1. 26 Apr, 2012 8 commits
    • Aneesh Kumar K.V's avatar
      hugetlbfs: lockdep annotate root inode properly · 65ed7601
      Aneesh Kumar K.V authored
      This fixes the below reported false lockdep warning.  e096d0c7
      ("lockdep: Add helper function for dir vs file i_mutex annotation") added
      a similar annotation for every other inode in hugetlbfs but missed the
      root inode because it was allocated by a separate function.
      
      For HugeTLB fs we allow taking i_mutex in mmap.  HugeTLB fs doesn't
      support file write and its file read callback is modified in a05b0855
      ("hugetlbfs: avoid taking i_mutex from hugetlbfs_read()") to not take
      i_mutex.  Hence for HugeTLB fs with regular files we really don't take
      i_mutex with mmap_sem held.
      
       ======================================================
       [ INFO: possible circular locking dependency detected ]
       3.4.0-rc1+ #322 Not tainted
       -------------------------------------------------------
       bash/1572 is trying to acquire lock:
        (&mm->mmap_sem){++++++}, at: [<ffffffff810f1618>] might_fault+0x40/0x90
      
       but task is already holding lock:
        (&sb->s_type->i_mutex_key#12){+.+.+.}, at: [<ffffffff81125f88>] vfs_readdir+0x56/0xa8
      
       which lock already depends on the new lock.
      
       the existing dependency chain (in reverse order) is:
      
       -> #1 (&sb->s_type->i_mutex_key#12){+.+.+.}:
              [<ffffffff810a09e5>] lock_acquire+0xd5/0xfa
              [<ffffffff816a2f5e>] __mutex_lock_common+0x48/0x350
              [<ffffffff816a3325>] mutex_lock_nested+0x2a/0x31
              [<ffffffff811fb8e1>] hugetlbfs_file_mmap+0x7d/0x104
              [<ffffffff810f859a>] mmap_region+0x272/0x47d
              [<ffffffff810f8a39>] do_mmap_pgoff+0x294/0x2ee
              [<ffffffff810f8b65>] sys_mmap_pgoff+0xd2/0x10e
              [<ffffffff8103d19e>] sys_mmap+0x1d/0x1f
              [<ffffffff816a5922>] system_call_fastpath+0x16/0x1b
      
       -> #0 (&mm->mmap_sem){++++++}:
              [<ffffffff810a0256>] __lock_acquire+0xa81/0xd75
              [<ffffffff810a09e5>] lock_acquire+0xd5/0xfa
              [<ffffffff810f1645>] might_fault+0x6d/0x90
              [<ffffffff81125d62>] filldir+0x6a/0xc2
              [<ffffffff81133a83>] dcache_readdir+0x5c/0x222
              [<ffffffff81125fa8>] vfs_readdir+0x76/0xa8
              [<ffffffff811260b6>] sys_getdents+0x79/0xc9
              [<ffffffff816a5922>] system_call_fastpath+0x16/0x1b
      
       other info that might help us debug this:
      
        Possible unsafe locking scenario:
      
              CPU0                    CPU1
              ----                    ----
         lock(&sb->s_type->i_mutex_key#12);
                                      lock(&mm->mmap_sem);
                                      lock(&sb->s_type->i_mutex_key#12);
         lock(&mm->mmap_sem);
      
        *** DEADLOCK ***
      
       1 lock held by bash/1572:
        #0:  (&sb->s_type->i_mutex_key#12){+.+.+.}, at: [<ffffffff81125f88>] vfs_readdir+0x56/0xa8
      
       stack backtrace:
       Pid: 1572, comm: bash Not tainted 3.4.0-rc1+ #322
       Call Trace:
        [<ffffffff81699a3c>] print_circular_bug+0x1f8/0x209
        [<ffffffff810a0256>] __lock_acquire+0xa81/0xd75
        [<ffffffff810f38aa>] ? handle_pte_fault+0x5ff/0x614
        [<ffffffff8109e622>] ? mark_lock+0x2d/0x258
        [<ffffffff810f1618>] ? might_fault+0x40/0x90
        [<ffffffff810a09e5>] lock_acquire+0xd5/0xfa
        [<ffffffff810f1618>] ? might_fault+0x40/0x90
        [<ffffffff816a3249>] ? __mutex_lock_common+0x333/0x350
        [<ffffffff810f1645>] might_fault+0x6d/0x90
        [<ffffffff810f1618>] ? might_fault+0x40/0x90
        [<ffffffff81125d62>] filldir+0x6a/0xc2
        [<ffffffff81133a83>] dcache_readdir+0x5c/0x222
        [<ffffffff81125cf8>] ? sys_ioctl+0x74/0x74
        [<ffffffff81125cf8>] ? sys_ioctl+0x74/0x74
        [<ffffffff81125cf8>] ? sys_ioctl+0x74/0x74
        [<ffffffff81125fa8>] vfs_readdir+0x76/0xa8
        [<ffffffff811260b6>] sys_getdents+0x79/0xc9
        [<ffffffff816a5922>] system_call_fastpath+0x16/0x1b
      Signed-off-by: default avatarAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
      Cc: Dave Jones <davej@redhat.com>
      Cc: Al Viro <viro@zeniv.linux.org.uk>
      Cc: Josh Boyer <jwboyer@redhat.com>
      Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
      Cc: Mimi Zohar <zohar@linux.vnet.ibm.com>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      65ed7601
    • Peter Feuerer's avatar
      acerhdf: lowered default temp fanon/fanoff values · 351963bb
      Peter Feuerer authored
      Due to new supported hardware, of which the actual temperature limits of
      processor, harddisk and other components are unknown, it feels safer with
      lower fanon / fanoff settings.
      
      It won't change much for most people, already using acerhdf, as they use
      their own fanon/fanoff variable settings when loading the module.
      
      Furthermore seems like kernel and userspace tools have been improved to
      work more efficient and netbooks don't get so hot anymore.
      Signed-off-by: default avatarPeter Feuerer <peter@piie.net>
      Acked-by: default avatarBorislav Petkov <petkovbb@gmail.com>
      Cc: Matthew Garrett <mjg@redhat.com>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      351963bb
    • Peter Feuerer's avatar
      acerhdf: add support for new hardware · 43ae1e32
      Peter Feuerer authored
      Add support for new hardware:
      Acer Aspire LT-10Q/531/751/1810/1825,
      Acer Travelmate 7730,
      Packard Bell ENBFT/DOTVR46
      Signed-off-by: default avatarPeter Feuerer <peter@piie.net>
      Acked-by: default avatarBorislav Petkov <petkovbb@gmail.com>
      Cc: Matthew Garrett <mjg@redhat.com>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      43ae1e32
    • Clay Carpenter's avatar
      acerhdf: add support for Aspire 1410 BIOS v1.3314 · e39a9ba2
      Clay Carpenter authored
      Add support for Aspire 1410 BIOS v1.3314.  Fixes the following error:
      
      acerhdf: unknown (unsupported) BIOS version Acer/Aspire 1410/v1.3314,
      please report, aborting!
      Signed-off-by: default avatarClay Carpenter <claycarpenter@gmail.com>
      Signed-off-by: default avatarPeter Feuerer <peter@piie.net>
      Cc: Matthew Garrett <mjg@redhat.com>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      e39a9ba2
    • Glauber Costa's avatar
      fs/buffer.c: remove BUG() in possible but rare condition · 61065a30
      Glauber Costa authored
      While stressing the kernel with with failing allocations today, I hit the
      following chain of events:
      
      alloc_page_buffers():
      
      	bh = alloc_buffer_head(GFP_NOFS);
      	if (!bh)
      		goto no_grow; <= path taken
      
      grow_dev_page():
              bh = alloc_page_buffers(page, size, 0);
              if (!bh)
                      goto failed;  <= taken, consequence of the above
      
      and then the failed path BUG()s the kernel.
      
      The failure is inserted a litte bit artificially, but even then, I see no
      reason why it should be deemed impossible in a real box.
      
      Even though this is not a condition that we expect to see around every
      time, failed allocations are expected to be handled, and BUG() sounds just
      too much.  As a matter of fact, grow_dev_page() can return NULL just fine
      in other circumstances, so I propose we just remove it, then.
      Signed-off-by: default avatarGlauber Costa <glommer@parallels.com>
      Cc: Michal Hocko <mhocko@suse.cz>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      61065a30
    • Ying Han's avatar
      mm: fix up the vmscan stat in vmstat · 904249aa
      Ying Han authored
      The "pgsteal" stat is confusing because it counts both direct reclaim as
      well as background reclaim.  However, we have "kswapd_steal" which also
      counts background reclaim value.
      
      This patch fixes it and also makes it match the existng "pgscan_" stats.
      
      Test:
      pgsteal_kswapd_dma32 447623
      pgsteal_kswapd_normal 42272677
      pgsteal_kswapd_movable 0
      pgsteal_direct_dma32 2801
      pgsteal_direct_normal 44353270
      pgsteal_direct_movable 0
      Signed-off-by: default avatarYing Han <yinghan@google.com>
      Reviewed-by: default avatarRik van Riel <riel@redhat.com>
      Acked-by: default avatarChristoph Lameter <cl@linux.com>
      Cc: Johannes Weiner <hannes@cmpxchg.org>
      Cc: Michal Hocko <mhocko@suse.cz>
      Cc: Mel Gorman <mel@csn.ul.ie>
      Acked-by: default avatarKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
      Cc: Hillf Danton <dhillf@gmail.com>
      Cc: Hugh Dickins <hughd@google.com>
      Cc: Dan Magenheimer <dan.magenheimer@oracle.com>
      Reviewed-by: default avatarMinchan Kim <minchan@kernel.org>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      904249aa
    • Jason Baron's avatar
      epoll: clear the tfile_check_list on -ELOOP · 13d51807
      Jason Baron authored
      An epoll_ctl(,EPOLL_CTL_ADD,,) operation can return '-ELOOP' to prevent
      circular epoll dependencies from being created.  However, in that case we
      do not properly clear the 'tfile_check_list'.  Thus, add a call to
      clear_tfile_check_list() for the -ELOOP case.
      Signed-off-by: default avatarJason Baron <jbaron@redhat.com>
      Reported-by: default avatarYurij M. Plotnikov <Yurij.Plotnikov@oktetlabs.ru>
      Cc: Nelson Elhage <nelhage@nelhage.com>
      Cc: Davide Libenzi <davidel@xmailserver.org>
      Tested-by: default avatarAlexandra N. Kossovsky <Alexandra.Kossovsky@oktetlabs.ru>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      13d51807
    • Konstantin Khlebnikov's avatar
      mm/hugetlb: fix warning in alloc_huge_page/dequeue_huge_page_vma · b1c12cbc
      Konstantin Khlebnikov authored
      Fix a gcc warning (and bug?) introduced in cc9a6c87 ("cpuset: mm: reduce
      large amounts of memory barrier related damage v3")
      
      Local variable "page" can be uninitialized if the nodemask from vma policy
      does not intersects with nodemask from cpuset.  Even if it doesn't happens
      it is better to initialize this variable explicitly than to introduce
      a kernel oops in a weird corner case.
      
      mm/hugetlb.c: In function `alloc_huge_page':
      mm/hugetlb.c:1135:5: warning: `page' may be used uninitialized in this function
      Signed-off-by: default avatarKonstantin Khlebnikov <khlebnikov@openvz.org>
      Acked-by: default avatarMel Gorman <mgorman@suse.de>
      Acked-by: default avatarDavid Rientjes <rientjes@google.com>
      Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      b1c12cbc
  2. 21 Apr, 2012 23 commits
  3. 20 Apr, 2012 9 commits