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

[alsa-devel] [PATCH] Fix pxa2xx-ac97-lib.c compilation
by Marc Zyngier 14 Oct '08

14 Oct '08

[This patch was posted to linux-arm-kernel ML yesterday, and reposted here in the hope that it will be merged soon] The last ALSA merge broke pxa2xx-ac97-lib.c, as it brought back references to cpu_is_pxa21x that Eric Miao removed in commit 0ffcbfd54ea81ca24c0749f55ca4fcf3e2bdc23e: [ARM] pxa: make cpu_is_pxa2* macros more consistent This patch gets rid of those references, and only keeps cpu_is_pxa25x(). Signed-off-by: Marc Zyngier <maz(a)misterjones.org> Acked-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> Acked-by: Eric Miao <eric.miao(a)marvell.com> diff --git a/sound/arm/pxa2xx-ac97-lib.c b/sound/arm/pxa2xx-ac97-lib.c index 99026df..34c1d94 100644 --- a/sound/arm/pxa2xx-ac97-lib.c +++ b/sound/arm/pxa2xx-ac97-lib.c @@ -50,7 +50,7 @@ unsigned short pxa2xx_ac97_read(struct snd_ac97 *ac97, unsigned short reg) mutex_lock(&car_mutex); /* set up primary or secondary codec space */ - if ((cpu_is_pxa21x() || cpu_is_pxa25x()) && reg == AC97_GPIO_STATUS) + if (cpu_is_pxa25x() && reg == AC97_GPIO_STATUS) reg_addr = ac97->num ? &SMC_REG_BASE : &PMC_REG_BASE; else reg_addr = ac97->num ? &SAC_REG_BASE : &PAC_REG_BASE; @@ -90,7 +90,7 @@ void pxa2xx_ac97_write(struct snd_ac97 *ac97, unsigned short reg, mutex_lock(&car_mutex); /* set up primary or secondary codec space */ - if ((cpu_is_pxa21x() || cpu_is_pxa25x()) && reg == AC97_GPIO_STATUS) + if (cpu_is_pxa25x() && reg == AC97_GPIO_STATUS) reg_addr = ac97->num ? &SMC_REG_BASE : &PMC_REG_BASE; else reg_addr = ac97->num ? &SAC_REG_BASE : &PAC_REG_BASE; @@ -200,7 +200,7 @@ static inline void pxa_ac97_cold_pxa3xx(void) bool pxa2xx_ac97_try_warm_reset(struct snd_ac97 *ac97) { #ifdef CONFIG_PXA25x - if (cpu_is_pxa21x() || cpu_is_pxa25x()) + if (cpu_is_pxa25x()) pxa_ac97_warm_pxa25x(); else #endif @@ -230,7 +230,7 @@ EXPORT_SYMBOL_GPL(pxa2xx_ac97_try_warm_reset); bool pxa2xx_ac97_try_cold_reset(struct snd_ac97 *ac97) { #ifdef CONFIG_PXA25x - if (cpu_is_pxa21x() || cpu_is_pxa25x()) + if (cpu_is_pxa25x()) pxa_ac97_cold_pxa25x(); else #endif @@ -301,7 +301,7 @@ EXPORT_SYMBOL_GPL(pxa2xx_ac97_hw_suspend); int pxa2xx_ac97_hw_resume(void) { - if (cpu_is_pxa21x() || cpu_is_pxa25x() || cpu_is_pxa27x()) { + if (cpu_is_pxa25x() || cpu_is_pxa27x()) { pxa_gpio_mode(GPIO31_SYNC_AC97_MD); pxa_gpio_mode(GPIO30_SDATA_OUT_AC97_MD); pxa_gpio_mode(GPIO28_BITCLK_AC97_MD); @@ -325,7 +325,7 @@ int __devinit pxa2xx_ac97_hw_probe(struct platform_device *dev) if (ret < 0) goto err; - if (cpu_is_pxa21x() || cpu_is_pxa25x() || cpu_is_pxa27x()) { + if (cpu_is_pxa25x() || cpu_is_pxa27x()) { pxa_gpio_mode(GPIO31_SYNC_AC97_MD); pxa_gpio_mode(GPIO30_SDATA_OUT_AC97_MD); pxa_gpio_mode(GPIO28_BITCLK_AC97_MD); -- A rat a day keeps the plague away.

2 1

[alsa-devel] [PATCH 2.6.28] Fix pxa2xx-ac97-lib.c compilation
by Mark Brown 14 Oct '08

14 Oct '08

From: Marc Zyngier <maz(a)misterjones.org> The last ALSA merge broke pxa2xx-ac97-lib.c, as it brought back references to cpu_is_pxa21x that Eric Miao removed in commit 0ffcbfd54ea81ca24c0749f55ca4fcf3e2bdc23e: [ARM] pxa: make cpu_is_pxa2* macros more consistent This patch gets rid of those references, and only keeps cpu_is_pxa25x(). Signed-off-by: Marc Zyngier <maz(a)misterjones.org> Acked-by: Eric Miao <eric.miao(a)marvell.com> Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- Please merge into 2.6.28 - this breaks the build on PXA platforms. sound/arm/pxa2xx-ac97-lib.c | 12 ++++++------ 1 files changed, 6 insertions(+), 6 deletions(-) diff --git a/sound/arm/pxa2xx-ac97-lib.c b/sound/arm/pxa2xx-ac97-lib.c index 99026df..34c1d94 100644 --- a/sound/arm/pxa2xx-ac97-lib.c +++ b/sound/arm/pxa2xx-ac97-lib.c @@ -50,7 +50,7 @@ unsigned short pxa2xx_ac97_read(struct snd_ac97 *ac97, unsigned short reg) mutex_lock(&car_mutex); /* set up primary or secondary codec space */ - if ((cpu_is_pxa21x() || cpu_is_pxa25x()) && reg == AC97_GPIO_STATUS) + if (cpu_is_pxa25x() && reg == AC97_GPIO_STATUS) reg_addr = ac97->num ? &SMC_REG_BASE : &PMC_REG_BASE; else reg_addr = ac97->num ? &SAC_REG_BASE : &PAC_REG_BASE; @@ -90,7 +90,7 @@ void pxa2xx_ac97_write(struct snd_ac97 *ac97, unsigned short reg, mutex_lock(&car_mutex); /* set up primary or secondary codec space */ - if ((cpu_is_pxa21x() || cpu_is_pxa25x()) && reg == AC97_GPIO_STATUS) + if (cpu_is_pxa25x() && reg == AC97_GPIO_STATUS) reg_addr = ac97->num ? &SMC_REG_BASE : &PMC_REG_BASE; else reg_addr = ac97->num ? &SAC_REG_BASE : &PAC_REG_BASE; @@ -200,7 +200,7 @@ static inline void pxa_ac97_cold_pxa3xx(void) bool pxa2xx_ac97_try_warm_reset(struct snd_ac97 *ac97) { #ifdef CONFIG_PXA25x - if (cpu_is_pxa21x() || cpu_is_pxa25x()) + if (cpu_is_pxa25x()) pxa_ac97_warm_pxa25x(); else #endif @@ -230,7 +230,7 @@ EXPORT_SYMBOL_GPL(pxa2xx_ac97_try_warm_reset); bool pxa2xx_ac97_try_cold_reset(struct snd_ac97 *ac97) { #ifdef CONFIG_PXA25x - if (cpu_is_pxa21x() || cpu_is_pxa25x()) + if (cpu_is_pxa25x()) pxa_ac97_cold_pxa25x(); else #endif @@ -301,7 +301,7 @@ EXPORT_SYMBOL_GPL(pxa2xx_ac97_hw_suspend); int pxa2xx_ac97_hw_resume(void) { - if (cpu_is_pxa21x() || cpu_is_pxa25x() || cpu_is_pxa27x()) { + if (cpu_is_pxa25x() || cpu_is_pxa27x()) { pxa_gpio_mode(GPIO31_SYNC_AC97_MD); pxa_gpio_mode(GPIO30_SDATA_OUT_AC97_MD); pxa_gpio_mode(GPIO28_BITCLK_AC97_MD); @@ -325,7 +325,7 @@ int __devinit pxa2xx_ac97_hw_probe(struct platform_device *dev) if (ret < 0) goto err; - if (cpu_is_pxa21x() || cpu_is_pxa25x() || cpu_is_pxa27x()) { + if (cpu_is_pxa25x() || cpu_is_pxa27x()) { pxa_gpio_mode(GPIO31_SYNC_AC97_MD); pxa_gpio_mode(GPIO30_SDATA_OUT_AC97_MD); pxa_gpio_mode(GPIO28_BITCLK_AC97_MD); -- 1.5.6.5

2 1

[alsa-devel] [PATCH] ASoC: Hide TLV320AIC26 configuration option for non-OpenFirwmare users
by Mark Brown 14 Oct '08

14 Oct '08

Since Kconfig ignores dependencies on select and other users should select SND_SOC_TLV320AIC26 we can add a dependency on the OpenFirmware helper to avoid exposing the menu entry to users who can't benefit from it. Thanks to Takashi for the suggestion. Signed-off-by: Mark Brown <broonie(a)opensource.wolfsonmicro.com> --- sound/soc/codecs/Kconfig | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig index 4975d85..9242889 100644 --- a/sound/soc/codecs/Kconfig +++ b/sound/soc/codecs/Kconfig @@ -70,6 +70,9 @@ config SND_SOC_TLV320AIC23 config SND_SOC_TLV320AIC26 tristate "TI TLV320AIC26 Codec support" depends on SPI + # Note the Kconfig doesn't enforce dependencies - this just + # prevents non-OF users seeing the tristate in menus. + depends on SND_SOC_OF_SIMPLE config SND_SOC_TLV320AIC3X tristate -- 1.5.6.5

3 6

[alsa-devel] OMAP3EVM ASOC TWL4030 Audio drivers
by naveen krishna ch 14 Oct '08

14 Oct '08

Hi All, I am using OMAP3 EVM + TWL4030 combination. The ASOC drivers this combination seems to work half way. My problem is when i do a aplay ***.wav it plays fine but when i try to record through the line in with arecord -c 2 -r 8000 -f 8000 -d 10 -t wav record.wav command it does not record any thing. I have used the Gumstix Overo sources as a reference for making the Asoc sources for this combination. Can any one tell me how can i solve this issue. -- Thanks, keep smiling... keep winning... (: Naveen Krishna Ch :)

1 0

[alsa-devel] [PATCH] hda: fix nid variable warning
by Matthew Ranostay 14 Oct '08

14 Oct '08

Fixed compiler warning with possible uninitialized variable 'nid'. CC [M] /home/mranostay/git/alsa-driver/pci/hda/patch_sigmatel.o /home/mranostay/git/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/patch_sigmatel.c: In function ‘stac92xx_parse_auto_config’: /home/mranostay/git/alsa-driver/pci/hda/../../alsa-kernel/pci/hda/patch_sigmatel.c:2815: warning: ‘nid’ may be used uninitialized in this function Signed-off-by: Matthew Ranostay <mranostay(a)embeddedalley.com> --- diff --git a/pci/hda/patch_sigmatel.c b/pci/hda/patch_sigmatel.c index c461baa..f8bfb4c 100644 --- a/pci/hda/patch_sigmatel.c +++ b/pci/hda/patch_sigmatel.c @@ -2812,7 +2812,7 @@ static int stac92xx_auto_create_multi_out_ctls(struct hda_codec *codec, static const char *chname[4] = { "Front", "Surround", NULL /*CLFE*/, "Side" }; - hda_nid_t nid; + hda_nid_t nid = 0; int i, err; struct sigmatel_spec *spec = codec->spec;

2 1

[alsa-devel] [PATCH] ALSA: ASoC: Fix compile-time warning for tlv320aic23.c
by Arun KS 13 Oct '08

13 Oct '08

Fixes this warning: sound/soc/codecs/tlv320aic23.c: In function 'tlv320aic23_write': sound/soc/codecs/tlv320aic23.c:104: warning: passing argument 2 of 'codec->hw_write' makes pointer from integer without a cast Replaces i2c smbus write function with standard i2c write function Signed-off-by: Arun KS <arunks(a)mistralsolutions.com> --- sound/soc/codecs/tlv320aic23.c | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) diff --git a/sound/soc/codecs/tlv320aic23.c b/sound/soc/codecs/tlv320aic23.c index bac7815..44308da 100644 --- a/sound/soc/codecs/tlv320aic23.c +++ b/sound/soc/codecs/tlv320aic23.c @@ -84,7 +84,7 @@ static int tlv320aic23_write(struct snd_soc_codec *codec, unsigned int reg, unsigned int value) { - u8 data; + u8 data[2]; /* TLV320AIC23 has 7 bit address and 9 bits of data * so we need to switch one data bit into reg and rest @@ -96,12 +96,12 @@ static int tlv320aic23_write(struct snd_soc_codec *codec, unsigned int reg, return -1; } - data = (reg << 1) | (value >> 8 & 0x01); + data[0] = (reg << 1) | (value >> 8 & 0x01); + data[1] = value & 0xff; tlv320aic23_write_reg_cache(codec, reg, value); - if (codec->hw_write(codec->control_data, data, - (value & 0xff)) == 0) + if (codec->hw_write(codec->control_data, data, 2) == 2) return 0; printk(KERN_ERR "%s cannot write %03x to register R%d\n", __func__, @@ -674,7 +674,7 @@ static int tlv320aic23_probe(struct platform_device *pdev) tlv320aic23_socdev = socdev; #if defined(CONFIG_I2C) || defined(CONFIG_I2C_MODULE) - codec->hw_write = (hw_write_t) i2c_smbus_write_byte_data; + codec->hw_write = (hw_write_t) i2c_master_send; codec->hw_read = NULL; ret = i2c_add_driver(&tlv320aic23_i2c_driver); if (ret != 0) -- 1.5.3.4

3 2

[alsa-devel] ASUS V1Sn can have digout
by asto 13 Oct '08

13 Oct '08

Regarding this change: http://git.alsa-project.org/?p=alsa-kernel.git;a=commit;h=2522d7359301efadf… (Added support for Asus V1Sn) This card can have digital output, and indeed does in my situation. The patch above added to patch_realtek.c: SND_PCI_QUIRK(0x1043, 0x1633, "Asus V1Sn", ALC861VD_LENOVO) However, ALC861VD_LENOVO definition has no digout Is it possible to have digital out for this device configured in auto-probe rather than map to lenovo ? For now, I add a definition for it based on the lenovo and 3stack-660-digout: to patch_realtek.c [ALC660VD_V1S] = { .mixers = { alc861vd_3st_mixer }, .init_verbs = { alc861vd_volume_init_verbs, alc861vd_3stack_init_verbs, alc861vd_eapd_verbs, alc861vd_lenovo_unsol_verbs }, .num_dacs = ARRAY_SIZE(alc660vd_dac_nids), .dac_nids = alc660vd_dac_nids, .dig_out_nid = ALC861VD_DIGOUT_NID, .num_channel_mode = ARRAY_SIZE(alc861vd_3stack_2ch_modes), .channel_mode = alc861vd_3stack_2ch_modes, .input_mux = &alc861vd_capture_source, .unsol_event = alc861vd_lenovo_unsol_event, .init_hook = alc861vd_lenovo_automute, },

2 1

[alsa-devel] missing subject in my previous email
by Philippe CARRIERE 13 Oct '08

13 Oct '08

Sorry, I realized I omitted the subject in my previous email. Read ehci-hcd patch required by snd-usb-us122l Regards Phil.

1 0

Re: [alsa-devel] [PATCH] alsa: Emagic Audiowerk8 low-level ALSA driver
by matthias.nyffenegger＠bluewin.ch 13 Oct '08

13 Oct '08

Hi Takashi, Thank you for your quick answer, I am not that fast ;-) >At the next time, it'd be appreciate if you post the patch itself, >then we can review and give comments on the code in a reply. >Even if it's too big, you can split to several pieces. Ok. >Just checking it now, and soon found out that many components look >similar as AW2. We have already AW2 driver (whose authors cc'ed). >Can you reuse the code? Hm, I'm afraid not at his point. The existing AW2 driver is IMHO a very straight-forward implementation - with focus on AW2 - and does not provide generic code parts (apart saa7146 register addr. definitions) for reuse. Maybe that's why it has its own directory not named after the chipset? Besides that I have put much emphasis on providing an abstraction-layer for saa7146 chipset audio functionality, on top of which the audiowerk specific stuff is built. Hence it does not make much sense to me trying to reuse code of the existing AW2 driver. I hope you see my point and would appreciate your opinion on this. regards, Matthias ----Ursprüngliche Nachricht---- Von: tiwai(a)suse.de Datum: 19.09.2008 17:55 An: <matthias.nyffenegger(a)bluewin.ch> Kopie: "Cedric Bregardis"<cedric.bregardis(a)free.fr>, "Jean-Christian Hassler"<jhassler(a)free.fr>, <alsa-devel@alsa- project.org> Betreff: Re: [PATCH] alsa: Emagic Audiowerk8 low-level ALSA driver At Fri, 19 Sep 2008 09:10:00 +0000 (GMT), matthias.nyffenegger(a)bluewin.ch wrote: > > From: Matthias Nyffenegger <matthias.nyffenegger(a)bluewin.ch> > > Low-level ALSA driver for Emagic Audiowerk8 sound card. > Project page: http://sourceforge.net/projects/aw8-alsa First off, thanks for your patch. At the next time, it'd be appreciate if you post the patch itself, then we can review and give comments on the code in a reply. Even if it's too big, you can split to several pieces. > Built and tested with Vanilla 2.6.25.16, ALSA 1.0.14a. > > Signed-off-by: Matthias Nyffenegger <matthias.nyffenegger(a)bluewin.ch> > --- > This is a request for submission to ALSA-tree. > URL: http://downloads.sourceforge.net/aw8-alsa/aw-alsa-patch-0.2 Just checking it now, and soon found out that many components look similar as AW2. We have already AW2 driver (whose authors cc'ed). Can you reuse the code? thanks, Takashi

2 1

Re: [alsa-devel] Backported sbxfi driver (UNTESTED!)
by Takashi Iwai 13 Oct '08

13 Oct '08

At Fri, 10 Oct 2008 20:39:48 +0400, The Source wrote: > > 2008/10/10 Takashi Iwai <tiwai(a)suse.de> > > At Fri, 10 Oct 2008 01:26:15 -0500, > Ted T. Logian wrote: > > > > > > > > On Fri, 2008-10-10 at 08:01 +0200, Takashi Iwai wrote: > > > At Thu, 09 Oct 2008 15:17:58 -0500, > > > Ted T. Logian wrote: > > > > > > > > Sorry for the multiple posts, but once I rmmod snd-sbxfi, running > mixer does > > > > fine. It seems if I have snd-sbxfi loaded and I run alsamixer, it > locks the > > > > entire system. > > > > > > So, do you mean loading the driver itself doesn't lock up? If so, > > > it's better than I expected. > > > > > > Did you run ever PCM playback/capture before that? This is more > > > dangerous :) > > > > > > Also, please give your hardware details, as there are different models > > > for X-Fi, and some are handled pretty differently. > > > > > > [BTW, please stop top-posting, and avoid HTML mails for ML. It's easy > > > to switch to normal (plain) mail mode on Gmail, just a click.] > > > > > > > > > thanks, > > > > > > Takashi > > > > > > I think perhaps from uninstalling pulseaudio I got further. I can use > > mixer now, and it even has a "Master" device, but nothing else... > > > > however, I do get a lock up later. > > > > I get this from running aplay, too, if this helps... > > Thanks! That's rather a stupid bug. > The patch is below. > > Takashi > > diff --git a/sound/pci/sbxfi/sbxfi.c b/sound/pci/sbxfi/sbxfi.c > index 8066bf4..8a0eea0 100644 > --- a/sound/pci/sbxfi/sbxfi.c > +++ b/sound/pci/sbxfi/sbxfi.c > @@ -831,7 +831,7 @@ static struct sbxfi_port *sbxfi_port_alloc(struct > sbxfi *chip, > port->src[0] = src; > port->src[1] = src + 1; > spin_lock_irq(&chip->port_lock); > - list_add(&chip->port_list, &port->list); > + list_add(&port->list, &chip->port_list); > spin_unlock_irq(&chip->port_lock); > return port; > } > _______________________________________________ > Alsa-devel mailing list > Alsa-devel(a)alsa-project.org > http://mailman.alsa-project.org/mailman/listinfo/alsa-devel > > Hello, I have tried the driver with the fix you provided, but I couldn't get > anything. Kernel panic on any attempt to use sound or mixer. Also kernel panic > several seconds after Xorg starts. Each crash causes either hang or reboot so > I couldn't get any error messages. If there are some logs left, please tell me > where they are and I'll send them to you. > My card is X-Fi Platinum Fatality Champion Series. For testing, first try without X but just use Linux console, so that you have a better chance to see Oops messages. It's pretty important to see whether it's an Oops or a panic. Also, double-check whether you really installed the corrected drivers. Often it's missing in the right place by stupid reasons. For first tests, don't use any sound server. Try with aplay like: % aplay -Dhw foo.wav Try first 96kHz samples, 48kHz samples, and then others. thanks, Takashi

5 9