It indeed does fix it. Doesn't seem to break anything along the way. Can you check it in?

2008/8/29 Takashi Iwai tiwai@suse.de:

More to follow, the patch below might fix the problem.

Takashi

diff --git a/sound/pci/ice1712/ice1724.c b/sound/pci/ice1712/ice1724.c index 60119d2..7bb99df 100644 --- a/sound/pci/ice1712/ice1724.c +++ b/sound/pci/ice1712/ice1724.c @@ -942,7 +942,7 @@ static int snd_vt1724_playback_pro_open(struct snd_pcm_substream *substream) { struct snd_pcm_runtime *runtime = substream->runtime; struct snd_ice1712 *ice = snd_pcm_substream_chip(substream);

•   int chs;
  

•   int chs, num_indeps;
  
   runtime->private_data = (void *)&vt1724_playback_pro_reg;
   ice->playback_pro_substream = substream;
  

@@ -952,7 +952,8 @@ static int snd_vt1724_playback_pro_open(struct snd_pcm_substream *substream) set_rate_constraints(ice, substream); mutex_lock(&ice->open_mutex); /* calculate the currently available channels */

•   for (chs = 0; chs < 3; chs++) {
  

•   num_indeps = ice->num_total_dacs / 2 - 1;
  

•   for (chs = 0; chs < num_indeps; chs++) {
           if (ice->pcm_reserved[chs])
                   break;
   }
  
  

At Fri, 29 Aug 2008 18:44:30 +0200, =?UTF-8?Q?Vedran_Mileti=C4=87?= wrote:

Wait, this isn't the one, my first mail is wrong. Need to figure out how git generates my mail and how to change that, take this one instead.

OK, I applied this now. Thanks!

Takashi

2008/8/29 Vedran Miletić rivanvx@gmail.com:

...

You guessed right :))

Well, this time... I also fixed lines longer than 80 chars along the way.

2008/8/29 Takashi Iwai tiwai@suse.de:

...

At Thu, 28 Aug 2008 18:50:15 +0200, =?UTF-8?Q?Vedran_Mileti=C4=87?= wrote:

...

Hopefully this is it.

Sorry, no. It's broken. I guess you edited the patch manually? Regenerate the patch via git diff please.

...

BTW, while at it, let's fix one more thing. In jack, this card (PHASE 22) shows playback_1, playback_2, playback_3...playback_8. Last 6 are useless because this card has only two analog outputs. They really don't output anywhere. I would say that it's probably that PHASE 22 and PHASE 28 share that part of code, while they shouldn't.

Can you point to the code which creates those devices?

ice1724.c.

thanks,

Takashi

-- Vedran Miletić

-- Vedran Miletić [2 ALSA-ice1724-Enable-MIDI-on-TerraTec-PHASE-22-and-28.patch <text/x-diff (base64)>] From afcb9c30edc69513c8f02c5d512836a1081f2d7a Mon Sep 17 00:00:00 2001 From: Vedran Miletic rivanvx@gmail.com Date: Fri, 29 Aug 2008 18:31:13 +0200 Subject: [PATCH] ALSA: ice1724: Enable MIDI on TerraTec PHASE 22 and PHASE 28

Even though MIDI was fixed on ice1724 chips a while ago, it wasn't yet enabled for some cards as it didn't get enough testing. This was tested with MIDI keyboard on PHASE 22 and with looping back output to input and it works stable, so it's safe to enable it.

Besides this, there are some more minor fixes, not exactly user visible:

• added info about PHASE 28 (collected, as I don't have a card)
• added info about TS22PCI and new revisions of PHASE 22
• disable 192k on PHASE 22 as AK4524 I2S doesn't support it
• enable SPDIF reciever on PHASE 22

Signed-off-by: Vedran Miletic rivanvx@gmail.com

pci/ice1712/phase.c | 23 +++++++++++++++++------ 1 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/pci/ice1712/phase.c b/pci/ice1712/phase.c index f5acdee..6a61472 100644 --- a/pci/ice1712/phase.c +++ b/pci/ice1712/phase.c @@ -22,15 +22,24 @@ */

/* PHASE 22 overview:

• • Audio controller: VIA Envy24HT-S (slightly trimmed down version of Envy24HT)

• • Audio controller: VIA Envy24HT-S (slightly trimmed down Envy24HT, 4in/4out)
  • Analog chip: AK4524 (partially via Philip's 74HCT125)

• • Digital receiver: CS8414-CS (not supported in this release)

• • Digital receiver: CS8414-CS (supported in this release)

• • PHASE 22 revision 2.0 and Terrasoniq/Musonik TS22PCI have CS8416
    

• • (support status unknown, please test and report)
    

  • Envy connects to AK4524
  • • CS directly from GPIO 10
  • • CCLK via 74HCT125's gate #4 from GPIO 4
  • • CDTI via 74HCT125's gate #2 from GPIO 5

• • CDTI may be completely blocked by 74HCT125's gate #1 controlled by GPIO 3
    

• • CDTI may be completely blocked by 74HCT125's gate #1
    

• • controlled by GPIO 3
    

• */

+/* PHASE 28 overview:

• • Audio controller: VIA Envy24HT (full untrimmed version, 8in/8out)
• • Analog chip: WM8770 (8 channel 192k DAC, 2 channel 96k ADC)
• • Digital receiver: CS8414-CS (supported in this release)
  */

#include <asm/io.h> @@ -161,9 +170,10 @@ static int __devinit phase22_add_controls(struct snd_ice1712 *ice) }

static unsigned char phase22_eeprom[] __devinitdata = {

• [ICE_EEP2_SYSCONF] = 0x00, /* 1xADC, 1xDACs */

• [ICE_EEP2_SYSCONF] = 0x28, /* clock 512, mpu 401,

• 			spdif-in/1xADC, 1xDACs */
  

  [ICE_EEP2_ACLINK] = 0x80, /* I2S */

• [ICE_EEP2_I2S] = 0xf8, /* vol, 96k, 24bit */

• [ICE_EEP2_I2S] = 0xf0, /* vol, 96k, 24bit */ [ICE_EEP2_SPDIF] = 0xc3, /* out-en, out-int, spdif-in */ [ICE_EEP2_GPIO_DIR] = 0xff, [ICE_EEP2_GPIO_DIR1] = 0xff,

@@ -177,7 +187,8 @@ static unsigned char phase22_eeprom[] __devinitdata = { };

static unsigned char phase28_eeprom[] __devinitdata = {

• [ICE_EEP2_SYSCONF] = 0x0b, /* clock 512, spdif-in/ADC, 4DACs */

• [ICE_EEP2_SYSCONF] = 0x2b, /* clock 512, mpu401,

• 			spdif-in/1xADC, 4xDACs */
  

  [ICE_EEP2_ACLINK] = 0x80, /* I2S */ [ICE_EEP2_I2S] = 0xfc, /* vol, 96k, 24bit, 192k */ [ICE_EEP2_SPDIF] = 0xc3, /* out-en, out-int, spdif-in */

-- 1.5.6.5