Bug 169722 - WebAssembly: spec-tests/f32.wast.js fails on ARM64
Summary: WebAssembly: spec-tests/f32.wast.js fails on ARM64
Status: NEW
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
Keywords: InRadar
Depends on:
Blocks: 161709
  Show dependency treegraph
Reported: 2017-03-15 17:14 PDT by JF Bastien
Modified: 2020-11-27 08:50 PST (History)
8 users (show)

See Also:


Note You need to log in before you can comment on or make changes to this bug.
Description JF Bastien 2017-03-15 17:14:02 PDT
# DYLD_FRAMEWORK_PATH=... .../jsc -m --useWebAssembly=1 ./spec-tests//f32.wast.js
Exception: Error: Unreachable code should not be executed (evaluating '$$.exports["assert_0"]()')
<wasm>@[wasm code]
<wasm>@[wasm code]
assert_0@[native code]
module code@/private/var/root/jsc-stress-results/.tests/wasm/spec-tests/f32.wast.js:128:14
evaluate@[native code]
moduleEvaluation@[native code]
[native code]
promiseReactionJob@[native code]
Comment 1 JF Bastien 2017-03-15 17:15:21 PDT
Same thing for:

Comment 2 JF Bastien 2017-03-22 09:57:14 PDT
I looked into this and it seems related to NaN handling. The spec tests try to validate NaNs are handled properly and compare their values.

The bug could be:
 - In our code
 - In the spec tests (they're not currently up to date, see https://github.com/WebAssembly/spec/pull/414)
 - In the code that translates .wast -> .wast.js

This is unlikely to be important for the near future, so I'll ignore it for now. Maybe we can just comment out the "assert_[0-9]+" invocations for now, and revisit when we update the tests.
Comment 3 Radar WebKit Bug Importer 2017-05-03 09:54:34 PDT
Comment 4 Sergey Rubanov 2020-11-27 08:50:43 PST
I think that this issue is not valid anymore