⚠ This page is served via a proxy. Original site: https://github.com
This service does not collect credentials or authentication data.
Skip to content

Conversation

@Khushmagrawal
Copy link

The previous implementation was introduced in commit 34eeedc to probably satisfy LLVM where the select instruction requires an i1 condition. However this could be incorrect for ARM NEON as it could result to raw 0x01 being passed, triggering the bit-mixing problem as discussed in JuliaSIMD/LoopVectorization.jl#564.

The change would force a Logical comparison to generate AbstractMask which would handle the select correctly.

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.

1 participant