https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77676
Martin Sebor <msebor at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |ASSIGNED Assignee|unassigned at gcc dot gnu.org |msebor at gcc dot gnu.org --- Comment #7 from Martin Sebor <msebor at gcc dot gnu.org> --- The optimization is meant to be independent of the warning. If it isn't it's a bug in the code. I successfully tested the pass on powerpc64le just last week, but I did not test it on any ILP32 targets. Let me look into this. In the meantime, the optimization can be disabled by initializing the -fprintf-return-value option to 0, without disabling anything else.