[llvm] [InstCombine] Avoid foldCmpLoadFromIndexedGlobal for multi-use load (PR #151524)

[Nikita Popov via llvm-commits]

nikic wrote:

The llvm-opt-benchmark results look fairly mixed to me. One very common case is where the load is multi-use, but ends up being sunk behind the check, which is likely beneficial.

For OOO cores the transform may be valuable even in multi-use cases, because the condition can be computed without waiting for the load to complete.

Overall I'm not sure whether we should change something here or not. Or possibly we should use less aggressive checks.

Thoughts? 

https://github.com/llvm/llvm-project/pull/151524