[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Wrapper dylib to fix OSX libBLAS & libLAPACK with -m64 (Re: 3.5.0+ c
From: |
Jarno Rajahalme |
Subject: |
Re: Wrapper dylib to fix OSX libBLAS & libLAPACK with -m64 (Re: 3.5.0+ compiled on Mac OSX 10.6.6) |
Date: |
Tue, 1 Feb 2011 10:51:54 +0200 |
On Jan 31, 2011, at 17:27 , ext Richard Campbell wrote:
>
> On Jan 31, 2011, at 2:42 AM, Jarno Rajahalme wrote:
>
>>
>> On Jan 31, 2011, at 9:28 , ext Jarno Rajahalme wrote:
>>
>>>
>>>
>>> I think there is NO chance to use the Apple-provided blas/lapack with
>>> --enable-64, as --enable-64 requires recompilation of all fortran
>>> sources/dependencies with the corrensponding integer size setting.
>>>
>>
>> Must add that the error above has nothing to do with the need to recompile
>> fortran code.
>>
>> However, my point is that if you want to use the (wrapped) blas/lapack
>> provided by Apple, then this error does not matter. When/if you get it to
>> compile, it will crash, when calling the incompatible fortran code.
>>
>> Jarno
>
> Is this a problem that has been reported to Apple (and/or Intel, since they
> wrote this BLAS for their chips?)
>
> Campbell
I don't think so. I do not think this is a problem per se, just a matter of
definition. These libs just use 32 bit ("int") indexes. Even in 64-bit code, it
is up to the definition of a function to specify what kind of ints are used
(8,16,32,64, or 128 bits) for which argument. The fact that fortran is not
type-safe (same as C) makes dealing with this a bit more complicated.
Jarno
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Re: Wrapper dylib to fix OSX libBLAS & libLAPACK with -m64 (Re: 3.5.0+ compiled on Mac OSX 10.6.6),
Jarno Rajahalme <=