Description

The AMD-K6 processor has a bug that prevents reliable operation when more than 32 MB of RAM is used.

The most common symptoms are segmentation violations (see The SIG11 FAQ) while compiling the Linux kernel.

It can be reproduced, up to now, only when doing heavy compilations, probably because only compilations stress the system enough. It is not a gcc problem, as it is sometimes the program that launches gcc (it can be make, or sh) that dies.

This bug has been seen by many people all around the world. The general consensus among them is that the bug only depends on the amount of memory used :

With 32 MB of RAM, or less, no problem.
With more than 32 MB, sporadic compile failures have been observed.

According to AMD :

this bug is documented in section 2.6.2 of the AMD-K6 MMX Enhanced Processor Revision Guide
it has been corrected in later chips produced in the B stepping.
current machines are shipped with the corrected chips.

(WARNING) If you have problems with 32 MB or less, read the The SIG11 FAQ, it is not the same problem.

References

This bug has already been described, and is still discussed in :

comp.os.linux.hardware :
- thread : "AMD K6 Problems"
- thread : "K6 is fucked!!"
- thread : "Re: K6 is fucked!! (Why not clearing this?)"
- thread : "ARGH! Help me hunt down a signal 11 error with linux and an AMD k6-233 CPU."
- thread : "AMD K6 Problems (Everyone with a K6 might want to try this...)"
- thread : "K6 Sig11 - Two More Datapoint"
- thread : "K6 and Linux problem SOLVED"
- thread : "K6 is...err...'having difficulties'"
- thread : "K6 Sig11 - Mine works!"
- thread : "Theories about the K6 Bug"
comp.os.linux.hardware & comp.os.linux.advocacy :
- thread : "All hail the Intel fan club!"
comp.sys.ibm.pc.hardware.chips :
- thread : "flaws in k6"
fr.comp.os.linux :
- thread : "Linux et K6"
de.comp.os.linux.hardware :
- thread : "Re: Linux-Probleme mit Hardware-Kombi?"
the linux kernel mailing list.
- thread : "Does the K6-PR233ANR work with Linux 2.0.30 and ASUS TX97-XE motherboard"
- thread : "AMD K6 and trampolines"
the freebsd-hardware mailing list.

How to reproduce the bug

The problem is hard to reproduce. It is probably why it only surfaces now, a few month after the K6 introduction.

ALL people that have tried to stress test their K6, except one, with the scripts that follow, have found the problem.

If your K6 seems to work well with more than 32 MB, it only means that you have not been faced the bug in the past. But, you may have to face corrupted data, segmentation violations and other nice things in the future.