ACPI + SiS IRQ Routing?

ACPI + SiS IRQ Routing?

[Joshua Schmidlkofer]

[-- Attachment #1: Type: text/plain, Size: 1126 bytes --]

What is the status of the SiS IRQ router?  I have two motherboards, one
with a 645, and one with a 648.  I have confirmed that all the way up to
[and including] 2.6.0-test6 I am having problems.  I am going to test
with mm4, but on another system I experienced a wierd Oops with mm4.  

I have historically had issue w/ the 1 interrupt/4 device High Point IDE
contollers unless I have full ACPI support enabled.  Prior to test6, I
had a conflict between my Nvidia [w/ or w/o binary driver].  Now, it is
my main USB device that suffers.  What can I do to help with the
process.

The symptom is that the ACPI interrupt begins acumlating events in
/proc/interrupts until it reaches something like 100000, then I get an
Oops about an interrupt that no one cared about [I.e. IRQ 9], and the
system continues to run fine.  Aside from the Oops, I don't see any
other problems, but I am a little afraid to run at that point.

thanks,
   Joshua


-- 
VB programmers ask why no one takes them seriously, 
it's somewhat akin to a McDonalds manager asking employees 
why they don't take their 'career' seriously.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: ACPI + SiS IRQ Routing?

[Joshua Schmidlkofer]

On Wed, 2003-10-08 at 17:17, Joshua Schmidlkofer wrote:
> What is the status of the SiS IRQ router?  I have two motherboards, one
> with a 645, and one with a 648.  I have confirmed that all the way up to
> [and including] 2.6.0-test6 I am having problems.  I am going to test
> with mm4, but on another system I experienced a wierd Oops with mm4.  
> 
> I have historically had issue w/ the 1 interrupt/4 device High Point IDE
> contollers unless I have full ACPI support enabled.  Prior to test6, I
> had a conflict between my Nvidia [w/ or w/o binary driver].  Now, it is
> my main USB device that suffers.  What can I do to help with the
> process.
> 
> The symptom is that the ACPI interrupt begins acumlating events in
> /proc/interrupts until it reaches something like 100000, then I get an
> Oops about an interrupt that no one cared about [I.e. IRQ 9], and the
> system continues to run fine.  Aside from the Oops, I don't see any
> other problems, but I am a little afraid to run at that point.

I wanted to follow up that the 648 chipset seems to be almost 100% with
test6.  I am building test7 now, and I will post if I have any news on
that.

js


-- 
VB programmers ask why no one takes them seriously, 
it's somewhat akin to a McDonalds manager asking employees 
why they don't take their 'career' seriously.



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php

Re: ACPI + SiS IRQ Routing?

[Joshua Schmidlkofer]

On Wed, 2003-10-08 at 21:24, Joshua Schmidlkofer wrote:
> On Wed, 2003-10-08 at 17:17, Joshua Schmidlkofer wrote:
> > What is the status of the SiS IRQ router?  I have two motherboards, one
> > with a 645, and one with a 648.  I have confirmed that all the way up to
> > [and including] 2.6.0-test6 I am having problems.  I am going to test
> > with mm4, but on another system I experienced a wierd Oops with mm4.  
> > 
> > I have historically had issue w/ the 1 interrupt/4 device High Point IDE
> > contollers unless I have full ACPI support enabled.  Prior to test6, I
> > had a conflict between my Nvidia [w/ or w/o binary driver].  Now, it is
> > my main USB device that suffers.  What can I do to help with the
> > process.
> > 
> > The symptom is that the ACPI interrupt begins acumlating events in
> > /proc/interrupts until it reaches something like 100000, then I get an
> > Oops about an interrupt that no one cared about [I.e. IRQ 9], and the
> > system continues to run fine.  Aside from the Oops, I don't see any
> > other problems, but I am a little afraid to run at that point.
> 
> I wanted to follow up that the 648 chipset seems to be almost 100% with
> test6.  I am building test7 now, and I will post if I have any news on
> that.
> 


   Test 7 and the 645 continue to be a problem.  It takes a lot longer
now, but something eventually begins causing more interrupts problems
with ACPI.  

   Does anyone have any comment?  Help?



> js
-- 
VB programmers ask why no one takes them seriously, 
it's somewhat akin to a McDonalds manager asking employees 
why they don't take their 'career' seriously.



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php

Re: ACPI + SiS IRQ Routing?

[Alan Cox]

On Iau, 2003-10-09 at 01:17, Joshua Schmidlkofer wrote:
> What is the status of the SiS IRQ router?  I have two motherboards, one
> with a 645, and one with a 648.  I have confirmed that all the way up to
> [and including] 2.6.0-test6 I am having problems.  I am going to test
> with mm4, but on another system I experienced a wierd Oops with mm4.  

Unless Linus and Andrew have forward ported the 2.4 PIRQ rewrite then
the SiS hardware won't work right with 2.6 in non ACPI mode at least.

Alan



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php

Re: ACPI + SiS IRQ Routing?

[Joshua Schmidlkofer]

On Fri, 2003-10-10 at 15:10, Alan Cox wrote:
> On Iau, 2003-10-09 at 01:17, Joshua Schmidlkofer wrote:
> > What is the status of the SiS IRQ router?  I have two motherboards, one
> > with a 645, and one with a 648.  I have confirmed that all the way up to
> > [and including] 2.6.0-test6 I am having problems.  I am going to test
> > with mm4, but on another system I experienced a wierd Oops with mm4.  
> 
> Unless Linus and Andrew have forward ported the 2.4 PIRQ rewrite then
> the SiS hardware won't work right with 2.6 in non ACPI mode at least.
> 
> Alan
> 

Does anyone have some hope about dealing with this going forward?  The
648 at least behaves better than the 645. I don't know how much of a
corner-case I represent, but advice would be cool.

js




-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php

Re: ACPI + SiS IRQ Routing?

[Len Brown]

Does your SiS 648 system behave like this one?:
http://bugme.osdl.org/show_bug.cgi?id=1283

Can you describe how the 645 behaviour differs -- hopefully in a bug
report?

thanks,
-Len




-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php

Re: ACPI + SiS IRQ Routing?

[Joshua Schmidlkofer]

Thanks! I will look this up. =).

js


On Fri, 2003-10-10 at 23:16, Len Brown wrote:
> Does your SiS 648 system behave like this one?:
> http://bugme.osdl.org/show_bug.cgi?id=1283
> 
> Can you describe how the 645 behaviour differs -- hopefully in a bug
> report?
> 
> thanks,
> -Len
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: SF.net Giveback Program.
> SourceForge.net hosts over 70,000 Open Source Projects.
> See the people who have HELPED US provide better services:
> Click here: http://sourceforge.net/supporters.php
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php

Re: ACPI + SiS IRQ Routing?

[Joshua Schmidlkofer]

Well first I have a little trouble sorting out the problem with that bug
report.  It looks like USB failure is the key symptom.  I see no
symptoms, except the oops.  I will enshrine this in a new bug report.

Also, my problems are with 2.6.0-testX, and I have not done much testing
with 2.4.xx.  Tell me, is pci=noacpi, or acpi=off a recommended way to
go?  I have a similar problem with the 645 mostly.  The 648 has given me
very few problems.  I will create a report for the 645, and try to
re-produce reliable problems with the 648.

js


On Fri, 2003-10-10 at 23:16, Len Brown wrote:
> Does your SiS 648 system behave like this one?:
> http://bugme.osdl.org/show_bug.cgi?id=1283
> 
> Can you describe how the 645 behaviour differs -- hopefully in a bug
> report?
> 
> thanks,
> -Len
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: SF.net Giveback Program.
> SourceForge.net hosts over 70,000 Open Source Projects.
> See the people who have HELPED US provide better services:
> Click here: http://sourceforge.net/supporters.php
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel
-- 
VB programmers ask why no one takes them seriously, 
it's somewhat akin to a McDonalds manager asking employees 
why they don't take their 'career' seriously.



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php