[clang] [llvm] [AMDGPU] Add support for `v_prng_b32` on gfx1250 (PR #149450)

[Shilei Tian via cfe-commits]

================
@@ -4007,7 +4007,8 @@ SDValue 
AMDGPUTargetLowering::performIntrinsicWOChainCombine(
   case Intrinsic::amdgcn_rcp_legacy:
   case Intrinsic::amdgcn_rsq_legacy:
   case Intrinsic::amdgcn_rsq_clamp:
-  case Intrinsic::amdgcn_tanh: {
+  case Intrinsic::amdgcn_tanh:
+  case Intrinsic::amdgcn_prng_b32: {
----------------
shiltian wrote:

And based on the LFSR algorithm, it looks safe to me to fold this.

https://github.com/llvm/llvm-project/pull/149450
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits