Is it time to take the middle-end stringop/array warnings out of -Wall?

Project / Subsystem

gcc / gcc

Date

2026-05-20

Proposer

Jonathan Wakely <jwakely.gcc@gmail.com>

Source type

public_inbox

Consensus

Under Review

Sentiment

/10

Technical tradeoffs

  • Moving the warnings to -Wextra might cause them to be ignored, leading to bitrot.
  • Removing the warnings entirely might eliminate the incentive to improve the compiler's optimization capabilities.
  • Keeping the warnings in -Wall can lead to user frustration due to false positives.

All attributes

project
gcc
subsystem
gcc
patch_id
discussion_id
CAH6eHdRfEVsGjbJ__kU5AR9BfVdLQZ_iGGDZa=yQnT4fgG2P6w@mail.gmail.com
source_type
public_inbox
title
Is it time to take the middle-end stringop/array warnings out of -Wall?
headline
Is it time to take the middle-end stringop/array warnings out of -Wall?
tldr
Jonathan Wakely argues for removing middle-end stringop/array warnings from -Wall due to high false positive rates and lack of actionable information.
proposer
Jonathan Wakely <jwakely.gcc@gmail.com>
consensus
Under Review
outcome
proposed
sentiment_score
technical_tradeoffs
  • Moving the warnings to -Wextra might cause them to be ignored, leading to bitrot.
  • Removing the warnings entirely might eliminate the incentive to improve the compiler's optimization capabilities.
  • Keeping the warnings in -Wall can lead to user frustration due to false positives.
series_id
series_role
standalone
series_parts
[]
tags
  • warnings
  • -Wall
  • middle-end
  • stringop
  • array
bugzilla_url
date
2026-05-20T00:00:00.000Z

Is it time to take the middle-end stringop/array warnings out of -Wall?

Jonathan Wakely argues that the false positive rate for middle-end stringop/array warnings in GCC has been unacceptably high for years. He claims the warnings often lie, are not actionable, and make GCC look bad. He suggests that using these warnings as a proxy for finding missed optimization opportunities is a terrible policy.