linaro-pkg team mailing list archive
-
linaro-pkg team
-
Mailing list archive
-
Message #01261
[Bug 1024409] Re: Kernel panic during streamline data capture in DS5 on vexpress Android platform.
I believe this is the same as the deadlock issue that Jon was working
on, and is already fixed in linux-linaro. The fix should show up in
builds soon, certainly by the time we pick up the next weekly build for
Monday.
** Also affects: linux-linaro
Importance: Undecided
Status: New
** Changed in: linux-linaro
Assignee: (unassigned) => Tixy (Jon Medhurst) (tixy)
--
You received this bug notification because you are a member of Linaro
Maintainers, which is the registrant for Linaro Linux.
https://bugs.launchpad.net/bugs/1024409
Title:
Kernel panic during streamline data capture in DS5 on vexpress Android
platform.
Status in Linaro Android:
New
Status in Linaro Linux:
New
Bug description:
For vexpress board with Linaro Android image:
https://android-build.linaro.org/builds/~linaro-android/vexpress-ics-
gcc47-armlt-tracking-open/#build=78
Kernel panic occurred during the streamline data capture in DS5
environment. Tried twice, same result.
root@android:/ # [ 286.209887] mmcblk0: error -5 transferring data, sector 2060
[ 286.239898] mmcblk0: retrying using single block transfer
[ 286.498473]
[ 286.502940] ======================================================
[ 286.521457] [ INFO: HARDIRQ-safe -> HARDIRQ-unsafe lock order detected ]
[ 286.541536] 3.5.0-rc3-00685-g2bd24f8 #1 Not tainted
[ 286.556139] ------------------------------------------------------
[ 286.574657] gatord-child/2313 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire:
[ 286.595558] (bdev_lock){+.+...}, at: [<c0104e68>] nr_blockdev_pages+0x10/0xc
[ 286.617193]
[ 286.617193] and this task is already holding:
[ 286.634700] (&rq->lock){-.-.-.}, at: [<c00552c0>] idle_balance+0x134/0x18c
[ 286.655552] which would create a new lock dependency:
[ 286.670705] (&rq->lock){-.-.-.} -> (bdev_lock){+.+...}
[ 286.686350]
[ 286.686350] but this new dependency connects a HARDIRQ-irq-safe lock:
[ 286.710084] (&rq->lock){-.-.-.}
[ 286.710084] ... which became HARDIRQ-irq-safe at:
[ 286.733840] [<c00691dc>] __lock_acquire+0x994/0x198c
[ 286.749238] [<c006a6c4>] lock_acquire+0x94/0x108
[ 286.763594] [<c04df4fc>] _raw_spin_lock+0x2c/0x3c
[ 286.778229] [<c004e1dc>] scheduler_tick+0x3c/0x17c
[ 286.793123] [<c002f39c>] update_process_times+0x5c/0x68
[ 286.809301] [<c00628f4>] tick_periodic+0x48/0xc0
[ 286.823654] [<c0062a04>] tick_handle_periodic+0x24/0x88
[ 286.839835] [<c001c7fc>] sp804_timer_interrupt+0x3c/0x4c
[ 286.856271] [<c0077e90>] handle_irq_event_percpu+0x5c/0x24c
[ 286.873486] [<c00780bc>] handle_irq_event+0x3c/0x5c
[ 286.888624] [<c007ad0c>] handle_fasteoi_irq+0x98/0x158
[ 286.904556] [<c0077860>] generic_handle_irq+0x20/0x30
[ 286.920225] [<c000f028>] handle_IRQ+0x4c/0xb0
[ 286.933796] [<c000848c>] gic_handle_irq+0x24/0x58
[ 286.948410] [<c000dcc4>] __irq_svc+0x44/0x60
[ 286.961724] [<c04d21b0>] calibrate_delay+0xc8/0x238
[ 286.976865] [<c068a780>] start_kernel+0x268/0x2f8
[ 286.991476] [<60008044>] 0x60008044
[ 287.002434]
[ 287.002434] to a HARDIRQ-irq-unsafe lock:
[ 287.018877] (bdev_lock){+.+...}
[ 287.018877] ... which became HARDIRQ-irq-unsafe at:
[ 287.043148] ... [<c00690e4>] __lock_acquire+0x89c/0x198c
[ 287.059326] [<c006a6c4>] lock_acquire+0x94/0x108
[ 287.073680] [<c04df4fc>] _raw_spin_lock+0x2c/0x3c
[ 287.088298] [<c0104e68>] nr_blockdev_pages+0x10/0x5c
[ 287.103709] [<c00a2d40>] si_meminfo+0x38/0x50
[ 287.117298] [<c069a944>] eventpoll_init+0x14/0xf4
[ 287.131911] [<c0008630>] do_one_initcall+0xfc/0x168
[ 287.147046] [<c068a90c>] kernel_init+0xfc/0x1c0
[ 287.161147] [<c000f278>] kernel_thread_exit+0x0/0x8
[ 287.176270]
[ 287.176270] other info that might help us debug this:
[ 287.176270]
[ 287.200255] Possible interrupt unsafe locking scenario:
[ 287.200255]
[ 287.220588] CPU0 CPU1
[ 287.234148] ---- ----
[ 287.247716] lock(bdev_lock);
[ 287.256850] local_irq_disable();
[ 287.274584] lock(&rq->lock);
[ 287.291276] lock(bdev_lock);
[ 287.307959] <Interrupt>
[ 287.315796] lock(&rq->lock);
[ 287.325448]
[ 287.325448] *** DEADLOCK ***
[ 287.325448]
[ 287.343185] 1 lock held by gatord-child/2313:
[ 287.356259] #0: (&rq->lock){-.-.-.}, at: [<c00552c0>] idle_balance+0x134/0c
[ 287.378421]
[ 287.378421] the dependencies between HARDIRQ-irq-safe lock and the holding l:
[ 287.405079] -> (&rq->lock){-.-.-.} ops: 670188 {
[ 287.418907] IN-HARDIRQ-W at:
[ 287.428327] [<c00691dc>] __lock_acquire+0x994/0x198c
[ 287.443464] [<c006a6c4>] lock_acquire+0x94/0x108
[ 287.457557] [<c04df4fc>] _raw_spin_lock+0x2c/0x3c
[ 287.471922] [<c004e1dc>] scheduler_tick+0x3c/0x17c
[ 287.486549] [<c002f39c>] update_process_times+0x5c/0x68
[ 287.502467] [<c00628f4>] tick_periodic+0x48/0xc0
[ 287.516560] [<c0062a04>] tick_handle_periodic+0x24/0x88
[ 287.532478] [<c001c7fc>] sp804_timer_interrupt+0x3c/0x4c
[ 287.548653] [<c0077e90>] handle_irq_event_percpu+0x5c/0x24c
[ 287.565608] [<c00780bc>] handle_irq_event+0x3c/0x5c
[ 287.580484] [<c007ad0c>] handle_fasteoi_irq+0x98/0x158
[ 287.596151] [<c0077860>] generic_handle_irq+0x20/0x30
[ 287.611552] [<c000f028>] handle_IRQ+0x4c/0xb0
[ 287.624861] [<c000848c>] gic_handle_irq+0x24/0x58
[ 287.639213] [<c000dcc4>] __irq_svc+0x44/0x60
To manage notifications about this bug go to:
https://bugs.launchpad.net/linaro-android/+bug/1024409/+subscriptions
