Wasmtime GitHub notifications bot (Dec 13 2022 at 23:47): Wasmtime GitHub notifications bot (Dec 13 2022 at 23:47):elliottt edited PR #5430 from trevor/fixed-nonallocatable-aliases to main:
The only situation where we use physical registers now is for fixed non-allocatable registers, all other uses should be virtual register. As such, it no longer makes sense to track aliases of registers that aren't virtual, as that would imply that it's acceptable to satisfy a constraint by using a non-allocatable register.
Fixes #5425
<!--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 (Dec 13 2022 at 23:48):elliottt edited PR #5430 from trevor/fixed-nonallocatable-aliases to main:
The only situation where we use physical registers now is for fixed non-allocatable registers, all other uses should be virtual register. As such, it no longer makes sense to report moves of registers that aren't virtual to regalloc2, as that would imply that it's acceptable to satisfy a constraint by using a non-allocatable register.
Fixes #5425
<!--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 (Dec 14 2022 at 00:51):elliottt updated PR #5430 from trevor/fixed-nonallocatable-aliases to main.
Wasmtime GitHub notifications bot (Dec 14 2022 at 00:54):elliottt submitted PR review.
Wasmtime GitHub notifications bot (Dec 14 2022 at 00:54):elliottt created PR review comment:
Reusing the cached decision about what instruction is a move meant that we keep the move instructions that involve fixed non-allocatable registers in the final assembly.
Wasmtime GitHub notifications bot (Dec 14 2022 at 00:59):elliottt submitted PR review.
Wasmtime GitHub notifications bot (Dec 14 2022 at 00:59):elliottt created PR review comment:
@uweigand would you mind checking these move changes? They all seemed reasonable to me, as the result of each function is
i32.
Wasmtime GitHub notifications bot (Dec 14 2022 at 01:09):cfallin submitted PR review.
Wasmtime GitHub notifications bot (Dec 14 2022 at 01:47):elliottt updated PR #5430 from trevor/fixed-nonallocatable-aliases to main.
Wasmtime GitHub notifications bot (Dec 14 2022 at 01:50):elliottt requested cfallin for a review on PR #5430.
Last updated: Dec 23 2024 at 12:05 UTC