Skip to content

[SPIR-V] Translate complex nested vector expressions instead of lowering them #5183

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Dec 23, 2021

Conversation

vmaksimo
Copy link
Contributor

@vmaksimo vmaksimo commented Dec 20, 2021

Allow translator to go deeper into complex nested instructions.
Enabled a possibility to translate expressions in extractelement and binary
operator instructions (e.g., fadd, fmul).

Also this change removes lowering for constant expression vector as the
alternative approach was introduced - to translate any complicated nested
instruction instead of lowering.

Original commit:
KhronosGroup/SPIRV-LLVM-Translator@c3c3c68

@bader
Copy link
Contributor

bader commented Dec 20, 2021

@vmaksimo, FYI. @alexbatashev fixed tests failing on AMD GPU here - intel/llvm-test-suite#642.

AlexeySotkin
AlexeySotkin previously approved these changes Dec 21, 2021
Allow translator to go deeper into complex nested instructions.
Enabled a possibility to translate expressions in `extractelement` and binary
operator instructions (e.g., `fadd`, `fmul`).

Also this change removes lowering for constant expression vector as the
alternative approach was introduced - to translate any complicated nested
instruction instead of lowering.

Original commit:
KhronosGroup/SPIRV-LLVM-Translator@c3c3c68b
@AlexeySotkin
Copy link
Contributor

The patch is ready for review. This is a cherry-pick of KhronosGroup/SPIRV-LLVM-Translator#1335

AFAIK the failures in the CI pre-commit test are not related to this patch.

@bader
Copy link
Contributor

bader commented Dec 23, 2021

The patch is ready for review.

Is it ready for review or merge? @AlexeySotkin, do you want @AlexeySachkov to approve it as well?

@AlexeySotkin
Copy link
Contributor

The patch is ready for review.

Is it ready for review or merge? @AlexeySotkin, do you want @AlexeySachkov to approve it as well?

I wrote "ready for review" because before that it was a draft PR. But yes, this patch looks good to me and I'd like to merge it.

I think @AlexeySachkov was added automatically as a codeowner(probably we should fix it)

@bader bader merged commit dc9bd3f into intel:sycl Dec 23, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants