From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paul Gortmaker Subject: Review comments/questions on preempt-irqs-i386.patch Date: Tue, 19 Aug 2008 20:04:11 -0400 Message-ID: <20080820000408.GA28774@windriver.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: linux-rt-users@vger.kernel.org Return-path: Received: from [128.224.252.2] ([128.224.252.2]:59829 "EHLO yow-pgortmak-d1.corp.ad.wrs.com" rhost-flags-FAIL-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1752423AbYHTAMh (ORCPT ); Tue, 19 Aug 2008 20:12:37 -0400 Content-Disposition: inline Sender: linux-rt-users-owner@vger.kernel.org List-ID: I'm looking at what is in "preempt-irqs-i386.patch" in the broken out series for 2.6.26-rt1; and I've spotted something that I think might be a merge/carry-forward error (not sure). Specifically this chunk: - if (auto_eoi) /* master does Auto EOI */ - outb_pic(MASTER_ICW4_DEFAULT | PIC_ICW4_AEOI, PIC_MASTER_IMR); - else /* master expects normal EOI */ - outb_pic(MASTER_ICW4_DEFAULT, PIC_MASTER_IMR); + if (!auto_eoi) /* master expects normal EOI */ + outb_p(MASTER_ICW4_DEFAULT, PIC_MASTER_IMR); + else /* master does Auto EOI */ + outb_p(MASTER_ICW4_DEFAULT | PIC_ICW4_AEOI, PIC_MASTER_IMR); I'm wondering (a) is the replacement of outb_pic with the more generic outb_p intentional, or a carry forward artefact, and (b) the outb aside, I think this change looks like it essentially does: - if (x) - x-action(); - else - not-x-action(); + if (!x) + not-x-action(); + else + x-action(); Is that a fair assesment, or am I missing something? Thanks, Paul.