From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: PIO interrupt handling Date: Thu, 15 Sep 2005 22:54:54 -0400 Message-ID: <432A33FE.4090703@pobox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.dvmed.net ([216.237.124.58]:28123 "EHLO mail.dvmed.net") by vger.kernel.org with ESMTP id S1030547AbVIPCzG (ORCPT ); Thu, 15 Sep 2005 22:55:06 -0400 Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: "linux-ide@vger.kernel.org" Cc: Albert Lee , Doug Maxey , Alan Cox Just wanted to point out positive feedback for Albert's patch: http://lkml.org/lkml/2005/9/15/358 The reason I mentioned Albert's patch in my reply, in that thread, is that sometimes SATA controllers may not emulate nIEN properly, or may depend on the device to set (or not) the 'I' bit in the SATA FIS. This success is an example of why I want to move libata to use interrupt-driven PIO by default, long term. We still want to keep the polling code, and perhaps even expand things a bit so that users with broken interrupt routing (read: no interrupt delivery for ATA controller) can still use their disks. Jeff