By the way, I also see a green on MIPSel. Shouldn't that be orange
given the criteria as spelled out. Or did Jason fix the tuning code
for this machine?

Bill.

On 9 June 2010 04:12, Bill Hart <[email protected]> wrote:
> To get a pass on fulvia we need to figure out where the 64 bit version
> of lib_gcc.so is on the system and add that path to LD_LIBRARY_PATH
> before trying to configure MPIR.
>
> It's usually in a directory lib64 somewhere, but it might even just be
> missing from the system altogether, in which case there is no point
> testing MPIR on fulvia.
>
> There's also no point testing MPIR again on systems with gcc 4.3.2.
>
> But it would be a great idea to test on all the other systems again
> with the actual source tarball. Of course ./mpirtest will still fail
> for a large number of systems due to odd configurations (e.g. CC=icc
> CXX=g++).
>
> Bill.
>
> On 9 June 2010 04:03, Minh Nguyen <[email protected]> wrote:
>> Hi Bill,
>>
>> I have updated the rules for colour coding at www.mpir.org. I see this
>> as positive direction towards better documentation for MPIR. At least
>> there is now clear documentation on how to interpret the results
>> reported in the test matrix. I find the previous documentation very
>> fuzzy.
>>
>> As a double check, I'll now again take the MPIR 2.1.0 source tarball
>> and do another round of:
>>
>> * ./configure
>> * make
>> * make check
>> * make tune
>> * ./mpirtest
>>
>> Based on the results, I'll update the test matrix some time in the
>> next 24 hours.
>>
>> --
>> Regards
>> Minh Van Nguyen
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"mpir-devel" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/mpir-devel?hl=en.

Reply via email to