Re: kernel 2.4.x + via-based kt266 mobo = IDE cdroms probls (revisited)

* Re: kernel 2.4.x + via-based kt266 mobo = IDE cdroms probls (revisited)
@ 2003-02-07 22:53 Laurent Grawet
  0 siblings, 0 replies; 3+ messages in thread
From: Laurent Grawet @ 2003-02-07 22:53 UTC (permalink / raw)
  To: linux-kernel

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii"; format=flowed, Size: 2444 bytes --]

Hi Andre,

I have exactly the same problem/symptoms like several other users !
Here is a interesting report that has been made on Viaarena Linux forum by 
Dam. This bring us useful information and better understanding of the 
situation :

http://forums.viaarena.com/messageview.cfm?catid=28&threadid=28308&STARTPAGE=1

--------- SNIP

  Thursday, February 06, 2003 9:56 PM

Hi there.

One more guy with the same problem. But I've got a kind of solution, but it 
is not a good one.

Using cdda2wav, you can use the -n option to tell cdda2wav the number of 
sector to read at a time.
I'm able to rip a cd using -n 1 . I can go up to -n 15 then it start hdc: 
lost interrupt ... Problem is that
performance is not great at -n 1.

I once contacted the cdrdao author, Andreas, and talk with him about this 
prog. I ended editing cdrdao sources
to try to reduce the number of sector read at a time to 1 (same as for 
cdda2wav). Problem disappeared but
performance were unacceptable.

Also there is the option to not write the rip to a file (for debugging 
purpose), using -N. Using that
option, I no more experience problem, but of course, I don't rip anything.

So it seemed to me that the problem is related to writing to a disk while 
ripping. So I tried cdda2wav with -N and
at the same time launched a cp -R /usr/src/linux /tmp. Everything happened 
correctly, slowly but without lost interrupts.

I think this is more of a kernel bug then a bios bug (why would it work 
under windows ?). Searching lkml gives many
results with people with this problem, but no solutions.

I think we should start bother people on the lkml.

---------- SNIP

 > I think we should start bother people on the lkml.
Well, this has just been done Dam   :-)

Exactly the same behaviour here...

Could Linux Guru's help us ?
We have this problem for too long time...

Thanks,

Laurent

** Please, CC me your answer to LKML **

P.S.
Sorry for my double post, but there was a problem in my message title...

-- 
| Laurent Grawet
| ON1MGL
|
| [e-mail]
|    Home...... mailto:laurent.grawet@ibelgique.com
|               mailto:on1mgl@ibelgique.com
|    Office.... mailto:laurent.grawet@fundp.ac.be
|
| [ax25]
|               on1mgl@on0cha.#clr.ht.bel.eu

_____________________________________________________________________
Envie de discuter en "live" avec vos amis ? Télécharger MSN Messenger
http://www.ifrance.com/_reloc/m la 1ère messagerie instantanée de France

^ permalink raw reply	[flat|nested] 3+ messages in thread