Re: Buggy HLT instruction in Cyrix M2 or error in ATAPI CDROM code ?

Frantisek Dufka (dufkaf@merit.cz)
Sun, 17 Oct 1999 16:52:40 +0200


Dne Ne, 17 říjen 1999 Alan Cox napsal(a):
> > So IMHO there's nothing wrong with my IDE interface.
>
> If it isnt running at 33Mhz it isnt running to spec.
>
> > I just thought linux-kernel people DO like bugreports.
>
> We do - but reports for probably overdriven hardware are not terribly useful
> because hardware does strange things and it becomes tricky to tell a kernel
> weirdness/bug from a hardware one

True, I had to write that I tried it also on 66Mhz, sorry.

I just double checked it yesterday evening,
It really does happen with the Cyrix clocked to 233(3.5x66) and 133(2x66)
It doesn't occurr with my old Pentium switched to 133(2x66)

The CPUs are not overclocked , I did compile it with gcc 2.7.2.3 from debian 2.1

I probably can borrow different CDROM drive if that helps.
I will report the result later.

I tried to find how many cycles it takes for the cyrix to wake up from 'hlt
mode'. There's good (approx. 150 pages) mii_all.pdf on the cyrix site, with
lot of information, but I failed to find precisely this :(
Anyway, I don't know it for pentium either.

My (humble) theory is that the CPU misses the interrupt or cannot react in
timely fashion when in 'hlt mode' and my Maxtor hard drive just waits a little
bit longer than the CDROM .

Frantisek Dufka

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/