From owner-freebsd-current Sun Mar 16 14:47:15 2003 Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5B08837B401 for ; Sun, 16 Mar 2003 14:47:14 -0800 (PST) Received: from mailman.zeta.org.au (mailman.zeta.org.au [203.26.10.16]) by mx1.FreeBSD.org (Postfix) with ESMTP id 12BD143F85 for ; Sun, 16 Mar 2003 14:47:13 -0800 (PST) (envelope-from bde@zeta.org.au) Received: from katana.zip.com.au (katana.zip.com.au [61.8.7.246]) by mailman.zeta.org.au (8.9.3/8.8.7) with ESMTP id JAA12301 for ; Mon, 17 Mar 2003 09:47:08 +1100 Date: Mon, 17 Mar 2003 09:47:05 +1100 (EST) From: Bruce Evans X-X-Sender: bde@gamplex.bde.org To: current@freebsd.org Subject: dubious sleep priorities in cv family Message-ID: <20030317080500.E15368@gamplex.bde.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Everything in kern_condvar.c uses a special form of sleeping which doesn't change the priority to one specified by the caller like for ordinary tsleep(). This affects at least select() and poll(). Processes sleeping in select() do it at priority PSOCK in RELENG_4 and at the user priority in -current. High user priorities decay to (PUSER + niceness) so the possibilities for priority inversion from this are limited and it's mainly a latency bug. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message