Skip to contentSkip to conversation

Crash 2.6.32.x, Echo 3G, Alsa, SCSI

DISCUSSION Frederick V Heitkamp 5 participants 14 messages Feb 2, 2010 – Jun 18, 2010 Sound

lore
Share & tools
lore
#0 lore raw

Repeatable Hard Crash.
What else do I need to provide?
Thanks!

Fred

[ 5484.995249] WriteControlReg: not written, no change
[ 5485.069621] divide error: 0000 [#1] pcm_hw_params ok
[ 5485.070159] Prepare rate=44100 format=2 channels=2
[ 5485.070161] set_audio_format[14] = 5
[ 5485.070166] Prepare rate=44100 format=2 channels=2
[ 5485.070167] set_audio_format[14] = 5
[ 5485.070003] PREEMPT SMP 
[ 5485.070003] last sysfs file: /sys/devices/pci0000:00/0000:00:1e.0/0000:0a:01.0/sound/card0/uevent
[ 5485.070003] Modules linked in: radeon ttm drm_kms_helper i2c_algo_bit jfs dme1737 hwmon_vid coretemp ipv6 snd_hda_codec_atihdmi snd_hda_intel snd_echo3g snd_hda_codec snd_rawmidi snd_seq_device snd_hwdep snd_pcm snd_timer snd st i3000_edac rtc_cmos parport_pc soundcore shpchp parport rtc_core snd_page_alloc i2c_i801 rtc_lib sr_mod pci_hotplug usblp container edac_core cdrom rng_core joydev sg evdev firewire_sbp2 firewire_ohci firewire_core crc_itu_t hangcheck_timer drm cryptoloop loop video output thermal processor fan button microcode cpuid usbhid uhci_hcd ehci_hcd
[ 5485.070003] 
[ 5485.070003] Pid: 5871, comm: rsync Not tainted (2.6.32.3 #14) empty
[ 5485.070003] EIP: 0060:[<f87aa037>] EFLAGS: 00010246 CPU: 0
[ 5485.070003] EIP is at pcm_pointer+0x37/0x70 [snd_echo3g]
[ 5485.070003] EAX: 00000000 EBX: 00000000 ECX: 00000000 EDX: 00000000
[ 5485.070003] ESI: f695e5c0 EDI: 00000000 EBP: cd11d2c0 ESP: eb39ff04
[ 5485.070003]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[ 5485.070003] Process rsync (pid: 5871, ti=eb39e000 task=f71b6180 task.ti=eb39e000)
[ 5485.070003] Stack:
[ 5485.070003]  c104764e 0000000e f695e5c0 f6b20000 00000000 f87ac9bf 00000024 00000009
[ 5485.070003] <0> f6a7f400 00000246 00000000 f66068c0 00000000 00000000 00000000 c107a5d5
[ 5485.070003] <0> c14f4a24 c107e375 c103fe7a 00000010 17a61f80 000004fd c14f5380 c14f53bc
[ 5485.070003] Call Trace:
[ 5485.070003]  [<c104764e>] ? run_timer_softirq+0x17e/0x2e0
[ 5485.070003]  [<f87ac9bf>] ? snd_echo_interrupt+0x11f/0x240 [snd_echo3g]
[ 5485.070003]  [<c107a5d5>] ? handle_IRQ_event+0x45/0x190
[ 5485.070003]  [<c107e375>] ? rcu_process_callbacks+0x35/0x40
[ 5485.070003]  [<c103fe7a>] ? __do_softirq+0xba/0x200
[ 5485.070003]  [<c107c235>] ? handle_fasteoi_irq+0x65/0xd0
[ 5485.070003]  [<c1005955>] ? handle_irq+0x15/0x30
[ 5485.070003]  [<c1004ec7>] ? do_IRQ+0x47/0xc0
[ 5485.070003]  [<c10401a5>] ? irq_exit+0x35/0x50
[ 5485.070003]  [<c1019584>] ? smp_apic_timer_interrupt+0x54/0x90
[ 5485.070003]  [<c10035e9>] ? common_interrupt+0x29/0x30
[ 5485.070003] Code: 40 70 8b a8 ac 00 00 00 8b 45 00 8b 00 89 c1 2b 4d 04 03 4d 08 89 45 04 89 4d 08 8b 46 70 8b 58 4c 8b 78 40 8d 04 cd 00 00 00 00 <f7> f3 39 c7 77 2a 89 fa 29 f8 f7 da 89 14 24 eb 08 8b 46 70 8b 
[ 5485.070003] EIP: [<f87aa037>] pcm_pointer+0x37/0x70 [snd_echo3g] SS:ESP 0068:eb39ff04
[ 5485.785994] ---[ end trace 6967d8f15086f11c ]---
[ 5485.799787] Kernel panic - not syncing: Fatal exception in interrupt
[ 5485.818780] Pid: 5871, comm: rsync Tainted: G      D    2.6.32.3 #14
[ 5485.837766] Call Trace:
[ 5485.845052]  [<c13c249e>] ? panic+0x4d/0xf1
[ 5485.857547]  [<c1006a8e>] ? oops_end+0x8e/0x90
[ 5485.870820]  [<c100423c>] ? do_divide_error+0x7c/0x90
[ 5485.885922]  [<f87aa037>] ? pcm_pointer+0x37/0x70 [snd_echo3g]
[ 5485.903349]  [<c1009268>] ? sched_clock+0x8/0x10
[ 5485.917147]  [<f870f039>] ? snd_pcm_update_hw_ptr_pos+0x19/0xd0 [snd_pcm]
[ 5485.937433]  [<c105d1d0>] ? ktime_get_ts+0xd0/0x100
[ 5485.951994]  [<c13c52a6>] ? error_code+0x66/0x6c
[ 5485.965788]  [<c10041c0>] ? do_divide_error+0x0/0x90
[ 5485.980619]  [<f87aa037>] ? pcm_pointer+0x37/0x70 [snd_echo3g]
[ 5485.998047]  [<c104764e>] ? run_timer_softirq+0x17e/0x2e0
[ 5486.014175]  [<f87ac9bf>] ? snd_echo_interrupt+0x11f/0x240 [snd_echo3g]
[ 5486.033940]  [<c107a5d5>] ? handle_IRQ_event+0x45/0x190
[ 5486.049537]  [<c107e375>] ? rcu_process_callbacks+0x35/0x40
[ 5486.066169]  [<c103fe7a>] ? __do_softirq+0xba/0x200
[ 5486.080727]  [<c107c235>] ? handle_fasteoi_irq+0x65/0xd0
[ 5486.096584]  [<c1005955>] ? handle_irq+0x15/0x30
[ 5486.110360]  [<c1004ec7>] ? do_IRQ+0x47/0xc0
[ 5486.123098]  [<c10401a5>] ? irq_exit+0x35/0x50
[ 5486.136357]  [<c1019584>] ? smp_apic_timer_interrupt+0x54/0x90
[ 5486.153770]  [<c10035e9>] ? common_interrupt+0x29/0x30
#1 lore raw
Quoted 18 lines · Frederick V Heitkamp

Repeatable Hard Crash.
What else do I need to provide?

[ 5484.995249] WriteControlReg: not written, no change
[ 5485.069621] divide error: 0000 [#1] pcm_hw_params ok
[ 5485.070159] Prepare rate=44100 format=2 channels=2
[ 5485.070161] set_audio_format[14] = 5
[ 5485.070166] Prepare rate=44100 format=2 channels=2
[ 5485.070167] set_audio_format[14] = 5
[ 5485.070003] PREEMPT SMP 
...
[ 5485.070003] EIP is at pcm_pointer+0x37/0x70 [snd_echo3g]
...
[ 5485.965788]  [<c10041c0>] ? do_divide_error+0x0/0x90
[ 5485.980619]  [<f87aa037>] ? pcm_pointer+0x37/0x70 [snd_echo3g]
[ 5485.998047]  [<c104764e>] ? run_timer_softirq+0x17e/0x2e0
[ 5486.014175]  [<f87ac9bf>] ? snd_echo_interrupt+0x11f/0x240 [snd_echo3g]
[ 5486.033940]  [<c107a5d5>] ? handle_IRQ_event+0x45/0x190

bytes_to_frames() divides by runtime->frame_bits which is not set until after the hw_params callback has succeeded, but the corresponding chip->substream[] entry is set in that callback, by init_engine(). It should probably have been set in the prepare callback.

Frederick, please try the patch below.

Takashi, is there any reason why the runtime-> fields are initialized after the hw_params callback?

Regards,
Clemens

sound/core/pcm_native.c +7 -6
@@ -403,12 +403,6 @@
 	if (err < 0)
 		goto _error;
 
-	if (substream->ops->hw_params != NULL) {
-		err = substream->ops->hw_params(substream, params);
-		if (err < 0)
-			goto _error;
-	}
-
 	runtime->access = params_access(params);
 	runtime->format = params_format(params);
 	runtime->subformat = params_subformat(params);
@@ -446,6 +440,13 @@
 		runtime->boundary *= 2;
 
 	snd_pcm_timer_resolution_change(substream);
+
+	if (substream->ops->hw_params != NULL) {
+		err = substream->ops->hw_params(substream, params);
+		if (err < 0)
+			goto _error;
+	}
+
 	runtime->status->state = SNDRV_PCM_STATE_SETUP;
 
 	pm_qos_remove_requirement(PM_QOS_CPU_DMA_LATENCY,
#2 lore raw
Quoted 10 lines · Clemens Ladisch
Quoted 5 lines
[ 5485.965788]  [<c10041c0>] ? do_divide_error+0x0/0x90
[ 5485.980619]  [<f87aa037>] ? pcm_pointer+0x37/0x70 [snd_echo3g]
[ 5485.998047]  [<c104764e>] ? run_timer_softirq+0x17e/0x2e0
[ 5486.014175]  [<f87ac9bf>] ? snd_echo_interrupt+0x11f/0x240 [snd_echo3g]
[ 5486.033940]  [<c107a5d5>] ? handle_IRQ_event+0x45/0x190

bytes_to_frames() divides by runtime->frame_bits which is not set until after the hw_params callback has succeeded, but the corresponding chip->substream[] entry is set in that callback, by init_engine(). It should probably have been set in the prepare callback.

How can an irq arrive before hw_params() and prepare() ?

Frederick, are you using the midi interface ?

--
Giuliano.

#4 lore raw
In reply to Clemens Ladisch
Quoted 20 lines · On Tue, 02 Feb 2010 16:30:27 +0100 Clemens Ladisch <[email protected]> wrote:
Quoted 18 lines · Frederick V Heitkamp

Repeatable Hard Crash.
What else do I need to provide?

[ 5484.995249] WriteControlReg: not written, no change
[ 5485.069621] divide error: 0000 [#1] pcm_hw_params ok
[ 5485.070159] Prepare rate=44100 format=2 channels=2
[ 5485.070161] set_audio_format[14] = 5
[ 5485.070166] Prepare rate=44100 format=2 channels=2
[ 5485.070167] set_audio_format[14] = 5
[ 5485.070003] PREEMPT SMP 
...
[ 5485.070003] EIP is at pcm_pointer+0x37/0x70 [snd_echo3g]
...
[ 5485.965788]  [<c10041c0>] ? do_divide_error+0x0/0x90
[ 5485.980619]  [<f87aa037>] ? pcm_pointer+0x37/0x70 [snd_echo3g]
[ 5485.998047]  [<c104764e>] ? run_timer_softirq+0x17e/0x2e0
[ 5486.014175]  [<f87ac9bf>] ? snd_echo_interrupt+0x11f/0x240 [snd_echo3g]
[ 5486.033940]  [<c107a5d5>] ? handle_IRQ_event+0x45/0x190
Quoted 4 lines

bytes_to_frames() divides by runtime->frame_bits which is not set until after the hw_params callback has succeeded, but the corresponding chip->substream[] entry is set in that callback, by init_engine(). It should probably have been set in the prepare callback.

I've just had another look at my code. Although it never happened to me, it is indeed possible when hw_params() completes if another substream is already running. The reason is that the card delivers an irq when it executes an irq instruction in any of the running s-g lists. The irq handler cannot know which substream caused it, so it has to call the pointer() function for each of the configured substreams (ie. the ones which completed one of the pcm_*_hw_params() callbacks.

There is another possible fix. I tested it briefly. It looks ok wrt race conditions because pipe->state is set only in the trigger callback. I hope I didn't overlook anything again...

alsa-kernel/pci/echoaudio/echoaudio.c renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig +3 -1
@@ -1821,7 +1821,9 @@ static irqreturn_t snd_echo_interrupt(in
 	/* The hardware doesn't tell us which substream caused the irq,
 	thus we have to check all running substreams. */
 	for (ss = 0; ss < DSP_MAXPIPES; ss++) {
-		if ((substream = chip->substream[ss])) {
+		substream = chip->substream[ss];
+		if (substream && ((struct audiopipe *)substream->runtime->
+				private_data)->state == PIPE_STATE_STARTED) {
 			period = pcm_pointer(substream) /
 				substream->runtime->period_size;
 			if (period != chip->last_period[ss]) {
Signature · 1 line
Giuliano.
#5 lore raw

There is another possible fix. I tested it briefly. It looks ok wrt race

conditions because pipe->state is set only in the trigger callback. I hope I didn't overlook anything again...

alsa-kernel/pci/echoaudio/echoaudio.c 2010-02-02 22:44:03.000000000 +0100 renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig 2010-02-02 22:37:33.000000000 +0100 +3 -1
@@ -1821,7 +1821,9 @@ static irqreturn_t snd_echo_interrupt(in
    /* The hardware doesn't tell us which substream caused the irq,
    thus we have to check all running substreams. */
    for (ss = 0; ss < DSP_MAXPIPES; ss++) {
-        if ((substream = chip->substream[ss])) {
+        substream = chip->substream[ss];
+        if (substream && ((struct audiopipe *)substream->runtime->
+                private_data)->state == PIPE_STATE_STARTED) {
            period = pcm_pointer(substream) /
                substream->runtime->period_size;
            if (period != chip->last_period[ss]) {
Signature · 6 lines
Giuliano.

I am using this patch.  So far, no crash.


Fred
#6 lore raw
In reply to Giuliano Pochini

There is another possible fix. I tested it briefly. It looks ok wrt race conditions because pipe->state is set only in the trigger callback. I hope I didn't overlook anything again...

alsa-kernel/pci/echoaudio/echoaudio.c 2010-02-02 22:44:03.000000000 +0100 renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig 2010-02-02 22:37:33.000000000 +0100 +3 -1
@@ -1821,7 +1821,9 @@ static irqreturn_t snd_echo_interrupt(in
    /* The hardware doesn't tell us which substream caused the irq,
    thus we have to check all running substreams. */
    for (ss = 0; ss < DSP_MAXPIPES; ss++) {
-        if ((substream = chip->substream[ss])) {
+        substream = chip->substream[ss];
+        if (substream && ((struct audiopipe *)substream->runtime->
+                private_data)->state == PIPE_STATE_STARTED) {
            period = pcm_pointer(substream) /
                substream->runtime->period_size;
            if (period != chip->last_period[ss]) {
Signature · 7 lines
Giuliano.


I am using this patch.  So far, no crash.


Fred
#7 lore raw
In reply to Giuliano Pochini
Quoted 51 lines · Giuliano Pochini
Quoted 21 lines · On Tue, 02 Feb 2010 16:30:27 +0100 Clemens…
Quoted 19 lines · Frederick V Heitkamp

Repeatable Hard Crash.
What else do I need to provide?

[ 5484.995249] WriteControlReg: not written, no change
[ 5485.069621] divide error: 0000 [#1] pcm_hw_params ok
[ 5485.070159] Prepare rate=44100 format=2 channels=2
[ 5485.070161] set_audio_format[14] = 5
[ 5485.070166] Prepare rate=44100 format=2 channels=2
[ 5485.070167] set_audio_format[14] = 5
[ 5485.070003] PREEMPT SMP
...
[ 5485.070003] EIP is at pcm_pointer+0x37/0x70 [snd_echo3g]
...
[ 5485.965788]  [<c10041c0>] ? do_divide_error+0x0/0x90
[ 5485.980619]  [<f87aa037>] ? pcm_pointer+0x37/0x70 [snd_echo3g]
[ 5485.998047]  [<c104764e>] ? run_timer_softirq+0x17e/0x2e0
[ 5486.014175]  [<f87ac9bf>] ? snd_echo_interrupt+0x11f/0x240 [snd_echo3g]
[ 5486.033940]  [<c107a5d5>] ? handle_IRQ_event+0x45/0x190
Quoted 5 lines

bytes_to_frames() divides by runtime->frame_bits which is not set until after the hw_params callback has succeeded, but the corresponding chip->substream[] entry is set in that callback, by init_engine(). It should probably have been set in the prepare callback.

I've just had another look at my code. Although it never happened to me, it is indeed possible when hw_params() completes if another substream is already running. The reason is that the card delivers an irq when it executes an irq instruction in any of the running s-g lists. The irq handler cannot know which substream caused it, so it has to call the pointer() function for each of the configured substreams (ie. the ones which completed one of the pcm_*_hw_params() callbacks.

There is another possible fix. I tested it briefly. It looks ok wrt race conditions because pipe->state is set only in the trigger callback. I hope I didn't overlook anything again...

alsa-kernel/pci/echoaudio/echoaudio.c static irqreturn_t snd_echo_interrupt(in · line 1829

Quoted 10 lines
alsa-kernel/pci/echoaudio/echoaudio.c renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig +3 -1
@@ -1821,7 +1821,9 @@ static irqreturn_t snd_echo_interrupt(in
  	/* The hardware doesn't tell us which substream caused the irq,
  	thus we have to check all running substreams. */
  	for (ss = 0; ss<  DSP_MAXPIPES; ss++) {
-		if ((substream = chip->substream[ss])) {
+		substream = chip->substream[ss];
+		if (substream&&  ((struct audiopipe *)substream->runtime->
+				private_data)->state == PIPE_STATE_STARTED) {
  			period = pcm_pointer(substream) /
  				substream->runtime->period_size;
  			if (period != chip->last_period[ss]) {

I've tried some newer kernels. Still having problems with the echo 3G. This is kernel version:
2.6.32.13. The above patches posted to the linux kernel list seemed to get rid of the crashes, but evidently did not go into the main kernel tree. Any help appreciated. I am willing to help to the best of my ability. Thanks!

Fred

This segment keeps looping to infinity:

[ 9331.528043] pcm_hw_free(0)
[ 9331.529200] free_pipes: Pipe 0
[ 9331.545230] pcm_hw_freed
[ 9331.552808] pcm_hw_freed
[ 9331.560365] pcm_close
[ 9331.567145] pcm_close  oc=0  cs=1  rs=1
[ 9331.578595] pcm_close2 oc=0  cs=1  rs=0
[ 9490.665630] pcm_analog_out_open
[ 9490.675009] max_channels=6
[ 9490.683109] pcm_analog_out_open  cs=1  oc=1  r=44100
[ 9490.698046] allocate_pipes: ch=0 int=2
[ 9490.702428] allocate_pipes: ok
[ 9490.718350] allocate_pipes()=0
[ 9490.727466] pcm_hw_params (bufsize=131072B periods=2 persize=65536B)
[ 9490.746450] SetSampleRate: 44100 clock d63
[ 9490.756870] WriteControlReg: Setting 0xd63, 0x3bfe
[ 9490.768468] WriteControlReg: not written, no change
[ 9490.787532] pcm_hw_params ok
[ 9490.796155] Prepare rate=44100 format=2 channels=2
[ 9490.810464] set_audio_format[0] = 5
[ 9490.820922] Prepare rate=44100 format=2 channels=2
[ 9490.835233] set_audio_format[0] = 5
[ 9490.845758] pcm_trigger start
[ 9490.847582] start_transport 1
[ 9497.317297] pcm_trigger stop
[ 9497.318517] stop_transport 1
[ 9497.334516] pcm_hw_free(0)
[ 9497.335483] free_pipes: Pipe 0
[ 9497.351709] pcm_hw_freed
[ 9497.359278] pcm_hw_freed
[ 9497.366836] pcm_close
[ 9497.373613] pcm_close  oc=0  cs=1  rs=1
[ 9497.385061] pcm_close2 oc=0  cs=1  rs=0
[ 9503.442232] pcm_analog_out_open
[ 9503.451611] max_channels=6
[ 9503.459713] pcm_analog_out_open  cs=1  oc=1  r=44100
[ 9503.474663] allocate_pipes: ch=0 int=2
[ 9503.484394] allocate_pipes: ok
[ 9503.494962] allocate_pipes()=0
[ 9503.504081] pcm_hw_params (bufsize=131072B periods=2 persize=65536B)
[ 9503.523069] SetSampleRate: 44100 clock d63
[ 9503.532174] WriteControlReg: Setting 0xd63, 0x3bfe
[ 9503.545805] WriteControlReg: not written, no change
[ 9503.564143] pcm_hw_params ok
[ 9503.572765] Prepare rate=44100 format=2 channels=2
[ 9503.587074] set_audio_format[0] = 5
[ 9503.597490] Prepare rate=44100 format=2 channels=2
[ 9503.611839] set_audio_format[0] = 5
[ 9503.622355] pcm_trigger start
[ 9503.623236] start_transport 1
[ 9510.095872] pcm_trigger stop
[ 9510.096578] stop_transport 1
[ 9510.113193] pcm_hw_free(0)
[ 9510.114163] free_pipes: Pipe 0
[ 9510.130405] pcm_hw_freed
[ 9510.137982] pcm_hw_freed
[ 9510.145541] pcm_close
[ 9510.152317] pcm_close  oc=0  cs=1  rs=1
#8 lore raw
Quoted 9 lines · On Mon, 31 May 2010 15:05:08 -0400 "F. Heitkamp" <[email protected]> wrote:
Quoted 8 lines · Giuliano Pochini
Quoted 4 lines · On Tue, 02 Feb 2010 16:30:27 +0100 Clemens Ladisch<[email protected]> wrote:
Quoted 2 lines · Frederick V Heitkamp

Repeatable Hard Crash.
What else do I need to provide?

[...]

Quoted 14 lines
Quoted 14 lines
alsa-kernel/pci/echoaudio/echoaudio.c renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig +3 -1
@@ -1821,7 +1821,9 @@ static irqreturn_t snd_echo_interrupt(in
  	/* The hardware doesn't tell us which substream caused the irq,
  	thus we have to check all running substreams. */
  	for (ss = 0; ss<  DSP_MAXPIPES; ss++) {
-		if ((substream = chip->substream[ss])) {
+		substream = chip->substream[ss];
+		if (substream&&  ((struct audiopipe *)substream->runtime->
+				private_data)->state == PIPE_STATE_STARTED) {
  			period = pcm_pointer(substream) /
  				substream->runtime->period_size;
Quoted 8 lines

alsa-kernel/pci/echoaudio/echoaudio.c static irqreturn_t snd_echo_interrupt(in · line 1829

Quoted 4 lines
alsa-kernel/pci/echoaudio/echoaudio.c renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig +0 -0
  			if (period != chip->last_period[ss]) {

I've tried some newer kernels. Still having problems with the echo 3G. This is kernel version:
2.6.32.13. The above patches posted to the linux kernel list seemed to get rid of the crashes, but evidently did not go into the main kernel tree.

Indeed. I've just checked 2.6.33 and the patch isn't there. It's in 2.6.34.

If you need 2.6.32.13 you can apply the above patch to the kernel tree or download ftp://ftp.alsa-project.org/pub/driver/alsa-driver-1.0.23.tar.bz2, otherwise just use the latest kernel.

Signature · 1 line
Giuliano.
#9 lore raw

At Mon, 31 May 2010 22:59:36 +0200,

Quoted 37 lines · Giuliano Pochini
Quoted 9 lines · On Mon, 31 May 2010 15:05:08 -0400 "F. Heitkamp" <[email protected]> wrote:
Quoted 8 lines · Giuliano Pochini
Quoted 4 lines · On Tue, 02 Feb 2010 16:30:27 +0100 Clemens Ladisch<[email protected]> wrote:
Quoted 2 lines · Frederick V Heitkamp

Repeatable Hard Crash.
What else do I need to provide?

[...]

Quoted 14 lines
Quoted 14 lines
alsa-kernel/pci/echoaudio/echoaudio.c renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig +3 -1
@@ -1821,7 +1821,9 @@ static irqreturn_t snd_echo_interrupt(in
  	/* The hardware doesn't tell us which substream caused the irq,
  	thus we have to check all running substreams. */
  	for (ss = 0; ss<  DSP_MAXPIPES; ss++) {
-		if ((substream = chip->substream[ss])) {
+		substream = chip->substream[ss];
+		if (substream&&  ((struct audiopipe *)substream->runtime->
+				private_data)->state == PIPE_STATE_STARTED) {
  			period = pcm_pointer(substream) /
  				substream->runtime->period_size;
Quoted 8 lines

alsa-kernel/pci/echoaudio/echoaudio.c static irqreturn_t snd_echo_interrupt(in · line 1829

Quoted 4 lines
alsa-kernel/pci/echoaudio/echoaudio.c renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig +0 -0
  			if (period != chip->last_period[ss]) {

I've tried some newer kernels. Still having problems with the echo 3G. This is kernel version:
2.6.32.13. The above patches posted to the linux kernel list seemed to get rid of the crashes, but evidently did not go into the main kernel tree.

Quoted 5 lines

Indeed. I've just checked 2.6.33 and the patch isn't there. It's in 2.6.34.

If you need 2.6.32.13 you can apply the above patch to the kernel tree or download ftp://ftp.alsa-project.org/pub/driver/alsa-driver-1.0.23.tar.bz2, otherwise just use the latest kernel.

Care to send the patch to stable kernel (with the upstream commit id)?

thanks,

Takashi

#10 lore raw
In reply to Giuliano Pochini
Quoted 43 lines · Giuliano Pochini
Quoted 13 lines · On Mon, 31 May 2010 15:05:08 -0400 "F. Heitkamp"<heitkamp-yWtbtysYrB+LZ21kGMrzw…
Quoted 11 lines · Giuliano Pochini
Quoted 6 lines · On Tue, 02 Feb 2010 16:30:27 +0100 Clemens…
Quoted 3 lines · Frederick V Heitkamp

Repeatable Hard Crash.
What else do I need to provide?

[...]

Quoted 15 lines
Quoted 15 lines
alsa-kernel/pci/echoaudio/echoaudio.c renamed from alsa-kernel/pci/echoaudio/echoaudio.c__orig +3 -1
@@ -1821,7 +1821,9 @@ static irqreturn_t snd_echo_interrupt(in
   	/* The hardware doesn't tell us which substream caused the irq,
   	thus we have to check all running substreams. */
   	for (ss = 0; ss<   DSP_MAXPIPES; ss++) {
-		if ((substream = chip->substream[ss])) {
+		substream = chip->substream[ss];
+		if (substream&&   ((struct audiopipe *)substream->runtime->
+				private_data)->state == PIPE_STATE_STARTED) {
   			period = pcm_pointer(substream) /
   				substream->runtime->period_size;
   			if (period != chip->last_period[ss]) {
Quoted 9 lines
Quoted 4 lines

I've tried some newer kernels. Still having problems with the echo 3G. This is kernel version:
2.6.32.13. The above patches posted to the linux kernel list seemed to get rid of the crashes, but evidently did not go into the main kernel tree.

Quoted 6 lines

Indeed. I've just checked 2.6.33 and the patch isn't there. It's in 2.6.34.

If you need 2.6.32.13 you can apply the above patch to the kernel tree or download ftp://ftp.alsa-project.org/pub/driver/alsa-driver-1.0.23.tar.bz2, otherwise just use the latest kernel.

I tried 2.6.34. I have jack 1.9..5 installed. I was able to use rhythmbox to play mp3s. I could not use the echomixer app. It complained
about a channels mismatch.

My goal is to be able to plug in a MIDI keyboard, or USB guitar interface and play along with songs etc.

I plugged in a MIDI keyboard, "/proc/asound/cards" showed the keyboard was recognized. The keyboard has a USB and MIDI interface. I am not sure if they both work at the same time. I had the MIDI cable plugged into the Gina3G, but I could not tell if the keyboard was recognized. I compiled csound5. When I engaged it the system immediately froze and started spitting out:

5831.768583] set_audio_format[0] = 86

[ 5831.768610] pcm_trigger start
[ 5831.768611] start_transport 4001
[ 5831.790538] pcm_trigger stop
[ 5831.790539] stop_transport 4001
[ 5831.790576] Prepare rate=48000 format=10 channels=2
[ 5831.790578] set_audio_format[14] = 7
[ 5831.790579] Prepare rate=48000 format=10 channels=6
[ 5831.790580] set_audio_format[0] = 86
[ 5831.790608] pcm_trigger start
[ 5831.790609] start_transport 4001
[ 5831.812538] pcm_trigger stop
[ 5831.812540] stop_transport 4001
[ 5831.812576] Prepare rate=48000 format=10 channels=2
[ 5831.812577] set_audio_format[14] = 7
[ 5831.812578] Prepare rate=48000 format=10 channels=6
[ 5831.812580] set_audio_format[0] = 86
[ 5831.812607] pcm_trigger start
[ 5831.812608] start_transport 4001
[ 5831.834537] pcm_trigger stop
[ 5831.834539] stop_transport 4001
[ 5831.834576] Prepare rate=48000 format=10 channels=2
[ 5831.834577] set_audio_format[14] = 7
[ 5831.834579] Prepare rate=48000 format=10 channels=6
[ 5831.834580] set_audio_format[0] = 86
[ 5831.834607] pcm_trigger start
[ 5831.834608] start_transport 4001
[ 5831.856536] pcm_trigger stop
[ 5831.856538] stop_transport 4001
[ 5831.856574] Prepare rate=48000 format=10 channels=2
[ 5831.856576] set_audio_format[14] = 7
[ 5831.856577] Prepare rate=48000 format=10 channels=6
[ 5831.856579] set_audio_format[0] = 86
[ 5831.856606] pcm_trigger start
[ 5831.856607] start_transport 4001
[ 5831.878536] pcm_trigger stop
[ 5831.878538] stop_transport 4001
[ 5831.878575] Prepare rate=48000 format=10 channels=2
[ 5831.878576] set_audio_format[14] = 7
[ 5831.878578] Prepare rate=48000 format=10 channels=6
[ 5831.878579] set_audio_format[0] = 86
[ 5831.878606] pcm_trigger start
[ 5831.878607] start_transport 4001
[ 5831.900535] pcm_trigger stop
[ 5831.900537] stop_transport 4001
[ 5831.900575] Prepare rate=48000 format=10 channels=2
[ 5831.900576] set_audio_format[14] = 7
[ 5831.900578] Prepare rate=48000 format=10 channels=6
[ 5831.900579] set_audio_format[0] = 86
[ 5831.900606] pcm_trigger start
[ 5831.900607] start_transport 4001
[ 5831.922535] pcm_trigger stop
[ 5831.922537] stop_transport 4001
[ 5831.922576] Prepare rate=48000 format=10 channels=2
[ 5831.922578] set_audio_format[14] = 7
[ 5831.922579] Prepare rate=48000 format=10 channels=6
[ 5831.922580] set_audio_format[0] = 86
[ 5831.922608] pcm_trigger start
[ 5831.922609] start_transport 4001
[ 5831.944535] pcm_trigger stop
[ 5831.944537] stop_transport 4001
[ 5831.944575] Prepare rate=48000 format=10 channels=2
[ 5831.944576] set_audio_format[14] = 7
[ 5831.944578] Prepare rate=48000 format=10 channels=6
[ 5831.944579] set_audio_format[0] = 86
[ 5831.944607] pcm_trigger start
[ 5831.944608] start_transport 4001
[ 5831.966536] pcm_trigger stop
[ 5831.966538] stop_transport 4001
[ 5831.966576] Prepare rate=48000 format=10 channels=2
[ 5831.966578] set_audio_format[14] = 7
[ 5831.966579] Prepare rate=48000 format=10 channels=6
[ 5831.966580] set_audio_format[0] = 86
[ 5831.966609] pcm_trigger start
[ 5831.966610] start_transport 4001
[ 5831.988533] pcm_trigger stop
[ 5831.988535] stop_transport 4001
[ 5831.988572] Prepare rate=48000 format=10 channels=2
[ 5831.988574] set_audio_format[14] = 7
[ 5831.988576] Prepare rate=48000 format=10 channels=6
[ 5831.988577] set_audio_format[0] = 86
[ 5831.988604] pcm_trigger start
[ 5831.988605] start_transport 4001
[ 5832.010532] pcm_trigger stop
[ 5832.010534] stop_transport 4001
[ 5832.010572] Prepare rate=48000 format=10 channels=2
[ 5832.010574] set_audio_format[14] = 7
[ 5832.010575] Prepare rate=48000 format=10 channels=6
[ 5832.010576] set_audio_format[0] = 86
[ 5832.010604] pcm_trigger start
[ 5832.010605] start_transport 4001
[ 5832.032531] pcm_trigger stop
[ 5832.032533] stop_transport 4001
[ 5832.032569] Prepare rate=48000 format=10 channels=2
[ 5832.032570] set_audio_format[14] = 7
[ 5832.032572] Prepare rate=48000 format=10 channels=6
[ 5832.032573] set_audio_format[0] = 86
[ 5832.032600] pcm_trigger start
[ 5832.032601] start_transport 4001
[ 5832.054532] pcm_trigger stop
[ 5832.054534] stop_transport 4001
[ 5832.054571] Prepare rate=48000 format=10 channels=2
[ 5832.054573] set_audio_format[14] = 7
[ 5832.054574] Prepare rate=48000 format=10 channels=6
[ 5832.054575] set_audio_format[0] = 86
[ 5832.054602] pcm_trigger start
[ 5832.054603] start_transport 4001
[ 5832.076529] pcm_trigger stop
[ 5832.076531] stop_transport 4001
[ 5832.076567] Prepare rate=48000 format=10 channels=2
[ 5832.076568] set_audio_format[14] = 7
[ 5832.076570] Prepare rate=48000 format=10 channels=6
[ 5832.076571] set_audio_format[0] = 86
[ 5832.076598] pcm_trigger start
[ 5832.076599] start_transport 4001
[ 5832.098530] pcm_trigger stop
[ 5832.098532] stop_transport 4001
[ 5832.098568] Prepare rate=48000 format=10 channels=2
[ 5832.098569] set_audio_format[14] = 7
[ 5832.098570] Prepare rate=48000 format=10 channels=6
[ 5832.098571] set_audio_format[0] = 86
[ 5832.098599] pcm_trigger start
[ 5832.098600] start_transport 4001
[ 5832.120532] pcm_trigger stop
[ 5832.120533] stop_transport 4001
[ 5832.120573] Prepare rate=48000 format=10 channels=2
[ 5832.120575] set_audio_format[14] = 7
[ 5832.120576] Prepare rate=48000 format=10 channels=6
[ 5832.120577] set_audio_format[0] = 86
[ 5832.120606] pcm_trigger start
[ 5832.120607] start_transport 4001
[ 5832.142529] pcm_trigger stop
[ 5832.142530] stop_transport 4001
[ 5832.142569] Prepare rate=48000 format=10 channels=2
[ 5832.142570] set_audio_format[14] = 7
[ 5832.142571] Prepare rate=48000 format=10 channels=6
[ 5832.142573] set_audio_format[0] = 86
[ 5832.142600] pcm_trigger start
[ 5832.142601] start_transport 4001
[ 5832.164529] pcm_trigger stop
[ 5832.164530] stop_transport 4001
[ 5832.164568] Prepare rate=48000 format=10 channels=2
[ 5832.164569] set_audio_format[14] = 7
[ 5832.164571] Prepare rate=48000 format=10 channels=6
[ 5832.164572] set_audio_format[0] = 86
[ 5832.164599] pcm_trigger start
[ 5832.164600] start_transport 4001
[ 5832.186528] pcm_trigger stop
[ 5832.186529] stop_transport 4001
[ 5832.186567] Prepare rate=48000 format=10 channels=2
[ 5832.186568] set_audio_format[14] = 7
[ 5832.186570] Prepare rate=48000 format=10 channels=6
[ 5832.186571] set_audio_format[0] = 86
[ 5832.186601] pcm_trigger start
[ 5832.186602] start_transport 4001
[ 5832.208527] pcm_trigger stop
[ 5832.208528] stop_transport 4001
[ 5832.208564] Prepare rate=48000 format=10 channels=2
[ 5832.208566] set_audio_format[14] = 7
[ 5832.208567] Prepare rate=48000 format=10 channels=6
[ 5832.208569] set_audio_format[0] = 86
[ 5832.208596] pcm_trigger start
[ 5832.208597] start_transport 4001
[ 5832.230526] pcm_trigger stop
[ 5832.230527] stop_transport 4001
[ 5832.230565] Prepare rate=48000 format=10 channels=2
[ 5832.230566] set_audio_format[14] = 7
[ 5832.230567] Prepare rate=48000 format=10 channels=6
[ 5832.230568] set_audio_format[0] = 86
[ 5832.230596] pcm_trigger start
[ 5832.230597] start_transport 4001
[ 5832.252545] pcm_trigger stop
[ 5832.252547] stop_transport 4001
[ 5832.252583] Prepare rate=48000 format=10 channels=2
[ 5832.252584] set_audio_format[14] = 7
[ 5832.252586] Prepare rate=48000 format=10 channels=6
[ 5832.252587] set_audio_format[0] = 86
[ 5832.252614] pcm_trigger start
[ 5832.252615] start_transport 4001
[ 5832.274524] pcm_trigger stop
[ 5832.274526] stop_transport 4001
[ 5832.274561] Prepare rate=48000 format=10 channels=2
[ 5832.274563] set_audio_format[14] = 7
[ 5832.274564] Prepare rate=48000 format=10 channels=6
[ 5832.274565] set_audio_format[0] = 86
[ 5832.274593] pcm_trigger start
[ 5832.274594] start_transport 4001
[ 5832.296536] pcm_trigger stop
[ 5832.296538] stop_transport 4001
[ 5832.296576] Prepare rate=48000 format=10 channels=2
[ 5832.296577] set_audio_format[14] = 7
[ 5832.296578] Prepare rate=48000 format=10 channels=6
[ 5832.296579] set_audio_format[0] = 86
[ 5832.296606] pcm_trigger start
[ 5832.296607] start_transport 4001
[ 5832.318525] pcm_trigger stop
[ 5832.318527] stop_transport 4001
[ 5832.318566] Prepare rate=48000 format=10 channels=2
[ 5832.318567] set_audio_format[14] = 7
[ 5832.318569] Prepare rate=48000 format=10 channels=6
[ 5832.318570] set_audio_format[0] = 86
[ 5832.318598] pcm_trigger start
[ 5832.318599] start_transport 4001
[ 5832.340523] pcm_trigger stop
[ 5832.340524] stop_transport 4001
[ 5832.340563] Prepare rate=48000 format=10 channels=2
[ 5832.340564] set_audio_format[14] = 7
[ 5832.340566] Prepare rate=48000 format=10 channels=6
[ 5832.340567] set_audio_format[0] = 86
[ 5832.340594] pcm_trigger start
[ 5832.340595] start_transport 4001
[ 5832.362522] pcm_trigger stop
[ 5832.362524] stop_transport 4001
[ 5834.003252] SysRq : Terminate All Tasks
[ 5846.410995] pcm_close
[ 5846.410997] pcm_close  oc=2  cs=0  rs=1
[ 5846.410998] pcm_close2 oc=2  cs=0  rs=1
[ 5846.449499] Prepare rate=48000 format=10 channels=2
[ 5846.473577] set_audio_format[14] = 7
[ 5846.484293] Prepare rate=48000 format=10 channels=6
[ 5846.503645] set_audio_format[0] = 86
[ 5846.514449] pcm_trigger start
[ 5846.515307] start_transport 4001
[ 5846.533291] pcm_trigger stop
[ 5846.534001] stop_transport 4001
[ 5846.567444] pcm_hw_free(14)
[ 5846.568002] free_pipes: Pipe 14
[ 5846.585146] pcm_hw_freed
[ 5846.592757] pcm_hw_freed
[ 5846.600307] pcm_close
[ 5846.607086] pcm_close  oc=1  cs=0  rs=1
[ 5846.618528] pcm_close2 oc=1  cs=1  rs=1
[ 5846.630021] pcm_hw_free(0)
[ 5846.631001] free_pipes: Pipe 0
[ 5846.647193] pcm_hw_freed
[ 5846.654751] pcm_hw_freed
[ 5846.662303] pcm_close
[ 5846.669072] pcm_close  oc=0  cs=1  rs=1
[ 5846.680516] pcm_close2 oc=0  cs=1  rs=0
[ 5846.706051] uhci_hcd 0000:00:1d.2: release dev 4 ep81-INT, period 8, phase 4s
[ 5846.783378] [drm] Resetting GPU
Quoted 2 lines
#11 lore raw
Quoted 7 lines · On Tue, 01 Jun 2010 08:25:35 -0400 "F. Heitkamp" <[email protected]> wrote:
Quoted 7 lines
Quoted 5 lines

I've tried some newer kernels. Still having problems with the echo 3G. This is kernel version:
2.6.32.13. The above patches posted to the linux kernel list seemed to get rid of the crashes, but evidently did not go into the main kernel tree.

Indeed. I've just checked 2.6.33 and the patch isn't there. It's in 2.6.34.

Quoted 7 lines
Quoted 4 lines

If you need 2.6.32.13 you can apply the above patch to the kernel tree or download ftp://ftp.alsa-project.org/pub/driver/alsa-driver-1.0.23.tar.bz2, otherwise just use the latest kernel.

I tried 2.6.34. I have jack 1.9..5 installed. I was able to use
rhythmbox to play mp3s. I could not use the echomixer app. It complained about a channels mismatch.

I'm running 2.6.34 now. Both a very old version of echomixer (2005!) and the latest version I just compiled from alsa-tools-1.0.23 work fine with my Gina3G. Please quote the error message.

Quoted 11 lines

My goal is to be able to plug in a MIDI keyboard, or USB guitar interface and play along with songs etc.

I plugged in a MIDI keyboard, "/proc/asound/cards" showed the keyboard was recognized. The keyboard has a USB and MIDI interface. I am not sure if they both work at the same time. I had the MIDI cable plugged into the Gina3G, but I could not tell if the keyboard was recognized.
I compiled csound5. When I engaged it the system immediately froze and started spitting out:
[...]

Ok, I compiled csound, but I don't know how to use it. How can I reproduce the problem ?

Signature · 1 line
Giuliano.
#12 lore raw
Quoted 33 lines · Giuliano Pochini
Quoted 7 lines · On Tue, 01 Jun 2010 08:25:35 -0400 "F. Heitkamp"<heitkamp-yWtbtysYrB+LZ21kGMrzw…
Quoted 7 lines
Quoted 5 lines

I've tried some newer kernels. Still having problems with the echo 3G. This is kernel version:
2.6.32.13. The above patches posted to the linux kernel list seemed to get rid of the crashes, but evidently did not go into the main kernel tree.

Indeed. I've just checked 2.6.33 and the patch isn't there. It's in 2.6.34.

Quoted 7 lines
Quoted 4 lines

If you need 2.6.32.13 you can apply the above patch to the kernel tree or download ftp://ftp.alsa-project.org/pub/driver/alsa-driver-1.0.23.tar.bz2, otherwise just use the latest kernel.

I tried 2.6.34. I have jack 1.9..5 installed. I was able to use
rhythmbox to play mp3s. I could not use the echomixer app. It complained about a channels mismatch.

I'm running 2.6.34 now. Both a very old version of echomixer (2005!) and the latest version I just compiled from alsa-tools-1.0.23 work fine with my Gina3G. Please quote the error message.

Quoted 11 lines

My goal is to be able to plug in a MIDI keyboard, or USB guitar interface and play along with songs etc.

I plugged in a MIDI keyboard, "/proc/asound/cards" showed the keyboard was recognized. The keyboard has a USB and MIDI interface. I am not sure if they both work at the same time. I had the MIDI cable plugged into the Gina3G, but I could not tell if the keyboard was recognized.
I compiled csound5. When I engaged it the system immediately froze and started spitting out:
[...]

Quoted 3 lines

Ok, I compiled csound, but I don't know how to use it. How can I reproduce the problem ?

Sorry I took so long to get back to you. It turns out Csound is not related to the problem.

In this case I am only using kernel 2.6.33.5-rt23-amd64, alsa lib 1.0.2, but 2.6.34 does it too.
I bring up firefox. Web pages attempt to play sound. The kernel begans doing:

[ 2879.766331] allocate_pipes: ch=1 int=2
[ 2879.766332] allocate_pipes: channel already open
[ 2879.766333] allocate_pipes(1) err=-11
[ 2879.766334] pcm_hw_freed
[ 2.821190] pcm_analog_out_open  cs=0  oc=2  r=44100
[ 2879.821266] allocate_pipes: ch=1 int=2
[ 2879.821267] allocate_pipes: channel already open

,infinitely. The system is unusable at this point and I have to Alt-SysRq-k to get back in.

Let me know what other information to provide.

Here is hwinfo on the system:
cpu:

                       Intel(R) Core(TM)2 CPU          6600  @ 2.40GHz, 
2400 MHz
                       Intel(R) Core(TM)2 CPU          6600  @ 2.40GHz, 

2400 MHz
keyboard:
/dev/input/input0 EzKEY USB-compliant keyboard
printer:

                       HP deskjet 5600

graphics card:
PC Partner VGA compatible controller
sound:
PC Partner Audio device
storage:
LSI Logic / Symbios Logic 53c1030 PCI-X
Fusion-MPT Dual Ultra320 SCSI
LSI Logic / Symbios Logic 53c1030 PCI-X
Fusion-MPT Dual Ultra320 SCSI
Intel 82801G (ICH7 Family) IDE Controller
network:
eth1 Tyan 82573E Gigabit Ethernet Controller (Copper)
eth0 Tyan 82573E Gigabit Ethernet Controller (Copper)
network interface:

  eth1                 Ethernet network interface
  eth0                 Ethernet network interface
  lo                   Loopback network interface
usb controller:
                       Intel 82801G (ICH7 Family) USB2 EHCI Controller
                       Intel 82801G (ICH7 Family) USB UHCI #4
                       Intel 82801G (ICH7 Family) USB UHCI #3
                       Intel 82801G (ICH7 Family) USB UHCI #2
                       Intel 82801G (ICH7 Family) USB UHCI #1

bios:
BIOS
bridge:
Texas Instruments PCI bridge
Intel 6702PXH PCI Express-to-PCI Bridge A Intel 82801GB/GR (ICH7 Family) LPC Interface Bridge Intel 82801 PCI Bridge
Intel 82801GR/GH/GHM (ICH7 Family) PCI Express
Port 6
Intel 82801GR/GH/GHM (ICH7 Family) PCI Express
Port 5
Intel 82801G (ICH7 Family) PCI Express Port 1 Intel 975X Express PCI Express Root Port Intel E7230 PCI Express Root Port
Intel E7230 Memory Controller Hub
hub:
ATEN International Hub
Linux 2.6.33.5-rt23-amd64 uhci_hcd UHCI Host
Controller

                       Linux 2.6.33.5-rt23-amd64 uhci_hcd UHCI Host 
Controller
                       Linux 2.6.33.5-rt23-amd64 uhci_hcd UHCI Host 
Controller
                       Linux 2.6.33.5-rt23-amd64 uhci_hcd UHCI Host 
Controller
                       Linux 2.6.33.5-rt23-amd64 ehci_hcd EHCI Host 

Controller
memory:
Main Memory
firewire controller:
Timedia TSB82AA2 IEEE-1394b Link Layer Controller
unknown:
FPU
DMA controller
PIC
Timer
Keyboard controller
/dev/lp0 Parallel controller
Echo Digital Audio 3G
Intel 82801G (ICH7 Family) SMBus Controller PnP Unclassified device
PnP Unclassified device
PnP Unclassified device
Unclassified device
PnP Unclassified device
PnP Unclassified device
PnP Unclassified device
PnP Unclassified device
PnP Unclassified device
PnP Unclassified device
PnP Unclassified device
PnP Unclassified device
Serial controller
/dev/input/input2 Logitech Trackball
/dev/input/input1 EzKEY USB-compliant keyboard
Belkin UPS

#13 lore raw
Quoted 30 lines · On Thu, 17 Jun 2010 08:11:06 -0400 "F. Heitkamp" <[email protected]> wrote:
Quoted 11 lines
Quoted 11 lines

My goal is to be able to plug in a MIDI keyboard, or USB guitar interface and play along with songs etc.

I plugged in a MIDI keyboard, "/proc/asound/cards" showed the keyboard was recognized. The keyboard has a USB and MIDI interface. I am not sure if they both work at the same time. I had the MIDI cable plugged into the Gina3G, but I could not tell if the keyboard was recognized.
I compiled csound5. When I engaged it the system immediately froze and started spitting out:
[...]

Quoted 3 lines

Ok, I compiled csound, but I don't know how to use it. How can I reproduce the problem ?

Sorry I took so long to get back to you. It turns out Csound is not related to the problem.

In this case I am only using kernel 2.6.33.5-rt23-amd64, alsa lib 1.0.2, but 2.6.34 does it too.
I bring up firefox. Web pages attempt to play sound. The kernel
begans doing:

[ 2879.766331] allocate_pipes: ch=1 int=2
[ 2879.766332] allocate_pipes: channel already open
[ 2879.766333] allocate_pipes(1) err=-11
[ 2879.766334] pcm_hw_freed
[ 2.821190] pcm_analog_out_open  cs=0  oc=2  r=44100
[ 2879.821266] allocate_pipes: ch=1 int=2
[ 2879.821267] allocate_pipes: channel already open
Quoted 4 lines

,infinitely. The system is unusable at this point and I have to Alt-SysRq-k to get back in.

Let me know what other information to provide.

Perhaps is it related with the OSS emulation ? Do you know if the browser plays the sound itself or if it is using an external program or plugin ?

Signature · 1 line
Giuliano.