[llvm] [LoopVectorizer] Prune VFs based on plan register pressure (PR #132190)

[Alexander Kornienko via llvm-commits]

alexfh wrote:

This is causing serious (~40-50%) regressions in the LCALS benchmarks from the llvm-test-suite (https://github.com/llvm/llvm-test-suite/tree/main/MicroBenchmarks/LCALS) on x86. `BM_VOL3D_CALC_LAMBDA` and `BM_VOL3D_CALC_RAW`  are affected.

Is this the same problem as the one being addressed in https://github.com/llvm/llvm-project/pull/141736?

https://github.com/llvm/llvm-project/pull/132190