From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jike Song Subject: Re: [PATCH 7/8] drm/i915: Create vgpu specific write MMIO to reduce traps Date: Fri, 19 Sep 2014 15:43:43 +0800 Message-ID: <541BDEAF.3020709@intel.com> References: <1411152428-7226-1-git-send-email-jike.song@intel.com> <1411152428-7226-8-git-send-email-jike.song@intel.com> <20140919065917.GA21738@nuc-i3427.alporthouse.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by gabe.freedesktop.org (Postfix) with ESMTP id CDE5789A0F for ; Fri, 19 Sep 2014 00:47:39 -0700 (PDT) In-Reply-To: <20140919065917.GA21738@nuc-i3427.alporthouse.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" To: Chris Wilson Cc: daniel.vetter@intel.com, intel-gfx@lists.freedesktop.org, "Zhang, Yu C" List-Id: intel-gfx@lists.freedesktop.org [adding cc. By mistake I set suppresscc=all in my ~/.gitconfig] On 09/19/2014 02:59 PM, Chris Wilson wrote: > On Sat, Sep 20, 2014 at 02:47:07AM +0800, Jike Song wrote: >> From: Yu Zhang >> >> In the virtualized environment, forcewake operations are not >> necessory for the driver, because mmio accesses will be trapped >> and emulated by the host side, and real forcewake operations are >> also done in the host. New mmio write handlers are added to directly >> call the __raw_i915_write, therefore will reduce many traps and >> increase the overall performance for drivers runing in the VM >> with Intel GVT-g enhancement > > Explain why this is not done inside uncore_init. > -Chris Hi Chris, The gpu_active flag is set by intel_detect_vgpu(), which uses I915_READxx, should be after intel_uncore_init. -- Thanks, Jike