[AMDGPU] Add tests for vector rebroadcast.

[PeddleSpam]

No description provided.

[llvmbot]

@llvm/pr-subscribers-backend-amdgpu

Author: Leon Clark (PeddleSpam)

Changes

---

Full diff: https://github.com/llvm/llvm-project/pull/91322.diff

1 Files Affected:

• (added) llvm/test/CodeGen/AMDGPU/vector_rebroadcast.ll (+39)

diff --git a/llvm/test/CodeGen/AMDGPU/vector_rebroadcast.ll b/llvm/test/CodeGen/AMDGPU/vector_rebroadcast.ll
new file mode 100644
index 0000000000000..50c5dadfcbb15
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/vector_rebroadcast.ll
@@ -0,0 +1,39 @@
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX9 %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1010 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX10 %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX11 %s
+
+define <4 x float> @rebroadcast_v4f32(ptr addrspace(1) %arg0) {
+; GFX9-LABEL: rebroadcast_v4f32:
+; GFX9:       ; %bb.0: ; %entry
+; GFX9-NEXT:  s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-NEXT:  global_load_dwordx4 v[0:3], v[0:1], off
+; GFX9-NEXT:  s_waitcnt vmcnt(0)
+; GFX9-NEXT:  v_mov_b32_e32 v0, v1
+; GFX9-NEXT:  v_mov_b32_e32 v2, v1
+; GFX9-NEXT:  v_mov_b32_e32 v3, v1
+; GFX9-NEXT:  s_setpc_b64 s[30:31]
+;
+; GFX10-LABEL: rebroadcast_v4f32:
+; GFX10:       ; %bb.0: ; %entry
+; GFX10-NEXT:  s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX10-NEXT:  global_load_dwordx4 v[0:3], v[0:1], off
+; GFX10-NEXT:  s_waitcnt vmcnt(0)
+; GFX10-NEXT:  v_mov_b32_e32 v0, v1
+; GFX10-NEXT:  v_mov_b32_e32 v2, v1
+; GFX10-NEXT:  v_mov_b32_e32 v3, v1
+; GFX10-NEXT:  s_setpc_b64 s[30:31]
+;
+; GFX11-LABEL: rebroadcast_v4f32:
+; GFX11:       ; %bb.0: ; %entry
+; GFX11-NEXT:  s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX11-NEXT:  global_load_b128 v[0:3], v[0:1], off
+; GFX11-NEXT:  s_waitcnt vmcnt(0)
+; GFX11-NEXT:  v_mov_b32_e32 v0, v1
+; GFX11-NEXT:  v_mov_b32_e32 v2, v1
+; GFX11-NEXT:  v_mov_b32_e32 v3, v1
+; GFX11-NEXT:  s_setpc_b64 s[30:31]
+entry:
+  %val0 = load <4 x float>, ptr addrspace(1) %arg0
+  %val1 = shufflevector <4 x float> %val0, <4 x float> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
+  ret <4 x float> %val1
+}

[arsenm]

Can this merge in with another test? Should it test more vector sizes? Probably should link to follow up patch context

[PeddleSpam]

Can this merge in with another test? Should it test more vector sizes? Probably should link to follow up patch context

I've added tests for more vector types/sizes. It's a lot to merge with another file but I can if you'd prefer.