Hello John,
> I've been having a lot of problems with two K6-350's. If I run
> one at full speed it segfaults very regularly during compiles,
> and under load (like X with Netscape).
I have exactly the same problem. Same CPU, but my system has
an ASUS PB-AB motherboard. The strange part is that all compiles
without '-pipe' are successful.
Even tar files are causing the same symptoms, but when I gunzip them
first and then extract them, everything is fine. I can also run massive
parallel makes without problems, but not with pipes.
> I bumped it down to 333Mhz
> and it seems to be more stable. I have tried two different
> processors, different motherboards, different memory, and all
> combinations of bios/bus setups.
I replaced the AMD with my old Pentium (no MMX) an the system was
absolutely stable. Maybe there're some design flaws in the CPU's
which are (still) unknown.
Remember the Win95 bugfix because of tight timing-loops ?
I guess that some instructions don't pair very well. I tried a few
combinations of memery access in tight loops using gas, but so far I
wasn't able to write a testcase to reproduce this behavior.
Cheers,
Patrick
-
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/