Re: Random process lockup on ARM board: alsa-lib-1.0.25, FUTEX_WAIT_PRIVATE

From: Huang Shijie
Date: Wed Feb 29 2012 - 04:13:01 EST


Hi ,

I meet a similar problem with the latest futex code.

I play the video and the processes will hang at the futex.

BR
Huang Shijie

On Fri, Feb 24, 2012 at 8:29 AM, Jonathan Andrews <jon@xxxxxxxxxxxxxxx> wrote:
> Using kernel 3.2.5 with alsa-lib 1.0.25, all compiled with generic
> Debian arm-linux-gnueabi toolchain.
>
> arm-linux-gnueabi-gcc (Debian 4.3.2-1.1) 4.3.2
> Was used to build kernel, alsa-lib and application.
>
> Changing gcc version, kernel version or alsa-lib version makes the
> problem worse or better, but ALL versions seem to suffer this problem. I
> have also seen it once on Intel (but only once so far).
>
> Something seeks broken at a lower layer than im using. ÂI simply don't
> have the skill to debug it.
>
> The hardware is a USB cm109 audio adapter, but the problem seems to show
> on more than this one driver.
>
> The audio application writing to alsa will freezes at random intervals,
> infrequent at the moment, last one was after runtime 20H 37M 29S. ÂTwo
> processes are running, one reading from the sound device and one writing
> to the sound device. I am not using threading or anything very clever
> just generic alsa functions.
>
> This is the only diagnostic I can generate so far as running the
> application under strace slows it to the point it no longer functions
> enough to generate the problem.
>
> ARM / # strace -p 417
> Process 417 attached - interrupt to quit
> futex(0x175734, FUTEX_WAIT_PRIVATE, 2, NULL^C <unfinished ...>
> Process 417 detached
>
> ARM / # uname -a
> Linux (none) 3.2.5 #2 Wed Feb 22 17:11:52 GMT 2012 armv4tl GNU/Linux
> ARM / # uptime
> Â22:36:19 up 22:36, Â0 users, Âload average: 0.15, 0.16, 0.18
> ARM / # cat /proc/cpuinfo
> Processor    : ARM920T rev 0 (v4l)
> BogoMIPS Â Â Â Â: 199.06
> Features    Â: swp half thumb crunch
> CPU implementer : 0x41
> CPU architecture: 4T
> CPU variant   : 0x1
> CPU part    Â: 0x920
> CPU revision  Â: 0
>
>
> Any help welcome.
>
> Thanks,
> Jon
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at Âhttp://vger.kernel.org/majordomo-info.html
> Please read the FAQ at Âhttp://www.tux.org/lkml/
--
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/