On Wed, 9 Sep 2020, Michael Matz wrote:
> Hello,
>
> On Tue, 8 Sep 2020, Richard Biener wrote:
>
> > CCing some people to double-check my graph partitioning algorithm.
>
> I seem to not know the pre-existing data structures enough to say anything
> about this, but I noticed small things which
Hello,
On Tue, 8 Sep 2020, Richard Biener wrote:
> CCing some people to double-check my graph partitioning algorithm.
I seem to not know the pre-existing data structures enough to say anything
about this, but I noticed small things which might or might not indicate
thinkos or incompleteness:
On Tue, 8 Sep 2020, Richard Biener wrote:
>
> This makes the BB vectorizer cost independent SLP subgraphs
> separately. While on pristine trunk and for x86_64 I failed to
> distill a testcase where the vectorizer would think _any_
> basic-block vectorization opportunity is not profitable I do
>
This makes the BB vectorizer cost independent SLP subgraphs
separately. While on pristine trunk and for x86_64 I failed to
distill a testcase where the vectorizer would think _any_
basic-block vectorization opportunity is not profitable I do
have pending work that would make the cost savings of
