Skip to content

L
linux
Switch branch/tag
  1. 28 Jul, 2014 5 commits
    • Hugh Dickins's avatar
      shmem: fix faulting into a hole while it's punched · 8685789b
      Hugh Dickins authored 
      commit f00cdc6d upstream.

Trinity finds that mmap access to a hole while it's punched from shmem
can prevent the madvise(MADV_REMOVE) or fallocate(FALLOC_FL_PUNCH_HOLE)
from completing, until the reader chooses to stop; with the puncher's
hold on i_mutex locking out all other writers until it can complete.

It appears that the tmpfs fault path is too light in comparison with its
hole-punching path, lacking an i_data_sem to obstruct it; but we don't
want to slow down the common case.

Extend shmem_fallocate()'s existing range notification mechanism, so
shmem_fault() can refrain from faulting pages into the hole while it's
punched, waiting instead on i_mutex (when safe to sleep; or repeatedly
faulting when not).

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: default avatarHugh Dickins <hughd@google.com>
Reported-by: default avatarSasha Levin <sasha.levin@oracle.com>
Tested-by: default avatarSasha Levin <sasha.levin@oracle.com>
Cc: Dave Jones <davej@redhat.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: default avatarJiri Slaby <jslaby@suse.cz>
      8685789b
    • Tyler Hall's avatar
      slip: Fix deadlock in write_wakeup · 7b10a526
      Tyler Hall authored 
      commit 661f7fda upstream.

Use schedule_work() to avoid potentially taking the spinlock in
interrupt context.

Commit cc9fa74e ("slip/slcan: added locking in wakeup function") added
necessary locking to the wakeup function and 367525c8/ddcde142 ("can:
slcan: Fix spinlock variant") converted it to spin_lock_bh() because the lock
is also taken in timers.

Disabling softirqs is not sufficient, however, as tty drivers may call
write_wakeup from interrupt context. This driver calls tty->ops->write() with
its spinlock held, which may immediately cause an interrupt on the same CPU and
subsequent spin_bug().

Simply converting to spin_lock_irq/irqsave() prevents this deadlock, but
causes lockdep to point out a possible circular locking dependency
between these locks:

(&(&sl->lock)->rlock){-.....}, at: slip_write_wakeup
(&port_lock_key){-.....}, at: serial8250_handle_irq.part.13

The slip transmit is holding the slip spinlock when calling the tty write.
This grabs the port lock. On an interrupt, the handler grabs the port
lock and calls write_wakeup which grabs the slip lock. This could be a
problem if a serial interrupt occurs on another CPU during the slip
transmit.

To deal with these issues, don't grab the lock in the wakeup function by
deferring the writeout to a workqueue. Also hold the lock during close
when de-assigning the tty pointer to safely disarm the worker and
timers.

This bug is easily reproducible on the first transmit when slip is
used with the standard 8250 serial driver.

[<c0410b7c>] (spin_bug+0x0/0x38) from [<c006109c>] (do_raw_spin_lock+0x60/0x1d0)
 r5:eab27000 r4:ec02754c
[<c006103c>] (do_raw_spin_lock+0x0/0x1d0) from [<c04185c0>] (_raw_spin_lock+0x28/0x2c)
 r10:0000001f r9:eabb814c r8:eabb8140 r7:40070193 r6:ec02754c r5:eab27000
 r4:ec02754c r3:00000000
[<c0418598>] (_raw_spin_lock+0x0/0x2c) from [<bf3a0220>] (slip_write_wakeup+0x50/0xe0 [slip])
 r4:ec027540 r3:00000003
[<bf3a01d0>] (slip_write_wakeup+0x0/0xe0 [slip]) from [<c026e420>] (tty_wakeup+0x48/0x68)
 r6:00000000 r5:ea80c480 r4:eab27000 r3:bf3a01d0
[<c026e3d8>] (tty_wakeup+0x0/0x68) from [<c028a8ec>] (uart_write_wakeup+0x2c/0x30)
 r5:ed68ea90 r4:c06790d8
[<c028a8c0>] (uart_write_wakeup+0x0/0x30) from [<c028dc44>] (serial8250_tx_chars+0x114/0x170)
[<c028db30>] (serial8250_tx_chars+0x0/0x170) from [<c028dffc>] (serial8250_handle_irq+0xa0/0xbc)
 r6:000000c2 r5:00000060 r4:c06790d8 r3:00000000
[<c028df5c>] (serial8250_handle_irq+0x0/0xbc) from [<c02933a4>] (dw8250_handle_irq+0x38/0x64)
 r7:00000000 r6:edd2f390 r5:000000c2 r4:c06790d8
[<c029336c>] (dw8250_handle_irq+0x0/0x64) from [<c028d2f4>] (serial8250_interrupt+0x44/0xc4)
 r6:00000000 r5:00000000 r4:c06791c4 r3:c029336c
[<c028d2b0>] (serial8250_interrupt+0x0/0xc4) from [<c0067fe4>] (handle_irq_event_percpu+0xb4/0x2b0)
 r10:c06790d8 r9:eab27000 r8:00000000 r7:00000000 r6:0000001f r5:edd52980
 r4:ec53b6c0 r3:c028d2b0
[<c0067f30>] (handle_irq_event_percpu+0x0/0x2b0) from [<c006822c>] (handle_irq_event+0x4c/0x6c)
 r10:c06790d8 r9:eab27000 r8:c0673ae0 r7:c05c2020 r6:ec53b6c0 r5:edd529d4
 r4:edd52980
[<c00681e0>] (handle_irq_event+0x0/0x6c) from [<c006b140>] (handle_level_irq+0xe8/0x100)
 r6:00000000 r5:edd529d4 r4:edd52980 r3:00022000
[<c006b058>] (handle_level_irq+0x0/0x100) from [<c00676f8>] (generic_handle_irq+0x30/0x40)
 r5:0000001f r4:0000001f
[<c00676c8>] (generic_handle_irq+0x0/0x40) from [<c000f57c>] (handle_IRQ+0xd0/0x13c)
 r4:ea997b18 r3:000000e0
[<c000f4ac>] (handle_IRQ+0x0/0x13c) from [<c00086c4>] (armada_370_xp_handle_irq+0x4c/0x118)
 r8:000003ff r7:ea997b18 r6:ffffffff r5:60070013 r4:c0674dc0
[<c0008678>] (armada_370_xp_handle_irq+0x0/0x118) from [<c0013840>] (__irq_svc+0x40/0x70)
Exception stack(0xea997b18 to 0xea997b60)
7b00:                                                       00000001 20070013
7b20: 00000000 0000000b 20070013 eab27000 20070013 00000000 ed10103e eab27000
7b40: c06790d8 ea997b74 ea997b60 ea997b60 c04186c0 c04186c8 60070013 ffffffff
 r9:eab27000 r8:ed10103e r7:ea997b4c r6:ffffffff r5:60070013 r4:c04186c8
[<c04186a4>] (_raw_spin_unlock_irqrestore+0x0/0x54) from [<c0288fc0>] (uart_start+0x40/0x44)
 r4:c06790d8 r3:c028ddd8
[<c0288f80>] (uart_start+0x0/0x44) from [<c028982c>] (uart_write+0xe4/0xf4)
 r6:0000003e r5:00000000 r4:ed68ea90 r3:0000003e
[<c0289748>] (uart_write+0x0/0xf4) from [<bf3a0d20>] (sl_xmit+0x1c4/0x228 [slip])
 r10:ed388e60 r9:0000003c r8:ffffffdd r7:0000003e r6:ec02754c r5:ea717eb8
 r4:ec027000
[<bf3a0b5c>] (sl_xmit+0x0/0x228 [slip]) from [<c0368d74>] (dev_hard_start_xmit+0x39c/0x6d0)
 r8:eaf163c0 r7:ec027000 r6:ea717eb8 r5:00000000 r4:00000000
Signed-off-by: default avatarTyler Hall <tylerwhall@gmail.com>
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Andre Naujoks <nautsch2@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
Signed-off-by: default avatarJiri Slaby <jslaby@suse.cz>
      7b10a526
    • Andy Whitcroft's avatar
      ACPI / resources: only reject zero length resources based at address zero · cff58058
      Andy Whitcroft authored 
      commit 867f9d46 upstream.

The recently merged change (in v3.14-rc6) to ACPI resource detection
(below) causes all zero length ACPI resources to be elided from the
table:

  commit b355cee8
  Author: Zhang Rui <rui.zhang@intel.com>
  Date:   Thu Feb 27 11:37:15 2014 +0800

    ACPI / resources: ignore invalid ACPI device resources

This change has caused a regression in (at least) serial port detection
for a number of machines (see LP#1313981 [1]).  These seem to represent
their IO regions (presumably incorrectly) as a zero length region.
Reverting the above commit restores these serial devices.

Only elide zero length resources which lie at address 0.

Fixes: b355cee8 (ACPI / resources: ignore invalid ACPI device resources)
Signed-off-by: default avatarAndy Whitcroft <apw@canonical.com>
Acked-by: default avatarZhang Rui <rui.zhang@intel.com>
Cc: 3.14+ <stable@vger.kernel.org> # 3.14+
Signed-off-by: default avatarRafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: default avatarJiri Slaby <jslaby@suse.cz>
      cff58058
    • Daniel Thompson's avatar
      serial: sirf: Fix compilation failure · a2c27add
      Daniel Thompson authored 
      commit 58eb97c9 upstream.

After 07d410e0) serial: sirf: fix spinlock deadlock issue it is no longer
possiblet to compile this driver. The rename of one of the spinlocks is
faulty. After looking at the original patch I believe this is the correct
fix.

Compile tested using ARM's multi_v7_defconfig
Reported-by: default avatarStephen Rothwell <sfr@canb.auug.org.au>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Qipan Li <Qipan.Li@csr.com>
Signed-off-by: default avatarDaniel Thompson <daniel.thompson@linaro.org>
Acked-by: default avatarBarry Song <baohua@kernel.org>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: default avatarJiri Slaby <jslaby@suse.cz>
      a2c27add
    • Qipan Li's avatar
      serial: sirf: fix spinlock deadlock issue · 67aedbd9
      Qipan Li authored 
      commit 07d410e0 upstream.

commit fb78b811 provide a workaround for
kernel panic, but bring potential deadlock risk. that is in
sirfsoc_rx_tmo_process_tl while enter into sirfsoc_uart_pio_rx_chars
cpu hold uart_port->lock, if uart interrupt comes cpu enter into
sirfsoc_uart_isr and deadlock occurs in getting uart_port->lock.

the patch replace spin_lock version to spin_lock_irq* version to avoid
spinlock dead lock issue. let function tty_flip_buffer_push in tasklet
outof spin_lock_irq* protect area to avoid add the pair of spin_lock and
spin_unlock for tty_flip_buffer_push.
BTW drop self defined unused spinlock protect of tx_lock/rx_lock.

56274.220464] BUG: spinlock lockup suspected on CPU#0, swapper/0/0
[56274.223648]  lock: 0xc05d9db0, .magic: dead4ead, .owner: swapper/0/0,
	.owner_cpu: 0
	[56274.231278] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G
	O 3.10.35 #1
	[56274.238241] [<c0015530>] (unwind_backtrace+0x0/0xf4) from
	[<c00120d8>] (show_stack+0x10/0x14)
	[56274.246742] [<c00120d8>] (show_stack+0x10/0x14) from
	[<c01b11b0>] (do_raw_spin_lock+0x110/0x184)
	[56274.255501] [<c01b11b0>] (do_raw_spin_lock+0x110/0x184) from
	[<c02124c8>] (sirfsoc_uart_isr+0x20/0x42c)
	[56274.264874] [<c02124c8>] (sirfsoc_uart_isr+0x20/0x42c) from
	[<c0075790>] (handle_irq_event_percpu+0x54/0x17c)
	[56274.274758] [<c0075790>] (handle_irq_event_percpu+0x54/0x17c)
	from [<c00758f4>] (handle_irq_event+0x3c/0x5c)
	[56274.284561] [<c00758f4>] (handle_irq_event+0x3c/0x5c) from
	[<c0077fa0>] (handle_level_irq+0x98/0xfc)
	[56274.293670] [<c0077fa0>] (handle_level_irq+0x98/0xfc) from
	[<c0074f44>] (generic_handle_irq+0x2c/0x3c)
	[56274.302952] [<c0074f44>] (generic_handle_irq+0x2c/0x3c) from
	[<c000ef80>] (handle_IRQ+0x40/0x90)
	[56274.311706] [<c000ef80>] (handle_IRQ+0x40/0x90) from
	[<c000dc80>] (__irq_svc+0x40/0x70)
	[56274.319697] [<c000dc80>] (__irq_svc+0x40/0x70) from
	[<c038113c>] (_raw_spin_unlock_irqrestore+0x10/0x48)
	[56274.329158] [<c038113c>]
	(_raw_spin_unlock_irqrestore+0x10/0x48) from [<c0200034>]
	(tty_port_tty_get+0x58/0x90)
	[56274.339213] [<c0200034>] (tty_port_tty_get+0x58/0x90) from
	[<c0212008>] (sirfsoc_uart_pio_rx_chars+0x1c/0xc8)
	[56274.349097] [<c0212008>]
	(sirfsoc_uart_pio_rx_chars+0x1c/0xc8) from [<c0212ef8>]
	(sirfsoc_rx_tmo_process_tl+0xe4/0x1fc)
	[56274.359853] [<c0212ef8>]
	(sirfsoc_rx_tmo_process_tl+0xe4/0x1fc) from [<c0027c04>]
	(tasklet_action+0x84/0x114)
	[56274.369739] [<c0027c04>] (tasklet_action+0x84/0x114) from
	[<c0027db4>] (__do_softirq+0x120/0x200)
	[56274.378585] [<c0027db4>] (__do_softirq+0x120/0x200) from
	[<c0027f44>] (do_softirq+0x54/0x5c)
	[56274.386998] [<c0027f44>] (do_softirq+0x54/0x5c) from
	[<c00281ec>] (irq_exit+0x9c/0xd0)
	[56274.394899] [<c00281ec>] (irq_exit+0x9c/0xd0) from
	[<c000ef84>] (handle_IRQ+0x44/0x90)
	[56274.402790] [<c000ef84>] (handle_IRQ+0x44/0x90) from
	[<c000dc80>] (__irq_svc+0x40/0x70)
	[56274.410774] [<c000dc80>] (__irq_svc+0x40/0x70) from
	[<c0288af4>] (cpuidle_enter_state+0x50/0xe0)
	[56274.419532] [<c0288af4>] (cpuidle_enter_state+0x50/0xe0) from
	[<c0288c34>] (cpuidle_idle_call+0xb0/0x148)
	[56274.429080] [<c0288c34>] (cpuidle_idle_call+0xb0/0x148) from
	[<c000f3ac>] (arch_cpu_idle+0x8/0x38)
	[56274.438016] [<c000f3ac>] (arch_cpu_idle+0x8/0x38) from
	[<c0059344>] (cpu_startup_entry+0xfc/0x140)
	[56274.446956] [<c0059344>] (cpu_startup_entry+0xfc/0x140) from
	[<c04a3a54>] (start_kernel+0x2d8/0x2e4)
Signed-off-by: default avatarQipan Li <Qipan.Li@csr.com>
Signed-off-by: default avatarBarry Song <Baohua.Song@csr.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: default avatarJiri Slaby <jslaby@suse.cz>
      67aedbd9
  3. 22 Jul, 2014 2 commits
    • Thomas Gleixner's avatar
      can: c_can: Remove EOB exit · 9b448624
      Thomas Gleixner authored 
      commit 710c5610 upstream.

The rx_poll code has the following gem:

	if (msg_ctrl_save & IF_MCONT_EOB)
		return num_rx_pkts;

The EOB bit is the indicator for the hardware that this is the last
configured FIFO object. But this object can contain valid data, if we
manage to free up objects before the overrun case hits.

Now if the code exits due to the EOB bit set, then this buffer is
stale and the interrupt bit and NewDat bit of the buffer are still
set. Results in a nice interrupt storm unless we come into an overrun
situation where the MSGLST bit gets set.

     ksoftirqd/0-3     [000] ..s.    79.124101: c_can_poll: rx_poll: val: 00008001 pend 00008001
     ksoftirqd/0-3     [000] ..s.    79.124176: c_can_poll: rx_poll: val: 00008000 pend 00008000
     ksoftirqd/0-3     [000] ..s.    79.124187: c_can_poll: rx_poll: val: 00008002 pend 00008002
     ksoftirqd/0-3     [000] ..s.    79.124256: c_can_poll: rx_poll: val: 00008000 pend 00008000
     ksoftirqd/0-3     [000] ..s.    79.124267: c_can_poll: rx_poll: val: 00008000 pend 00008000

The amazing thing is that the check of the MSGLST (aka overrun bit)
used to be after the check of the EOB bit. That was "fixed" in commit
5d0f801a(can: c_can: Fix RX message handling, handle lost message
before EOB). But the author of this "fix" did not even understand that
the EOB check is broken as well.

Again a simple solution: Remove
Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
[mkl: adjusted subject and commit message]
Signed-off-by: default avatarMarc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: default avatarJiri Slaby <jslaby@suse.cz>
      9b448624
    • Linus Torvalds's avatar
      Don't trigger congestion wait on dirty-but-not-writeout pages · e43bbc2c
      Linus Torvalds authored 
      commit b738d764 upstream.

shrink_inactive_list() used to wait 0.1s to avoid congestion when all
the pages that were isolated from the inactive list were dirty but not
under active writeback.  That makes no real sense, and apparently causes
major interactivity issues under some loads since 3.11.

The ostensible reason for it was to wait for kswapd to start writing
pages, but that seems questionable as well, since the congestion wait
code seems to trigger for kswapd itself as well.  Also, the logic behind
delaying anything when we haven't actually started writeback is not
clear - it only delays actually starting that writeback.

We'll still trigger the congestion waiting if

 (a) the process is kswapd, and we hit pages flagged for immediate
     reclaim

 (b) the process is not kswapd, and the zone backing dev writeback is
     actually congested.

This probably needs to be revisited, but as it is this fixes a reported
regression.

[mhocko@suse.cz: backport to 3.12 stable tree]
Fixes: e2be15f6 ('mm: vmscan: stall page reclaim and writeback pages based on dirty/writepage pages encountered')
Reported-by: default avatarFelipe Contreras <felipe.contreras@gmail.com>
Pinpointed-by: default avatarHillf Danton <dhillf@gmail.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: default avatarMichal Hocko <mhocko@suse.cz>
Signed-off-by: default avatarJiri Slaby <jslaby@suse.cz>
      e43bbc2c
  5. 18 Jul, 2014 33 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7