From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH/RFC 7/9] libata: pdc_freeze() semantic change Date: Mon, 18 Jun 2007 14:14:55 +0900 Message-ID: <467614CF.5070006@gmail.com> References: <46720353.90209@tw.ibm.com> <467207BF.9020409@tw.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from wa-out-1112.google.com ([209.85.146.179]:16283 "EHLO wa-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752948AbXFRFPA (ORCPT ); Mon, 18 Jun 2007 01:15:00 -0400 Received: by wa-out-1112.google.com with SMTP id v27so2037580wah for ; Sun, 17 Jun 2007 22:15:00 -0700 (PDT) In-Reply-To: <467207BF.9020409@tw.ibm.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: albertl@mail.com Cc: Linux IDE , Doug Maxey , Mikael Pettersson Albert Lee wrote: > Patch 7/9: > > After checking the current implementations of freeze()/thaw(), it seems only pdc_freeze() > do more than simple irq masking. Remove the DMA disable code from pdc_freeze(). > > The question is the design/semantic of freeze()/thaw(). > Maybe we should limit them to simple irq on/off? Yeap, we can do that but if we do so it would be better to use ->irq_off/->irq_on hooks and kill ->freeze/->thaw. -- tejun