https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88576
--- Comment #21 from rguenther at suse dot de <rguenther at suse dot de> --- > Am 20.03.2026 um 15:58 schrieb bugdal at aerifal dot cx > <[email protected]>: > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88576 > > --- Comment #20 from Rich Felker <bugdal at aerifal dot cx> --- > Is the actual bug here fixed yet or planned to be fixed? Speculation about > better optimizations around otherwise-pure/const functions and errno is nice I > guess, but the root issue here is an existing wrong-optimization that breaks > real code (and that would still be formally incorrect even if it didn't). > > We can worry about improving situations that can be improved (if there are > any) > after fixing the actual bug. The bug is not fixed. But the idea is to at least have a plan to void regressions when fixing it. Those regressions (which need to be re-confirmed!) are the whole reason why it is not fixed.
