Permalink
Switch branches/tags
v2_4_0-1dkms v2_3_65-1dkms v2_3_64-1dkms orion-2_3_49_92_1-71chaos orion-2_3_49_92_1-70chaos orion-2_3_49_54_2-75chaos orion-2_3_49_54_2-69chaos orion-2_3_49_54_2-68chaos orion-2_3_49_54_2-63chaos orion-2_3_49_54_2-62chaos orion-2_3_49_54_2-61chaos orion-2_3_49_54_2-60chaos orion-2_3_49_54_2-59chaos orion-2_3_49_54_1-58chaos orion-2_3_49_54_1-57chaos orion-2_3_49_54_1-56chaos orion-2_3_49_54_1-55chaos orion-2_3_49_54_1-54chaos orion-2_3_49_54_1-53chaos orion-2_3_49_54_1-52chaos orion-2_2_49_57_2-45chaos orion-2_2_49_57_1-44chaos orion-2_2_49_57_1-43chaos orion-2_2_49_57_1-42chaos orion-2_2_49_57_1-41chaos orion-2_2_49_56_9-40chaos orion-2_2_49_56_8-39chaos orion-2_2_49_56_7-38chaos orion-2_2_49_56_7-37chaos orion-2_2_49_56_7-36chaos orion-2_2_49_56_7-35chaos orion-2_2_49_56_7-34chaos orion-2_2_49_56_7-33chaos orion-2_2_49_56_7-32chaos orion-2_2_49_56_6-31chaos orion-2_2_49_56_6-30chaos orion-2_2_49_56_5-29chaos orion-2_2_49_56_5-28chaos orion-2_2_49_56_5-27chaos orion-2_2_49_56_5-26chaos orion-2_2_49_56_4-25chaos orion-2_2_49_56_4-24chaos orion-2_2_49_56_3-23chaos orion-2_2_49_56_3-22chaos orion-2_2_49_56_2 orion-2_2_49_56_2-21chaos orion-2_2_49_56_1 orion-2_2_49_56_1-20chaos orion-2_2_49_56_1-19chaos orion-2_2_49_50_7 orion-2_2_49_50_7-18chaos orion-2_2_49_50_6 orion-2_2_49_50_6-17chaos orion-2_2_49_50_6-16chaos orion-2_2_49_50_5 orion-2_2_49_50_5-15chaos orion-2_2_49_50_5-14chaos orion-2_2_49_50_4 orion-2_2_49_50_4-13chaos orion-2_2_49_50_4-12chaos orion-2_2_49_50_4-11chaos orion-2_2_49_50_3 orion-2_2_49_50_3-10chaos orion-2_2_49_50_3-9chaos orion-2_2_49_50_3-8chaos orion-2_2_49_50_3-7chaos orion-2_2_49_50_3-6chaos orion-2_2_49_50_3-5chaos orion-2_2_49_50_2 orion-2_2_49_50_2-4chaos orion-2_2_49_50_1 orion-2_2_49_50_1-3chaos orion-2_2_49_50_1-2chaos orion-2_2_49_50_1-1chaos 2.10.3_3.chaos 2.10.3_2.chaos 2.10.3_1.chaos 2.10.2_1.chaos 2.8.2_1.chaos 2.5.5-15chaos 2.5.3-6chaos 2.5.3-5chaos 2.5.3-4chaos 2.5.3-3chaos 2.5.3-2chaos 2.5.3-1zfs 2.5.3-1chaos 2.4.2-17chaos 2.4.2-16chaos 2.4.2-15chaos 2.4.2-14.1chaos 2.4.2-14chaos 2.4.2-13chaos 2.4.2-12chaos 2.4.2-11chaos 2.4.2-10chaos 2.4.2-9chaos 2.4.2-8chaos 2.4.2-7chaos 2.4.2-6chaos
Nothing to show
Commits on Jul 13, 2012
  1. LU-1621 capa: disable Lustre capa by force

    Fan Yong authored and morrone committed Jul 11, 2012
    Current Lustre capa framework has some security holes, and cannot
    work as expected. It also affects the performance because of the
    overhead of calculating Lustre capa. So disable it by force until
    patched or new Lustre capa implemented.
    
    Signed-off-by: Fan Yong <yong.fan@whamcloud.com>
    Change-Id: If6d7514d0b888c589adcc50e397f53395b7c2b0d
Commits on Jul 12, 2012
  1. LU-1597 obdfilter: honor disabled capa

    morrone committed Jul 12, 2012
    Make obdfilter recognize when capabilities are disabled
    in one additional place where it was missed.
    
    Signed-off-by: Christopher J. Morrone <morrone2@llnl.gov>
    Change-Id: I28f10ad1227759a5ea725fec7aaa885970d09f24
Commits on Jul 6, 2012
  1. LU-1442 llite: cleanup if a page failed to add into cache

    Jinshan Xiong authored and morrone committed Jun 27, 2012
    In lustre, we assume that a dirty page must be queued in osc cache
    for writing. However, in vvp_io_commit_write(), if a page failed to
    add into cache, page dirty flag isn't cleared this will cause the
    page will never be added into cache again.
    
    Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Change-Id: I4e2eb61d25e97daf9d32ba2f5c57223d77763b5b
    Patch-Set: 1
  2. LU-1576 llite: correct page usage count

    Bobi Jam authored and morrone committed Jul 2, 2012
    If kernel has add_to_page_cache_lru(), the ll_pagevec_add() is defined
    as an empty function, while page_cache_get(page) only makes sense if
    ll_pagevec_add() is defined.
    
    This patch moves page_cache_get into ll_pagevec_add() macro
    definition.
    
    Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
    Change-Id: Iad98aacff43beec3e7a64fd1a778f549250aa5b8
Commits on Jun 21, 2012
  1. LU-1428 ldlm: fix a race in ldlm_lock_destroy_internal

    liangzhen authored and morrone committed Jun 5, 2012
    ldlm_lock::l_exp_hash should be protected by internal lock of
    cfs_hash, but we called cfs_hlist_unhashed(lock::l_exp_hash)
    w/o holding cfs_hash lock in ldlm_lock_destroy_internal,
    which means if someone called ldlm_lock_cancel on a lock while
    export::exp_lock_hash is in progress of rehashing (thread context of
    cfs_workitem), there could be tiny window between deleting this lock
    from bucket[A] and re-adding it to bucket[B] of l_exp_hash, and
    cfs_hlist_unhashed(lock::l_exp_hash) will return 1 in this window,
    then we destroyed a lock but left it on l_exp_hash forever because
    lock::l_destroyed has been set to 1 and ldlm_lock_destroy_internal()
    wouldn't be able to remove the lock from l_exp_hash even it's called
    infinite times in ldlm_cancel_locks_for_export_cb().
    
    This patch also added some debug information to
    ldlm_cancel_locks_for_export_cb in case this patch can't fix this
    problem.
    
    Signed-off-by: Liang Zhen <liang@whamcloud.com>
    Change-Id: Ia0932658b3f085a55535e36bee4fb833e74fa242
Commits on Jun 18, 2012
  1. LU-1467 ost: ASSERTION(lock->l_req_mode == lock->l_granted_mode)

    yangsheng authored and morrone committed Jun 6, 2012
    The lock may be cancel while ost_prolong_lock_one invoked, so
    just return in this case.
    
    Signed-off-by: yang sheng <ys@whamcloud.com>
    Change-Id: Ica6ad9199e4b210145e99d2420925803b18a7edd
    Reviewed-on: http://review.whamcloud.com/3042
    Tested-by: Hudson
    Reviewed-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Reviewed-by: Niu Yawei <niu@whamcloud.com>
    Tested-by: Maloo <whamcloud.maloo@gmail.com>
    Reviewed-by: Oleg Drokin <green@whamcloud.com>
Commits on Jun 5, 2012
  1. LU-1299 clio: a combo patch to fix cl_lock

    Jinshan Xiong authored and morrone committed May 2, 2012
    In this patch, the following issues are fixed:
    * redefine cl_lock_error(). There are two kinds of error when
      enqueuing a cl_lock: internal errors return from server side and
      local error for example interrupted by signal when waiting for a
      lock. Obviously the latter one is not fatal so we shouldn't err the
      lock out;
    * for fault handling, only signal SIGKILL and SIGTERM is not allowed;
    * handle return code of ll_fault correctly;
    * redefine ->unuse() to release resource it held if enqueuing lock
      failed;
    * in cl_lock_hold_release(), only CLS_CACHED mode lock will be cached,
      for locks in other states, they will be cancelled voluntarily;
    * fix the problem that cfs_block_sigs() is wrongly used as
      cfs_restore_sigs() and implementation of cfs_block_sigs() was wrong
    * other minor fixes.
    
    Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Change-Id: Iffce8be356723781b8f33ec9bdc2cf73e9e07138
    Patch-Set: 10
  2. Revert "LU-1299 clio: a combo patch to fix cl_lock"

    morrone committed Jun 5, 2012
    This reverts commit 7952d97.
Commits on May 29, 2012
  1. LU-1234 dcache: don't drop invalid dentry arbitrarily

    Lai Siyao authored and morrone committed Mar 28, 2012
    This is a backport of part of LU-506 dcache scalability support:
    * remove super hack d_rehash_cond(), and treats DCACHE_LUSTRE_INVALID
      similar to DCACHE_DISCONNECTED, therefore dentry doesn't need to
      be dropped and rehashed frequently.
    * .lookup(LOOKUP_CREATE) calls d_add() dentry directly, and .create
      only needs to d_instantiate() this dentry.
    * other cleanups.
    
    Signed-off-by: Lai Siyao <laisiyao@whamcloud.com>
    Change-Id: Ie169bc7e763e6891084999041aac9f62c8dee9f0
    Patch-Set: 4
Commits on May 21, 2012
  1. LU-1280 ldiskfs: remove EXT_ASSERT from ext3_ext_new_extent_cb()

    Yu Jian authored and morrone committed May 17, 2012
    The EXT_ASSERT() in ext3_ext_new_extent_cb() is invalid since
    new locking is introduced in ext4_ext_walk_space().
    
    Signed-off-by: Yu Jian <yujian@whamcloud.com>
    Change-Id: I00d81b6fe89bd0b7a95eae4b336eb10927240e01
Commits on May 18, 2012
  1. LU-980 llog: cleanup return value in llog_client_create

    Hongchao Zhang authored and morrone committed Jan 12, 2012
    in llog_client_create, the newly allocated llog_handle is
    return by parameter res, but it doesn't be cleaned up
    if the following operations failed and the corresponding
    llog_handle is already freed.
    
    Signed-off-by: Hongchao Zhang <hongchao.zhang@whamcloud.com>
    Signed-off-by: Bob Glossman <bogl@whamcloud.com>
    Change-Id: Ib8c40c53b071fff7de3550a39f009915cb8511a7
    Reviewed-on: http://review.whamcloud.com/2806
    Tested-by: Hudson
    Reviewed-by: Fan Yong <yong.fan@whamcloud.com>
    Reviewed-by: Bobi Jam <bobijam@whamcloud.com>
    Tested-by: Maloo <whamcloud.maloo@gmail.com>
    Reviewed-by: Oleg Drokin <green@whamcloud.com>
  2. LU-1102 crypto: correctly check crypto_alloc_blkcipher returns

    Bobi Jam authored and morrone committed May 9, 2012
    ll_crypto_alloc_blkcipher() returns error value as well as possible
    NULL pointer, should check its return value carefully.
    
    Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
    Change-Id: I181b236406e2649580a04940886f849ad6071078
    Reviewed-on: http://review.whamcloud.com/2703
    Reviewed-by: Lai Siyao <laisiyao@whamcloud.com>
    Tested-by: Hudson
    Tested-by: Maloo <whamcloud.maloo@gmail.com>
    Reviewed-by: Oleg Drokin <green@whamcloud.com>
  3. LU-1366 utils: disable ldiskfs extents feature for MDT

    Bobi Jam authored and morrone committed May 15, 2012
    Explicitly disable "extents" for MDT filesystem if it's based on ext4,
    it provides no benifit for MDT.
    
    Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
    Change-Id: I284c6c207fb8cc79537bebd60b6ab8d836fd4ed9
    Reviewed-on: http://review.whamcloud.com/2798
    Tested-by: Hudson
    Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
    Tested-by: Maloo <whamcloud.maloo@gmail.com>
    Reviewed-by: Oleg Drokin <green@whamcloud.com>
Commits on May 15, 2012
  1. LU-1299 clio: a combo patch to fix cl_lock

    Jinshan Xiong authored and morrone committed May 2, 2012
    In this patch, the following issues are fixed:
    * redefine cl_lock_error(). There are two kinds of error when
      enqueuing a cl_lock: internal errors return from server side and
      local error for example interrupted by signal when waiting for a
      lock. Obviously the latter one is not fatal so we shouldn't err the
      lock out;
    * for fault handling, signal is not allowed;
    * handle return code of ll_fault correctly;
    * redefine ->unuse() to release resource it held if enqueuing lock
      failed;
    * in cl_lock_hold_release(), only CLS_CACHED mode lock will be
    * cached,
      for locks in other states, they will be cancelled voluntarily;
    * other minor fixes.
    
    Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Change-Id: Iffce8be356723781b8f33ec9bdc2cf73e9e07138
    Patch-Set: 7
Commits on May 14, 2012
  1. LU-1376 ldlm: Add ldlm_poold module options

    morrone committed May 14, 2012
    Add ldlm_poold_enabled module option to allow disabling the
    ldlm_poold thread.
    
    Add ldlm_poold_period module option to adjust the length of
    time that the ldlm_poold sleeps between calling ldlm_pool_recalc().
    
    The ldlm_poold currently creates far too much system noise and
    is severely impacting tightly synchronized application.  It
    currently reduces some of our real applications' performance by 75%
    at 8192 tasks (on 512 nodes), and possibly even more at larger scales.
    
    Disabling the thread or increasing the sleep time allows us to
    temporarily work around the problem.
    
    Signed-off-by: Christopher J. Morrone <morrone2@llnl.gov>
    Change-Id: Ie528c516793c5241ac932b0793ce12bb788dadb2
Commits on May 10, 2012
  1. LU-1378 fid: Add console info for super seq allocation

    wangdi authored and morrone committed May 9, 2012
    Add console information for super sequence allocation. Because
    one super sequence will include 1 billion sequences, it rarely
    happens in reality, so it will not cause the flood of console
    msg.
    
    Signed-off-by: Di Wang <di.wang@whamcloud.com>
    Change-Id: I5154ee2a03006680b6a08d588287bf3941149457
  2. LU-1353 recovery: debug patch for open replay

    Lai Siyao authored and morrone committed May 8, 2012
    Open replay may fail on -ENOENT, add debug message to print fid and
    etc.
    
    Signed-off-by: Lai Siyao <laisiyao@whamcloud.com>
    Change-Id: Ic7b2e796429934d8996621f648107d11cf9f3373
Commits on Apr 30, 2012
  1. Revert "LU-1299 clio: handle signal correctly for cl_lock"

    morrone committed Apr 30, 2012
    This reverts commit 4861c39.
    
    This may have introduced new bugs.
Commits on Apr 27, 2012
  1. LU-1217 osc: to not check a cl_lock's state w/o protection

    Jinshan Xiong authored and morrone committed Mar 26, 2012
    osc_page_putref_lock() used to check cl_lock's refcount and
    corresponding osc_lock's ols_hold without any protection, this
    is racy because other process can change the lock state so as to
    make the assertion be false.
    
    Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Change-Id: I65fe1fa7fc55e8642fea6789784d7bb92a45d56f
  2. Revert "LU-1217 clio: debug patch"

    morrone committed Apr 27, 2012
    This reverts commit 78b1f90.
Commits on Apr 19, 2012
  1. LU-1320 llite: fix a race between readpage and releasepage

    Jinshan Xiong authored and morrone committed Apr 18, 2012
    This is a race between page stealing and readpage. If a just read
    page is stolen, readpage will find the page is not uptodate, this
    makes it panic so -EIO is returned to the reading application.
    
    Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Change-Id: Ib16d12d3bc3cc8c0545aa27f0836e4fd89c3a809
    
    NOTE: This is patchset 5 from http://review.whamcloud.com/2564
  2. LU-1299 clio: handle signal correctly for cl_lock

    Jinshan Xiong authored and nedbass committed Apr 18, 2012
    Two issues are fixed in this patch:
    * for fault handling, signal is not allowed;
    * if a process is interrupted while waiting for a cl_lock, the lock
      shouldn't be erred out.
    
    Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Change-Id: Iffce8be356723781b8f33ec9bdc2cf73e9e07138
  3. LU-1282 lprocfs: disable interrupts in lprocfs_stats_lock()

    nedbass committed Apr 18, 2012
    Deadlock will occur if we try to take stats->ls_lock under an
    interrupt while already holding the lock.  Use the irqsave version
    of the spinlock to avoid this.
    
    Signed-off-by: Ned Bass <bass6@llnl.gov>
    Change-Id: I9f978f93ff0da446b1632f25ac849143218e64a3
  4. LU-1092 ptlrpc: take export refcount during connect

    Lai Siyao authored and morrone committed Mar 19, 2012
    In the process of (re)connect,  a refcount of export should be taken,
    otherwise disconnect of this export may be called, and it will put
    the last refcount of this export and make access to this export
    invalid.
    
    Signed-off-by: Lai Siyao <laisiyao@whamcloud.com>
    Change-Id: Iaf27e842ed516b8968c90bfce396609e39f52c85
Commits on Apr 17, 2012
  1. LU-1245 lprocfs: use correct cpu number

    Bobi Jam authored and morrone committed Mar 27, 2012
    Take care of correct cpu number in lprocfs_stats_collector().
    
    Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
    Change-Id: Ifb149f64ee6d5b67a029331c0d0452fc29533c6b
Commits on Apr 16, 2012
Commits on Apr 13, 2012
  1. LU-1282 lprocfs: Add a module param to disable percpu stats

    Bobi Jam authored and morrone committed Apr 12, 2012
    Add an obdclass module option to choose to use a single lprocfs stats
    structure rather than percpu data.
    
    Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
    Change-Id: I45d5a05029197e629d4f7d161a5e4e5d01a93bf5
Commits on Apr 10, 2012
  1. LU-81 deadlock of changelog adding vs. changelog cancelling

    Niu Yawei authored and morrone committed Aug 18, 2011
    This is a workaround for the deadlock of changelog adding vs.
    changelog cancelling. Changelog adding always start transaction
    before acquiring the catlog lock(lgh_lock), whereas, changelog
    cancelling do start transaction after holding the catlog lock.
    
    We start transaction earlier to avoid above deadlock.
    
    Signed-off-by: Niu Yawei <niu@whamcloud.com>
    Change-Id: I9647b9a559f68a27dc0d4b4885857d3cf73b5b8e
    Reviewed-on: http://review.whamcloud.com/1260
    Tested-by: Hudson
    Tested-by: Maloo <whamcloud.maloo@gmail.com>
    Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
    Reviewed-by: Oleg Drokin <green@whamcloud.com>
Commits on Apr 7, 2012
  1. LU-1282 misc: Use present cpu numbers to save memory.

    Bobi Jam authored and morrone committed Apr 5, 2012
    lprocfs stats data should allocated by the number of present cpus in
    stead of by possible cpu numbers which wastes a lot of memory.
    
    OSS minimum thread number also better be decided by present cpu
    numbers.
    
    Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
    Change-Id: Id1690f185f4f83fae75be7eddb756e413cbc4fba
Commits on Apr 6, 2012
  1. LU-1234 dcache: don't drop invalid dentry arbitrarily

    Lai Siyao authored and morrone committed Mar 28, 2012
    This is a backport of part of LU-506 dcache scalability support:
    * remove super hack d_rehash_cond(), and treats DCACHE_LUSTRE_INVALID
      similar to DCACHE_DISCONNECTED, therefore dentry doesn't need to
      be dropped and rehashed frequently.
    * .lookup(LOOKUP_CREATE) calls d_add() dentry directly, and .create
      only needs to d_instantiate() this dentry.
    * other cleanups.
    
    Signed-off-by: Lai Siyao <laisiyao@whamcloud.com>
    Change-Id: Ie169bc7e763e6891084999041aac9f62c8dee9f0
Commits on Apr 5, 2012
  1. LU-1280 ldiskfs: remove LASSERTF from ext3_ext_new_extent_cb()

    Yu Jian authored and nedbass committed Apr 5, 2012
    The LASSERTF() in ext3_ext_new_extent_cb() was injected for
    debugging purpose to make sure the race really happened but
    was forgotten to be removed from the original patch in
    http://review.whamcloud.com/1618 .
    
    Signed-off-by: Yu Jian <yujian@whamcloud.com>
    Change-Id: I978b8ab88cc4413c7ac00db838f7578f8011b192
Commits on Mar 26, 2012
  1. LU-931 mdd: store lu_fid instead of pointer in md_capainfo

    Hongchao Zhang authored and morrone committed Jan 17, 2012
    in md_capainfo, mc_fid contains at most 5 pointers to lu_fid,
    and if the corresponding lu_fid is freed, the pointer isn't notified
    about it, then the pointer will point to freed memory!
    
    Signed-off-by: Hongchao Zhang <hongchao.zhang@whamcloud.com>
    Change-Id: I00088cbfeb145ceac0477467a8b2436f6cf1e530
  2. LU-1217 clio: debug patch

    Jinshan Xiong authored and morrone committed Mar 26, 2012
    debug patch to print lock state.
    
    Signed-off-by: Jinshan Xiong <jinshan.xiong@whamcloud.com>
    Change-Id: Ia6ff85a0fdaffb6d03216ad0d69953fb339417a9
Commits on Mar 16, 2012
  1. LU-1206 mdt: Fix error handling in mdt_mfd_open

    Oleg Drokin authored and morrone committed Mar 15, 2012
    In mdt_mfd_open if the mo_open() call failed or we could not allocate
    mfd, we also need to undo write/exec reference count in order to
    not mess up with subsequent exec/write accesses.
    
    Signed-off-by: Prakash Surya <surya1@llnl.gov>
    Signed-off-by: Oleg Drokin <green@whamcloud.com>
    Change-Id: I3bd98bd68368b48f2afaa7bb450d3a9947c992ac