[llvm] [AMDGPU][GISel] Fold 'min(min(x,y),z)' and 'max(max(x,y),z)' into min3 and max3 (PR #124263)

[Alan Li via llvm-commits]

================
@@ -175,10 +185,9 @@ def AMDGPUPostLegalizerCombiner: GICombiner<
   let CombineAllMethodName = "tryCombineAllImpl";
 }
 
-def AMDGPURegBankCombiner : GICombiner<
-  "AMDGPURegBankCombinerImpl",
-  [unmerge_merge, unmerge_cst, unmerge_undef,
-   zext_trunc_fold, int_minmax_to_med3, ptr_add_immed_chain,
-   fp_minmax_to_clamp, fp_minmax_to_med3, fmed3_intrinsic_to_clamp,
-   redundant_and]> {
-}
+def AMDGPURegBankCombiner
+    : GICombiner<"AMDGPURegBankCombinerImpl",
+                 [unmerge_merge, unmerge_cst, unmerge_undef, zext_trunc_fold,
+                  int_minmax_to_med3, ptr_add_immed_chain, fp_minmax_to_clamp,
+                  fp_minmax_to_med3, fmed3_intrinsic_to_clamp,
+                  minmax_to_minmax3, redundant_and]> {}
----------------
lialan wrote:

This lint is irrelevant, so probably we should exclude it from the PR.

https://github.com/llvm/llvm-project/pull/124263