From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [git patches] libata fixes for .26 Date: Sat, 05 Jul 2008 12:18:28 +0900 Message-ID: <486EE804.6020202@kernel.org> References: <20080704131005.GA30768@havoc.gtf.org> <486EE6CA.2040908@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from hera.kernel.org ([140.211.167.34]:40732 "EHLO hera.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751162AbYGEDS4 (ORCPT ); Fri, 4 Jul 2008 23:18:56 -0400 In-Reply-To: <486EE6CA.2040908@kernel.org> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Linus Torvalds Cc: Jeff Garzik , Andrew Morton , linux-ide@vger.kernel.org, LKML Tejun Heo wrote: > so, it basically behaves like level triggered IRQ latch for ports, > which also creates an easy-to-miss requirement for MSI implementation > where the controller should generate a new MSI message when the driver > clears some of the IRQ pending bits but not all. Oops, confused. The easy-to-miss requirement is the controller needing to resend the message if the irq_stat is cleared but port IRQ is still pending. Resending on partial clear is always necessary. Thanks. -- tejun