On Tue, Jun 02, 1998 at 12:22:15PM +0400, Eugene Sevinian wrote: > Recently, there were a lot of post in linux-kernel@vger.rutgers.edu > concerning new K6 bug. Tests was done by 'crashme'. Here is excerption > from Andreas Haumer <andreas@xss.co.at> report. > > # OS CPU result > ==================================== > 11 Linux-2.0 K6 crash > 0 Linux-2.0 K6 no crash > 0 Linux-2.1 K6 crash > 3 Linux-2.1 K6 no crash > 0 Linux-2.0 Pentium crash > 6 Linux-2.0 Pentium no crash > 1 Linux-2.0 K6 unclear > ===================================== > > Sincerly speaking I have now idea how cricual is this, but as far as > I consider K6 as a choice for upgrating my PC, I would like to know how > reliable is this assuming that I have no intention to run this testing > program very often. I followed this thread from when it started and my conclusions up to now are: 1) It appears to show up only under circumstances not easily found in real life. Someone now posted a short test program that will crash a K6. This is not worse than the Intel F0 0F bug. 2) 2.1 kernels are immune to this crash, and as these are in almost as deep freeze as debian 2.0 is now :-), 2.2 is not too far away and the problem goes away. I do not yet see a reason to buy an Intel instead. Nils -- *-----------------------------------------------------------------------------* | Quotes from the net: L> Linus Torvalds, W> Winfried Truemper | | L>this is the special easter release of linux, more mundanely called 1.3.84 | | W>Umh, oh. What do you mean by "special easter release"?. Will it quit | * W>working today and rise on easter? *
Attachment:
pgp69NEos2HLF.pgp
Description: PGP signature