[BPF] Visit nested map array during BTF generation (#150608)

Fixes missing inner map struct type definitions [^1]. We should visit
the type of nested array of maps like we do for global maps. This patch
adds a boolean to convey the information to visitTypeEntry and
visitDerivedType that the pointee is a map definition and should be
treated as such.

It ressembles and works with commit 0d21c956a5c1 ("[BPF] Handle nested
wrapper structs in BPF map definition traversal (#144097)") which
focused on directly nested wrapper structs.

Before that patch, this ARRAY_OF_MAPS definition would lead to the BTF
information include the 'missing_type' as "FWD 'missing_type'
fwd_kind=struct":

	struct missing_type { uint64_t foo; };
	struct {
		__uint(type, BPF_MAP_TYPE_ARRAY_OF_MAPS);
		[...]
		__array(
			values, struct {
				[...]
				__type(value, struct missing_type);
			});
	} map SEC(".maps");

Which lead to errors while trying to load the map:

libbpf: map 'outer_map.inner': can't determine value size for type [N]:
-22.

To solve this issue, users had to use the struct in a dummy variable or
in a dummy function for the BTF to be generated correctly [^2].

[^1]: https://lore.kernel.org/netdev/aH_cGvgC20iD8qs9@gmail.com/T/#u
[^2]:
https://github.com/cilium/ebpf/discussions/1658#discussioncomment-12491339

---------

Signed-off-by: Mahe Tardy <mahe.tardy@gmail.com>
Co-authored-by: Eduard Zingerman <eddyz87@gmail.com>

4 files changed, 136 insertions, 116 deletions

diff --git a/llvm/lib/Target/BPF/BTFDebug.cpp b/llvm/lib/Target/BPF/BTFDebug.cpp
index a87b9a2..bed6bc9 100644
--- a/llvm/lib/Target/BPF/BTFDebug.cpp
+++ b/llvm/lib/Target/BPF/BTFDebug.cpp
@@ -957,47 +957,47 @@ void BTFDebug::visitMapDefType(const DIType *Ty, uint32_t &TypeId) {
     return;
   }
 
-  // MapDef type may be a struct type or a non-pointer derived type
-  const DIType *OrigTy = Ty;
-  while (auto *DTy = dyn_cast<DIDerivedType>(Ty)) {
-    auto Tag = DTy->getTag();
-    if (Tag != dwarf::DW_TAG_typedef && Tag != dwarf::DW_TAG_const_type &&
-        Tag != dwarf::DW_TAG_volatile_type &&
-        Tag != dwarf::DW_TAG_restrict_type)
-      break;
-    Ty = DTy->getBaseType();
-  }
-
-  const auto *CTy = dyn_cast<DICompositeType>(Ty);
-  if (!CTy)
-    return;
-
-  auto Tag = CTy->getTag();
-  if (Tag != dwarf::DW_TAG_structure_type || CTy->isForwardDecl())
-    return;
-
-  // Visit all struct members to ensure their types are visited.
-  const DINodeArray Elements = CTy->getElements();
-  for (const auto *Element : Elements) {
-    const auto *MemberType = cast<DIDerivedType>(Element);
-    const DIType *MemberBaseType = MemberType->getBaseType();
-
-    // If the member is a composite type, that may indicate the currently
-    // visited composite type is a wrapper, and the member represents the
-    // actual map definition.
-    // In that case, visit the member with `visitMapDefType` instead of
-    // `visitTypeEntry`, treating it specifically as a map definition rather
-    // than as a regular composite type.
-    const auto *MemberCTy = dyn_cast<DICompositeType>(MemberBaseType);
-    if (MemberCTy) {
-      visitMapDefType(MemberBaseType, TypeId);
-    } else {
-      visitTypeEntry(MemberBaseType);
+  uint32_t TmpId;
+  switch (Ty->getTag()) {
+  case dwarf::DW_TAG_typedef:
+  case dwarf::DW_TAG_const_type:
+  case dwarf::DW_TAG_volatile_type:
+  case dwarf::DW_TAG_restrict_type:
+  case dwarf::DW_TAG_pointer_type:
+    visitMapDefType(dyn_cast<DIDerivedType>(Ty)->getBaseType(), TmpId);
+    break;
+  case dwarf::DW_TAG_array_type:
+    // Visit nested map array and jump to the element type
+    visitMapDefType(dyn_cast<DICompositeType>(Ty)->getBaseType(), TmpId);
+    break;
+  case dwarf::DW_TAG_structure_type: {
+    // Visit all struct members to ensure their types are visited.
+    const auto *CTy = cast<DICompositeType>(Ty);
+    const DINodeArray Elements = CTy->getElements();
+    for (const auto *Element : Elements) {
+      const auto *MemberType = cast<DIDerivedType>(Element);
+      const DIType *MemberBaseType = MemberType->getBaseType();
+      // If the member is a composite type, that may indicate the currently
+      // visited composite type is a wrapper, and the member represents the
+      // actual map definition.
+      // In that case, visit the member with `visitMapDefType` instead of
+      // `visitTypeEntry`, treating it specifically as a map definition rather
+      // than as a regular composite type.
+      const auto *MemberCTy = dyn_cast<DICompositeType>(MemberBaseType);
+      if (MemberCTy) {
+        visitMapDefType(MemberBaseType, TmpId);
+      } else {
+        visitTypeEntry(MemberBaseType);
+      }
     }
+    break;
+  }
+  default:
+    break;
   }
 
   // Visit this type, struct or a const/typedef/volatile/restrict type
-  visitTypeEntry(OrigTy, TypeId, false, false);
+  visitTypeEntry(Ty, TypeId, false, false);
 }
 
 /// Read file contents from the actual file or from the source
diff --git a/llvm/test/CodeGen/BPF/BTF/map-def-2.ll b/llvm/test/CodeGen/BPF/BTF/map-def-2.ll
index 5f971ec..d4c836f 100644
--- a/llvm/test/CodeGen/BPF/BTF/map-def-2.ll
+++ b/llvm/test/CodeGen/BPF/BTF/map-def-2.ll
@@ -1,5 +1,6 @@
-; RUN: llc -mtriple=bpfel -filetype=asm -o - %s | FileCheck -check-prefixes=CHECK %s
-; RUN: llc -mtriple=bpfeb -filetype=asm -o - %s | FileCheck -check-prefixes=CHECK %s
+; RUN: llc -mtriple=bpfel -mcpu=v3 -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 code:
 ;   struct key_type {
@@ -18,51 +19,17 @@
 
 @hash_map = dso_local local_unnamed_addr global %struct.map_type zeroinitializer, section ".maps", align 8, !dbg !0
 
-; CHECK:             .long   0                               # BTF_KIND_PTR(id = 1)
-; CHECK-NEXT:        .long   33554432                        # 0x2000000
-; CHECK-NEXT:        .long   2
-; CHECK-NEXT:        .long   1                               # BTF_KIND_STRUCT(id = 2)
-; CHECK-NEXT:        .long   67108865                        # 0x4000001
-; CHECK-NEXT:        .long   4
-; CHECK-NEXT:        .long   10
-; CHECK-NEXT:        .long   3
-; CHECK-NEXT:        .long   0                               # 0x0
-; CHECK-NEXT:        .long   13                              # BTF_KIND_INT(id = 3)
-; CHECK-NEXT:        .long   16777216                        # 0x1000000
-; CHECK-NEXT:        .long   4
-; CHECK-NEXT:        .long   16777248                        # 0x1000020
-; CHECK-NEXT:        .long   17                              # BTF_KIND_TYPEDEF(id = 4)
-; CHECK-NEXT:        .long   134217728                       # 0x8000000
-; CHECK-NEXT:        .long   5
-; CHECK-NEXT:        .long   28                              # BTF_KIND_TYPEDEF(id = 5)
-; CHECK-NEXT:        .long   134217728                       # 0x8000000
-; CHECK-NEXT:        .long   6
-; CHECK-NEXT:        .long   38                              # BTF_KIND_STRUCT(id = 6)
-; CHECK-NEXT:        .long   67108865                        # 0x4000001
-; CHECK-NEXT:        .long   8
-; CHECK-NEXT:        .long   47
-; CHECK-NEXT:        .long   1
-; CHECK-NEXT:        .long   0                               # 0x0
-; CHECK-NEXT:        .long   51                              # BTF_KIND_VAR(id = 7)
-; CHECK-NEXT:        .long   234881024                       # 0xe000000
-; CHECK-NEXT:        .long   4
-; CHECK-NEXT:        .long   1
-; CHECK-NEXT:        .long   60                              # BTF_KIND_DATASEC(id = 8)
-; CHECK-NEXT:        .long   251658241                       # 0xf000001
-; CHECK-NEXT:        .long   0
-; CHECK-NEXT:        .long   7
-; CHECK-NEXT:        .long   hash_map
-; CHECK-NEXT:        .long   8
-
-; CHECK:             .ascii  "key_type"                      # string offset=1
-; CHECK:             .ascii  "a1"                            # string offset=10
-; CHECK:             .ascii  "int"                           # string offset=13
-; CHECK:             .ascii  "__map_type"                    # string offset=17
-; CHECK:             .ascii  "_map_type"                     # string offset=28
-; CHECK:             .ascii  "map_type"                      # string offset=38
-; CHECK:             .ascii  "key"                           # string offset=47
-; CHECK:             .ascii  "hash_map"                      # string offset=51
-; CHECK:             .ascii  ".maps"                         # string offset=60
+; CHECK-BTF: [1] PTR '(anon)' type_id=2
+; CHECK-BTF-NEXT: [2] STRUCT 'key_type' size=4 vlen=1
+; CHECK-BTF-NEXT:         'a1' type_id=3 bits_offset=0
+; CHECK-BTF-NEXT: [3] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED
+; CHECK-BTF-NEXT: [4] STRUCT 'map_type' size=8 vlen=1
+; CHECK-BTF-NEXT:         'key' type_id=1 bits_offset=0
+; CHECK-BTF-NEXT: [5] TYPEDEF '_map_type' type_id=4
+; CHECK-BTF-NEXT: [6] TYPEDEF '__map_type' type_id=5
+; CHECK-BTF-NEXT: [7] VAR 'hash_map' type_id=6, linkage=global
+; CHECK-BTF-NEXT: [8] DATASEC '.maps' size=0 vlen=1
+; CHECK-BTF-NEXT:         type_id=7 offset=0 size=8
 
 !llvm.dbg.cu = !{!2}
 !llvm.module.flags = !{!16, !17, !18}
diff --git a/llvm/test/CodeGen/BPF/BTF/map-def-3.ll b/llvm/test/CodeGen/BPF/BTF/map-def-3.ll
index 6aa8af9..1d95f03 100644
--- a/llvm/test/CodeGen/BPF/BTF/map-def-3.ll
+++ b/llvm/test/CodeGen/BPF/BTF/map-def-3.ll
@@ -1,5 +1,6 @@
-; RUN: llc -mtriple=bpfel -filetype=asm -o - %s | FileCheck -check-prefixes=CHECK %s
-; RUN: llc -mtriple=bpfeb -filetype=asm -o - %s | FileCheck -check-prefixes=CHECK %s
+; RUN: llc -mtriple=bpfel -mcpu=v3 -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 code:
 ;   struct key_type {
@@ -13,36 +14,13 @@
 
 @hash_map = dso_local local_unnamed_addr constant %struct.key_type zeroinitializer, section ".maps", align 4, !dbg !0
 
-; CHECK:             .long   1                               # BTF_KIND_INT(id = 1)
-; CHECK-NEXT:        .long   16777216                        # 0x1000000
-; CHECK-NEXT:        .long   4
-; CHECK-NEXT:        .long   16777248                        # 0x1000020
-; CHECK-NEXT:        .long   0                               # BTF_KIND_CONST(id = 2)
-; CHECK-NEXT:        .long   167772160                       # 0xa000000
-; CHECK-NEXT:        .long   3
-; CHECK-NEXT:        .long   5                               # BTF_KIND_STRUCT(id = 3)
-; CHECK-NEXT:        .long   67108865                        # 0x4000001
-; CHECK-NEXT:        .long   4
-; CHECK-NEXT:        .long   14
-; CHECK-NEXT:        .long   1
-; CHECK-NEXT:        .long   0                               # 0x0
-; CHECK-NEXT:        .long   17                              # BTF_KIND_VAR(id = 4)
-; CHECK-NEXT:        .long   234881024                       # 0xe000000
-; CHECK-NEXT:        .long   2
-; CHECK-NEXT:        .long   1
-; CHECK-NEXT:        .long   26                              # BTF_KIND_DATASEC(id = 5)
-; CHECK-NEXT:        .long   251658241                       # 0xf000001
-; CHECK-NEXT:        .long   0
-; CHECK-NEXT:        .long   4
-; CHECK-NEXT:        .long   hash_map
-; CHECK-NEXT:        .long   4
-
-; CHECK:             .ascii  "int"                           # string offset=1
-; CHECK:             .ascii  "key_type"                      # string offset=5
-; CHECK:             .ascii  "a1"                            # string offset=14
-; CHECK:             .ascii  "hash_map"                      # string offset=17
-; CHECK:             .ascii  ".maps"                         # string offset=26
-
+; CHECK-BTF: [1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED
+; CHECK-BTF-NEXT: [2] STRUCT 'key_type' size=4 vlen=1
+; CHECK-BTF-NEXT:         'a1' type_id=1 bits_offset=0
+; CHECK-BTF-NEXT: [3] CONST '(anon)' type_id=2
+; CHECK-BTF-NEXT: [4] VAR 'hash_map' type_id=3, linkage=global
+; CHECK-BTF-NEXT: [5] DATASEC '.maps' size=0 vlen=1
+; CHECK-BTF-NEXT:         type_id=4 offset=0 size=4
 
 !llvm.dbg.cu = !{!2}
 !llvm.module.flags = !{!11, !12, !13}
diff --git a/llvm/test/CodeGen/BPF/BTF/map-def-nested-array.ll b/llvm/test/CodeGen/BPF/BTF/map-def-nested-array.ll
new file mode 100644
index 0000000..fc95daf
--- /dev/null
+++ b/llvm/test/CodeGen/BPF/BTF/map-def-nested-array.ll
@@ -0,0 +1,75 @@
+; RUN: llc -mtriple=bpfel -mcpu=v3 -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-SHORT %s
+; RUN: %python %p/print_btf.py %t2 | FileCheck -check-prefixes=CHECK-BTF %s
+; Source:
+;  struct nested_value_type {
+;  	int a1;
+;  };
+;  struct map_type {
+;  	struct {
+;  		struct nested_value_type *value;
+;  	} *values[];
+;  };
+; Compilation flags:
+;   clang -target bpf -g -O2 -S -emit-llvm prog.c
+
+; ModuleID = 'prog.c'
+source_filename = "prog.c"
+target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n32:64-S128"
+target triple = "bpf"
+
+%struct.map_type = type { [0 x ptr] }
+
+@array_of_maps = dso_local local_unnamed_addr global %struct.map_type zeroinitializer, section ".maps", align 8, !dbg !0
+
+; We expect no forward declarations.
+;
+; CHECK-BTF-SHORT-NOT: FWD
+
+; Assert the whole BTF.
+;
+; CHECK-BTF: [1] PTR '(anon)' type_id=2
+; CHECK-BTF-NEXT: [2] STRUCT 'nested_value_type' size=4 vlen=1
+; CHECK-BTF-NEXT:         'a1' type_id=3 bits_offset=0
+; CHECK-BTF-NEXT: [3] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED
+; CHECK-BTF-NEXT: [4] STRUCT '(anon)' size=8 vlen=1
+; CHECK-BTF-NEXT:         'value' type_id=1 bits_offset=0
+; CHECK-BTF-NEXT: [5] PTR '(anon)' type_id=4
+; CHECK-BTF-NEXT: [6] ARRAY '(anon)' type_id=5 index_type_id=7 nr_elems=0
+; CHECK-BTF-NEXT: [7] INT '__ARRAY_SIZE_TYPE__' size=4 bits_offset=0 nr_bits=32 encoding=(none)
+; CHECK-BTF-NEXT: [8] STRUCT 'map_type' size=0 vlen=1
+; CHECK-BTF-NEXT:         'values' type_id=6 bits_offset=0
+; CHECK-BTF-NEXT: [9] VAR 'array_of_maps' type_id=8, linkage=global
+; CHECK-BTF-NEXT: [10] DATASEC '.maps' size=0 vlen=1
+; CHECK-BTF-NEXT:         type_id=9 offset=0 size=0
+
+!llvm.dbg.cu = !{!2}
+!llvm.module.flags = !{!20, !21, !22, !23}
+!llvm.ident = !{!24}
+
+!0 = !DIGlobalVariableExpression(var: !1, expr: !DIExpression())
+!1 = distinct !DIGlobalVariable(name: "array_of_maps", scope: !2, file: !3, line: 9, type: !5, isLocal: false, isDefinition: true)
+!2 = distinct !DICompileUnit(language: DW_LANG_C11, file: !3, producer: "clang version 22.0.0git (git@github.com:llvm/llvm-project.git ed93eaa421b714028b85cc887d80c45991d7207f)", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug, globals: !4, splitDebugInlining: false, nameTableKind: None)
+!3 = !DIFile(filename: "prog.c", directory: "/home/mtardy/llvm-bug-repro", checksumkind: CSK_MD5, checksum: "9381d9e83e9c0b235a14704224815e96")
+!4 = !{!0}
+!5 = distinct !DICompositeType(tag: DW_TAG_structure_type, name: "map_type", file: !3, line: 4, elements: !6)
+!6 = !{!7}
+!7 = !DIDerivedType(tag: DW_TAG_member, name: "values", scope: !5, file: !3, line: 7, baseType: !8)
+!8 = !DICompositeType(tag: DW_TAG_array_type, baseType: !9, elements: !18)
+!9 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !10, size: 64)
+!10 = distinct !DICompositeType(tag: DW_TAG_structure_type, scope: !5, file: !3, line: 5, size: 64, elements: !11)
+!11 = !{!12}
+!12 = !DIDerivedType(tag: DW_TAG_member, name: "value", scope: !10, file: !3, line: 6, baseType: !13, size: 64)
+!13 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !14, size: 64)
+!14 = distinct !DICompositeType(tag: DW_TAG_structure_type, name: "nested_value_type", file: !3, line: 1, size: 32, elements: !15)
+!15 = !{!16}
+!16 = !DIDerivedType(tag: DW_TAG_member, name: "a1", scope: !14, file: !3, line: 2, baseType: !17, size: 32)
+!17 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
+!18 = !{!19}
+!19 = !DISubrange(count: -1)
+!20 = !{i32 7, !"Dwarf Version", i32 5}
+!21 = !{i32 2, !"Debug Info Version", i32 3}
+!22 = !{i32 1, !"wchar_size", i32 4}
+!23 = !{i32 7, !"frame-pointer", i32 2}
+!24 = !{!"clang version 22.0.0git (git@github.com:llvm/llvm-project.git ed93eaa421b714028b85cc887d80c45991d7207f)"}