Date: Mon, 18 May 1998 19:17:31 -0700 From: John Polstra <jdp@polstra.com> To: chuckr@glue.umd.edu Cc: hackers@FreeBSD.ORG Subject: Re: g++ new allocator Message-ID: <199805190217.TAA08741@austin.polstra.com> In-Reply-To: <Pine.BSF.3.96.980516230958.16302l-100000@localhost> References: <Pine.BSF.3.96.980516230958.16302l-100000@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <Pine.BSF.3.96.980516230958.16302l-100000@localhost>, Chuck Robey <chuckr@glue.umd.edu> wrote: > On Sun, 17 May 1998, Greg Lehey wrote: > > > On Sat, 16 May 1998 at 10:20:38 -0400, Chuck Robey wrote: > > > Does anyone know if the "new" operator, using g++, uses it's own > > > allocator, or PHK's libc malloc? I have a problem I'm trying to shoot > > > that uses the libstdc++ and sigbus's, and I was wondering which route > > > the code is going. > > > > My understanding is that it uses whichever malloc is bound in with the > > executable. > > ... Umm ... it's built using our libstdc++ (I just compiled it). The > question is what malloc does the g++ using our own libstdc++ (on > current, actually) use? I didn't post this to current because I didn't > think it applied only to current. > > I guess I don't understand what you mean by "whichever malloc is bound > in with the executable", if I'm doing the binding here. I'm 99% certain it uses malloc. But you can easily find out for sure with a simple experiment. Create a tiny C++ program that allocates something with "new". Compile it for debugging. Run it under the debugger and set a breakpoint on "malloc". Look at the stack trace and see whether it's being called from new. -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199805190217.TAA08741>