Wasmtime GitHub notifications bot (Apr 22 2020 at 17:14): Wasmtime GitHub notifications bot (Apr 22 2020 at 17:14):bnjbvr opened PR #1578 from aarch64-float-to-int-conversions to master:
These are inherited from wasm semantics:
- if the input is a NaN, early trap
- if the input is out of bounds, trap
- truncating has towards-0 semantics, e.g. all the values in ]-1, 0] are possible inputs for unsigned truncate.
This makes Spidermonkey pass one more test entirely.
@cfallin PTAL!
Wasmtime GitHub notifications bot (Apr 22 2020 at 17:14):bnjbvr requested cfallin for a review on PR #1578.
Wasmtime GitHub notifications bot (Apr 22 2020 at 19:09):cfallin submitted PR Review.
Wasmtime GitHub notifications bot (Apr 22 2020 at 19:09):cfallin created PR Review Comment:
Perhaps I'm missing a subtle reason here -- but shouldn't the unsigned-case lower bound be
0.0? Otherwise the>=check will allow e.g.-0.9.Ah -- does this have to do with the truncation rounding mode? A clarification comment would help if so!
Wasmtime GitHub notifications bot (Apr 22 2020 at 19:09):cfallin submitted PR Review.
Wasmtime GitHub notifications bot (Apr 23 2020 at 09:15):bnjbvr submitted PR Review.
Wasmtime GitHub notifications bot (Apr 23 2020 at 09:15):bnjbvr created PR Review Comment:
Yes, the "truncate towards zero" was meant to explain it, it might be a bit terse, so I'll reword it.
Wasmtime GitHub notifications bot (Apr 23 2020 at 09:44):bnjbvr updated PR #1578 from aarch64-float-to-int-conversions to master:
These are inherited from wasm semantics:
- if the input is a NaN, early trap
- if the input is out of bounds, trap
- truncating has towards-0 semantics, e.g. all the values in ]-1, 0] are possible inputs for unsigned truncate.
This makes Spidermonkey pass one more test entirely.
@cfallin PTAL!
Wasmtime GitHub notifications bot (Apr 23 2020 at 11:28):bnjbvr updated PR #1578 from aarch64-float-to-int-conversions to master:
These are inherited from wasm semantics:
- if the input is a NaN, early trap
- if the input is out of bounds, trap
- truncating has towards-0 semantics, e.g. all the values in ]-1, 0] are possible inputs for unsigned truncate.
This makes Spidermonkey pass one more test entirely.
@cfallin PTAL!
Wasmtime GitHub notifications bot (Apr 23 2020 at 12:03):bnjbvr updated PR #1578 from aarch64-float-to-int-conversions to master:
These are inherited from wasm semantics:
- if the input is a NaN, early trap
- if the input is out of bounds, trap
- truncating has towards-0 semantics, e.g. all the values in ]-1, 0] are possible inputs for unsigned truncate.
This makes Spidermonkey pass one more test entirely.
@cfallin PTAL!
Wasmtime GitHub notifications bot (Apr 24 2020 at 08:53):bnjbvr updated PR #1578 from aarch64-float-to-int-conversions to master:
These are inherited from wasm semantics:
- if the input is a NaN, early trap
- if the input is out of bounds, trap
- truncating has towards-0 semantics, e.g. all the values in ]-1, 0] are possible inputs for unsigned truncate.
This makes Spidermonkey pass one more test entirely.
@cfallin PTAL!
Wasmtime GitHub notifications bot (Apr 24 2020 at 09:09):bnjbvr updated PR #1578 from aarch64-float-to-int-conversions to master:
These are inherited from wasm semantics:
- if the input is a NaN, early trap
- if the input is out of bounds, trap
- truncating has towards-0 semantics, e.g. all the values in ]-1, 0] are possible inputs for unsigned truncate.
This makes Spidermonkey pass one more test entirely.
@cfallin PTAL!
Wasmtime GitHub notifications bot (Apr 24 2020 at 09:38):bjorn3 submitted PR Review.
Wasmtime GitHub notifications bot (Apr 24 2020 at 09:38):bjorn3 created PR Review Comment:
self.block_state.stack = stack;
Wasmtime GitHub notifications bot (Apr 24 2020 at 09:38):bjorn3 submitted PR Review.
Wasmtime GitHub notifications bot (Apr 24 2020 at 09:38):bjorn3 created PR Review Comment:
let session.op_offset_map = op_offset_map;
Wasmtime GitHub notifications bot (Apr 24 2020 at 09:51):bnjbvr merged PR #1578.
Last updated: Nov 22 2024 at 16:03 UTC