Stream: git-wasmtime

Topic: wasmtime / PR #8860 Initial `f16` and `f128` support

view this post on Zulip Wasmtime GitHub notifications bot (Jun 21 2024 at 21:50):

beetrees requested elliottt for a review on PR #8860.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 21 2024 at 21:50):

beetrees requested wasmtime-compiler-reviewers for a review on PR #8860.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 21 2024 at 21:50):

beetrees opened PR #8860 from beetrees:f16-f128-mvp to bytecodealliance:main:

Issue #8312

This PR adds initial basic support for f16 and f128 to Cranelift. This is enough to allow the interpreter to bitcast to and from f16s and f128s, as shown in the added filetest.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 21 2024 at 21:54):

beetrees updated PR #8860.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 21:12):

jameysharp submitted PR review:

This looks great!

I didn't check the Ieee16/Ieee128 implementations super carefully and it would be nice to find ways to deduplicate the code for those four Ieee* types a bit, but based on the tests I'm confident enough to merge these.

I have a couple quick questions but overall I think this covers everything necessary to get started implementing these types. Thank you!

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 21:12):

jameysharp submitted PR review:

This looks great!

I didn't check the Ieee16/Ieee128 implementations super carefully and it would be nice to find ways to deduplicate the code for those four Ieee* types a bit, but based on the tests I'm confident enough to merge these.

I have a couple quick questions but overall I think this covers everything necessary to get started implementing these types. Thank you!

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 21:12):

jameysharp created PR review comment:

Perhaps this test should verify that floats(16..64) doubles to floats(32..128).

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 21:12):

jameysharp created PR review comment:

I don't remember exactly how we defined the bit-packed type IDs. Why is renumbering these lane types okay? Do we not use 4/5 somewhere else?

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 21:12):

jameysharp created PR review comment:

Maybe this should test that half-width of floats(32..128) is floats(16..64)? I'm not sure the separate 32..32 case is necessary to test.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:27):

beetrees submitted PR review.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:27):

beetrees created PR review comment:

According to constants.rs, 0x70-0x7d are reserved for lane types (LANE_BASE is 0x70), with the vector type bit-packing using 4 bits to identify the lane type. This means that LANE_BASE + 0 to LANE_BASE + 13 inclusive are the possible valid lane types.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:28):

beetrees updated PR #8860.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:28):

beetrees created PR review comment:

Done as suggested

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:28):

beetrees submitted PR review.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:28):

beetrees submitted PR review.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:28):

beetrees created PR review comment:

Done

view this post on Zulip Wasmtime GitHub notifications bot (Jun 26 2024 at 23:33):

beetrees updated PR #8860.

view this post on Zulip Wasmtime GitHub notifications bot (Jun 27 2024 at 00:13):

jameysharp commented on PR #8860:

Awesome, thank you!

view this post on Zulip Wasmtime GitHub notifications bot (Jun 27 2024 at 00:28):

jameysharp merged PR #8860.

Last updated: Nov 22 2024 at 16:03 UTC