On Thu, Sep 24, 2015 at 09:46:37PM -0400, Michael Shell wrote:
> On Fri, 25 Sep 2015 01:06:12 +0100
> Ken Moffat <zarniwh...@ntlworld.com> wrote:
> 
> > An ICE (internal compiler error) means something *apparently* went
> > wrong in your compiler.  On one of my machines (an AMD phenom - they
> > are notorious for this), building with -j4 often provokes this sort
> > of error, particularly when building a new kernel.
> 
> 
>   Ken,
> 
> Do you have any knowledge about how the newer AMD processors
> (Phenom II, A10, etc.) hold up in this regard?

My phenom is actually AMD Phenom(tm) II X4 965 Processor

I know of one person, who sometimes posts on lkml, who also has a
similar processor and now never goes beyond make -j4.

My own A4, when it was alive, showed no problems (apart from only
having two cores).  My A10 (kaveri) is fine, and there I use make
-j5 because it has an SSD.  So, from my experience only the Phenom
is a problem.

OTOH, in many things they are slower than my SandyBridge i3, and in
particular an SBU (single threaded) is much slower - some of that,
but not all, might be because the acpi cpufreq governor works less
well on AMD CPUs than the old K10 code did.
> 
> As to the problem of the OP, is it possible that gcc was not complied
> for the correct target such that in unusual circumstances it really
> will attempt to execute an unsupported CPU instruction?
> 
> If a reattempt is made to build atk (after cleaning the source tree)
> does the exact same error occur at the same place? If so, a hardware
> problem is less likely.
> 

Yes, those both sound sensible.

> Another possibility is some kind of stack corruption (caused by
> a bug in gcc or one of the libraries it depends on) where data
> is somehow overwriting exec code.
> 

For that, I hope not ;-)  That does sound suspiciously like what the
phenom appears to do on its bad days.  But we don't know what CPU the
OP was using.

ĸen
-- 
Il Porcupino Nil Sodomy Est! (if you will excuse my latatian)
  aka "The hedgehog song"
-- 
http://lists.linuxfromscratch.org/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to