Skip to content

GitLab

"...composable_kernel.git" did not exist on "823657ed120144943b7db87c07fe3e647128db56"
Unverified Commit 5fc1bee4 authored by Bartłomiej Kocot's avatar Bartłomiej Kocot Committed by GitHub
Browse files

Fix nhwgc f16 wmma instances (#1328)

parent ce66277a
Hide whitespace changes
Inline Side-by-side
Showing with 11 additions and 0 deletions
library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
View file @ 5fc1bee4
...@@ -402,6 +402,17 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe ...@@ -402,6 +402,17 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe
if constexpr(NumDimSpatial == 2 && is_same_v<InLayout, NHWGC> && if constexpr(NumDimSpatial == 2 && is_same_v<InLayout, NHWGC> &&
is_same_v<WeiLayout, GKYXC> && is_same_v<OutLayout, NHWGK>) is_same_v<WeiLayout, GKYXC> && is_same_v<OutLayout, NHWGK>)
{ {
#ifdef CK_ENABLE_FP16
if constexpr(is_same_v<InDataType, half_t> && is_same_v<WeiDataType, half_t> &&
is_same_v<OutDataType, half_t> && is_same_v<AComputeType, half_t> &&
is_same_v<BComputeType, half_t>)
{
add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_instances(op_ptrs);
add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1p0_instances(op_ptrs);
add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1s1p0_instances(op_ptrs);
add_device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instances(op_ptrs);
}
#endif
#ifdef CK_ENABLE_INT8 #ifdef CK_ENABLE_INT8
if constexpr(is_same_v<InDataType, int8_t> && is_same_v<WeiDataType, int8_t> && if constexpr(is_same_v<InDataType, int8_t> && is_same_v<WeiDataType, int8_t> &&
is_same_v<OutDataType, int8_t> && is_same_v<AComputeType, int8_t> && is_same_v<OutDataType, int8_t> && is_same_v<AComputeType, int8_t> &&
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment