Stream: git-wasmtime

Topic: wasmtime / PR #4157 Upgrade to regalloc2 0.1.3.


view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 05:52):

cfallin opened PR #4157 from use-regalloc2-0.1.3 to main:

This pulls in bytecodealliance/regalloc2#49, which slightly improves
codegen in soem cases where a safepoint (for reference-typed values)
occurs in the same liverange as a register-constraineed use. For
example, in bytecodealliance/wasmtime#3785, an extra move instruction
appeared and a callee-save register was used (necessitating a more
expensive prologue) because of suboptimal splitting heuristics, which
this PR fixes. The updated RA2 heuristics appear to have no measured
downsides in existing benchmarks and improve the manually-observed
codegen issue.

<!--

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

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

view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 05:53):

cfallin updated PR #4157 from use-regalloc2-0.1.3 to main.

view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 05:53):

cfallin edited PR #4157 from use-regalloc2-0.1.3 to main:

This pulls in bytecodealliance/regalloc2#49, which slightly improves
codegen in some cases where a safepoint (for reference-typed values)
occurs in the same liverange as a register-constraineed use. For
example, in bytecodealliance/wasmtime#3785, an extra move instruction
appeared and a callee-save register was used (necessitating a more
expensive prologue) because of suboptimal splitting heuristics, which
this PR fixes. The updated RA2 heuristics appear to have no measured
downsides in existing benchmarks and improve the manually-observed
codegen issue.

<!--

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

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

view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 05:54):

cfallin edited PR #4157 from use-regalloc2-0.1.3 to main:

This pulls in bytecodealliance/regalloc2#49, which slightly improves
codegen in some cases where a safepoint (for reference-typed values)
occurs in the same liverange as a register-constrained use. For
example, in bytecodealliance/wasmtime#3785, an extra move instruction
appeared and a callee-save register was used (necessitating a more
expensive prologue) because of suboptimal splitting heuristics, which
this PR fixes. The updated RA2 heuristics appear to have no measured
downsides in existing benchmarks and improve the manually-observed
codegen issue.

<!--

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

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

view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 05:54):

cfallin updated PR #4157 from use-regalloc2-0.1.3 to main.

view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 17:36):

cfallin updated PR #4157 from use-regalloc2-0.1.3 to main.

view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 17:36):

cfallin requested fitzgen for a review on PR #4157.

view this post on Zulip Wasmtime GitHub notifications bot (May 17 2022 at 17:36):

cfallin requested abrown for a review on PR #4157.

view this post on Zulip Wasmtime GitHub notifications bot (May 18 2022 at 18:37):

abrown submitted PR review.

view this post on Zulip Wasmtime GitHub notifications bot (May 18 2022 at 18:48):

cfallin merged PR #4157.


Last updated: Nov 22 2024 at 16:03 UTC