[llvm] [LiveDebugVariables] Stop holding SlotIndexes for erased instructions (PR #224467)

[Petar Jovanovic via llvm-commits]

================
@@ -0,0 +1,57 @@
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx90a -O3 -stop-after=virtregrewriter,2 < %s | FileCheck %s
+
+; LiveDebugVariables is built before the SGPR allocator and still holds
+; SlotIndexes when the VGPR allocator spills %v1 several passes later. The two
+; intervals recorded for %v1 around the spill resolve onto the same position, so
+; only one DBG_VALUE belongs there. Without canonicalizeIndexes() each interval
+; emits its own and the last CHECK-NEXT lands on the duplicate.
+
+; CHECK-LABEL: name: partial_copy
+; CHECK:      DBG_VALUE $vgpr0_vgpr1, $noreg, ![[V1:[0-9]+]], !DIExpression()
+; CHECK-NEXT: SI_SPILL_AV64_SAVE
+; CHECK-NEXT: DBG_VALUE %stack.0, 0, ![[V1]], !DIExpression()
+; CHECK-NEXT: GLOBAL_STORE_DWORDX4
+
+define amdgpu_kernel void @partial_copy(<4 x i32> %arg) #0 !dbg !5 {
+  call void asm sideeffect "; use $0", "a"(i32 poison), !dbg !13
+  %v0 = call <4 x i32> asm sideeffect "; def $0", "=v"(), !dbg !14
+    #dbg_value(<4 x i32> %v0, !9, !DIExpression(), !14)
+  %v1 = call <2 x i32> asm sideeffect "; def $0", "=v"(), !dbg !15
+    #dbg_value(<2 x i32> %v1, !11, !DIExpression(), !15)
+  %mai = tail call <4 x i32> @llvm.amdgcn.mfma.i32.4x4x4i8(i32 1, i32 2, <4 x i32> %arg, i32 0, i32 0, i32 0), !dbg !16
+    #dbg_value(<4 x i32> %mai, !12, !DIExpression(), !16)
+  store volatile <4 x i32> %v0, ptr addrspace(1) poison, align 16, !dbg !17
+  store volatile <2 x i32> %v1, ptr addrspace(1) poison, align 8, !dbg !18
+  store volatile <4 x i32> %mai, ptr addrspace(1) poison, align 16, !dbg !19
+  ret void, !dbg !20
+}
+
+declare <4 x i32> @llvm.amdgcn.mfma.i32.4x4x4i8(i32, i32, <4 x i32>, i32, i32, i32)
+
+; The VGPR budget is what forces %v1 to be spilled.
+attributes #0 = { nounwind "amdgpu-num-vgpr"="5" "amdgpu-no-dispatch-id" "amdgpu-no-dispatch-ptr" "amdgpu-no-implicitarg-ptr" "amdgpu-no-lds-kernel-id" "amdgpu-no-queue-ptr" "amdgpu-no-workgroup-id-x" "amdgpu-no-cluster-id-x" "amdgpu-no-workgroup-id-y" "amdgpu-no-cluster-id-y" "amdgpu-no-workgroup-id-z" "amdgpu-no-cluster-id-z" "amdgpu-no-workitem-id-x" "amdgpu-no-workitem-id-y" "amdgpu-no-workitem-id-z" }
----------------
petar-jovanovic wrote:

Done, make sense. Reduced it all to "amdgpu-num-vgpr"="5".

https://github.com/llvm/llvm-project/pull/224467