On Fri, Mar 11, 2011 at 2:44 PM, Takashi Iwai tiwai@suse.de wrote:

At Fri, 11 Mar 2011 11:04:27 +0000, Amir Shamsuddin wrote:

...

Hi,

My laptop occasionally fails to resume from suspend, and when it does, I see (only) this error message on screen:

[nnn.nnn] atiixp: codec reset timeout

I had a look at the code where this comes from, and while I'm not sure it's related to my resume problem, it does (to my untrained eyes) look like there is a small bug in the function snd_atiixp_aclink_reset in file sound/pci/atiixp.c:

(lines 500 - 536 in latest git) static int snd_atiixp_aclink_reset(struct atiixp *chip) {        int timeout; [snip]        timeout = 10;        while (! (atiixp_read(chip, CMD) & ATI_REG_CMD_ACLINK_ACTIVE)) {                /* do a hard reset */ [snip]                if (--timeout) {                        snd_printk(KERN_ERR "atiixp: codec reset timeout\n");                        break;                }        } [snip] }

Surely the timeout condition should be negated, so the timeout message and loop break is only hit when 'timeout' reaches zero? ie.

if (! --timeout) {

Or am I missing something?

Indeed it's a bug.  Does the message go away when you change it?

I haven't tried a new kernel with it fixed yet (I'm not usually a kernel developer), but I'll give it a go and try and trigger the resume bug and let you know what happens.

regards,

Amir

thanks,

Takashi

At Mon, 14 Mar 2011 09:35:08 +0000, Amir Shamsuddin wrote:

On Fri, Mar 11, 2011 at 2:50 PM, Amir Shamsuddin AmirS2+alsa@gmail.com wrote:

...

On Fri, Mar 11, 2011 at 2:44 PM, Takashi Iwai tiwai@suse.de wrote:

...

At Fri, 11 Mar 2011 11:04:27 +0000, Amir Shamsuddin wrote:

...

Hi,

My laptop occasionally fails to resume from suspend, and when it does, I see (only) this error message on screen:

[nnn.nnn] atiixp: codec reset timeout

I had a look at the code where this comes from, and while I'm not sure it's related to my resume problem, it does (to my untrained eyes) look like there is a small bug in the function snd_atiixp_aclink_reset in file sound/pci/atiixp.c:

(lines 500 - 536 in latest git) static int snd_atiixp_aclink_reset(struct atiixp *chip) {        int timeout; [snip]        timeout = 10;        while (! (atiixp_read(chip, CMD) & ATI_REG_CMD_ACLINK_ACTIVE)) {                /* do a hard reset */ [snip]                if (--timeout) {                        snd_printk(KERN_ERR "atiixp: codec reset timeout\n");                        break;                }        } [snip] }

Surely the timeout condition should be negated, so the timeout message and loop break is only hit when 'timeout' reaches zero? ie.

if (! --timeout) {

Or am I missing something?

Indeed it's a bug.  Does the message go away when you change it?

I haven't tried a new kernel with it fixed yet (I'm not usually a kernel developer), but I'll give it a go and try and trigger the resume bug and let you know what happens.

I've tried it now, the message does go away with the fix in.

Unfortunately it doesn't help my laptop resume, but I think that's a separate problem!

Thanks for testing.

Takashi