Mailing-List: contact cygwin-developers-help AT cygwin DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-developers-owner AT cygwin DOT com Delivered-To: mailing list cygwin-developers AT cygwin DOT com Message-ID: <010101c2205f$669c81b0$6132bc3e@BABEL> From: "Conrad Scott" To: References: <001901c21eee$8aadf060$0200a8c0 AT lifelesswks> <034301c21f49$a989f6e0$6132bc3e AT BABEL> <002e01c21fca$f0b19560$1800a8c0 AT LAPTOP> <004401c22031$7f167620$6132bc3e AT BABEL> <20020630163635 DOT GA32201 AT redhat DOT com> Subject: Re: Threads and the C++ new and delete operators Date: Sun, 30 Jun 2002 18:56:02 +0100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 "Christopher Faylor" wrote: > You may notice that cygwin does not use new/delete very much. This is > one of the reasons why. Ahh. Okay. So, to be *completely* thread-safe (until gcc 3.1), I should avoid any use of new and delete in multi-threaded programs (unless I can confine it to one thread)? And that implies no use of any C++ classes that might themselves use new and delete (e.g. iostreams)? This is no big deal and I'm most of the way there already (i.e. there's no explicit use of new/delete now in cygserver). BTW I'm not being facetious here or anything, just checking the score. It's been a slightly stressful few days narrowing this one down and I'm getting paranoid. Thanks, // Conrad