[llvm] [LLVM][AMDGPU] AMDGPUInstCombineIntrinsic for *lane intrinsics (PR #99878)

[Jay Foad via llvm-commits]

jayfoad wrote:

> Compile-time: http://llvm-compile-time-tracker.com/compare.php?from=a7fb25dd1fcc2e5afcc65cccfa83b7b381b48906&to=18cda5b39d709a861cea69e690d674c4b49789b4&stat=instructions:u

I wonder if the slow down is due to running CycleAnalysis, and whether we could somehow avoid that on targets that don't care about divergence.

https://github.com/llvm/llvm-project/pull/99878