Merge remote-tracking branch 'origin/develop' into migx-flash-attn

4100d1d8 · Alan Turner · 48717006 · c8a8385f · 4100d1d8 · 4100d1d8
Commit 4100d1d8 authored Aug 23, 2023 by Alan Turner
20 changed files
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_data/device_grouped_conv2d_bwd_data_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_bwd_weight/device_grouped_conv2d_bwd_weight_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_instance.hpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_xdl_instance.hpp