From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: Re: [PATCH 1/2] ahci: preserve PORTS_IMPL over host resets Date: Fri, 06 Oct 2006 00:02:39 +0900 Message-ID: <45251E8F.5000401@gmail.com> References: <20060930122121.GR25800@htj.dyndns.org> <4524EC09.2070608@pobox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from py-out-1112.google.com ([64.233.166.176]:30409 "EHLO py-out-1112.google.com") by vger.kernel.org with ESMTP id S1751472AbWJEPCc (ORCPT ); Thu, 5 Oct 2006 11:02:32 -0400 Received: by py-out-1112.google.com with SMTP id n25so770975pyg for ; Thu, 05 Oct 2006 08:02:31 -0700 (PDT) In-Reply-To: <4524EC09.2070608@pobox.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Jeff Garzik Cc: linux-ide@vger.kernel.org, robbat2@gentoo.org Jeff Garzik wrote: > Tejun Heo wrote: >> Instead of writing 0xf blindly, preserve the content of write-once >> PORTS_IMPL register over host resets. >> >> This patch is taken from Jeff Garzik's AHCI init update patch. >> >> Signed-off-by: Tejun Heo >> Cc: Jeff Garzik > > One reason why I didn't go ahead and apply this is for > uninitalized-by-BIOS cases that are the reason I originally hardcoded a > 0xf value into the register. > > This patch is OK, iff we also include a test to see if impl_save==0 > prior to global reset. If that is the case, we should hardcode 0xf or > (1<