[llvm] [AMDGPU] Use LSH for lowering ctlz_zero_undef.i8/i16 (PR #88512)

[Matt Arsenault via llvm-commits]

================
@@ -1270,13 +1270,22 @@ AMDGPULegalizerInfo::AMDGPULegalizerInfo(const GCNSubtarget &ST_,
     .custom();
 
   // The 64-bit versions produce 32-bit results, but only on the SALU.
-  getActionDefinitionsBuilder({G_CTLZ_ZERO_UNDEF, G_CTTZ_ZERO_UNDEF})
-    .legalFor({{S32, S32}, {S32, S64}})
-    .clampScalar(0, S32, S32)
-    .clampScalar(1, S32, S64)
-    .scalarize(0)
-    .widenScalarToNextPow2(0, 32)
-    .widenScalarToNextPow2(1, 32);
+  getActionDefinitionsBuilder(G_CTLZ_ZERO_UNDEF)
+      .legalFor({{S32, S32}, {S32, S64}})
+      .customFor({{S32, S8}, {S32, S16}})
----------------
arsenm wrote:

This shouldn't really need to specially handle illegal typed sources. I would expect to be able to recover the promotion that would happen anyway with KnownBits. If you're going to special case the illegal types, might as well handle arbitrary bit width sources here 

https://github.com/llvm/llvm-project/pull/88512