Allow reifying intrinsics to fn pointers. (rebase of #86699)

[GrigorenkoPV]

The original PR (#86699) died after merge conflicts due to inactivity. This PR is just a rebase of the original with some minor changes to make it fit into the now-nearly-3-years-older rustc. Such as using //@ instead of // for test directives, the removal of Abi::PlatformIntrinsic in #121516, etc.

I am not very familiar with the code surrounding the changes, so review this with caution. There might have been some new problems that popped up during those 3 years that I have no idea about.

Might allow to re-revert #81238.

[rustbot]

r? @pnkfelix

rustbot has assigned @pnkfelix.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[rustbot]

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

[compiler-errors]

I'll review this soon

[compiler-errors]

Why was this added?

[compiler-errors]

Actually, I know why. I wonder if it's better to place this in build_call_shim, or even make our own function like build_reified_intrinsic.

[compiler-errors]

Generally looks fine.

I think this needs a T-types FCP since AFAICT the only way to hit this is on nightly code. Is that last part correct?

Regardless, do you mind writing up a short blurb motivating why we want to do this and explaining the general approach?

[compiler-errors]

First this needs a rebase and also please fix the CI issues

@rustbot author

[GrigorenkoPV]

First this needs a rebase

Done

and also please fix the CI issues

Sorry, I've tried, but couldn't figure out what the problem was and why it didn't pick up the mir test. I would appreciate if somebody could help me here.

[GrigorenkoPV]

@rustbot label +E-help-wanted

[bors]

☔ The latest upstream changes (presumably #127831) made this pull request unmergeable. Please resolve the merge conflicts.

[jieyouxu]

Maybe

// EMIT_MIR lower_intrinsics.reify_intrinsics.LowerIntrinsics.mir

?

[GrigorenkoPV]

Well, I think this should be a .diff to preserve the original intentions. Plus I was not able to make it generate the same code as in #86699.

[GrigorenkoPV]

I've rebased and found a place where this ICEs. I am not sure whether the ICE was there originally or it got introduced after one of the rebases.

[rust-log-analyzer]

The job mingw-check-tidy failed! Check out the build log: (web) (plain)

Click to see the possible cause of the failure (guessed by this bot)

COPY host-x86_64/mingw-check/validate-toolstate.sh /scripts/
COPY host-x86_64/mingw-check/validate-error-codes.sh /scripts/

# NOTE: intentionally uses python2 for x.py so we can test it still works.
# validate-toolstate only runs in our CI, so it's ok for it to only support python3.
ENV SCRIPT TIDY_PRINT_DIFF=1 python2.7 ../x.py test \
           --stage 0 src/tools/tidy tidyselftest --extra-checks=py:lint,cpp:fmt
# This file is autogenerated by pip-compile with Python 3.10
# by the following command:
#
#    pip-compile --allow-unsafe --generate-hashes reuse-requirements.in
---
#13 3.083 Building wheels for collected packages: reuse
#13 3.084   Building wheel for reuse (pyproject.toml): started
#13 3.339   Building wheel for reuse (pyproject.toml): finished with status 'done'
#13 3.340   Created wheel for reuse: filename=reuse-4.0.3-cp310-cp310-manylinux_2_35_x86_64.whl size=132720 sha256=026f3bb0f1aa8090b861fd0a0939cb1a782396d84c8aab7875096557d637a0f6
#13 3.341   Stored in directory: /tmp/pip-ephem-wheel-cache-uid128zr/wheels/3d/8d/0a/e0fc6aba4494b28a967ab5eaf951c121d9c677958714e34532
#13 3.343 Installing collected packages: boolean-py, binaryornot, tomlkit, reuse, python-debian, markupsafe, license-expression, jinja2, chardet, attrs
#13 3.749 Successfully installed attrs-23.2.0 binaryornot-0.4.4 boolean-py-4.0 chardet-5.2.0 jinja2-3.1.4 license-expression-30.3.0 markupsafe-2.1.5 python-debian-0.1.49 reuse-4.0.3 tomlkit-0.13.0
#13 3.749 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv
#13 4.304 Collecting virtualenv
#13 4.304 Collecting virtualenv
#13 4.389   Downloading virtualenv-20.27.0-py3-none-any.whl (3.1 MB)
#13 4.518      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.1/3.1 MB 24.6 MB/s eta 0:00:00
#13 4.574 Collecting filelock<4,>=3.12.2
#13 4.593   Downloading filelock-3.16.1-py3-none-any.whl (16 kB)
#13 4.614 Collecting distlib<1,>=0.3.7
#13 4.631   Downloading distlib-0.3.9-py2.py3-none-any.whl (468 kB)
#13 4.640      ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 469.0/469.0 KB 72.0 MB/s eta 0:00:00
#13 4.675 Collecting platformdirs<5,>=3.9.1
#13 4.724   Downloading platformdirs-4.3.6-py3-none-any.whl (18 kB)
#13 4.811 Installing collected packages: distlib, platformdirs, filelock, virtualenv
#13 4.990 Successfully installed distlib-0.3.9 filelock-3.16.1 platformdirs-4.3.6 virtualenv-20.27.0
#13 DONE 5.1s

#14 [7/8] COPY host-x86_64/mingw-check/validate-toolstate.sh /scripts/
#14 DONE 0.0s
---
DirectMap4k:      184256 kB
DirectMap2M:     9252864 kB
DirectMap1G:     9437184 kB
##[endgroup]
Executing TIDY_PRINT_DIFF=1 python2.7 ../x.py test            --stage 0 src/tools/tidy tidyselftest --extra-checks=py:lint,cpp:fmt
+ TIDY_PRINT_DIFF=1 python2.7 ../x.py test --stage 0 src/tools/tidy tidyselftest --extra-checks=py:lint,cpp:fmt
    Finished `dev` profile [unoptimized] target(s) in 0.04s
##[endgroup]
downloading https://static.rust-lang.org/dist/2024-10-16/rustfmt-nightly-x86_64-unknown-linux-gnu.tar.xz
extracting /checkout/obj/build/cache/2024-10-16/rustfmt-nightly-x86_64-unknown-linux-gnu.tar.xz to /checkout/obj/build/x86_64-unknown-linux-gnu/rustfmt
---
##[endgroup]
fmt check
fmt: checked 5635 files
tidy check
tidy error: the following output file is not associated with any mir-opt test, you can remove it: /checkout/tests/mir-opt/lower_intrinsics.core.intrinsics-#1-wrapping_add.LowerIntrinsics.diff
tidy error: the following output file is not associated with any mir-opt test, you can remove it: /checkout/tests/mir-opt/lower_intrinsics.core.intrinsics-#1-discriminant_value.LowerIntrinsics.diff
tidy error: the following output file is not associated with any mir-opt test, you can remove it: /checkout/tests/mir-opt/lower_intrinsics.core.intrinsics-#1-unreachable.LowerIntrinsics.diff
tidy error: the following output file is not associated with any mir-opt test, you can remove it: /checkout/tests/mir-opt/lower_intrinsics.core.intrinsics-#1-size_of.LowerIntrinsics.diff
tidy error: the following output file is not associated with any mir-opt test, you can remove it: /checkout/tests/mir-opt/lower_intrinsics.core.intrinsics-#1-forget.LowerIntrinsics.diff
removing old virtual environment
removing old virtual environment
creating virtual environment at '/checkout/obj/build/venv' using 'python3.10'
Requirement already satisfied: pip in ./build/venv/lib/python3.10/site-packages (24.2)
  Downloading pip-24.3.1-py3-none-any.whl.metadata (3.7 kB)
Downloading pip-24.3.1-py3-none-any.whl (1.8 MB)
   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.8/1.8 MB 35.1 MB/s eta 0:00:00
Installing collected packages: pip
---
Successfully installed pip-24.3.1
linting python files
All checks passed!
checking C++ file formatting
some tidy checks failed
Command has failed. Rerun with -v to see more details.
  local time: Mon Oct 28 13:07:18 UTC 2024
  network time: Mon, 28 Oct 2024 13:07:18 GMT
##[error]Process completed with exit code 1.
Post job cleanup.

[GrigorenkoPV]

I've rebased and found a place where this ICEs. I am not sure whether the ICE was there originally or it got introduced after one of the rebases.

I've checked and this ICE has been present since the first version of this PR. I was not able to check whether it existed in the original #86699, because I couldn't get that old version of the compiler to build.