From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755046Ab1LIXxQ (ORCPT ); Fri, 9 Dec 2011 18:53:16 -0500 Received: from home.keithp.com ([63.227.221.253]:59599 "EHLO keithp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752428Ab1LIXxO (ORCPT ); Fri, 9 Dec 2011 18:53:14 -0500 X-Amavis-Alert: BAD HEADER SECTION, Duplicate header field: "References" From: Keith Packard To: intel-gfx@lists.freedesktop.org Cc: linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, Keith Packard Subject: drm/i915: Enabling RC6 where possible Date: Fri, 9 Dec 2011 15:53:47 -0800 Message-Id: <1323474829-2501-1-git-send-email-keithp@keithp.com> X-Mailer: git-send-email 1.7.7.3 In-Reply-To: <1321684889-18691-1-git-send-email-keithp@keithp.com> References: <1321684889-18691-1-git-send-email-keithp@keithp.com> References: <1322526298-2746-1-git-send-email-eugeni.dodonov@intel.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Ok, here's a "final" patch set to enable RC6 where possible on SNB and IVB machines. The first patch creates a new variable, intel_iommu_enabled, that is exported by the intel iommu code and set when that code has successfully initialized itself. The old plan of using no_iommu || dmar_disabled would work -- those variables are set only by kernel parameters and don't reflect what the system is actually doing about virtualization. The second patch uses that value on SNB to tell whether RC6 can be enabled by default. On IVB, RC6 is always enabled. Of course, in all cases, you can override the RC6 setting with the i915 module parameter. For those of you who have experienced the delights of RC6 crashing your machines, please test as this will be heading to 3.2 unless you find something wrong with it. -keith