[Buildroot] AT91SAM9G20-EK and aplay

[Buildroot] AT91SAM9G20-EK and aplay

[Carlo Rajko]

Hello,

I have a board AT91SAM9G20-EK Rev C.

With buildroot-2013.02 use kernel 3.1.6 and use toolchain
CODESOURCERY_ARM201203

The kernel detects sound card, but when use prg aplay or use mpg123 to
play a file, receive a "Segmentation fault"

if I run the command "cat file.wav > /dev/dsp" the sound card works fine.

With buildroot-2011.11 is all ok.

Can you help me ?


Best regards,
carlo

[Buildroot] AT91SAM9G20-EK and aplay

[Thomas Petazzoni]

Dear Carlo Rajko,

On Fri, 24 May 2013 09:36:11 +0200, Carlo Rajko wrote:

> With buildroot-2013.02 use kernel 3.1.6 and use toolchain
> CODESOURCERY_ARM201203
> 
> The kernel detects sound card, but when use prg aplay or use mpg123 to
> play a file, receive a "Segmentation fault"
> 
> if I run the command "cat file.wav > /dev/dsp" the sound card works fine.
> 
> With buildroot-2011.11 is all ok.
> 
> Can you help me ?

Do you have a few more messages in the kernel logs that given more
details about the segmentation fault?

Otherwise, do "ulimit -c unlimited", run the crashing application, it
will generate a 'core' file. Transfer it to your development PC, and
run cross-gdb on it:

	./output/host/usr/bin/arm-linux-gdb -c core ./output/target/usr/bin/aplay

and then see what's the reason of the segmentation fault.

Also, could you post on http://code.bulix.org your Buildroot .config
and give us the link? Just to check there isn't anything obviously
wrong in your configuration file.

Best regards,

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[Buildroot] AT91SAM9G20-EK and aplay

[Carlo Rajko]

Il 24/05/2013 10:05, Thomas Petazzoni ha scritto:
> Dear Carlo Rajko,
> 
> On Fri, 24 May 2013 09:36:11 +0200, Carlo Rajko wrote:
> 
>> With buildroot-2013.02 use kernel 3.1.6 and use toolchain
>> CODESOURCERY_ARM201203
>>
>> The kernel detects sound card, but when use prg aplay or use mpg123 to
>> play a file, receive a "Segmentation fault"
>>
>> if I run the command "cat file.wav > /dev/dsp" the sound card works fine.
>>
>> With buildroot-2011.11 is all ok.
>>
>> Can you help me ?
> 
> Do you have a few more messages in the kernel logs that given more
> details about the segmentation fault?
no message

> 
> Otherwise, do "ulimit -c unlimited", run the crashing application, it
> will generate a 'core' file. Transfer it to your development PC, and
> run cross-gdb on it:
> 
> 	./output/host/usr/bin/arm-linux-gdb -c core ./output/target/usr/bin/aplay
> 
> and then see what's the reason of the segmentation fault.

I do not have arm-linux-gdb but have arm-none-linux-gnueabi-gdb.
But when use the command
./output/host/opt/ext-toolchain/bin/arm-none-linux-gnueabi-gdb -c core
./output/target/usr/bin/aplay and use command "run" the debugger show:
Don't know how to run.  Try "help target".

I think the problem is not in aply because I have the same problem with
mpg123.
Could be the library alsa?

> 
> Also, could you post on http://code.bulix.org your Buildroot .config
> and give us the link? Just to check there isn't anything obviously
> wrong in your configuration file.

when I press buttom "PASTE" I have error "Antispam verification failed"

Best regards,

Carlo

[Buildroot] AT91SAM9G20-EK and aplay

[Thomas Petazzoni]

Dear Carlo Rajko,

On Fri, 24 May 2013 11:03:23 +0200, Carlo Rajko wrote:


> I do not have arm-linux-gdb but have arm-none-linux-gnueabi-gdb.
> But when use the command
> ./output/host/opt/ext-toolchain/bin/arm-none-linux-gnueabi-gdb -c core
> ./output/target/usr/bin/aplay and use command "run" the debugger show:
> Don't know how to run.  Try "help target".

Don't use "run", "run" is to start the program. Here we want to do a
post-mortem analysis of the crash using the core file. Google "gdb core
file" to have more details about what a core file is.

Can you run the gdb command, and copy/paste what you see? Without
entering any command *inside* gdb, it should already tell you which
instruction has faulted, and for what reason.

> I think the problem is not in aply because I have the same problem with
> mpg123.
> Could be the library alsa?

It could. But at least gdb will tell us if it's an undefined
instruction, an invalid memory access, an unaligned access, or
something else.

> > Also, could you post on http://code.bulix.org your Buildroot .config
> > and give us the link? Just to check there isn't anything obviously
> > wrong in your configuration file.
> 
> when I press buttom "PASTE" I have error "Antispam verification failed"

Right left to the "Paste" button there is a little box to train your
mental calculation. You should pay attention to it :-)

Thanks,

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[Buildroot] AT91SAM9G20-EK and aplay

[Carlo Rajko]

Il 24/05/2013 11:23, Thomas Petazzoni ha scritto:
> Dear Carlo Rajko,
> 
> On Fri, 24 May 2013 11:03:23 +0200, Carlo Rajko wrote:
> 
> 
>> I do not have arm-linux-gdb but have arm-none-linux-gnueabi-gdb.
>> But when use the command
>> ./output/host/opt/ext-toolchain/bin/arm-none-linux-gnueabi-gdb -c core
>> ./output/target/usr/bin/aplay and use command "run" the debugger show:
>> Don't know how to run.  Try "help target".
> 
> Don't use "run", "run" is to start the program. Here we want to do a
> post-mortem analysis of the crash using the core file. Google "gdb core
> file" to have more details about what a core file is.
> 
> Can you run the gdb command, and copy/paste what you see? Without
> entering any command *inside* gdb, it should already tell you which
> instruction has faulted, and for what reason.
> 
OK.

GNU gdb (Sourcery CodeBench Lite 2012.03-57) 7.2.50.20100908-cvs
Copyright (C) 2010 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "--host=i686-pc-linux-gnu
--target=arm-none-linux-gnueabi".
For bug reporting instructions, please see:
<https://support.codesourcery.com/GNUToolchain/>...
Reading symbols from /home/carlo/debug/aplay...(no debugging symbols
found)...done.

warning: exec file is newer than core file.
[New LWP 524]
Reading symbols from /home/carlo/debug/librt.so.1...(no debugging
symbols found)...done.
Loaded symbols for /home/carlo/debug/librt.so.1
Reading symbols from /home/carlo/debug/libasound.so.2...(no debugging
symbols found)...done.
Loaded symbols for /home/carlo/debug/libasound.so.2
Reading symbols from /home/carlo/debug/libm.so.6...(no debugging symbols
found)...done.
Loaded symbols for /home/carlo/debug/libm.so.6
Reading symbols from /home/carlo/debug/libdl.so.2...(no debugging
symbols found)...done.
Loaded symbols for /home/carlo/debug/libdl.so.2
Reading symbols from /home/carlo/debug/libpthread.so.0...(no debugging
symbols found)...done.
Loaded symbols for /home/carlo/debug/libpthread.so.0
Reading symbols from /home/carlo/debug/libgcc_s.so.1...(no debugging
symbols found)...done.
Loaded symbols for /home/carlo/debug/libgcc_s.so.1
Reading symbols from /home/carlo/debug/libc.so.6...(no debugging symbols
found)...done.
Loaded symbols for /home/carlo/debug/libc.so.6
Reading symbols from /home/carlo/debug/ld-linux.so.3...(no debugging
symbols found)...done.
Loaded symbols for /home/carlo/debug/ld-linux.so.3
Core was generated by `/mnt/2/sox/aplay'.
Program terminated with signal 11, Segmentation fault.
#0  0x400f31cc in ?? () from /home/carlo/debug/ld-linux.so.3
(gdb)


>> I think the problem is not in aply because I have the same problem with
>> mpg123.
>> Could be the library alsa?
> 
> It could. But at least gdb will tell us if it's an undefined
> instruction, an invalid memory access, an unaligned access, or
> something else.
> 
>>> Also, could you post on http://code.bulix.org your Buildroot .config
>>> and give us the link? Just to check there isn't anything obviously
>>> wrong in your configuration file.
>>
>> when I press buttom "PASTE" I have error "Antispam verification failed"
> 
> Right left to the "Paste" button there is a little box to train your
> mental calculation. You should pay attention to it :-)
OK, sorry.

http://code.bulix.org/bazvmv-83592?raw

Best regards,

Carlo

[Buildroot] AT91SAM9G20-EK and aplay

[Gustavo Zacarias]

On 05/24/2013 07:47 AM, Carlo Rajko wrote:
>>> I think the problem is not in aply because I have the same problem with
>>> mpg123.
>>> Could be the library alsa?
>>
>> It could. But at least gdb will tell us if it's an undefined
>> instruction, an invalid memory access, an unaligned access, or
>> something else.

Actually it's the toolchain, seems CS 2012.03 isn't emitting proper
arm926ej-s code.
I've managed to reproduce it with the qemu_arm_versatile sample
configuration and adding a bunch of packages (mpg123, vorbis-tools, ...)
It did strike me as odd that i've got the same failures as yours (aplay,
mpg123) but ogg123 worked fine with alsa output.
So switching to an internal toolchain or CS 2011.09 magically fixed the
issues.
Regards.

[Buildroot] AT91SAM9G20-EK and aplay

[Carlo Rajko]

Il 24/05/2013 14:50, Gustavo Zacarias ha scritto:
> On 05/24/2013 07:47 AM, Carlo Rajko wrote:
>>>> I think the problem is not in aply because I have the same problem with
>>>> mpg123.
>>>> Could be the library alsa?
>>>
>>> It could. But at least gdb will tell us if it's an undefined
>>> instruction, an invalid memory access, an unaligned access, or
>>> something else.
> 
> Actually it's the toolchain, seems CS 2012.03 isn't emitting proper
> arm926ej-s code.
> I've managed to reproduce it with the qemu_arm_versatile sample
> configuration and adding a bunch of packages (mpg123, vorbis-tools, ...)
> It did strike me as odd that i've got the same failures as yours (aplay,
> mpg123) but ogg123 worked fine with alsa output.
> So switching to an internal toolchain or CS 2011.09 magically fixed the
> issues.
> Regards.
> 

I switch to CS 2011.09 and now is ok, thanks.

Regards
Carlo

[Buildroot] AT91SAM9G20-EK and aplay

[Thomas Petazzoni]

Dear Carlo Rajko,

On Mon, 27 May 2013 10:03:50 +0200, Carlo Rajko wrote:

> > Actually it's the toolchain, seems CS 2012.03 isn't emitting proper
> > arm926ej-s code.
> > I've managed to reproduce it with the qemu_arm_versatile sample
> > configuration and adding a bunch of packages (mpg123, vorbis-tools, ...)
> > It did strike me as odd that i've got the same failures as yours (aplay,
> > mpg123) but ogg123 worked fine with alsa output.
> > So switching to an internal toolchain or CS 2011.09 magically fixed the
> > issues.
> 
> I switch to CS 2011.09 and now is ok, thanks.

It's a problem worth reporting to Mentor Graphics, IMO.

Even though I doubt anything will happen, last time I reported a bug,
nothing really happened, unfortunately.

Best regards,

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[Buildroot] AT91SAM9G20-EK and aplay

[Carlo Rajko]

Il 27/05/2013 11:16, Thomas Petazzoni ha scritto:
> Dear Carlo Rajko,
> 
> On Mon, 27 May 2013 10:03:50 +0200, Carlo Rajko wrote:
> 
>>> Actually it's the toolchain, seems CS 2012.03 isn't emitting proper
>>> arm926ej-s code.
>>> I've managed to reproduce it with the qemu_arm_versatile sample
>>> configuration and adding a bunch of packages (mpg123, vorbis-tools, ...)
>>> It did strike me as odd that i've got the same failures as yours (aplay,
>>> mpg123) but ogg123 worked fine with alsa output.
>>> So switching to an internal toolchain or CS 2011.09 magically fixed the
>>> issues.
>>
>> I switch to CS 2011.09 and now is ok, thanks.
> 
> It's a problem worth reporting to Mentor Graphics, IMO.
> 
> Even though I doubt anything will happen, last time I reported a bug,
> nothing really happened, unfortunately.

I'll try with the new version 2013.05, just available in the package
buildroot, maybe the problem is already solved.

Regards,
Carlo

[Buildroot] AT91SAM9G20-EK and aplay

[Arnout Vandecappelle]

On 27/05/13 12:28, Carlo Rajko wrote:
> Il 27/05/2013 11:16, Thomas Petazzoni ha scritto:
>> Dear Carlo Rajko,
>>
>> On Mon, 27 May 2013 10:03:50 +0200, Carlo Rajko wrote:
>>
>>>> Actually it's the toolchain, seems CS 2012.03 isn't emitting proper
>>>> arm926ej-s code.
>>>> I've managed to reproduce it with the qemu_arm_versatile sample
>>>> configuration and adding a bunch of packages (mpg123, vorbis-tools, ...)
>>>> It did strike me as odd that i've got the same failures as yours (aplay,
>>>> mpg123) but ogg123 worked fine with alsa output.
>>>> So switching to an internal toolchain or CS 2011.09 magically fixed the
>>>> issues.
>>>
>>> I switch to CS 2011.09 and now is ok, thanks.
>>
>> It's a problem worth reporting to Mentor Graphics, IMO.
>>
>> Even though I doubt anything will happen, last time I reported a bug,
>> nothing really happened, unfortunately.
>
> I'll try with the new version 2013.05, just available in the package
> buildroot, maybe the problem is already solved.

  We should probably blacklist the failing CS toolchains for the 
arm926ej-s, then.

  Regards,
  Arnout


-- 
Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F

[Buildroot] AT91SAM9G20-EK and aplay

[Carlo Rajko]

Il 28/05/2013 19:02, Arnout Vandecappelle ha scritto:
> On 27/05/13 12:28, Carlo Rajko wrote:
>> Il 27/05/2013 11:16, Thomas Petazzoni ha scritto:
>>> Dear Carlo Rajko,
>>>
>>> On Mon, 27 May 2013 10:03:50 +0200, Carlo Rajko wrote:
>>>
>>>>> Actually it's the toolchain, seems CS 2012.03 isn't emitting proper
>>>>> arm926ej-s code.
>>>>> I've managed to reproduce it with the qemu_arm_versatile sample
>>>>> configuration and adding a bunch of packages (mpg123, vorbis-tools,
>>>>> ...)
>>>>> It did strike me as odd that i've got the same failures as yours
>>>>> (aplay,
>>>>> mpg123) but ogg123 worked fine with alsa output.
>>>>> So switching to an internal toolchain or CS 2011.09 magically fixed
>>>>> the
>>>>> issues.
>>>>
>>>> I switch to CS 2011.09 and now is ok, thanks.
>>>
>>> It's a problem worth reporting to Mentor Graphics, IMO.
>>>
>>> Even though I doubt anything will happen, last time I reported a bug,
>>> nothing really happened, unfortunately.
>>
>> I'll try with the new version 2013.05, just available in the package
>> buildroot, maybe the problem is already solved.
> 
>  We should probably blacklist the failing CS toolchains for the
> arm926ej-s, then.

I tried version CS 2013.05 successfully for my problem.
But there could be other hidden problems

Regards,
Carlo

[Buildroot] AT91SAM9G20-EK and aplay

[Aleksandar Zivkovic]

On 5/24/2013 2:50 PM, Gustavo Zacarias wrote:
> On 05/24/2013 07:47 AM, Carlo Rajko wrote:
>>>> I think the problem is not in aply because I have the same problem with
>>>> mpg123.
>>>> Could be the library alsa?
>>>
>>> It could. But at least gdb will tell us if it's an undefined
>>> instruction, an invalid memory access, an unaligned access, or
>>> something else.
>
> Actually it's the toolchain, seems CS 2012.03 isn't emitting proper
> arm926ej-s code.
> I've managed to reproduce it with the qemu_arm_versatile sample
> configuration and adding a bunch of packages (mpg123, vorbis-tools, ...)
> It did strike me as odd that i've got the same failures as yours (aplay,
> mpg123) but ogg123 worked fine with alsa output.
> So switching to an internal toolchain or CS 2011.09 magically fixed the
> issues.
> Regards.
>
Hi,

I have segfault issue with aplay on Buildroot generated image x86 with 
Linux kernel 2.6.35.
I found some post where modification of Linux kernel source helps 
http://permalink.gmane.org/gmane.linux.alsa.devel/86868. Could it be 
related to my problem?

Here are my logs, please ask for more details you need:

*******************************************************************************
GDB backtrace:
*******************************************************************************
prs at none:~$ gdb aplay
GNU gdb (GDB) 7.5.1
Copyright (C) 2012 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later 
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "i686-buildroot-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>...
Reading symbols from /usr/bin/aplay...done.
(gdb) r
Starting program: /usr/bin/aplay
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/libthread_db.so.1".

Program received signal SIGSEGV, Segmentation fault.
0xb774f39d in ?? () from /lib/ld-linux.so.2
(gdb) bt
#0  0xb774f39d in ?? () from /lib/ld-linux.so.2
#1  0xb774fb09 in ?? () from /lib/ld-linux.so.2
#2  0xb758e5a0 in ?? () from /lib/libc.so.6
#3  0xb758ea27 in _dl_sym () from /lib/libc.so.6
#4  0xb761cd36 in ?? () from /lib/libdl.so.2
#5  0xb775479f in ?? () from /lib/ld-linux.so.2
#6  0xb761d30a in ?? () from /lib/libdl.so.2
#7  0xb761cdb4 in dlsym () from /lib/libdl.so.2
#8  0xb767595a in snd_dlsym_verify () from /usr/lib/libasound.so.2
#9  0xb7675a3f in snd_dlsym () from /usr/lib/libasound.so.2
#10 0xb7671ce2 in _snd_config_evaluate () from /usr/lib/libasound.so.2
#11 0xb767019d in snd_config_walk () from /usr/lib/libasound.so.2
#12 0xb76701ff in snd_config_walk () from /usr/lib/libasound.so.2
#13 0xb76710c7 in snd_config_expand () from /usr/lib/libasound.so.2
#14 0xb767129e in snd_config_hook_load () from /usr/lib/libasound.so.2
#15 0xb76718fd in snd_config_hook_load_for_all_cards () from 
/usr/lib/libasound.so.2
#16 0xb76722c5 in snd_config_hooks.constprop.24 () from 
/usr/lib/libasound.so.2
#17 0xb76724d6 in snd_config_searcha_hooks () from /usr/lib/libasound.so.2
#18 0xb7672652 in snd1_config_search_alias_hooks () from 
/usr/lib/libasound.so.2
#19 0xb76719cb in snd_config_search_definition () from 
/usr/lib/libasound.so.2
#20 0xb768946f in snd_pcm_open_noupdate () from /usr/lib/libasound.so.2
#21 0x08050a09 in main ()
(gdb)

*******************************************************************************
soundcard details:
*******************************************************************************
$cat /proc/asound/cards
  0 [Intel          ]: HDA-Intel - HDA Intel
                       HDA Intel at 0xfdff4000 irq 44
  1 [USB            ]: USB-Audio - Aureon 7.1 USB
                       Aureon 7.1 USB at usb-0000:00:1d.2-2, full speed



Regards,
Aleksandar