From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jike Song <jike.song@intel.com>
Subject: Re: [PATCH 8/8] drm/i915: Support alias ppgtt in VM if
 ppgtt is enabled
Date: Mon, 22 Sep 2014 19:17:41 +0800
Message-ID: <54200555.4050008@intel.com>
References: <1411152428-7226-1-git-send-email-jike.song@intel.com>
 <1411152428-7226-9-git-send-email-jike.song@intel.com>
 <20140919082539.GG21738@nuc-i3427.alporthouse.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; Format="flowed"
Content-Transfer-Encoding: 7bit
Return-path: <intel-gfx-bounces@lists.freedesktop.org>
Received: from mga11.intel.com (mga11.intel.com [192.55.52.93])
 by gabe.freedesktop.org (Postfix) with ESMTP id 765C06E143
 for <intel-gfx@lists.freedesktop.org>; Mon, 22 Sep 2014 04:21:37 -0700 (PDT)
In-Reply-To: <20140919082539.GG21738@nuc-i3427.alporthouse.com>
List-Unsubscribe: <http://lists.freedesktop.org/mailman/options/intel-gfx>,
 <mailto:intel-gfx-request@lists.freedesktop.org?subject=unsubscribe>
List-Archive: <http://lists.freedesktop.org/archives/intel-gfx>
List-Post: <mailto:intel-gfx@lists.freedesktop.org>
List-Help: <mailto:intel-gfx-request@lists.freedesktop.org?subject=help>
List-Subscribe: <http://lists.freedesktop.org/mailman/listinfo/intel-gfx>,
 <mailto:intel-gfx-request@lists.freedesktop.org?subject=subscribe>
Errors-To: intel-gfx-bounces@lists.freedesktop.org
Sender: "Intel-gfx" <intel-gfx-bounces@lists.freedesktop.org>
To: Chris Wilson <chris@chris-wilson.co.uk>, daniel.vetter@intel.com, intel-gfx@lists.freedesktop.org
List-Id: intel-gfx@lists.freedesktop.org

On 09/19/2014 04:25 PM, Chris Wilson wrote:
>
> This should be moved to sanitize_enable_ppgtt(), probably by expanding
> HAS_PPGTT(), e.g.:
>
> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> index 07dafa2c2d8c..b1fa13942d14 100644
> --- a/drivers/gpu/drm/i915/i915_drv.h
> +++ b/drivers/gpu/drm/i915/i915_drv.h
> @@ -2139,8 +2139,6 @@ struct drm_i915_cmd_table {
>
>   #define HAS_HW_CONTEXTS(dev)   (INTEL_INFO(dev)->gen >= 6)
>   #define HAS_LOGICAL_RING_CONTEXTS(dev) (INTEL_INFO(dev)->gen >= 8)
> -#define HAS_ALIASING_PPGTT(dev)        (INTEL_INFO(dev)->gen >= 6)
> -#define HAS_PPGTT(dev)         (INTEL_INFO(dev)->gen >= 7 && !IS_GEN8(dev))
>   #define USES_PPGTT(dev)                (i915.enable_ppgtt)
>   #define USES_FULL_PPGTT(dev)   (i915.enable_ppgtt == 2)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
> index a234446a8678..3bea0bdfd276 100644
> --- a/drivers/gpu/drm/i915/i915_gem_gtt.c
> +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
> @@ -35,13 +35,23 @@ static void chv_setup_private_ppat(struct drm_i915_private *dev_priv);
>
>   static int sanitize_enable_ppgtt(struct drm_device *dev, int enable_ppgtt)
>   {
> -       if (enable_ppgtt == 0 || !HAS_ALIASING_PPGTT(dev))
> +       bool has_aliasing_ppgtt;
> +       bool has_full_ppgtt;
> +
> +       has_aliasing_ppgtt = INTEL_INFO(dev)->gen >= 6;
> +       has_full_ppgtt = INTEL_INFO(dev)->gen >= 7;
> +       if (IS_GEN8(dev))
> +               has_full_ppgtt = false; /* XXX why? */
> +       if (intel_vgpu_active(dev))
> +               has_full_ppgtt = false; /* emulation is too hard */
> +
> +       if (enable_ppgtt == 0 || !has_aliasing_ppgtt)
>                  return 0;
>
>          if (enable_ppgtt == 1)
>                  return 1;
>
> -       if (enable_ppgtt == 2 && HAS_PPGTT(dev))
> +       if (enable_ppgtt == 2 && has_full_ppgtt)
>                  return 2;
>
>   #ifdef CONFIG_INTEL_IOMMU
> @@ -59,7 +69,7 @@ static int sanitize_enable_ppgtt(struct drm_device *dev, int enable_ppgtt)
>                  return 0;
>          }
>
> -       return HAS_PPGTT(dev) ? 2 : HAS_ALIASING_PPGTT(dev) ? 1 : 0;
> +       return has_full_ppgtt ? 2 : has_aliasing_ppgtt ? 1 : 0;
>   }

Thanks for the demo. Currently sanitize_enable_ppgtt() is called by i915_gem_gtt_init(), before
intel_check_vgpu().

I'm trying to detect VGPU as early as possible, maybe between intel_detect_pch() and intel_uncore_init(),
using bare readq/readw instead of I915_READxx.


>
> -Chris
>
--
Thanks,
Jike