Mailing-List:	contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe:	<mailto:cygwin-subscribe AT sources DOT redhat DOT com>
List-Archive:	<http://sources.redhat.com/ml/cygwin/>
List-Post:	<mailto:cygwin AT sources DOT redhat DOT com>
List-Help:	<mailto:cygwin-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender:	cygwin-owner AT sources DOT redhat DOT com
Delivered-To:	mailing list cygwin AT sources DOT redhat DOT com
Message-ID:	<410A7080D2D2D11185F800A0C9663B823E8954@pulse-bdc.pulse.com>
From:	Ted Soo-Hoo <Ted DOT Soo-Hoo AT go DOT ecitele DOT com>
To:	"'cygwin AT sourceware DOT cygnus DOT com'" <cygwin AT sourceware DOT cygnus DOT com>
Subject:	Re: 1.1.4 select and poll false positive past 32 descriptors
Date:	Thu, 10 Aug 2000 15:06:11 -0400
MIME-Version:	1.0
X-Mailer:	Internet Mail Service (5.5.2650.21)

Chris Faylor wrote:

> Thanks for the extra details.  I'll fix this.

Thanks for taking a look at select. Here are some other comments.
I've looked at the source code some more and also tried to learn a
little about Win32 programming. I still don't understand all of what's
going on in the select.cc code so please do as you like with these.
I don't really need an answer. I'll just be pleased if any of this helps.

  HANDLE w4[MAXIMUM_WAIT_OBJECTS];
This constant is 64 whenever I've seen it. The loop in select.cc is
not doing a check for overrunning the array. I hope users who try
wide selects won't cause that. I don't quite understand the w4 use.
I know that in my test where I set a lot of read bits, m was small.

I was also wondering if poll() was implemented with select() as is
sometimes done, so had a look at poll.cc. In this file my concern
is about width of the fd_set being used. The select.cc code tries
to dynamically adapt that by allocfd_set()  to the needed bit-width.
If poll.cc is using the standard 64-bit fd_set (and FD_ZERO, etc.)
it may write past the allocated bits if it gets a descriptor over 63.

Still, even if poll has a limit for awhile it would be nice if a really
wide select could be supported, as the code seems to try to do.

Thanks once again for your help!

Regards,
Ted


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com

- Raw text -