- Alsa-devel - mailman.alsa-project.org

[alsa-devel] Issues with alsa-lib-1.0.13 built using arm gcc 4.1.1
by sandeep lahane 18 Mar '08

18 Mar '08

Hello, I have written an external io type plugin for an arm based board. While debugging the plugin, I encountered unexpected errors which I doubt are because of gcc optimizations, and after looking up on web it seems that arm-gcc 4.1.1 which we are using has some known bugs due to which alsa-library functions behave unexpectedly. E.g. snd_mask_refine() reportedly behaves differently when it is compiled using different optimization levels, I am facing the same issue where snd_mask_refine() returns unexpected (-EINVAL) and after I modify the code as in src/pcm/mask_inline.h MASK_INLINE void snd_mask_copy(snd_mask_t *mask, const snd_mask_t *v) { - *mask = *v; + memcpy(mask->bits, v->bits, MASK_SIZE * 4); } snd_mask_refine() returns 0 which is valid but I get segfaults just after returning from snd_pcm_ioplug_hw_refine(). Is this a known bug and if yes/no then is there any list of known bugs/quirks when using alsa-lib with arm gcc 4.1.1? I am attaching a log below, pardon for non-alsa flavour of debug messages in the log. # aplay -D myplug etc/startup.wav AlsaVersion=1.0.13, Opened for Playback ----- INFO (ALSA) : Successfully Registered the IO plugin ALSA-OPENMAX PLUGIN Playing WAVE 'etc/startup.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Ste reo ####aplay.c set_params calling snd_pcm_hw_params_any ###### snd_pcm_hw_params_any 1 ##### _snd_pcm_hw_params_any 0 ##### _snd_pcm_hw_params_any 2 ##### _snd_pcm_hw_params_any 8 ##### _snd_pcm_hw_params_any 19 _snd_pcm_hw_param_any in hw_is_mask _snd_pcm_hw_param_any in hw_is_mask _snd_pcm_hw_param_any in hw_is_mask _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval _snd_pcm_hw_param_any in hw_is_interval ##### snd_pcm_hw_params_any 1 #####snd_pcm_hw_refine #####snd_pcm_hw_refine calling hw_refine #### snd_pcm_ioplug_hw_refine 1 snd_ext_parm_mask_refine 1 snd_mask_refine 1 snd_ext_parm_mask_refine return val is = 1 snd_ext_parm_mask_refine 1 snd_mask_refine 1 snd_ext_parm_mask_refine return val is = 1 snd_pcm_ioplug_hw_refine 2 snd_ext_parm_interval_refine 1 snd_ext_parm_interval_refine 1 snd_pcm_ioplug_hw_refine 3 snd_mask_refine 1 snd_mask_refine 1 snd_mask_refine 1 snd_pcm_ioplug_hw_refine 4 snd_ext_parm_interval_refine 1 snd_pcm_ioplug_hw_refine 5 snd_ext_parm_interval_refine 1 snd_ext_parm_interval_refine 1 snd_ext_parm_interval_refine 1 snd_ext_parm_interval_refine 1 snd_pcm_ioplug_hw_refine 5 change is = 1 ##### return val of hw_refine = 1 #####snd_pcm_hw_refine #####snd_pcm_hw_refine calling hw_refine #### snd_pcm_ioplug_hw_refine 1 snd_ext_parm_mask_refine 1 snd_mask_refine 1 snd_ext_parm_mask_refine return val is = 0 snd_ext_parm_mask_refine 1 snd_mask_refine 1 snd_ext_parm_mask_refine return val is = 0 snd_pcm_ioplug_hw_refine 2 snd_ext_parm_interval_refine 1 snd_ext_parm_interval_refine 1 snd_pcm_ioplug_hw_refine 3 snd_mask_refine 1 snd_pcm_ioplug_hw_refine 4 snd_ext_parm_interval_refine 1 snd_pcm_ioplug_hw_refine 5 snd_ext_parm_interval_refine 1 snd_ext_parm_interval_refine 1 snd_pcm_ioplug_hw_refine 5 change is = 0 ##### return val of hw_refine = 0 Dec 31 16:27:59 kernel: aplay: unhandled page fault (11) at 0x00000022, code 0x0 17 Dec 31 16:27:59 kernel: pgd = c30d8000 Dec 31 16:27:59 kernel: Dec 31 16:27:59 kernel: Dec 31 16:27:59 kernel: Pid: 206, comm: aplay Dec 31 16:27:59 kernel: CPU: 0 Not tainted (2.6.23.14-BROADCOM #1) Dec 31 16:27:59 kernel: PC is at 0x40170770 Dec 31 16:27:59 kernel: LR is at 0x40168210 Dec 31 16:27:59 kernel: pc : [<40170770>] lr : [<40168210>] psr: a0000010 Dec 31 16:27:59 kernel: sp : bee6f698 ip : 401855f8 fp : bee6fb64 Dec 31 16:27:59 kernel: r10: 401852b0 r9 : 00000006 r8 : bee6f894 Dec 31 16:27:59 kernel: r7 : 0001cc64 r6 : 00000022 r5 : 000130fc r4 : bee6f7 04 Dec 31 16:27:59 kernel: r3 : ffffffff r2 : 00000022 r1 : fffffffe r0 : 000000 22 Dec 31 16:27:59 kernel: Flags: NzCv IRQs on FIQs on Mode USER_32 ISA ARM Se gment user Dec 31 16:27:59 kernel: Control: 00e5387f Table: c30d8008 DAC: 00000015 Dec 31 16:27:59 kernel: (__do_user_fault+0x5c/0xa4) Dec 31 16:27:59 kernel: Dec 31 16:27:59 kernel: (do_page_fault+0x200/0x238) Dec 31 16:27:59 kernel: Dec 31 16:27:59 kernel: (do_DataAbort+0x3c/0xa0) Dec 31 16:27:59 kernel: (ret_from_exception+0x0/0x10) Dec 31 16:27:59 kernel: Exception stack(0xc31a1fb0 to 0xc31a1ff8) Dec 31 16:27:59 kernel: Dec 31 16:27:59 kernel: Dec 31 16:27:59 kernel: Dec 31 16:27:59 kernel: Segmentation fault Regards, Sandeep.

1 1

[alsa-devel] If you have a copy callback, do you need to have a silence?
by Barclay, Bruce 18 Mar '08

18 Mar '08

I was finally able to get my driver working. I had to use the copy callback since my memory is external to Alsa. I have noticed that I commented out the silence callback and the ack. All seems to work and just for grins, I put them in with some printk's and they were never invoked. I don't fully understand these callbacks. When are they called and am I just lucky? Bruce --------------------- Bruce Barclay Intel Corporation ( Phone : 480-554-0553 ( Cell : 480-274-7493 * Email:bruce.barclay@intel.com * MS : CH7-308 "Real Gold Fears No Fire" - Li Quan

1 0

[alsa-devel] arecord | aplay expected delay
by van der Linden 18 Mar '08

18 Mar '08

Hi, I'm running arecord and aplay under uCLinux on a Blackfin DSP. In this way I loopback my audio. Does anyone know what delay I can expect between audio input and audio output? I measure at my test board 50ms. I want to have about 1ms! E-J

2 2

[alsa-devel] [PATCH] 1/2 alsa-python: API coverage documentation tool
by Aldrin Martoq 18 Mar '08

18 Mar '08

HG Log: ------- Added python tool of alsa-lib C API coverage for alsa-python binding. Description: ------------ The following patch is a python tool that shows how much of the asoundlib API is being covered by the alsa-python binding. It works by parsing the C source code and comparing with the official Doxygen documentation from alsa website. It will help the following users: - alsa-python developers: know how is mapped the original C API, so which python function/variable may be used. - python binding developers: know how much of the C API is covered. It could help for mistakes (free() instead of using snd_ctl_card_info_free), know what is missing, and statistic interest ;) Sample output: -------------- Functions --------- 1 const char* snd_asoundlib_version ( void ) Returns the ALSA sound library version in ASCII format. => Used in : alsacard.c: asoundlib_version Defines ------- N/A #define SND_CONTROL_DLSYM_VERSION _dlsym_control_001 dlsym version for interface entry callback **** NOT AVAILABLE/USED SND_CONTROL_DLSYM_VERSION **** 2 #define SND_CTL_ASYNC Async notification (flag for open mode) => As constant: alsacontrol.c: open_mode["ASYNC"] => As constant: alsahcontrol.c: open_mode["ASYNC"] 1 #define snd_ctl_card_info_alloca ( ptr ) allocate an invalid snd_ctl_card_info_t using standard alloca => Used in : alsacontrol.c: pyalsacontrol_cardinfo Stats: ------ After every section, the program shows some stats about usage. This is the current usage of each doxygen module: Global defines and functions 16 missing ( 94%) 0 excluded ( 0%) of 17 total ( 6% covered). Error handling 9 missing ( 90%) 0 excluded ( 0%) of 10 total ( 10% covered). Control Interface 137 missing ( 53%) 0 excluded ( 0%) of 257 total ( 47% covered). High level Control Interface 21 missing ( 46%) 0 excluded ( 0%) of 46 total ( 54% covered). Setup Control Interface 4 missing (100%) 0 excluded ( 0%) of 4 total ( 0% covered). MIDI Sequencer 25 missing ( 61%) 7 excluded ( 17%) of 41 total ( 39% covered). Sequencer Client Interface 15 missing ( 35%) 14 excluded ( 33%) of 43 total ( 65% covered). Sequencer Port Interface 48 missing ( 74%) 6 excluded ( 9%) of 65 total ( 26% covered). Sequencer Port Subscription 3 missing ( 7%) 15 excluded ( 33%) of 46 total ( 93% covered). Sequencer Queue Interface 58 missing ( 84%) 0 excluded ( 0%) of 69 total ( 16% covered). Sequencer Event API 42 missing ( 93%) 0 excluded ( 0%) of 45 total ( 7% covered). Sequencer Miscellaneous 0 missing ( N/A) 0 excluded ( N/A) of 0 total ( N/A covered). Sequencer Event Type Checks 7 missing ( 23%) 0 excluded ( 0%) of 30 total ( 77% covered). Sequencer Event Definitions 35 missing ( 40%) 0 excluded ( 0%) of 87 total ( 60% covered). Sequencer Middle Level Interfa 38 missing ( 78%) 0 excluded ( 0%) of 49 total ( 22% covered). Sequencer event <-> MIDI byte 11 missing (100%) 0 excluded ( 0%) of 11 total ( 0% covered). Mixer Interface 44 missing ( 75%) 0 excluded ( 0%) of 59 total ( 25% covered). Simple Mixer Interface 14 missing ( 18%) 0 excluded ( 0%) of 78 total ( 82% covered). Cheers! -- Aldrin Martoq <amartoq(a)dcc.uchile.cl>

2 1

[alsa-devel] Query regarding SND_PCM_PLUGIN_DEFINE_FUNC()
by Pharaoh . 18 Mar '08

18 Mar '08

Hi, What does the mode field in SND_PCM_PLUGIN_DEFINE_FUNC() used for? is it for determining blocking mode or non-blocking mode of opening? -pharaoh

2 1

[alsa-devel] [PATCH] Removed deprecated sound/driver.h from Freescale MPC8610 drivers
by Timur Tabi 18 Mar '08

18 Mar '08

With commit 9004acc70e8c49c50c4c7b652f906f1e0ed5709d, include/sound/driver.h is deprecated. This patch removes the #include from fsl_ssi.c and fsl_dma.c. Signed-off-by: Timur Tabi <timur(a)freescale.com> --- sound/soc/fsl/fsl_dma.c | 1 - sound/soc/fsl/fsl_ssi.c | 1 - 2 files changed, 0 insertions(+), 2 deletions(-) diff --git a/sound/soc/fsl/fsl_dma.c b/sound/soc/fsl/fsl_dma.c index 652514f..78de716 100644 --- a/sound/soc/fsl/fsl_dma.c +++ b/sound/soc/fsl/fsl_dma.c @@ -20,7 +20,6 @@ #include <linux/interrupt.h> #include <linux/delay.h> -#include <sound/driver.h> #include <sound/core.h> #include <sound/pcm.h> #include <sound/pcm_params.h> diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 145ad13..b2a11b0 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -15,7 +15,6 @@ #include <linux/device.h> #include <linux/delay.h> -#include <sound/driver.h> #include <sound/core.h> #include <sound/pcm.h> #include <sound/pcm_params.h> -- 1.5.4

2 1

[alsa-devel] [PATCH] usbmidi.c: Call usb_autopm functions from rawmidi open/close callbacks
by Karsten Wiese 18 Mar '08

18 Mar '08

Lets "usb auto suspend" work for devices that set their usb_driver::supports_autosuspend to 1. Has no effect for other devices. Signed-off-by: Karsten Wiese <fzu(a)wemgehoertderstaat.de> --- sound/usb/usbmidi.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/sound/usb/usbmidi.c b/sound/usb/usbmidi.c index 6c616c7..842486f 100644 --- a/sound/usb/usbmidi.c +++ b/sound/usb/usbmidi.c @@ -860,11 +860,14 @@ static int snd_usbmidi_output_open(struct snd_rawmidi_substream *substream) } substream->runtime->private_data = port; port->state = STATE_UNKNOWN; + usb_autopm_get_interface(umidi->iface); return 0; } static int snd_usbmidi_output_close(struct snd_rawmidi_substream *substream) { + struct snd_usb_midi* umidi = substream->rmidi->private_data; + usb_autopm_put_interface(umidi->iface); return 0; } -- 1.5.3.3 -------------------------------------------------------

2 1

[alsa-devel] /usr/share/alsa/alsa.conf :and file extention
by Alexander Indenbaum 18 Mar '08

18 Mar '08

Hello, I'm playing with embedded system without any hardware sound device. I'd like to grab sound from ALSA application, for instance Firefox displaying YouTube clip inside flash plugin and to store output PCM stream into file on local disk. I got an advice to use File extension. So I defined following /usr/share/alsa/alsa.conf pcm.default{ type file file /tmp/pcm.out } But when alsa lib still complains about "Unknown PCM plughw:0,0" "Cannot open shared library /usr/lib/alsa-lib/libasound_module_pcm_file.so" I have a nasty feeling that syntax of my /usr/share/alsa/alsa.conf is broken beyond recognition. I wonder what is the correct way to define my situation: * No real audio devices * All sound is output goes to file as PCM stream Do you have a clue? Alexander Indenbaum

5 17

[alsa-devel] ~ALSA Audio coordinator/manager
by Harsha priya gupta 18 Mar '08

18 Mar '08

Hi all, I am looking to see if anyone is working on an Audio manager kind of solution as a part of ALSA. 'By Audio manager, I mean a coordinator of all audio input and output streams. A coordinator which is application and devices aware . Example functionality of an audio manager would to direct different steams to different devices like (voice output to headset and music output to speakers). Or to pause an media player when a voice recording starts.' If anyone is doing any work on these lines, please let me know about the work and its integration into ALSA as well. Thanks in Advance, Harsha.

2 1

[alsa-devel] basic questions
by Harsha priya gupta 18 Mar '08

18 Mar '08

Hi, I am ramping up on ALSA architecture and framework. I have a couple of basic questions. I apologize if these questions are very basic to be raised in this mailing list. If anyone can help me find answers, it would be great. Thanks in advance. 1. My assumption is that ALSA handles only PCM streams today. Does it handle encoded streams from application to sound card (Which can do hardware decoding)? 2. What are the hwdep files in ALSA used for? Where do the concerned IOCTLS land to? Thanks and Regards, Harsha.

3 4