Wasmtime GitHub notifications bot (Oct 17 2023 at 22:00): Wasmtime GitHub notifications bot (Oct 17 2023 at 22:00):cfallin requested fitzgen for a review on PR #7274.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:00):cfallin opened PR #7274 from cfallin:pcc-wasm to bytecodealliance:main:
This PR wires up end-to-end support for validating the compilation of accesses to static memories in Wasmtime. It creates "memory types" for the vmcontext struct and the memories it points to, and it adds PCC annotations to the relevant memory pointers. On aarch64 (where instruction semantics for PCC are implemented), this validates a simple
i32.loadfunction body with default memory configuration.Note that facts are not yet propagated through egraph rewrites, so optimization has to be disabled for this to work; that support will come next.
Co-authored-by: Nick Fitzgerald <fitzgen@gmail.com>
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:00):cfallin requested wasmtime-compiler-reviewers for a review on PR #7274.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:00):cfallin requested wasmtime-core-reviewers for a review on PR #7274.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:01):cfallin updated PR #7274.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:47):cfallin updated PR #7274.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:55):cfallin updated PR #7274.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:58):fitzgen submitted PR review.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:58):fitzgen submitted PR review.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:58):fitzgen created PR review comment:
Can we also test when the
vmctxgv has a fact but the block param doesn't, so we exercise the fact copying in tests?
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:58):fitzgen created PR review comment:
We can actually use
heap.index_typealready, we don't need another parameter. I think we should at least assert on that value now, if we don't update the actual max now, so that it isn't silently incorrect.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:58):fitzgen created PR review comment:
I think this is done? Or is this referring to something else we still haven't done yet?
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:58):fitzgen created PR review comment:
Not 100% sure we want to leave this trace in.
Wasmtime GitHub notifications bot (Oct 17 2023 at 22:58):fitzgen created PR review comment:
Similarly, I don't think this trace will be useful long term.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin updated PR #7274.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin submitted PR review.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin created PR review comment:
Removed!
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin submitted PR review.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin created PR review comment:
Yep, just stale and I missed it -- thanks!
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin submitted PR review.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin created PR review comment:
Ah, yes, good point -- added that assert.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin submitted PR review.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin created PR review comment:
Added!
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin submitted PR review.
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:17):cfallin created PR review comment:
Removed!
Wasmtime GitHub notifications bot (Oct 17 2023 at 23:24):cfallin has enabled auto merge for PR #7274.
Wasmtime GitHub notifications bot (Oct 18 2023 at 00:32):cfallin merged PR #7274.
Last updated: Oct 23 2024 at 20:03 UTC