[llvm-branch-commits] [llvm] [llvm-objdump] Support --symbolize-operand on AArch64 (PR #145009)

[Alexis Engelke via llvm-branch-commits]

================
@@ -0,0 +1,67 @@
+# RUN: yaml2obj %s -o %t
+# RUN: llvm-objdump %t -d --symbolize-operands --no-show-raw-insn --no-leading-addr | \
+# RUN:   FileCheck %s --match-full-lines -DABS_ADRP_VAL=0x6000
+# RUN: llvm-objdump %t -d --symbolize-operands --no-show-raw-insn --no-leading-addr --adjust-vma=0x2000 | \
+# RUN:   FileCheck %s --match-full-lines -DABS_ADRP_VAL=0x8000
+
+## Expect to find the branch labels and global variable name.
+# CHECK:      <_start>:
+# CHECK-NEXT:   ldr x0, <symbol>
+# CHECK-NEXT: <L0>:
+# CHECK-NEXT:   adrp x1, [[ABS_ADRP_VAL]] <symbol+0xff4>
+# CHECK-NEXT:   adr x2, <symbol>
+# CHECK-NEXT:   cmp x1, x2
+# CHECK-NEXT:   b.eq <L1>
+# CHECK-NEXT:   b <L0>
+# CHECK-NEXT: <L1>:
+# CHECK-NEXT:   cbz x2, <L0>
+# CHECK-NEXT:   ret
+
+## Machine code generated with:
----------------
aengelke wrote:

obj2yaml produces loads of unnecessary content (program headers, dynamic sections (dynsym/dynstr/hash/dynamic), its output is twice as long as this test currently is. I can do that, but I don't think it's worth the effort.

https://github.com/llvm/llvm-project/pull/145009