Corrected setting IRBuilder debug location in HLSignatureLower::GenerateDxilComputeAndNodeCommonInputs() (#8022)

As a result of
#7799, the HLSL
entry function can now contain `dbg.value` calls with !dbg location
metadata with scopes corresponding to functions which have been inlined.
The `HLSignatureLower::GenerateDxilComputeAndNodeCommonInputs()` creates
`@dx.op.threadId` intrinsic calls, using the default dbg loc when
creating an `IRBuilder`. When the first instruction in the entry block
is one of the `dbg,value` calls, its dbg location is copied to the
`@dx.op.threadId` calls. That makes the DXIL unreadable by the modern
LLVM IR reader since it fails the IR module verification because of !dbg
pointing to a different subprogram scope.

This change sets the dbg location for IRBuilder to be the !dbg node of
the first non-PHI and non-debug instruction in the block.

---------

Co-authored-by: Konstantin <konstantin.pyzhov@amd.com>

Author: kpyzhov

lib/HLSL/HLSignatureLower.cpp

@@ -1253,6 +1253,8 @@ void HLSignatureLower::GenerateDxilComputeAndNodeCommonInputs() {
   DXASSERT(funcAnnotation, "must find annotation for entry function");
   auto &funcProps = HLM.GetDxilFunctionProps(Entry);
   IRBuilder<> Builder(Entry->getEntryBlock().getFirstInsertionPt());
+  Builder.SetCurrentDebugLocation(
+      Entry->getEntryBlock().getFirstNonPHIOrDbg()->getDebugLoc());
 
   for (Argument &arg : Entry->args()) {
     DxilParameterAnnotation &paramAnnotation =

tools/clang/test/HLSLFileCheck/dxil/debug/dxop_threadid_dbgloc.hlsl

@@ -0,0 +1,24 @@
+// RUN: %dxc -T cs_6_5 -E CSMain -O0 -Zi -enable-16bit-types %s | FileCheck %s
+
+// CHECK-LABEL:  @CSMain()
+// CHECK: %{{[0-9]+}} =  call i32 @dx.op.threadId.i32(i32 {{[0-9]+}}, i32 {{[0-9]+}}), !dbg [[DBG_MD:![0-9]+]]
+// CHECK-DAG: [[SUBPGM:![0-9]+]] = !DISubprogram(name: "CSMain"
+// CHECK-DAG: [[DBG_MD]] = !DILocation(line: 21, column: 15, scope: [[SUBPGM]])
+
+RWBuffer<float> u0 : register(u0);
+RWBuffer<float> u1 : register(u1);
+
+static float my_var;
+static float my_var2;
+
+void foo() {
+  my_var2 = my_var * 2;
+}
+
+[RootSignature("DescriptorTable(UAV(u0,numDescriptors=2))")]
+[numthreads(64,1,1)]
+void CSMain(uint3 dtid : SV_DispatchThreadID) {
+  my_var = u0[dtid.x];
+  foo();
+  u1[dtid.x] = my_var2;
+}