"Illegal instruction" traps on smp clients - 2.4.19

[Benjamin Herrenschmidt]

On Thu, 2003-02-27 at 16:57, Ethan Weinstein wrote:
> Rudy Klinksiek wrote:
>
> | 	Smp and UP kernels work fine on the "server".
> |
> | 	Has anyone else seen this type of problem or something similar?
> |
>
> I have. I run a dual G4(7400), also with CONFIG_HIMEM and got plenty of
> random SIGILL's, SIGABRT's and deamons crashing upon startup until I
> started using Ben's bk tree.  I've been using 2.4.20-ben(x) for quite
> some time now without this problem.
>
> |         This appears to me to be an smp problem.
>
> | 	And then, what can I do about it?  I'm willing to try things
> | 	as my time permits.  I have looked at 2.5.60 memory.c/mmap.c
> | 	and related functions, and trying to port the new methods
> | 	back to 2.4.19 seems to be a rather daunting task.
> |
> |         Comments, suggestions?
> |
> Try rsyncing ben's bk tree and using it, it cleared up this issue for me.
>
> rsync -avz --delete rsync.penguinppc.org::linux-2.4-benh

This is actually my "stable" tree (2.4.20-ben7 at this time). Please
let me know if it's also fixed by the linuxppc_2_4 PPC bk tree (see
penguinppc.org for details on how to get these).

Ben.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/