This is the mail archive of the
cygwin-developers@cygwin.com
mailing list for the Cygwin project.
RE: Quick testfeedback...
- To: <cygwin-developers at cygwin dot com>
- Subject: RE: Quick testfeedback...
- From: "Robert Collins" <robert dot collins at itdomain dot com dot au>
- Date: Wed, 12 Sep 2001 11:52:50 +1000
> -----Original Message-----
> From: Christopher Faylor [mailto:cgf@redhat.com]
> Sent: Wednesday, September 12, 2001 12:04 PM
> To: cygwin-developers@cygwin.com
> Subject: Re: Quick testfeedback...
>
>
> On Wed, Sep 12, 2001 at 11:43:31AM +1000, Robert Collins wrote:
> >This is broadcast.c:
>
> So, if I read that correctly, sleep can be called from a thread.
>
> I checked in a patch that should fix the
> call_signal_handler_now SIGSEGV
> but I don't exactly understand why a signal handler was being called
> here at all. It doesn't look like anything in broadcast.c should be
> generating a signal unless there is another
> SIGSEGV/SIGBUS/SIGLL lurking
> elsewhere.
>
> If that is the case, then the signal handler bug may have been masking
> another problem. With luck, my checkin will allow us to see what the
> problem is. Or Robert will tell me why there should have
> been a signal
> generated in broadcast.c...
There shouldn't have been any errors from pthread, certainly not because
sleep() was called :|. I've never seen broadcast fail once the first
version was up and running...
Which is why I was so mystified and concerned about it.
Rob