offsetof: Support sub-field

Since bpftrace PR [1] support offsetof() function. As previously discussed [2], the current offsetof() does not support sub-fields, and this PR implements offsetof(struct foo, a.b.c), implement [1] unfinished features. Example: struct Foo { struct e { int a; int b; struct c { int a, b, c; } c; } e; } struct offsetof { struct { int comm; int ustack; } kstack; } BEGIN { printf("struct Foo e.b offset %ld\n", offsetof(struct Foo, e.b)); printf("struct Foo e.c offset %ld\n", offsetof(struct Foo, e.c)); printf("struct Foo e.c.b offset %ld\n", offsetof(struct Foo, e.c.b)); printf("struct offsetof kstack.ustack offset %ld\n", offsetof(struct offsetof, kstack.ustack)); exit(); } Link: bpftrace#2579 [1] merged Link: bpftrace#2216 [2] Signed-off-by: Rong Tao <rongtao@cestc.cn>
Rtoax · Feb 1, 2025 · 036a6b0 · 036a6b0
1 parent 04c87f6
commit 036a6b0
Show file tree

Hide file tree

Showing 7 changed files with 197 additions and 15 deletions.
diff --git a/src/ast/passes/codegen_llvm.cpp b/src/ast/passes/codegen_llvm.cpp
@@ -1522,8 +1522,21 @@ void CodegenLLVM::visit(Sizeof &szof)
 
 void CodegenLLVM::visit(Offsetof &ofof)
 {
-  auto &field = ofof.record.GetField(ofof.field);
-  expr_ = b_.getInt64(field.offset);
+  ssize_t offset = 0;
+  std::string token;
+  std::stringstream ss(ofof.field);
+  std::vector<std::string> tokens;
+
+  while (std::getline(ss, token, '.')) {
+    tokens.push_back(token);
+  }
+
+  const SizedType *record = &ofof.record;
+  for (const auto& token : tokens) {
+    offset += record->GetField(token).offset;
+    record = &record->GetField(token).type;
+  }
+  expr_ = b_.getInt64(offset);
 }
 
 void CodegenLLVM::visit(Map &map)

diff --git a/src/ast/passes/semantic_analyser.cpp b/src/ast/passes/semantic_analyser.cpp
@@ -1615,17 +1615,33 @@ void SemanticAnalyser::visit(Offsetof &ofof)
     Visit(ofof.expr);
     ofof.record = ofof.expr->type;
   }
+
   resolve_struct_type(ofof.record, ofof.loc);
 
-  if (!ofof.record.IsRecordTy()) {
-    LOG(ERROR, ofof.loc, err_)
-        << "offsetof() 1st argument is not of a record type.";
-  } else if (!bpftrace_.structs.Has(ofof.record.GetName())) {
-    LOG(ERROR, ofof.loc, err_) << "'" << ofof.record << "' does not exist.";
-  } else if (!ofof.record.HasField(ofof.field)) {
-    LOG(ERROR, ofof.loc, err_) << "'" << ofof.record << "' "
-                               << "has no field named "
-                               << "'" << ofof.field << "'";
+  std::string token;
+  std::stringstream ss(ofof.field);
+  std::vector<std::string> tokens;
+  while (std::getline(ss, token, '.')) {
+    tokens.push_back(token);
+  }
+
+  // Check if all sub-fields are present.
+  SizedType record = ofof.record;
+  for (const auto& token : tokens) {
+    if (!record.IsRecordTy()) {
+      LOG(ERROR, ofof.loc, err_) << "'" << record << "' "
+                                 << "is not of a record type.";
+    } else if (!bpftrace_.structs.Has(record.GetName())) {
+      LOG(ERROR, ofof.loc, err_) << "'" << record.GetName()
+                                 << "' does not exist.";
+    } else if (!record.HasField(token)) {
+      LOG(ERROR, ofof.loc, err_) << "'" << record.GetName() << "' "
+                                 << "has no field named "
+                                 << "'" << token << "'";
+    } else {
+      // Get next sub-field
+      record = record.GetField(token).type;
+    }
   }
 }
 

diff --git a/src/parser.yy b/src/parser.yy
@@ -131,7 +131,7 @@ void yyerror(bpftrace::Driver &driver, const char *s);
 
 
 %type <ast::Operator> unary_op compound_op
-%type <std::string> attach_point_def c_definitions ident keyword external_name
+%type <std::string> attach_point_def c_definitions ident keyword external_name struct_field
 
 %type <ast::AttachPoint *> attach_point
 %type <ast::AttachPointList> attach_points
@@ -626,10 +626,15 @@ sizeof_expr:
         |       SIZEOF "(" expr ")"                         { $$ = driver.ctx.make_node<ast::Sizeof>($3, @$); }
                 ;
 
+struct_field:
+                external_name                       { $$ = $1; }
+        |       struct_field DOT external_name      { $$ = $1 + "." + $3; }
+        ;
+
 offsetof_expr:
-                OFFSETOF "(" struct_type "," external_name ")"      { $$ = driver.ctx.make_node<ast::Offsetof>($3, $5, @$); }
-/* For example: offsetof(*curtask, comm) */
-        |       OFFSETOF "(" expr "," external_name ")"             { $$ = driver.ctx.make_node<ast::Offsetof>($3, $5, @$); }
+                OFFSETOF "(" struct_type "," struct_field ")"      { $$ = driver.ctx.make_node<ast::Offsetof>($3, $5, @$); }
+                /* For example: offsetof(*curtask, comm) */
+        |       OFFSETOF "(" expr "," struct_field ")"             { $$ = driver.ctx.make_node<ast::Offsetof>($3, $5, @$); }
                 ;
 
 int:

diff --git a/tests/codegen/call_offsetof.cpp b/tests/codegen/call_offsetof.cpp
@@ -11,6 +11,13 @@ TEST(codegen, call_offsetof)
        NAME);
 }
 
+TEST(codegen, call_offsetof_sub_field)
+{
+  test("struct Foo { struct Bar { int a; } d; }"
+       "BEGIN { @x = offsetof(struct Foo, d.a); exit(); }",
+       NAME);
+}
+
 } // namespace codegen
 } // namespace test
 } // namespace bpftrace
diff --git a/tests/codegen/llvm/call_offsetof_sub_field.ll b/tests/codegen/llvm/call_offsetof_sub_field.ll
@@ -0,0 +1,125 @@
+; ModuleID = 'bpftrace'
+source_filename = "bpftrace"
+target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n32:64-S128"
+target triple = "bpf-pc-linux"
+
+%"struct map_t" = type { i8*, i8*, i8*, i8* }
+%"struct map_t.0" = type { i8*, i8* }
+%"struct map_t.1" = type { i8*, i8*, i8*, i8* }
+%exit_t = type <{ i64, i8 }>
+
+@LICENSE = local_unnamed_addr global [4 x i8] c"GPL\00", section "license"
+@AT_x = dso_local global %"struct map_t" zeroinitializer, section ".maps", !dbg !0
+@ringbuf = dso_local global %"struct map_t.0" zeroinitializer, section ".maps", !dbg !16
+@event_loss_counter = dso_local global %"struct map_t.1" zeroinitializer, section ".maps", !dbg !30
+
+define i64 @BEGIN_1(i8* nocapture readnone %0) local_unnamed_addr section "s_BEGIN_1" !dbg !45 {
+entry:
+  %key = alloca i32, align 4
+  %exit = alloca %exit_t, align 8
+  %"@x_val" = alloca i64, align 8
+  %"@x_key" = alloca i64, align 8
+  %1 = bitcast i64* %"@x_key" to i8*
+  call void @llvm.lifetime.start.p0i8(i64 -1, i8* nonnull %1)
+  store i64 0, i64* %"@x_key", align 8
+  %2 = bitcast i64* %"@x_val" to i8*
+  call void @llvm.lifetime.start.p0i8(i64 -1, i8* nonnull %2)
+  store i64 0, i64* %"@x_val", align 8
+  %update_elem = call i64 inttoptr (i64 2 to i64 (%"struct map_t"*, i64*, i64*, i64)*)(%"struct map_t"* nonnull @AT_x, i64* nonnull %"@x_key", i64* nonnull %"@x_val", i64 0)
+  call void @llvm.lifetime.end.p0i8(i64 -1, i8* nonnull %2)
+  call void @llvm.lifetime.end.p0i8(i64 -1, i8* nonnull %1)
+  %3 = bitcast %exit_t* %exit to i8*
+  call void @llvm.lifetime.start.p0i8(i64 -1, i8* nonnull %3)
+  %4 = getelementptr inbounds %exit_t, %exit_t* %exit, i64 0, i32 0
+  store i64 30000, i64* %4, align 8
+  %5 = getelementptr inbounds %exit_t, %exit_t* %exit, i64 0, i32 1
+  store i8 0, i8* %5, align 8
+  %ringbuf_output = call i64 inttoptr (i64 130 to i64 (%"struct map_t.0"*, %exit_t*, i64, i64)*)(%"struct map_t.0"* nonnull @ringbuf, %exit_t* nonnull %exit, i64 9, i64 0)
+  %ringbuf_loss = icmp slt i64 %ringbuf_output, 0
+  br i1 %ringbuf_loss, label %event_loss_counter, label %counter_merge
+
+event_loss_counter:                               ; preds = %entry
+  %6 = bitcast i32* %key to i8*
+  call void @llvm.lifetime.start.p0i8(i64 -1, i8* nonnull %6)
+  store i32 0, i32* %key, align 4
+  %lookup_elem = call i8* inttoptr (i64 1 to i8* (%"struct map_t.1"*, i32*)*)(%"struct map_t.1"* nonnull @event_loss_counter, i32* nonnull %key)
+  %map_lookup_cond.not = icmp eq i8* %lookup_elem, null
+  br i1 %map_lookup_cond.not, label %lookup_merge, label %lookup_success
+
+counter_merge:                                    ; preds = %lookup_merge, %entry
+  call void @llvm.lifetime.end.p0i8(i64 -1, i8* nonnull %3)
+  ret i64 0
+
+lookup_success:                                   ; preds = %event_loss_counter
+  %7 = bitcast i8* %lookup_elem to i64*
+  %8 = atomicrmw add i64* %7, i64 1 seq_cst, align 8
+  br label %lookup_merge
+
+lookup_merge:                                     ; preds = %event_loss_counter, %lookup_success
+  call void @llvm.lifetime.end.p0i8(i64 -1, i8* nonnull %6)
+  br label %counter_merge
+}
+
+; Function Attrs: argmemonly mustprogress nofree nosync nounwind willreturn
+declare void @llvm.lifetime.start.p0i8(i64 immarg %0, i8* nocapture %1) #0
+
+; Function Attrs: argmemonly mustprogress nofree nosync nounwind willreturn
+declare void @llvm.lifetime.end.p0i8(i64 immarg %0, i8* nocapture %1) #0
+
+attributes #0 = { argmemonly mustprogress nofree nosync nounwind willreturn }
+
+!llvm.dbg.cu = !{!42}
+!llvm.module.flags = !{!44}
+
+!0 = !DIGlobalVariableExpression(var: !1, expr: !DIExpression())
+!1 = distinct !DIGlobalVariable(name: "AT_x", linkageName: "global", scope: !2, file: !2, type: !3, isLocal: false, isDefinition: true)
+!2 = !DIFile(filename: "bpftrace.bpf.o", directory: ".")
+!3 = !DICompositeType(tag: DW_TAG_structure_type, scope: !2, file: !2, size: 256, elements: !4)
+!4 = !{!5, !11, !12, !15}
+!5 = !DIDerivedType(tag: DW_TAG_member, name: "type", scope: !2, file: !2, baseType: !6, size: 64)
+!6 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !7, size: 64)
+!7 = !DICompositeType(tag: DW_TAG_array_type, baseType: !8, size: 32, elements: !9)
+!8 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
+!9 = !{!10}
+!10 = !DISubrange(count: 1, lowerBound: 0)
+!11 = !DIDerivedType(tag: DW_TAG_member, name: "max_entries", scope: !2, file: !2, baseType: !6, size: 64, offset: 64)
+!12 = !DIDerivedType(tag: DW_TAG_member, name: "key", scope: !2, file: !2, baseType: !13, size: 64, offset: 128)
+!13 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !14, size: 64)
+!14 = !DIBasicType(name: "int64", size: 64, encoding: DW_ATE_signed)
+!15 = !DIDerivedType(tag: DW_TAG_member, name: "value", scope: !2, file: !2, baseType: !13, size: 64, offset: 192)
+!16 = !DIGlobalVariableExpression(var: !17, expr: !DIExpression())
+!17 = distinct !DIGlobalVariable(name: "ringbuf", linkageName: "global", scope: !2, file: !2, type: !18, isLocal: false, isDefinition: true)
+!18 = !DICompositeType(tag: DW_TAG_structure_type, scope: !2, file: !2, size: 128, elements: !19)
+!19 = !{!20, !25}
+!20 = !DIDerivedType(tag: DW_TAG_member, name: "type", scope: !2, file: !2, baseType: !21, size: 64)
+!21 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !22, size: 64)
+!22 = !DICompositeType(tag: DW_TAG_array_type, baseType: !8, size: 864, elements: !23)
+!23 = !{!24}
+!24 = !DISubrange(count: 27, lowerBound: 0)
+!25 = !DIDerivedType(tag: DW_TAG_member, name: "max_entries", scope: !2, file: !2, baseType: !26, size: 64, offset: 64)
+!26 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !27, size: 64)
+!27 = !DICompositeType(tag: DW_TAG_array_type, baseType: !8, size: 8388608, elements: !28)
+!28 = !{!29}
+!29 = !DISubrange(count: 262144, lowerBound: 0)
+!30 = !DIGlobalVariableExpression(var: !31, expr: !DIExpression())
+!31 = distinct !DIGlobalVariable(name: "event_loss_counter", linkageName: "global", scope: !2, file: !2, type: !32, isLocal: false, isDefinition: true)
+!32 = !DICompositeType(tag: DW_TAG_structure_type, scope: !2, file: !2, size: 256, elements: !33)
+!33 = !{!34, !11, !39, !15}
+!34 = !DIDerivedType(tag: DW_TAG_member, name: "type", scope: !2, file: !2, baseType: !35, size: 64)
+!35 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !36, size: 64)
+!36 = !DICompositeType(tag: DW_TAG_array_type, baseType: !8, size: 64, elements: !37)
+!37 = !{!38}
+!38 = !DISubrange(count: 2, lowerBound: 0)
+!39 = !DIDerivedType(tag: DW_TAG_member, name: "key", scope: !2, file: !2, baseType: !40, size: 64, offset: 128)
+!40 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !41, size: 64)
+!41 = !DIBasicType(name: "int32", size: 32, encoding: DW_ATE_signed)
+!42 = distinct !DICompileUnit(language: DW_LANG_C, file: !2, producer: "bpftrace", isOptimized: false, runtimeVersion: 0, emissionKind: LineTablesOnly, globals: !43)
+!43 = !{!0, !16, !30}
+!44 = !{i32 2, !"Debug Info Version", i32 3}
+!45 = distinct !DISubprogram(name: "BEGIN_1", linkageName: "BEGIN_1", scope: !2, file: !2, type: !46, flags: DIFlagPrototyped, spFlags: DISPFlagDefinition, unit: !42, retainedNodes: !50)
+!46 = !DISubroutineType(types: !47)
+!47 = !{!14, !48}
+!48 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !49, size: 64)
+!49 = !DIBasicType(name: "int8", size: 8, encoding: DW_ATE_signed)
+!50 = !{!51}
+!51 = !DILocalVariable(name: "ctx", arg: 1, scope: !45, file: !2, type: !48)
diff --git a/tests/parser.cpp b/tests/parser.cpp
@@ -1963,6 +1963,15 @@ TEST(Parser, offsetof_type)
        "  offsetof: \n"
        "   struct Foo\n"
        "   x\n");
+  test("struct Foo { struct Bar { int x; } bar; } "
+       "BEGIN { offsetof(struct Foo, bar.x); }",
+       "struct Foo { struct Bar { int x; } bar; };\n"
+       "\n"
+       "Program\n"
+       " BEGIN\n"
+       "  offsetof: \n"
+       "   struct Foo\n"
+       "   bar.x\n");
 }
 
 TEST(Parser, offsetof_expression)

diff --git a/tests/semantic_analyser.cpp b/tests/semantic_analyser.cpp
@@ -3741,9 +3741,16 @@ TEST(semantic_analyser, call_offsetof)
           long l; \
         } \
         BEGIN { @x = offsetof(struct Ano, a); }");
+  test("struct Foo { struct Bar { int a; } bar; } \
+        BEGIN { @x = offsetof(struct Foo, bar.a); }");
+  test("struct Foo { struct Bar { struct { int a; } anon; } bar; } \
+        BEGIN { @x = offsetof(struct Foo, bar.anon.a); }");
   test("struct Foo { int x; long l; char c; } \
         BEGIN { @x = offsetof(struct Foo, __notexistfield__); }",
        1);
+  test("struct Foo { struct { int a; } bar; } \
+        BEGIN { @x = offsetof(struct Foo, bar.__notexist_subfield__); }",
+       1);
   test("BEGIN { @x = offsetof(__passident__, x); }", 1);
   test("BEGIN { @x = offsetof(struct __notexiststruct__, x); }", 1);
 }