This repository has been archived by the owner on Dec 22, 2021. It is now read-only.
any_true instructions should be untyped #416
Comments
|
I more or less agree - in Firefox the any_true implementations boil away and uniformly become v128.any_bit_set. There's no clear reason to have multiple ops. |
|
No objections to v128.any_true operation, V8 generates the same code for all the variants as well. |
This was referenced Jan 8, 2021
|
Was implemented in #423 |
ngzhian
added a commit
to ngzhian/simd
that referenced
this issue
Jan 22, 2021
This was accepted in WebAssembly#416. The test code is manually modified (this is not generated code), to always call v128.any_true. The exported function names are unchanged, and no tests are removed, since it is useful to test v128 of different constants.
ngzhian
added a commit
that referenced
this issue
Jan 26, 2021
This was accepted in #416. The test code is manually modified (this is not generated code), to always call v128.any_true. The exported function names are unchanged, and no tests are removed, since it is useful to test v128 of different constants.
ngzhian
added a commit
to ngzhian/simd
that referenced
this issue
Jan 26, 2021
The renaming was accepted in WebAssembly#416. Interpreter was changed in WebAssembly#426. This makes similar changes to the formal text.
ngzhian
added a commit
to ngzhian/simd
that referenced
this issue
Jan 26, 2021
The renaming was accepted in WebAssembly#416. Interpreter was changed in WebAssembly#426. This makes similar changes to the formal text.
tlively
added a commit
to tlively/emscripten
that referenced
this issue
Feb 2, 2021
These instructions were removed from the spec proposal and from V8 because they were redundant with i8x16.any_true, which has identical semantics. Since they were removed from V8, these tests started failing. This PR fixes the breakage by removing the obsolete tests. See WebAssembly/simd#416.
tlively
added a commit
to emscripten-core/emscripten
that referenced
this issue
Feb 2, 2021
These instructions were removed from the spec proposal and from V8 because they were redundant with i8x16.any_true, which has identical semantics. Since they were removed from V8, these tests started failing. This PR fixes the breakage by removing the obsolete tests. See WebAssembly/simd#416.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
any_trueSIMD instruction returns 1 if any lane is non-zero and returns 0 otherwise. Notably, the result does not depend on whether we look at 8-bit, 16-bit, 32-bit, or 64-bit lanes: if the SIMD vector has any bit set, the result is 1; otherwise, the result is 0. Therefore, it would make sense to replacei8x16.any_true,i16x8.any_true,i32x4.any_true, andi64x2.any_truewith a singlev128.any_trueinstruction.The text was updated successfully, but these errors were encountered: