[Intel-gfx] [PATCH 3/3] drm/i915: Use insert_page for pwrite_fast

[Chris Wilson]

On Thu, Nov 05, 2015 at 05:15:59PM +0530, ankitprasad.r.sharma at intel.com wrote:
> From: Ankitprasad Sharma <ankitprasad.r.sharma at intel.com>
> 
> In pwrite_fast, map an object page by page if obj_ggtt_pin fails. First,
> we try a nonblocking pin for the whole object (since that is fastest if
> reused), then failing that we try to grab one page in the mappable
> aperture. It also allows us to handle objects larger than the mappable
> aperture (e.g. if we need to pwrite with vGPU restricting the aperture
> to a measely 8MiB or something like that).
> 
> Signed-off-by: Ankitprasad Sharma <ankitprasad.r.sharma at intel.com>
> ---
>  drivers/gpu/drm/i915/i915_gem.c | 92 ++++++++++++++++++++++++++++++-----------
>  1 file changed, 69 insertions(+), 23 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index bf5ef7a..9132240 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -766,14 +766,26 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev,
>  			 struct drm_file *file)
>  {
>  	struct drm_i915_private *dev_priv = dev->dev_private;
> +	struct drm_mm_node node;
>  	ssize_t remain;
>  	loff_t offset, page_base;
>  	char __user *user_data;
> -	int page_offset, page_length, ret;
> +	int page_offset, page_length, ret, i;
> +	bool pinned = true;
>  
>  	ret = i915_gem_obj_ggtt_pin(obj, 0, PIN_MAPPABLE | PIN_NONBLOCK);
> -	if (ret)
> -		goto out;
> +	if (ret) {
> +		pinned = false;
> +		memset(&node, 0, sizeof(node));
> +		ret = drm_mm_insert_node_in_range_generic(&dev_priv->gtt.base.mm,
> +							  &node, 4096, 0,
> +							  I915_CACHE_NONE, 0,
> +							  dev_priv->gtt.mappable_end,
> +							  DRM_MM_SEARCH_DEFAULT,
> +							  DRM_MM_CREATE_DEFAULT);
> +		if (ret)
> +			goto out;

Prefer to refer to my original patch as to why this wrong.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre