Stream: git-wasmtime

Topic: wasmtime / PR #1335 Infer REX prefix for SIMD operations;...

Wasmtime GitHub notifications bot (Mar 17 2020 at 01:28):

abrown opened PR #1335 from infer-rex to master:

• Convert recipes to have necessary size calculator
• Add a missing binemit function, put_dynrexmp3
• Modify the meta-encodings of x86 SIMD instructions to use infer_rex(), mostly through the enc_both_inferred() helper
• Fix up tests that previously always emitted a REX prefix

<!--

Please ensure that the following steps are all taken care of before submitting
the PR.

• [ ] This has been discussed in issue #..., or if not, please tell us why
  here.

• [ ] A short description of what this does, why it is needed; if the
  description becomes long, the matter should probably be discussed in an issue
  first.

• [ ] This PR contains test cases, if meaningful.

• [ ] A reviewer from the core maintainer team has been assigned for this PR.
  If you don't know who could review this, please indicate so. The list of
  suggested reviewers on the right can help you.

Please ensure all communication adheres to the code of conduct.
-->

Wasmtime GitHub notifications bot (Mar 17 2020 at 01:28):

abrown requested bnjbvr for a review on PR #1335.

Wasmtime GitHub notifications bot (Mar 17 2020 at 01:28):

abrown requested bnjbvr, and iximeow for a review on PR #1335.

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr submitted PR Review.

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr submitted PR Review.

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr created PR Review Comment:

As discussed, whenever we need the REX.W bit to be set, we can just use rex() instead of infer_rex(), so it avoids the runtime decision.

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr created PR Review Comment:

ditto here

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr created PR Review Comment:

Do we need the clone on template.clone()? infer_rex takes &self and returns a copy, if I'm not mistaken.

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr created PR Review Comment:

ditto here

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr created PR Review Comment:

nit: ditto here

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr created PR Review Comment:

Ditto here.

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:26):

bnjbvr created PR Review Comment:

If that's not too much hassle, can you, for each different recipe involved, duplicate the test with a register that would require a REX prefix, please?

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:30):

abrown submitted PR Review.

Wasmtime GitHub notifications bot (Mar 17 2020 at 16:30):

abrown created PR Review Comment:

I will change this and I opened #1342 to avoid this type of thing altogether.

Wasmtime GitHub notifications bot (Mar 18 2020 at 16:35):

abrown updated PR #1335 from infer-rex to master:

• Convert recipes to have necessary size calculator
• Add a missing binemit function, put_dynrexmp3
• Modify the meta-encodings of x86 SIMD instructions to use infer_rex(), mostly through the enc_both_inferred() helper
• Fix up tests that previously always emitted a REX prefix

<!--

Please ensure that the following steps are all taken care of before submitting
the PR.

• [ ] This has been discussed in issue #..., or if not, please tell us why
  here.

• [ ] A short description of what this does, why it is needed; if the
  description becomes long, the matter should probably be discussed in an issue
  first.

• [ ] This PR contains test cases, if meaningful.

• [ ] A reviewer from the core maintainer team has been assigned for this PR.
  If you don't know who could review this, please indicate so. The list of
  suggested reviewers on the right can help you.

Please ensure all communication adheres to the code of conduct.
-->

Wasmtime GitHub notifications bot (Mar 18 2020 at 16:40):

abrown submitted PR Review.

Wasmtime GitHub notifications bot (Mar 18 2020 at 16:40):

abrown created PR Review Comment:

Ok, I added some REX-inducing tests; it's not 100% coverage but that is a rather high bar with all of the variations of these instructions.

Wasmtime GitHub notifications bot (Mar 18 2020 at 17:12):

abrown merged PR #1335.

Last updated: Oct 23 2024 at 20:03 UTC