Yes, the original one could fix my problem. I found it is actually called several times for the NULL bo in some application. And this patch will improve the performance a little.
-----Original Message----- From: Libva [mailto:libva-boun...@lists.freedesktop.org] On Behalf Of Xiang, Haihao Sent: Friday, December 30, 2016 4:26 PM To: Zhao, Yakui <yakui.z...@intel.com>; libva@lists.freedesktop.org Subject: Re: [Libva] [PATCH V1][libva-intel-driver] Return if the surface bo is NULL On Fri, 2016-12-30 at 16:18 +0800, Zhao Yakui wrote: > On 12/30/2016 04:08 PM, peng.chen wrote: > > Signed-off-by: peng.chen<peng.c.c...@intel.com> > > In fact the dri_bo_unreference function can handle the scenario that > bo is NULL. Yes, it is like as free() with NULL pointer. > > Of course it also looks good to me that the explicit check is added in > the i965_destroy_surface_storage. I think the original code is clean and prefer the original code. > > Add: Reviewed-by: Zhao Yakui <yakui.z...@intel.com> > > Thanks > > --- > > src/i965_drv_video.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/src/i965_drv_video.c b/src/i965_drv_video.c index > > bf59374..ded3d93 100644 > > --- a/src/i965_drv_video.c > > +++ b/src/i965_drv_video.c > > @@ -1210,7 +1210,7 @@ VAStatus > > i965_QueryConfigAttributes(VADriverContextP ctx, > > void > > i965_destroy_surface_storage(struct object_surface *obj_surface) > > { > > - if (!obj_surface) > > + if (!obj_surface || !obj_surface->bo) > > return; > > > > dri_bo_unreference(obj_surface->bo); > > _______________________________________________ > Libva mailing list > Libva@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/libva _______________________________________________ Libva mailing list Libva@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/libva _______________________________________________ Libva mailing list Libva@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/libva