Merge branch 'uif2-initial' into uif2-migraphx

client_example/07_grouped_convnd_fwd/grouped_conv1d_fwd.cpp

@@ -100,7 +100,7 @@ int main()
     SimpleDeviceMem wei(sizeof(WeiDataType) * G * K * X * C);
     SimpleDeviceMem out(sizeof(OutDataType) * G * N * Wo * K);
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<NumDimSpatial,
                                                                                    InLayout,
                                                                                    WeiLayout,
                                                                                    ck::Tuple<>,

client_example/07_grouped_convnd_fwd/grouped_conv2d_fwd.cpp

@@ -71,7 +71,7 @@ int main()
     SimpleDeviceMem wei(sizeof(WeiDataType) * G * K * Y * X * C);
     SimpleDeviceMem out(sizeof(OutDataType) * N * Ho * Wo * G * K);
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<NumDimSpatial,
                                                                                    InLayout,
                                                                                    WeiLayout,
                                                                                    ck::Tuple<>,

client_example/09_quantization/conv2d_fwd_bias_relu_perchannel_quantization.cpp

@@ -80,7 +80,7 @@ int main(int argc, char* argv[])
     SimpleDeviceMem requant_scale(sizeof(RequantScaleDataType) * G * K);
     SimpleDeviceMem out(sizeof(OutDataType) * N * Ho * Wo * G * K);
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<
         NumDimSpatial,
         InLayout,
         WeiLayout,

client_example/09_quantization/conv2d_fwd_bias_relu_perlayer_quantization.cpp

@@ -78,7 +78,7 @@ int main(int argc, char* argv[])
     SimpleDeviceMem out(sizeof(OutDataType) * N * Ho * Wo * G * K);
 
     using DeviceOp =
-        ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+        ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<NumDimSpatial,
                                                                       InLayout,
                                                                       WeiLayout,
                                                                       ck::Tuple<BiasLayout>,

client_example/09_quantization/conv2d_fwd_bias_tanh_perchannel_quantization.cpp

@@ -83,7 +83,7 @@ int main(int argc, char* argv[])
     SimpleDeviceMem requant_scale(sizeof(RequantScaleDataType) * G * K);
     SimpleDeviceMem out(sizeof(OutDataType) * N * Ho * Wo * G * K);
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<
         NumDimSpatial,
         InLayout,
         WeiLayout,

client_example/09_quantization/conv2d_fwd_bias_tanh_perlayer_quantization.cpp

@@ -79,7 +79,7 @@ int main(int argc, char* argv[])
     SimpleDeviceMem out(sizeof(OutDataType) * N * Ho * Wo * G * K);
 
     using DeviceOp =
-        ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+        ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<NumDimSpatial,
                                                                       InLayout,
                                                                       WeiLayout,
                                                                       ck::Tuple<BiasLayout>,

client_example/09_quantization/conv2d_fwd_perchannel_quantization.cpp

@@ -76,8 +76,8 @@ int main(int argc, char* argv[])
     SimpleDeviceMem requant_scale(sizeof(RequantScaleDataType) * G * K);
     SimpleDeviceMem out(sizeof(OutDataType) * N * Ho * Wo * G * K);
 
-    using DeviceOp =
-        ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<
+        NumDimSpatial,
         InLayout,
         WeiLayout,
         ck::Tuple<RequantScaleLayout>,

client_example/09_quantization/conv2d_fwd_perlayer_quantization.cpp

@@ -72,7 +72,7 @@ int main(int argc, char* argv[])
     SimpleDeviceMem wei(sizeof(WeiDataType) * G * K * Y * X * C);
     SimpleDeviceMem out(sizeof(OutDataType) * N * Ho * Wo * G * K);
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<NumDimSpatial,
                                                                                    InLayout,
                                                                                    WeiLayout,
                                                                                    ck::Tuple<>,

client_example/16_convnd_fwd/common.hpp

@@ -11,7 +11,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp"
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_d.hpp"
+#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_abd.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 using PassThrough = ck::tensor_operation::element_wise::PassThrough;
@@ -174,7 +174,7 @@ bool run_grouped_conv_fwd(std::array<ck::index_t, NumDimSpatial + NumNonSpatialD
     std::size_t flop      = GetFlops<NumDimSpatial>(out_lengths, wei_lengths);
     std::size_t num_bytes = in_mem_size + wei_mem_size + out_mem_size;
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<NumDimSpatial,
                                                                                    InLayout,
                                                                                    WeiLayout,
                                                                                    ck::Tuple<>,

client_example/23_grouped_convnd_fwd_scaleadd_scaleadd_relu/grouped_conv_fwd_scaleadd_scaleadd_relu.inc

@@ -76,7 +76,7 @@ int execute_conv_fwd_scaleadd_scaleadd_relu()
     SimpleDeviceMem d0(sizeof(std::tuple_element_t<0, DDataTypes>) * N * Do * Ho * Wo * G * K);
     SimpleDeviceMem d1(sizeof(std::tuple_element_t<1, DDataTypes>) * N * Do * Ho * Wo * G * K);
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<
         NumDimSpatial,
         InLayout,
         WeiLayout,

client_example/24_grouped_convnd_fwd_scaleadd_ab/grouped_conv_fwd_scaleadd_ab.inc

@@ -86,7 +86,7 @@ int execute_conv_fwd_scaleadd_ab()
     SimpleDeviceMem wei_bias(sizeof(WeightBiasDtype) * G * K * Z * Y * X * C);
     SimpleDeviceMem out(sizeof(OutDataType) * N * Do * Ho * Wo * G * K);
 
-    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD<NumDimSpatial,
+    using DeviceOp = ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD<NumDimSpatial,
                                                                                    InLayout,
                                                                                    WeiLayout,
                                                                                    ck::Tuple<>,

docs/sphinx/requirements.txt

@@ -48,6 +48,12 @@ idna==3.4
     # via requests
 imagesize==1.4.1
     # via sphinx
+importlib-metadata==6.8.0
+    # via
+    #   sphinx
+    #   sphinxcontrib-bibtex
+importlib-resources==6.1.0
+    # via rocm-docs-core
 jinja2==3.1.2
     # via
     #   myst-parser
@@ -90,9 +96,13 @@ pygments==2.14.0
     #   pydata-sphinx-theme
     #   sphinx
 pyjwt[crypto]==2.6.0
-    # via pygithub
+    # via
+    #   pygithub
+    #   pyjwt
 pynacl==1.5.0
     # via pygithub
+pytz==2023.3.post1
+    # via babel
 pyyaml==6.0
     # via
     #   myst-parser
@@ -103,7 +113,7 @@ requests==2.28.2
     # via
     #   pygithub
     #   sphinx
-rocm-docs-core==0.26.0
+rocm-docs-core==0.27.0
     # via -r requirements.in
 six==1.16.0
     # via
@@ -157,3 +167,7 @@ urllib3==1.26.15
     # via requests
 wrapt==1.15.0
     # via deprecated
+zipp==3.17.0
+    # via
+    #   importlib-metadata
+    #   importlib-resources

example/09_convnd_fwd/convnd_fwd_xdl_bf16.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 
@@ -27,7 +27,7 @@ static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecializatio
 
 template <ck::index_t NDimSpatial, typename InLayout, typename WeiLayout, typename OutLayout>
 using DeviceGroupedConvNDFwdInstance =
-    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD_Xdl_CShuffle<
+    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<
         NDimSpatial,
         InLayout,
         WeiLayout,

example/09_convnd_fwd/convnd_fwd_xdl_fp16.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 
@@ -27,7 +27,7 @@ static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecializatio
 
 template <ck::index_t NDimSpatial, typename InLayout, typename WeiLayout, typename OutLayout>
 using DeviceGroupedConvNDFwdInstance =
-    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD_Xdl_CShuffle<
+    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<
         NDimSpatial,
         InLayout,
         WeiLayout,

example/09_convnd_fwd/convnd_fwd_xdl_fp32.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 
@@ -27,7 +27,7 @@ static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecializatio
 
 template <ck::index_t NDimSpatial, typename InLayout, typename WeiLayout, typename OutLayout>
 using DeviceGroupedConvNDFwdInstance =
-    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD_Xdl_CShuffle<
+    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<
         NDimSpatial,
         InLayout,
         WeiLayout,

example/09_convnd_fwd/convnd_fwd_xdl_fp64.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 
@@ -27,7 +27,7 @@ static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecializatio
 
 template <ck::index_t NDimSpatial, typename InLayout, typename WeiLayout, typename OutLayout>
 using DeviceGroupedConvNDFwdInstance =
-    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD_Xdl_CShuffle<
+    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<
         NDimSpatial,
         InLayout,
         WeiLayout,

example/09_convnd_fwd/convnd_fwd_xdl_int8.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 
@@ -27,7 +27,7 @@ static constexpr auto GemmSpec = ck::tensor_operation::device::GemmSpecializatio
 
 template <ck::index_t NDimSpatial, typename InLayout, typename WeiLayout, typename OutLayout>
 using DeviceGroupedConvNDFwdInstance =
-    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD_Xdl_CShuffle<
+    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<
         NDimSpatial,
         InLayout,
         WeiLayout,

example/30_grouped_conv_fwd_multiple_d/README.md

@@ -26,5 +26,5 @@ out: dim 5, lengths {1, 128, 256, 36, 36}, strides {256, 331776, 1, 9216, 256}
 launch_and_time_kernel: grid_dim {1296, 1, 1}, block_dim {256, 1, 1}
 Warm up 1 time
 Start running 10 times...
-Perf: 1.55981 ms, 94.0927 TFlops, 213.868 GB/s, DeviceGroupedConvFwdMultipleD_Xdl_CShuffle<256, 128, 256, 16, Default>
+Perf: 1.55981 ms, 94.0927 TFlops, 213.868 GB/s, DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<256, 128, 256, 16, Default>
 ```

example/30_grouped_conv_fwd_multiple_d/common.hpp

@@ -12,7 +12,7 @@
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/convolution_forward_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_abd_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 

example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_bias_relu_add_example.inc

@@ -34,7 +34,7 @@ using ResidualLayout = typename LayoutSettingSelector<NDimSpatial>::ResidualLayo
 
 template <ck::index_t NDimSpatial>
 using DeviceConvFwdInstance =
-    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleD_Xdl_CShuffle<
+    ck::tensor_operation::device::DeviceGroupedConvFwdMultipleABD_Xdl_CShuffle<
         NDimSpatial,
         InputLayout<NDimSpatial>,
         WeightLayout<NDimSpatial>,