[llvm] [NVPTX] Customize getScalarizationOverhead (PR #128077)

via llvm-commits llvm-commits at lists.llvm.org
Fri Feb 21 12:40:14 PST 2025


peterbell10 wrote:

@Artem-B I don't follow your point. Yes, `ptxas` can optimize the insert-element sequence itself, but that doesn't have any effect on the slp-vectorizer's cost heuristics.

https://github.com/llvm/llvm-project/pull/128077


More information about the llvm-commits mailing list