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

Re: [alsa-devel] Bug in sound-unstable-2.6 ice1724
by Vedran Miletić 02 Feb '09

02 Feb '09

On Thu, Jan 1, 2009 at 10:27 AM, Takashi Iwai <tiwai(a)suse.de> wrote: > At Fri, 26 Dec 2008 21:34:47 +0100, > =?UTF-8?Q?Vedran_Mileti=C4=87?= wrote: >> >> On Wed, Nov 26, 2008 at 4:51 PM, Takashi Iwai <tiwai(a)suse.de> wrote: >> > Doh, I forgot to attach the patch. The below is the one. >> > >> > Please, if anyone can test it, let me know whether it works. >> > This would be nice to have in the next kernel. >> > >> > >> > thanks, >> > >> > Takashi >> > >> >> Do you still need this to be tested? I can do it. > > Yes, please. > > > Takashi > [vedran@phenomche ~]$ aplaymidi -l Port Client name Port name 14:0 Midi Through Midi Through Port-0 16:0 SB Live! 5.1 Dell OEM [SB0228] EMU10K1 MPU-401 (UART) 17:0 Emu10k1 WaveTable Emu10k1 Port 0 17:1 Emu10k1 WaveTable Emu10k1 Port 1 17:2 Emu10k1 WaveTable Emu10k1 Port 2 17:3 Emu10k1 WaveTable Emu10k1 Port 3 20:0 Terratec PHASE 22 ICE1724 MIDI [vedran@phenomche ~]$ aplaymidi -p 20:0 blah.mid [vedran@phenomche ~]$ cat /proc/asound/card1/midi0 ICE1724 MIDI Output 0 Tx bytes : 4379 Mode : native Buffer size : 4096 Avail : 4096 Input 0 Rx bytes : Works without problems. I'm using yestarday's sound-unstable git snapshot. -- Vedran Miletić

2 1

[alsa-devel] Confusion about whether snd_pcm_poll_descriptors_revents()'s revents field is a single integer or an array
by Lennart Poettering 02 Feb '09

02 Feb '09

Heya! If we look into the pcm example how snd_pcm_poll_descriptors_revents() is used then we can see that the revents parameter apparently is supposed to be a single integer. (which makes a lot of sense to me) http://www.alsa-project.org/alsa-doc/alsa-lib/_2test_2pcm_8c-example.html#a… However, snd_pcm_wait_nocheck() calls the same function and assumes it is a complete array! http://git.alsa-project.org/?p=alsa-lib.git;a=blob;f=src/pcm/pcm.c;h=74d1d1… So what's it now? It makes more sense to me if it would be a single fd. The problem becomnes visible if someone writes a plugin that wants wakeups from more than one fd because either snd_pcm_wait() using applications start to act weirdly because the revents are not fully initialized or snd_pcm_poll_descriptors_revents() using applications get invalid memory accesses. This has become visible in the Bluetooth plugin. I guess nobody has written a plugin that has more than one fd to sleep on yet, right? Lennart -- Lennart Poettering Red Hat, Inc. lennart [at] poettering [dot] net ICQ# 11060553 http://0pointer.net/lennart/ GnuPG 0x1A015CC4

3 10

[alsa-devel] Dell Latitude D830 internal speaker volume degraded on 2.6.29-rc*
by Ben Gamari 02 Feb '09

02 Feb '09

Hey all, For quite a while now, the internal speaker volume on my Dell Latitude D830 has been very soft. While usually the case can be made to resonate with all mixers at 100%, recently I have noticed a substantial decrease in output. Figuring this must be caused by a codec mis-configuration, I booted into a working kernel 2.6.27.9-134.fc10.x86_64 and collected a dump from /proc/asound/card0/codec#0. After booting back into my own vanilla kernel from master, I collected another trace. When I was looking through the diff -u of the two dumps, one particular difference seemed notable, Node 0x11 [Audio Output] wcaps 0xd0c05: Stereo Amp-Out R/L Amp-Out caps: N/A - Amp-Out vals: [0x7f 0x7f] - Converter: stream=5, channel=0 + Amp-Out vals: [0xff 0xff] + Converter: stream=0, channel=0 Power: setting=D0, actual=D0 Delay: 13 samples Perhaps this is significant? Regardless, I have attached both of the original dumps. Let me know if anything else is needed. Thanks, - Ben

2 3

[alsa-devel] Enumerating PCM devices
by Bankim Bhavsar 02 Feb '09

02 Feb '09

Hi Alsa developers/users, For some reason my message got truncated earlier. So here I go again. I'm using snd_device_name_hint("pcm"..) and then snd_device_name_get_hint(.."NAME") to get the NAME parameter of pcm devices. Is it true that all sound cards that support Alsa will *always* have "default" pcm device? I'm more inclined towards using "default:CARD=<vendor>" as the parameter to snd_pcm_open(). This way if the program configures hardware params for any number of channels then the appropriate PCM device will be selected on the sound card, correct? Thanks, Bankim.

2 1

[alsa-devel] snd-hda-intel option for Toshiba A200-1MB
by Vivia Nikolaidou 02 Feb '09

02 Feb '09

Dear ALSA developers, I recently installed linux on a Toshiba Satellite A200-1MB laptop. Its sound card is ALC861VD, using the hda_intel module. The problem was that the earphone jack sense wasn't working. It turns out that I had to use the Lenovo option. It would be nice if you could add this model (and probably other A200 models as well) to ALSA-Configuration.txt . Thank you, Vivia Nikolaidou

2 1

[alsa-devel] ASoC updates for 2.6.30
by Mark Brown 02 Feb '09

02 Feb '09

The following changes since commit 7393958f630ac91e591e62058f2bdb61523ec60c: Peter Ujfalusi (1): ASoC: TWL4030: Add analog loopback support are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound-2.6.git for-2.6.30 Eero Nurkkala (1): OMAP: ASoC: Fix spinlock misuse in omap-pcm.c Grazvydas Ignotas (1): ASoC: Update OMAP3 pandora board file Mark Brown (1): Merge branch 'for-2.6.29' into for-2.6.30 Timur Tabi (2): ASoC: fix initialization order of the CS4270 codec driver ASoC: fix documentation in CS4270 codec driver sound/soc/codecs/cs4270.c | 332 ++++++++++++++++++++++++----------------- sound/soc/omap/omap-pcm.c | 5 +- sound/soc/omap/omap3pandora.c | 49 ++++--- 3 files changed, 230 insertions(+), 156 deletions(-)

2 1

[alsa-devel] ASoC update for 2.6.29
by Mark Brown 02 Feb '09

02 Feb '09

The following changes since commit ef390c0b6e3f4d2d2d43f53f4bd35e1884571a14: Misael Lopez Cruz (1): ASoC: OMAP: Initialize XCCR and RCCR registers in McBSP DAI driver are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound-2.6.git for-2.6.29 Eero Nurkkala (1): OMAP: ASoC: Fix spinlock misuse in omap-pcm.c sound/soc/omap/omap-pcm.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-)

2 1

[alsa-devel] [PATCH] msnd: add module description and license for the snd-msnd-lib
by Krzysztof Helt 02 Feb '09

02 Feb '09

From: Krzysztof Helt <krzysztof.h1(a)wp.pl> The missing module license generates warning during module loading. Signed-off-by: Krzysztof Helt <krzysztof.h1(a)wp.pl> --- sound/isa/msnd/msnd.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/sound/isa/msnd/msnd.c b/sound/isa/msnd/msnd.c index 264e082..9064544 100644 --- a/sound/isa/msnd/msnd.c +++ b/sound/isa/msnd/msnd.c @@ -700,3 +700,6 @@ int snd_msnd_pcm(struct snd_card *card, int device, } EXPORT_SYMBOL(snd_msnd_pcm); +MODULE_DESCRIPTION("Common routines for Turtle Beach Multisound drivers"); +MODULE_LICENSE("GPL"); + -- 1.5.2.2 ---------------------------------------------------------------------- Konkurs Walentynkowy! Wybierz nagrode i wygraj! Sprawdz >> http://link.interia.pl/f2039

2 1

[alsa-devel] [PATCH] sscape: drop redundant fields from soundscape struct
by Krzysztof Helt 02 Feb '09

02 Feb '09

From: Krzysztof Helt <krzysztof.h1(a)wp.pl> The wss_base is disguised parameter for one function. It is converted to a function parameter. The code_type is only set but never read. It is removed. The midi_vol is set only to 0 so it does not work as detection of change in midi volume. It is removed. The xport variable is alias to the port[dev]. Use the port[dev] directly to increase readability. Signed-off-by: Krzysztof Helt <krzysztof.h1(a)wp.pl> --- sound/isa/sscape.c | 42 ++++++++++++++---------------------------- 1 files changed, 14 insertions(+), 28 deletions(-) diff --git a/sound/isa/sscape.c b/sound/isa/sscape.c index 681e223..c6a90f0 100644 --- a/sound/isa/sscape.c +++ b/sound/isa/sscape.c @@ -135,8 +135,6 @@ enum card_type { struct soundscape { spinlock_t lock; unsigned io_base; - unsigned wss_base; - int codec_type; int ic_type; enum card_type type; struct resource *io_res; @@ -152,7 +150,6 @@ struct soundscape { spinlock_t fwlock; int hw_in_use; unsigned long midi_usage; - unsigned char midi_vol; }; #define INVALID_IRQ ((unsigned)-1) @@ -760,14 +757,9 @@ static int sscape_midi_put(struct snd_kcontrol *kctl, * and then perform another volume-related command. Perhaps the * first command is an "open" and the second command is a "close"? */ - if (s->midi_vol == ((unsigned char) uctl->value.integer. value[0] & 127)) { - change = 0; - goto __skip_change; - } change = (host_write_ctrl_unsafe(s->io_base, CMD_SET_MIDI_VOL, 100) && host_write_ctrl_unsafe(s->io_base, ((unsigned char) uctl->value.integer. value[0]) & 127, 100) && host_write_ctrl_unsafe(s->io_base, CMD_XXX_MIDI_VOL, 100)); - __skip_change: /* * Take the board out of HOST mode and back into MIDI mode ... @@ -809,12 +801,11 @@ static unsigned __devinit get_irq_config(int irq) * Perform certain arcane port-checks to see whether there * is a SoundScape board lurking behind the given ports. */ -static int __devinit detect_sscape(struct soundscape *s) +static int __devinit detect_sscape(struct soundscape *s, long wss_io) { unsigned long flags; unsigned d; int retval = 0; - int codec = s->wss_base; spin_lock_irqsave(&s->lock, flags); @@ -830,13 +821,11 @@ static int __devinit detect_sscape(struct soundscape *s) if ((d & 0x80) != 0) goto _done; - if (d == 0) { - s->codec_type = 1; + if (d == 0) s->ic_type = IC_ODIE; - } else if ((d & 0x60) != 0) { - s->codec_type = 2; + else if ((d & 0x60) != 0) s->ic_type = IC_OPUS; - } else + else goto _done; outb(0xfa, ODIE_ADDR_IO(s->io_base)); @@ -856,10 +845,10 @@ static int __devinit detect_sscape(struct soundscape *s) sscape_write_unsafe(s->io_base, GA_HMCTL_REG, d | 0xc0); if (s->type == SSCAPE_VIVO) - codec += 4; + wss_io += 4; /* wait for WSS codec */ for (d = 0; d < 500; d++) { - if ((inb(codec) & 0x80) == 0) + if ((inb(wss_io) & 0x80) == 0) break; spin_unlock_irqrestore(&s->lock, flags); msleep(1); @@ -1057,7 +1046,6 @@ static int __devinit create_sscape(int dev, struct snd_card *card) unsigned dma_cfg; unsigned irq_cfg; unsigned mpu_irq_cfg; - unsigned xport; struct resource *io_res; struct resource *wss_res; unsigned long flags; @@ -1077,15 +1065,15 @@ static int __devinit create_sscape(int dev, struct snd_card *card) printk(KERN_ERR "sscape: Invalid IRQ %d\n", mpu_irq[dev]); return -ENXIO; } - xport = port[dev]; /* * Grab IO ports that we will need to probe so that we * can detect and control this hardware ... */ - io_res = request_region(xport, 8, "SoundScape"); + io_res = request_region(port[dev], 8, "SoundScape"); if (!io_res) { - snd_printk(KERN_ERR "sscape: can't grab port 0x%x\n", xport); + snd_printk(KERN_ERR + "sscape: can't grab port 0x%lx\n", port[dev]); return -EBUSY; } wss_res = NULL; @@ -1112,10 +1100,9 @@ static int __devinit create_sscape(int dev, struct snd_card *card) spin_lock_init(&sscape->fwlock); sscape->io_res = io_res; sscape->wss_res = wss_res; - sscape->io_base = xport; - sscape->wss_base = wss_port[dev]; + sscape->io_base = port[dev]; - if (!detect_sscape(sscape)) { + if (!detect_sscape(sscape, wss_port[dev])) { printk(KERN_ERR "sscape: hardware not detected at 0x%x\n", sscape->io_base); err = -ENODEV; goto _release_dma; @@ -1188,11 +1175,11 @@ static int __devinit create_sscape(int dev, struct snd_card *card) } #define MIDI_DEVNUM 0 if (sscape->type != SSCAPE_VIVO) { - err = create_mpu401(card, MIDI_DEVNUM, xport, mpu_irq[dev]); + err = create_mpu401(card, MIDI_DEVNUM, port[dev], mpu_irq[dev]); if (err < 0) { printk(KERN_ERR "sscape: Failed to create " - "MPU-401 device at 0x%x\n", - xport); + "MPU-401 device at 0x%lx\n", + port[dev]); goto _release_dma; } @@ -1204,7 +1191,6 @@ static int __devinit create_sscape(int dev, struct snd_card *card) /* * Initialize mixer */ - sscape->midi_vol = 0; host_write_ctrl_unsafe(sscape->io_base, CMD_SET_MIDI_VOL, 100); host_write_ctrl_unsafe(sscape->io_base, 0, 100); host_write_ctrl_unsafe(sscape->io_base, CMD_XXX_MIDI_VOL, 100); -- 1.5.2.2 ---------------------------------------------------------------------- Doladuj telefon! Sprawdz >> http://link.interia.pl/f2047

2 1

[alsa-devel] Recalcitrant input pin on ALC262
by Tony Vroon 02 Feb '09

02 Feb '09

Good evening, Currently I am implementing full support for the Tyan Thunder n6650W and have noticed a pin complex that advertises input, but can't seem to be set to it: Node 0x1c [Pin Complex] wcaps 0x400001: Stereo Pincap 0x00000020: IN Pin Default 0x411111f0: [N/A] Speaker at Ext Rear Conn = 1/8, Color = Black DefAssociation = 0xf, Sequence = 0x0 Misc = NO_PRESENCE Pin-ctls: 0x00: As you can see the pin capabilities reflect input. However, even after issuing this in the model-specific verbs: {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, the pin-ctls firmly remains 0x00. Even the python-based hda-analyzer appears unable to flip the pin-ctls over to 0x20 and reports no error messages in the GUI or in the terminal. What am I missing please? Slightly related, is it considered acceptable to modify pin defaults if they are blatantly wrong? (Reported colour, N/A vs Fixed, etc) Regards, Tony V.

2 1