Wasmtime GitHub notifications bot (Feb 22 2024 at 18:41): Wasmtime GitHub notifications bot (Feb 22 2024 at 18:41):elliottt opened PR #7979 from elliottt:trevor/winch-trampoline-stack-checks to bytecodealliance:main:
Stack overflow checking in trampolines is slightly different than when translating arbitrary wasm functions, as the vmctx isn't necessarily in the pinned register yet. To avoid bloating the function prologue, I opted to parameterize the
MacroAssembler::check_stackfunction on the register that currently contains thevmctx. This change allows the trampolines to pass the current location ofvmctxinstead of forcing it to live in the pinned register.
<!--
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
-->
Wasmtime GitHub notifications bot (Feb 22 2024 at 18:41):elliottt requested fitzgen for a review on PR #7979.
Wasmtime GitHub notifications bot (Feb 22 2024 at 18:41):elliottt requested wasmtime-compiler-reviewers for a review on PR #7979.
Wasmtime GitHub notifications bot (Feb 22 2024 at 18:42):elliottt requested saulecabrera for a review on PR #7979.
Wasmtime GitHub notifications bot (Feb 22 2024 at 18:44):elliottt edited PR #7979:
Stack overflow checking in trampolines is slightly different than when translating arbitrary wasm functions, as the vmctx isn't necessarily in the pinned register yet. To avoid bloating the function prologue, I opted to parameterize the
MacroAssembler::check_stackfunction on the register that currently contains thevmctx. This change allows the trampolines to pass the current location ofvmctxinstead of forcing it to live in the pinned register.The first commit implements this change, while the second was a cleanup that I made while I was editing these files: I wanted to make more consistent use of the derived float register byte-size used while determining the space required for clobbers.
<!--
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
-->
Wasmtime GitHub notifications bot (Feb 22 2024 at 18:53):saulecabrera submitted PR review:
Looks great to me, thanks!
Wasmtime GitHub notifications bot (Feb 22 2024 at 19:21):elliottt merged PR #7979.
Last updated: Dec 23 2024 at 12:05 UTC