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>

CHANGELOG.md

@@ -10,6 +10,8 @@ and this project adheres to
 
 #### Breaking Changes
 #### Added
+- `offsetof()` now supports sub fields e.g. `offsetof(struct Foo, bar.a.b);`
+  - [#3761](https://github.com/bpftrace/bpftrace/pull/3761)
 - Pointers may now be used in if conditions, tenary conditions and as operands in logical AND and OR expressions
   - [#3656](https://github.com/bpftrace/bpftrace/pull/3656)
 #### Changed

man/adoc/bpftrace.adoc

@@ -1789,14 +1789,32 @@ You can also pass the address type (e.g. AF_INET) explicitly as the first parame
 === offsetof
 
 .variants
-* `offsetof(STRUCT, FIELD)`
-* `offsetof(EXPRESSION, FIELD)`
+* `offsetof(STRUCT, FIELD[.SUBFIELD])`
+* `offsetof(EXPRESSION, FIELD[.SUBFIELD])`
 
 *compile time*
 
 Returns offset of the field offset bytes in struct.
 Similar to kernel `offsetof` operator.
-Note that subfields are not yet supported.
+
+Support any number of sub field levels, for example:
+
+----
+struct Foo {
+  struct {
+    struct {
+      struct {
+        int d;
+      } c;
+    } b;
+  } a;
+}
+BEGIN {
+  @x = offsetof(struct Foo, a.b.c.d);
+  exit();
+}
+----
+
 
 [#functions-override]
 === override

src/ast/ast.cpp

@@ -105,12 +105,16 @@ Sizeof::Sizeof(Expression *expr, location loc) : Expression(loc), expr(expr)
 {
 }
 
-Offsetof::Offsetof(SizedType record, std::string &field, location loc)
+Offsetof::Offsetof(SizedType record,
+                   std::vector<std::string> &field,
+                   location loc)
     : Expression(loc), record(record), field(field)
 {
 }
 
-Offsetof::Offsetof(Expression *expr, std::string &field, location loc)
+Offsetof::Offsetof(Expression *expr,
+                   std::vector<std::string> &field,
+                   location loc)
     : Expression(loc), expr(expr), field(field)
 {
 }

src/ast/ast.h

@@ -219,12 +219,12 @@ class Offsetof : public Expression {
 public:
   DEFINE_ACCEPT
 
-  Offsetof(SizedType record, std::string &field, location loc);
-  Offsetof(Expression *expr, std::string &field, location loc);
+  Offsetof(SizedType record, std::vector<std::string> &field, location loc);
+  Offsetof(Expression *expr, std::vector<std::string> &field, location loc);
 
   SizedType record;
   Expression *expr = nullptr;
-  std::string field;
+  std::vector<std::string> field;
 
 private:
   Offsetof(const Offsetof &other) = default;

src/ast/passes/codegen_llvm.cpp

@@ -1522,8 +1522,13 @@ 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;
+  const SizedType *record = &ofof.record;
+  for (const auto &field : ofof.field) {
+    offset += record->GetField(field).offset;
+    record = &record->GetField(field).type;
+  }
+  expr_ = b_.getInt64(offset);
 }
 
 void CodegenLLVM::visit(Map &map)

src/ast/passes/printer.cpp

@@ -141,7 +141,9 @@ void Printer::visit(Offsetof &ofof)
     out_ << indentParam << ofof.record << std::endl;
   }
 
-  out_ << indentParam << ofof.field << std::endl;
+  for (const auto &field : ofof.field) {
+    out_ << indentParam << field << std::endl;
+  }
   --depth_;
 }
 

src/ast/passes/semantic_analyser.cpp

@@ -1615,17 +1615,26 @@ 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 << "'";
+  // Check if all sub-fields are present.
+  SizedType record = ofof.record;
+  for (const auto &field : ofof.field) {
+    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(field)) {
+      LOG(ERROR, ofof.loc, err_) << "'" << record.GetName() << "' "
+                                 << "has no field named "
+                                 << "'" << field << "'";
+    } else {
+      // Get next sub-field
+      record = record.GetField(field).type;
+    }
   }
 }
 

src/parser.yy

@@ -132,6 +132,7 @@ 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::vector<std::string>> struct_field
 
 %type <ast::AttachPoint *> attach_point
 %type <ast::AttachPointList> attach_points
@@ -627,9 +628,9 @@ sizeof_expr:
                 ;
 
 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:
@@ -662,6 +663,11 @@ ident:
         |       STACK_MODE    { $$ = $1; }
                 ;
 
+struct_field:
+                external_name                       { $$.push_back($1); }
+        |       struct_field DOT external_name      { $$ = std::move($1); $$.push_back($3); }
+        ;
+
 external_name:
                 keyword       { $$ = $1; }
         |       ident         { $$ = $1; }

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

tests/codegen/llvm/call_offsetof_sub_field.ll

@@ -0,0 +1,130 @@
+; 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 { ptr, ptr, ptr, ptr }
+%"struct map_t.0" = type { ptr, ptr }
+%"struct map_t.1" = type { ptr, ptr, ptr, ptr }
+%exit_t = type <{ i64, i8 }>
+
+@LICENSE = 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
+
+; Function Attrs: nounwind
+declare i64 @llvm.bpf.pseudo(i64 %0, i64 %1) #0
+
+define i64 @BEGIN_1(ptr %0) 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
+  call void @llvm.lifetime.start.p0(i64 -1, ptr %"@x_key")
+  store i64 0, ptr %"@x_key", align 8
+  call void @llvm.lifetime.start.p0(i64 -1, ptr %"@x_val")
+  store i64 0, ptr %"@x_val", align 8
+  %update_elem = call i64 inttoptr (i64 2 to ptr)(ptr @AT_x, ptr %"@x_key", ptr %"@x_val", i64 0)
+  call void @llvm.lifetime.end.p0(i64 -1, ptr %"@x_val")
+  call void @llvm.lifetime.end.p0(i64 -1, ptr %"@x_key")
+  call void @llvm.lifetime.start.p0(i64 -1, ptr %exit)
+  %1 = getelementptr %exit_t, ptr %exit, i64 0, i32 0
+  store i64 30000, ptr %1, align 8
+  %2 = getelementptr %exit_t, ptr %exit, i64 0, i32 1
+  store i8 0, ptr %2, align 1
+  %ringbuf_output = call i64 inttoptr (i64 130 to ptr)(ptr @ringbuf, ptr %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
+  call void @llvm.lifetime.start.p0(i64 -1, ptr %key)
+  store i32 0, ptr %key, align 4
+  %lookup_elem = call ptr inttoptr (i64 1 to ptr)(ptr @event_loss_counter, ptr %key)
+  %map_lookup_cond = icmp ne ptr %lookup_elem, null
+  br i1 %map_lookup_cond, label %lookup_success, label %lookup_failure
+
+counter_merge:                                    ; preds = %lookup_merge, %entry
+  call void @llvm.lifetime.end.p0(i64 -1, ptr %exit)
+  ret i64 0
+
+lookup_success:                                   ; preds = %event_loss_counter
+  %3 = atomicrmw add ptr %lookup_elem, i64 1 seq_cst, align 8
+  br label %lookup_merge
+
+lookup_failure:                                   ; preds = %event_loss_counter
+  br label %lookup_merge
+
+lookup_merge:                                     ; preds = %lookup_failure, %lookup_success
+  call void @llvm.lifetime.end.p0(i64 -1, ptr %key)
+  br label %counter_merge
+
+deadcode:                                         ; No predecessors!
+  ret i64 0
+}
+
+; Function Attrs: nocallback nofree nosync nounwind willreturn memory(argmem: readwrite)
+declare void @llvm.lifetime.start.p0(i64 immarg %0, ptr nocapture %1) #1
+
+; Function Attrs: nocallback nofree nosync nounwind willreturn memory(argmem: readwrite)
+declare void @llvm.lifetime.end.p0(i64 immarg %0, ptr nocapture %1) #1
+
+attributes #0 = { nounwind }
+attributes #1 = { nocallback nofree nosync nounwind willreturn memory(argmem: readwrite) }
+
+!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)

tests/parser.cpp

@@ -1963,6 +1963,23 @@ 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\n"
+       "   x\n");
+  test_parse_failure("struct Foo { struct Bar { int x; } *bar; } "
+                     "BEGIN { offsetof(struct Foo, bar->x); }",
+                     R"(
+stdin:1:74-79: ERROR: syntax error, unexpected ->, expecting ) or .
+struct Foo { struct Bar { int x; } *bar; } BEGIN { offsetof(struct Foo, bar->x); }
+                                                                         ~~~~~
+)");
 }
 
 TEST(Parser, offsetof_expression)

tests/runtime/builtin

@@ -174,8 +174,8 @@ EXPECT_REGEX ^size=
 REQUIRES_FEATURE btf
 
 NAME offsetof
-PROG struct Foo { int x; long l; char c; } BEGIN { printf("%ld\n", offsetof(struct Foo, x)); exit(); }
-EXPECT_REGEX ^0$
+PROG struct Foo { int x; struct Bar { int x; } bar; } BEGIN { printf("%ld %ld\n", offsetof(struct Foo, x), offsetof(struct Foo, bar.x)); exit(); }
+EXPECT_REGEX ^0 4$
 
 NAME print args in fentry
 PROG fentry:vfs_open { print(args); exit(); }