alexcrichton opened PR #6372 from alexcrichton:x64-sse2-sshr
to bytecodealliance:main
:
The previous implementation fell back to a scalar implementation by extracting lanes and re-inserting them. These operations are significantly more complicated without SSE 4.1, however. Comparing against LLVM's lowerings, as well, it looks like a different strategy is used which avoids extracting and inserting lanes. This commit updates to follow the strategy matched by LLVM's lowerings which additionally has the benefit of providing a non-SSE4.1 implementation.
<!--
Please make sure you include the following information:
If this work has been discussed elsewhere, please include a link to that
conversation. If it was discussed in an issue, just mention "issue #...".Explain why this change is needed. If the details are in an issue already,
this can be brief.Our development process is documented in the Wasmtime book:
https://docs.wasmtime.dev/contributing-development-process.htmlPlease ensure all communication follows the code of conduct:
https://github.com/bytecodealliance/wasmtime/blob/main/CODE_OF_CONDUCT.md
-->
alexcrichton requested wasmtime-compiler-reviewers for a review on PR #6372.
alexcrichton requested fitzgen for a review on PR #6372.
alexcrichton requested abrown for a review on PR #6372.
alexcrichton requested wasmtime-compiler-reviewers for a review on PR #6372.
abrown submitted PR review.
abrown created PR review comment:
I guess since we're here we could remove this comment: "but this backend does not currently have support for EVEX encodings." That is no longer true and
vpsraq
is an instruction lowering that could be added eventually (an opposite direction from this PR which is adding lowerings for older features).
abrown submitted PR review.
alexcrichton updated PR #6372.
alexcrichton has enabled auto merge for PR #6372.
alexcrichton merged PR #6372.
Last updated: Dec 23 2024 at 12:05 UTC