Skip to contentSkip to conversation

ice1712: fix: lock samplerate when samplerate locking is enabled.

PATCH In review Sebastien Alaiwan 2 participants 5 messages Feb 3, 2010 – Feb 5, 2010 Sound

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ice1712: fix: lock samplerate when samplerate locking is enabled.

Hello,

I found that the sampling rate locking setting of the ice1712 sound driver was only half-respected : when the driver was locked to, let's say, 44100Hz, and a usermode app was requesting 48000Hz playback, the request was succesful although the soundcard would continue to run at 44100Hz.

Here's a patch that will make those requests to fail.

sound/pci/ice1712/ice1712.c +5 -0
index c7cff6f..4775626 100644
@@ -1181,6 +1181,8 @@ static int snd_ice1712_playback_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
 
+	if (is_pro_rate_locked(ice))
+		snd_pcm_hw_constraint_minmax(runtime, SNDRV_PCM_HW_PARAM_RATE, PRO_RATE_DEFAULT, PRO_RATE_DEFAULT);
 	if (ice->spdif.ops.open)
 		ice->spdif.ops.open(ice, substream);
 
@@ -1197,6 +1199,9 @@ static int snd_ice1712_capture_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+		snd_pcm_hw_constraint_minmax(runtime, SNDRV_PCM_HW_PARAM_RATE, PRO_RATE_DEFAULT, PRO_RATE_DEFAULT);
+
 	return 0;
 }
 
#1 lore raw

At Wed, 03 Feb 2010 21:45:34 +0100,

Quoted 9 lines · Sebastien Alaiwan

Hello,

I found that the sampling rate locking setting of the ice1712 sound driver was only half-respected : when the driver was locked to, let's say, 44100Hz, and a usermode app was requesting 48000Hz playback, the request was succesful although the soundcard would continue to run at 44100Hz.

Here's a patch that will make those requests to fail.

Thanks, the patch looks OK. But changing runtime->hw.rate_min and rate_max would be cheaper. It won't involve with kmalloc(). In theory, snd_pcm_hw_constraint*() can fail.

Could you fix in that way and give your sign-off to merge to the upstream? (And don't forget to add me to Cc.)

thanks,

Takashi

Quoted 29 lines
sound/pci/ice1712/ice1712.c +5 -1
index c7cff6f..4775626 100644
@@ -1181,6 +1181,8 @@ static int snd_ice1712_playback_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
 
+	if (is_pro_rate_locked(ice))
+		snd_pcm_hw_constraint_minmax(runtime, SNDRV_PCM_HW_PARAM_RATE, PRO_RATE_DEFAULT, PRO_RATE_DEFAULT);
 	if (ice->spdif.ops.open)
 		ice->spdif.ops.open(ice, substream);
 
@@ -1197,6 +1199,9 @@ static int snd_ice1712_capture_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+		snd_pcm_hw_constraint_minmax(runtime, SNDRV_PCM_HW_PARAM_RATE, PRO_RATE_DEFAULT, PRO_RATE_DEFAULT);
+
 	return 0;
 }
 
--

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#2 lore raw
Quoted 9 lines · Takashi Iwai

At Wed, 03 Feb 2010 21:45:34 +0100,

Quoted 7 lines · Sebastien Alaiwan

Hello,

I found that the sampling rate locking setting of the ice1712 sound driver was only half-respected : when the driver was locked to, let's say, 44100Hz, and a usermode app was requesting 48000Hz playback, the request was succesful although the soundcard would continue to run at 44100Hz.

Quoted 8 lines
Quoted 1 line

Here's a patch that will make those requests to fail.

Thanks, the patch looks OK. But changing runtime->hw.rate_min and rate_max would be cheaper. It won't involve with kmalloc(). In theory, snd_pcm_hw_constraint*() can fail.

Could you fix in that way and give your sign-off to merge to the upstream? (And don't forget to add me to Cc.)

Thank you for reviewing my patch. Here's an updated version according to your advice. You can merge to the upstream if it's okay.

Sebastien Alaiwan

------

sound/pci/ice1712/ice1712.c +11 -0
index c7cff6f..319df17 100644
@@ -1180,6 +1180,11 @@ static int snd_ice1712_playback_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+	{
+		runtime->hw.rate_min = PRO_RATE_DEFAULT;
+		runtime->hw.rate_max = PRO_RATE_DEFAULT;
+	}
 
 	if (ice->spdif.ops.open)
 		ice->spdif.ops.open(ice, substream);
@@ -1197,6 +1202,12 @@ static int snd_ice1712_capture_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+	{
+		runtime->hw.rate_min = PRO_RATE_DEFAULT;
+		runtime->hw.rate_max = PRO_RATE_DEFAULT;
+	}
+
 	return 0;
 }
 
#3 lore raw

Sebastien Alaiwan wrote:

Quoted 9 lines · Takashi Iwai

At Wed, 03 Feb 2010 21:45:34 +0100,

Quoted 7 lines · Sebastien Alaiwan

Hello,

I found that the sampling rate locking setting of the ice1712 sound driver was only half-respected : when the driver was locked to, let's say, 44100Hz, and a usermode app was requesting 48000Hz playback, the request was succesful although the soundcard would continue to run at 44100Hz.

Quoted 7 lines
Quoted 1 line

Here's a patch that will make those requests to fail.

Thanks, the patch looks OK. But changing runtime->hw.rate_min and rate_max would be cheaper. It won't involve with kmalloc(). In theory, snd_pcm_hw_constraint*() can fail.

Could you fix in that way and give your sign-off to merge to the upstream? (And don't forget to add me to Cc.)

Thank you for reviewing my patch. Here's an updated version according to your advice. You can merge to the upstream if it's okay. Sorry for the double mail, here's a signed-off version.

------

sound/pci/ice1712/ice1712.c +11 -0
index c7cff6f..319df17 100644
@@ -1180,6 +1180,11 @@ static int snd_ice1712_playback_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+	{
+		runtime->hw.rate_min = PRO_RATE_DEFAULT;
+		runtime->hw.rate_max = PRO_RATE_DEFAULT;
+	}
 
 	if (ice->spdif.ops.open)
 		ice->spdif.ops.open(ice, substream);
@@ -1197,6 +1202,12 @@ static int snd_ice1712_capture_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+	{
+		runtime->hw.rate_min = PRO_RATE_DEFAULT;
+		runtime->hw.rate_max = PRO_RATE_DEFAULT;
+	}
+
 	return 0;
 }
 
#4 lore raw

At Thu, 04 Feb 2010 22:20:11 +0100,

Quoted 17 lines · Sebastien Alaiwan

Sebastien Alaiwan wrote:

Quoted 14 lines · Takashi Iwai

At Wed, 03 Feb 2010 21:45:34 +0100,

Quoted 8 lines · Sebastien Alaiwan

Hello,

I found that the sampling rate locking setting of the ice1712 sound driver was only half-respected : when the driver was locked to, let's say, 44100Hz, and a usermode app was requesting 48000Hz playback, the request was succesful although the soundcard would continue to run at 44100Hz.

Here's a patch that will make those requests to fail.

Thanks, the patch looks OK. But changing runtime->hw.rate_min and rate_max would be cheaper. It won't involve with kmalloc(). In theory, snd_pcm_hw_constraint*() can fail.

Quoted 8 lines
Quoted 2 lines

Could you fix in that way and give your sign-off to merge to the upstream? (And don't forget to add me to Cc.)

Thank you for reviewing my patch. Here's an updated version according to your advice. You can merge to the upstream if it's okay. Sorry for the double mail, here's a signed-off version.

Thanks! Applied now (with minor coding-style fixes).

Takashi

Quoted 31 lines

------

sound/pci/ice1712/ice1712.c +11 -0
index c7cff6f..319df17 100644
@@ -1180,6 +1180,11 @@ static int snd_ice1712_playback_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+	{
+		runtime->hw.rate_min = PRO_RATE_DEFAULT;
+		runtime->hw.rate_max = PRO_RATE_DEFAULT;
+	}
 
 	if (ice->spdif.ops.open)
 		ice->spdif.ops.open(ice, substream);
@@ -1197,6 +1202,12 @@ static int snd_ice1712_capture_pro_open(struct snd_pcm_substream *substream)
 	snd_pcm_set_sync(substream);
 	snd_pcm_hw_constraint_msbits(runtime, 0, 32, 24);
 	snd_pcm_hw_constraint_list(runtime, 0, SNDRV_PCM_HW_PARAM_RATE, &hw_constraints_rates);
+	if (is_pro_rate_locked(ice))
+	{
+		runtime->hw.rate_min = PRO_RATE_DEFAULT;
+		runtime->hw.rate_max = PRO_RATE_DEFAULT;
+	}
+
 	return 0;
 }