Terry Lambert wrote: > "Daniel C. Sobral" wrote: > > > I can definitely reproduce this here and also fairly angry about it. > > > In addition to what you mentioned, fstat() gives an incorrect st_size > > > result now and it's likely that non-NOTE_LOWAT low watermarks are > > > firing too early as well. > > > > > > Ugly test program @ http://people.freebsd.org/~tjr/kq.c > > > > If I ensure the write only happens after the child is waiting on kq, the > > bug does not show up for me. > > > So that's the kqueue bug, right? That an event is not generated > for pending input at the time that the event is registered?
Folks, *PLEASE WATCH THE CC: LIST*! re@ does not need to be cc'ed on the speculation. They've been made aware of the issue, and if a conclusion is reached then they need to know that too. But overflowing re@ with email is not a good way to get the release polished off. Cheers, -Peter -- Peter Wemm - [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
