From: sos@prospect.com.ru (Sergey Okhapkin) Subject: RE: Sergey's latest cygwin.dll and pdksh 16 Jun 1997 07:06:22 -0700 Sender: mail@cygnus.com Approved: cygnus.gnu-win32@cygnus.com Distribution: cygnus Message-ID: <01BC7A72.D9971970.cygnus.gnu-win32@gater.krystalbank.msk.ru> Original-To: "gnu-win32@cygnus.com" , "'cgf@bbc.com'" Encoding: 38 TEXT Original-Sender: owner-gnu-win32@cygnus.com Chris Faylor wrote: > >It's easy for me to fix this bug, if anyone will answer my question - > >should sigsuspend() (and the others blocking calls) terminate when > >receiving blocked or uncatched signal? > > Sergey, > If I run the code below on a variety of UNIX systems and tell you how it > works, will that answer your question? > > #include > main() > { > sigset_t nada = 0L; > if (fork() == 0) { > sleep(2); > exit(0); > } > > sigsuspend(&nada); > puts("returned"); > exit(0); > } > > FYI, I ran this on one system here (Digital UNIX 3.2C) and it blocked > forever. If this isn't quite right, then, if it is possible, could you > modify this (or throw it away and start from scratch) to show the answer > to your question? I've tried something like it on linux with the same result - blocked, ignored or uncatched signals with ignore as a default disposition, does not terminates wait calls. So, I have no ideas now why zsh's SIGCHLD behavior differs on unixes and on cygwin... -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help".