Open MPI logo

Hardware Locality Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Hardware Locality Development mailing list

Subject: Re: [hwloc-devel] hwloc-1.3.1 and 1.4 failures on MIPS64/qemu
From: Paul H. Hargrove (PHHargrove_at_[hidden])
Date: 2012-02-03 14:03:12

On 2/3/2012 10:01 AM, Samuel Thibault wrote:
> Paul H. Hargrove, le Wed 01 Feb 2012 20:18:59 +0100, a écrit :
>> So if qemu's emulation were at fault, then one would expect more wide-spread
>> problems and a generally unstable system.
>> While -mabi=64 is not the default in user-space, it IS the way the linux
>> kernel is built and thus should be well tested.
> Except that it does not use floating point operations. I've commited a
> "fix". It makes the failures change. I got fun things with the mips64
> libc:
> fprintf(stderr,"%f\n", min_distance);
> fprintf(stderr,"%f\n", accuracy);
> fprintf(stderr,"%f %f\n", min_distance, min_distance);
> fprintf(stderr,"%f %f\n", min_distance, accuracy);
> will print
> 2.0
> 0
> 2.0 2.0
> 0.0 0.0
> ! It seems lenny on qemu-mips64 is not really stable with floats yet :)
> Samuel

Now I am confused. My qeum is simulating the 5Kc processor.
I have been configuring w/ CFLAGS="-march=5kc -mabi=64"
5Kc = single issue no floating-point
5Kf = adds floating-point unit that can issue indep. of integer unit.

So, are the references to floating-point some thing *I* should retest on
my platform?
Perhaps the "lenny on qemu-mips64" is saying the SOFT FP support is broken?
Or perhaps this is something discovered in testing that is unrelated to
my failure?
I could use a clue.

-Paul [who might try again today to get qemu/mips64/squeeze working]

Paul H. Hargrove                          PHHargrove_at_[hidden]
Future Technologies Group
HPC Research Department                   Tel: +1-510-495-2352
Lawrence Berkeley National Laboratory     Fax: +1-510-486-6900