From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Jeffrey W. Baker" Subject: AHCI power down unused ports merge status? Date: Sat, 27 Sep 2008 12:39:15 -0700 Message-ID: <1222544355.26413.6.camel@stinklet> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from wa-out-1112.google.com ([209.85.146.177]:55785 "EHLO wa-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753159AbYI0TjX (ORCPT ); Sat, 27 Sep 2008 15:39:23 -0400 Received: by wa-out-1112.google.com with SMTP id v27so702333wah.21 for ; Sat, 27 Sep 2008 12:39:21 -0700 (PDT) Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: linux-ide@vger.kernel.org I was just poking around in the ICH8 documentation and /dev/mem and I notice that all three of the SATA PHYs on my laptop are enabled, when I have only one SATA target device. AHCI allows the port to be disabled, and this saves a lot of power, but this functionality is not exposed in Linux. There are patches floating around on this mailing list to disable the port by echoing "power_off" to ALPM in sysfs, but this patch isn't merged in libata-dev and I don't see it in 2.6.27-rc7 either. Is there any hope of this getting merged? Is it dangerous or otherwise not advised? I think users would really appreciate having this. Disabling unused ports results in +10% extra battery life on my laptop. cf http://marc.info/?l=linux-ide&m=121504056904748&w=2 I think part 1 is severable from part 2, so perhaps the sysfs bits can be merged without the default policy part that seems objectionable. -jwb