From owner-svn-src-head@FreeBSD.ORG Mon Apr 1 16:30:12 2013 Return-Path: Delivered-To: svn-src-head@freebsd.org Received: from mx1.freebsd.org (mx1.FreeBSD.org [8.8.178.115]) by hub.freebsd.org (Postfix) with ESMTP id 6A4A32C7; Mon, 1 Apr 2013 16:30:12 +0000 (UTC) (envelope-from jhb@freebsd.org) Received: from bigwig.baldwin.cx (bigwig.baldwin.cx [IPv6:2001:470:1f11:75::1]) by mx1.freebsd.org (Postfix) with ESMTP id 4885415D; Mon, 1 Apr 2013 16:30:12 +0000 (UTC) Received: from jhbbsd.localnet (unknown [209.249.190.124]) by bigwig.baldwin.cx (Postfix) with ESMTPSA id B686EB960; Mon, 1 Apr 2013 12:30:11 -0400 (EDT) From: John Baldwin To: "Alexander V. Chernikov" Subject: Re: svn commit: r248705 - head/sys/dev/ipmi Date: Mon, 1 Apr 2013 11:54:09 -0400 User-Agent: KMail/1.13.5 (FreeBSD/8.2-CBSD-20110714-p25; KDE/4.5.5; amd64; ; ) References: <201303251430.r2PEUYcx045864@svn.freebsd.org> In-Reply-To: <201303251430.r2PEUYcx045864@svn.freebsd.org> MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201304011154.09795.jhb@freebsd.org> X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.7 (bigwig.baldwin.cx); Mon, 01 Apr 2013 12:30:11 -0400 (EDT) Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org X-BeenThere: svn-src-head@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: SVN commit messages for the src tree for head/-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 01 Apr 2013 16:30:12 -0000 On Monday, March 25, 2013 10:30:34 am Alexander V. Chernikov wrote: > Author: melifaro > Date: Mon Mar 25 14:30:34 2013 > New Revision: 248705 > URL: http://svnweb.freebsd.org/changeset/base/248705 > > Log: > Unlock IPMI sc while performing requests via KCS and SMIC interfaces. > It is already done in SSIF interface code. > This reduces contention/spinning reported by many users. Eh, this was on purpose to prevent concurrent access to the hardware. SSIF doesn't do this because the smbus driver itself does locking internally. There are no followups in the PR to state how this patch helps (or if it was tested). OTOH, we should probably make KCS and SMIC perform their requests synchronously rather than kicking them over to a worker thread and only use a worker thread for SSIF. This has the advantage that you could make it interruptible so you could Ctrl-C ipmiutil and have it do something useful. Also, the current KCS/SMIC code has a timeout, it shouldn't spin forever. The fact that it is spinning forever is a different bug (possibly recently fixed by making 'ticks' volatile). That different bug is probably what should be fixed instead. -- John Baldwin