[Lldb-commits] [PATCH] D107213: Disassemble AArch64 pc-relative address calculations & symbolicate

[Jason Molenda via Phabricator via lldb-commits]

jasonmolenda created this revision.
jasonmolenda added a project: LLDB.
Herald added subscribers: omjavaid, danielkiss, JDevlieghere, kristof.beyls.
jasonmolenda requested review of this revision.

lldb's disassembler for arm64 / aarch64 doesn't recognize the ADRP+ADD instruction pair often used to calculate pc-relative addresses, and symbolicate what is being pointed to, making it hard to understand the assembly; doing the computation manually is annoying.  This patch adds a bit of arm64-specific knowledge to the disassembly symbolicator to remember the state and compute the address being specified when the target is aarch64.

I'm fixing the other thing that I dislike the most about lldb's arm64 disassembly in the llvm aarch64 instruction printer, via https://reviews.llvm.org/D107196 .


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D107213

Files:
  lldb/source/Plugins/Disassembler/LLVMC/DisassemblerLLVMC.cpp
  lldb/source/Plugins/Disassembler/LLVMC/DisassemblerLLVMC.h
  lldb/test/API/functionalities/disassemble/aarch64-adrp-add/TestAArch64AdrpAdd.py
  lldb/test/API/functionalities/disassemble/aarch64-adrp-add/a.out.yaml
  lldb/test/API/functionalities/disassemble/aarch64-adrp-add/main.c

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D107213.363274.patch
Type: text/x-patch
Size: 17635 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/lldb-commits/attachments/20210731/5804ce97/attachment-0001.bin>