From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 871 invoked by alias); 29 Aug 2010 14:17:26 -0000 Received: (qmail 855 invoked by uid 22791); 29 Aug 2010 14:17:25 -0000 X-Spam-Check-By: sourceware.org Received: from aquarius.hirmke.de (HELO calimero.vinschen.de) (217.91.18.234) by sourceware.org (qpsmtpd/0.83/v0.83-20-g38e4449) with ESMTP; Sun, 29 Aug 2010 14:17:19 +0000 Received: by calimero.vinschen.de (Postfix, from userid 500) id 9A7966D4364; Sun, 29 Aug 2010 16:17:16 +0200 (CEST) Date: Sun, 29 Aug 2010 14:41:00 -0000 From: Corinna Vinschen To: cygwin@cygwin.com, cygwin-xfree@cygwin.com Subject: Re: /dev/windows and select() [was Re: Slow response to keypresses in xorg-server-1.8.0-1] Message-ID: <20100829141716.GG6726@calimero.vinschen.de> Mail-Followup-To: cygwin@cygwin.com, cygwin-xfree@cygwin.com References: <4BDC3161.9070101@cornell.edu> <4BDDE5F1.9080204@cornell.edu> <4C2B8171.5020409@dronecode.org.uk> <4C5B08AE.4080902@pobox.com> <4C5DD910.9050809@dronecode.org.uk> <4C7A62FD.10506@dronecode.org.uk> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <4C7A62FD.10506@dronecode.org.uk> User-Agent: Mutt/1.5.20 (2009-06-14) Mailing-List: contact cygwin-xfree-help@cygwin.com; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-xfree-owner@cygwin.com Reply-To: cygwin-xfree@cygwin.com Mail-Followup-To: cygwin-xfree@cygwin.com X-SW-Source: 2010-08/txt/msg00130.txt.bz2 On Aug 29 14:39, Jon TURNEY wrote: > On 08/08/2010 12:04, Andy Koppe wrote: > >On 7 August 2010 23:07, Jon TURNEY wrote: > >>Hmmm, looking again at the implementation of select(), I don't immediately > >>see that when waiting on /dev/windows, it checks that the message queue has > >>old messages on it before waiting. The MSDN documentation for > >>MsgWaitForMultipleObjects() seems to says that messages which had arrived > >>before the last PeekMessage() etc. aren't considered new and so don't end > >>the wait? > > > >I think you're right, a call to PeekMessage is needed for proper > >select() semantics: it shouldn't block if data is available for > >reading. > > Attached is a small test-case which seems to demonstrate this problem. > > Run ./dev-windows-select-test and observe select() blocks for the > full timeout, despite the fact that the /dev/windows fd is ready for > reading (and it reported as such as the end of the timeout) > > If you run './dev-windows-select-test -skip' to skip the > PeekMessage(), select() returns immediately, indicating the > /dev/windows fd is ready for reading. > [...] Thanks for the testcase. I examined this and I think I have a workaround. MSDN states that there's a flag QS_ALLPOSTMESSAGE for MsgWaitForMultipleObjects, which is not cleared by PeekMessage, if the wMsgFilterMin and wMsgFilterMax arguments are not both 0. So, what I did was to add the QS_ALLPOSTMESSAGE flag to the MsgWaitForMultipleObjects call in select.cc, and to change the PeekMessage call in select.cc:peek_windows() from PeekMessage (&m, (HWND) h, 0, 0, PM_NOREMOVE) to PeekMessage (&m, (HWND) h, 1, UINT_MAX, PM_NOREMOVE) Same in your above test application. This appears to do the trick. However, I'm not exactly sure if that's a valid fix. Patch below. Corinna Index: select.cc =================================================================== RCS file: /cvs/src/src/winsup/cygwin/select.cc,v retrieving revision 1.160 diff -u -p -r1.160 select.cc --- select.cc 2 Apr 2010 22:36:44 -0000 1.160 +++ select.cc 29 Aug 2010 14:16:18 -0000 @@ -287,7 +287,8 @@ select_stuff::wait (fd_set *readfds, fd_ if (!windows_used) wait_ret = WaitForMultipleObjects (m, w4, FALSE, ms); else - wait_ret = MsgWaitForMultipleObjects (m, w4, FALSE, ms, QS_ALLINPUT); + wait_ret = MsgWaitForMultipleObjects (m, w4, FALSE, ms, + QS_ALLINPUT | QS_ALLPOSTMESSAGE); switch (wait_ret) { @@ -1531,7 +1532,7 @@ peek_windows (select_record *me, bool) if (me->read_selected && me->read_ready) return 1; - if (PeekMessage (&m, (HWND) h, 0, 0, PM_NOREMOVE)) + if (PeekMessage (&m, (HWND) h, 1, UINT_MAX, PM_NOREMOVE)) { me->read_ready = true; select_printf ("window %d(%p) ready", me->fd, me->fh->get_handle ()); -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://x.cygwin.com/docs/ FAQ: http://x.cygwin.com/docs/faq/