[Lldb-commits] [lldb] [lldb][DWARF] Fix implicit widening bug in DW_OP_plus_uconst (PR #204896)

[via lldb-commits]

================
@@ -722,6 +722,58 @@ TEST(DWARFExpression, DW_OP_plus_uconst) {
   EXPECT_THAT_EXPECTED(Evaluate({DW_OP_const1s, static_cast<uint8_t>(-10),
                                  DW_OP_plus_uconst, 5}),
                        ExpectScalar(static_cast<int32_t>(-5)));
+  const char *yamldata = R"(
+--- !ELF
+FileHeader:
+  Class:   ELFCLASS64
+  Data:    ELFDATA2LSB
+  Type:    ET_EXEC
+  Machine: EM_386
+DWARF:
+  debug_abbrev:
+    - Table:
+        - Code:            0x00000001
+          Tag:             DW_TAG_compile_unit
+          Children:        DW_CHILDREN_yes
+          Attributes:
+            - Attribute:       DW_AT_language
+              Form:            DW_FORM_data2
+        - Code:            0x00000002
+          Tag:             DW_TAG_base_type
+          Children:        DW_CHILDREN_no
+          Attributes:
+            - Attribute:       DW_AT_encoding
+              Form:            DW_FORM_data1
+            - Attribute:       DW_AT_byte_size
+              Form:            DW_FORM_data1
+  debug_info:
+    - Version:         4
----------------
firmiana402 wrote:

Since this test uses `DW_OP_convert` and is specifically checking typed-expression semantics, should this fixture be a DWARF v5 unit (with the corresponding `UnitType` and updated DIE offset)? `DW_OP_convert` is a DWARF v5 opcode, so the current v4 fixture is not a valid typed-expression input.

https://github.com/llvm/llvm-project/pull/204896