mirror of
https://anongit.gentoo.org/git/repo/gentoo.git
synced 2025-06-13 00:28:15 +02:00
Closes: https://github.com/gentoo/gentoo/pull/39854 Closes: https://github.com/gentoo/gentoo/pull/40320 Closes: https://bugs.gentoo.org/950261 Signed-off-by: Paul Zander <negril.nx+gentoo@gmail.com> Closes: https://github.com/gentoo/gentoo/pull/40390 Signed-off-by: Sam James <sam@gentoo.org>
199 lines
13 KiB
Diff
199 lines
13 KiB
Diff
diff --git a/devices/hip/CMakeLists.txt b/devices/hip/CMakeLists.txt
|
|
index e9a9bec..b738d5b 100644
|
|
--- a/devices/hip/CMakeLists.txt
|
|
+++ b/devices/hip/CMakeLists.txt
|
|
@@ -53,12 +53,6 @@ set_target_properties(OpenImageDenoise_device_hip PROPERTIES
|
|
CXX_STANDARD 17
|
|
)
|
|
|
|
-target_include_directories(OpenImageDenoise_device_hip
|
|
- PRIVATE
|
|
- "${OIDN_ROOT_SOURCE_DIR}/external/composable_kernel/include"
|
|
- "${OIDN_ROOT_SOURCE_DIR}/external/composable_kernel/library/include"
|
|
-)
|
|
-
|
|
set_source_files_properties(
|
|
ck_conv_wmma.cpp
|
|
PROPERTIES COMPILE_FLAGS "-mcumode -mno-wavefrontsize64" # Navi3x optimizations
|
|
diff --git a/devices/hip/ck_conv_dl.cpp b/devices/hip/ck_conv_dl.cpp
|
|
index 1aacfe8..e99da53 100644
|
|
--- a/devices/hip/ck_conv_dl.cpp
|
|
+++ b/devices/hip/ck_conv_dl.cpp
|
|
@@ -3,7 +3,7 @@
|
|
|
|
#include "hip_conv.h"
|
|
#include "ck_conv.h"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
OIDN_NAMESPACE_BEGIN
|
|
|
|
diff --git a/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp16.cpp b/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp16.cpp
|
|
index db5a7f0..2f384cb 100644
|
|
--- a/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp16.cpp
|
|
+++ b/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp16.cpp
|
|
@@ -3,7 +3,7 @@
|
|
|
|
#include "convnd_fwd_dl_common.hpp"
|
|
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
#include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
|
|
|
|
diff --git a/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp32.cpp b/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp32.cpp
|
|
index 964d784..aa56666 100644
|
|
--- a/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp32.cpp
|
|
+++ b/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_fp32.cpp
|
|
@@ -3,7 +3,7 @@
|
|
|
|
#include "convnd_fwd_dl_common.hpp"
|
|
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
#include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
|
|
|
|
diff --git a/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_int8.cpp b/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_int8.cpp
|
|
index b0cd88f..15b554c 100644
|
|
--- a/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_int8.cpp
|
|
+++ b/external/composable_kernel/example/09_convnd_fwd/convnd_fwd_dl_int8.cpp
|
|
@@ -3,7 +3,7 @@
|
|
|
|
#include "convnd_fwd_dl_common.hpp"
|
|
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
#include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
|
|
|
|
diff --git a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perchannel_quantization_int8.cpp b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perchannel_quantization_int8.cpp
|
|
index 5c445d9..49d2ba3 100644
|
|
--- a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perchannel_quantization_int8.cpp
|
|
+++ b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perchannel_quantization_int8.cpp
|
|
@@ -2,7 +2,7 @@
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include "common.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
using InDataType = int8_t;
|
|
using WeiDataType = int8_t;
|
|
diff --git a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perlayer_quantization_int8.cpp b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perlayer_quantization_int8.cpp
|
|
index 0ff85f0..661d744 100644
|
|
--- a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perlayer_quantization_int8.cpp
|
|
+++ b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_relu_perlayer_quantization_int8.cpp
|
|
@@ -2,7 +2,7 @@
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include "common.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
using InDataType = int8_t;
|
|
using WeiDataType = int8_t;
|
|
diff --git a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perchannel_quantization_int8.cpp b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perchannel_quantization_int8.cpp
|
|
index f8f996d..65cded3 100644
|
|
--- a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perchannel_quantization_int8.cpp
|
|
+++ b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perchannel_quantization_int8.cpp
|
|
@@ -2,7 +2,7 @@
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include "common.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
using InDataType = int8_t;
|
|
using WeiDataType = int8_t;
|
|
diff --git a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perlayer_quantization_int8.cpp b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perlayer_quantization_int8.cpp
|
|
index 3b25fec..ca5f217 100644
|
|
--- a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perlayer_quantization_int8.cpp
|
|
+++ b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_bias_tanh_perlayer_quantization_int8.cpp
|
|
@@ -2,7 +2,7 @@
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include "common.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
using InDataType = int8_t;
|
|
using WeiDataType = int8_t;
|
|
diff --git a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perchannel_quantization_int8.cpp b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perchannel_quantization_int8.cpp
|
|
index a98a1e2..88b9ccc 100644
|
|
--- a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perchannel_quantization_int8.cpp
|
|
+++ b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perchannel_quantization_int8.cpp
|
|
@@ -2,7 +2,7 @@
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include "common.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
using InDataType = int8_t;
|
|
using WeiDataType = int8_t;
|
|
diff --git a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perlayer_quantization_int8.cpp b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perlayer_quantization_int8.cpp
|
|
index 262594d..f5678e7 100644
|
|
--- a/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perlayer_quantization_int8.cpp
|
|
+++ b/external/composable_kernel/example/40_conv2d_fwd_quantization/conv2d_fwd_dl_perlayer_quantization_int8.cpp
|
|
@@ -2,7 +2,7 @@
|
|
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include "common.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
using InDataType = int8_t;
|
|
using WeiDataType = int8_t;
|
|
diff --git a/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp b/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
|
|
index fc18b3c..020d93e 100644
|
|
--- a/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
|
|
+++ b/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
|
|
@@ -5,7 +5,7 @@
|
|
|
|
#include "ck/ck.hpp"
|
|
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
|
|
#include "ck/tensor_operation/gpu/device/convolution_forward_specialization.hpp"
|
|
#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
|
|
diff --git a/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp b/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
|
|
index 648b396..42825ce 100644
|
|
--- a/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
|
|
+++ b/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
|
|
@@ -5,7 +5,7 @@
|
|
|
|
#include "ck/ck.hpp"
|
|
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
|
|
#include "ck/tensor_operation/gpu/device/convolution_forward_specialization.hpp"
|
|
#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
|
|
diff --git a/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_int8_instance.cpp b/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_int8_instance.cpp
|
|
index 1cb5d06..ab74235 100644
|
|
--- a/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_int8_instance.cpp
|
|
+++ b/external/composable_kernel/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_int8_instance.cpp
|
|
@@ -5,7 +5,7 @@
|
|
|
|
#include "ck/ck.hpp"
|
|
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
|
|
#include "ck/tensor_operation/gpu/device/convolution_forward_specialization.hpp"
|
|
#include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
|
|
diff --git a/external/composable_kernel/library/src/tensor_operation_instance/gpu/quantization/conv2d_fwd/device_conv2d_dl_int8_instance.hpp b/external/composable_kernel/library/src/tensor_operation_instance/gpu/quantization/conv2d_fwd/device_conv2d_dl_int8_instance.hpp
|
|
index 3c4987f..62ee1a4 100644
|
|
--- a/external/composable_kernel/library/src/tensor_operation_instance/gpu/quantization/conv2d_fwd/device_conv2d_dl_int8_instance.hpp
|
|
+++ b/external/composable_kernel/library/src/tensor_operation_instance/gpu/quantization/conv2d_fwd/device_conv2d_dl_int8_instance.hpp
|
|
@@ -4,7 +4,7 @@
|
|
#pragma once
|
|
|
|
#include "conv2d_quantization_common.hpp"
|
|
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_dl_multiple_d_nhwc_kyxc_nhwk.hpp"
|
|
|
|
namespace ck {
|
|
namespace tensor_operation {
|