release/21.x: [BPF] Backport backend fixes related to BTF

llvm/lib/Target/BPF/BTFDebug.cpp

@@ -14,6 +14,7 @@
 #include "BPF.h"
 #include "BPFCORE.h"
 #include "MCTargetDesc/BPFMCTargetDesc.h"
+#include "llvm/BinaryFormat/Dwarf.h"
 #include "llvm/BinaryFormat/ELF.h"
 #include "llvm/CodeGen/AsmPrinter.h"
 #include "llvm/CodeGen/MachineModuleInfo.h"
@@ -23,6 +24,7 @@
 #include "llvm/MC/MCObjectFileInfo.h"
 #include "llvm/MC/MCSectionELF.h"
 #include "llvm/MC/MCStreamer.h"
+#include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/LineIterator.h"
 #include "llvm/Support/MemoryBuffer.h"
 #include "llvm/Target/TargetLoweringObjectFile.h"
@@ -93,7 +95,24 @@ void BTFTypeDerived::completeType(BTFDebug &BDebug) {
     return;
   IsCompleted = true;
 
-  BTFType.NameOff = BDebug.addString(Name);
+  switch (Kind) {
+  case BTF::BTF_KIND_PTR:
+  case BTF::BTF_KIND_CONST:
+  case BTF::BTF_KIND_VOLATILE:
+  case BTF::BTF_KIND_RESTRICT:
+    // Debug info might contain names for these types, but given that we want
+    // to keep BTF minimal and naming reference types doesn't bring any value
+    // (what matters is the completeness of the base type), we don't emit them.
+    //
+    // Furthermore, the Linux kernel refuses to load BPF programs that contain
+    // BTF with these types named:
+    // https://elixir.bootlin.com/linux/v6.17.1/source/kernel/bpf/btf.c#L2586
+    BTFType.NameOff = 0;
+    break;
+  default:
+    BTFType.NameOff = BDebug.addString(Name);
+    break;
+  }
 
   if (NeedsFixup || !DTy)
     return;
@@ -301,21 +320,59 @@ void BTFTypeStruct::completeType(BTFDebug &BDebug) {
 
   BTFType.NameOff = BDebug.addString(STy->getName());
 
+  if (STy->getTag() == dwarf::DW_TAG_variant_part) {
+    // Variant parts might have a discriminator, which has its own memory
+    // location, and variants, which share the memory location afterwards. LLVM
+    // DI doesn't consider discriminator as an element and instead keeps
+    // it as a separate reference.
+    // To keep BTF simple, let's represent the structure as an union with
+    // discriminator as the first element.
+    // The offsets inside variant types are already handled correctly in the
+    // DI.
+    const auto *DTy = STy->getDiscriminator();
+    if (DTy) {
+      struct BTF::BTFMember Discriminator;
+
+      Discriminator.NameOff = BDebug.addString(DTy->getName());
+      Discriminator.Offset = DTy->getOffsetInBits();
+      const auto *BaseTy = DTy->getBaseType();
+      Discriminator.Type = BDebug.getTypeId(BaseTy);
+
+      Members.push_back(Discriminator);
+    }
+  }
+
   // Add struct/union members.
   const DINodeArray Elements = STy->getElements();
   for (const auto *Element : Elements) {
     struct BTF::BTFMember BTFMember;
-    const auto *DDTy = cast<DIDerivedType>(Element);
 
-    BTFMember.NameOff = BDebug.addString(DDTy->getName());
-    if (HasBitField) {
-      uint8_t BitFieldSize = DDTy->isBitField() ? DDTy->getSizeInBits() : 0;
-      BTFMember.Offset = BitFieldSize << 24 | DDTy->getOffsetInBits();
-    } else {
-      BTFMember.Offset = DDTy->getOffsetInBits();
+    switch (Element->getTag()) {
+    case dwarf::DW_TAG_member: {
+      const auto *DDTy = cast<DIDerivedType>(Element);
+
+      BTFMember.NameOff = BDebug.addString(DDTy->getName());
+      if (HasBitField) {
+        uint8_t BitFieldSize = DDTy->isBitField() ? DDTy->getSizeInBits() : 0;
+        BTFMember.Offset = BitFieldSize << 24 | DDTy->getOffsetInBits();
+      } else {
+        BTFMember.Offset = DDTy->getOffsetInBits();
+      }
+      const auto *BaseTy = tryRemoveAtomicType(DDTy->getBaseType());
+      BTFMember.Type = BDebug.getTypeId(BaseTy);
+      break;
+    }
+    case dwarf::DW_TAG_variant_part: {
+      const auto *DCTy = dyn_cast<DICompositeType>(Element);
+
+      BTFMember.NameOff = BDebug.addString(DCTy->getName());
+      BTFMember.Offset = DCTy->getOffsetInBits();
+      BTFMember.Type = BDebug.getTypeId(DCTy);
+      break;
+    }
+    default:
+      llvm_unreachable("Unexpected DI tag of a struct/union element");
     }
-    const auto *BaseTy = tryRemoveAtomicType(DDTy->getBaseType());
-    BTFMember.Type = BDebug.getTypeId(BaseTy);
     Members.push_back(BTFMember);
   }
 }
@@ -672,16 +729,28 @@ void BTFDebug::visitStructType(const DICompositeType *CTy, bool IsStruct,
                                uint32_t &TypeId) {
   const DINodeArray Elements = CTy->getElements();
   uint32_t VLen = Elements.size();
+  // Variant parts might have a discriminator. LLVM DI doesn't consider it as
+  // an element and instead keeps it as a separate reference. But we represent
+  // it as an element in BTF.
+  if (CTy->getTag() == dwarf::DW_TAG_variant_part) {
+    const auto *DTy = CTy->getDiscriminator();
+    if (DTy) {
+      visitTypeEntry(DTy);
+      VLen++;
+    }
+  }
   if (VLen > BTF::MAX_VLEN)
     return;
 
   // Check whether we have any bitfield members or not
   bool HasBitField = false;
   for (const auto *Element : Elements) {
-    auto E = cast<DIDerivedType>(Element);
-    if (E->isBitField()) {
-      HasBitField = true;
-      break;
+    if (Element->getTag() == dwarf::DW_TAG_member) {
+      auto E = cast<DIDerivedType>(Element);
+      if (E->isBitField()) {
+        HasBitField = true;
+        break;
+      }
     }
   }
 
@@ -696,9 +765,22 @@ void BTFDebug::visitStructType(const DICompositeType *CTy, bool IsStruct,
   // Visit all struct members.
   int FieldNo = 0;
   for (const auto *Element : Elements) {
-    const auto Elem = cast<DIDerivedType>(Element);
-    visitTypeEntry(Elem);
-    processDeclAnnotations(Elem->getAnnotations(), TypeId, FieldNo);
+    switch (Element->getTag()) {
+    case dwarf::DW_TAG_member: {
+      const auto Elem = cast<DIDerivedType>(Element);
+      visitTypeEntry(Elem);
+      processDeclAnnotations(Elem->getAnnotations(), TypeId, FieldNo);
+      break;
+    }
+    case dwarf::DW_TAG_variant_part: {
+      const auto Elem = cast<DICompositeType>(Element);
+      visitTypeEntry(Elem);
+      processDeclAnnotations(Elem->getAnnotations(), TypeId, FieldNo);
+      break;
+    }
+    default:
+      llvm_unreachable("Unexpected DI tag of a struct/union element");
+    }
     FieldNo++;
   }
 }
@@ -781,16 +863,25 @@ void BTFDebug::visitFwdDeclType(const DICompositeType *CTy, bool IsUnion,
 void BTFDebug::visitCompositeType(const DICompositeType *CTy,
                                   uint32_t &TypeId) {
   auto Tag = CTy->getTag();
-  if (Tag == dwarf::DW_TAG_structure_type || Tag == dwarf::DW_TAG_union_type) {
+  switch (Tag) {
+  case dwarf::DW_TAG_structure_type:
+  case dwarf::DW_TAG_union_type:
+  case dwarf::DW_TAG_variant_part:
     // Handle forward declaration differently as it does not have members.
     if (CTy->isForwardDecl())
       visitFwdDeclType(CTy, Tag == dwarf::DW_TAG_union_type, TypeId);
     else
       visitStructType(CTy, Tag == dwarf::DW_TAG_structure_type, TypeId);
-  } else if (Tag == dwarf::DW_TAG_array_type)
+    break;
+  case dwarf::DW_TAG_array_type:
     visitArrayType(CTy, TypeId);
-  else if (Tag == dwarf::DW_TAG_enumeration_type)
+    break;
+  case dwarf::DW_TAG_enumeration_type:
     visitEnumType(CTy, TypeId);
+    break;
+  default:
+    llvm_unreachable("Unexpected DI tag of a composite type");
+  }
 }
 
 bool BTFDebug::IsForwardDeclCandidate(const DIType *Base) {

llvm/test/CodeGen/BPF/BTF/ptr-named-2.ll

@@ -0,0 +1,59 @@
+; RUN: llc -mtriple=bpfel -filetype=obj -o %t1 %s
+; RUN: llvm-objcopy --dump-section='.BTF'=%t2 %t1
+; RUN: %python %p/print_btf.py %t2 | FileCheck -check-prefixes=CHECK-BTF %s
+; RUN: llc -mtriple=bpfeb -filetype=obj -o %t1 %s
+; RUN: llvm-objcopy --dump-section='.BTF'=%t2 %t1
+; RUN: %python %p/print_btf.py %t2 | FileCheck -check-prefixes=CHECK-BTF %s
+;
+; This IR is hand-written.
+
+; ModuleID = 'ptr-named-2.ll'
+source_filename = "ptr-named-2.ll"
+target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n32:64-S128"
+target triple = "bpfel-unknown-none"
+
+%struct.TypeExamples = type { i32*, i32, i32, i32* }
+
+@type_examples = internal global %struct.TypeExamples zeroinitializer, align 8, !dbg !0
+
+!llvm.dbg.cu = !{!1}
+!llvm.module.flags = !{!2, !3, !4}
+!llvm.ident = !{!21}
+
+; CHECK-BTF:      [1] STRUCT 'TypeExamples' size=32 vlen=4
+; CHECK-BTF-NEXT:         'ptr' type_id=2 bits_offset=0
+; CHECK-BTF-NEXT:         'volatile' type_id=4 bits_offset=64
+; CHECK-BTF-NEXT:         'const' type_id=5 bits_offset=128
+; CHECK-BTF-NEXT:         'restrict_ptr' type_id=6 bits_offset=192
+; CHECK-BTF-NEXT: [2] PTR '(anon)' type_id=3
+; CHECK-BTF-NEXT: [3] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED
+; CHECK-BTF-NEXT: [4] VOLATILE '(anon)' type_id=3
+; CHECK-BTF-NEXT: [5] CONST '(anon)' type_id=3
+; CHECK-BTF-NEXT: [6] RESTRICT '(anon)' type_id=7
+; CHECK-BTF-NEXT: [7] PTR '(anon)' type_id=3
+; CHECK-BTF-NEXT: [8] VAR 'type_examples' type_id=1, linkage=static
+; CHECK-BTF-NEXT: [9] DATASEC '.bss' size=0 vlen=1
+; CHECK-BTF-NEXT:         type_id=8 offset=0 size=24
+
+!0 = !DIGlobalVariableExpression(var: !5, expr: !DIExpression())
+!1 = distinct !DICompileUnit(language: DW_LANG_C99, file: !6, isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !7, globals: !8, splitDebugInlining: false, nameTableKind: None)
+!2 = !{i32 2, !"Dwarf Version", i32 4}
+!3 = !{i32 2, !"Debug Info Version", i32 3}
+!4 = !{i32 1, !"wchar_size", i32 4}
+!5 = distinct !DIGlobalVariable(name: "type_examples", scope: !1, file: !6, line: 12, type: !9, isLocal: true, isDefinition: true)
+!6 = !DIFile(filename: "ptr-named-2.ll", directory: "/tmp")
+!7 = !{}
+!8 = !{!0}
+!9 = distinct !DICompositeType(tag: DW_TAG_structure_type, name: "TypeExamples", file: !6, line: 5, size: 256, elements: !10)
+!10 = !{!11, !12, !13, !14}
+!11 = !DIDerivedType(tag: DW_TAG_member, name: "ptr", scope: !9, file: !6, line: 6, baseType: !15, size: 64)
+!12 = !DIDerivedType(tag: DW_TAG_member, name: "volatile", scope: !9, file: !6, line: 7, baseType: !17, size: 64, offset: 64)
+!13 = !DIDerivedType(tag: DW_TAG_member, name: "const", scope: !9, file: !6, line: 8, baseType: !18, size: 64, offset: 128)
+!14 = !DIDerivedType(tag: DW_TAG_member, name: "restrict_ptr", scope: !9, file: !6, line: 9, baseType: !19, size: 64, offset: 192)
+!15 = !DIDerivedType(tag: DW_TAG_pointer_type, name: "*int", baseType: !16, size: 64)
+!16 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
+!17 = !DIDerivedType(tag: DW_TAG_volatile_type, name: "volatile int", baseType: !16)
+!18 = !DIDerivedType(tag: DW_TAG_const_type, name: "const int", baseType: !16)
+!19 = !DIDerivedType(tag: DW_TAG_restrict_type, name: "*int restrict", baseType: !20)
+!20 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !16, size: 64)
+!21 = !{!"my hand-written IR"}

llvm/test/CodeGen/BPF/BTF/ptr-named.ll

@@ -0,0 +1,75 @@
+; RUN: llc -mtriple=bpfel -filetype=obj -o %t1 %s
+; RUN: llvm-objcopy --dump-section='.BTF'=%t2 %t1
+; RUN: %python %p/print_btf.py %t2 | FileCheck -check-prefixes=CHECK-BTF %s
+; RUN: llc -mtriple=bpfeb -filetype=obj -o %t1 %s
+; RUN: llvm-objcopy --dump-section='.BTF'=%t2 %t1
+; RUN: %python %p/print_btf.py %t2 | FileCheck -check-prefixes=CHECK-BTF %s
+;
+; Source:
+;   #![no_std]
+;   #![no_main]
+;
+;   pub struct MyType {
+;       ptr: *const u32,
+;   }
+;
+;   impl MyType {
+;       pub const fn new() -> Self {
+;           let ptr = core::ptr::null();
+;           Self { ptr }
+;       }
+;   }
+;
+;   unsafe impl Sync for MyType {}
+;
+;   #[unsafe(no_mangle)]
+;   pub static X: MyType = MyType::new();
+;
+;   #[cfg(not(test))]
+;   #[panic_handler]
+;   fn panic(_info: &core::panic::PanicInfo) -> ! {
+;       loop {}
+;   }
+; Compilation flag:
+;   cargo +nightly rustc -Zbuild-std=core --target=bpfel-unknown-none -- --emit=llvm-bc
+;   llvm-extract --glob=X $(find target/ -name "*.bc" | head -n 1) -o ptr-named.bc
+;   llvm-dis ptr-named.bc -o ptr-named.ll
+
+; ModuleID = 'ptr-named.bc'
+source_filename = "1m2uqe50qkwxmo53ydydvou91"
+target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n32:64-S128"
+target triple = "bpfel"
+
+@X = constant [8 x i8] zeroinitializer, align 8, !dbg !0
+
+!llvm.module.flags = !{!11, !12, !13, !14}
+!llvm.ident = !{!15}
+!llvm.dbg.cu = !{!16}
+
+; CHECK-BTF:      [1] STRUCT 'MyType' size=8 vlen=1
+; CHECK-BTF-NEXT:         'ptr' type_id=2 bits_offset=0
+; CHECK-BTF-NEXT: [2] PTR '(anon)' type_id=3
+; CHECK-BTF-NEXT: [3] INT 'u32' size=4 bits_offset=0 nr_bits=32 encoding=(none)
+; CHECK-BTF-NEXT: [4] VAR 'X' type_id=1, linkage=global
+; CHECK-BTF-NEXT: [5] DATASEC '.rodata' size=0 vlen=1
+; CHECK-BTF-NEXT:         type_id=4 offset=0 size=8
+
+!0 = !DIGlobalVariableExpression(var: !1, expr: !DIExpression())
+!1 = distinct !DIGlobalVariable(name: "X", scope: !2, file: !3, line: 19, type: !4, isLocal: false, isDefinition: true, align: 64)
+!2 = !DINamespace(name: "ptr_named", scope: null)
+!3 = !DIFile(filename: "ptr-named/src/main.rs", directory: "/tmp/ptr-named", checksumkind: CSK_MD5, checksum: "e37168304600b30cbb5ba168f0384932")
+!4 = !DICompositeType(tag: DW_TAG_structure_type, name: "MyType", scope: !2, file: !5, size: 64, align: 64, flags: DIFlagPublic, elements: !6, templateParams: !10, identifier: "7609fa40332dd486922f074276a171c3")
+!5 = !DIFile(filename: "<unknown>", directory: "")
+!6 = !{!7}
+!7 = !DIDerivedType(tag: DW_TAG_member, name: "ptr", scope: !4, file: !5, baseType: !8, size: 64, align: 64, flags: DIFlagPrivate)
+!8 = !DIDerivedType(tag: DW_TAG_pointer_type, name: "*const u32", baseType: !9, size: 64, align: 64, dwarfAddressSpace: 0)
+!9 = !DIBasicType(name: "u32", size: 32, encoding: DW_ATE_unsigned)
+!10 = !{}
+!11 = !{i32 8, !"PIC Level", i32 2}
+!12 = !{i32 7, !"PIE Level", i32 2}
+!13 = !{i32 7, !"Dwarf Version", i32 4}
+!14 = !{i32 2, !"Debug Info Version", i32 3}
+!15 = !{!"rustc version 1.92.0-nightly (c8905eaa6 2025-09-28)"}
+!16 = distinct !DICompileUnit(language: DW_LANG_Rust, file: !17, producer: "clang LLVM (rustc version 1.92.0-nightly (c8905eaa6 2025-09-28))", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, globals: !18, splitDebugInlining: false, nameTableKind: None)
+!17 = !DIFile(filename: "ptr-named/src/main.rs/@/1m2uqe50qkwxmo53ydydvou91", directory: "/tmp/ptr-named")
+!18 = !{!0}