[PATCH] drm/i915/gvt: Fix an logic issue in ppgtt_allocate_oos_page

[Zhi Wang]

Please take this one. :P Thanks. I forget to change the author in the 
previous patch after re-basing it to staging branch.

On 04/01/17 13:04, Zhi Wang wrote:
> From: Min He <min.he at intel.com>
>
> In ppgtt_allocate_oos_page, when there's no free oos_page, we'll sync an
> old oos page and detach it. But that oos page may not belong to current
> gvt, so we need to find corresponding gvt and sync to it.
>
> Fixes: 2707e44 ("drm/i915/gvt: vGPU graphics memory virtualization")
> Signed-off-by: Min He <min.he at intel.com>
> Signed-off-by: Zhi Wang <zhi.a.wang at intel.com>
> ---
>   drivers/gpu/drm/i915/gvt/gtt.c | 7 +++++--
>   1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
> index 6da4e44..cb13bc7 100644
> --- a/drivers/gpu/drm/i915/gvt/gtt.c
> +++ b/drivers/gpu/drm/i915/gvt/gtt.c
> @@ -1160,6 +1160,7 @@ static int ppgtt_allocate_oos_page(struct intel_vgpu *vgpu,
>   	struct intel_gvt *gvt = vgpu->gvt;
>   	struct intel_gvt_gtt *gtt = &gvt->gtt;
>   	struct intel_vgpu_oos_page *oos_page = gpt->oos_page;
> +	struct intel_vgpu_ppgtt_spt *spt = NULL;
>   	int ret;
>
>   	WARN(oos_page, "shadow PPGTT page has already has a oos page\n");
> @@ -1167,10 +1168,12 @@ static int ppgtt_allocate_oos_page(struct intel_vgpu *vgpu,
>   	if (list_empty(&gtt->oos_page_free_list_head)) {
>   		oos_page = container_of(gtt->oos_page_use_list_head.next,
>   			struct intel_vgpu_oos_page, list);
> -		ret = ppgtt_set_guest_page_sync(vgpu, oos_page->guest_page);
> +		spt = guest_page_to_ppgtt_spt(oos_page->guest_page);
> +		ret = ppgtt_set_guest_page_sync(spt->vgpu,
> +				oos_page->guest_page);
>   		if (ret)
>   			return ret;
> -		ret = detach_oos_page(vgpu, oos_page);
> +		ret = detach_oos_page(spt->vgpu, oos_page);
>   		if (ret)
>   			return ret;
>   	} else
>