptlrpc_pinger_sending_on_import
- change obd_increase_kms to obd_adjust_kms (up or down) (5654)
nic [Fri, 4 Mar 2005 17:21:22 +0000 (17:21 +0000)]
all 2.6 platforms need the -objdir/Module.symvers built
nic [Fri, 4 Mar 2005 16:11:50 +0000 (16:11 +0000)]
set LINUX26 in 2.6-rhel4 to get correct building instructions
only grab Kerntypes and symvers on suse, rhel4 does not have them
alex [Thu, 3 Mar 2005 19:23:12 +0000 (19:23 +0000)]
b=5599
serialize orphan cleanups between OSTs
eeb [Thu, 3 Mar 2005 15:44:06 +0000 (15:44 +0000)]
* Fix for 5806: vibnal NULL arp_data deref
eeb [Thu, 3 Mar 2005 12:11:18 +0000 (12:11 +0000)]
* fix for 5722 openibnal: NULL dereference in kibnal_close_conn_locked
jacob [Wed, 2 Mar 2005 21:26:01 +0000 (21:26 +0000)]
Patch from HP to support external qsnet.
jacob [Wed, 2 Mar 2005 21:16:30 +0000 (21:16 +0000)]
b=5786
r=phil
create an awesome super macro for nicely handling --with-foo,
--with-foo-includes, and --with-foo-libs for some situations we
encounter. make gm and cray portals use it.
nic [Wed, 2 Mar 2005 20:21:14 +0000 (20:21 +0000)]
remove the lustre version from the kernel EXTRA_VERSION to help get long branch
named nonrelease kernels under the 64 char uname limit
eeb [Wed, 2 Mar 2005 19:10:12 +0000 (19:10 +0000)]
* Fixed 5708: openib reconnect on client reboot issues
green [Wed, 2 Mar 2005 18:06:51 +0000 (18:06 +0000)]
lustre b1_4_rhel4 requires portals b1_4_rhel4
yury [Wed, 2 Mar 2005 17:30:19 +0000 (17:30 +0000)]
- fixes from b1_4:
- in filter_setattr() for test_17 from sanityN.sh.
- in ll_truncate() do not call truncate on server if new inode size the same as
previous one. Note: using lov_merge_size() in llite is clear layering violation.
- fixes in ll_setattr_raw()
yury [Wed, 2 Mar 2005 13:42:51 +0000 (13:42 +0000)]
- fixed sanity test 55.
- fixed sanity test 57a. It did not work with LMV setup due to unsupported features flag.
yury [Wed, 2 Mar 2005 07:51:40 +0000 (07:51 +0000)]
- fixed md_getattr() in lustre_common-fill_super()
adilger [Wed, 2 Mar 2005 00:15:31 +0000 (00:15 +0000)]
Merge b1_4_smallfix from b1_4 (20050301_1341)
- don't LASSERT in ll_release on NULL lld with NFS export (4655, 5760)
- hold NS lock when calling handle_ast_error->del_waiting_lock (5746)
- fix setattr mtime regression from lovcleanup merge (4829, 5669)
- workaround for 2.6 crash in ll_unhash_aliases (5687, 5210)
- small ext3 extents cleanups and fixes (5733)
- improved mballoc code, several small races and bugs fixed (5733, 5638)
- kernel version 43: fix remove_suid bugs in both 2.4 and 2.6 (5695)
- avoid needless client->OST connect, fix handle mismatch (5317)
- fix lustre/lustrefs init scripts for SuSE (patch from Scali, 5702)
- don't hold the pinger_sem in ptlrpc_pinger_sending_on_import
- change obd_increase_kms to obd_adjust_kms (up or down) (5654)
jacob [Tue, 1 Mar 2005 21:04:20 +0000 (21:04 +0000)]
b=1987
r=phil,adilger
Convert on-disk llog to native endian as needed.
yury [Tue, 1 Mar 2005 11:43:59 +0000 (11:43 +0000)]
- fixed comment of OBD_MD_FID.
eeb [Tue, 1 Mar 2005 10:45:10 +0000 (10:45 +0000)]
* Fixed ranal setup/teardown bug that ran a device's scheduler even
if the device wasn't present
nathan [Tue, 1 Mar 2005 02:37:26 +0000 (02:37 +0000)]
b=4834
merge from b1_4
phil [Tue, 1 Mar 2005 01:12:36 +0000 (01:12 +0000)]
More HP patch merging; part of misc_sfs_build_changes.patch
b=5785
Info:
Also correct a logic bug in the number of cpus calculation which
can lead to a shell error if the egrep fails and thus reports no
output.
phil [Tue, 1 Mar 2005 01:09:15 +0000 (01:09 +0000)]
b=5783
r=jacob
More HP patch integration. I will note jacob's reluctance to approve the
Requires hunk, commenting that he "fear[s] change"
Info:
HP SFS 1550/CFS 5783
Fix to ensure that the lustre-modules rpm is dependent on the
kernel rpm it is built against, and also to ensure it runs
depmod against the appropriate kernel install area to allow
for installing the lustre-modules rpm before rebooting on the
new kernel.
phil [Tue, 1 Mar 2005 00:35:07 +0000 (00:35 +0000)]
Some, but not all, of the GMnal updates submitted by HP.
b=5786
gmnal_kernel_changes.patch:
Info:
Kernel code changes for updated GMNAL as developed by Morgan
Doyle and maintained in the HP SFS product release.
gmnal_utils_changes.patch:
Info:
Userspace utils changes for updated GMNAL as developed by Morgan
Doyle and maintained in the HP SFS product release.
phil [Tue, 1 Mar 2005 00:33:10 +0000 (00:33 +0000)]
Turn a CERROR into a CWARN. HP patch quieten_console_errors-sfs1530.patch
b=4971
Info:
HP SFS 1530/CFS 4971
Patches to convert some LustreError messages to simple warning
Lustre messages where it seemed appropriate.
yury [Mon, 28 Feb 2005 19:21:46 +0000 (19:21 +0000)]
- returned back OBD_MD_FID. For many reasons its removal was wrong.
- fixed declation of OBD_MD.. flags. They should be declared like the following:
#define OBD_MD_FLALL (0xffffffffffffffffLL)
to show compiller explicitly that they are 64bit value and should considered as such.
- fixed md_reint_rename(). It should first check if objects are the same then check if tgt is subdir of src because for the same inodes is_subdir() returns 1 and mds_reint_rename() for renaming object into itself it returned -EINVAL;
- added asserts in LMV checking if body->valid contains valid fid before accessing it to check if object is splitted.
- in lmv_intent.c after md_intent_lock() checking not only error code, but also disposition, because there are cases when we're interested in name which does not exist, lock is not taken and policy_res2 is cleared in mds_intent_policy(). In this case we should do nothing (especially access body->id1 in LMV), but exit. This is because body->id1 is zero and (body->valid & OBD_MD_FID) == 0 because say mds_getattr_lock() failed before filling out as inode is not found.
yury [Mon, 28 Feb 2005 16:17:26 +0000 (16:17 +0000)]
- fixed ROOT setting up. If it is not empty, it should not be rewritten by "/r".
yury [Mon, 28 Feb 2005 12:25:39 +0000 (12:25 +0000)]
- small fix in ll_update_inode()
tappro [Mon, 28 Feb 2005 06:58:23 +0000 (06:58 +0000)]
initial release
green [Sun, 27 Feb 2005 20:17:08 +0000 (20:17 +0000)]
b=5230
r=adilger
Do not rehash dentries in ll_inode_revalidate_it(), it is only supposed to
update inodes.
yury [Sat, 26 Feb 2005 10:57:40 +0000 (10:57 +0000)]
- returning correct error code from mds_statfs() amd mds_sync(), as otherwise it may cause segfault in mds_handle().
tappro [Sat, 26 Feb 2005 08:56:34 +0000 (08:56 +0000)]
HLD document for SMFS(Storage Management Filesystem)
adilger [Sat, 26 Feb 2005 08:14:35 +0000 (08:14 +0000)]
Add b_cmd* to buildcvs to get correct portals version.
yury [Fri, 25 Feb 2005 19:45:18 +0000 (19:45 +0000)]
- locks and dentries should be released in the case request is being handled by wrong MDS
and should be restared with correct one in mds_getattr_lock().
eeb [Thu, 24 Feb 2005 17:32:39 +0000 (17:32 +0000)]
* vibnal HIGHMEM hacking
eeb [Thu, 24 Feb 2005 15:35:33 +0000 (15:35 +0000)]
* simpified vibnal rdma setup
eeb [Thu, 24 Feb 2005 14:57:37 +0000 (14:57 +0000)]
* Fixed the vibnal resilience-in-the-face-of-peer-crashes issue
alex [Thu, 24 Feb 2005 11:57:06 +0000 (11:57 +0000)]
- tuning ported from b1_4
eeb [Thu, 24 Feb 2005 11:01:55 +0000 (11:01 +0000)]
* lustre updates for recent portals HEAD changes
- added vib
- zeroconf vib, ra
- lconf/lmc openib,vib changes
eeb [Wed, 23 Feb 2005 21:43:25 +0000 (21:43 +0000)]
* First cut working vibnal
eeb [Wed, 23 Feb 2005 21:34:42 +0000 (21:34 +0000)]
* removed unused #defines in RANAL
* fixed a misleading comment in RANAL
yury [Wed, 23 Feb 2005 14:59:02 +0000 (14:59 +0000)]
- fix from b1_4 in MDS_CHECK_RESENT()
- more fixes about fid extracting on MDS. By now correct fid is always should
be returned to client.
adilger [Wed, 23 Feb 2005 12:05:27 +0000 (12:05 +0000)]
Branch: b1_4
Sort list alphabetically, like it says in the comment.
liuduo [Wed, 23 Feb 2005 06:59:26 +0000 (06:59 +0000)]
add some code related to setxid
liuduo [Wed, 23 Feb 2005 03:35:36 +0000 (03:35 +0000)]
to make b_hd_remote_uid portals based on b_hd_sec portals
alex [Tue, 22 Feb 2005 19:28:49 +0000 (19:28 +0000)]
- commit from zzeng
yury [Tue, 22 Feb 2005 16:37:48 +0000 (16:37 +0000)]
- changes with names on exports and another fields in llite and mds. lov_exp is
renamed to dt_exp and lmv_exp is renamed to md_exp. Related changes.
- removed OBD_MD_FID. And fid is now fetched on MDS on all constructive requests. It
is not fetched on unlink(), rename(), etc. This is because OBD_MD_FID flag lost
the sense as lmv layer needs it for all operations (to check if object is splitted).
adilger [Fri, 18 Feb 2005 23:38:51 +0000 (23:38 +0000)]
Branch: b1_4
Ignore generated files.
nic [Thu, 17 Feb 2005 20:43:11 +0000 (20:43 +0000)]
ia64 needs make vmlinux on 2.6-suse, and it works on rhel-2.4 as well
jacob [Tue, 15 Feb 2005 21:31:18 +0000 (21:31 +0000)]
b=5684
r=adilger
First stab at improving some user visible error messages.
nic [Tue, 15 Feb 2005 06:07:26 +0000 (06:07 +0000)]
need these changes on HEAD for build too
eeb [Mon, 14 Feb 2005 15:30:19 +0000 (15:30 +0000)]
* Added loopback optimisation to lib-move.c
lctl --net ??? loopback # show current state
lctl --net ??? loopback on # enable loopback optimisation
lctl --net ??? loopback off # disable loopback optimisation
By default loopback optimisation is turned OFF
eeb [Sat, 12 Feb 2005 11:11:53 +0000 (11:11 +0000)]
* fixed bug in handling timed out RapkSetRiParams()
phil [Sat, 12 Feb 2005 06:23:38 +0000 (06:23 +0000)]
I keep finding these trees that I have, with little uncommitted copyright
changes, and wondering how that happened.
eeb [Sat, 12 Feb 2005 03:23:36 +0000 (03:23 +0000)]
* Coded for new connection handshake.
eeb [Fri, 11 Feb 2005 20:22:44 +0000 (20:22 +0000)]
* ghastly hack to serialise RapkSetRiParams with comms scheduler thread
* Ranal "working" @ 1.2GBytes/sec single channel, 2.4GBytes/sec double
nic [Fri, 11 Feb 2005 02:52:13 +0000 (02:52 +0000)]
only build ksocknal with modules enabled
jacob [Fri, 11 Feb 2005 01:24:30 +0000 (01:24 +0000)]
originally by phil
do not dist non-linux (darwin) files by default.
phil [Thu, 10 Feb 2005 23:09:22 +0000 (23:09 +0000)]
removed some empty files (authors, news, readme)
added a license file
fixed up some stray copyright notices
adilger [Thu, 10 Feb 2005 22:23:29 +0000 (22:23 +0000)]
Fix LC_COLLATE so that [A-Z] only matches upper-case letters.
eeb [Wed, 9 Feb 2005 12:33:10 +0000 (12:33 +0000)]
* Bug 5676 fix: keep extra ref on peer until all done in
ksocknal_del_peer_locked CVS:
eeb [Tue, 8 Feb 2005 01:15:18 +0000 (01:15 +0000)]
* removed spurious 'inline' from extern portal_ioctl_getdata() decl
liangzhen [Sat, 5 Feb 2005 08:47:01 +0000 (08:47 +0000)]
b_port_step is a branch only for portals
jacob [Sat, 5 Feb 2005 01:18:13 +0000 (01:18 +0000)]
i did not mean for this to get committed
jacob [Sat, 5 Feb 2005 00:53:52 +0000 (00:53 +0000)]
sort alphabetically
jacob [Sat, 5 Feb 2005 00:53:16 +0000 (00:53 +0000)]
remove arch symlink