tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: 4.0/i386 auich_calibrate() and GENERIC.NOACPI failure (microtime?)



In article <20080111155333.GA979%ece.cmu.edu@localhost>,
Chuck Cranor  <chuck%ece.cmu.edu@localhost> wrote:
>hi-
>
>    I updated my i386/desktop to 4.0 yesterday, and discovered that the 
>audio auich0 does not configure properly with the GENERIC.NOACPI kernel.
>It might be timecounter related?
>
>GENERIC kernel (correct audio):
>  auich0: measured ac97 link rate at 55069 Hz, will use 55000 Hz
>
>
>GENERIC.NOACPI (audio plays at wrong rate):
>  auich0: measured ac97 link rate at 43170 Hz, will use 48000 Hz
>
>
>55000 Hz is the correct value, 48000 Hz sounds wrong (music too fast).
>
>the auich driver uses the hardware and microtime() to compute the 
>ac97 link rate.   The GENERIC kernel uses kern.timecounter.hardware=ACPI-Fast,
>where as the GENERIC.NOACPI uses kern.timecounter.hardware=i8254.
>
>
>    Anyone have any suggestions on how to debug this?   Here are
>some links to dmesg output:
>
>  GENERIC:        http://yogi.pdl.cmu.edu/~chuck/40/dmesg-gen.txt
>  GENERIC.NOACPI: http://yogi.pdl.cmu.edu/~chuck/40/dmesg-noacpi.txt

The clock on my alpha with /current runs too fast. Could it be related?

christos




Home | Main Index | Thread Index | Old Index