Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Robert Hardy]

For some reason I can't comment on an existing bug which I didn't author. 
Perhaps this requires a higher level of access?

I'm user rhardy on Mantis...

Basically I was trying to point out I am also affected by this bug:
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1938

Two possibilities exist:
1. I missed some of the changes tiwai made. Please point out what if this is
the case.

2. It really isn't fixed in alsa-lib CVS (or isn't visible in viewcvs.)

Testing out the recent changes made by tiwai here:
http://cvs.sourceforge.net/viewcvs.py/alsa/alsa-lib/src/pcm/?sortby=date#dirlist

I took alsa-lib-1.0.11rc4 and applied patches extracted from view cvs above
to update the follow files to the listed versions below:

pcm_dmix.c: 1.102
pcm_plugin.h: 1.41
pcm.c: 1.328
pcm_hw.c: 1.149
pcm_rate.c: 1.113
pcm_direct.c: 1.60
pcm_plugin.c: 1.101
pcm_softvol.c: 1.18

This is the resulting patch:
http://webcon.ca/~rhardy/patches/alsa-lib-1.0.11rc4-pcm_fixes.patch.gz

This unfortunately made no difference here. I'm running 2.6.16-rh3 which is 2.6.16.1 with
custom patches including the recent git commit for [ALSA] version 1.0.11rc4 from here:
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=1c2e02750b992703a8a18634e08b04353face243

My sound card:
02:07.0 Multimedia audio controller: Creative Labs SB0400 Audigy2 Value
         Subsystem: Creative Labs: Unknown device 1001
         Flags: bus master, medium devsel, latency 32, IRQ 201
         I/O ports at a000 [size=64]
         Capabilities: [dc] Power Management version 2

This card once worked fine unfortunately I'm not sure with what
mplayer/alsa-lib/alsa-driver/kernel....

I'm running a i686/32bit mplayer in a 64 bit environment.

After applying the patch above and upgrading alsa-lib, I still get the same
errors from mplayer when I try to seek 6 channel audio using surround51:

alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match \
with appl_ptr (216531) % buf_size (16384)
alsa-play: write error: Broken pipe
alsa-play: trying to reset soundcard
alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match \
with appl_ptr (216531) % buf_size (16384)
alsa-play: write error: Broken pipe
alsa-play: trying to reset soundcard

[same 3 lines repeat forever at 100% cpu usage until I kill mplayer]

Regards,
Rob

-- 
---------------------"Happiness is understanding."----------------------
Robert Hardy, B.Eng Computer Systems                  C.E.O. Webcon Inc.
rhardy <at> webcon <dot> ca    GPG Key available          (613) 276-7327


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Takashi Iwai]

At Wed, 29 Mar 2006 17:18:59 -0500 (EST),
Robert Hardy wrote:
> 
> For some reason I can't comment on an existing bug which I didn't author. 
> Perhaps this requires a higher level of access?
> 
> I'm user rhardy on Mantis...
> 
> Basically I was trying to point out I am also affected by this bug:
> https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1938
> 
> Two possibilities exist:
> 1. I missed some of the changes tiwai made. Please point out what if this is
> the case.
> 
> 2. It really isn't fixed in alsa-lib CVS (or isn't visible in viewcvs.)
> 
> Testing out the recent changes made by tiwai here:
> http://cvs.sourceforge.net/viewcvs.py/alsa/alsa-lib/src/pcm/?sortby=date#dirlist
> 
> I took alsa-lib-1.0.11rc4 and applied patches extracted from view cvs above
> to update the follow files to the listed versions below:
> 
> pcm_dmix.c: 1.102
> pcm_plugin.h: 1.41
> pcm.c: 1.328
> pcm_hw.c: 1.149
> pcm_rate.c: 1.113
> pcm_direct.c: 1.60
> pcm_plugin.c: 1.101
> pcm_softvol.c: 1.18
> 
> This is the resulting patch:
> http://webcon.ca/~rhardy/patches/alsa-lib-1.0.11rc4-pcm_fixes.patch.gz
> 
> This unfortunately made no difference here. I'm running 2.6.16-rh3 which is 2.6.16.1 with
> custom patches including the recent git commit for [ALSA] version 1.0.11rc4 from here:
> http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=1c2e02750b992703a8a18634e08b04353face243
> 
> My sound card:
> 02:07.0 Multimedia audio controller: Creative Labs SB0400 Audigy2 Value
>          Subsystem: Creative Labs: Unknown device 1001
>          Flags: bus master, medium devsel, latency 32, IRQ 201
>          I/O ports at a000 [size=64]
>          Capabilities: [dc] Power Management version 2
> 
> This card once worked fine unfortunately I'm not sure with what
> mplayer/alsa-lib/alsa-driver/kernel....
> 
> I'm running a i686/32bit mplayer in a 64 bit environment.

Strange.  It works fine on my test syetm now with 32bit mplayer on
64bit kernel.


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Takashi Iwai]

At Thu, 30 Mar 2006 11:45:09 +0200,
I wrote:
> 
> At Wed, 29 Mar 2006 17:18:59 -0500 (EST),
> Robert Hardy wrote:
> > 
> > For some reason I can't comment on an existing bug which I didn't author. 
> > Perhaps this requires a higher level of access?
> > 
> > I'm user rhardy on Mantis...
> > 
> > Basically I was trying to point out I am also affected by this bug:
> > https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1938
> > 
> > Two possibilities exist:
> > 1. I missed some of the changes tiwai made. Please point out what if this is
> > the case.
> > 
> > 2. It really isn't fixed in alsa-lib CVS (or isn't visible in viewcvs.)
> > 
> > Testing out the recent changes made by tiwai here:
> > http://cvs.sourceforge.net/viewcvs.py/alsa/alsa-lib/src/pcm/?sortby=date#dirlist
> > 
> > I took alsa-lib-1.0.11rc4 and applied patches extracted from view cvs above
> > to update the follow files to the listed versions below:
> > 
> > pcm_dmix.c: 1.102
> > pcm_plugin.h: 1.41
> > pcm.c: 1.328
> > pcm_hw.c: 1.149
> > pcm_rate.c: 1.113
> > pcm_direct.c: 1.60
> > pcm_plugin.c: 1.101
> > pcm_softvol.c: 1.18
> > 
> > This is the resulting patch:
> > http://webcon.ca/~rhardy/patches/alsa-lib-1.0.11rc4-pcm_fixes.patch.gz
> > 
> > This unfortunately made no difference here. I'm running 2.6.16-rh3 which is 2.6.16.1 with
> > custom patches including the recent git commit for [ALSA] version 1.0.11rc4 from here:
> > http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=1c2e02750b992703a8a18634e08b04353face243
> > 
> > My sound card:
> > 02:07.0 Multimedia audio controller: Creative Labs SB0400 Audigy2 Value
> >          Subsystem: Creative Labs: Unknown device 1001
> >          Flags: bus master, medium devsel, latency 32, IRQ 201
> >          I/O ports at a000 [size=64]
> >          Capabilities: [dc] Power Management version 2
> > 
> > This card once worked fine unfortunately I'm not sure with what
> > mplayer/alsa-lib/alsa-driver/kernel....
> > 
> > I'm running a i686/32bit mplayer in a 64 bit environment.
> 
> Strange.  It works fine on my test syetm now with 32bit mplayer on
> 64bit kernel.

Maybe a dumb question: Did you properly install 32bit library?

I often compile/install 64bit lib for debugging 32bit problems, and
throw precious hours aways...


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Robert Hardy]

On Thu, 30 Mar 2006, Takashi Iwai wrote:
> At Thu, 30 Mar 2006 11:45:09 +0200,
>> At Wed, 29 Mar 2006 17:18:59 -0500 (EST),
>> Robert Hardy wrote:
>>> Basically I was trying to point out I am also affected by this bug:
>>> https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1938
>>>
>>> This unfortunately made no difference here. I'm running 2.6.16-rh3 which is 2.6.16.1 with
>>> custom patches including the recent git commit for [ALSA] version 1.0.11rc4 from here:
>>> http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=1c2e02750b992703a8a18634e08b04353face243
>>>
>>> My sound card:
>>> 02:07.0 Multimedia audio controller: Creative Labs SB0400 Audigy2 Value
>>>          Subsystem: Creative Labs: Unknown device 1001
>>>
>>> This card once worked fine unfortunately I'm not sure with what
>>> mplayer/alsa-lib/alsa-driver/kernel....
>>>
>>> I'm running a i686/32bit mplayer in a 64 bit environment.

>> Strange.  It works fine on my test syetm now with 32bit mplayer on
>> 64bit kernel.
>
> Maybe a dumb question: Did you properly install 32bit library?
>
> I often compile/install 64bit lib for debugging 32bit problems, and
> throw precious hours aways...

I rolled those patches into my alsa-lib rpm before I tested them. Both my
i686 and x86_64 rpms have the patches. My currrent system has:

> rpm -qa --qf='%{name}-%{version}-%{release}-%{arch}\n' | grep alsa
alsa-lib-devel-1.0.11-3.rc4.2.el4-x86_64
alsa-utils-1.0.11-2.rc4.1.el4-x86_64
alsa-lib-1.0.11-3.rc4.2.el4-i686
alsa-lib-devel-1.0.11-3.rc4.2.el4-i686
alsa-lib-1.0.11-3.rc4.2.el4-x86_64

Wanting to make sure it wasn't some how getting confused by the x86_64
libraries, I just now tested only the i686 rpms in my chrooted el4 i386
environment and the same problem happened.

In Chroot> rpm -qa --qf='%{name}-%{version}-%{release}-%{arch}\n' | grep alsa
alsa-utils-1.0.11-2.rc4.1.el4-i686
alsa-lib-devel-1.0.11-3.rc4.2.el4-i686
alsa-lib-1.0.11-3.rc4.2.el4-i686

Regards,
Rob

-- 
---------------------"Happiness is understanding."----------------------
Robert Hardy, B.Eng Computer Systems                  C.E.O. Webcon Inc.
rhardy <at> webcon <dot> ca    GPG Key available          (613) 276-7327


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Takashi Iwai]

At Thu, 30 Mar 2006 11:25:08 -0500 (EST),
Robert Hardy wrote:
> 
> On Thu, 30 Mar 2006, Takashi Iwai wrote:
> > At Thu, 30 Mar 2006 11:45:09 +0200,
> >> At Wed, 29 Mar 2006 17:18:59 -0500 (EST),
> >> Robert Hardy wrote:
> >>> Basically I was trying to point out I am also affected by this bug:
> >>> https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1938
> >>>
> >>> This unfortunately made no difference here. I'm running 2.6.16-rh3 which is 2.6.16.1 with
> >>> custom patches including the recent git commit for [ALSA] version 1.0.11rc4 from here:
> >>> http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=1c2e02750b992703a8a18634e08b04353face243
> >>>
> >>> My sound card:
> >>> 02:07.0 Multimedia audio controller: Creative Labs SB0400 Audigy2 Value
> >>>          Subsystem: Creative Labs: Unknown device 1001
> >>>
> >>> This card once worked fine unfortunately I'm not sure with what
> >>> mplayer/alsa-lib/alsa-driver/kernel....
> >>>
> >>> I'm running a i686/32bit mplayer in a 64 bit environment.
> 
> >> Strange.  It works fine on my test syetm now with 32bit mplayer on
> >> 64bit kernel.
> >
> > Maybe a dumb question: Did you properly install 32bit library?
> >
> > I often compile/install 64bit lib for debugging 32bit problems, and
> > throw precious hours aways...
> 
> I rolled those patches into my alsa-lib rpm before I tested them. Both my
> i686 and x86_64 rpms have the patches. My currrent system has:
> 
> > rpm -qa --qf='%{name}-%{version}-%{release}-%{arch}\n' | grep alsa
> alsa-lib-devel-1.0.11-3.rc4.2.el4-x86_64
> alsa-utils-1.0.11-2.rc4.1.el4-x86_64
> alsa-lib-1.0.11-3.rc4.2.el4-i686
> alsa-lib-devel-1.0.11-3.rc4.2.el4-i686
> alsa-lib-1.0.11-3.rc4.2.el4-x86_64
> 
> Wanting to make sure it wasn't some how getting confused by the x86_64
> libraries, I just now tested only the i686 rpms in my chrooted el4 i386
> environment and the same problem happened.

OK, which problem exactly?  All tests failed?


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Robert Hardy]

On Thu, 30 Mar 2006, Takashi Iwai wrote:
> At Thu, 30 Mar 2006 11:25:08 -0500 (EST),> Robert Hardy wrote:
>> On Thu, 30 Mar 2006, Takashi Iwai wrote:
>>> At Thu, 30 Mar 2006 11:45:09 +0200,
>>>> At Wed, 29 Mar 2006 17:18:59 -0500 (EST),
>>>> Robert Hardy wrote:
>>>>> Basically I was trying to point out I am also affected by this bug:
>>>>> https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1938
>>>>>
>>>>> This unfortunately made no difference here. I'm running 2.6.16-rh3 which is 2.6.16.1 with
>>>>> custom patches including the recent git commit for [ALSA] version 1.0.11rc4 from here:
>>>>> http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=1c2e02750b992703a8a18634e08b04353face243
>>>>>
>>>>> My sound card:
>>>>> 02:07.0 Multimedia audio controller: Creative Labs SB0400 Audigy2 Value
>>>>>          Subsystem: Creative Labs: Unknown device 1001
>>>>>
>>>>> This card once worked fine unfortunately I'm not sure with what
>>>>> mplayer/alsa-lib/alsa-driver/kernel....
>>>>>
>>>>> I'm running a i686/32bit mplayer in a 64 bit environment.
>>
>>>> Strange.  It works fine on my test syetm now with 32bit mplayer on
>>>> 64bit kernel.
>>>
>>> Maybe a dumb question: Did you properly install 32bit library?
>>>
>>> I often compile/install 64bit lib for debugging 32bit problems, and
>>> throw precious hours aways...
>>
>> I rolled those patches into my alsa-lib rpm before I tested them. Both my
>> i686 and x86_64 rpms have the patches. My currrent system has:
>>
>>> rpm -qa --qf='%{name}-%{version}-%{release}-%{arch}\n' | grep alsa
>> alsa-lib-devel-1.0.11-3.rc4.2.el4-x86_64
>> alsa-utils-1.0.11-2.rc4.1.el4-x86_64
>> alsa-lib-1.0.11-3.rc4.2.el4-i686
>> alsa-lib-devel-1.0.11-3.rc4.2.el4-i686
>> alsa-lib-1.0.11-3.rc4.2.el4-x86_64
>>
>> Wanting to make sure it wasn't some how getting confused by the x86_64
>> libraries, I just now tested only the i686 rpms in my chrooted el4 i386
>> environment and the same problem happened.
>
> OK, which problem exactly?  All tests failed?

cd /el4_i386_chroot
setarch i386 chroot .

# mplayer -vo xv -ao alsa -channels 6 An_AC3_movie.avi |& less

MPlayer dev-Fedora-GS-CVS-060328-14:45-3.4.5 (C) 2000-2006 MPlayer Team
CPU: Advanced Micro Devices Athlon 64 Newcastle,Winchester,San Diego,Venice;
Sempron Palermo (Family: 15, Stepping: 0)
CPUflags:  MMX: 1 MMX2: 1 3DNow: 1 3DNow2: 1 SSE: 1 SSE2: 1
Compiled for x86 CPU with extensions: MMX MMX2 3DNow 3DNowEx SSE SSE2

Playing An_AC3_movie.avi.
AVI file format detected.
VIDEO:  [XVID]  640x272  12bpp  25.000 fps  861.2 kbps (105.1 kbyte/s)
Clip info:
  Software: Nandub v1.0rc2
==========================================================================
Opening audio decoder: [liba52] AC3 decoding with liba52
Using SSE optimized IMDCT transform
AC3: 5.1 (3f+2r+lfe)  48000 Hz  448.0 kbit/s
Using MMX optimized resampler
AUDIO: 48000 Hz, 6 ch, s16le, 448.0 kbit/9.72% (ratio: 56000->576000)
Selected audio codec: [a52] afm: liba52 (AC3-liba52)
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffodivx] vfm: ffmpeg (FFmpeg MPEG-4)
==========================================================================
alsa-init: using device surround51
alsa: 48000 Hz/6 channels/12 bpf/196608 bytes buffer/Signed 16 bit Little
Endian
AO: [alsa] 48000Hz 6ch s16le (2 bytes per sample)
Starting playback...
VDec: vo config request - 640 x 272 (preferred colorspace: Planar YV12)
VDec: using Planar YV12 as output csp (no 0)
Movie-Aspect is 2.35:1 - prescaling to correct movie aspect.
VO: [xv] 640x272 => 640x272 Planar YV12
[ a bunch of position info trimmed ]
[ attempt to seek with right arrow ]
alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)
alsa-play: write error: Broken pipe
alsa-play: trying to reset soundcard
alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)
alsa-play: write error: Broken pipe
alsa-play: trying to reset soundcard
[ mplayer hangs w/ 100% cpu usage; same 3 messages repeat forever until mplayer is killed ]

Is there some way I can comment directly on bug 1938 in Mantis?

We seem to lose context in email (and Mantis doesn't seem to let me comment)?

Regards,
Rob

-- 
---------------------"Happiness is understanding."----------------------
Robert Hardy, B.Eng Computer Systems                  C.E.O. Webcon Inc.
rhardy <at> webcon <dot> ca    GPG Key available          (613) 276-7327


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Takashi Iwai]

At Thu, 30 Mar 2006 12:09:06 -0500 (EST),
Robert Hardy wrote:
> 
> On Thu, 30 Mar 2006, Takashi Iwai wrote:
> > At Thu, 30 Mar 2006 11:25:08 -0500 (EST),> Robert Hardy wrote:
> >> On Thu, 30 Mar 2006, Takashi Iwai wrote:
> >>> At Thu, 30 Mar 2006 11:45:09 +0200,
> >>>> At Wed, 29 Mar 2006 17:18:59 -0500 (EST),
> >>>> Robert Hardy wrote:
> >>>>> Basically I was trying to point out I am also affected by this bug:
> >>>>> https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1938
> >>>>>
> >>>>> This unfortunately made no difference here. I'm running 2.6.16-rh3 which is 2.6.16.1 with
> >>>>> custom patches including the recent git commit for [ALSA] version 1.0.11rc4 from here:
> >>>>> http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=1c2e02750b992703a8a18634e08b04353face243
> >>>>>
> >>>>> My sound card:
> >>>>> 02:07.0 Multimedia audio controller: Creative Labs SB0400 Audigy2 Value
> >>>>>          Subsystem: Creative Labs: Unknown device 1001
> >>>>>
> >>>>> This card once worked fine unfortunately I'm not sure with what
> >>>>> mplayer/alsa-lib/alsa-driver/kernel....
> >>>>>
> >>>>> I'm running a i686/32bit mplayer in a 64 bit environment.
> >>
> >>>> Strange.  It works fine on my test syetm now with 32bit mplayer on
> >>>> 64bit kernel.
> >>>
> >>> Maybe a dumb question: Did you properly install 32bit library?
> >>>
> >>> I often compile/install 64bit lib for debugging 32bit problems, and
> >>> throw precious hours aways...
> >>
> >> I rolled those patches into my alsa-lib rpm before I tested them. Both my
> >> i686 and x86_64 rpms have the patches. My currrent system has:
> >>
> >>> rpm -qa --qf='%{name}-%{version}-%{release}-%{arch}\n' | grep alsa
> >> alsa-lib-devel-1.0.11-3.rc4.2.el4-x86_64
> >> alsa-utils-1.0.11-2.rc4.1.el4-x86_64
> >> alsa-lib-1.0.11-3.rc4.2.el4-i686
> >> alsa-lib-devel-1.0.11-3.rc4.2.el4-i686
> >> alsa-lib-1.0.11-3.rc4.2.el4-x86_64
> >>
> >> Wanting to make sure it wasn't some how getting confused by the x86_64
> >> libraries, I just now tested only the i686 rpms in my chrooted el4 i386
> >> environment and the same problem happened.
> >
> > OK, which problem exactly?  All tests failed?
> 
> cd /el4_i386_chroot
> setarch i386 chroot .
> 
> # mplayer -vo xv -ao alsa -channels 6 An_AC3_movie.avi |& less
> 
> MPlayer dev-Fedora-GS-CVS-060328-14:45-3.4.5 (C) 2000-2006 MPlayer Team
> CPU: Advanced Micro Devices Athlon 64 Newcastle,Winchester,San Diego,Venice;
> Sempron Palermo (Family: 15, Stepping: 0)
> CPUflags:  MMX: 1 MMX2: 1 3DNow: 1 3DNow2: 1 SSE: 1 SSE2: 1
> Compiled for x86 CPU with extensions: MMX MMX2 3DNow 3DNowEx SSE SSE2
> 
> Playing An_AC3_movie.avi.
> AVI file format detected.
> VIDEO:  [XVID]  640x272  12bpp  25.000 fps  861.2 kbps (105.1 kbyte/s)
> Clip info:
>   Software: Nandub v1.0rc2
> ==========================================================================
> Opening audio decoder: [liba52] AC3 decoding with liba52
> Using SSE optimized IMDCT transform
> AC3: 5.1 (3f+2r+lfe)  48000 Hz  448.0 kbit/s
> Using MMX optimized resampler
> AUDIO: 48000 Hz, 6 ch, s16le, 448.0 kbit/9.72% (ratio: 56000->576000)
> Selected audio codec: [a52] afm: liba52 (AC3-liba52)
> ==========================================================================
> Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
> Selected video codec: [ffodivx] vfm: ffmpeg (FFmpeg MPEG-4)
> ==========================================================================
> alsa-init: using device surround51
> alsa: 48000 Hz/6 channels/12 bpf/196608 bytes buffer/Signed 16 bit Little
> Endian
> AO: [alsa] 48000Hz 6ch s16le (2 bytes per sample)
> Starting playback...
> VDec: vo config request - 640 x 272 (preferred colorspace: Planar YV12)
> VDec: using Planar YV12 as output csp (no 0)
> Movie-Aspect is 2.35:1 - prescaling to correct movie aspect.
> VO: [xv] 640x272 => 640x272 Planar YV12
> [ a bunch of position info trimmed ]
> [ attempt to seek with right arrow ]
> alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)
> alsa-play: write error: Broken pipe
> alsa-play: trying to reset soundcard
> alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)

That looks bad.  Could you set $LIBASOUND_DEBUG=2 and run it from gdb?
It will send abort() at that place, then you can see backtrace what is
the original caller.

does it happen if you play via aplay like

	% aplay -Dplug:surround51 some.wav

too?

> Is there some way I can comment directly on bug 1938 in Mantis?

I don't know.  I heard it should be fixed now.

> We seem to lose context in email (and Mantis doesn't seem to let me comment)?

Honestly, I hate bugtracking systems.  Bug reports and communcations
on ML motivates me much better :)


Takashi


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Takashi Iwai]

At Thu, 30 Mar 2006 18:53:23 -0500 (EST),
Robert Hardy wrote:
> 
> On Thu, 30 Mar 2006, Takashi Iwai wrote:
> > At Thu, 30 Mar 2006 12:09:06 -0500 (EST), Robert Hardy wrote:
> >> cd /el4_i386_chroot
> >> setarch i386 chroot .
> >>
> >> # mplayer -vo xv -ao alsa -channels 6 An_AC3_movie.avi |& less
> >>
> >> MPlayer dev-Fedora-GS-CVS-060328-14:45-3.4.5 (C) 2000-2006 MPlayer Team
> >> [...]
> >> [ attempt to seek with right arrow ]
> >> alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)
> >> alsa-play: write error: Broken pipe
> >> alsa-play: trying to reset soundcard
> >> alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)
> >
> > That looks bad.  Could you set $LIBASOUND_DEBUG=2 and run it from gdb?
> > It will send abort() at that place, then you can see backtrace what is
> > the original caller.
> >
> > does it happen if you play via aplay like
> > 	% aplay -Dplug:surround51 some.wav
> > too?
> 
> No it only happens when I seek 6 channel audio. I can play something with
> mplayer -ao alsa -channels 6  and it will play fine _UNTIL_ I seek. Then it
> blows up.
> 
> So something like this will immediately fail:
> mplayer -ao alsa -channels 6 -ss 00:01:00 An_AC3_movie.avi
> 
> With this stuff:
> alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)
> alsa-play: write error: Broken pipe
> alsa-play: trying to reset soundcard
> 
> If I switch to a 64 bit chrooted environment with a 64 bit mplayer the audio
> works properly. But then I can't use any of the 32 bit codecs.
> 
> >> Is there some way I can comment directly on bug 1938 in Mantis?
> >
> > I don't know.  I heard it should be fixed now.
> 
> The author of the bug seems to think it's still broken and seems to have
> beat me to posting a backtrace. It's been a busy day here. If you still want
> me to do this let me know.
> 
> > Honestly, I hate bugtracking systems.  Bug reports and communcations
> > on ML motivates me much better :)
> 
> k email it is 8o) I'd love to get this fixed...

I think I got it.  It looks like the old applptr that are not updated
by the prepare call due to pcm links.

Try the patch below.


Takashi

Index: alsa-lib/src/pcm/pcm_hw.c
===================================================================
RCS file: /home/iwai/cvs/alsa/alsa-lib/src/pcm/pcm_hw.c,v
retrieving revision 1.114
diff -u -r1.114 pcm_hw.c
--- alsa-lib/src/pcm/pcm_hw.c	29 Mar 2006 09:30:49 -0000	1.114
+++ alsa-lib/src/pcm/pcm_hw.c	31 Mar 2006 10:04:17 -0000
@@ -624,6 +624,9 @@
 		SYSMSG("SNDRV_PCM_IOCTL_REWIND failed");
 		return err;
 	}
+	err = sync_ptr(hw, SNDRV_PCM_SYNC_PTR_APPL);
+	if (err < 0)
+		return err;
 	return frames;
 }
 
@@ -637,6 +640,9 @@
 			SYSMSG("SNDRV_PCM_IOCTL_FORWARD failed");
 			return err;
 		}
+		err = sync_ptr(hw, SNDRV_PCM_SYNC_PTR_APPL);
+		if (err < 0)
+			return err;
 		return frames;
 	} else {
 		snd_pcm_sframes_t avail;
Index: alsa-lib/src/pcm/pcm_multi.c
===================================================================
RCS file: /home/iwai/cvs/alsa/alsa-lib/src/pcm/pcm_multi.c,v
retrieving revision 1.72
diff -u -r1.72 pcm_multi.c
--- alsa-lib/src/pcm/pcm_multi.c	27 Feb 2006 12:39:29 -0000	1.72
+++ alsa-lib/src/pcm/pcm_multi.c	31 Mar 2006 10:04:17 -0000
@@ -424,8 +424,9 @@
 	int err = 0;
 	unsigned int i;
 	for (i = 0; i < multi->slaves_count; ++i) {
-		if (multi->slaves[i].linked)
-			continue;
+		/* We call prepare to each slave even if it's linked.
+		 * This is to make sure to sync non-mmaped control/status.
+		 */
 		err = snd_pcm_prepare(multi->slaves[i].pcm);
 		if (err < 0)
 			return err;
@@ -439,8 +440,7 @@
 	int err = 0;
 	unsigned int i;
 	for (i = 0; i < multi->slaves_count; ++i) {
-		if (multi->slaves[i].linked)
-			continue;
+		/* Reset each slave, as well as in prepare */
 		err = snd_pcm_reset(multi->slaves[i].pcm);
 		if (err < 0)
 			return err;


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642

Re: Re: [ALSA - driver 0001938]: EPIPE on Audigy with surround sound when seeking in 6 channel sound file

[Robert Hardy]

On Fri, 31 Mar 2006, Takashi Iwai wrote:
> At Thu, 30 Mar 2006 18:53:23 -0500 (EST), Robert Hardy wrote:
>> No it only happens when I seek 6 channel audio. I can play something with
>> mplayer -ao alsa -channels 6  and it will play fine _UNTIL_ I seek. Then it
>> blows up.
>>
>> So something like this will immediately fail:
>> mplayer -ao alsa -channels 6 -ss 00:01:00 An_AC3_movie.avi
>>
>> With this stuff:
>> alsa-lib: pcm.c:6300:(snd_pcm_mmap_commit) commit offset (0) doesn't match with appl_ptr (291840) % buf_size (16384)
>> alsa-play: write error: Broken pipe
>> alsa-play: trying to reset soundcard
>>
>> If I switch to a 64 bit chrooted environment with a 64 bit mplayer the audio
>> works properly. But then I can't use any of the 32 bit codecs.
>>
>>>> Is there some way I can comment directly on bug 1938 in Mantis?
>>>
>>> I don't know.  I heard it should be fixed now.
>>
>> The author of the bug seems to think it's still broken and seems to have
>> beat me to posting a backtrace. It's been a busy day here. If you still want
>> me to do this let me know.
>
> I think I got it.  It looks like the old applptr that are not updated
> by the prepare call due to pcm links.
>
> Try the patch below.

*smiles happily*

Your two most recently patches creating the files below fixed my seeking
problems and likely closes bug 1938 on Mantis too.

Thanks a lot for your help!

RCS file: /home/iwai/cvs/alsa/alsa-lib/src/pcm/pcm_hw.c,v
retrieving revision 1.114

RCS file: /home/iwai/cvs/alsa/alsa-lib/src/pcm/pcm_multi.c,v
retrieving revision 1.72


Regards,
Rob

-- 
---------------------"Happiness is understanding."----------------------
Robert Hardy, B.Eng Computer Systems                  C.E.O. Webcon Inc.
rhardy <at> webcon <dot> ca    GPG Key available          (613) 276-7327


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642