My GCC Summit paper about Fortran Front End optimizations is accepted ...

[Toon Moene]

FX Coudert wrote:
>>                          T
>> c. Special treatment of A . B
>>
>>    FX Coudert ?

> Also, the handling of matmul(transpose(a),b) was 
> preexistent (although it's reused in the case of BLAS, which itself 
> knows how to perform this operation). Now, of course, it still gain us 
> significant time for large arrays (there's a compile-time threshold the 
> user can tweak to suit his system); I remember having done & posted some 
> benchmarking at the time, but can't find it on the list archives.

Thanks - my, as yet shallow, recollection of this feature didn't figure 
out the pre-existence of matmul(transpose(a),b).

Anyone wants to claim it ?  Is it primordial (i.e., done by Paul Brook 
and Steven Bosscher when they wrote the gfortran-internal-trees -> 
generic interface - the trans*.c files ?)

-- 
Toon Moene - e-mail: toon@moene.indiv.nluug.nl - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
At home: http://moene.indiv.nluug.nl/~toon/
Who's working on GNU Fortran: 
http://gcc.gnu.org/ml/gcc/2007-01/msg00059.html