Date: Tue, 28 May 2002 23:50:18 -0400 (EDT) From: John Baldwin <jhb@FreeBSD.org> To: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> Cc: cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org Subject: Re: cvs commit: src/sys/kern uipc_socket.c Message-ID: <XFMail.20020528235018.jhb@FreeBSD.org> In-Reply-To: <200205290337.g4T3bNDc010112@khavrinen.lcs.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On 29-May-2002 Garrett Wollman wrote: > <<On Tue, 28 May 2002 23:05:06 -0400 (EDT), John Baldwin <jhb@FreeBSD.org> said: > >> Define "an appropriate constructor" that can somehow make up for broken >> code on another process that is knowingly accessing stale pointers. > > An appropriate constructor is one which doesn't blow away the version > number, which allows us to detect when the object has been recycled. > Please go back and read up on non-blocking synchronization before > continuing this thread. This ignores the fact that you can be reading stale values anyway without a lock of some sort on modern architectures. This also makes the assumption that a chunk of memory is once-a-socket always-a-socket which doesn't let UMA properly adjust to spikes in system load by shuffling memory resources around to subsystems that need it most from other subsystems that don't. -- John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.20020528235018.jhb>