[llvm] [AMDGPU] Capping max number of registers for indirect calls (PR #199765)

[Janek van Oirschot via llvm-commits]

================
@@ -0,0 +1,251 @@
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 < %s | FileCheck %s --check-prefixes=CHECK
+
----------------
JanekvO wrote:

Does this file showcase what normally would show an erroneous value?
Also, is there an agpr equivalent?

https://github.com/llvm/llvm-project/pull/199765