On 09/29/2015 06:26 PM, Sebastian Pop wrote:
This code is not used anymore after we removed the previous loop optimizer (not
based on the ISL scheduler.) We will add back the detection of commutative
reductions after we improve the code generation of scalar dependences (by not
going out of SSA for scalar dependences just to expose them to the data
dependence graph.)
Patch passed bootstrap and check on x86_64-linux with ISL-0.15.
I will commit this patch to trunk.
LGTM.
Regarding the handling of scalars, Polly does this now by only virtually
modeling
them as memory dependences, but leaving them as registers until code generation.
The final code generation is still done by alloca(ing) a memory slot and then
generating loads/stores from this memory slot. This is significantly easier than
trying to directly generate SSA again.
Best,
Tobias
