tschneidereit opened PR #13887 from tschneidereit:current-thread-fast-path to bytecodealliance:main:
Splitting current_thread into an #[inline] fast path plus #[cold] outlined deferred-promotion slow path reduces call overhead by about 20% (measured in a Linux VM on an M5 Max MBP): sync calls go from about 83ns to about 65ns, immediately ready async calls from 142ns to 117ns.
tschneidereit added the label wasmtime on PR #13887.
tschneidereit requested pchickey for a review on PR #13887.
tschneidereit added the label performance on PR #13887.
tschneidereit requested wasmtime-core-reviewers for a review on PR #13887.
tschneidereit updated PR #13887.
github-actions[bot] added the label wasmtime:api on PR #13887.
:thumbs_up: alexcrichton submitted PR review.
alexcrichton added PR #13887 Split current_thread into an inline fast path and a cold slow path to the merge queue.
github-merge-queue[bot] removed PR #13887 Split current_thread into an inline fast path and a cold slow path from the merge queue.
alexcrichton added PR #13887 Split current_thread into an inline fast path and a cold slow path to the merge queue.
github-merge-queue[bot] removed PR #13887 Split current_thread into an inline fast path and a cold slow path from the merge queue.
alexcrichton added PR #13887 Split current_thread into an inline fast path and a cold slow path to the merge queue.
github-merge-queue[bot] removed PR #13887 Split current_thread into an inline fast path and a cold slow path from the merge queue.
alexcrichton added PR #13887 Split current_thread into an inline fast path and a cold slow path to the merge queue.
:check: alexcrichton merged PR #13887.
alexcrichton removed PR #13887 Split current_thread into an inline fast path and a cold slow path from the merge queue.
Last updated: Jul 29 2026 at 05:03 UTC