[llvm] AMDGPU: Treat uint32_max as the default value for amdgpu-max-num-workgroups (PR #113751)

[Matt Arsenault via llvm-commits]

================
@@ -54,16 +81,25 @@ attributes #4 = {"amdgpu-max-num-workgroups"="1024,1024,1024"}
 
 ; CHECK: - .args:
 ; CHECK:        .max_flat_workgroup_size: 1024
+; CHECK-NEXT:   .max_num_workgroups_x: 0
----------------
arsenm wrote:

That's on the emitted directive, it's not directly documented on the IR attribute.

I think I've seen that it's lower in the y and z dimensions but I don't remember where and can't find documentation for it 

https://github.com/llvm/llvm-project/pull/113751