[llvm-dev] New pass manager for optimization pipeline status and questions

Thu Jul 23 09:27:29 PDT 2020

While I'm also excited about moving to the new pass manager, I do
think you may be overestimating how many users have switched when you
write that the vast majority of testing is now using the NPM.

My guess would instead be that the vast majority of projects using
LLVM are using default pass manager. I know Chrome uses the default. I
know that's how the upstream releases are built and tested. Defaults
tend to be popular. All our docs about writing passes and such seem to
assume the default pm.

I guess what I'm saying is I think the transition has to be more
careful than biting the bullet and throwing the switch today.

On Thu, Jul 23, 2020 at 12:15 AM Philip Reames via llvm-dev
<llvm-dev at lists.llvm.org> wrote:
>
> (I'm probably going to derail your thread, sorry about that.)
>
> I think at this point, we should just bite the bullet and make the switch to NPM by default for Clang's optimization pipeline.  Today.
>
> Why?  Because many of our downstream consumers have already switched.  Google has.  We (Azul) have.  I think I've heard the same for a couple other major contributors.  Why does this matter?  Testing.  At the current moment, the vast majority of testing the project gets is exercising NPM, not LPM.
>
> NPM is functionally complete for Clang optimization.  There might be a few missing cases around the sanitizers, but last I heard those were on the edge of being fixed.
>
> I think we should make the switch, and deal with any fall out as regressions.  If we made the change immediately after a release branch, we'd have several months to address any major issues before the next release.
>
> Philip
>
> On 7/22/20 2:39 PM, Arthur Eubanks via llvm-dev wrote:
>
> Hi all,
>
> I wanted to give a quick update on the status of NPM for the IR optimization pipeline and ask some questions.
>
> In the past I believe there were thoughts that NPM was basically ready because all of check-llvm and check-clang passed when -DENABLE_EXPERIMENTAL_NEW_PASS_MANAGER=ON was specified. But that CMake flag did not apply to opt and any tests running something like `opt -foo-pass -bar-pass` (which is the vast majority of check-llvm tests) were still using the legacy PM. The intended way to use NPM was to use the -passes flag, e.g. `opt -passes='foo,bar'`.
>
> I've added a -enable-new-pm flag to opt to force running NPM passes even when `opt -foo-pass` is used. This is because I didn't want to go through every single test and figure out which ones should be using both -foo-pass and -passes=foo. Switching on -enable-new-pm currently leads to ~1800 check-llvm failures. I've documented the failed tests count per directory in https://bugs.llvm.org/show_bug.cgi?id=46651 (some have been fixed since that was posted).
>
> This has led to real bugs in NPM being discovered and fixed (e.g. some optnone issues).
>
> But a large portion of the remaining failures are because codegen-only passes haven't been ported to NPM yet. That's fine for the optimization pipeline NPM transition since it doesn't affect the optimization pipeline, but it does present an issue with the approach of the -enable-new-pm flag (which would by default become true alongside the NPM transition). Lots of tests are testing codegen-specific passes via opt (e.g. `opt -amdgpu-lower-intrinsics`) and they can't use NPM (yet).
>
> I was thinking either we have a way of identifying codegen-only passes and revert back to the legacy PM in opt whenever we see one, or we go back to considering the originally intended approach of adding an equivalent `-passes=` RUN to all tests that should be also running under NPM.
>
> I'm not sure of a nice and clean solution to identify codegen-only passes. We could go and update every instance of INITIALIZE_PASS to take another parameter indicating if it's codegen-only. Or we could just have a central list somewhere where we check if the pass is in some hardcoded list or has some prefix (e.g. "x86-").
>
> The approach of adding equivalent `-passes=` RUN lines to all relevant tests seems daunting, but not exactly sure how daunting. Maybe it's possible to script something and see what fails? We'd still need some way to identify codegen-only passes to make sure we don't miss anything, and we'd need to distinguish between analyses and normal passes. Also, it would slow down test execution since we'd run a lot more tests twice, but maybe that's not such a big deal? Maybe it's good to have most tests running against the legacy PM even when NPM is on by default?
>
> Thoughts?
>
> This is split off from http://lists.llvm.org/pipermail/llvm-dev/2020-July/143395.html.
>
>
>
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev