afonso360 opened PR #4403 from aarch64-of-fix
to main
:
@yuyang-ok reported via zulip that i128 overflow tests were wrong.
It turns out that both the tests and the
aarch64
codegen are wrong.I've altered the implementation to match what LLVM produces, and added the interpreter to that test suite.
bjorn3 submitted PR review.
bjorn3 created PR review comment:
Does x86_64 support this too?
afonso360 submitted PR review.
afonso360 created PR review comment:
Hmm, It does but produces wrong results
Running `C:\Users\Afonso\CLionProjects\wasmtime\target\debug\clif-util.exe test .\filetests\filetests\runtests\i128-icmp-overflow.clif` skipped .\filetests\filetests\runtests\i128-icmp-overflow.clif: host can't run Aarch64(Aarch64) programs FAIL .\filetests\filetests\runtests\i128-icmp-overflow.clif: run Caused by: Failed test: run: %icmp_of_i128(-170141183460469231731687303715884105728, 1) == true, actual: false 1 tests Error: 1 failure error: process didn't exit successfully: `C:\Users\Afonso\CLionProjects\wasmtime\target\debug\clif-util.exe test .\filetests\filetests\runtests\i128-icmp-overflow.clif` (exit code: 1)
afonso360 created PR review comment:
And on the current
main
it fails in a different test:Failed test: run: %icmp_of_i128(1, 170141183460469231731687303715884105727) == true, actual: false
afonso360 submitted PR review.
afonso360 updated PR #4403 from aarch64-of-fix
to main
.
afonso360 updated PR #4403 from aarch64-of-fix
to main
.
cfallin submitted PR review.
cfallin merged PR #4403.
Last updated: Dec 23 2024 at 13:07 UTC