[Beignet] [PATCH] runtime: fix one bug in BDW image.
He Junyan
junyan.he at inbox.com
Tue Nov 11 23:16:16 PST 2014
It's good for me
On δΈ‰, 2014-11-12 at 14:12 +0800, Zhigang Gong wrote:
> As we still have the image 1d array workaround, we need to
> fix it for BDW as well.
>
> Signed-off-by: Zhigang Gong <zhigang.gong at intel.com>
> ---
> src/intel/intel_gpgpu.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/src/intel/intel_gpgpu.c b/src/intel/intel_gpgpu.c
> index c6ea17f..b6e19db 100644
> --- a/src/intel/intel_gpgpu.c
> +++ b/src/intel/intel_gpgpu.c
> @@ -1028,8 +1028,10 @@ intel_get_surface_type(cl_mem_object_type type)
> static uint32_t get_surface_type(intel_gpgpu_t *gpgpu, int index, cl_mem_object_type type)
> {
> uint32_t surface_type;
> - if (((IS_IVYBRIDGE(gpgpu->drv->device_id) || IS_HASWELL(gpgpu->drv->device_id))) &&
> - index >= 128 + BTI_RESERVED_NUM &&
> + if (((IS_IVYBRIDGE(gpgpu->drv->device_id) ||
> + IS_HASWELL(gpgpu->drv->device_id) ||
> + IS_BROADWELL(gpgpu->drv->device_id))) &&
> + index >= BTI_MAX_IMAGE_NUM + BTI_RESERVED_NUM &&
> type == CL_MEM_OBJECT_IMAGE1D_ARRAY)
> surface_type = I965_SURFACE_2D;
> else
More information about the Beignet
mailing list