On Tue, Jul 09, 2024 at 06:38:42PM +0200, Alejandro Colomar wrote: > On Mon, Jul 08, 2024 at 04:03:40AM GMT, наб wrote: > > Applies to Linux and NetBSD. > Is this non-standard behavior? In the case of setitimer(): "obviously yes" because the interface is not part of the standard. In the case of alarm(): the standard doesn't mention anything here. But alarm() is allowed to interact with a lot of stuff. The only description of SI_TIMER (which repeats twice) is The signal was generated by the expiration of a timer set by timer_settime( ). > Should we mention something about it? All systems seem to agree that "when you get a timer signal, it'll be due to SI_TIMER", so "meh" imo. Best,
Attachment:
signature.asc
Description: PGP signature
