Skip to content

L
linux
Switch branch/tag
  1. 22 Dec, 2011 5 commits
    • Srivatsa S. Bhat's avatar
      VFS: Fix race between CPU hotplug and lglocks · e30e2fdf
      Srivatsa S. Bhat authored 
      Currently, the *_global_[un]lock_online() routines are not at all synchronized
with CPU hotplug. Soft-lockups detected as a consequence of this race was
reported earlier at https://lkml.org/lkml/2011/8/24/185. (Thanks to Cong Meng
for finding out that the root-cause of this issue is the race condition
between br_write_[un]lock() and CPU hotplug, which results in the lock states
getting messed up).

Fixing this race by just adding {get,put}_online_cpus() at appropriate places
in *_global_[un]lock_online() is not a good option, because, then suddenly
br_write_[un]lock() would become blocking, whereas they have been kept as
non-blocking all this time, and we would want to keep them that way.

So, overall, we want to ensure 3 things:
1. br_write_lock() and br_write_unlock() must remain as non-blocking.
2. The corresponding lock and unlock of the per-cpu spinlocks must not happen
   for different sets of CPUs.
3. Either prevent any new CPU online operation in between this lock-unlock, or
   ensure that the newly onlined CPU does not proceed with its corresponding
   per-cpu spinlock unlocked.

To achieve all this:
(a) We introduce a new spinlock that is taken by the *_global_lock_online()
    routine and released by the *_global_unlock_online() routine.
(b) We register a callback for CPU hotplug notifications, and this callback
    takes the same spinlock as above.
(c) We maintain a bitmap which is close to the cpu_online_mask, and once it is
    initialized in the lock_init() code, all future updates to it are done in
    the callback, under the above spinlock.
(d) The above bitmap is used (instead of cpu_online_mask) while locking and
    unlocking the per-cpu locks.

The callback takes the spinlock upon the CPU_UP_PREPARE event. So, if the
br_write_lock-unlock sequence is in progress, the callback keeps spinning,
thus preventing the CPU online operation till the lock-unlock sequence is
complete. This takes care of requirement (3).

The bitmap that we maintain remains unmodified throughout the lock-unlock
sequence, since all updates to it are managed by the callback, which takes
the same spinlock as the one taken by the lock code and released only by the
unlock routine. Combining this with (d) above, satisfies requirement (2).

Overall, since we use a spinlock (mentioned in (a)) to prevent CPU hotplug
operations from racing with br_write_lock-unlock, requirement (1) is also
taken care of.

By the way, it is to be noted that a CPU offline operation can actually run
in parallel with our lock-unlock sequence, because our callback doesn't react
to notifications earlier than CPU_DEAD (in order to maintain our bitmap
properly). And this means, since we use our own bitmap (which is stale, on
purpose) during the lock-unlock sequence, we could end up unlocking the
per-cpu lock of an offline CPU (because we had locked it earlier, when the
CPU was online), in order to satisfy requirement (2). But this is harmless,
though it looks a bit awkward.
Debugged-by: default avatarCong Meng <mc@linux.vnet.ibm.com>
Signed-off-by: default avatarSrivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: default avatarAl Viro <viro@zeniv.linux.org.uk>
Cc: stable@vger.kernel.org
      e30e2fdf
    • Linus Torvalds's avatar
      Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net · ecefc36b
      Linus Torvalds authored 
      * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
  net: Add a flow_cache_flush_deferred function
  ipv4: reintroduce route cache garbage collector
  net: have ipconfig not wait if no dev is available
  sctp: Do not account for sizeof(struct sk_buff) in estimated rwnd
  asix: new device id
  davinci-cpdma: fix locking issue in cpdma_chan_stop
  sctp: fix incorrect overflow check on autoclose
  r8169: fix Config2 MSIEnable bit setting.
  llc: llc_cmsg_rcv was getting called after sk_eat_skb.
  net: bpf_jit: fix an off-one bug in x86_64 cond jump target
  iwlwifi: update SCD BC table for all SCD queues
  Revert "Bluetooth: Revert: Fix L2CAP connection establishment"
  Bluetooth: Clear RFCOMM session timer when disconnecting last channel
  Bluetooth: Prevent uninitialized data access in L2CAP configuration
  iwlwifi: allow to switch to HT40 if not associated
  iwlwifi: tx_sync only on PAN context
  mwifiex: avoid double list_del in command cancel path
  ath9k: fix max phy rate at rate control init
  nfc: signedness bug in __nci_request()
  iwlwifi: do not set the sequence control bit is not needed
      ecefc36b
    • Linus Torvalds's avatar
      Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound · d5ed5e48
      Linus Torvalds authored 
      * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: atmel/ac97c: using software reset instead hardware reset if not available
      d5ed5e48
    • Linus Torvalds's avatar
      Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6 · 0703c680
      Linus Torvalds authored 
      * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sameo/mfd-2.6:
  mfd: Include linux/io.h to jz4740-adc
  mfd: Use request_threaded_irq for twl4030-irq instead of irq_set_chained_handler
  mfd: Base interrupt for twl4030-irq must be one-shot
  mfd: Handle tps65910 clear-mask correctly
  mfd: add #ifdef CONFIG_DEBUG_FS guard for ab8500_debug_resources
  mfd: Fix twl-core oops while calling twl_i2c_* for unbound driver
  mfd: include linux/module.h for ab5500-debugfs
  mfd: Update wm8994 active device checks for WM1811
  mfd: Set tps6586x bits if new value is different from the old one
  mfd: Set da903x bits if new value is different from the old one
  mfd: Set adp5520 bits if new value is different from the old one
  mfd: Add missed free_irq in da903x_remove
      0703c680
    • Dave Kleikamp's avatar
      vfs: __read_cache_page should use gfp argument rather than GFP_KERNEL · e6f67b8c
      Dave Kleikamp authored 
      lockdep reports a deadlock in jfs because a special inode's rw semaphore
is taken recursively.  The mapping's gfp mask is GFP_NOFS, but is not
used when __read_cache_page() calls add_to_page_cache_lru().
Signed-off-by: default avatarDave Kleikamp <dave.kleikamp@oracle.com>
Acked-by: default avatarHugh Dickins <hughd@google.com>
Acked-by: default avatarAl Viro <viro@zeniv.linux.org.uk>
Cc: stable@kernel.org
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      e6f67b8c
  3. 21 Dec, 2011 8 commits
  5. 20 Dec, 2011 27 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7