linux.git/net/sunrpc, branch v3.7-rc3

SUNRPC: Get rid of the xs_error_report socket callback

2012-10-24T14:46:15+00:00

Chris Perl reports that we're seeing races between the wakeup call in
xs_error_report and the connect attempts. Basically, Chris has shown
that in certain circumstances, the call to xs_error_report causes the
rpc_task that is responsible for reconnecting to wake up early, thus
triggering a disconnect and retry.

Since the sk->sk_error_report() calls in the socket layer are always
followed by a tcp_done() in the cases where we care about waking up
the rpc_tasks, just let the state_change callbacks take responsibility
for those wake ups.

Reported-by: Chris Perl 
Signed-off-by: Trond Myklebust 
Cc: stable@vger.kernel.org
Tested-by: Chris Perl

SUNRPC: Prevent races in xs_abort_connection()

2012-10-24T14:46:08+00:00

The call to xprt_disconnect_done() that is triggered by a successful
connection reset will trigger another automatic wakeup of all tasks
on the xprt->pending rpc_wait_queue. In particular it will cause an
early wake up of the task that called xprt_connect().

All we really want to do here is clear all the socket-specific state
flags, so we split that functionality out of xs_sock_mark_closed()
into a helper that can be called by xs_abort_connection()

Reported-by: Chris Perl 
Signed-off-by: Trond Myklebust 
Cc: stable@vger.kernel.org
Tested-by: Chris Perl

Revert "SUNRPC: Ensure we close the socket on EPIPE errors too..."

2012-10-24T14:45:39+00:00

This reverts commit 55420c24a0d4d1fce70ca713f84aa00b6b74a70e.
Now that we clear the connected flag when entering TCP_CLOSE_WAIT,
the deadlock described in this commit is no longer possible.
Instead, the resulting call to xs_tcp_shutdown() can interfere
with pending reconnection attempts.

Reported-by: Chris Perl 
Signed-off-by: Trond Myklebust 
Cc: stable@vger.kernel.org
Tested-by: Chris Perl

SUNRPC: Clear the connect flag when socket state is TCP_CLOSE_WAIT

2012-10-24T14:44:49+00:00

This is needed to ensure that we call xprt_connect() upon the next
call to call_connect().

Signed-off-by: Trond Myklebust 
Cc: stable@vger.kernel.org
Tested-by: Chris Perl

SUNRPC: Prevent kernel stack corruption on long values of flush

2012-10-17T18:59:10+00:00

The buffer size in read_flush() is too small for the longest possible values
for it. This can lead to a kernel stack corruption:

[   43.047329] Kernel panic - not syncing: stack-protector: Kernel stack is corrupted in: ffffffff833e64b4
[   43.047329]
[   43.049030] Pid: 6015, comm: trinity-child18 Tainted: G        W    3.5.0-rc7-next-20120716-sasha #221
[   43.050038] Call Trace:
[   43.050435]  [] panic+0xcd/0x1f4
[   43.050931]  [] ? read_flush.isra.7+0xe4/0x100
[   43.051602]  [] __stack_chk_fail+0x16/0x20
[   43.052206]  [] read_flush.isra.7+0xe4/0x100
[   43.052951]  [] ? read_flush_pipefs+0x30/0x30
[   43.053594]  [] read_flush_procfs+0x2c/0x30
[   43.053596]  [] proc_reg_read+0x9c/0xd0
[   43.053596]  [] ? proc_reg_write+0xd0/0xd0
[   43.053596]  [] do_loop_readv_writev+0x4b/0x90
[   43.053596]  [] do_readv_writev+0xf6/0x1d0
[   43.053596]  [] vfs_readv+0x3e/0x60
[   43.053596]  [] sys_readv+0x48/0xb0
[   43.053596]  [] system_call_fastpath+0x1a/0x1f

Signed-off-by: Sasha Levin 
Cc: stable@kernel.org
Signed-off-by: J. Bruce Fields

Merge branch 'for-3.7' of git://linux-nfs.org/~bfields/linux

2012-10-13T01:53:54+00:00

Pull nfsd update from J Bruce Fields:
 "Another relatively quiet cycle.  There was some progress on my
  remaining 4.1 todo's, but a couple of them were just of the form
  "check that we do X correctly", so didn't have much affect on the
  code.

  Other than that, a bunch of cleanup and some bugfixes (including an
  annoying NFSv4.0 state leak and a busy-loop in the server that could
  cause it to peg the CPU without making progress)."

* 'for-3.7' of git://linux-nfs.org/~bfields/linux: (46 commits)
  UAPI: (Scripted) Disintegrate include/linux/sunrpc
  UAPI: (Scripted) Disintegrate include/linux/nfsd
  nfsd4: don't allow reclaims of expired clients
  nfsd4: remove redundant callback probe
  nfsd4: expire old client earlier
  nfsd4: separate session allocation and initialization
  nfsd4: clean up session allocation
  nfsd4: minor free_session cleanup
  nfsd4: new_conn_from_crses should only allocate
  nfsd4: separate connection allocation and initialization
  nfsd4: reject bad forechannel attrs earlier
  nfsd4: enforce per-client sessions/no-sessions distinction
  nfsd4: set cl_minorversion at create time
  nfsd4: don't pin clientids to pseudoflavors
  nfsd4: fix bind_conn_to_session xdr comment
  nfsd4: cast readlink() bug argument
  NFSD: pass null terminated buf to kstrtouint()
  nfsd: remove duplicate init in nfsd4_cb_recall
  nfsd4: eliminate redundant nfs4_free_stateid
  fs/nfsd/nfs4idmap.c: adjust inconsistent IS_ERR and PTR_ERR
  ...

Merge Trond's bugfixes

2012-10-11T16:41:05+00:00

Merge branch 'bugfixes' of git://linux-nfs.org/~trondmy/nfs-2.6 into
for-3.7-incoming.  Mainly needed for Bryan's "SUNRPC: Set alloc_slot for
backchannel tcp ops", without which the 4.1 server oopses.

Merge tag 'nfs-for-3.7-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

2012-10-10T14:52:35+00:00

Pull NFS client updates from Trond Myklebust:
 "Features include:

   - Remove CONFIG_EXPERIMENTAL dependency from NFSv4.1
     Aside from the issues discussed at the LKS, distros are shipping
     NFSv4.1 with all the trimmings.
   - Fix fdatasync()/fsync() for the corner case of a server reboot.
   - NFSv4 OPEN access fix: finally distinguish correctly between
     open-for-read and open-for-execute permissions in all situations.
   - Ensure that the TCP socket is closed when we're in CLOSE_WAIT
   - More idmapper bugfixes
   - Lots of pNFS bugfixes and cleanups to remove unnecessary state and
     make the code easier to read.
   - In cases where a pNFS read or write fails, allow the client to
     resume trying layoutgets after two minutes of read/write-
     through-mds.
   - More net namespace fixes to the NFSv4 callback code.
   - More net namespace fixes to the NFSv3 locking code.
   - More NFSv4 migration preparatory patches.
     Including patches to detect network trunking in both NFSv4 and
     NFSv4.1
   - pNFS block updates to optimise LAYOUTGET calls."

* tag 'nfs-for-3.7-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (113 commits)
  pnfsblock: cleanup nfs4_blkdev_get
  NFS41: send real read size in layoutget
  NFS41: send real write size in layoutget
  NFS: track direct IO left bytes
  NFSv4.1: Cleanup ugliness in pnfs_layoutgets_blocked()
  NFSv4.1: Ensure that the layout sequence id stays 'close' to the current
  NFSv4.1: Deal with seqid wraparound in the pNFS return-on-close code
  NFSv4 set open access operation call flag in nfs4_init_opendata_res
  NFSv4.1: Remove the dependency on CONFIG_EXPERIMENTAL
  NFSv4 reduce attribute requests for open reclaim
  NFSv4: nfs4_open_done first must check that GETATTR decoded a file type
  NFSv4.1: Deal with wraparound when updating the layout "barrier" seqid
  NFSv4.1: Deal with wraparound issues when updating the layout stateid
  NFSv4.1: Always set the layout stateid if this is the first layoutget
  NFSv4.1: Fix another refcount issue in pnfs_find_alloc_layout
  NFSv4: don't put ACCESS in OPEN compound if O_EXCL
  NFSv4: don't check MAY_WRITE access bit in OPEN
  NFS: Set key construction data for the legacy upcall
  NFSv4.1: don't do two EXCHANGE_IDs on mount
  NFS: nfs41_walk_client_list(): re-lock before iterating
  ...

nfs: disintegrate UAPI for nfs

2012-10-09T22:35:22+00:00

This is to complete part of the Userspace API (UAPI) disintegration for which
the preparatory patches were pulled recently.  After these patches, userspace
headers will be segregated into:

        include/uapi/linux/.../foo.h

for the userspace interface stuff, and:

        include/linux/.../foo.h

for the strictly kernel internal stuff.

Signed-off-by: J. Bruce Fields

Merge branch 'for-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq

2012-10-02T16:54:49+00:00

Pull workqueue changes from Tejun Heo:
 "This is workqueue updates for v3.7-rc1.  A lot of activities this
  round including considerable API and behavior cleanups.

   * delayed_work combines a timer and a work item.  The handling of the
     timer part has always been a bit clunky leading to confusing
     cancelation API with weird corner-case behaviors.  delayed_work is
     updated to use new IRQ safe timer and cancelation now works as
     expected.

   * Another deficiency of delayed_work was lack of the counterpart of
     mod_timer() which led to cancel+queue combinations or open-coded
     timer+work usages.  mod_delayed_work[_on]() are added.

     These two delayed_work changes make delayed_work provide interface
     and behave like timer which is executed with process context.

   * A work item could be executed concurrently on multiple CPUs, which
     is rather unintuitive and made flush_work() behavior confusing and
     half-broken under certain circumstances.  This problem doesn't
     exist for non-reentrant workqueues.  While non-reentrancy check
     isn't free, the overhead is incurred only when a work item bounces
     across different CPUs and even in simulated pathological scenario
     the overhead isn't too high.

     All workqueues are made non-reentrant.  This removes the
     distinction between flush_[delayed_]work() and
     flush_[delayed_]_work_sync().  The former is now as strong as the
     latter and the specified work item is guaranteed to have finished
     execution of any previous queueing on return.

   * In addition to the various bug fixes, Lai redid and simplified CPU
     hotplug handling significantly.

   * Joonsoo introduced system_highpri_wq and used it during CPU
     hotplug.

  There are two merge commits - one to pull in IRQ safe timer from
  tip/timers/core and the other to pull in CPU hotplug fixes from
  wq/for-3.6-fixes as Lai's hotplug restructuring depended on them."

Fixed a number of trivial conflicts, but the more interesting conflicts
were silent ones where the deprecated interfaces had been used by new
code in the merge window, and thus didn't cause any real data conflicts.

Tejun pointed out a few of them, I fixed a couple more.

* 'for-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq: (46 commits)
  workqueue: remove spurious WARN_ON_ONCE(in_irq()) from try_to_grab_pending()
  workqueue: use cwq_set_max_active() helper for workqueue_set_max_active()
  workqueue: introduce cwq_set_max_active() helper for thaw_workqueues()
  workqueue: remove @delayed from cwq_dec_nr_in_flight()
  workqueue: fix possible stall on try_to_grab_pending() of a delayed work item
  workqueue: use hotcpu_notifier() for workqueue_cpu_down_callback()
  workqueue: use __cpuinit instead of __devinit for cpu callbacks
  workqueue: rename manager_mutex to assoc_mutex
  workqueue: WORKER_REBIND is no longer necessary for idle rebinding
  workqueue: WORKER_REBIND is no longer necessary for busy rebinding
  workqueue: reimplement idle worker rebinding
  workqueue: deprecate __cancel_delayed_work()
  workqueue: reimplement cancel_delayed_work() using try_to_grab_pending()
  workqueue: use mod_delayed_work() instead of __cancel + queue
  workqueue: use irqsafe timer for delayed_work
  workqueue: clean up delayed_work initializers and add missing one
  workqueue: make deferrable delayed_work initializer names consistent
  workqueue: cosmetic whitespace updates for macro definitions
  workqueue: deprecate system_nrt[_freezable]_wq
  workqueue: deprecate flush[_delayed]_work_sync()
  ...