Gcc 3.2.2 vs gcc 3.2.3

[Jens Rehsack]

Holger Kipp wrote:

...

> imho memtest seems to be suitable for this-is-real-broken-memory(tm) only.

Yes, memtest86 could only detect bad memory. No reported failure doesn't
says anything except memtest86 didn't found any error.

> It might be that due to differnent optimisations (gcc 3.2.2 vs gcc 3.2.3) 
> you might escape the bitpatterns that would otherwise trigger the sig 11.

There're several bugs known in gcc-3.2, some of them are removed in
gcc-3.3, some of them are scheduled to be fixed in gcc-3.4.
The best recommendation to optimization (if you don't want to research
atr gcc.gnu.org) is using -O and not more.

Regards,
Jens