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

Re: [patches] powerpc 8xx dcbz problem



Mark Mitchell wrote:
> Steven Munroe wrote:
>
>   
>> So any fix/work-around should go in ports and should use --with-cpu. You
>> can override libc-start.c from ports/sysdeps/powerpc/powerpc32/8xx with
>> a version that zeros __cache_line_size for 8xx and sets its from
>> AT_DCACHEBSIZE otherwise.
>> .
>>     
>
> I don't understand what argument you would give to --with-cpu in this
> case?  I'm not an expert on Power, so forgive me if the particular CPUs
> I pick below doesn't make sense.
>
> Suppose that we want a GLIBC that runs on 603 or later processors,
> including 8xx.  So, we configure with --with-cpu=603.  Are you
> suggesting that we change the 603 libc-start.c do the check for 8xx?
> And, presumably, all other CPUs X such that code for X should run on an
> 8xx?  Won't that result in duplication along another dimension?
>
>   
Well for power4/5/6 I have to use lots of Implies files to pick up
usable optimizition from older processors.

I cant seem to find any documentation on the 8xx. Is it PowerPC IAS 1.0
or better? Then you could put this code in /powerpc/powerpc32/603  and
then Implies back to it from the various 7xx, 8xx processors you want to
support.

Or you to compromise I suggested to Nathan.