From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Jones Subject: Re: [PATCH] Poll-based IDE driver Date: Wed, 8 Oct 2003 17:48:51 +0100 Sender: linux-ide-owner@vger.kernel.org Message-ID: <20031008164851.GT29736@redhat.com> References: <20031008115051.GD705@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from pix-525-pool.redhat.com ([66.187.233.200]:13412 "EHLO lacrosse.corp.redhat.com") by vger.kernel.org with ESMTP id S261592AbTJHQtg (ORCPT ); Wed, 8 Oct 2003 12:49:36 -0400 Content-Disposition: inline In-Reply-To: List-Id: linux-ide@vger.kernel.org To: Andre Hedrick Cc: Srivatsa Vaddagiri , Alan Cox , lkcd-devel@lists.sourceforge.net, Linux Kernel Mailing List , linux-ide@vger.kernel.org On Wed, Oct 08, 2003 at 09:36:36AM -0700, Andre Hedrick wrote: > > Does not matter, priority is to get content to platter and the hell with > everything else. I don't buy this. Without correct udelay()'s, how is code like this.. for (i = 0; i < 10; i++) { dump_udelay(1); if (OK_STAT((stat = hwif->INB(IDE_STATUS_REG)), good, bad)) return 0; } expected to work ? It won't wait for 10usec at all, but be over almost instantly. Ramming commands at the drive before its status has settled doesn't strike me as a particularly safe thing to do. Dave -- Dave Jones http://www.codemonkey.org.uk