[NFC] [DirectX] Update DirectX codegen test CBufferAccess/gep-ce-two-uses.ll due to changes to ReplaceConstant

[Icohedron]

Fixes an LLVM DirectX codegen test after it broke due to #169141

The CBuffer loads and GEPs are no longer duplicated when there are two or more accesses within the same basic block.
This PR removes the duplicate check for CBuffer load and GEP from the original test function @f and adds a new test function @g which places duplicate CBuffer loads into separate basic blocks.

[llvmbot]

@llvm/pr-subscribers-backend-directx

Author: Deric C. (Icohedron)

Changes

Fixes an LLVM DirectX codegen test after it broke due to #169141

The CBuffer loads and GEPs are no longer duplicated when in there are two or more accesses within the same basic block.
This PR removes the duplicate CBuffer load and GEP from the original test function @<!-- -->f and adds a new test function @<!-- -->g which places duplicate CBuffer loads into separate basic blocks.

---

Full diff: https://github.com/llvm/llvm-project/pull/169848.diff

1 Files Affected:

• (modified) llvm/test/CodeGen/DirectX/CBufferAccess/gep-ce-two-uses.ll (+15)

diff --git a/llvm/test/CodeGen/DirectX/CBufferAccess/gep-ce-two-uses.ll b/llvm/test/CodeGen/DirectX/CBufferAccess/gep-ce-two-uses.ll
index 4eda6353f47ed..751f1d3ebcfa9 100644
--- a/llvm/test/CodeGen/DirectX/CBufferAccess/gep-ce-two-uses.ll
+++ b/llvm/test/CodeGen/DirectX/CBufferAccess/gep-ce-two-uses.ll
@@ -21,6 +21,21 @@ entry:
   %a1 = load float, ptr addrspace(2) getelementptr inbounds nuw (i8, ptr addrspace(2) @a1, i32 16), align 4
   store float %a1, ptr %dst, align 32
 
+  %a2 = load float, ptr addrspace(2) getelementptr inbounds nuw (i8, ptr addrspace(2) @a1, i32 16), align 4
+  store float %a2, ptr %dst, align 32
+
+  ret void
+}
+
+define void @g(ptr %dst) {
+entry:
+  ; CHECK: [[PTR:%.*]] = call ptr addrspace(2) @llvm.dx.resource.getpointer.{{.*}}(target("dx.CBuffer", %__cblayout_CB) {{%.*}}, i32 0)
+  ; CHECK: getelementptr inbounds nuw i8, ptr addrspace(2) [[PTR]], i32 16
+  %a1 = load float, ptr addrspace(2) getelementptr inbounds nuw (i8, ptr addrspace(2) @a1, i32 16), align 4
+  store float %a1, ptr %dst, align 32
+  br label %next
+
+next:
   ; CHECK: [[PTR:%.*]] = call ptr addrspace(2) @llvm.dx.resource.getpointer.{{.*}}(target("dx.CBuffer", %__cblayout_CB) {{%.*}}, i32 0)
   ; CHECK: getelementptr inbounds nuw i8, ptr addrspace(2) [[PTR]], i32 16
   %a2 = load float, ptr addrspace(2) getelementptr inbounds nuw (i8, ptr addrspace(2) @a1, i32 16), align 4

[bogner]

Should probably add checks for the two stores to show that they're reusing the same GEP

[bogner]

and I guess for completeness checking that the two stores use their own GEPs here might make sense.

[llvm-ci]

LLVM Buildbot has detected a new failure on builder ppc64le-mlir-rhel-clang running on ppc64le-mlir-rhel-test while building llvm at step 6 "test-build-check-mlir-build-only-check-mlir".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/129/builds/33995

Here is the relevant piece of the build log for the reference

Step 6 (test-build-check-mlir-build-only-check-mlir) failure: 1200 seconds without output running [b'ninja', b'check-mlir'], attempting to kill
...
PASS: MLIR-Unit :: IR/./MLIRIRTests/100/130 (3689 of 3700)
PASS: MLIR-Unit :: IR/./MLIRIRTests/101/130 (3690 of 3700)
PASS: MLIR-Unit :: IR/./MLIRIRTests/0/130 (3691 of 3700)
PASS: MLIR-Unit :: IR/./MLIRIRTests/38/130 (3692 of 3700)
PASS: MLIR-Unit :: Interfaces/./MLIRInterfacesTests/12/22 (3693 of 3700)
PASS: MLIR-Unit :: Pass/./MLIRPassTests/10/13 (3694 of 3700)
PASS: MLIR-Unit :: Interfaces/./MLIRInterfacesTests/13/22 (3695 of 3700)
PASS: MLIR-Unit :: IR/./MLIRIRTests/39/130 (3696 of 3700)
PASS: MLIR-Unit :: Interfaces/./MLIRInterfacesTests/11/22 (3697 of 3700)
PASS: MLIR :: mlir-reduce/dce-test.mlir (3698 of 3700)
command timed out: 1200 seconds without output running [b'ninja', b'check-mlir'], attempting to kill
process killed by signal 9
program finished with exit code -1
elapsedTime=4210.605909