[llvm] [LSV] Enhance LoadStoreVectorizer to Handle Disjoint Flag in OR Instructions and Restore Vectorization Opportunities (PR #96495)

Hao Li via llvm-commits llvm-commits at lists.llvm.org
Fri Jun 28 01:16:41 PDT 2024


================

----------------
LiHao217 wrote:

You are right, the or disjoint is missing when -mtriple=nvptx64-nvidia-cuda is enabled, I will check this issue。
I found the reason is that the https://github.com/llvm/llvm-project/commit/cd564c6b4638e654edb68acf7f7d985c08edf468 this commit mention LSV optimization is not mature, So the default is to turn this optimization off, so I think for now I can make some small changes on LSV and then continue to let the user choose whether to enable LSV optimization or not.

https://github.com/llvm/llvm-project/pull/96495


More information about the llvm-commits mailing list