Whamcloud - gitweb
fs/lustre-release.git
16 years agoReplace /usr/sbin and /usr/bin filelists with wildcards.
brian [Wed, 30 Mar 2005 19:40:10 +0000 (19:40 +0000)]
Replace /usr/sbin and /usr/bin filelists with wildcards.

16 years ago* Removed '-Wall' from ranal build
eeb [Wed, 30 Mar 2005 18:06:51 +0000 (18:06 +0000)]
*    Removed '-Wall' from ranal build

16 years ago* Fixed trailing ';' on #define PORTALS_CFG_VERSION
eeb [Wed, 30 Mar 2005 18:06:03 +0000 (18:06 +0000)]
*    Fixed trailing ';' on #define PORTALS_CFG_VERSION

16 years ago* Changed --with-iib to allow a path to the installed Infinicon headers
eeb [Wed, 30 Mar 2005 18:05:07 +0000 (18:05 +0000)]
*    Changed --with-iib to allow a path to the installed Infinicon headers

16 years agoall aboard! next stop, testing your build before submitting requests to buffalo!
jacob [Wed, 30 Mar 2005 17:08:43 +0000 (17:08 +0000)]
all aboard!  next stop, testing your build before submitting requests to buffalo!

16 years agoBranch: b_cray
adilger [Wed, 30 Mar 2005 02:42:53 +0000 (02:42 +0000)]
Branch: b_cray
Remove l_getgroups from RPM build, as this is not present in any branch
except HEAD.  We need some sort of simple configure test to determine if
this should be included in the RPM or not.

16 years agoBranch: b_cray
adilger [Wed, 30 Mar 2005 01:03:26 +0000 (01:03 +0000)]
Branch: b_cray
Quiet liblustre compile warning.

16 years agoBranch: b1_4
adilger [Tue, 29 Mar 2005 20:09:03 +0000 (20:09 +0000)]
Branch: b1_4
Ignore backup series files.

16 years agoEnable SMFS build.
buffalo-pull [Tue, 29 Mar 2005 09:08:01 +0000 (09:08 +0000)]
Enable SMFS build.
Include /usr/sbin/l_getgroups in the RPM.

16 years ago- update from HEAD.
yury [Tue, 29 Mar 2005 08:45:22 +0000 (08:45 +0000)]
- update from HEAD.

16 years agoNeed to include smfs_api.h in the distributed sources (i.e. make dist).
buffalo-pull [Tue, 29 Mar 2005 08:14:23 +0000 (08:14 +0000)]
Need to include smfs_api.h in the distributed sources (i.e. make dist).

16 years agoHEAD is now at 1.7.0. Mostly cosmetic for building RPMS and so forth.
buffalo-pull [Tue, 29 Mar 2005 08:09:33 +0000 (08:09 +0000)]
HEAD is now at 1.7.0.  Mostly cosmetic for building RPMS and so forth.

16 years agoFix needed so make dist works again.
buffalo-pull [Tue, 29 Mar 2005 07:49:57 +0000 (07:49 +0000)]
Fix needed so make dist works again.

16 years agoget rid of old wrong patch
alex [Tue, 29 Mar 2005 07:09:20 +0000 (07:09 +0000)]
get rid of old wrong patch

16 years agob1_4_quotaoff needs same portals as b1_4 too
green [Tue, 29 Mar 2005 06:00:51 +0000 (06:00 +0000)]
b1_4_quotaoff needs same portals as b1_4 too

16 years agor=phil
jacob [Tue, 29 Mar 2005 05:27:18 +0000 (05:27 +0000)]
r=phil

remove unused files.

16 years agoClean out this change in my tree; don't install the liblustre file.
jacob [Tue, 29 Mar 2005 05:02:16 +0000 (05:02 +0000)]
Clean out this change in my tree; don't install the liblustre file.

16 years agoChangelog:
zhaoqiang [Tue, 29 Mar 2005 03:00:34 +0000 (03:00 +0000)]
Changelog:
- fill lib_qry_user()
- define ptl_netid_t

16 years agothe basic code/patch for remote permission, including the fix for
wangchao [Mon, 28 Mar 2005 06:22:02 +0000 (06:22 +0000)]
the basic code/patch for remote permission, including the fix for
defects 36-40 (because b_hd_acl was always not ready on b_hd_remote_uid,
the fix of the defects were on local).
d = 36-40

16 years agoBranch: b_cray
adilger [Sat, 26 Mar 2005 01:40:44 +0000 (01:40 +0000)]
Branch: b_cray
Fix liblustre test compilation because kmap is not defined (nor should it
be) for user-space NALs.
b=5878
r=gcc

16 years agoBranch: b1_4
adilger [Fri, 25 Mar 2005 22:51:03 +0000 (22:51 +0000)]
Branch: b1_4
The compatibility macro for cpu_online() had reversed parameters.
b=5929
r=fergal (original patch)

16 years ago- fixed ability of variables being used uninitialized in mds_get_parent_child(ren...
yury [Fri, 25 Mar 2005 13:46:03 +0000 (13:46 +0000)]
- fixed ability of variables being used uninitialized in mds_get_parent_child(ren)_locked()
d=85

16 years ago- fixed ability of ret_mode being used uninitialized in mds_lock_mode_for_dir()
yury [Fri, 25 Mar 2005 13:44:27 +0000 (13:44 +0000)]
- fixed ability of ret_mode being used uninitialized in mds_lock_mode_for_dir()
d=84

16 years agoMerge b1_4_quota from b1_4 (20050325_1039)
lsy [Fri, 25 Mar 2005 06:48:39 +0000 (06:48 +0000)]
Merge b1_4_quota from b1_4 (20050325_1039)

+ - mdc_enqueue does not need max_mds_easize request buffer on send (5707)
+ - swab llog records of type '0' so we get proper header size/idx (5861)
+ - send llog cancel req to DLM cancel portal instead of cb portal (5515)
+ - allow --write-conf on an MDS with different nettype than client (5619)
+ - don't write config llogs to MDS for mounts not from that MDS (5617)
+ - lconf should create multiple TCP connections from a client (5201)
+ - init scripts are now turned off by default; run chkconfig --on
+   lustre and chkconfig --on lustrefs to use them
+ - fix deadlock in obdfilter statistics vs. object create (5811)
+ - use time_{before,after} to avoid timer jiffies wrap (5882)
+ - shutdown --force/--failover stability (3607,3651,4797,5203,4834)
+ - Do not leak request if server was not able to process it (5154)
+ - If mds_open unable to find parent dir, make that negative lookup(5154)
+ - support for RHEL4 kernel on i686 (5773)
+ - provide error messages when incompatible logs are encountered (5898)

since b1_4_bgl landed on b1_4, b1_4_quota starts to use the same branch of portals.

16 years agodon't break things on purpose
jacob [Thu, 24 Mar 2005 22:58:03 +0000 (22:58 +0000)]
don't break things on purpose

16 years agob1_4 lustre needs b1_4_bgl portals until after 1.4.2 is released
jacob [Thu, 24 Mar 2005 22:51:10 +0000 (22:51 +0000)]
b1_4 lustre needs b1_4_bgl portals until after 1.4.2 is released

16 years agobranch: b_hd_acl
ericm [Thu, 24 Mar 2005 19:22:56 +0000 (19:22 +0000)]
branch: b_hd_acl
re-parent b_hd_acl to b_hd_sec, not test yet.

16 years agoBranch: HEAD
adilger [Wed, 23 Mar 2005 18:54:34 +0000 (18:54 +0000)]
Branch: HEAD
Move clearpatches.sh and confirmpatches.sh into build from lustre-core.

16 years agojust set portals tag on any v1_4 build
nic [Wed, 23 Mar 2005 04:16:57 +0000 (04:16 +0000)]
just set portals tag on any v1_4 build

16 years agob=5881
tappro [Mon, 21 Mar 2005 22:29:24 +0000 (22:29 +0000)]
b=5881
r=alex
d=17

fix design defect. plugins are needed struct inode to check its flags

16 years agob=5881
tappro [Mon, 21 Mar 2005 13:27:18 +0000 (13:27 +0000)]
b=5881
r=wangdi

small fixes to provide correct operation order, fix double messages due to using both RETURN and EXIT macro, always initialize several s_ops, plugin framework was added.

16 years agoedit code that was changed by latest commits
tappro [Mon, 21 Mar 2005 08:10:41 +0000 (08:10 +0000)]
edit code that was changed by latest commits

16 years agob=5881
tappro [Sat, 19 Mar 2005 12:21:22 +0000 (12:21 +0000)]
b=5881
r=wangdi

smfs fixes according with dld.

16 years ago* forgot to re-do wirecheck on last commit (vibnal 5890 fix)
eeb [Fri, 18 Mar 2005 15:34:08 +0000 (15:34 +0000)]
*    forgot to re-do wirecheck on last commit (vibnal 5890 fix)

16 years ago* "proper" fix for 5890: fixed the misplaced vibnal wire struct
eeb [Fri, 18 Mar 2005 11:48:47 +0000 (11:48 +0000)]
*   "proper" fix for 5890: fixed the misplaced vibnal wire struct
    __attribute__((packed)) statements, bumped wire protocol version and
    rebuilt wire struct assertions.

*   Added new vibnal_wire.h into autoMakefile.am

16 years agoIgnore generated file.
adilger [Fri, 18 Mar 2005 08:55:55 +0000 (08:55 +0000)]
Ignore generated file.

16 years agoBranch: b1_4_rhel4
adilger [Fri, 18 Mar 2005 00:27:56 +0000 (00:27 +0000)]
Branch: b1_4_rhel4
Use the HEAD portals branch for RHEL4.

16 years agoBranch: HEAD
adilger [Thu, 17 Mar 2005 23:22:39 +0000 (23:22 +0000)]
Branch: HEAD
Build fix for RHEL4.
b=5733
r=green (original patch)

16 years ago* fix for 5890: different compilers produce different wire struct offsets
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.

16 years ago* vibnal fix + earlier arp failure reporting for bug 5852
eeb [Thu, 17 Mar 2005 18:54:57 +0000 (18:54 +0000)]
*    vibnal fix + earlier arp failure reporting for bug 5852

16 years ago- added check for existing TAG file before catting it
yury [Thu, 17 Mar 2005 08:50:46 +0000 (08:50 +0000)]
- added check for existing TAG file before catting it

16 years ago*** empty log message ***
zhaoqiang [Thu, 17 Mar 2005 03:11:10 +0000 (03:11 +0000)]
*** empty log message ***

16 years ago- find_assigned_loop() should check all loop devices and do not return after first...
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.

16 years ago- fix in test_16 in conf sanity
yury [Wed, 16 Mar 2005 11:49:32 +0000 (11:49 +0000)]
- fix in test_16 in conf sanity

16 years ago* fix for 5876 to change vibnal connection request/response message
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.

16 years agob=5881
tappro [Wed, 16 Mar 2005 10:57:38 +0000 (10:57 +0000)]
b=5881
r=wangdi

fixes in accordance with SMFS changes

16 years agob=5881
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.

16 years ago- fix in mds_cancel_cookie_cb() with freeing lsm.
yury [Wed, 16 Mar 2005 08:05:07 +0000 (08:05 +0000)]
- fix in mds_cancel_cookie_cb() with freeing lsm.

16 years agoadd debug mask for quota
niu [Wed, 16 Mar 2005 07:37:57 +0000 (07:37 +0000)]
add debug mask for quota

16 years agob=5874
alex [Tue, 15 Mar 2005 14:20:10 +0000 (14:20 +0000)]
b=5874

  additional debug info

16 years ago- returned back old schema of mds_verify_child()
yury [Tue, 15 Mar 2005 07:41:35 +0000 (07:41 +0000)]
- returned back old schema of mds_verify_child()

16 years ago- fixed wrongly inverted condition in lmv_create_obj()
yury [Tue, 15 Mar 2005 07:25:03 +0000 (07:25 +0000)]
- fixed wrongly inverted condition in lmv_create_obj()

16 years agob=5765
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

16 years ago- memory leak fixes from b_fid.
yury [Mon, 14 Mar 2005 19:16:37 +0000 (19:16 +0000)]
- memory leak fixes from b_fid.

16 years ago- more checks and verbosity in test_16 of conf-sanity.sh to see what happens when...
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.

16 years ago- mds_verify_child() is converted to use fid component of lustre_id instead of pair...
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.

16 years agob=5863
alex [Fri, 11 Mar 2005 22:17:22 +0000 (22:17 +0000)]
b=5863

- assertion was wrong

16 years ago- fixed using not initialized ctxt with pop_ctxt() in mds_lock_and_check_slave()
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()

16 years ago* fixed bug tracefile.c: tage_alloc() as encountered in bug 5857
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

16 years ago* fixed 5000: socknal autoconnect race (outgoing v. incoming)
eeb [Fri, 11 Mar 2005 11:10:46 +0000 (11:10 +0000)]
*    fixed 5000: socknal autoconnect race (outgoing v. incoming)

*    minor socknal cleanups

16 years agoexp might be NULL
wangdi [Fri, 11 Mar 2005 09:37:41 +0000 (09:37 +0000)]
exp might be NULL
B: HEAD

16 years ago- update from HEAD
yury [Fri, 11 Mar 2005 09:17:21 +0000 (09:17 +0000)]
- update from HEAD

16 years agob=5209
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

16 years ago* fix for 5809: vibnal tx_sending race
eeb [Thu, 10 Mar 2005 16:25:36 +0000 (16:25 +0000)]
*   fix for 5809: vibnal tx_sending race

16 years agoturn debugging on by default for now
nikita [Wed, 9 Mar 2005 11:39:00 +0000 (11:39 +0000)]
turn debugging on by default for now

16 years agoBranch: HEAD
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

16 years ago- fix against credits leak in journal_release_buffer()
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)

16 years agouptodate with HEAD, resovle lots of conflictions
liuduo [Tue, 8 Mar 2005 03:43:35 +0000 (03:43 +0000)]
uptodate with HEAD, resovle lots of conflictions

16 years agoBranch: b1_2
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).

16 years ago* further lctl debug_file fixes
eeb [Mon, 7 Mar 2005 12:24:23 +0000 (12:24 +0000)]
*   further lctl debug_file fixes
    - truncate output file
    - close streams

16 years ago* vibnal fixes
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

16 years ago- different fixes in MDS
yury [Sun, 6 Mar 2005 18:58:33 +0000 (18:58 +0000)]
- different fixes in MDS

16 years agoBranch: b1_4
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

16 years agoMerge b1_4_bgl from b1_4 (20050304_1757)
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)

16 years agoall 2.6 platforms need the -objdir/Module.symvers built
nic [Fri, 4 Mar 2005 17:21:22 +0000 (17:21 +0000)]
all 2.6 platforms need the -objdir/Module.symvers built

16 years agoset LINUX26 in 2.6-rhel4 to get correct building instructions
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

16 years agob=5599
alex [Thu, 3 Mar 2005 19:23:12 +0000 (19:23 +0000)]
b=5599

serialize orphan cleanups between OSTs

16 years ago* Fix for 5806: vibnal NULL arp_data deref
eeb [Thu, 3 Mar 2005 15:44:06 +0000 (15:44 +0000)]
*  Fix for 5806: vibnal NULL arp_data deref

16 years ago* fix for 5722 openibnal: NULL dereference in kibnal_close_conn_locked
eeb [Thu, 3 Mar 2005 12:11:18 +0000 (12:11 +0000)]
*   fix for 5722 openibnal: NULL dereference in kibnal_close_conn_locked

16 years agoPatch from HP to support external qsnet.
jacob [Wed, 2 Mar 2005 21:26:01 +0000 (21:26 +0000)]
Patch from HP to support external qsnet.

16 years agob=5786
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.

16 years agoremove the lustre version from the kernel EXTRA_VERSION to help get long branch
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

16 years ago* Fixed 5708: openib reconnect on client reboot issues
eeb [Wed, 2 Mar 2005 19:10:12 +0000 (19:10 +0000)]
*   Fixed 5708: openib reconnect on client reboot issues

16 years agolustre b1_4_rhel4 requires portals b1_4_rhel4
green [Wed, 2 Mar 2005 18:06:51 +0000 (18:06 +0000)]
lustre b1_4_rhel4 requires portals b1_4_rhel4

16 years ago- fixes from b1_4:
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()

16 years ago- fixed sanity test 55.
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.

16 years ago- fixed md_getattr() in lustre_common-fill_super()
yury [Wed, 2 Mar 2005 07:51:40 +0000 (07:51 +0000)]
- fixed md_getattr() in lustre_common-fill_super()

16 years agoMerge b1_4_smallfix from b1_4 (20050301_1341)
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)

16 years agob=1987
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.

16 years ago- fixed comment of OBD_MD_FID.
yury [Tue, 1 Mar 2005 11:43:59 +0000 (11:43 +0000)]
- fixed comment of OBD_MD_FID.

16 years ago* Fixed ranal setup/teardown bug that ran a device's scheduler even
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

16 years agob=4834
nathan [Tue, 1 Mar 2005 02:37:26 +0000 (02:37 +0000)]
b=4834
merge from b1_4

16 years agoMore HP patch merging; part of misc_sfs_build_changes.patch
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.

16 years agob=5783
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.

16 years agoSome, but not all, of the GMnal updates submitted by HP.
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.

16 years agoTurn a CERROR into a CWARN. HP patch quieten_console_errors-sfs1530.patch
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.

16 years ago- returned back OBD_MD_FID. For many reasons its removal was wrong.
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.

16 years ago- fixed ROOT setting up. If it is not empty, it should not be rewritten by "/r".
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".

16 years ago- small fix in ll_update_inode()
yury [Mon, 28 Feb 2005 12:25:39 +0000 (12:25 +0000)]
- small fix in ll_update_inode()

16 years agoinitial release
tappro [Mon, 28 Feb 2005 06:58:23 +0000 (06:58 +0000)]
initial release