Re: [RFC] Remove VRP threader passes in exchange for better threading pre-VRP.

[Richard Biener <richard.guenther@gmail.com>]

On Tue, Oct 19, 2021 at 9:33 AM Aldy Hernandez <aldyh@redhat.com> wrote:
>
> On Tue, Oct 19, 2021 at 8:52 AM Richard Biener
> <richard.guenther@gmail.com> wrote:
> >
> > On Mon, Oct 18, 2021 at 4:03 PM Aldy Hernandez <aldyh@redhat.com> wrote:
> > >
> > >
> > >
> > > On 10/18/21 3:41 PM, Aldy Hernandez wrote:
> > >
> > > > I've been experimenting with reducing the total number of threading
> > > > passes, and I'd like to see if there's consensus/stomach for altering
> > > > the pipeline.  Note, that the goal is to remove forward threader clients,
> > > > not the other way around.  So, we should prefer to remove a VRP threader
> > > > instance over a *.thread one immediately before VRP.
> > > >
> > > > After some playing, it looks like if we enable fully-resolving mode in
> > > > the *.thread passes immediately preceeding VRP, we can remove the VRP
> > > > threading passes altogether, thus removing 2 threading passes (and
> > > > forward threading passes at that!).
> > >
> > > It occurs to me that we could also remove the threading before VRP
> > > passes, and enable a fully-resolving backward threader after VRP.  I
> > > haven't played with this scenario, but it should be just as good.  That
> > > being said, I don't know the intricacies of why we had both pre and post
> > > VRP threading passes, and if one is ideally better than the other.
> >
> > It was done because they were different threaders.  Since the new threader
> > uses built-in VRP it shouldn't really matter whether it's before or after
> > VRP _for the threading_, but it might be that if threading runs before VRP
> > then VRP itself can do a better job on cleaning up the IL.
>
> Good point.
>
> FWIW, earlier this season I played with replacing the VRPs with evrp
> instances (which fold far more conditionals) and I found that the
> threaders can actually find LESS opportunities after *vrp fold away
> things.  I don't know if this is a good or a bad thing.

Probably a sign that either threading theads stuff that's pointless
(does not consider conditions on the path that always evaluate false?)
or that after VRP and removing redundant conditions blocks are now
bigger and we run into some --param limit (that would suggest the
limits behave odd if it would thread when we'd artificially split blocks).

Examples might be interesting to look at to understand what's going "wrong".

> Perhaps we
> should benchmark three alternatives:
>
> 1. Mainline
> 2. Fully resolving threader -> VRP -> No threading.
> 3. No threading -> VRP -> Full resolving threader.
>
> ...and see what the actual effect is, regardless of number of threaded paths.

As said, only 2. makes "sense" to me unless examples show why we really
have the usual pass ordering issue.  As said, I think threading exposes new
VRP (esp. constant/copy prop) opportunities but VRP shouldn't expose new
threading opportunities.

> Speak of which, what's the blessed way of benchmarking performance
> nowadays?  I've seen some PRs fly that measure some more lightweight
> benchmarks (open source?) than a full blown SPEC.

The answer is SPEC.  The other answer would be GCC bootstrap time
and resulting code size.

> >
> > +      /* ?? Is this still needed.  ?? */
> >        /* Threading can leave many const/copy propagations in the IL.
> >          Clean them up.  Instead of just copy_prop, we use ccp to
> >          compute alignment and nonzero bits.  */
> >
> > Yes, it's still needed but not for the stated reason - the VRP
> > substitution and folding stage should deal with copy/constant propagation
> > but we replaced the former copy propagation with CCP to re-compute
> > nonzero bits & alignment so I'd change the comment to
> >
> >    /* Run CCP to compute alignment and nonzero bits.  */
>
> Ahh..
>
> There's another similar comment after DOM.  Is this comment still relevant?

Yes, since DOM still does threading the threaded paths lack const/copy
propagation.

>       NEXT_PASS (pass_dominator, true /* may_peel_loop_headers_p */);
>       /* Threading can leave many const/copy propagations in the IL.
>      Clean them up.  Failure to do so well can lead to false
>      positives from warnings for erroneous code.  */
>       NEXT_PASS (pass_copy_prop);
>       /* Identify paths that should never be executed in a conforming
>      program and isolate those paths.  */
>
> Thanks.
> Aldy
>