A Weird K6 Bug : FAQ

A Weird K6 Bug : FAQ


Table of contents

  1. Which configuration have been tested ?
  2. To whom has this happened ?
  3. Do you have a re-marked chip ?
  4. Is your CPU too hot ?
  5. Is your voltage regulators too hot ?
  6. Are your RAM or motherboard corrupted ?
  7. What is AMD's analysis of the problem ? (HOT)
  8. Read the SIG11 FAQ !
  9. My K6 works, you are a moron !
  10. What about serial numbers ?
  11. Has the problem been seen with less than 64 MB ?
  12. My K6 has the bug. What should I do now ? (HOT)

Q1: Which configuration have been tested ?

The bug has been observed with :

Motherboards :

RAM :

Clock speeds :

Operating systems :

Problems disappear as soon as :

Q2: To whom has this happened ?

The following persons have a CPU that has the bug : The following persons have a CPU that may have the bug : The following persons have a CPU that does not have the bug :

Q3: Do you have a re-marked chip ?

It does not matter because any K6 that AMD sells has been tested at 166 MHz (there are no slower K6), so it should work at 166 MHz. My K6 does not work at 100 Mhz with 64 MB, but works at 233 Mhz with only 32 MB, so it is not an overcloking problem. Others have also underclock their CPUs, but the problem remained.

Q4: Is your CPU too hot ?

I don't know how a K6/100 with 64 MB would fail because of temperature, when the same K6 at 233 MHz works perfectly with only 32 MB. Anyway, the TX97 has a heat sensor under the CPU, and with a correct cooling system, the K6 is well under its limits. See The chessboard test if you still believe it is a heat problem.

Q5: Is your voltage regulators too hot ?

No, on the TX97 they are barely warm. See The chessboard test if you still believe it is a heat problem.

Q6: Is your RAM or motherboard corrupted ?

They work perfectly with a P100 and a P200+ and 64 MB. They also work perfectly with the K6 with only 32 MB. Many motherboards and many RAM modules have been tested. Some of them were probably OK, but as the problem remained, it is probably the K6's fault...

Q7: What is AMD's analysis of the problem ? (HOT)

AMD says :

Q8: Read the SIG11 FAQ !

Thanks for the advice. I've read it many times, it does not contain the solution of this very problem.

Q9: My K6 works, you are a moron !

Thanks for the information. If you don't mind helping a fast growing set of morons :
  1. Test your K6 with the script found in this page.
  2. If it works :
  3. If it does not work :

Q10: What about serial numbers ?

I have received failure reports of K6 with the following serial numbers : I have received success reports of K6 with the following serial numbers : From now, only informative serial numbers will be added to these lists.

Q11: Has the problem been seen with less than 64 MB ?

Under Linux, I have has been able to reproduce it with 36 MB but not with 34 MB or less.

Q12: My K6 has the bug. What should I do now ? (HOT)

Here is AMD's answer to this question :
AMD recently received reports from a limited number of users having
intermittent problems while running core re-compiles of the Linux
shareware operating system. Our systems engineering group has duplicated
the observation and determined that it is related to a previously know
erratum. Full technical details of this erratum are documented in
section 2.6.2 of the AMD-K6 MMX Enhanced Processor Revision Guide
on our website, www.amd.com. Users that feel they are being affected by
this problem, should contact AMD's support line at (408) 749-3060 and
ask for Dan Hingle or Glen Garcia.
People living outside of the USA should contact their local AMD technical support.
[Main page] [News page] [Tests page]
Updated: 97/09/23.
Send comments to poulot@wanadoo.fr