wasmtime / issue #5036 cranelift: Support `bnot`, `band`,... · git-wasmtime

Stream: git-wasmtime

Topic: wasmtime / issue #5036 cranelift: Support `bnot`, `band`,...

Wasmtime GitHub notifications bot (Oct 18 2022 at 20:19):

I need to make test for float to vec. xor operation may generate incorrect code because current xor implementation does use vector_all_ones which will flip all unused bits.

Wasmtime GitHub notifications bot (Oct 18 2022 at 20:20):

ArtBlnd edited a comment on issue #5036:

I need to make test for float to vec. xor operation may generate incorrect code because current xor implementation does use vector_all_ones which will flip all unused bits.

I am finding a efficient way to generate single slot bit mask.

Wasmtime GitHub notifications bot (Oct 18 2022 at 20:20):

ArtBlnd edited a comment on issue #5036:

first of all, I need to make test for float to vec to check that unused bits are filtered out. xor operation may generate incorrect code because current xor implementation does use vector_all_ones which will flip all unused bits.

I am finding a efficient way to generate single slot bit mask.

Wasmtime GitHub notifications bot (Oct 18 2022 at 20:21):

ArtBlnd edited a comment on issue #5036:

first of all, I need to make test for float to vec to check that unused bits are filtered out. xor operation may generate incorrect code because current xor implementation does use vector_all_ones which will flip all unused bits.

so I am finding a efficient way to generate single slot bit mask. (not using load ops. that will need extra stack allocations.)

Wasmtime GitHub notifications bot (Oct 18 2022 at 20:22):

ArtBlnd edited a comment on issue #5036:

first of all, I need to make test for float to vec to check that unused bits are filtered out. bnot operation may generate incorrect code because current bnot implementation does use vector_all_ones which will flip all unused bits.

so I am finding a efficient way to generate single slot bit mask. (not using load ops. that will need extra stack allocations.)

Wasmtime GitHub notifications bot (Oct 18 2022 at 20:22):

ArtBlnd edited a comment on issue #5036:

first of all, I need to make test for float to vec to check that unused bits are filtered out. bnot operation may generate incorrect code because current bnot implementation does use vector_all_ones which will flip all unused bits.

so I am finding a efficient way to generate single slot bit mask. (not using load ops. that will need extra stack allocations.)

https://github.com/bytecodealliance/wasmtime/pull/5036/files#diff-4fb30cf23cc13cba2f11079dc6d0305f972170585aab27b27d143c2f3252e4faR1270-R1274

Wasmtime GitHub notifications bot (Oct 18 2022 at 20:29):

ArtBlnd edited a comment on issue #5036:

first of all, I need to make test for float to vec to check that unused bits are filtered out. bnot operation may generate incorrect code because current bnot implementation does use vector_all_ones which will flip all unused bits. means it has side-effect.

so I am finding a efficient way to generate single slot bit mask. (not using load ops. that will need extra stack allocations.)

https://github.com/bytecodealliance/wasmtime/pull/5036/files#diff-4fb30cf23cc13cba2f11079dc6d0305f972170585aab27b27d143c2f3252e4faR1270-R1274

Wasmtime GitHub notifications bot (Oct 18 2022 at 21:02):

abrown commented on issue #5036:

You can tag me once you're ready for review; I glanced over this and it made sense.

Wasmtime GitHub notifications bot (Oct 22 2022 at 18:50):

ArtBlnd commented on issue #5036:

I think there is no cheap way to generate masks...

Wasmtime GitHub notifications bot (Oct 22 2022 at 18:51):

ArtBlnd edited a comment on issue #5036:

I think there is no cheap way to generate masks...

maybe we need to truncate all unused vector slots when using scalar_to_vector

Wasmtime GitHub notifications bot (Oct 22 2022 at 18:54):

ArtBlnd edited a comment on issue #5036:

I think there is no cheap way to generate masks...

maybe we need to truncate all unused vector slots when using scalar_to_vector
I'll make seperate PR for this task.

Wasmtime GitHub notifications bot (Oct 22 2022 at 19:15):

ArtBlnd edited a comment on issue #5036:

I think there is no cheap way to generate masks...

~~maybe we need to truncate all unused vector slots when using scalar_to_vector~~
~~I'll make seperate PR for this task.~~

nevermind. scalar_to_vector is using movss, movsd.

Wasmtime GitHub notifications bot (Oct 22 2022 at 19:22):

ArtBlnd commented on issue #5036:

@abrown I think its ready to review.

Wasmtime GitHub notifications bot (Nov 03 2022 at 02:56):

ArtBlnd commented on issue #5036:

sorry for the very late response for the review.

Wasmtime GitHub notifications bot (Nov 10 2022 at 19:22):

abrown commented on issue #5036:

@ArtBlnd, I believe the CI failure here is due to some rustc changes between when this PR was open and now. Can you rebase this PR and see if that resolves things?

Wasmtime GitHub notifications bot (Nov 18 2022 at 04:05):

ArtBlnd commented on issue #5036:

@ArtBlnd, I believe the CI failure here is due to some rustc changes between when this PR was open and now. Can you rebase this PR and see if that resolves things?

Done.

Last updated: Nov 22 2024 at 17:03 UTC