November 2008 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [RFC][PATCH] pxa2xx-ac97: switch AC unit to correct state before probing
by Dmitry Baryshkov 11 Nov '08

11 Nov '08

Hi, I've stumbled upon a problem with my tosa unit. The bootloader leaves ac97 in some partially enabled state, so when request_irq enables the AC97 irq, I see a unending storm of irq requests. I've come with the following patch (against 2.6.27, if the idea is OK, I'll rebase against 2.6.28-rc). From: Dmitry Baryshkov <dbaryshkov(a)gmail.com> If AC97 unit is in partially enabled state, early request_irq can trigger IRQ storm or even full hang up. Workaround this by forcibly switching ACLINK off at the start of the probe. Signed-off-by: Dmitry Baryshkov <dbaryshkov(a)gmail.com> --- sound/soc/pxa/pxa2xx-ac97.c | 30 +++++++++++++++++------------- 1 files changed, 17 insertions(+), 13 deletions(-) diff --git a/sound/soc/pxa/pxa2xx-ac97.c b/sound/soc/pxa/pxa2xx-ac97.c index d94a495..259c6e1 100644 --- a/sound/soc/pxa/pxa2xx-ac97.c +++ b/sound/soc/pxa/pxa2xx-ac97.c @@ -315,10 +315,6 @@ static int pxa2xx_ac97_probe(struct platform_device *pdev, { int ret; - ret = request_irq(IRQ_AC97, pxa2xx_ac97_irq, IRQF_DISABLED, "AC97", NULL); - if (ret < 0) - goto err; - pxa_gpio_mode(GPIO31_SYNC_AC97_MD); pxa_gpio_mode(GPIO30_SDATA_OUT_AC97_MD); pxa_gpio_mode(GPIO28_BITCLK_AC97_MD); @@ -331,28 +327,36 @@ static int pxa2xx_ac97_probe(struct platform_device *pdev, if (IS_ERR(ac97conf_clk)) { ret = PTR_ERR(ac97conf_clk); ac97conf_clk = NULL; - goto err_irq; + goto err_conf; } #endif + GCR = GCR_ACLINK_OFF; + ac97_clk = clk_get(&pdev->dev, "AC97CLK"); if (IS_ERR(ac97_clk)) { ret = PTR_ERR(ac97_clk); ac97_clk = NULL; - goto err_irq; + goto err_clk; } clk_enable(ac97_clk); + + ret = request_irq(IRQ_AC97, pxa2xx_ac97_irq, IRQF_DISABLED, "AC97", NULL); + if (ret < 0) + goto err_irq; + return 0; - err_irq: +err_irq: GCR |= GCR_ACLINK_OFF; + clk_disable(ac97_clk); + clk_put(ac97_clk); + ac97_clk = NULL; +err_clk: #ifdef CONFIG_PXA27x - if (ac97conf_clk) { - clk_put(ac97conf_clk); - ac97conf_clk = NULL; - } + clk_put(ac97conf_clk); + ac97conf_clk = NULL; +err_conf: #endif - free_irq(IRQ_AC97, NULL); - err: return ret; } -- 1.5.6.5 -- With best wishes Dmitry

2 3

[alsa-devel] [PATCH] Make the HP EliteBook 8530p use AD1884A model 'laptop' by default.
by Travis Place 11 Nov '08

11 Nov '08

Changelog: Added a QUIRK to patch_analog.c for the HP Elitebook 8530p (IDs 0x103c:0x30e7) to use AD1884A model 'laptop' by default. Playback and Capture confirmed working. Signed-off-by: Travis Place

2 1

Re: [alsa-devel] [Intel-gfx] G45 user experience
by Wu Fengguang 11 Nov '08

11 Nov '08

On Mon, Nov 10, 2008 at 08:15:01PM -0800, Shane W wrote: > On Tue, Nov 11, 2008 at 10:32:40AM +0800, Fu Michael wrote: > > Do you have any HDMI sink device that can support 5.1 or 7.1 audio? I'm > > looking for those kind of devices as well... > > I'm not really sure what you mean by HDMI sink but my > receiver which connects to the G45 does support 5.1 and 7.1 > HDMI decoding along with TRUEHD, DTSHD and DSD so I can > certainly test anything that comes down the pipe as far as > patches go. Wow, thank you very much! > By the way, had a bit more time to test out these HDMI sync > xorg patches. Under mplayer, after playing audio for about > 10 or so minutes with a video stream, the audio starts to > drop out and then stops altogether. Stopping and starting Do you mean the volume goes down gradually? > mplayer doesn't fix it, stopping mplayer and playing a wav > via aplay gives silence. Stopping Xorg does correct it and > once Xorg is restarted, it'll run with video for another 10 > or so minutes and stop again. This is using Saturday's GIT OK, then restarting Xorg fixed it temporarily. I wonder if restarting the monitor restores the sound as well? > on ALSA, Xorg and Intel with the IALSA Intel hdmi patch, > and the HDMI patches for Xorg and the Intel driver applied. Thank you, Fengguang

1 0

[alsa-devel] HDA-Intel - Patch to support detecting HD Audio devices with PCI class code
by Yang, Libin 11 Nov '08

11 Nov '08

Sorry, adding the title... Thanks & Regards, Libin > -----Original Message----- > From: Yang, Libin > Sent: Tuesday, November 11, 2008 10:43 AM > To: 'Takashi Iwai' > Cc: 'alsa-devel(a)alsa-project.org' > Subject: > > Hi Takashi, > > This patch is to support detecting new AMD HD Audio devices with HDA PCI > class code. Would you please review it? Thanks a lot! > -------------------------------------------------------------------- > > Signed-off-by: Libin Yang <libin.yang(a)amd.com> > > --- alsa-driver-1.0.18.orig/alsa-kernel/pci/hda/hda_intel.c 2008-10-29 > 20:41:35.000000000 +0800 > +++ alsa-driver-1.0.18/alsa-kernel/pci/hda/hda_intel.c 2008-11-11 > 18:33:14.000000000 +0800 > @@ -291,6 +291,7 @@ > /* Define VIA HD Audio Device ID*/ > #define VIA_HDAC_DEVICE_ID 0x3288 > > +#define PCI_CLASS_MULTIMEDIA_HDA 0x040300 > > /* > */ > @@ -410,6 +411,7 @@ > AZX_DRIVER_ULI, > AZX_DRIVER_NVIDIA, > AZX_DRIVER_TERA, > + AZX_DRIVER_AMD_AUTO, > AZX_NUM_DRIVERS, /* keep this as last entry */ > }; > > @@ -423,6 +425,7 @@ > [AZX_DRIVER_ULI] = "HDA ULI M5461", > [AZX_DRIVER_NVIDIA] = "HDA NVidia", > [AZX_DRIVER_TERA] = "HDA Teradici", > + [AZX_DRIVER_AMD_AUTO] = "HDA AMD", > }; > > /* > @@ -1198,6 +1201,7 @@ > static unsigned int azx_default_codecs[AZX_NUM_DRIVERS] __devinitdata = { > [AZX_DRIVER_ICH] = 3, > [AZX_DRIVER_ATI] = 3, > + [AZX_DRIVER_AMD_AUTO] = 3, > }; > > static int __devinit azx_codec_create(struct azx *chip, const char *model, > @@ -2146,6 +2150,7 @@ > chip->playback_streams = ULI_NUM_PLAYBACK; > chip->capture_streams = ULI_NUM_CAPTURE; > break; > + case AZX_DRIVER_AMD_AUTO: > case AZX_DRIVER_ATIHDMI: > chip->playback_streams = ATIHDMI_NUM_PLAYBACK; > chip->capture_streams = ATIHDMI_NUM_CAPTURE; > @@ -2373,6 +2378,9 @@ > { PCI_DEVICE(0x10de, 0x0bd7), .driver_data = AZX_DRIVER_NVIDIA }, > /* Teradici */ > { PCI_DEVICE(0x6549, 0x1200), .driver_data = AZX_DRIVER_TERA }, > + /* AMD Generic, PCI class code and Vendor ID for HD Audio */ > + { PCI_VENDOR_ID_ATI, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, > + PCI_CLASS_MULTIMEDIA_HDA, 0xffffff, AZX_DRIVER_AMD_AUTO }, > { 0, } > }; > MODULE_DEVICE_TABLE(pci, azx_ids); > > Thanks & Regards, > Libin

1 0

Re: [alsa-devel] Bugs on aspire one A150
by Maxim Levitsky 10 Nov '08

10 Nov '08

Bob Copeland wrote: > On Fri, Nov 7, 2008 at 10:26 PM, Maxim Levitsky <maximlevitsky(a)gmail.com> wrote: >> ** 2 - wireless: not to mention the fact that ath5k wasn't installed by >> default in ubuntu... >> wireless more or less works, but kernel log is full of backtraces. > > As you've seen, these should at least be gone now... > >> Was able to connect to my WPA2 access point. >> Sometimes wireless fails completely, especially after suspend to ram. > > Did you get noise calibration failures in dmesg at this point? Felix posted > a patch recently that handles them better. Yep, but they seems to be gone in latest -git too. Wireless work fine now, it seems like my iwl3945. > >> Advanced features like monitor/injection work, but when I changed the card's >> mac address it stopped working. >> I also noticed that if I then start airodump, then wireless works with new >> mac. Why it doesn't accept new mac?, can this be fixed? > >> ** 4 - wireless led doesn't work. >> ath5k devs, can you fix this? > > Currently all the LED code we have in there is for setting a particular gpio > for particular laptops (and they differ between models). We only have quirks > for IBM and HP, none for Acer. Also, the legacy-hal seems to have a different > strategy for 2425 chips. I can try to hack up some code for you to test. > I have seen on the web that madwifi did support the led with some gpio settings sysctl -w dev.wifi0.ledpin=3 sysctl -w dev.wifi0.softled=1 from https://help.ubuntu.com/community/AspireOne Best regards, Maxim Levitsky

1 0

[alsa-devel] MS-7350 Realtek ALC888
by Philipp Matthias Hahn 10 Nov '08

10 Nov '08

Hello! I have an MSI MS-7350 motherborad with an onboard Realtek ALC888 HDA codec. I need to specify model=6stack-dig to get the internal CD connector to show up, else automatic BIOS scanning doesn't find it. Might be that you want to add the appropriate DMI/PCI quirk stuff, so other users get the benefit of automatic detection. $ lspci -s 0:10.1 00:10.1 Audio device: nVidia Corporation MCP51 High Definition Audio (rev a2) $ lspci -s 0:10.1 -n -vvv 00:10.1 0403: 10de:026c (rev a2) Subsystem: 1462:7350 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Latency: 0 (500ns min, 1250ns max) Interrupt: pin B routed to IRQ 22 Region 0: Memory at f9ff8000 (32-bit, non-prefetchable) [size=16K] Capabilities: <access denied> Kernel driver in use: HDA Intel Kernel modules: snd-hda-intel BYtE Philipp -- / / (_)__ __ ____ __ Philipp Hahn / /__/ / _ \/ // /\ \/ / /____/_/_//_/\_,_/ /_/\_\ pmhahn(a)titan.lahn.de

1 0

[alsa-devel] [PATCH] ALSA SOC driver for s3c24xx: 8 bit sound fix
by Christian Pellegrin 10 Nov '08

10 Nov '08

fixes playing/recording of 8 bit audio files. Generated on 20081108 against v2.6.27 Signed-off-by: Christian Pellegrin <chripell(a)fsfe.org> --- sound/soc/s3c24xx/s3c24xx-i2s.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/sound/soc/s3c24xx/s3c24xx-i2s.c b/sound/soc/s3c24xx/s3c24xx-i2s.c index ba4476b..c18977b 100644 --- a/sound/soc/s3c24xx/s3c24xx-i2s.c +++ b/sound/soc/s3c24xx/s3c24xx-i2s.c @@ -261,10 +261,17 @@ static int s3c24xx_i2s_hw_params(struct snd_pcm_substream *substream, switch (params_format(params)) { case SNDRV_PCM_FORMAT_S8: + iismod &= ~S3C2410_IISMOD_16BIT; + ((struct s3c24xx_pcm_dma_params *) + rtd->dai->cpu_dai->dma_data)->dma_size = 1; break; case SNDRV_PCM_FORMAT_S16_LE: iismod |= S3C2410_IISMOD_16BIT; + ((struct s3c24xx_pcm_dma_params *) + rtd->dai->cpu_dai->dma_data)->dma_size = 2; break; + default: + return -EINVAL; } writel(iismod, s3c24xx_i2s.regs + S3C2410_IISMOD); -- 1.4.4.4

4 4

[alsa-devel] [PATCH] gusextreme: Fix build errors
by Ville Syrjala 10 Nov '08

10 Nov '08

gusextreme depends on opl3 support. Add the approriate select to Kconfig. Also remove the unnecessary hwdep select. Relevant build errors: ERROR: "snd_opl3_hwdep_new" [sound/isa/gus/snd-gusextreme.ko] undefined! ERROR: "snd_opl3_create" [sound/isa/gus/snd-gusextreme.ko] undefined! Signed-off-by: Ville Syrjala <syrjala(a)sci.fi> --- sound/isa/Kconfig | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/sound/isa/Kconfig b/sound/isa/Kconfig index 660beb4..ce0aa04 100644 --- a/sound/isa/Kconfig +++ b/sound/isa/Kconfig @@ -211,7 +211,7 @@ config SND_GUSCLASSIC config SND_GUSEXTREME tristate "Gravis UltraSound Extreme" - select SND_HWDEP + select SND_OPL3_LIB select SND_MPU401_UART select SND_PCM help -- 1.5.6.4

2 1

[alsa-devel] [PATCH] Renamed the struct soc_enum field .mask to .count. The field refers to the number of enumerable entries instead of being a bit mask. This change makes the interface a more clear.
by Marc Singer 10 Nov '08

10 Nov '08

Signed-off-by: Marc Singer <elf(a)buici.com> --- include/sound/soc-codec.h | 14 +++++++------- sound/soc/soc-core.c | 20 ++++++++++---------- sound/soc/soc-dapm.c | 14 +++++++------- 3 files changed, 24 insertions(+), 24 deletions(-) diff --git a/include/sound/soc-codec.h b/include/sound/soc-codec.h index d8595ef..d59015a 100644 --- a/include/sound/soc-codec.h +++ b/include/sound/soc-codec.h @@ -79,13 +79,13 @@ .get = snd_soc_get_volsw_2r, .put = snd_soc_put_volsw_2r, \ .private_value = (reg_left) | ((shift) << 8) | \ ((max) << 12) | ((invert) << 20) | ((reg_right) << 24) } -#define SOC_ENUM_DOUBLE(xreg, xshift_l, xshift_r, xmask, xtexts) \ +#define SOC_ENUM_DOUBLE(xreg, xshift_l, xshift_r, xcount, xtexts) \ { .reg = xreg, .shift_l = xshift_l, .shift_r = xshift_r, \ - .mask = xmask, .texts = xtexts } -#define SOC_ENUM_SINGLE(xreg, xshift, xmask, xtexts) \ - SOC_ENUM_DOUBLE(xreg, xshift, xshift, xmask, xtexts) -#define SOC_ENUM_SINGLE_EXT(xmask, xtexts) \ -{ .mask = xmask, .texts = xtexts } + .count = xcount, .texts = xtexts } +#define SOC_ENUM_SINGLE(xreg, xshift, xcount, xtexts) \ + SOC_ENUM_DOUBLE(xreg, xshift, xshift, xcount, xtexts) +#define SOC_ENUM_SINGLE_EXT(xcount, xtexts) \ +{ .count = xcount, .texts = xtexts } #define SOC_ENUM(xname, xenum) \ { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname,\ .info = snd_soc_info_enum_double, \ @@ -180,7 +180,7 @@ struct soc_enum { unsigned short reg2; unsigned char shift_l; unsigned char shift_r; - unsigned int mask; + unsigned int count; /* Count of options */ const char **texts; void *dapm; }; diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 706f98d..b8efb4b 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -1262,10 +1262,10 @@ int snd_soc_info_enum_double(struct snd_kcontrol *kcontrol, uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; uinfo->count = e->shift_l == e->shift_r ? 1 : 2; - uinfo->value.enumerated.items = e->mask; + uinfo->value.enumerated.items = e->count; - if (uinfo->value.enumerated.item > e->mask - 1) - uinfo->value.enumerated.item = e->mask - 1; + if (uinfo->value.enumerated.item > e->count - 1) + uinfo->value.enumerated.item = e->count - 1; strcpy(uinfo->value.enumerated.name, e->texts[uinfo->value.enumerated.item]); return 0; @@ -1288,7 +1288,7 @@ int snd_soc_get_enum_double(struct snd_kcontrol *kcontrol, struct soc_enum *e = (struct soc_enum *)kcontrol->private_value; unsigned short val, bitmask; - for (bitmask = 1; bitmask < e->mask; bitmask <<= 1) + for (bitmask = 1; bitmask < e->count; bitmask <<= 1) ; val = snd_soc_read(codec, e->reg); ucontrol->value.enumerated.item[0] = (val >> e->shift_l) & (bitmask - 1); @@ -1317,14 +1317,14 @@ int snd_soc_put_enum_double(struct snd_kcontrol *kcontrol, unsigned short val; unsigned short mask, bitmask; - for (bitmask = 1; bitmask < e->mask; bitmask <<= 1) + for (bitmask = 1; bitmask < e->count; bitmask <<= 1) ; - if (ucontrol->value.enumerated.item[0] > e->mask - 1) + if (ucontrol->value.enumerated.item[0] > e->count - 1) return -EINVAL; val = ucontrol->value.enumerated.item[0] << e->shift_l; mask = (bitmask - 1) << e->shift_l; if (e->shift_l != e->shift_r) { - if (ucontrol->value.enumerated.item[1] > e->mask - 1) + if (ucontrol->value.enumerated.item[1] > e->count - 1) return -EINVAL; val |= ucontrol->value.enumerated.item[1] << e->shift_r; mask |= (bitmask - 1) << e->shift_r; @@ -1351,10 +1351,10 @@ int snd_soc_info_enum_ext(struct snd_kcontrol *kcontrol, uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; uinfo->count = 1; - uinfo->value.enumerated.items = e->mask; + uinfo->value.enumerated.items = e->count; - if (uinfo->value.enumerated.item > e->mask - 1) - uinfo->value.enumerated.item = e->mask - 1; + if (uinfo->value.enumerated.item > e->count - 1) + uinfo->value.enumerated.item = e->count - 1; strcpy(uinfo->value.enumerated.name, e->texts[uinfo->value.enumerated.item]); return 0; diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index f203693..5b83329 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -119,13 +119,13 @@ static void dapm_set_path_status(struct snd_soc_dapm_widget *w, struct soc_enum *e = (struct soc_enum *)w->kcontrols[i].private_value; int val, item, bitmask; - for (bitmask = 1; bitmask < e->mask; bitmask <<= 1) + for (bitmask = 1; bitmask < e->count; bitmask <<= 1) ; val = snd_soc_read(w->codec, e->reg); item = (val >> e->shift_l) & (bitmask - 1); p->connect = 0; - for (i = 0; i < e->mask; i++) { + for (i = 0; i < e->count; i++) { if (!(strcmp(p->name, e->texts[i])) && item == i) p->connect = 1; } @@ -162,7 +162,7 @@ static int dapm_connect_mux(struct snd_soc_card *soc_card, struct soc_enum *e = (struct soc_enum *)kcontrol->private_value; int i; - for (i = 0; i < e->mask; i++) { + for (i = 0; i < e->count; i++) { if (!(strcmp(control_name, e->texts[i]))) { list_add(&path->list, &soc_card->dapm_paths); list_add(&path->list_sink, &dest->sources); @@ -1197,7 +1197,7 @@ int snd_soc_dapm_get_enum_double(struct snd_kcontrol *kcontrol, struct soc_enum *e = (struct soc_enum *)kcontrol->private_value; unsigned short val, bitmask; - for (bitmask = 1; bitmask < e->mask; bitmask <<= 1) + for (bitmask = 1; bitmask < e->count; bitmask <<= 1) ; val = snd_soc_read(widget->codec, e->reg); ucontrol->value.enumerated.item[0] = (val >> e->shift_l) & (bitmask - 1); @@ -1227,15 +1227,15 @@ int snd_soc_dapm_put_enum_double(struct snd_kcontrol *kcontrol, unsigned short mask, bitmask; int ret = 0; - for (bitmask = 1; bitmask < e->mask; bitmask <<= 1) + for (bitmask = 1; bitmask < e->count; bitmask <<= 1) ; - if (ucontrol->value.enumerated.item[0] > e->mask - 1) + if (ucontrol->value.enumerated.item[0] > e->count - 1) return -EINVAL; mux = ucontrol->value.enumerated.item[0]; val = mux << e->shift_l; mask = (bitmask - 1) << e->shift_l; if (e->shift_l != e->shift_r) { - if (ucontrol->value.enumerated.item[1] > e->mask - 1) + if (ucontrol->value.enumerated.item[1] > e->count - 1) return -EINVAL; val |= ucontrol->value.enumerated.item[1] << e->shift_r; mask |= (bitmask - 1) << e->shift_r; -- 1.5.6.5

2 1

[alsa-devel] Non-functional Altec Lansing FX5051 USB-soundcard + speakers
by Jeremy Nickurak 10 Nov '08

10 Nov '08

About 14 months ago I purchased a nice new ~$300 set of altec lansing 5.1 speakers, which include a USB-sound card. Since that time they've largely gone unused, or at least not properly used. I brought this up on the mailing list at the time, but didn't get much response, so I thought I'd poke my head up again and see if anybody has any insight. First off, the set works correctly, out-of-the-box under windows, without having to install any special drivers or software at all. Under linux, plugging them in instantly freezes the usb sound card up, and locks it at 100% volume, in such a way that there's absolutely no way to turn the volume down: not from alsamixer, and not from the speaker's hardware control dial. Sound does play, but again, it's unbearably loud, to the point where I don't do it for risk of damaging my speakers and/or ears. I filed the bug away on the bug-tracker, but haven't heard much out of it since. The bug contains my lsusb -v information, the "contents" output from amixer for this card, and some preliminary results showing some ability to tweak the volume with amixer, but not reliably. Any suggestions are welcome. I'm happy to experiment with this set more if it can help, run experimental code, etc. Heck, I'd go at the thing with a screwdriver for chipset information if I know what I was looking for or thought it could help.

2 1