Playing audio it sounds like a record recorded ay 45 RPM playing at 33 1/3

[rwklein@…]

Iiittt souunds sloooow. I let you hear this at the Developer Workshop Vladest, you play music and it plays slow. This also does not occur with the Soundmax driver.

[vladest]

How can I test it at my side? the only difference I see it's different ac97 codec rates btw, I will not at dev workshop

[anonymous]

Well I have had this problem a Thinkpad with a Crystal AC 4297 chipset. And I have had this on a T23 and a T40. It seems to be something with AC 97 on Intel side. How can I provide more information ?

[vladest]

provide me T23 or T40. Or even T34 - I'll come to nl on it :) w/o jokes: I wonna you to test it via uniaud api

[rwklein@…]

Finaly getting time to get back to this. Its not the UNIAUD API that will make a difference, its a timing issue on hardware level. Rudiger Ihle confirmed it was something in the AC 97 Intel (and Intel compatible code).

[anonymous]

I see this on my T42 with the latest uniaud netlabs testcase dated-6-23-2006 when my system boots off of battery. This is not an issue here if I boot while on power. I do not recall seeing this on other versions of the driver however.

[Steve Wendt]

Today for the first time I have slow audio on the T42 after booting on AC. Same drivers dated 6-23-2006.

[Steve Wendt]

OK, the 7-09-06 testcase release is working fine when booting under battery or AC on the T42 now.

[abwillis]

On T42 I am now experiencing this with the uniaud_28012007_testcase.zip build. This is on power, haven't tried battery. Previous drivers versions haven't done this. Will attach log. Vendor 8086h Intel Corporation Device 24C5h 82801DB/DBL AC'97 Audio Controller (ICH4/ICH4-L A1 step) Subsystem ID 05541014h SoundMAX Integrated Digital Audio

[abwillis]

I had skipped the msv_ich_test build but went back to double check that the problem was not introduced there. With that build sound is fine so it was indeed introduced with the uniaud_28012007_testcase build.

[abwillis]

OK, just determined that in my case the slow down is caused by the power saver changes done in the latest release. I was set to battery optimized even on AC and I just changed that and it works fine when not in battery optimized as far as sound goes.

[Brendan Oakley]

This problem does not occur with a build of the source from revision 85 (which needs slight modification to allow it to build). Date 10/2006. Revision 86 causes a TRAP E on the effected hardware, and Revision 88 does not TRAP but does have this slow sound problem with power saving enabled.

[Brendan Oakley]

Selectively updating files from revision 85 to revision 88 and testing reveals that this problem occurs only when the AC'97 code is updated. This includes 4 files: /alsa-kernel/pci/ac97/ac97_codec.c /alsa-kernel/pci/ac97/ac97_patch.c /alsa-kernel/pci/ac97/ac97_patch.h /alsa-kernel/include/sound/ac97_codec.h

[Brendan Oakley]

I believe this is fixed now in Revision 229. Leaving the ticket open for now until tested and confirmed.

[Yoda]

please test:
​ftp://ftp.ecomstation.dk/ac97-1.zip[[BR]]

And see if it changes anything....
and give us "uniinfo" logs

[Brendan Oakley]

All recent testing with this has been done by Andy. Roderick, can you please confirm that this is fixed in Uniaud 1.1.4RC4? (The fix applied to 1.1.4RC3.)

[Yoda]

No response for too long...