Subject: Re: [linux-audio-user] APIC is bad?
From: Mark Knecht (mknecht_AT_controlnet.com)
Date: Fri Jul 16 2004 - 18:19:48 EEST
R Parker wrote:
> --- Mark Knecht <mknecht_AT_controlnet.com> wrote:
>>Personally I have done no optimization work on APIC
>>systems and cannot
>>tell you what would be best. However, TTBOMK, there
>>is no simple way
>>today to optimize APIC interrupts in a Linux system
>>anyway, so as far as
>>I can tell you basically take what you get. If you
>>get too many xruns
>>then I guess you go back to non-APIC mode.
> My recolector is awful faulty but I seem to recall
> Clemens Ladisch having written something for APIC
> interupt ordering. Is my condition even worse than I'm
> aware of?
No, you are exactly right. Clemens and I have some small technical
disagreements on this subject. We had a conversation that was
interesting but I think didn't yield any clear answers, at least in my
mind, so I freely admit I'm still confused on the subject.
I design PC hardware for a living and have worked on chipsets while at
AMD. I understand the older style interrupt controller structure very
well. APIC stuff came along after I left to work on 1394 stuff so I
don't know it as well. As I remember the conversation the interrupt
order was from the highest interrupt number and going down from there,
so if Florin sees 25 interrupts, then #25 gets serviced first and that
would be where I'd want my sound card, if I could get it there. (How you
do this is not clear...)
If I remember Clemens POV on this subject he said it didn't matter what
APIC IRQ you choose. I somehow felt that choosign a bad one had to cause
stuff like xrun problems, but I haven't tried to test this hypothesis.
This archive was generated by hypermail 2b28 : Fri Jul 16 2004 - 18:27:19 EEST