[llvm] [LoopVectorize] Don't discount instructions scalarized due to tail folding (PR #109289)

David Sherwood via llvm-commits llvm-commits at lists.llvm.org
Fri Oct 18 06:09:31 PDT 2024

Previous message: [llvm] [LoopVectorize] Don't discount instructions scalarized due to tail folding (PR #109289)
Next message: [compiler-rt] 9cc6d6e - [compiler-rt] Explicitly enable C extensions for profile (#110555)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

david-arm wrote:

Sorry, please ignore my comment about `icmp ult <8 x i64> %vec.ind, <i64 15, i64 15, i64 15, i64 15, i64 15, i64 15, i64 15, i64 15>` being folded away - that's bogus because the last lane will be false.

https://github.com/llvm/llvm-project/pull/109289

Previous message: [llvm] [LoopVectorize] Don't discount instructions scalarized due to tail folding (PR #109289)
Next message: [compiler-rt] 9cc6d6e - [compiler-rt] Explicitly enable C extensions for profile (#110555)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the llvm-commits mailing list