BUG: soft lockup when recording audio on MX28EVK with ASoC sgtl5000

From: Hector Palacios
Date: Mon Mar 25 2013 - 07:01:53 EST

Hello,

I just tried recording audio on Freescale's MX28EVK that uses ASoC sgtl5000 (kernel v3.8) with:

arecord -M -f cd sound.wav --duration 10

and got a scheduler message:

[ 789.041847] [sched_delayed] sched: RT throttling activated

The system then becomes nearly unresponsive and after some seconds the kernel complains of a soft lockup.

[ 904.211821] BUG: soft lockup - CPU#0 stuck for 22s! [mdev:279]
[ 904.217693] Modules linked in:
[ 904.220776] irq event stamp: 1576636
[ 904.224363] hardirqs last enabled at (1576635): [<c000ee28>] __irq_svc+0x48/0x54
[ 904.231903] hardirqs last disabled at (1576636): [<c000ee14>] __irq_svc+0x34/0x54
[ 904.239422] softirqs last enabled at (1575782): [<c0026138>] __do_softirq+0x13c/0x220
[ 904.247388] softirqs last disabled at (1575771): [<c002630c>] irq_exit+0x8c/0x94
[ 904.254821]
[ 904.256328] Pid: 279, comm: mdev
[ 904.260967] CPU: 0 Not tainted (3.8.0-00041-gbdf34c0-dirty #49)
[ 904.267272] PC is at cpu_arm926_switch_mm+0x8/0x20
[ 904.272089] LR is at flush_old_exec+0x340/0x5c0
[ 904.276644] pc : [<c001a5c8>] lr : [<c00da6a0>] psr: 00000013
[ 904.276644] sp : c6f73e60 ip : 00000000 fp : c0634c10
[ 904.288138] r10: c779dd20 r9 : c77ac654 r8 : c779d5a0
[ 904.293378] r7 : c6f0ac00 r6 : c77ac400 r5 : c6f72000 r4 : c779dd20
[ 904.299918] r3 : 60000013 r2 : 00000000 r1 : c779d5a0 r0 : 46f7c000
[ 904.306461] Flags: nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
[ 904.313613] Control: 0005317f Table: 46f78000 DAC: 00000015
[ 904.319427] [<c001510c>] (unwind_backtrace+0x0/0xf4) from [<c006c4a8>] (watchdog_timer_fn+0x124/0x160)
[ 904.328795] [<c006c4a8>] (watchdog_timer_fn+0x124/0x160) from [<c0042784>] (__run_hrtimer+0x7c/0x1e8)
[ 904.338061] [<c0042784>] (__run_hrtimer+0x7c/0x1e8) from [<c0042c9c>] (hrtimer_interrupt+0x110/0x310)
[ 904.347329] [<c0042c9c>] (hrtimer_interrupt+0x110/0x310) from [<c001ab80>] (mxs_timer_interrupt+0x1c/0x2
8)
[ 904.357033] [<c001ab80>] (mxs_timer_interrupt+0x1c/0x28) from [<c006cb38>] (handle_irq_event_percpu+0x5c
/0x26c)
[ 904.367169] [<c006cb38>] (handle_irq_event_percpu+0x5c/0x26c) from [<c006cd84>] (handle_irq_event+0x3c/0
x5c)
[ 904.377039] [<c006cd84>] (handle_irq_event+0x3c/0x5c) from [<c006f3b0>] (handle_level_irq+0x8c/0x118)
[ 904.386300] [<c006f3b0>] (handle_level_irq+0x8c/0x118) from [<c006cacc>] (generic_handle_irq+0x28/0x30)
[ 904.395741] [<c006cacc>] (generic_handle_irq+0x28/0x30) from [<c001009c>] (handle_IRQ+0x30/0x84)
[ 904.404568] [<c001009c>] (handle_IRQ+0x30/0x84) from [<c00086ec>] (icoll_handle_irq+0x30/0x44)
[ 904.413220] [<c00086ec>] (icoll_handle_irq+0x30/0x44) from [<c000ee24>] (__irq_svc+0x44/0x54)
[ 904.421761] Exception stack(0xc6f73e18 to 0xc6f73e60)
[ 904.426834] 3e00: 46f7c000 c779d5a0
[ 904.435043] 3e20: 00000000 60000013 c779dd20 c6f72000 c77ac400 c6f0ac00 c779d5a0 c77ac654
[ 904.443250] 3e40: c779dd20 c0634c10 00000000 c6f73e60 c00da6a0 c001a5c8 00000013 ffffffff
[ 904.451472] [<c000ee24>] (__irq_svc+0x44/0x54) from [<c001a5c8>] (cpu_arm926_switch_mm+0x8/0x20)
[ 904.460298] [<c001a5c8>] (cpu_arm926_switch_mm+0x8/0x20) from [<c6e740c0>] (0xc6e740c0)
[ 936.211821] BUG: soft lockup - CPU#0 stuck for 21s! [arecord:264]
[ 936.217952] Modules linked in:
[ 936.221038] irq event stamp: 6749868
[ 936.224624] hardirqs last enabled at (6749867): [<c000ee28>] __irq_svc+0x48/0x54
[ 936.232163] hardirqs last disabled at (6749868): [<c000ee14>] __irq_svc+0x34/0x54
[ 936.239681] softirqs last enabled at (6749012): [<c0026138>] __do_softirq+0x13c/0x220
[ 936.247644] softirqs last disabled at (6748999): [<c002630c>] irq_exit+0x8c/0x94
[ 936.255077]
[ 936.256586] Pid: 264, comm: arecord
[ 936.261223] CPU: 0 Not tainted (3.8.0-00041-gbdf34c0-dirty #49)
[ 936.267528] PC is at cpu_arm926_switch_mm+0x8/0x20
[ 936.272352] LR is at T.1273+0xf4/0x150
[ 936.276125] pc : [<c001a5c8>] lr : [<c004c9d4>] psr: 00000013
[ 936.276125] sp : c6ed1d98 ip : 00000000 fp : c6ed1dc4
[ 936.287618] r10: c0455e0c r9 : c6ec91e0 r8 : 00000001
[ 936.292857] r7 : c7430000 r6 : 00000000 r5 : c062e870 r4 : c6ed0000
[ 936.299398] r3 : c6ec91e0 r2 : 20000013 r1 : c6ec91e0 r0 : 46ec4000
[ 936.305942] Flags: nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
[ 936.313092] Control: 0005317f Table: 4776c000 DAC: 00000015
[ 936.318906] [<c001510c>] (unwind_backtrace+0x0/0xf4) from [<c006c4a8>] (watchdog_timer_fn+0x124/0x160)
[ 936.328273] [<c006c4a8>] (watchdog_timer_fn+0x124/0x160) from [<c0042784>] (__run_hrtimer+0x7c/0x1e8)
[ 936.337539] [<c0042784>] (__run_hrtimer+0x7c/0x1e8) from [<c0042c9c>] (hrtimer_interrupt+0x110/0x310)
[ 936.346806] [<c0042c9c>] (hrtimer_interrupt+0x110/0x310) from [<c001ab80>] (mxs_timer_interrupt+0x1c/0x2
8)
[ 936.356509] [<c001ab80>] (mxs_timer_interrupt+0x1c/0x28) from [<c006cb38>] (handle_irq_event_percpu+0x5c
/0x26c)
[ 936.366647] [<c006cb38>] (handle_irq_event_percpu+0x5c/0x26c) from [<c006cd84>] (handle_irq_event+0x3c/0
x5c)
[ 936.376519] [<c006cd84>] (handle_irq_event+0x3c/0x5c) from [<c006f3b0>] (handle_level_irq+0x8c/0x118)
[ 936.385781] [<c006f3b0>] (handle_level_irq+0x8c/0x118) from [<c006cacc>] (generic_handle_irq+0x28/0x30)
[ 936.395222] [<c006cacc>] (generic_handle_irq+0x28/0x30) from [<c001009c>] (handle_IRQ+0x30/0x84)
[ 936.404050] [<c001009c>] (handle_IRQ+0x30/0x84) from [<c00086ec>] (icoll_handle_irq+0x30/0x44)
[ 936.412701] [<c00086ec>] (icoll_handle_irq+0x30/0x44) from [<c000ee24>] (__irq_svc+0x44/0x54)
[ 936.421243] Exception stack(0xc6ed1d50 to 0xc6ed1d98)
[ 936.426320] 1d40: 46ec4000 c6ec91e0 20000013 c6ec91e0
[ 936.434527] 1d60: c6ed0000 c062e870 00000000 c7430000 00000001 c6ec91e0 c0455e0c c6ed1dc4
[ 936.442727] 1d80: 00000000 c6ed1d98 c004c9d4 c001a5c8 00000013 ffffffff
[ 936.449389] [<c000ee24>] (__irq_svc+0x44/0x54) from [<c001a5c8>] (cpu_arm926_switch_mm+0x8/0x20)
[ 936.458217] [<c001a5c8>] (cpu_arm926_switch_mm+0x8/0x20) from [<c6e303c0>] (0xc6e303c0)

The system does not completely die but it is extremely slow and almost unusable,
I checked that the mxs-saif interrupt (184) counter suddenly increased very quickly (from 288 to 45427 in one second, 94471 the next second, and so on).
I was wondering if this happens in other platforms using this codec or if it is an MXS issue only.

Best regards,
--
Héctor Palacios
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/