From owner-freebsd-current Tue Aug 25 07:11:08 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA02277 for freebsd-current-outgoing; Tue, 25 Aug 1998 07:11:08 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from khavrinen.lcs.mit.edu (khavrinen.lcs.mit.edu [18.24.4.193]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA02265; Tue, 25 Aug 1998 07:11:00 -0700 (PDT) (envelope-from wollman@khavrinen.lcs.mit.edu) Received: (from wollman@localhost) by khavrinen.lcs.mit.edu (8.9.1/8.9.1) id KAA01928; Tue, 25 Aug 1998 10:10:12 -0400 (EDT) (envelope-from wollman) Date: Tue, 25 Aug 1998 10:10:12 -0400 (EDT) From: Garrett Wollman Message-Id: <199808251410.KAA01928@khavrinen.lcs.mit.edu> To: "Gary Palmer" Cc: Chuck Robey , freebsd-current@FreeBSD.ORG Subject: Re: Threads across processors In-Reply-To: <3686.904031598@gjp.erols.com> References: <3686.904031598@gjp.erols.com> Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG < said: > Theres a bunch of other stuff out there that *should* be threaded, but > unfortunately our threads support, while complete, seems to be a bit slow at > times It's a bit slow because of all the work necessary to ensure mutual exclusion in those parts of the thread-kernel which ought to be doable in the kernel-kernel. I've watched Cyclone hit 150,000 syscalls a SECOND when it's not doing much of anything. It's impressive that it can do that, but I'd like to have some CPU left over for useful work... Most of those system calls were to sigprocmask() to protect some critical section from the signals that drive the thred scheduler. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same wollman@lcs.mit.edu | O Siem / The fires of freedom Opinions not those of| Dance in the burning flame MIT, LCS, CRS, or NSA| - Susan Aglukark and Chad Irschick To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message