Skip to content

[DirectX] Add GlobalDCE pass after finalize linkage pass in DirectX backend #151071

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 4 commits into
base: main
Choose a base branch
from

Conversation

kmpeng
Copy link
Contributor

@kmpeng kmpeng commented Jul 29, 2025

Fixes #139023.

This PR essentially removes unused global variables:

  • Restores the GlobalDCE Legacy pass and adds it to the DirectX backend after the finalize linkage pass
  • Converts external global variables with no usage to internal linkage in the finalize linkage pass
    • (so they can be removed by GlobalDCE)
  • Makes the dxil-finalize-linkage pass usable using the new pass manager flag syntax
  • Adds tests to finalize_linkage.ll that make sure unused global variables are removed
  • Adds a use for variable @CBV in opaque-value_as_metadata.ll so it isn't removed
  • Changes the scalar-data.ll run command to avoid removing its global variables

@llvmbot
Copy link
Member

llvmbot commented Jul 29, 2025

@llvm/pr-subscribers-backend-directx

Author: Kaitlin Peng (kmpeng)

Changes

Fixes #139023.

This PR essentially removes unused global variables:

  • Restores the GlobalDCE Legacy pass and adds it to the DirectX backend after the finalize linkage pass
  • Converts external global variables with no usage to internal linkage in the finalize linkage pass
    • (so they can be removed by GlobalDCE)
  • Makes the dxil-finalize-linkage pass usable using the new pass manager flag syntax
  • Adds tests to finalize_linkage.ll that make sure unused global variables are removed
  • Adds a use for variable @<!-- -->CBV in opaque-value_as_metadata.ll so it isn't removed
  • Changes the scalar-data.ll run command to avoid removing its global variables

Full diff: https://github.com/llvm/llvm-project/pull/151071.diff

12 Files Affected:

  • (modified) llvm/include/llvm/InitializePasses.h (+1)
  • (modified) llvm/include/llvm/LinkAllPasses.h (+2)
  • (modified) llvm/include/llvm/Transforms/IPO/GlobalDCE.h (+2)
  • (modified) llvm/lib/Target/DirectX/CMakeLists.txt (+1)
  • (modified) llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp (+4-4)
  • (modified) llvm/lib/Target/DirectX/DirectXPassRegistry.def (+1)
  • (modified) llvm/lib/Target/DirectX/DirectXTargetMachine.cpp (+4)
  • (modified) llvm/lib/Transforms/IPO/GlobalDCE.cpp (+32)
  • (modified) llvm/test/CodeGen/DirectX/finalize_linkage.ll (+20-2)
  • (modified) llvm/test/CodeGen/DirectX/llc-pipeline.ll (+1)
  • (modified) llvm/test/CodeGen/DirectX/scalar-data.ll (+1-1)
  • (modified) llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll (+1)
diff --git a/llvm/include/llvm/InitializePasses.h b/llvm/include/llvm/InitializePasses.h
index 31801daa126ad..0e48d80936270 100644
--- a/llvm/include/llvm/InitializePasses.h
+++ b/llvm/include/llvm/InitializePasses.h
@@ -127,6 +127,7 @@ LLVM_ABI void initializeGCEmptyBasicBlocksPass(PassRegistry &);
 LLVM_ABI void initializeGCMachineCodeAnalysisPass(PassRegistry &);
 LLVM_ABI void initializeGCModuleInfoPass(PassRegistry &);
 LLVM_ABI void initializeGVNLegacyPassPass(PassRegistry &);
+LLVM_ABI void initializeGlobalDCELegacyPassPass(PassRegistry &);
 LLVM_ABI void initializeGlobalMergeFuncPassWrapperPass(PassRegistry &);
 LLVM_ABI void initializeGlobalMergePass(PassRegistry &);
 LLVM_ABI void initializeGlobalsAAWrapperPassPass(PassRegistry &);
diff --git a/llvm/include/llvm/LinkAllPasses.h b/llvm/include/llvm/LinkAllPasses.h
index f82a43967e67a..7b33ec7f09124 100644
--- a/llvm/include/llvm/LinkAllPasses.h
+++ b/llvm/include/llvm/LinkAllPasses.h
@@ -38,6 +38,7 @@
 #include "llvm/Support/Valgrind.h"
 #include "llvm/Transforms/IPO.h"
 #include "llvm/Transforms/IPO/AlwaysInliner.h"
+#include "llvm/Transforms/IPO/GlobalDCE.h"
 #include "llvm/Transforms/InstCombine/InstCombine.h"
 #include "llvm/Transforms/ObjCARC.h"
 #include "llvm/Transforms/Scalar.h"
@@ -83,6 +84,7 @@ struct ForcePassLinking {
     (void)llvm::createDomOnlyViewerWrapperPassPass();
     (void)llvm::createDomViewerWrapperPassPass();
     (void)llvm::createAlwaysInlinerLegacyPass();
+    (void)llvm::createGlobalDCEPass();
     (void)llvm::createGlobalMergeFuncPass();
     (void)llvm::createGlobalsAAWrapperPass();
     (void)llvm::createInstSimplifyLegacyPass();
diff --git a/llvm/include/llvm/Transforms/IPO/GlobalDCE.h b/llvm/include/llvm/Transforms/IPO/GlobalDCE.h
index f8f775b8ea96e..55dff64889270 100644
--- a/llvm/include/llvm/Transforms/IPO/GlobalDCE.h
+++ b/llvm/include/llvm/Transforms/IPO/GlobalDCE.h
@@ -32,6 +32,7 @@ class GlobalVariable;
 class Metadata;
 class Module;
 class Value;
+class ModulePass;
 
 /// Pass to remove unused function declarations.
 class GlobalDCEPass : public PassInfoMixin<GlobalDCEPass> {
@@ -80,6 +81,7 @@ class GlobalDCEPass : public PassInfoMixin<GlobalDCEPass> {
   void ComputeDependencies(Value *V, SmallPtrSetImpl<GlobalValue *> &U);
 };
 
+ModulePass *createGlobalDCEPass();
 }
 
 #endif // LLVM_TRANSFORMS_IPO_GLOBALDCE_H
diff --git a/llvm/lib/Target/DirectX/CMakeLists.txt b/llvm/lib/Target/DirectX/CMakeLists.txt
index c7c09caf43667..8100f941c8d94 100644
--- a/llvm/lib/Target/DirectX/CMakeLists.txt
+++ b/llvm/lib/Target/DirectX/CMakeLists.txt
@@ -49,6 +49,7 @@ add_llvm_target(DirectXCodeGen
   DirectXInfo
   DirectXPointerTypeAnalysis
   FrontendHLSL
+  IPO
   MC
   ScalarOpts
   SelectionDAG
diff --git a/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp b/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp
index 5f331dbd2d826..13e3408815bba 100644
--- a/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp
+++ b/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp
@@ -20,13 +20,13 @@ using namespace llvm;
 static bool finalizeLinkage(Module &M) {
   bool MadeChange = false;
 
-  // Convert private global variables to internal linkage.
-  for (GlobalVariable &GV : M.globals()) {
-    if (GV.hasPrivateLinkage()) {
+  // Convert private globals and external globals with no usage to internal
+  // linkage.
+  for (GlobalVariable &GV : M.globals())
+    if (GV.hasPrivateLinkage() || (GV.hasExternalLinkage() && GV.use_empty())) {
       GV.setLinkage(GlobalValue::InternalLinkage);
       MadeChange = true;
     }
-  }
 
   SmallVector<Function *> Funcs;
 
diff --git a/llvm/lib/Target/DirectX/DirectXPassRegistry.def b/llvm/lib/Target/DirectX/DirectXPassRegistry.def
index d5069541642b5..b4b48a166800e 100644
--- a/llvm/lib/Target/DirectX/DirectXPassRegistry.def
+++ b/llvm/lib/Target/DirectX/DirectXPassRegistry.def
@@ -24,6 +24,7 @@ MODULE_ANALYSIS("dxil-root-signature-analysis", dxil::RootSignatureAnalysis())
 #define MODULE_PASS(NAME, CREATE_PASS)
 #endif
 MODULE_PASS("dxil-cbuffer-access", DXILCBufferAccess())
+MODULE_PASS("dxil-finalize-linkage", DXILFinalizeLinkage())
 MODULE_PASS("dxil-data-scalarization", DXILDataScalarization())
 MODULE_PASS("dxil-flatten-arrays", DXILFlattenArrays())
 MODULE_PASS("dxil-intrinsic-expansion", DXILIntrinsicExpansion())
diff --git a/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp b/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp
index 84751d2db2266..d7609d596e444 100644
--- a/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp
+++ b/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp
@@ -14,6 +14,7 @@
 #include "DirectXTargetMachine.h"
 #include "DXILCBufferAccess.h"
 #include "DXILDataScalarization.h"
+#include "DXILFinalizeLinkage.h"
 #include "DXILFlattenArrays.h"
 #include "DXILForwardHandleAccesses.h"
 #include "DXILIntrinsicExpansion.h"
@@ -45,6 +46,7 @@
 #include "llvm/Support/Compiler.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Target/TargetLoweringObjectFile.h"
+#include "llvm/Transforms/IPO/GlobalDCE.h"
 #include "llvm/Transforms/Scalar/Scalarizer.h"
 #include <optional>
 
@@ -62,6 +64,7 @@ extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeDirectXTarget() {
   initializeEmbedDXILPassPass(*PR);
   initializeWriteDXILPassPass(*PR);
   initializeDXContainerGlobalsPass(*PR);
+  initializeGlobalDCELegacyPassPass(*PR);
   initializeDXILOpLoweringLegacyPass(*PR);
   initializeDXILResourceAccessLegacyPass(*PR);
   initializeDXILResourceImplicitBindingLegacyPass(*PR);
@@ -103,6 +106,7 @@ class DirectXPassConfig : public TargetPassConfig {
   FunctionPass *createTargetRegisterAllocator(bool) override { return nullptr; }
   void addCodeGenPrepare() override {
     addPass(createDXILFinalizeLinkageLegacyPass());
+    addPass(createGlobalDCEPass());
     addPass(createDXILResourceAccessLegacyPass());
     addPass(createDXILIntrinsicExpansionLegacyPass());
     addPass(createDXILCBufferAccessLegacyPass());
diff --git a/llvm/lib/Transforms/IPO/GlobalDCE.cpp b/llvm/lib/Transforms/IPO/GlobalDCE.cpp
index 45fb1f5212b70..3812032488035 100644
--- a/llvm/lib/Transforms/IPO/GlobalDCE.cpp
+++ b/llvm/lib/Transforms/IPO/GlobalDCE.cpp
@@ -21,6 +21,8 @@
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/Module.h"
+#include "llvm/InitializePasses.h"
+#include "llvm/Pass.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Transforms/IPO.h"
 #include "llvm/Transforms/Utils/CtorUtils.h"
@@ -30,6 +32,36 @@ using namespace llvm;
 
 #define DEBUG_TYPE "globaldce"
 
+namespace {
+class GlobalDCELegacyPass : public ModulePass {
+public:
+  static char ID; // Pass identification, replacement for typeid
+  GlobalDCELegacyPass() : ModulePass(ID) {
+    initializeGlobalDCELegacyPassPass(*PassRegistry::getPassRegistry());
+  }
+  bool runOnModule(Module &M) override {
+    if (skipModule(M))
+      return false;
+    // Note: GlobalDCEPass does not use MAM. That
+    // means we can get away with init and pass
+    // as arg.
+    ModuleAnalysisManager MAM;
+    auto PA = Impl.run(M, MAM);
+    return !PA.areAllPreserved();
+  }
+
+private:
+  GlobalDCEPass Impl;
+};
+} // namespace
+
+char GlobalDCELegacyPass::ID = 0;
+INITIALIZE_PASS(GlobalDCELegacyPass, "globaldce", "Dead Global Elimination",
+                false, false)
+
+// Public interface to the GlobalDCEPass.
+ModulePass *llvm::createGlobalDCEPass() { return new GlobalDCELegacyPass(); }
+
 static cl::opt<bool>
     ClEnableVFE("enable-vfe", cl::Hidden, cl::init(true),
                 cl::desc("Enable virtual function elimination"));
diff --git a/llvm/test/CodeGen/DirectX/finalize_linkage.ll b/llvm/test/CodeGen/DirectX/finalize_linkage.ll
index dc1140f1c9160..0e45257ba1824 100644
--- a/llvm/test/CodeGen/DirectX/finalize_linkage.ll
+++ b/llvm/test/CodeGen/DirectX/finalize_linkage.ll
@@ -1,10 +1,17 @@
-; RUN: opt -S -dxil-finalize-linkage -mtriple=dxil-unknown-shadermodel6.5-compute %s | FileCheck %s
+; RUN: opt -S -passes='dxil-finalize-linkage,globaldce' -mtriple=dxil-unknown-shadermodel6.5-compute %s | FileCheck %s
 ; RUN: llc %s --filetype=asm -o - | FileCheck %s --check-prefixes=CHECK-LLC
 
 target triple = "dxilv1.5-pc-shadermodel6.5-compute"
 
 ; DXILFinalizeLinkage changes linkage of all functions that are hidden to
-; internal, and converts private global variables to internal linkage.
+; internal, converts private global variables to internal linkage, and removes
+; unused global variables.
+
+; CHECK-NOT: @aTile
+@aTile = hidden addrspace(3) global [4 x [1 x i32]] zeroinitializer, align 4
+
+; CHECK-NOT: @bTile
+@bTile = hidden addrspace(3) global [1 x <1 x i32>] zeroinitializer, align 4
 
 ; CHECK: @switch.table = internal unnamed_addr constant [4 x i32]
 @switch.table = private unnamed_addr constant [4 x i32] [i32 1, i32 257, i32 65793, i32 16843009], align 4
@@ -27,6 +34,17 @@ target triple = "dxilv1.5-pc-shadermodel6.5-compute"
 ; CHECK: @hidden_var = hidden global i32
 @hidden_var = hidden global i32 1, align 4
 
+define void @anchor_function() #0 {
+entry:
+  %0 = load i32, ptr @switch.table, align 4
+  %1 = load [3 x float], ptr @private_array, align 4
+  %2 = load i32, ptr @private_var, align 4
+  %3 = load i32, ptr @internal_var, align 4
+  %4 = load i32, ptr @external_var, align 4
+  %5 = load i32, ptr @hidden_var, align 4
+  ret void
+}
+
 ; CHECK-NOT: define internal void @"?f1@@YAXXZ"()
 define void @"?f1@@YAXXZ"() #0 {
 entry:
diff --git a/llvm/test/CodeGen/DirectX/llc-pipeline.ll b/llvm/test/CodeGen/DirectX/llc-pipeline.ll
index 151603a7161c5..a3a214338536f 100644
--- a/llvm/test/CodeGen/DirectX/llc-pipeline.ll
+++ b/llvm/test/CodeGen/DirectX/llc-pipeline.ll
@@ -14,6 +14,7 @@
 
 ; CHECK-NEXT: ModulePass Manager
 ; CHECK-NEXT:   DXIL Finalize Linkage
+; CHECK-NEXT:   Dead Global Elimination
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     DXIL Resource Access
 ; CHECK-NEXT:   DXIL Intrinsic Expansion
diff --git a/llvm/test/CodeGen/DirectX/scalar-data.ll b/llvm/test/CodeGen/DirectX/scalar-data.ll
index 4861a0890f136..d9c8df9bc169c 100644
--- a/llvm/test/CodeGen/DirectX/scalar-data.ll
+++ b/llvm/test/CodeGen/DirectX/scalar-data.ll
@@ -1,4 +1,4 @@
-; RUN: llc %s -mtriple=dxil-pc-shadermodel6.3-library --filetype=asm -o - | FileCheck %s
+; RUN: opt -S -passes='dxil-data-scalarization,dxil-flatten-arrays' -mtriple=dxil-unknown-shadermodel6.5-compute %s | FileCheck %s
 
 ; Make sure we don't touch arrays without vectors and that can recurse and flatten multiple-dimension arrays of vectors
 
diff --git a/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll b/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll
index 165d5d2e520df..c53bb2127ebcc 100644
--- a/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll
+++ b/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll
@@ -7,6 +7,7 @@ target triple = "dxil-unknown-shadermodel6.7-library"
 @CBV = external constant %"$Globals"
 
 define void @main() #0 {
+  %1 = load float, ptr @CBV, align 4
   ret void
 }
 

@llvmbot
Copy link
Member

llvmbot commented Jul 29, 2025

@llvm/pr-subscribers-llvm-transforms

Author: Kaitlin Peng (kmpeng)

Changes

Fixes #139023.

This PR essentially removes unused global variables:

  • Restores the GlobalDCE Legacy pass and adds it to the DirectX backend after the finalize linkage pass
  • Converts external global variables with no usage to internal linkage in the finalize linkage pass
    • (so they can be removed by GlobalDCE)
  • Makes the dxil-finalize-linkage pass usable using the new pass manager flag syntax
  • Adds tests to finalize_linkage.ll that make sure unused global variables are removed
  • Adds a use for variable @<!-- -->CBV in opaque-value_as_metadata.ll so it isn't removed
  • Changes the scalar-data.ll run command to avoid removing its global variables

Full diff: https://github.com/llvm/llvm-project/pull/151071.diff

12 Files Affected:

  • (modified) llvm/include/llvm/InitializePasses.h (+1)
  • (modified) llvm/include/llvm/LinkAllPasses.h (+2)
  • (modified) llvm/include/llvm/Transforms/IPO/GlobalDCE.h (+2)
  • (modified) llvm/lib/Target/DirectX/CMakeLists.txt (+1)
  • (modified) llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp (+4-4)
  • (modified) llvm/lib/Target/DirectX/DirectXPassRegistry.def (+1)
  • (modified) llvm/lib/Target/DirectX/DirectXTargetMachine.cpp (+4)
  • (modified) llvm/lib/Transforms/IPO/GlobalDCE.cpp (+32)
  • (modified) llvm/test/CodeGen/DirectX/finalize_linkage.ll (+20-2)
  • (modified) llvm/test/CodeGen/DirectX/llc-pipeline.ll (+1)
  • (modified) llvm/test/CodeGen/DirectX/scalar-data.ll (+1-1)
  • (modified) llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll (+1)
diff --git a/llvm/include/llvm/InitializePasses.h b/llvm/include/llvm/InitializePasses.h
index 31801daa126ad..0e48d80936270 100644
--- a/llvm/include/llvm/InitializePasses.h
+++ b/llvm/include/llvm/InitializePasses.h
@@ -127,6 +127,7 @@ LLVM_ABI void initializeGCEmptyBasicBlocksPass(PassRegistry &);
 LLVM_ABI void initializeGCMachineCodeAnalysisPass(PassRegistry &);
 LLVM_ABI void initializeGCModuleInfoPass(PassRegistry &);
 LLVM_ABI void initializeGVNLegacyPassPass(PassRegistry &);
+LLVM_ABI void initializeGlobalDCELegacyPassPass(PassRegistry &);
 LLVM_ABI void initializeGlobalMergeFuncPassWrapperPass(PassRegistry &);
 LLVM_ABI void initializeGlobalMergePass(PassRegistry &);
 LLVM_ABI void initializeGlobalsAAWrapperPassPass(PassRegistry &);
diff --git a/llvm/include/llvm/LinkAllPasses.h b/llvm/include/llvm/LinkAllPasses.h
index f82a43967e67a..7b33ec7f09124 100644
--- a/llvm/include/llvm/LinkAllPasses.h
+++ b/llvm/include/llvm/LinkAllPasses.h
@@ -38,6 +38,7 @@
 #include "llvm/Support/Valgrind.h"
 #include "llvm/Transforms/IPO.h"
 #include "llvm/Transforms/IPO/AlwaysInliner.h"
+#include "llvm/Transforms/IPO/GlobalDCE.h"
 #include "llvm/Transforms/InstCombine/InstCombine.h"
 #include "llvm/Transforms/ObjCARC.h"
 #include "llvm/Transforms/Scalar.h"
@@ -83,6 +84,7 @@ struct ForcePassLinking {
     (void)llvm::createDomOnlyViewerWrapperPassPass();
     (void)llvm::createDomViewerWrapperPassPass();
     (void)llvm::createAlwaysInlinerLegacyPass();
+    (void)llvm::createGlobalDCEPass();
     (void)llvm::createGlobalMergeFuncPass();
     (void)llvm::createGlobalsAAWrapperPass();
     (void)llvm::createInstSimplifyLegacyPass();
diff --git a/llvm/include/llvm/Transforms/IPO/GlobalDCE.h b/llvm/include/llvm/Transforms/IPO/GlobalDCE.h
index f8f775b8ea96e..55dff64889270 100644
--- a/llvm/include/llvm/Transforms/IPO/GlobalDCE.h
+++ b/llvm/include/llvm/Transforms/IPO/GlobalDCE.h
@@ -32,6 +32,7 @@ class GlobalVariable;
 class Metadata;
 class Module;
 class Value;
+class ModulePass;
 
 /// Pass to remove unused function declarations.
 class GlobalDCEPass : public PassInfoMixin<GlobalDCEPass> {
@@ -80,6 +81,7 @@ class GlobalDCEPass : public PassInfoMixin<GlobalDCEPass> {
   void ComputeDependencies(Value *V, SmallPtrSetImpl<GlobalValue *> &U);
 };
 
+ModulePass *createGlobalDCEPass();
 }
 
 #endif // LLVM_TRANSFORMS_IPO_GLOBALDCE_H
diff --git a/llvm/lib/Target/DirectX/CMakeLists.txt b/llvm/lib/Target/DirectX/CMakeLists.txt
index c7c09caf43667..8100f941c8d94 100644
--- a/llvm/lib/Target/DirectX/CMakeLists.txt
+++ b/llvm/lib/Target/DirectX/CMakeLists.txt
@@ -49,6 +49,7 @@ add_llvm_target(DirectXCodeGen
   DirectXInfo
   DirectXPointerTypeAnalysis
   FrontendHLSL
+  IPO
   MC
   ScalarOpts
   SelectionDAG
diff --git a/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp b/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp
index 5f331dbd2d826..13e3408815bba 100644
--- a/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp
+++ b/llvm/lib/Target/DirectX/DXILFinalizeLinkage.cpp
@@ -20,13 +20,13 @@ using namespace llvm;
 static bool finalizeLinkage(Module &M) {
   bool MadeChange = false;
 
-  // Convert private global variables to internal linkage.
-  for (GlobalVariable &GV : M.globals()) {
-    if (GV.hasPrivateLinkage()) {
+  // Convert private globals and external globals with no usage to internal
+  // linkage.
+  for (GlobalVariable &GV : M.globals())
+    if (GV.hasPrivateLinkage() || (GV.hasExternalLinkage() && GV.use_empty())) {
       GV.setLinkage(GlobalValue::InternalLinkage);
       MadeChange = true;
     }
-  }
 
   SmallVector<Function *> Funcs;
 
diff --git a/llvm/lib/Target/DirectX/DirectXPassRegistry.def b/llvm/lib/Target/DirectX/DirectXPassRegistry.def
index d5069541642b5..b4b48a166800e 100644
--- a/llvm/lib/Target/DirectX/DirectXPassRegistry.def
+++ b/llvm/lib/Target/DirectX/DirectXPassRegistry.def
@@ -24,6 +24,7 @@ MODULE_ANALYSIS("dxil-root-signature-analysis", dxil::RootSignatureAnalysis())
 #define MODULE_PASS(NAME, CREATE_PASS)
 #endif
 MODULE_PASS("dxil-cbuffer-access", DXILCBufferAccess())
+MODULE_PASS("dxil-finalize-linkage", DXILFinalizeLinkage())
 MODULE_PASS("dxil-data-scalarization", DXILDataScalarization())
 MODULE_PASS("dxil-flatten-arrays", DXILFlattenArrays())
 MODULE_PASS("dxil-intrinsic-expansion", DXILIntrinsicExpansion())
diff --git a/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp b/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp
index 84751d2db2266..d7609d596e444 100644
--- a/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp
+++ b/llvm/lib/Target/DirectX/DirectXTargetMachine.cpp
@@ -14,6 +14,7 @@
 #include "DirectXTargetMachine.h"
 #include "DXILCBufferAccess.h"
 #include "DXILDataScalarization.h"
+#include "DXILFinalizeLinkage.h"
 #include "DXILFlattenArrays.h"
 #include "DXILForwardHandleAccesses.h"
 #include "DXILIntrinsicExpansion.h"
@@ -45,6 +46,7 @@
 #include "llvm/Support/Compiler.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Target/TargetLoweringObjectFile.h"
+#include "llvm/Transforms/IPO/GlobalDCE.h"
 #include "llvm/Transforms/Scalar/Scalarizer.h"
 #include <optional>
 
@@ -62,6 +64,7 @@ extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeDirectXTarget() {
   initializeEmbedDXILPassPass(*PR);
   initializeWriteDXILPassPass(*PR);
   initializeDXContainerGlobalsPass(*PR);
+  initializeGlobalDCELegacyPassPass(*PR);
   initializeDXILOpLoweringLegacyPass(*PR);
   initializeDXILResourceAccessLegacyPass(*PR);
   initializeDXILResourceImplicitBindingLegacyPass(*PR);
@@ -103,6 +106,7 @@ class DirectXPassConfig : public TargetPassConfig {
   FunctionPass *createTargetRegisterAllocator(bool) override { return nullptr; }
   void addCodeGenPrepare() override {
     addPass(createDXILFinalizeLinkageLegacyPass());
+    addPass(createGlobalDCEPass());
     addPass(createDXILResourceAccessLegacyPass());
     addPass(createDXILIntrinsicExpansionLegacyPass());
     addPass(createDXILCBufferAccessLegacyPass());
diff --git a/llvm/lib/Transforms/IPO/GlobalDCE.cpp b/llvm/lib/Transforms/IPO/GlobalDCE.cpp
index 45fb1f5212b70..3812032488035 100644
--- a/llvm/lib/Transforms/IPO/GlobalDCE.cpp
+++ b/llvm/lib/Transforms/IPO/GlobalDCE.cpp
@@ -21,6 +21,8 @@
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/Module.h"
+#include "llvm/InitializePasses.h"
+#include "llvm/Pass.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Transforms/IPO.h"
 #include "llvm/Transforms/Utils/CtorUtils.h"
@@ -30,6 +32,36 @@ using namespace llvm;
 
 #define DEBUG_TYPE "globaldce"
 
+namespace {
+class GlobalDCELegacyPass : public ModulePass {
+public:
+  static char ID; // Pass identification, replacement for typeid
+  GlobalDCELegacyPass() : ModulePass(ID) {
+    initializeGlobalDCELegacyPassPass(*PassRegistry::getPassRegistry());
+  }
+  bool runOnModule(Module &M) override {
+    if (skipModule(M))
+      return false;
+    // Note: GlobalDCEPass does not use MAM. That
+    // means we can get away with init and pass
+    // as arg.
+    ModuleAnalysisManager MAM;
+    auto PA = Impl.run(M, MAM);
+    return !PA.areAllPreserved();
+  }
+
+private:
+  GlobalDCEPass Impl;
+};
+} // namespace
+
+char GlobalDCELegacyPass::ID = 0;
+INITIALIZE_PASS(GlobalDCELegacyPass, "globaldce", "Dead Global Elimination",
+                false, false)
+
+// Public interface to the GlobalDCEPass.
+ModulePass *llvm::createGlobalDCEPass() { return new GlobalDCELegacyPass(); }
+
 static cl::opt<bool>
     ClEnableVFE("enable-vfe", cl::Hidden, cl::init(true),
                 cl::desc("Enable virtual function elimination"));
diff --git a/llvm/test/CodeGen/DirectX/finalize_linkage.ll b/llvm/test/CodeGen/DirectX/finalize_linkage.ll
index dc1140f1c9160..0e45257ba1824 100644
--- a/llvm/test/CodeGen/DirectX/finalize_linkage.ll
+++ b/llvm/test/CodeGen/DirectX/finalize_linkage.ll
@@ -1,10 +1,17 @@
-; RUN: opt -S -dxil-finalize-linkage -mtriple=dxil-unknown-shadermodel6.5-compute %s | FileCheck %s
+; RUN: opt -S -passes='dxil-finalize-linkage,globaldce' -mtriple=dxil-unknown-shadermodel6.5-compute %s | FileCheck %s
 ; RUN: llc %s --filetype=asm -o - | FileCheck %s --check-prefixes=CHECK-LLC
 
 target triple = "dxilv1.5-pc-shadermodel6.5-compute"
 
 ; DXILFinalizeLinkage changes linkage of all functions that are hidden to
-; internal, and converts private global variables to internal linkage.
+; internal, converts private global variables to internal linkage, and removes
+; unused global variables.
+
+; CHECK-NOT: @aTile
+@aTile = hidden addrspace(3) global [4 x [1 x i32]] zeroinitializer, align 4
+
+; CHECK-NOT: @bTile
+@bTile = hidden addrspace(3) global [1 x <1 x i32>] zeroinitializer, align 4
 
 ; CHECK: @switch.table = internal unnamed_addr constant [4 x i32]
 @switch.table = private unnamed_addr constant [4 x i32] [i32 1, i32 257, i32 65793, i32 16843009], align 4
@@ -27,6 +34,17 @@ target triple = "dxilv1.5-pc-shadermodel6.5-compute"
 ; CHECK: @hidden_var = hidden global i32
 @hidden_var = hidden global i32 1, align 4
 
+define void @anchor_function() #0 {
+entry:
+  %0 = load i32, ptr @switch.table, align 4
+  %1 = load [3 x float], ptr @private_array, align 4
+  %2 = load i32, ptr @private_var, align 4
+  %3 = load i32, ptr @internal_var, align 4
+  %4 = load i32, ptr @external_var, align 4
+  %5 = load i32, ptr @hidden_var, align 4
+  ret void
+}
+
 ; CHECK-NOT: define internal void @"?f1@@YAXXZ"()
 define void @"?f1@@YAXXZ"() #0 {
 entry:
diff --git a/llvm/test/CodeGen/DirectX/llc-pipeline.ll b/llvm/test/CodeGen/DirectX/llc-pipeline.ll
index 151603a7161c5..a3a214338536f 100644
--- a/llvm/test/CodeGen/DirectX/llc-pipeline.ll
+++ b/llvm/test/CodeGen/DirectX/llc-pipeline.ll
@@ -14,6 +14,7 @@
 
 ; CHECK-NEXT: ModulePass Manager
 ; CHECK-NEXT:   DXIL Finalize Linkage
+; CHECK-NEXT:   Dead Global Elimination
 ; CHECK-NEXT:   FunctionPass Manager
 ; CHECK-NEXT:     DXIL Resource Access
 ; CHECK-NEXT:   DXIL Intrinsic Expansion
diff --git a/llvm/test/CodeGen/DirectX/scalar-data.ll b/llvm/test/CodeGen/DirectX/scalar-data.ll
index 4861a0890f136..d9c8df9bc169c 100644
--- a/llvm/test/CodeGen/DirectX/scalar-data.ll
+++ b/llvm/test/CodeGen/DirectX/scalar-data.ll
@@ -1,4 +1,4 @@
-; RUN: llc %s -mtriple=dxil-pc-shadermodel6.3-library --filetype=asm -o - | FileCheck %s
+; RUN: opt -S -passes='dxil-data-scalarization,dxil-flatten-arrays' -mtriple=dxil-unknown-shadermodel6.5-compute %s | FileCheck %s
 
 ; Make sure we don't touch arrays without vectors and that can recurse and flatten multiple-dimension arrays of vectors
 
diff --git a/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll b/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll
index 165d5d2e520df..c53bb2127ebcc 100644
--- a/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll
+++ b/llvm/test/tools/dxil-dis/opaque-value_as_metadata.ll
@@ -7,6 +7,7 @@ target triple = "dxil-unknown-shadermodel6.7-library"
 @CBV = external constant %"$Globals"
 
 define void @main() #0 {
+  %1 = load float, ptr @CBV, align 4
   ret void
 }
 

// Convert private globals and external globals with no usage to internal
// linkage.
for (GlobalVariable &GV : M.globals())
if (GV.hasPrivateLinkage() || (GV.hasExternalLinkage() && GV.use_empty())) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For anyone else looking this is the crux of the change right here. GlobalDCE is conservative so it won't remove globals marked external. This is because Linker would handle these globals and usually something like the internalizer pass would run and do a similar conversion to what we are doing here.

In HLSL's case after finalize linkage we should be able to assume that it is safe to mark all external globals with no uses as internal. If we then run GlobalDCE after this these globals will get cleaned up.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we're just concerned with deleting unused external variables... we could do that as part of DXILFinalizeLinkage instead of running GlobalDCE in the backend.

Are there other cases that GlobalDCE catches that we need to catch late?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Justin suggested we not do special casing of global_ctors and that GlobalDCE was a more complete solution. Hence why we closed the previous attempt at fixing this issue.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We need to special case global constructors for library shaders. That's not optional.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we use GlobalDCE we don't need to special case, things just work by converting to internal.

Copy link
Contributor

@bogner bogner left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This looks basically good to me. A few minor comments.

@@ -1,4 +1,4 @@
; RUN: llc %s -mtriple=dxil-pc-shadermodel6.3-library --filetype=asm -o - | FileCheck %s
; RUN: opt -S -passes='dxil-data-scalarization,dxil-flatten-arrays' -mtriple=dxil-unknown-shadermodel6.5-compute %s | FileCheck %s
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would it be better to modify the linkage of the globals in this test to be external rather than changing how we test this?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@farzonl suggested changing it to opt - do you have thoughts on this?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah switched to opt b\c globalDCE would eliminate globals that were being used in this test.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I meant do you have thoughts on if it would be better to modify the linkages instead of changing to opt?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This test case is intended to be very limited to the effects of the array flattening and vector scalarization passes. Even if we change the linkage to external it would still get cleaned up by Global DCE. Since we know which two passes are responsible for the data transformations I think its fine to just run those two passes.

@kmpeng kmpeng force-pushed the bugfix/unused-globals branch from 0688a7e to dcf76b9 Compare August 7, 2025 18:22
@kmpeng kmpeng force-pushed the bugfix/unused-globals branch from dcf76b9 to d1b498e Compare August 7, 2025 18:23
@farzonl
Copy link
Member

farzonl commented Aug 11, 2025

I did a run of the DML Shaders today. There are currently only two validator errors left and none of them are Internal declaration is unused . I'm conflicted now on moving forward with this PR.

The remaining two validation issues:

40x error: Bitcast on struct types is not allowed.
32x error: Pointer type bitcast must be have same size.

@Icohedron are you aware of any changes that went in that could have impacted these validation issues? Is it possible the bitcast errors are masking the unused errors?

@bogner should we hold off on this pr?

@Icohedron
Copy link
Contributor

@Icohedron are you aware of any changes that went in that could have impacted these validation issues? Is it possible the bitcast errors are masking the unused errors?

No, I am not aware of any changes that could have impacted these validation issues. I also don't think the bitcast errors are masking the unused errors because they used to coexist.

@Icohedron
Copy link
Contributor

Icohedron commented Aug 11, 2025

The remaining two validation issues:

40x error: Bitcast on struct types is not allowed.
32x error: Pointer type bitcast must be have same size.

Also there is a third remaining validation issue: 120 occurrences of Load/Store operand is not a pointer type
The validation failure message doesn't contain "error:" so your script didn't pick up on it.

This validation issue could be masking the unused errors.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[DirectX] Validator is erroring with Internal declaration '.*' is unused
6 participants