Merge pull request #834 from ROCmSoftwarePlatform/mha-train-develop-tinyfix

Remove un-necessary including in the FlashAttention device operators

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_kloop_v1.hpp

@@ -21,8 +21,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_kloop_v2.hpp

@@ -20,8 +20,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_light_v1.hpp

@@ -10,7 +10,6 @@
 #include "ck/utility/philox_rand.hpp"
 #include "ck/tensor_description/tensor_descriptor.hpp"
 #include "ck/tensor_description/tensor_descriptor_helper.hpp"
-// #include "ck/tensor_operation/gpu/device/device_batched_multihead_attention_backward.hpp" // TODO
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/masking_specialization.hpp"
@@ -22,8 +21,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_light_v2.hpp

@@ -21,8 +21,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_v1.hpp

@@ -10,7 +10,6 @@
 #include "ck/utility/philox_rand.hpp"
 #include "ck/tensor_description/tensor_descriptor.hpp"
 #include "ck/tensor_description/tensor_descriptor_helper.hpp"
-// #include "ck/tensor_operation/gpu/device/device_batched_multihead_attention_backward.hpp" // TODO
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/masking_specialization.hpp"
@@ -21,8 +20,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_v2.hpp

@@ -20,8 +20,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_multiple_r_xdl_cshuffle.hpp

@@ -22,7 +22,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 #include "ck/host_utility/io.hpp"
-#include "ck/library/utility/numeric.hpp"
 
 namespace ck {
 namespace tensor_operation {

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_kloop_v1.hpp

@@ -10,7 +10,6 @@
 #include "ck/utility/philox_rand.hpp"
 #include "ck/tensor_description/tensor_descriptor.hpp"
 #include "ck/tensor_description/tensor_descriptor_helper.hpp"
-// #include "ck/tensor_operation/gpu/device/device_batched_multihead_attention_backward.hpp" // TODO
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/masking_specialization.hpp"
@@ -21,8 +20,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_kloop_v2.hpp

@@ -10,7 +10,6 @@
 #include "ck/utility/philox_rand.hpp"
 #include "ck/tensor_description/tensor_descriptor.hpp"
 #include "ck/tensor_description/tensor_descriptor_helper.hpp"
-// #include "ck/tensor_operation/gpu/device/device_batched_multihead_attention_backward.hpp" // TODO
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/masking_specialization.hpp"
@@ -21,8 +20,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_light_v1.hpp

@@ -22,8 +22,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_light_v2.hpp

@@ -10,7 +10,6 @@
 #include "ck/utility/philox_rand.hpp"
 #include "ck/tensor_description/tensor_descriptor.hpp"
 #include "ck/tensor_description/tensor_descriptor_helper.hpp"
-// #include "ck/tensor_operation/gpu/device/device_batched_multihead_attention_backward.hpp" // TODO
 #include "ck/tensor_operation/gpu/device/device_base.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 #include "ck/tensor_operation/gpu/device/masking_specialization.hpp"
@@ -22,8 +21,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_v1.hpp

@@ -21,8 +21,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_v2.hpp

@@ -21,8 +21,6 @@
 #include "ck/host_utility/device_prop.hpp"
 #include "ck/host_utility/kernel_launch.hpp"
 
-#include "ck/library/utility/host_tensor.hpp"
-
 namespace ck {
 namespace tensor_operation {
 namespace device {

library/include/ck/library/utility/fill.hpp

@@ -10,6 +10,7 @@
 #include <type_traits>
 #include <utility>
 
+#include "ck/utility/type_convert.hpp"
 #include "ck/utility/data_type.hpp"
 
 namespace ck {