From owner-freebsd-questions Sat Apr 29 22:23: 5 2000 Delivered-To: freebsd-questions@freebsd.org Received: from mail.hcisp.net (Stargate.hcisp.net [208.60.89.18]) by hub.freebsd.org (Postfix) with SMTP id 9EEB337B6C9 for ; Sat, 29 Apr 2000 22:23:01 -0700 (PDT) (envelope-from tim@mysql.com) Received: (qmail 19042 invoked from network); 30 Apr 2000 04:31:58 -0000 Received: from modem5.hcisp.net (HELO threads.polyesthetic.msg) (208.60.89.71) by stargate.hcisp.net with SMTP; 30 Apr 2000 04:31:58 -0000 Received: (qmail 71863 invoked by uid 1001); 30 Apr 2000 05:22:41 -0000 From: "Thimble Smith" Date: Sun, 30 Apr 2000 01:22:41 -0400 To: freebsd-questions@freebsd.org Subject: SMP, FreeBSD + Linuxthreads Message-ID: <20000430012241.M2418@threads.polyesthetic.msg> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Hi, all of you extremely helpful and cheery people. Sometimes on the MySQL mailing list we get questions like the following: ----- Begin Forwarded message ----- I'd like to use SMP system for our Mysql database, but from what I found, FreeBSD native threads just can't make use of multiple processors (maybe it's time to mention this in FreeBSD specific section in the MySQL manual). ----- End forwarded message ----- I'm wondering what the best response is for this (assume they are using FreeBSD 4-STABLE). o No, our native threads are just fine for applications like the MySQL server. o Use the LinuxThreads port, which makes better use of SMP machines. o ... something else. Also, is this idea because syscalls are serialized per process (so I/O is serialized)? Or is there another way to account for this idea? Are there plans (in the works, down the road) to change whatever the problem is (assuming there really is a problem) that I can mention? Are there papers I should read that describe the current (and future) architecture? For example, I see this possibly unrelated note in the vfork(2) man page: BUGS This system call will be eliminated when proper system sharing mechanisms are implemented. Users should not depend on the memory sharing semantics of vfork(2) as it will, in that case, be made synonymous to fork(2). Thanks, Tim (your FreeBSD advocate in MySQL-land) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message