Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: (AGP) related panic

On Mon, Sep 22, 2008 at 09:32:08PM +0400, Aleksej Saushev wrote:
> (...)
> Additional information: in another discussion thread,
> it was pointed, that this may be connected with erroneous
> "makeoptions CPUFLAGS", which allowed usage of SSE.
> Yet I have never allowed SSE in CPUFLAGS.
> Even more, I have rebuilt kernel to avoid that, the panic
> occurred in that case too. Following options were probed:
> makeoptions   CPUFLAGS="-march=prescott"
> makeoptions   CPUFLAGS="-march=i486 -mtune=prescott"
> Thus, hardly the problem is connected with SSE, in my opinion.
> Am I wrong?

Quote from gcc manpage:

"   -mtune=cpu-type
       Tune to cpu-type everything applicable about the generated code,
       except for the ABI and the set of available instructions.  The
       choices for cpu-type are:


       pentium4, pentium4m
           Intel Pentium4 CPU with MMX, SSE and SSE2 instruction set sup-

           Improved version of Intel Pentium4 CPU with MMX, SSE, SSE2 and
           SSE3 instruction set support.

           Improved version of Intel Pentium4 CPU with 64-bit extensions,
           MMX, SSE, SSE2 and SSE3 instruction set support.

To me it /does/ sound as if still sse / mmx code is involved. How about trying
a kernel with the plain base CPUFLAGS as they are in GENERIC
(-march=i486 -mtune=pentiumpro) [as I have suggested to you literally] - note 
that the gcc manpage doesn't mention SSE or MMX with even one syllable 
when describing the mtune=pentiumpro option. At least then you can rule out
that source of errors.


Home | Main Index | Thread Index | Old Index