- May 13, 2008
-
-
Andreas Dilger authored
Update build version to 1.9.50, to avoid any confusion that this is a pre-1.8 branch.
-
Pravin Shelar authored
b=14340 i=h.huang i=nikita.danilov lu_dirent structure was inconsistent after lu_dirent->lde_hash element extension to 64 bit change patch. size of structure is different on 32 and 64 bit platform. patch fixes that bug.
-
Bobi Jam authored
b=12888 i=johann, green Description: mds_mfd_close() ASSERTION(rc == 0) Details : In mds_mfd_close(), we need protect inode's writecount change within its orphan write semaphore to prevent possible races.
-
- May 12, 2008
-
-
Elena Gryaznova authored
i=Adilger keep conf-sanity test_32a and test_32b in sync with b1_6
-
Alexey Lyashkov authored
-
kalpak authored
Reinitialize optind to 0 on each lfs_* call, in case lfs is being used interactively. Most lfs_* functions do this already, but a few (find, setstripe, quotachown) do not. b=15212 i=rupesh i=johann
-
Yong Fan authored
b=15563 update buildcvs
-
Andreas Dilger authored
Fix typo in error message.
-
- May 11, 2008
-
-
bwzhou authored
b=15521 r=adilger, grev the test case for bug 15521
-
- May 10, 2008
-
-
Isaac Huang authored
i=liangzhen,i=maxim: - ensure UNLINK event when MD unlinked via LNetMEUnlink. - removed trailing spaces.
-
Walter Poxon authored
i=walter Land Andreas' original fix for catamount build errors
-
Walter Poxon authored
i=adilger fix compile warnings which are treated as errors on Cray XT3 catamount build
-
- May 09, 2008
-
-
bwzhou authored
b=15521 r=adilger, bobijam fix the wrong lprcfs stats
-
Andreas Dilger authored
Allow mds2, mds3, mds4 to be specified on other hosts.
-
bwzhou authored
b=14740 r=johann, bobijam allocating one percpu struct each time instead of all in one time
-
bwzhou authored
b=14219 r=adilger, shadow use obd_punch_rqset() instead of obd_setattr() to change size
-
bwzhou authored
b=14034 r=adilger, bobijam error handling for the ENOMEM case when failing to allocate ldlm resource
-
Terry Rutledge authored
-
komaln authored
r=Wangdi, Adilger To remove direct instances of proc file system.
-
Alexey Lyashkov authored
b=14929 i=zam
-
Alexey Lyashkov authored
b=14645 i=johan i=alex
-
bwzhou authored
b=13707 r=bobijam, johann exported debug_file_path as a module param of libcfs
-
Elena Gryaznova authored
i=Adilger disable LIBLUSTRE test for NETTYPE != tcp
-
- May 08, 2008
-
-
Brian J. Murrell authored
Small logic thinko that fails the case where no o2iblnd tree is specifed.
-
Elena Gryaznova authored
i=Adilger sanity test_118* fixes; set fail loc on all oss nodes
-
Alexey Lyashkov authored
b=14949
-
Alexey Lyashkov authored
b=14949 i=johan i=deen
-
Alexander Zarochentsev authored
i=vitaly.vertman i=oleg.drokin Lockless i/o and lockless truncate code and sanityN tests.
-
- May 07, 2008
-
-
Walter Poxon authored
i=he.huang i=adilger fix for build errors on Cray catamount builds
-
Walter Poxon authored
i=he.huang i=adilger fix for user-tcpip.c build errors on Cray catamount builds
-
Brian J. Murrell authored
i=cliffw i=wangyb Build OFED 1.3 kernel-ib* RPMs for the vendor kernels we support patchless clients on (RHEL5 and SLES10). Build Lustre with the above OFED 1.3 support (i.e. for patchless clients).
-
Johann Lombardi authored
b=15459 i=adilger i=bzzz disable preallocation for non-regular files. although our patchsets only enable extents on regular files, we should prevent crashing.
-
Bobi Jam authored
b=15575 i=wangdi, johann Description: Stack overflow during MDS log replay Details : ease stack pressure by using a thread dealing llog_process.
-
- May 06, 2008
-
-
Alexey Lyashkov authored
b=15278 i=johann i=deen
-
komaln authored
Added Changelog entry.
-
komaln authored
r=Nathan Rutman, Alexey Lyashkov Ported the patch 'to organize the stats on per-nid basis instead of the existing per-export basis' to HEAD. This patch also includes all its related bug fixes(bug 14016, bug 14370, bug 14873).
-
- May 05, 2008
-
-
Isaac Huang authored
forgot to update changelog.
-
Liang Zhen authored
Backout patch of b13983, checking and updating ksnp_proto and adding conn to ksnp_conns should always happen atomically, keep extra reference count of socket during establishing of connection. b=15712 i=adilger i=liangzhen
-
- May 02, 2008
-
-
Johann Lombardi authored
b=15517 i=johann i=umka Fix compile warnings on i686.
-
- Apr 30, 2008
-
-
Mikhail Pershin authored
-