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

[alsa-devel] [patch 0/6] kirkwood openrd client audio support - v4
by apatard＠mandriva.com 31 May '10

31 May '10

Hi, This patchset is an other try to get merged openrd audio support mainline. The main change is going back to using kirkwood instead of orion, as the only orion having i2s is not merged mainline (if I got it right). I've also fixed the lastests comments on the cs42l51 codec patch. Arnaud Changes : - changed back from orion to kirkwood - removed cs42l51 adc input register initialisation - moved the DAPM event macro to soc-dapm - fixed cs42l51_set_dai_fmt() master/slave check - removed cs42l51 startup message - use snd_soc_update_bits() in cs42l51_pdn_event()

4 18

[alsa-devel] arecord and DMA misaligned error with device
by Ruchi Sirauthiya 31 May '10

31 May '10

Hi all, I am working on an OMAP board with Linux kernel 2.6.22.18-omap3 and ALSA. I got the following stack trace when running 'arecord' continuously for a long time (~30 minutes). What could be the reason for "page allocation failure" and "DMA misaligned error"? How can I solve this issue? Thanks and Regards, Ruchi Sirauthiya ---- LOG START ---- page allocation failure. order:5, mode:0xd1 [ 1249.046875] [<c0050380>] (dump_stack+0x0/0x14) from [<c00ae260>] (__alloc_pages+0x288/0x29c) [ 1249.054687] [<c00adfd8>] (__alloc_pages+0x0/0x29c) from [<c0051cac>] (__dma_alloc+0x17c/0x3d4) [ 1249.062500] [<c0051b30>] (__dma_alloc+0x0/0x3d4) from [<c0051f58>] (dma_alloc_coherent+0x24/0x30) [ 1249.070312] [<c0051f34>] (dma_alloc_coherent+0x0/0x30) from [<c02717b4>] (snd_omap_alsa_hw_params+0x54/0x7c) [ 1249.085937] [<c0271760>] (snd_omap_alsa_hw_params+0x0/0x7c) from [<c02683dc>] (snd_pcm_hw_params+0x90/0x2e0) [ 1249.093750] r5:c19d6000 r4:c15da000 [ 1249.093750] [<c026834c>] (snd_pcm_hw_params+0x0/0x2e0) from [<c02686b0>] (snd_pcm_hw_params_user+0x84/0xe4) [ 1249.109375] r9:cb8f4000 r8:c004c028 r7:c077e360 r6:c19d6000 r5:c077e360 [ 1249.109375] r4:001bdb48 [ 1249.117187] [<c026862c>] (snd_pcm_hw_params_user+0x0/0xe4) from [<c02689c0>] (snd_pcm_common_ioctl1+0x1cc/0xf04) [ 1249.125000] r7:c077e360 r6:001bdb48 r5:001bdb48 r4:c25c4111 [ 1249.132812] [<c02687f4>] (snd_pcm_common_ioctl1+0x0/0xf04) from [<c0269b44>] (snd_pcm_capture_ioctl1+0x44c/0x470) [ 1249.140625] [<c02696f8>] (snd_pcm_capture_ioctl1+0x0/0x470) from [<c0269b9c>] (snd_pcm_capture_ioctl+0x34/0x38) [ 1249.148437] [<c0269b68>] (snd_pcm_capture_ioctl+0x0/0x38) from [<c00d0adc>] (do_ioctl+0x34/0x78) [ 1249.164062] [<c00d0aa8>] (do_ioctl+0x0/0x78) from [<c00d0d80>] (vfs_ioctl+0x260/0x278) [ 1249.171875] r5:001bdb48 r4:ccb9a4e0 [ 1249.171875] [<c00d0b20>] (vfs_ioctl+0x0/0x278) from [<c00d0dd8>] (sys_ioctl+0x40/0x64) [ 1249.179687] r7:ccb9a4e0 r6:c25c4111 r5:001bdb48 r4:00000039 [ 1249.187500] [<c00d0d98>] (sys_ioctl+0x0/0x64) from [<c004be80>] (ret_fast_syscall+0x0/0x2c) [ 1249.195312] r7:00000036 r6:001ec5c8 r5:413b3bd4 r4:001bdb48 [ 1249.203125] Mem-info: [ 1249.203125] DMA per-cpu: [ 1249.203125] CPU 0: Hot: hi: 90, btch: 15 usd: 66 Cold: hi: 30, btch: 7 usd: 24 [ 1249.210937] Active:10183 inactive:35469 dirty:4 writeback:0 unstable:0 [ 1249.210937] free:739 slab:5167 mapped:2540 pagetables:125 bounce:0 [ 1249.226562] DMA free:2956kB min:1888kB low:2360kB high:2832kB active:40732kB inactive:141876kB present:223520kB pages_scanned:0 all_unreclaimable? no [ 1249.242187] lowmem_reserve[]: 0 0 [ 1249.242187] DMA: 13*4kB 33*8kB 27*16kB 3*32kB 31*64kB 1*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB 0*8192kB = 2956kB [ 1249.257812] Free swap: 0kB [ 1249.265625] 56320 pages of RAM [ 1249.273437] 1016 free pages [ 1249.273437] 3616 reserved pages [ 1249.273437] 5167 slab pages [ 1249.281250] 41756 pages shared [ 1249.281250] 0 pages swap cached [ 1249.343750] DMA misaligned error with device 1 [ 1249.343750] DMA misaligned error with device 1 [ 1249.351562] DMA misaligned error with device 1 [ 1249.359375] DMA misaligned error with device 1 [ 1249.359375] DMA misaligned error with device 1 [ 1249.367187] DMA misaligned error with device 1 [ 1249.367187] DMA misaligned error with device 1 [ 1249.375000] DMA misaligned error with device 1 [ 1249.375000] DMA misaligned error with device 1 [ 1249.382812] DMA misaligned error with device 1 [ 1249.390625] DMA misaligned error with device 1 [ 1249.390625] DMA misaligned error with device 1 [ 1249.398437] DMA misaligned error with device 1 [ 1249.398437] DMA misaligned error with device 1 [ 1249.406250] DMA misaligned error with device 1 [ 1249.406250] DMA misaligned error with device 1 ... ... log truncated ... ... ---- LOG END ----

2 1

[alsa-devel] [PATCH v2 0/8] ASoC: TWL4030: put the codec to off when not active
by Peter Ujfalusi 31 May '10

31 May '10

Hello, Changes since v1: - Patch 3: removed the shifting of offset cancellation value - Added register content reset functionality as suggested by Liam and Mark I have added Mark's Acked-by to patch 1-7, patch 8 is a new one, which shall address the comments regarding to the register content reset. To-do after this series: Revisit the register reset functionality to use burst write. Intro mail from the original series: The following series at the end will let the twl4030 codec to use BIAS_OFF instead of BIAS_STANDBY. The difference in power consumption is about 0.5mA. To achieve this: - The regcache has been reseted to codec default - The codec initialization has been optimized, it is no longer writes all 73 registers at startup, but only modifies few selected one. - The power related code has been cleaned up, and optimized - Support added for machine drivers to select the offset cancellation path - debug support for checking the codec default registers (machine drivers can ask for checking, but shall be disabled in production). - Added codec register reset functionality I guess that's it. The driver has been tested on a custom board with twl5031. It passed all of our internal test cases covering much of the codec features. --- Peter Ujfalusi (8): ASoC: TWL4030: Revisit codec defaults ASoC: TWL4030: Remove wrapper for power down ASoC: TWL4030: Make offset cancellation path configurable ASoC: TWL4030: Optimize the power up sequence ASoC: TWL4030: Helper to check chip default registers ASoC: TWL4030: Correct the ARXR2_APGA_CTL chip default ASoC: TWL4030: Use BIAS_OFF instead of BIAS_STANDBY, when not in use ASoC: TWL4030: Add functionalty to reset the registers sound/soc/codecs/twl4030.c | 291 +++++++++++++++++++++++++------------------- sound/soc/codecs/twl4030.h | 3 + 2 files changed, 169 insertions(+), 125 deletions(-)

3 10

[alsa-devel] How to support software mixing on QUALCOMM 72XX without mmap?
by Berg 31 May '10

31 May '10

Hello,guys. I'm porting ASLA1.0.23 to qualcomm 72xx base on ARM11. The first instance,"aplay test.wav", works now, but the second instance should fail with "device busy" error message. I have tried plug dmix and dshare "aply -D plug:mycard test.wav", but met a fail again for the soc chip doesn't support mmap. Of course,pcm_mycard is defined in /etc/asound.conf. In my pc ,both above cmds work well . Though,the document shows that: NOTE: For ALSA 1.0.9rc2 and higher you don't need to setup dmix for analogue output. Dmix is enabled by default for soundcards which don't support hardware mixing. You still need to set it up for digital outputs. My questions are: 1)Is software mixing implemented in ALSA Driver,or in ALSA LIB? 2)What could I do to make the software mixing work? Is there any special configure which need to adjust? And when testing on pc whth HD audio, I noticed that A)when aplay instances runs WITHOUT dmix, the result of"cat /etc/one_of_my_aplay_pid/maps" should like: ... b331d000-b731e000 rw-s 00000000 00:15 862346 /dev/shm/pulse-shm-2882420837 ... B)when aplay instances runs WITH dmix,the result is: ... b7511000-b7519000 rw-s 00000000 00:0f 4560 /dev/snd/pcmC0D0p b7519000-b751a000 rw-s 81000000 00:0f 4560 /dev/snd/pcmC0D0p b751a000-b751b000 r--s 80000000 00:0f 4560 /dev/snd/pcmC0D0p ... So,the third question: 3) From A) and B),can I make a result that the mmap feacture is not necessary for the buildin software mixing ? Expect your help,please berg

3 2

[alsa-devel] [PATCH] ALSA: hda: Use LPIB for another mainboard
by Daniel Chen 31 May '10

31 May '10

Hi, Attached is a patch against my latest LPIB patches that makes the original reporter's playback and capture audible by default. Please apply. This patch is applicable to all 2.6.3x stable series. Best, -Dan

2 1

[alsa-devel] [PATCH] ALSA: hda: Use mbp31 quirk for an iMac model
by Daniel Chen 31 May '10

31 May '10

Hi, Attached is a patch against sound-2.6 master enabling audible playback and capture for the original reporter's iMac. Please apply. This patch is applicable to all 2.6.3x stable series. Best, -Dan

2 2

[alsa-devel] [PATCH] ALSA: hda: Use LPIB for an ASUS device
by Daniel Chen 31 May '10

31 May '10

Hi, Attached is a patch against sound-2.6 master that enables the original reporter to adjust capture volume through PulseAudio without the daemon crashing. This patch is applicable to all 2.6.3x stable series. Best, -Dan

2 1

[alsa-devel] [PATCH] pcm: Define G723 3-bit and 5-bit formats
by Ben Collins 31 May '10

31 May '10

pcm: Define G723 3-bit and 5-bit formats This defines the 24bps and 40bps (8khz sample rate) G.723 codec formats. They are going to be used once I submit the driver for an mpeg4/g723 compression card. I've updated the signed value to -1 as per Takashi's comments since these are non-linear formats. Signed-off-by: Ben Collins <bcollins(a)bluecherry.net> Acked-by: Takashi Iwai <tiwai(a)suse.de> diff --git a/include/sound/asound.h b/include/sound/asound.h index 0985955..62e0857 100644 --- a/include/sound/asound.h +++ b/include/sound/asound.h @@ -212,7 +212,11 @@ typedef int __bitwise snd_pcm_format_t; #define SNDRV_PCM_FORMAT_S18_3BE ((__force snd_pcm_format_t) 41) /* in three bytes */ #define SNDRV_PCM_FORMAT_U18_3LE ((__force snd_pcm_format_t) 42) /* in three bytes */ #define SNDRV_PCM_FORMAT_U18_3BE ((__force snd_pcm_format_t) 43) /* in three bytes */ -#define SNDRV_PCM_FORMAT_LAST SNDRV_PCM_FORMAT_U18_3BE +#define SNDRV_PCM_FORMAT_G723_24 ((__force snd_pcm_format_t) 44) /* 8 samples in 3 bytes */ +#define SNDRV_PCM_FORMAT_G723_24_1B ((__force snd_pcm_format_t) 45) /* 1 sample in 1 byte */ +#define SNDRV_PCM_FORMAT_G723_40 ((__force snd_pcm_format_t) 46) /* 8 Samples in 5 bytes */ +#define SNDRV_PCM_FORMAT_G723_40_1B ((__force snd_pcm_format_t) 47) /* 1 sample in 1 byte */ +#define SNDRV_PCM_FORMAT_LAST SNDRV_PCM_FORMAT_G723_40_1B #ifdef SNDRV_LITTLE_ENDIAN #define SNDRV_PCM_FORMAT_S16 SNDRV_PCM_FORMAT_S16_LE diff --git a/include/sound/pcm.h b/include/sound/pcm.h index 8b611a5..37b5c3c 100644 --- a/include/sound/pcm.h +++ b/include/sound/pcm.h @@ -173,6 +173,10 @@ struct snd_pcm_ops { #define SNDRV_PCM_FMTBIT_U18_3LE (1ULL << SNDRV_PCM_FORMAT_U18_3LE) #define SNDRV_PCM_FMTBIT_S18_3BE (1ULL << SNDRV_PCM_FORMAT_S18_3BE) #define SNDRV_PCM_FMTBIT_U18_3BE (1ULL << SNDRV_PCM_FORMAT_U18_3BE) +#define SNDRV_PCM_FMTBIT_G723_24 (1ULL << SNDRV_PCM_FORMAT_G723_24) +#define SNDRV_PCM_FMTBIT_G723_24_1B (1ULL << SNDRV_PCM_FORMAT_G723_24_1B) +#define SNDRV_PCM_FMTBIT_G723_40 (1ULL << SNDRV_PCM_FORMAT_G723_40) +#define SNDRV_PCM_FMTBIT_G723_40_1B (1ULL << SNDRV_PCM_FORMAT_G723_40_1B) #ifdef SNDRV_LITTLE_ENDIAN #define SNDRV_PCM_FMTBIT_S16 SNDRV_PCM_FMTBIT_S16_LE diff --git a/sound/core/pcm_misc.c b/sound/core/pcm_misc.c index ea2bf82..434af3c 100644 --- a/sound/core/pcm_misc.c +++ b/sound/core/pcm_misc.c @@ -128,6 +128,14 @@ static struct pcm_format_data pcm_formats[SNDRV_PCM_FORMAT_LAST+1] = { .width = 4, .phys = 4, .le = -1, .signd = -1, .silence = {}, }, + [SNDRV_PCM_FORMAT_G723_24] = { + .width = 3, .phys = 3, .le = -1, .signd = -1, + .silence = {}, + }, + [SNDRV_PCM_FORMAT_G723_40] = { + .width = 5, .phys = 5, .le = -1, .signd = -1, + .silence = {}, + }, /* FIXME: the following three formats are not defined properly yet */ [SNDRV_PCM_FORMAT_MPEG] = { .le = -1, .signd = -1, @@ -186,6 +194,14 @@ static struct pcm_format_data pcm_formats[SNDRV_PCM_FORMAT_LAST+1] = { .width = 18, .phys = 24, .le = 0, .signd = 0, .silence = { 0x02, 0x00, 0x00 }, }, + [SNDRV_PCM_FORMAT_G723_24_1B] = { + .width = 3, .phys = 8, .le = -1, .signd = -1, + .silence = {}, + }, + [SNDRV_PCM_FORMAT_G723_40_1B] = { + .width = 5, .phys = 8, .le = -1, .signd = -1, + .silence = {}, + }, }; -- Bluecherry: http://www.bluecherrydvr.com/ SwissDisk : http://www.swissdisk.com/ Ubuntu : http://www.ubuntu.com/ My Blog : http://ben-collins.blogspot.com/

2 1

[alsa-devel] [PATCH 1/4] snd-usb-caiaq: Restore 'Control vinyl' input mode on A4DJ
by Mark Hills 31 May '10

31 May '10

This feature was undocumented on early A4DJ units. It is indicated by lighting both the 'line' and 'phono' lamps at the same time. Newer units document this and the newer Windows drivers enable this for all units, so restore the functionality. This patch simplifies the code and changes the mode mapping to match the A8DJ, favouring simpler code and consistency over keeping the existing mapping. Both 'Control vinyl' and 'Phono' input modes enable the hardware preamp. The difference is the input impedance. This reverts commit 9a9527e. Acked-by: Daniel Mack <daniel(a)caiaq.de> Signed-off-by: Mark Hills <mark(a)pogo.org.uk> --- sound/usb/caiaq/control.c | 30 ++---------------------------- 1 files changed, 2 insertions(+), 28 deletions(-) diff --git a/sound/usb/caiaq/control.c b/sound/usb/caiaq/control.c index 36ed703..70c3866 100644 --- a/sound/usb/caiaq/control.c +++ b/sound/usb/caiaq/control.c @@ -42,21 +42,12 @@ static int control_info(struct snd_kcontrol *kcontrol, switch (dev->chip.usb_id) { case USB_ID(USB_VID_NATIVEINSTRUMENTS, USB_PID_AUDIO8DJ): - if (pos == 0) { - /* current input mode of A8DJ */ - uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER; - uinfo->value.integer.min = 0; - uinfo->value.integer.max = 2; - return 0; - } - break; - case USB_ID(USB_VID_NATIVEINSTRUMENTS, USB_PID_AUDIO4DJ): if (pos == 0) { - /* current input mode of A4DJ */ + /* current input mode of A8DJ and A4DJ */ uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER; uinfo->value.integer.min = 0; - uinfo->value.integer.max = 1; + uinfo->value.integer.max = 2; return 0; } break; @@ -86,14 +77,6 @@ static int control_get(struct snd_kcontrol *kcontrol, struct snd_usb_caiaqdev *dev = caiaqdev(chip->card); int pos = kcontrol->private_value; - if (dev->chip.usb_id == - USB_ID(USB_VID_NATIVEINSTRUMENTS, USB_PID_AUDIO4DJ)) { - /* A4DJ has only one control */ - /* do not expose hardware input mode 0 */ - ucontrol->value.integer.value[0] = dev->control_state[0] - 1; - return 0; - } - if (pos & CNT_INTVAL) ucontrol->value.integer.value[0] = dev->control_state[pos & ~CNT_INTVAL]; @@ -113,15 +96,6 @@ static int control_put(struct snd_kcontrol *kcontrol, unsigned char cmd = EP1_CMD_WRITE_IO; switch (dev->chip.usb_id) { - case USB_ID(USB_VID_NATIVEINSTRUMENTS, USB_PID_AUDIO4DJ): { - /* A4DJ has only one control */ - /* do not expose hardware input mode 0 */ - dev->control_state[0] = ucontrol->value.integer.value[0] + 1; - snd_usb_caiaq_send_command(dev, EP1_CMD_WRITE_IO, - dev->control_state, sizeof(dev->control_state)); - return 1; - } - case USB_ID(USB_VID_NATIVEINSTRUMENTS, USB_PID_TRAKTORKONTROLX1): cmd = EP1_CMD_DIMM_LEDS; break; -- 1.7.1

2 4

[alsa-devel] [PATCH] ALSA: hda: Use LPIB for a Shuttle device
by Daniel Chen 31 May '10

31 May '10

Hi, Attached please find a patch against current sound-2.6 master that fixes audio anomalies and PulseAudio crashing by specifying LPIB method explicitly. Please apply. This patch is applicable to all 2.6.3x stable series. Best, -Dan

2 1