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

[alsa-devel] Endiannes and alsa (in an embedded device)
by Ciaccia 01 May '07

01 May '07

Dear all, I have a question that really puzzles me, I wonder if someone can explain to me what I'm doing wrong... :-| I have programmed an ALSA driver for an embedded device, everything works fine and all the application use my driver without any glitches. The snd_pcm_hardware for this device is defined as following: static const struct snd_pcm_hardware ep93xx_pcm_hardware = { .info= SNDRV_PCM_INFO_MMAP | SNDRV_PCM_INFO_MMAP_VALID | SNDRV_PCM_INFO_INTERLEAVED | SNDRV_PCM_INFO_BLOCK_TRANSFER, .rates = SNDRV_PCM_RATE_8000_48000, .rate_min = 8000, .rate_max = 48000, .channels_min = 2, .channels_max = 2, .formats = SNDRV_PCM_FMTBIT_S24_LE, .buffer_bytes_max = 131072, .period_bytes_min = 512, .period_bytes_max = 32768, .periods_min = 1, .periods_max = MAX_PERIODS, .fifo_size = 32, }; Then I tried to program a test application that uses the hardware device directly (without using alsa-lib to convert the format and so on). I open the device as following: if ((err = snd_pcm_open (&playback_handle, "hw:0,0", SND_PCM_STREAM_PLAYBACK, 0)) < 0) { Everything works fine until I try to write data to the stream. I tried both SND_PCM_ACCESS_RW_INTERLEAVED and SND_PCM_ACCESS_MMAP_INTERLEAVED, but I always have the following problem. When I want to write data to the stream, I have to "reverse" the byte order, otherwise I hear noise on the speaker. The system I am working on is little-endian (it has cirrus ep93xx arm core), so I assume I could write data directly to the buffer. Here is what I have to do: // fill the buffer for (i = 0; i < 512; i++) { int sample = (int)(20000.0f * sin((float)i * 2.0f * M_PI / 512.0f)); int reversed; char *src = (char *)(&sample); char *dst = (char *)(&reversed); dst[0] = src[3]; dst[1] = src[2]; dst[2] = src[1]; dst[3] = src[0]; *(buffer++) = reversed; *(buffer++) = reversed; } It this normal? I thought that since the system is little-endian and the format of the driver is S24_LE, I could copy the data directly to the stream (either using snd_pcm_writei or snd_pcm_mmap_begin). I think the driver is ok, since all the applications run with no problem at all. Can someone explain to me how/why the applications such as aplay/arecord work ok? Every help is greatly appreciated Thanks in advance Andrea __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

1 0

[alsa-devel] Implementing Priority Based Device ALSA device interface. ALSA drivers
by Nobin Mathew 01 May '07

01 May '07

Hi Friends, I am a new to ALSA, I am developing a ALSA ASoC driver. Now driver can support both audio playback and capture. Now the problem is handling of application with different priority. We have three applications say app1, app2 and app3; they are in increasing order of priority. One scenario of app1 and app2 is below If app2 is accessing the device while app1 playback is going on, then app1 should stop (clearly do pause) playback and allow app2 to playback, and app1 should continue playback when app2 exit. The requirement is that there should be no changes to apps, except the devices each app will access. There will be no daemons in user space which will handle priority. For that i implemented three interfaces like pcmC0p0, pcmC0p1 and pcmC0p2 there are in increaing order of priority. All these interfaces will access the same audio hardware. Only thing is each will have separate ALSA kernel buffer. In this implementation app1 will always access pcmC0p0, and app2 pcmC0p1 like that. What i am doing is, in driver i will pause the pcmC0p0 substream using snd_pcm_pause() when pcmC0p1 is accessed. Same for pcmc0p2. This call will change the state of pcmc0p0 substream to SNDRV_PCM_STATE_PAUSED, and this call should notify the user app also. When app2 playback is over i will change the state of pcmC0p0 substream to SNDRV_PCM_STATE_RUNNING by another call to snd_pcm_pause(). This implementation requires some minor modifications in the ALSA lib and ALSA app (handling of PAUSE from kernel). But what i found is kernel is not notifying the user using snd_timer_notify() (user call back) function because i am using the substream timer snd_pcm_timer which is there in pcm_timer.c. For this timer SNDRV_TIMER_HW_SLAVE flag is set so it will not notify the user. What is the use of these timers in ALSA? Am i going wrong? Is my design correct? Can u suggest some alternate design? Sorry for my poor English. Thanks Nobin Mathew

1 0

[alsa-devel] Set fragment size for USB sound card
by Jason Heeris 01 May '07

01 May '07

System: NEC Versa M540 laptop Dist: Debian Etch/Lenny Kernel: 2.6.18 Alsa: using drivers in kernel, although no difference when compiled from alsa-source package. Card in question: C-Media USB sound card (more info below) Hi there, I realise that this may not be strictly a developer list question, but someone on the alsa-user list suggested I try here. Any help would be appreciated. The question is a continuation of: http://sourceforge.net/mailarchive/forum.php?thread_name=1176015858.4497.9.… and http://www.linuxquestions.org/questions/showthread.php?t=540205 Essentially, I have a C-Media USB audio adaptor with which to use my headphones, and it only works with a fragment size of 2048 or less (discovered using 'alsaplayer -f 2048'). I would like to be able to plug it in, have ALSA recognise it and use this fragment size for it. Without specifying this fragment size, I get: -- $ alsaplayer -i text -o alsa -d hw:2 /usr/share/sounds/generic.wav Unable to install hw params: ACCESS: RW_INTERLEAVED FORMAT: S16_LE SUBFORMAT: STD SAMPLE_BITS: 16 FRAME_BITS: 32 CHANNELS: 2 RATE: 44100 PERIOD_TIME: (23219 23220) PERIOD_SIZE: 1024 PERIOD_BYTES: 4096 PERIODS: 8 BUFFER_TIME: (185759 185760) BUFFER_SIZE: 8192 BUFFER_BYTES: 32768 TICK_TIME: 1000 failed to configure output device...trying OSS ...done playing -- ...and the sound comes through the laptop speakers. I tried defining a slave PCM: -- pcm.usbtest { type dmix ipc_key 192837465 ipc_key_add_uid true slave { pcm "hw:2,0" # period_size 2048 buffer_size 2048 } } -- But trying 'alsaplayer -i text -o alsa -d usbtest /usr/share/sounds/generic.wav' gave exactly the same error. >From my previous post: -- ...I also tried forcing the index of the card by adding the following to /etc/modprobe.d/alsa-base, but even this only works half of the time: -- # Prevent abnormal drivers from grabbing index 0 options snd-bt87x index=-2 options snd-cx88_alsa index=-2 options snd-atiixp-modem index=-2 # options snd-intel8x0m index=-2 options snd-via82xx-modem index=-2 options snd-intel8x0 index=0 options snd-intel8x0m index=1 options snd-usb-audio index=2 vid=0d8c pid=000c options snd-usb-audio index=3 -- The contents of '/proc/asound/cards' are: -- 0 [I82801DBICH4 ]: ICH4 - Intel 82801DB-ICH4 Intel 82801DB-ICH4 with ALC655 at 0xf0000400, irq 5 1 [Modem ]: ICH-MODEM - Intel 82801DB-ICH4 Modem Intel 82801DB-ICH4 Modem at 0xe200, irq 5 2 [default ]: USB-Audio - C-Media USB Headphone Set C-Media USB Headphone Set at usb-0000:00:1d.1-1, full speed -- ...although sometimes the ICH-MODEM is missing and the C-Media card is at #1. Card information from 'lsusb': Bus 002 Device 003: ID 0d8c:000c C-Media Electronics, Inc. Audio Adapter Usually I boot with the card plugged in, but if I plug it in after booting, the new modules loaded are: snd_hwdep, snd_rawmidi, snd_usb_audio and snd_usb_lib. -- Anyway, as I said, any help would be appreciated. Cheers, Jason

1 0

[alsa-devel] Possible bug in snd_mixer_selem_set_playback_switch_all
by Tim Burrell 30 Apr '07

30 Apr '07

Hi there, I've encountered a situation where if I very quickly, and repeatedly make calls to snd_mixer_selem_set_playback_switch_all I get some strange undefined behavior. The situation is I'm looping over all mixer elements (on all sound cards) to do a system wide mute / unmute. If I make numerous calls to this function it ends up that some mixers only toggle switches on specific channels (right or left), and sometimes the volume on certain (random) channels gets set to some also random (usually 0, but can be anywhere between 0 and max) level. I've been able to remedy the situation by inserting a 10ms delay in between calls to snd_mixer_selem_set_playback_switch_all. Clearly though this is undesirable. I don't have more than one sound card to test on so I can't be sure if it's a problem with all drivers or just mine (hda-intel). I can certainly provide more information if requested, just let me know what I can do. Or if I'm doing something wrong I'd love to be made aware of something I can do to fix this behavior. Thanks, Tim.

1 0

[alsa-devel] List of all Apple Mac's hda-intel pin configurations
by Ivan N. Zlatev 30 Apr '07

30 Apr '07

I have attached all windows drivers' .ini files (taken from latest windows drivers cd of Boot Camp), which include all pin configuration as well as probably some more useful information (GPIO and stuff). I hope this will help you to make a nice update. [Subsystem ID - configuration] 106B0800 - Board1.ini 106B0600 - Board2.ini 106B0700 - Board2.ini 106B0E00 - Board3.ini 106B0F00 - Board3.ini 106B1600 - Board3.ini 106B1700 - Board3.ini 106B0200 - Board3.ini 106B1E00 - Board3.ini 106B0400 - Board4.ini 106B1A00 - Board4.ini 106B0A00 - Board5.ini 106B2200 - Board5.ini 00000100 - Board4.ini Regards. -- Ivan N. Zlatev Web: http://www.i-nZ.net "It's all some kind of whacked out conspiracy."

1 0

Re: [alsa-devel] [Alsa-user] ALSA Newbie - Getting Ensoniq SoundscapeVivo to work? [Solved]
by Cody Jung 30 Apr '07

30 Apr '07

On 4/29/07, Rene Herman <rene.herman(a)gmail.com> wrote: > On 04/29/2007 09:31 AM, Rene Herman wrote: > > > I've attached (to try and avoid linewrap) the lines that you will need in > > /etc/modprobe.conf. > > No I haven't. > > Rene. > > alias snd-card-0 snd-cs4231 > alias sound-slot-0 snd-cs4231 > > options snd-cs4231 index=0 port=0x534 irq=5 dma1=1 dma2=0 > > install snd-cs4231 echo activate >/sys/bus/pnp/devices/01\:01.00/resources &&\ > sleep 1 &&\ > /sbin/modprobe --ignore-install snd-cs4231 > > remove snd-cs4231 /sbin/modprobe -r --ignore-remove snd-cs4231 &&\ > echo disable >/sys/bus/pnp/devices/01\:01.00/resources > > Wahoo! Thank you, thank you! I ended up not going the /etc/modules way, partially because it wasn't working and partially because I didn't know what I was doing. Instead I added the following 4 lines to a file and included it in my init scripts. echo activate > /sys/bus/pnp/devices/01\:01.00/resources sleep 1 modprobe snd modprobe snd-cs4231 port=0x534 irq=5 dma1=1 dma2=0 And it works great! Thanks again for putting up with my incompetence. ~Cody Jung

2 1

Re: [alsa-devel] Boot failure with 2.6.21 with snd-maestro3
by Michal Piotrowski 29 Apr '07

29 Apr '07

Hi, On 29/04/07, Giridhar Pemmasani <pgiri(a)yahoo.com> wrote: > When snd-maestro3 (ESS Maestro 3 driver) is enabled, kernel 2.6.21 fails to > boot - it seems to stop at various points depending on what other modules are > enabled. Disabling snd-maestro3 fixes all issues. Moreover, with everything > else being same, using maestro3.c from 2.6.20 kernel and using it in 2.6.21 > kernel also works - i.e., system boots fine. I will try to isolate the issue > and follow up. Please revert this patches http://www.stardust.webpages.pl/files/patches/alsa_revert/ and test kernel after each revert. Takashi, I'll add this to known regressions for -stable team. > > Giri Regards, Michal -- Michal K. K. Piotrowski Kernel Monkeys (http://kernel.wikidot.com/start)

1 0

[alsa-devel] Can't record sound on Aser Aspire 5100. ATI SB450, RelTek ALC 883
by xinyun liu 29 Apr '07

29 Apr '07

hi, I have an Acer Aspire 5100. I can play sound but with the build in mic, I can't record sound. The driver is the latest HG code. http://pastebin.ca/464406 [root@localhost ~]# lspci -nnv -d1002:437b 00:14.2 0403: 1002:437b (rev 01) Subsystem: 1025:009f Flags: bus master, slow devsel, latency 64, IRQ 19 Memory at c0000000 (64-bit, non-prefetchable) [size=16K] Capabilities: [50] Power Management version 2 Capabilities: [60] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable- [root@localhost ~]# dmesg | fgrep ALSA > /tmp/1 [root@localhost ~]# vi /tmp/1 ALSA ~/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/hda_intel.c:698: codec_mask = 0xb ALSA ~/alsa-driver/pci/hda/hda_codec.c:1820: hda_codec: model 'acer' is selected for config 1025:0 (Acer laptop) ALSA ~/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/hda_intel.c:698: codec_mask = 0xb ALSA ~/alsa-driver/pci/hda/hda_codec.c:1820: hda_codec: model 'acer' is selected for config 1025:0 (Acer laptop) ALSA ~/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/hda_intel.c:1188: azx_pcm_prepare: bufsize=0x10000, fragsize=0x1000, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x4, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x5, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x2, stream=0x0, channel=0, format=0x0 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x4, stream=0x0, channel=0, format=0x0 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x3, stream=0x0, channel=0, format=0x0 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x5, stream=0x0, channel=0, format=0x0 ALSA ~/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/hda_intel.c:1188: azx_pcm_prepare: bufsize=0x10000, fragsize=0x1000, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x8, stream=0x1, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x8, stream=0x0, channel=0, format=0x0 ALSA ~/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/hda_intel.c:1188: azx_pcm_prepare: bufsize=0x10000, fragsize=0x1000, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x4, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x5, stream=0x5, channel=0, format=0x11 ALSA ~/alsa-driver/pci/hda/hda_codec.c:630: hda_codec_setup_stream: NID=0x2, stream=0x0, channel=0, format=0x0

1 1

Re: [alsa-devel] [Alsa-user] ALSA Newbie - Getting Ensoniq SoundscapeVivo to work?
by Rene Herman 29 Apr '07

29 Apr '07

On 04/29/2007 04:18 AM, Cody Jung wrote: > Er...that appears to be a .diff file. Do I have to do something with it > for it to be workable, or...? Yes. You'd have to apply the patch (as it's called) to a recent kernel source tree, reconfigure the kernel adding this new VIVO driver to your config and then recompile and reinstall that kernel. Oh well, you posting to alsa-user and not alsa-devel should've been a hint. Rest assured the driver would've worked nicely :-/ There is another thing you can do. What was posted was just a wrapper around cs4231-lib in much the same way that the existing snd-cs4231 module is and that module you _will_ already have installed. That one doesn't do anything automatic though, so you will have to know how to manually enable your card first. First, open a root shell and look through the /sys/bus/pnp/devices directory for your VIVO. Under the "xx:yy.zz" sub-directories there you will find "id" files, and one of them will say "ENS1011". Example: # cat /sys/bus/pnp/devices/01\:01.00/id ENS1011 After having found it in there, activate the card: # echo activate >/sys/bus/pnp/devices/01\:01.00/resources and look at what resources were assigned to it by the PnP layer: # cat /sys/bus/pnp/devices/01\:01.00/resources state = active io 0x330-0x33f io 0x530-0x537 io 0x2100-0x213f irq 5 irq 9 dma 1 dma 0 You can now load the snd-cs4231 module specifying those card resources manually: # modprobe snd-cs4231 port=0x534 irq=5 dma1=1 dma2=0 Yes, 0x534. The second io resource + 4. This should've loaded without problem. Start up alsamixer: # alsamixer and up and unmute the PCM volume. With -> and <- you can walk through the sliders; ^ ups the one you're on, and "m" toggles the mute (the "MM" at the bottom should turn to a highlighted "OO"). PCM is the first slider. 90 seems to be an okay value. Quit alsamixer again with Alt-Q and now you should be able to hear things. If you have a plain "wav" file around somewhere, that makes for the quickest test: # aplay foo.wav No, I'm not going to tell you to plug in your speakers. If anything in the above doesn't work as expected, your system is not as expected and there's isn't much I can help with. If everything does work, you can automate it for the next time. I've attached (to try and avoid linewrap) the lines that you will need in /etc/modprobe.conf. Ubuntu has no doubt found a way to make that dark and muddy again so you might need to add it to an /etc/modprobe/whatever file instead. Please ask a fellow ubuntu victim for advice about translating this generic linux advice to your install. Once you do have that all setup, sound should "Just Work". You probably need to adjust the volume as specified above once only after which ubuntu will save it for the next boot, but if you don't hear anything ater rebooting, please check that first. Rene.

1 1

Re: [alsa-devel] [Alsa-user] ALSA Newbie - Getting Ensoniq SoundscapeVivo to work?
by Rene Herman 29 Apr '07

29 Apr '07

On 04/28/2007 07:06 AM, Cody Jung wrote: > As it suggests in the subject, I'm both a Linux-newbie and an > ALSA-newbie. I have an old clunker of a machine running Ubuntu Feisty > that I built out of parts left on curbs during spring cleanup, and that's > how I came across my sound card. It's an Ensoniq SoundscapeVivo (Not a > Vivo90, AFAIK). As far as I'm aware, there are no other VIVO's than the VIVO90. > I've gone through install and troubleshooting guides galore, but I can't > seem to get ALSA to recognize my card, and I'm confused and rather tired > of it all. > > Do you think you can help, and if so, what information do you need from > me? I probably can; I have both a Soundscape VIVO90 and a much older Soundscape S-2000. My VIVO90 is an ISA-PnP card and has a Ensoniq MARK5-RO, an OTTOR2C and a Analog Devices AD1845 codec: http://en.wikipedia.org/wiki/Ensoniq_Soundscape_VIVO90 I tested that when I acquired it a while ago and if I remember correctly, it just worked... mmm. come to think of it, it might be the case that I needed to add its PnP ID to the driver. Not sure anymore. Could you confirm whether or not your Soundscape is an ISA-PnP card and if so (that's also a way to find out if you don't know if it is) run "pnpdump" with it installed and look for a PnP-ID for the card in it? Something of the form "ENSxxxx". If your card is one of the older non-PnP soundscapes it's going to be a bit harder; they also need firmware and all that and I haven't tested that card myself yet. (please use the @alsa-project.org list addresses by the way; in the case of alsa-devel, the sourceforge one is gone even). Rene.

3 4