X-Recipient: archive-cygwin AT delorie DOT com
X-SWARE-Spam-Status: No, hits=-2.6 required=5.0	tests=AWL,BAYES_00,GENERIC_IXHASH,KHOP_RCVD_UNTRUST,KHOP_THREADED,RCVD_IN_HOSTKARMA_W,RCVD_IN_HOSTKARMA_WL,RP_MATCHES_RCVD,SPF_HELO_PASS
X-Spam-Check-By: sourceware.org
X-IronPortListener: Outbound_SMTP
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AqkJALKur1CcKEep/2dsb2JhbABEhie5G4EKbAeCHgEBAQMBEhERSgsCAQgaAgYgAgICHRMVEAEBBBsah2UGok6KB5JMgSKOQzJhA401mQ6Cb4Id
From: "Lavrentiev, Anton (NIH/NLM/NCBI) [C]" <lavr AT ncbi DOT nlm DOT nih DOT gov>
To: "cygwin AT cygwin DOT com" <cygwin AT cygwin DOT com>
Subject: RE: Possible race in SYSV IPC (semaphores)
Date: Fri, 23 Nov 2012 17:16:02 +0000
Message-ID: <5F8AAC04F9616747BC4CC0E803D5907D0162C6@MLBXv04.nih.gov>
References: <5F8AAC04F9616747BC4CC0E803D5907D012856 AT MLBXV09 DOT nih DOT gov> <20121123113605 DOT GN17347 AT calimero DOT vinschen DOT de> <20121123131020 DOT GR17347 AT calimero DOT vinschen DOT de> <20121123133332 DOT GU17347 AT calimero DOT vinschen DOT de> <20121123164404 DOT GX17347 AT calimero DOT vinschen DOT de>
In-Reply-To: <20121123164404.GX17347@calimero.vinschen.de>
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
X-IsSubscribed: yes
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Id: <cygwin.cygwin.com>
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from base64 to 8bit by delorie.com id qANHGP0N025282

> This should avoid the race (*and* work...)
> Please give it a try.

Thanks!  I've tried both patches (pipe race + semadj), and they do seem to work!

Although (not being a party pooper :-), I think that all the logic
around "pipe_instance" can now be dropped entirely, and benefit from eliminating
an unnecessary synchronization point (pipe_instance is now only used
for logging a proper pipe ID, which is not even seen in regular use,
and for an assert(), which again is only there in debug mode builds).
So there is no need to arrange Enter'ing/Leave'ing of the critical section
to do just the hidden things, yet the working code would be a bit cleaner.

Anton Lavrentiev
Contractor NIH/NLM/NCBI