[llvm] [X86] Avoid shl->mul lowbit-isolate fold on Zen when unprofitable (PR #217776)

[Simon Pilgrim via llvm-commits]

================
@@ -0,0 +1,42 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=alderlake | FileCheck %s --check-prefix=BASELINE
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=znver5 | FileCheck %s --check-prefix=ZNVER5
----------------
RKSimon wrote:

why did you chose znver5? that is the one znver cpu that has weird BLSI and TZCNT false dependencies

https://github.com/llvm/llvm-project/pull/217776