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/pool_fwd/device_avg_pool3d_fwd_ndhwc_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/pool_fwd_instance_common.hpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/pool_fwd_instance_common.hpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/quantization/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/quantization/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/quantization/conv2d_fwd/device_conv2d_dl_int8_instance.hpp
+++ b/library/src/tensor_operation_instance/gpu/quantization/conv2d_fwd/device_conv2d_dl_int8_instance.hpp
--- a/library/src/tensor_operation_instance/gpu/softmax/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/softmax/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank3_reduce1.cpp
+++ b/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank3_reduce1.cpp
--- a/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank3_reduce2.cpp
+++ b/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank3_reduce2.cpp
--- a/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank3_reduce3.cpp
+++ b/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank3_reduce3.cpp
--- a/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank4_reduce1.cpp
+++ b/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank4_reduce1.cpp
--- a/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank4_reduce2.cpp
+++ b/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank4_reduce2.cpp
--- a/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank4_reduce3.cpp
+++ b/library/src/tensor_operation_instance/gpu/softmax/device_softmax_f16_f16_instance_rank4_reduce3.cpp