[llvm-dev] [LoopVectorizer] Improving the performance of dot product reduction loop

[Hal Finkel via llvm-dev]

On 07/24/2018 07:58 AM, Renato Golin wrote:
> On Tue, 24 Jul 2018 at 13:46, Hal Finkel via llvm-dev
> <llvm-dev at lists.llvm.org> wrote:
>> Ah, interesting. I was indeed wondering whether this was another case where we'd benefit from having vectorizer-maximize-bandwidth on by default. I recall that our defaults have long been to prefer smaller VF in cases where the costs are equal to avoid extra legalization costs (and, potentially, to retain more freedom for interleaving later). Should the costs here be equal, or should we have some extra target information to distinguish them? It sounds like they're not really equal in practice.
> Hi Hal,
>
> Enabling vectorizer-maximize-bandwidth by default has been explored
> before (see history in https://reviews.llvm.org/D46283) and it created
> a handful of problems.
>
> With the correctness problems fixed, we still were blocked on high
> negative performance hits, which was probably to do with the cost
> model. Adhemerval sent this patch instead:
> https://reviews.llvm.org/D48332, which dealt with the specific case in
> hand.

This matches my recollection: the correctness problems had been
addressed and there were still some outstanding cost-modeling problems.

>
> I remember hearing that some complex loops get worse performance for
> wide loads (ex. AVX512) because it makes the loop too short and the
> shuffles in and out too long, or increase the number of shuffles if
> the loads are not trivial.

Sounds like another potential cost-modeling problem?

>
> So, while enabling it by default is probably a good idea in a lot of
> cases, we probably need to be careful with its usage as a wide scope
> default. I don't have more info on the individual cases, though, so
> just is more of an FYI. :)

I don't really see any long-term path to addressing these kinds of
problems aside from working through the cost-modeling problems and
getting this enabled. We'll see ;)

 -Hal

>

-- 
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory