[PATCH 3/3] x86/disasm: rework comment handling

Project / Subsystem

binutils / x86/disasm

Date

2026-05-15

Proposer

Jan Beulich <jbeulich@suse.com>

Source type

public_inbox

Consensus

Proposed

Sentiment

/10

Technical tradeoffs

  • Unifying %rip-relative comments with other comments is currently not possible due to the `print_address_func()` hook.
  • Moving the test suite might require adjustments to the testing infrastructure.
  • The new approach might introduce subtle changes in the generated disassembly output.

All attributes

project
binutils
subsystem
x86/disasm
patch_id
discussion_id
916dde10-59ed-4180-86e6-a42677b90c20@suse.com
source_type
public_inbox
title
[PATCH 3/3] x86/disasm: rework comment handling
headline
x86/disasm: Rework comment handling
tldr
Refactors x86 disassembler to handle comments in the same order as operands, improving annotation consistency.
proposer
Jan Beulich <jbeulich@suse.com>
consensus
Proposed
outcome
proposed
sentiment_score
technical_tradeoffs
  • Unifying %rip-relative comments with other comments is currently not possible due to the `print_address_func()` hook.
  • Moving the test suite might require adjustments to the testing infrastructure.
  • The new approach might introduce subtle changes in the generated disassembly output.
series_id
binutils:x86/disasm: rework comment handling
series_role
reply
series_parts
[]
tags
  • x86
  • disassembler
  • comments
  • annotation
  • refactoring
bugzilla_url
date
2026-05-15T00:00:00.000Z

[PATCH 3/3] x86/disasm: rework comment handling

This patch refactors the x86 disassembler’s comment handling to align it with operand handling, allowing comments to be emitted in the same order as operands. It aims to provide a more consistent approach to annotation. The author discusses the limitations of unifying %rip-relative comments due to the print_address_func() hook and proposes moving the test suite to a more generic location.