diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cublas/include/__init__.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cublas/include/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/CL/cl.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/CL/cl.h new file mode 100644 index 0000000000000000000000000000000000000000..290495a68f1be43916c468b9a3881c6972085e89 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/CL/cl.h @@ -0,0 +1,1578 @@ +/******************************************************************************* + * Copyright (c) 2008-2020 The Khronos Group Inc. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + ******************************************************************************/ + +#ifndef __OPENCL_CL_H +#define __OPENCL_CL_H + +#ifdef __APPLE__ +#include +#else +#include +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +/******************************************************************************/ + +typedef struct _cl_platform_id * cl_platform_id; +typedef struct _cl_device_id * cl_device_id; +typedef struct _cl_context * cl_context; +typedef struct _cl_command_queue * cl_command_queue; +typedef struct _cl_mem * cl_mem; +typedef struct _cl_program * cl_program; +typedef struct _cl_kernel * cl_kernel; +typedef struct _cl_event * cl_event; +typedef struct _cl_sampler * cl_sampler; + +typedef cl_uint cl_bool; /* WARNING! Unlike cl_ types in cl_platform.h, cl_bool is not guaranteed to be the same size as the bool in kernels. */ +typedef cl_ulong cl_bitfield; +typedef cl_ulong cl_properties; +typedef cl_bitfield cl_device_type; +typedef cl_uint cl_platform_info; +typedef cl_uint cl_device_info; +typedef cl_bitfield cl_device_fp_config; +typedef cl_uint cl_device_mem_cache_type; +typedef cl_uint cl_device_local_mem_type; +typedef cl_bitfield cl_device_exec_capabilities; +typedef cl_bitfield cl_device_svm_capabilities; +typedef cl_bitfield cl_command_queue_properties; +typedef intptr_t cl_device_partition_property; +typedef cl_bitfield cl_device_affinity_domain; + +typedef intptr_t cl_context_properties; +typedef cl_uint cl_context_info; +typedef cl_properties cl_queue_properties; +typedef cl_uint cl_command_queue_info; +typedef cl_uint cl_channel_order; +typedef cl_uint cl_channel_type; +typedef cl_bitfield cl_mem_flags; +typedef cl_bitfield cl_svm_mem_flags; +typedef cl_uint cl_mem_object_type; +typedef cl_uint cl_mem_info; +typedef cl_bitfield cl_mem_migration_flags; +typedef cl_uint cl_image_info; +typedef cl_uint cl_buffer_create_type; +typedef cl_uint cl_addressing_mode; +typedef cl_uint cl_filter_mode; +typedef cl_uint cl_sampler_info; +typedef cl_bitfield cl_map_flags; +typedef intptr_t cl_pipe_properties; +typedef cl_uint cl_pipe_info; +typedef cl_uint cl_program_info; +typedef cl_uint cl_program_build_info; +typedef cl_uint cl_program_binary_type; +typedef cl_int cl_build_status; +typedef cl_uint cl_kernel_info; +typedef cl_uint cl_kernel_arg_info; +typedef cl_uint cl_kernel_arg_address_qualifier; +typedef cl_uint cl_kernel_arg_access_qualifier; +typedef cl_bitfield cl_kernel_arg_type_qualifier; +typedef cl_uint cl_kernel_work_group_info; +typedef cl_uint cl_kernel_sub_group_info; +typedef cl_uint cl_event_info; +typedef cl_uint cl_command_type; +typedef cl_uint cl_profiling_info; +typedef cl_properties cl_sampler_properties; +typedef cl_uint cl_kernel_exec_info; +typedef cl_bitfield cl_device_atomic_capabilities; +typedef cl_bitfield cl_device_device_enqueue_capabilities; +typedef cl_uint cl_khronos_vendor_id; +typedef cl_properties cl_mem_properties; +typedef cl_uint cl_version; + +typedef struct _cl_image_format { + cl_channel_order image_channel_order; + cl_channel_type image_channel_data_type; +} cl_image_format; + +typedef struct _cl_image_desc { + cl_mem_object_type image_type; + size_t image_width; + size_t image_height; + size_t image_depth; + size_t image_array_size; + size_t image_row_pitch; + size_t image_slice_pitch; + cl_uint num_mip_levels; + cl_uint num_samples; +#ifdef __GNUC__ + __extension__ /* Prevents warnings about anonymous union in -pedantic builds */ +#endif + union { + cl_mem buffer; + cl_mem mem_object; + }; +} cl_image_desc; + +typedef struct _cl_buffer_region { + size_t origin; + size_t size; +} cl_buffer_region; + +#define CL_NAME_VERSION_MAX_NAME_SIZE 64 + +typedef struct _cl_name_version { + cl_version version; + char name[CL_NAME_VERSION_MAX_NAME_SIZE]; +} cl_name_version; + +/******************************************************************************/ + +/* Error Codes */ +#define CL_SUCCESS 0 +#define CL_DEVICE_NOT_FOUND -1 +#define CL_DEVICE_NOT_AVAILABLE -2 +#define CL_COMPILER_NOT_AVAILABLE -3 +#define CL_MEM_OBJECT_ALLOCATION_FAILURE -4 +#define CL_OUT_OF_RESOURCES -5 +#define CL_OUT_OF_HOST_MEMORY -6 +#define CL_PROFILING_INFO_NOT_AVAILABLE -7 +#define CL_MEM_COPY_OVERLAP -8 +#define CL_IMAGE_FORMAT_MISMATCH -9 +#define CL_IMAGE_FORMAT_NOT_SUPPORTED -10 +#define CL_BUILD_PROGRAM_FAILURE -11 +#define CL_MAP_FAILURE -12 +#define CL_MISALIGNED_SUB_BUFFER_OFFSET -13 +#define CL_EXEC_STATUS_ERROR_FOR_EVENTS_IN_WAIT_LIST -14 +#define CL_COMPILE_PROGRAM_FAILURE -15 +#define CL_LINKER_NOT_AVAILABLE -16 +#define CL_LINK_PROGRAM_FAILURE -17 +#define CL_DEVICE_PARTITION_FAILED -18 +#define CL_KERNEL_ARG_INFO_NOT_AVAILABLE -19 + +#define CL_INVALID_VALUE -30 +#define CL_INVALID_DEVICE_TYPE -31 +#define CL_INVALID_PLATFORM -32 +#define CL_INVALID_DEVICE -33 +#define CL_INVALID_CONTEXT -34 +#define CL_INVALID_QUEUE_PROPERTIES -35 +#define CL_INVALID_COMMAND_QUEUE -36 +#define CL_INVALID_HOST_PTR -37 +#define CL_INVALID_MEM_OBJECT -38 +#define CL_INVALID_IMAGE_FORMAT_DESCRIPTOR -39 +#define CL_INVALID_IMAGE_SIZE -40 +#define CL_INVALID_SAMPLER -41 +#define CL_INVALID_BINARY -42 +#define CL_INVALID_BUILD_OPTIONS -43 +#define CL_INVALID_PROGRAM -44 +#define CL_INVALID_PROGRAM_EXECUTABLE -45 +#define CL_INVALID_KERNEL_NAME -46 +#define CL_INVALID_KERNEL_DEFINITION -47 +#define CL_INVALID_KERNEL -48 +#define CL_INVALID_ARG_INDEX -49 +#define CL_INVALID_ARG_VALUE -50 +#define CL_INVALID_ARG_SIZE -51 +#define CL_INVALID_KERNEL_ARGS -52 +#define CL_INVALID_WORK_DIMENSION -53 +#define CL_INVALID_WORK_GROUP_SIZE -54 +#define CL_INVALID_WORK_ITEM_SIZE -55 +#define CL_INVALID_GLOBAL_OFFSET -56 +#define CL_INVALID_EVENT_WAIT_LIST -57 +#define CL_INVALID_EVENT -58 +#define CL_INVALID_OPERATION -59 +#define CL_INVALID_GL_OBJECT -60 +#define CL_INVALID_BUFFER_SIZE -61 +#define CL_INVALID_MIP_LEVEL -62 +#define CL_INVALID_GLOBAL_WORK_SIZE -63 +#define CL_INVALID_PROPERTY -64 +#define CL_INVALID_IMAGE_DESCRIPTOR -65 +#define CL_INVALID_COMPILER_OPTIONS -66 +#define CL_INVALID_LINKER_OPTIONS -67 +#define CL_INVALID_DEVICE_PARTITION_COUNT -68 +#define CL_INVALID_PIPE_SIZE -69 +#define CL_INVALID_DEVICE_QUEUE -70 +#define CL_INVALID_SPEC_ID -71 +#define CL_MAX_SIZE_RESTRICTION_EXCEEDED -72 + +/* OpenCL Version */ +#define CL_VERSION_1_0 1 +#define CL_VERSION_1_1 1 +#define CL_VERSION_1_2 1 +#define CL_VERSION_2_0 1 +#define CL_VERSION_2_1 1 +#define CL_VERSION_2_2 1 +#define CL_VERSION_3_0 1 + +/* cl_bool */ +#define CL_FALSE 0 +#define CL_TRUE 1 +#define CL_BLOCKING CL_TRUE +#define CL_NON_BLOCKING CL_FALSE + +/* cl_platform_info */ +#define CL_PLATFORM_PROFILE 0x0900 +#define CL_PLATFORM_VERSION 0x0901 +#define CL_PLATFORM_NAME 0x0902 +#define CL_PLATFORM_VENDOR 0x0903 +#define CL_PLATFORM_EXTENSIONS 0x0904 +#define CL_PLATFORM_HOST_TIMER_RESOLUTION 0x0905 +#define CL_PLATFORM_NUMERIC_VERSION 0x0906 +#define CL_PLATFORM_EXTENSIONS_WITH_VERSION 0x0907 + +/* cl_device_type - bitfield */ +#define CL_DEVICE_TYPE_DEFAULT (1 << 0) +#define CL_DEVICE_TYPE_CPU (1 << 1) +#define CL_DEVICE_TYPE_GPU (1 << 2) +#define CL_DEVICE_TYPE_ACCELERATOR (1 << 3) +#define CL_DEVICE_TYPE_CUSTOM (1 << 4) +#define CL_DEVICE_TYPE_ALL 0xFFFFFFFF + +/* cl_device_info */ +#define CL_DEVICE_TYPE 0x1000 +#define CL_DEVICE_VENDOR_ID 0x1001 +#define CL_DEVICE_MAX_COMPUTE_UNITS 0x1002 +#define CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS 0x1003 +#define CL_DEVICE_MAX_WORK_GROUP_SIZE 0x1004 +#define CL_DEVICE_MAX_WORK_ITEM_SIZES 0x1005 +#define CL_DEVICE_PREFERRED_VECTOR_WIDTH_CHAR 0x1006 +#define CL_DEVICE_PREFERRED_VECTOR_WIDTH_SHORT 0x1007 +#define CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT 0x1008 +#define CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG 0x1009 +#define CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT 0x100A +#define CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE 0x100B +#define CL_DEVICE_MAX_CLOCK_FREQUENCY 0x100C +#define CL_DEVICE_ADDRESS_BITS 0x100D +#define CL_DEVICE_MAX_READ_IMAGE_ARGS 0x100E +#define CL_DEVICE_MAX_WRITE_IMAGE_ARGS 0x100F +#define CL_DEVICE_MAX_MEM_ALLOC_SIZE 0x1010 +#define CL_DEVICE_IMAGE2D_MAX_WIDTH 0x1011 +#define CL_DEVICE_IMAGE2D_MAX_HEIGHT 0x1012 +#define CL_DEVICE_IMAGE3D_MAX_WIDTH 0x1013 +#define CL_DEVICE_IMAGE3D_MAX_HEIGHT 0x1014 +#define CL_DEVICE_IMAGE3D_MAX_DEPTH 0x1015 +#define CL_DEVICE_IMAGE_SUPPORT 0x1016 +#define CL_DEVICE_MAX_PARAMETER_SIZE 0x1017 +#define CL_DEVICE_MAX_SAMPLERS 0x1018 +#define CL_DEVICE_MEM_BASE_ADDR_ALIGN 0x1019 +#define CL_DEVICE_MIN_DATA_TYPE_ALIGN_SIZE 0x101A +#define CL_DEVICE_SINGLE_FP_CONFIG 0x101B +#define CL_DEVICE_GLOBAL_MEM_CACHE_TYPE 0x101C +#define CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE 0x101D +#define CL_DEVICE_GLOBAL_MEM_CACHE_SIZE 0x101E +#define CL_DEVICE_GLOBAL_MEM_SIZE 0x101F +#define CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE 0x1020 +#define CL_DEVICE_MAX_CONSTANT_ARGS 0x1021 +#define CL_DEVICE_LOCAL_MEM_TYPE 0x1022 +#define CL_DEVICE_LOCAL_MEM_SIZE 0x1023 +#define CL_DEVICE_ERROR_CORRECTION_SUPPORT 0x1024 +#define CL_DEVICE_PROFILING_TIMER_RESOLUTION 0x1025 +#define CL_DEVICE_ENDIAN_LITTLE 0x1026 +#define CL_DEVICE_AVAILABLE 0x1027 +#define CL_DEVICE_COMPILER_AVAILABLE 0x1028 +#define CL_DEVICE_EXECUTION_CAPABILITIES 0x1029 +#define CL_DEVICE_QUEUE_PROPERTIES 0x102A /* deprecated */ +#define CL_DEVICE_QUEUE_ON_HOST_PROPERTIES 0x102A +#define CL_DEVICE_NAME 0x102B +#define CL_DEVICE_VENDOR 0x102C +#define CL_DRIVER_VERSION 0x102D +#define CL_DEVICE_PROFILE 0x102E +#define CL_DEVICE_VERSION 0x102F +#define CL_DEVICE_EXTENSIONS 0x1030 +#define CL_DEVICE_PLATFORM 0x1031 +#define CL_DEVICE_DOUBLE_FP_CONFIG 0x1032 +/* 0x1033 reserved for CL_DEVICE_HALF_FP_CONFIG which is already defined in "cl_ext.h" */ +#define CL_DEVICE_PREFERRED_VECTOR_WIDTH_HALF 0x1034 +#define CL_DEVICE_HOST_UNIFIED_MEMORY 0x1035 /* deprecated */ +#define CL_DEVICE_NATIVE_VECTOR_WIDTH_CHAR 0x1036 +#define CL_DEVICE_NATIVE_VECTOR_WIDTH_SHORT 0x1037 +#define CL_DEVICE_NATIVE_VECTOR_WIDTH_INT 0x1038 +#define CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG 0x1039 +#define CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT 0x103A +#define CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE 0x103B +#define CL_DEVICE_NATIVE_VECTOR_WIDTH_HALF 0x103C +#define CL_DEVICE_OPENCL_C_VERSION 0x103D +#define CL_DEVICE_LINKER_AVAILABLE 0x103E +#define CL_DEVICE_BUILT_IN_KERNELS 0x103F +#define CL_DEVICE_IMAGE_MAX_BUFFER_SIZE 0x1040 +#define CL_DEVICE_IMAGE_MAX_ARRAY_SIZE 0x1041 +#define CL_DEVICE_PARENT_DEVICE 0x1042 +#define CL_DEVICE_PARTITION_MAX_SUB_DEVICES 0x1043 +#define CL_DEVICE_PARTITION_PROPERTIES 0x1044 +#define CL_DEVICE_PARTITION_AFFINITY_DOMAIN 0x1045 +#define CL_DEVICE_PARTITION_TYPE 0x1046 +#define CL_DEVICE_REFERENCE_COUNT 0x1047 +#define CL_DEVICE_PREFERRED_INTEROP_USER_SYNC 0x1048 +#define CL_DEVICE_PRINTF_BUFFER_SIZE 0x1049 +#define CL_DEVICE_IMAGE_PITCH_ALIGNMENT 0x104A +#define CL_DEVICE_IMAGE_BASE_ADDRESS_ALIGNMENT 0x104B +#define CL_DEVICE_MAX_READ_WRITE_IMAGE_ARGS 0x104C +#define CL_DEVICE_MAX_GLOBAL_VARIABLE_SIZE 0x104D +#define CL_DEVICE_QUEUE_ON_DEVICE_PROPERTIES 0x104E +#define CL_DEVICE_QUEUE_ON_DEVICE_PREFERRED_SIZE 0x104F +#define CL_DEVICE_QUEUE_ON_DEVICE_MAX_SIZE 0x1050 +#define CL_DEVICE_MAX_ON_DEVICE_QUEUES 0x1051 +#define CL_DEVICE_MAX_ON_DEVICE_EVENTS 0x1052 +#define CL_DEVICE_SVM_CAPABILITIES 0x1053 +#define CL_DEVICE_GLOBAL_VARIABLE_PREFERRED_TOTAL_SIZE 0x1054 +#define CL_DEVICE_MAX_PIPE_ARGS 0x1055 +#define CL_DEVICE_PIPE_MAX_ACTIVE_RESERVATIONS 0x1056 +#define CL_DEVICE_PIPE_MAX_PACKET_SIZE 0x1057 +#define CL_DEVICE_PREFERRED_PLATFORM_ATOMIC_ALIGNMENT 0x1058 +#define CL_DEVICE_PREFERRED_GLOBAL_ATOMIC_ALIGNMENT 0x1059 +#define CL_DEVICE_PREFERRED_LOCAL_ATOMIC_ALIGNMENT 0x105A +#define CL_DEVICE_IL_VERSION 0x105B +#define CL_DEVICE_MAX_NUM_SUB_GROUPS 0x105C +#define CL_DEVICE_SUB_GROUP_INDEPENDENT_FORWARD_PROGRESS 0x105D +#define CL_DEVICE_NUMERIC_VERSION 0x105E +#define CL_DEVICE_EXTENSIONS_WITH_VERSION 0x1060 +#define CL_DEVICE_ILS_WITH_VERSION 0x1061 +#define CL_DEVICE_BUILT_IN_KERNELS_WITH_VERSION 0x1062 +#define CL_DEVICE_ATOMIC_MEMORY_CAPABILITIES 0x1063 +#define CL_DEVICE_ATOMIC_FENCE_CAPABILITIES 0x1064 +#define CL_DEVICE_NON_UNIFORM_WORK_GROUP_SUPPORT 0x1065 +#define CL_DEVICE_OPENCL_C_ALL_VERSIONS 0x1066 +#define CL_DEVICE_PREFERRED_WORK_GROUP_SIZE_MULTIPLE 0x1067 +#define CL_DEVICE_WORK_GROUP_COLLECTIVE_FUNCTIONS_SUPPORT 0x1068 +#define CL_DEVICE_GENERIC_ADDRESS_SPACE_SUPPORT 0x1069 +/* 0x106A to 0x106E - Reserved for upcoming KHR extension */ +#define CL_DEVICE_OPENCL_C_FEATURES 0x106F +#define CL_DEVICE_DEVICE_ENQUEUE_CAPABILITIES 0x1070 +#define CL_DEVICE_PIPE_SUPPORT 0x1071 +#define CL_DEVICE_LATEST_CONFORMANCE_VERSION_PASSED 0x1072 + +/* cl_device_fp_config - bitfield */ +#define CL_FP_DENORM (1 << 0) +#define CL_FP_INF_NAN (1 << 1) +#define CL_FP_ROUND_TO_NEAREST (1 << 2) +#define CL_FP_ROUND_TO_ZERO (1 << 3) +#define CL_FP_ROUND_TO_INF (1 << 4) +#define CL_FP_FMA (1 << 5) +#define CL_FP_SOFT_FLOAT (1 << 6) +#define CL_FP_CORRECTLY_ROUNDED_DIVIDE_SQRT (1 << 7) + +/* cl_device_mem_cache_type */ +#define CL_NONE 0x0 +#define CL_READ_ONLY_CACHE 0x1 +#define CL_READ_WRITE_CACHE 0x2 + +/* cl_device_local_mem_type */ +#define CL_LOCAL 0x1 +#define CL_GLOBAL 0x2 + +/* cl_device_exec_capabilities - bitfield */ +#define CL_EXEC_KERNEL (1 << 0) +#define CL_EXEC_NATIVE_KERNEL (1 << 1) + +/* cl_command_queue_properties - bitfield */ +#define CL_QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE (1 << 0) +#define CL_QUEUE_PROFILING_ENABLE (1 << 1) +#define CL_QUEUE_ON_DEVICE (1 << 2) +#define CL_QUEUE_ON_DEVICE_DEFAULT (1 << 3) + +/* cl_context_info */ +#define CL_CONTEXT_REFERENCE_COUNT 0x1080 +#define CL_CONTEXT_DEVICES 0x1081 +#define CL_CONTEXT_PROPERTIES 0x1082 +#define CL_CONTEXT_NUM_DEVICES 0x1083 + +/* cl_context_properties */ +#define CL_CONTEXT_PLATFORM 0x1084 +#define CL_CONTEXT_INTEROP_USER_SYNC 0x1085 + +/* cl_device_partition_property */ +#define CL_DEVICE_PARTITION_EQUALLY 0x1086 +#define CL_DEVICE_PARTITION_BY_COUNTS 0x1087 +#define CL_DEVICE_PARTITION_BY_COUNTS_LIST_END 0x0 +#define CL_DEVICE_PARTITION_BY_AFFINITY_DOMAIN 0x1088 + +/* cl_device_affinity_domain */ +#define CL_DEVICE_AFFINITY_DOMAIN_NUMA (1 << 0) +#define CL_DEVICE_AFFINITY_DOMAIN_L4_CACHE (1 << 1) +#define CL_DEVICE_AFFINITY_DOMAIN_L3_CACHE (1 << 2) +#define CL_DEVICE_AFFINITY_DOMAIN_L2_CACHE (1 << 3) +#define CL_DEVICE_AFFINITY_DOMAIN_L1_CACHE (1 << 4) +#define CL_DEVICE_AFFINITY_DOMAIN_NEXT_PARTITIONABLE (1 << 5) + +/* cl_device_svm_capabilities */ +#define CL_DEVICE_SVM_COARSE_GRAIN_BUFFER (1 << 0) +#define CL_DEVICE_SVM_FINE_GRAIN_BUFFER (1 << 1) +#define CL_DEVICE_SVM_FINE_GRAIN_SYSTEM (1 << 2) +#define CL_DEVICE_SVM_ATOMICS (1 << 3) + +/* cl_command_queue_info */ +#define CL_QUEUE_CONTEXT 0x1090 +#define CL_QUEUE_DEVICE 0x1091 +#define CL_QUEUE_REFERENCE_COUNT 0x1092 +#define CL_QUEUE_PROPERTIES 0x1093 +#define CL_QUEUE_SIZE 0x1094 +#define CL_QUEUE_DEVICE_DEFAULT 0x1095 +#define CL_QUEUE_PROPERTIES_ARRAY 0x1098 + +/* cl_mem_flags and cl_svm_mem_flags - bitfield */ +#define CL_MEM_READ_WRITE (1 << 0) +#define CL_MEM_WRITE_ONLY (1 << 1) +#define CL_MEM_READ_ONLY (1 << 2) +#define CL_MEM_USE_HOST_PTR (1 << 3) +#define CL_MEM_ALLOC_HOST_PTR (1 << 4) +#define CL_MEM_COPY_HOST_PTR (1 << 5) +/* reserved (1 << 6) */ +#define CL_MEM_HOST_WRITE_ONLY (1 << 7) +#define CL_MEM_HOST_READ_ONLY (1 << 8) +#define CL_MEM_HOST_NO_ACCESS (1 << 9) +#define CL_MEM_SVM_FINE_GRAIN_BUFFER (1 << 10) /* used by cl_svm_mem_flags only */ +#define CL_MEM_SVM_ATOMICS (1 << 11) /* used by cl_svm_mem_flags only */ +#define CL_MEM_KERNEL_READ_AND_WRITE (1 << 12) + +/* cl_mem_migration_flags - bitfield */ +#define CL_MIGRATE_MEM_OBJECT_HOST (1 << 0) +#define CL_MIGRATE_MEM_OBJECT_CONTENT_UNDEFINED (1 << 1) + +/* cl_channel_order */ +#define CL_R 0x10B0 +#define CL_A 0x10B1 +#define CL_RG 0x10B2 +#define CL_RA 0x10B3 +#define CL_RGB 0x10B4 +#define CL_RGBA 0x10B5 +#define CL_BGRA 0x10B6 +#define CL_ARGB 0x10B7 +#define CL_INTENSITY 0x10B8 +#define CL_LUMINANCE 0x10B9 +#define CL_Rx 0x10BA +#define CL_RGx 0x10BB +#define CL_RGBx 0x10BC +#define CL_DEPTH 0x10BD +#define CL_DEPTH_STENCIL 0x10BE +#define CL_sRGB 0x10BF +#define CL_sRGBx 0x10C0 +#define CL_sRGBA 0x10C1 +#define CL_sBGRA 0x10C2 +#define CL_ABGR 0x10C3 + +/* cl_channel_type */ +#define CL_SNORM_INT8 0x10D0 +#define CL_SNORM_INT16 0x10D1 +#define CL_UNORM_INT8 0x10D2 +#define CL_UNORM_INT16 0x10D3 +#define CL_UNORM_SHORT_565 0x10D4 +#define CL_UNORM_SHORT_555 0x10D5 +#define CL_UNORM_INT_101010 0x10D6 +#define CL_SIGNED_INT8 0x10D7 +#define CL_SIGNED_INT16 0x10D8 +#define CL_SIGNED_INT32 0x10D9 +#define CL_UNSIGNED_INT8 0x10DA +#define CL_UNSIGNED_INT16 0x10DB +#define CL_UNSIGNED_INT32 0x10DC +#define CL_HALF_FLOAT 0x10DD +#define CL_FLOAT 0x10DE +#define CL_UNORM_INT24 0x10DF +#define CL_UNORM_INT_101010_2 0x10E0 + +/* cl_mem_object_type */ +#define CL_MEM_OBJECT_BUFFER 0x10F0 +#define CL_MEM_OBJECT_IMAGE2D 0x10F1 +#define CL_MEM_OBJECT_IMAGE3D 0x10F2 +#define CL_MEM_OBJECT_IMAGE2D_ARRAY 0x10F3 +#define CL_MEM_OBJECT_IMAGE1D 0x10F4 +#define CL_MEM_OBJECT_IMAGE1D_ARRAY 0x10F5 +#define CL_MEM_OBJECT_IMAGE1D_BUFFER 0x10F6 +#define CL_MEM_OBJECT_PIPE 0x10F7 + +/* cl_mem_info */ +#define CL_MEM_TYPE 0x1100 +#define CL_MEM_FLAGS 0x1101 +#define CL_MEM_SIZE 0x1102 +#define CL_MEM_HOST_PTR 0x1103 +#define CL_MEM_MAP_COUNT 0x1104 +#define CL_MEM_REFERENCE_COUNT 0x1105 +#define CL_MEM_CONTEXT 0x1106 +#define CL_MEM_ASSOCIATED_MEMOBJECT 0x1107 +#define CL_MEM_OFFSET 0x1108 +#define CL_MEM_USES_SVM_POINTER 0x1109 +#define CL_MEM_PROPERTIES 0x110A + +/* cl_image_info */ +#define CL_IMAGE_FORMAT 0x1110 +#define CL_IMAGE_ELEMENT_SIZE 0x1111 +#define CL_IMAGE_ROW_PITCH 0x1112 +#define CL_IMAGE_SLICE_PITCH 0x1113 +#define CL_IMAGE_WIDTH 0x1114 +#define CL_IMAGE_HEIGHT 0x1115 +#define CL_IMAGE_DEPTH 0x1116 +#define CL_IMAGE_ARRAY_SIZE 0x1117 +#define CL_IMAGE_BUFFER 0x1118 +#define CL_IMAGE_NUM_MIP_LEVELS 0x1119 +#define CL_IMAGE_NUM_SAMPLES 0x111A + +/* cl_pipe_info */ +#define CL_PIPE_PACKET_SIZE 0x1120 +#define CL_PIPE_MAX_PACKETS 0x1121 +#define CL_PIPE_PROPERTIES 0x1122 + +/* cl_addressing_mode */ +#define CL_ADDRESS_NONE 0x1130 +#define CL_ADDRESS_CLAMP_TO_EDGE 0x1131 +#define CL_ADDRESS_CLAMP 0x1132 +#define CL_ADDRESS_REPEAT 0x1133 +#define CL_ADDRESS_MIRRORED_REPEAT 0x1134 + +/* cl_filter_mode */ +#define CL_FILTER_NEAREST 0x1140 +#define CL_FILTER_LINEAR 0x1141 + +/* cl_sampler_info */ +#define CL_SAMPLER_REFERENCE_COUNT 0x1150 +#define CL_SAMPLER_CONTEXT 0x1151 +#define CL_SAMPLER_NORMALIZED_COORDS 0x1152 +#define CL_SAMPLER_ADDRESSING_MODE 0x1153 +#define CL_SAMPLER_FILTER_MODE 0x1154 +/* These enumerants are for the cl_khr_mipmap_image extension. + They have since been added to cl_ext.h with an appropriate + KHR suffix, but are left here for backwards compatibility. */ +#define CL_SAMPLER_MIP_FILTER_MODE 0x1155 +#define CL_SAMPLER_LOD_MIN 0x1156 +#define CL_SAMPLER_LOD_MAX 0x1157 +#define CL_SAMPLER_PROPERTIES 0x1158 + +/* cl_map_flags - bitfield */ +#define CL_MAP_READ (1 << 0) +#define CL_MAP_WRITE (1 << 1) +#define CL_MAP_WRITE_INVALIDATE_REGION (1 << 2) + +/* cl_program_info */ +#define CL_PROGRAM_REFERENCE_COUNT 0x1160 +#define CL_PROGRAM_CONTEXT 0x1161 +#define CL_PROGRAM_NUM_DEVICES 0x1162 +#define CL_PROGRAM_DEVICES 0x1163 +#define CL_PROGRAM_SOURCE 0x1164 +#define CL_PROGRAM_BINARY_SIZES 0x1165 +#define CL_PROGRAM_BINARIES 0x1166 +#define CL_PROGRAM_NUM_KERNELS 0x1167 +#define CL_PROGRAM_KERNEL_NAMES 0x1168 +#define CL_PROGRAM_IL 0x1169 +#define CL_PROGRAM_SCOPE_GLOBAL_CTORS_PRESENT 0x116A +#define CL_PROGRAM_SCOPE_GLOBAL_DTORS_PRESENT 0x116B + +/* cl_program_build_info */ +#define CL_PROGRAM_BUILD_STATUS 0x1181 +#define CL_PROGRAM_BUILD_OPTIONS 0x1182 +#define CL_PROGRAM_BUILD_LOG 0x1183 +#define CL_PROGRAM_BINARY_TYPE 0x1184 +#define CL_PROGRAM_BUILD_GLOBAL_VARIABLE_TOTAL_SIZE 0x1185 + +/* cl_program_binary_type */ +#define CL_PROGRAM_BINARY_TYPE_NONE 0x0 +#define CL_PROGRAM_BINARY_TYPE_COMPILED_OBJECT 0x1 +#define CL_PROGRAM_BINARY_TYPE_LIBRARY 0x2 +#define CL_PROGRAM_BINARY_TYPE_EXECUTABLE 0x4 + +/* cl_build_status */ +#define CL_BUILD_SUCCESS 0 +#define CL_BUILD_NONE -1 +#define CL_BUILD_ERROR -2 +#define CL_BUILD_IN_PROGRESS -3 + +/* cl_kernel_info */ +#define CL_KERNEL_FUNCTION_NAME 0x1190 +#define CL_KERNEL_NUM_ARGS 0x1191 +#define CL_KERNEL_REFERENCE_COUNT 0x1192 +#define CL_KERNEL_CONTEXT 0x1193 +#define CL_KERNEL_PROGRAM 0x1194 +#define CL_KERNEL_ATTRIBUTES 0x1195 + +/* cl_kernel_arg_info */ +#define CL_KERNEL_ARG_ADDRESS_QUALIFIER 0x1196 +#define CL_KERNEL_ARG_ACCESS_QUALIFIER 0x1197 +#define CL_KERNEL_ARG_TYPE_NAME 0x1198 +#define CL_KERNEL_ARG_TYPE_QUALIFIER 0x1199 +#define CL_KERNEL_ARG_NAME 0x119A + +/* cl_kernel_arg_address_qualifier */ +#define CL_KERNEL_ARG_ADDRESS_GLOBAL 0x119B +#define CL_KERNEL_ARG_ADDRESS_LOCAL 0x119C +#define CL_KERNEL_ARG_ADDRESS_CONSTANT 0x119D +#define CL_KERNEL_ARG_ADDRESS_PRIVATE 0x119E + +/* cl_kernel_arg_access_qualifier */ +#define CL_KERNEL_ARG_ACCESS_READ_ONLY 0x11A0 +#define CL_KERNEL_ARG_ACCESS_WRITE_ONLY 0x11A1 +#define CL_KERNEL_ARG_ACCESS_READ_WRITE 0x11A2 +#define CL_KERNEL_ARG_ACCESS_NONE 0x11A3 + +/* cl_kernel_arg_type_qualifier */ +#define CL_KERNEL_ARG_TYPE_NONE 0 +#define CL_KERNEL_ARG_TYPE_CONST (1 << 0) +#define CL_KERNEL_ARG_TYPE_RESTRICT (1 << 1) +#define CL_KERNEL_ARG_TYPE_VOLATILE (1 << 2) +#define CL_KERNEL_ARG_TYPE_PIPE (1 << 3) + +/* cl_kernel_work_group_info */ +#define CL_KERNEL_WORK_GROUP_SIZE 0x11B0 +#define CL_KERNEL_COMPILE_WORK_GROUP_SIZE 0x11B1 +#define CL_KERNEL_LOCAL_MEM_SIZE 0x11B2 +#define CL_KERNEL_PREFERRED_WORK_GROUP_SIZE_MULTIPLE 0x11B3 +#define CL_KERNEL_PRIVATE_MEM_SIZE 0x11B4 +#define CL_KERNEL_GLOBAL_WORK_SIZE 0x11B5 + +/* cl_kernel_sub_group_info */ +#define CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE 0x2033 +#define CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE 0x2034 +#define CL_KERNEL_LOCAL_SIZE_FOR_SUB_GROUP_COUNT 0x11B8 +#define CL_KERNEL_MAX_NUM_SUB_GROUPS 0x11B9 +#define CL_KERNEL_COMPILE_NUM_SUB_GROUPS 0x11BA + +/* cl_kernel_exec_info */ +#define CL_KERNEL_EXEC_INFO_SVM_PTRS 0x11B6 +#define CL_KERNEL_EXEC_INFO_SVM_FINE_GRAIN_SYSTEM 0x11B7 + +/* cl_event_info */ +#define CL_EVENT_COMMAND_QUEUE 0x11D0 +#define CL_EVENT_COMMAND_TYPE 0x11D1 +#define CL_EVENT_REFERENCE_COUNT 0x11D2 +#define CL_EVENT_COMMAND_EXECUTION_STATUS 0x11D3 +#define CL_EVENT_CONTEXT 0x11D4 + +/* cl_command_type */ +#define CL_COMMAND_NDRANGE_KERNEL 0x11F0 +#define CL_COMMAND_TASK 0x11F1 +#define CL_COMMAND_NATIVE_KERNEL 0x11F2 +#define CL_COMMAND_READ_BUFFER 0x11F3 +#define CL_COMMAND_WRITE_BUFFER 0x11F4 +#define CL_COMMAND_COPY_BUFFER 0x11F5 +#define CL_COMMAND_READ_IMAGE 0x11F6 +#define CL_COMMAND_WRITE_IMAGE 0x11F7 +#define CL_COMMAND_COPY_IMAGE 0x11F8 +#define CL_COMMAND_COPY_IMAGE_TO_BUFFER 0x11F9 +#define CL_COMMAND_COPY_BUFFER_TO_IMAGE 0x11FA +#define CL_COMMAND_MAP_BUFFER 0x11FB +#define CL_COMMAND_MAP_IMAGE 0x11FC +#define CL_COMMAND_UNMAP_MEM_OBJECT 0x11FD +#define CL_COMMAND_MARKER 0x11FE +#define CL_COMMAND_ACQUIRE_GL_OBJECTS 0x11FF +#define CL_COMMAND_RELEASE_GL_OBJECTS 0x1200 +#define CL_COMMAND_READ_BUFFER_RECT 0x1201 +#define CL_COMMAND_WRITE_BUFFER_RECT 0x1202 +#define CL_COMMAND_COPY_BUFFER_RECT 0x1203 +#define CL_COMMAND_USER 0x1204 +#define CL_COMMAND_BARRIER 0x1205 +#define CL_COMMAND_MIGRATE_MEM_OBJECTS 0x1206 +#define CL_COMMAND_FILL_BUFFER 0x1207 +#define CL_COMMAND_FILL_IMAGE 0x1208 +#define CL_COMMAND_SVM_FREE 0x1209 +#define CL_COMMAND_SVM_MEMCPY 0x120A +#define CL_COMMAND_SVM_MEMFILL 0x120B +#define CL_COMMAND_SVM_MAP 0x120C +#define CL_COMMAND_SVM_UNMAP 0x120D +#define CL_COMMAND_SVM_MIGRATE_MEM 0x120E + +/* command execution status */ +#define CL_COMPLETE 0x0 +#define CL_RUNNING 0x1 +#define CL_SUBMITTED 0x2 +#define CL_QUEUED 0x3 + +/* cl_buffer_create_type */ +#define CL_BUFFER_CREATE_TYPE_REGION 0x1220 + +/* cl_profiling_info */ +#define CL_PROFILING_COMMAND_QUEUED 0x1280 +#define CL_PROFILING_COMMAND_SUBMIT 0x1281 +#define CL_PROFILING_COMMAND_START 0x1282 +#define CL_PROFILING_COMMAND_END 0x1283 +#define CL_PROFILING_COMMAND_COMPLETE 0x1284 + +/* cl_device_atomic_capabilities - bitfield */ +#define CL_DEVICE_ATOMIC_ORDER_RELAXED (1 << 0) +#define CL_DEVICE_ATOMIC_ORDER_ACQ_REL (1 << 1) +#define CL_DEVICE_ATOMIC_ORDER_SEQ_CST (1 << 2) +#define CL_DEVICE_ATOMIC_SCOPE_WORK_ITEM (1 << 3) +#define CL_DEVICE_ATOMIC_SCOPE_WORK_GROUP (1 << 4) +#define CL_DEVICE_ATOMIC_SCOPE_DEVICE (1 << 5) +#define CL_DEVICE_ATOMIC_SCOPE_ALL_DEVICES (1 << 6) + +/* cl_device_device_enqueue_capabilities - bitfield */ +#define CL_DEVICE_QUEUE_SUPPORTED (1 << 0) +#define CL_DEVICE_QUEUE_REPLACEABLE_DEFAULT (1 << 1) + +/* cl_khronos_vendor_id */ +#define CL_KHRONOS_VENDOR_ID_CODEPLAY 0x10004 + +/* cl_version */ +#define CL_VERSION_MAJOR_BITS (10) +#define CL_VERSION_MINOR_BITS (10) +#define CL_VERSION_PATCH_BITS (12) + +#define CL_VERSION_MAJOR_MASK ((1 << CL_VERSION_MAJOR_BITS) - 1) +#define CL_VERSION_MINOR_MASK ((1 << CL_VERSION_MINOR_BITS) - 1) +#define CL_VERSION_PATCH_MASK ((1 << CL_VERSION_PATCH_BITS) - 1) + +#define CL_VERSION_MAJOR(version) \ + ((version) >> (CL_VERSION_MINOR_BITS + CL_VERSION_PATCH_BITS)) + +#define CL_VERSION_MINOR(version) \ + (((version) >> CL_VERSION_PATCH_BITS) & CL_VERSION_MINOR_MASK) + +#define CL_VERSION_PATCH(version) ((version) & CL_VERSION_PATCH_MASK) + +#define CL_MAKE_VERSION(major, minor, patch) \ + ((((major) & CL_VERSION_MAJOR_MASK) \ + << (CL_VERSION_MINOR_BITS + CL_VERSION_PATCH_BITS)) | \ + (((minor) & CL_VERSION_MINOR_MASK) << CL_VERSION_PATCH_BITS) | \ + ((patch) & CL_VERSION_PATCH_MASK)) + +/********************************************************************************************************/ + +/* Platform API */ +extern CL_API_ENTRY cl_int CL_API_CALL +clGetPlatformIDs(cl_uint num_entries, + cl_platform_id * platforms, + cl_uint * num_platforms) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetPlatformInfo(cl_platform_id platform, + cl_platform_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +/* Device APIs */ +extern CL_API_ENTRY cl_int CL_API_CALL +clGetDeviceIDs(cl_platform_id platform, + cl_device_type device_type, + cl_uint num_entries, + cl_device_id * devices, + cl_uint * num_devices) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetDeviceInfo(cl_device_id device, + cl_device_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clCreateSubDevices(cl_device_id in_device, + const cl_device_partition_property * properties, + cl_uint num_devices, + cl_device_id * out_devices, + cl_uint * num_devices_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainDevice(cl_device_id device) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseDevice(cl_device_id device) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetDefaultDeviceCommandQueue(cl_context context, + cl_device_id device, + cl_command_queue command_queue) CL_API_SUFFIX__VERSION_2_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetDeviceAndHostTimer(cl_device_id device, + cl_ulong* device_timestamp, + cl_ulong* host_timestamp) CL_API_SUFFIX__VERSION_2_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetHostTimer(cl_device_id device, + cl_ulong * host_timestamp) CL_API_SUFFIX__VERSION_2_1; + +/* Context APIs */ +extern CL_API_ENTRY cl_context CL_API_CALL +clCreateContext(const cl_context_properties * properties, + cl_uint num_devices, + const cl_device_id * devices, + void (CL_CALLBACK * pfn_notify)(const char * errinfo, + const void * private_info, + size_t cb, + void * user_data), + void * user_data, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_context CL_API_CALL +clCreateContextFromType(const cl_context_properties * properties, + cl_device_type device_type, + void (CL_CALLBACK * pfn_notify)(const char * errinfo, + const void * private_info, + size_t cb, + void * user_data), + void * user_data, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainContext(cl_context context) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseContext(cl_context context) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetContextInfo(cl_context context, + cl_context_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetContextDestructorCallback(cl_context context, + void (CL_CALLBACK* pfn_notify)(cl_context context, + void* user_data), + void* user_data) CL_API_SUFFIX__VERSION_3_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueTask(cl_command_queue command_queue, + cl_kernel kernel, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +/* Command Queue APIs */ +extern CL_API_ENTRY cl_command_queue CL_API_CALL +clCreateCommandQueue(cl_context context, + cl_device_id device, + cl_command_queue_properties properties, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_command_queue CL_API_CALL +clCreateCommandQueueWithProperties(cl_context context, + cl_device_id device, + const cl_queue_properties * properties, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainCommandQueue(cl_command_queue command_queue) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseCommandQueue(cl_command_queue command_queue) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetCommandQueueInfo(cl_command_queue command_queue, + cl_command_queue_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +/* Memory Object APIs */ +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreateBuffer(cl_context context, + cl_mem_flags flags, + size_t size, + void * host_ptr, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreateSubBuffer(cl_mem buffer, + cl_mem_flags flags, + cl_buffer_create_type buffer_create_type, + const void * buffer_create_info, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_1; + +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreateImage(cl_context context, + cl_mem_flags flags, + const cl_image_format * image_format, + const cl_image_desc * image_desc, + void * host_ptr, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreatePipe(cl_context context, + cl_mem_flags flags, + cl_uint pipe_packet_size, + cl_uint pipe_max_packets, + const cl_pipe_properties * properties, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreateBufferWithProperties(cl_context context, + const cl_mem_properties * properties, + cl_mem_flags flags, + size_t size, + void * host_ptr, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_3_0; + +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreateImageWithProperties(cl_context context, + const cl_mem_properties * properties, + cl_mem_flags flags, + const cl_image_format * image_format, + const cl_image_desc * image_desc, + void * host_ptr, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_3_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainMemObject(cl_mem memobj) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseMemObject(cl_mem memobj) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetSupportedImageFormats(cl_context context, + cl_mem_flags flags, + cl_mem_object_type image_type, + cl_uint num_entries, + cl_image_format * image_formats, + cl_uint * num_image_formats) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetMemObjectInfo(cl_mem memobj, + cl_mem_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetImageInfo(cl_mem image, + cl_image_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetPipeInfo(cl_mem pipe, + cl_pipe_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetMemObjectDestructorCallback(cl_mem memobj, + void (CL_CALLBACK * pfn_notify)(cl_mem memobj, + void * user_data), + void * user_data) CL_API_SUFFIX__VERSION_1_1; + +/* SVM Allocation APIs */ +extern CL_API_ENTRY void * CL_API_CALL +clSVMAlloc(cl_context context, + cl_svm_mem_flags flags, + size_t size, + cl_uint alignment) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY void CL_API_CALL +clSVMFree(cl_context context, + void * svm_pointer) CL_API_SUFFIX__VERSION_2_0; + +/* Sampler APIs */ +extern CL_API_ENTRY cl_sampler CL_API_CALL +clCreateSampler(cl_context context, + cl_bool normalized_coords, + cl_addressing_mode addressing_mode, + cl_filter_mode filter_mode, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_sampler CL_API_CALL +clCreateSamplerWithProperties(cl_context context, + const cl_sampler_properties * sampler_properties, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainSampler(cl_sampler sampler) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseSampler(cl_sampler sampler) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetSamplerInfo(cl_sampler sampler, + cl_sampler_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +/* Program Object APIs */ +extern CL_API_ENTRY cl_program CL_API_CALL +clCreateProgramWithSource(cl_context context, + cl_uint count, + const char ** strings, + const size_t * lengths, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_program CL_API_CALL +clCreateProgramWithBinary(cl_context context, + cl_uint num_devices, + const cl_device_id * device_list, + const size_t * lengths, + const unsigned char ** binaries, + cl_int * binary_status, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_program CL_API_CALL +clCreateProgramWithBuiltInKernels(cl_context context, + cl_uint num_devices, + const cl_device_id * device_list, + const char * kernel_names, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_program CL_API_CALL +clCreateProgramWithIL(cl_context context, + const void* il, + size_t length, + cl_int* errcode_ret) CL_API_SUFFIX__VERSION_2_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainProgram(cl_program program) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseProgram(cl_program program) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clBuildProgram(cl_program program, + cl_uint num_devices, + const cl_device_id * device_list, + const char * options, + void (CL_CALLBACK * pfn_notify)(cl_program program, + void * user_data), + void * user_data) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clCompileProgram(cl_program program, + cl_uint num_devices, + const cl_device_id * device_list, + const char * options, + cl_uint num_input_headers, + const cl_program * input_headers, + const char ** header_include_names, + void (CL_CALLBACK * pfn_notify)(cl_program program, + void * user_data), + void * user_data) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_program CL_API_CALL +clLinkProgram(cl_context context, + cl_uint num_devices, + const cl_device_id * device_list, + const char * options, + cl_uint num_input_programs, + const cl_program * input_programs, + void (CL_CALLBACK * pfn_notify)(cl_program program, + void * user_data), + void * user_data, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_2_2_DEPRECATED cl_int CL_API_CALL +clSetProgramReleaseCallback(cl_program program, + void (CL_CALLBACK * pfn_notify)(cl_program program, + void * user_data), + void * user_data) CL_EXT_SUFFIX__VERSION_2_2_DEPRECATED; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetProgramSpecializationConstant(cl_program program, + cl_uint spec_id, + size_t spec_size, + const void* spec_value) CL_API_SUFFIX__VERSION_2_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clUnloadPlatformCompiler(cl_platform_id platform) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetProgramInfo(cl_program program, + cl_program_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetProgramBuildInfo(cl_program program, + cl_device_id device, + cl_program_build_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +/* Kernel Object APIs */ +extern CL_API_ENTRY cl_kernel CL_API_CALL +clCreateKernel(cl_program program, + const char * kernel_name, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clCreateKernelsInProgram(cl_program program, + cl_uint num_kernels, + cl_kernel * kernels, + cl_uint * num_kernels_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_kernel CL_API_CALL +clCloneKernel(cl_kernel source_kernel, + cl_int* errcode_ret) CL_API_SUFFIX__VERSION_2_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainKernel(cl_kernel kernel) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseKernel(cl_kernel kernel) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetKernelArg(cl_kernel kernel, + cl_uint arg_index, + size_t arg_size, + const void * arg_value) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetKernelArgSVMPointer(cl_kernel kernel, + cl_uint arg_index, + const void * arg_value) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetKernelExecInfo(cl_kernel kernel, + cl_kernel_exec_info param_name, + size_t param_value_size, + const void * param_value) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetKernelInfo(cl_kernel kernel, + cl_kernel_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetKernelArgInfo(cl_kernel kernel, + cl_uint arg_indx, + cl_kernel_arg_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetKernelWorkGroupInfo(cl_kernel kernel, + cl_device_id device, + cl_kernel_work_group_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetKernelSubGroupInfo(cl_kernel kernel, + cl_device_id device, + cl_kernel_sub_group_info param_name, + size_t input_value_size, + const void* input_value, + size_t param_value_size, + void* param_value, + size_t* param_value_size_ret) CL_API_SUFFIX__VERSION_2_1; + +/* Event Object APIs */ +extern CL_API_ENTRY cl_int CL_API_CALL +clWaitForEvents(cl_uint num_events, + const cl_event * event_list) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetEventInfo(cl_event event, + cl_event_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_event CL_API_CALL +clCreateUserEvent(cl_context context, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainEvent(cl_event event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseEvent(cl_event event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetUserEventStatus(cl_event event, + cl_int execution_status) CL_API_SUFFIX__VERSION_1_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetEventCallback(cl_event event, + cl_int command_exec_callback_type, + void (CL_CALLBACK * pfn_notify)(cl_event event, + cl_int event_command_status, + void * user_data), + void * user_data) CL_API_SUFFIX__VERSION_1_1; + +/* Profiling APIs */ +extern CL_API_ENTRY cl_int CL_API_CALL +clGetEventProfilingInfo(cl_event event, + cl_profiling_info param_name, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_API_SUFFIX__VERSION_1_0; + +/* Flush and Finish APIs */ +extern CL_API_ENTRY cl_int CL_API_CALL +clFlush(cl_command_queue command_queue) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clFinish(cl_command_queue command_queue) CL_API_SUFFIX__VERSION_1_0; + +/* Enqueued Commands APIs */ +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueReadBuffer(cl_command_queue command_queue, + cl_mem buffer, + cl_bool blocking_read, + size_t offset, + size_t size, + void * ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueReadBufferRect(cl_command_queue command_queue, + cl_mem buffer, + cl_bool blocking_read, + const size_t * buffer_offset, + const size_t * host_offset, + const size_t * region, + size_t buffer_row_pitch, + size_t buffer_slice_pitch, + size_t host_row_pitch, + size_t host_slice_pitch, + void * ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueWriteBuffer(cl_command_queue command_queue, + cl_mem buffer, + cl_bool blocking_write, + size_t offset, + size_t size, + const void * ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueWriteBufferRect(cl_command_queue command_queue, + cl_mem buffer, + cl_bool blocking_write, + const size_t * buffer_offset, + const size_t * host_offset, + const size_t * region, + size_t buffer_row_pitch, + size_t buffer_slice_pitch, + size_t host_row_pitch, + size_t host_slice_pitch, + const void * ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueFillBuffer(cl_command_queue command_queue, + cl_mem buffer, + const void * pattern, + size_t pattern_size, + size_t offset, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueCopyBuffer(cl_command_queue command_queue, + cl_mem src_buffer, + cl_mem dst_buffer, + size_t src_offset, + size_t dst_offset, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueCopyBufferRect(cl_command_queue command_queue, + cl_mem src_buffer, + cl_mem dst_buffer, + const size_t * src_origin, + const size_t * dst_origin, + const size_t * region, + size_t src_row_pitch, + size_t src_slice_pitch, + size_t dst_row_pitch, + size_t dst_slice_pitch, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueReadImage(cl_command_queue command_queue, + cl_mem image, + cl_bool blocking_read, + const size_t * origin, + const size_t * region, + size_t row_pitch, + size_t slice_pitch, + void * ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueWriteImage(cl_command_queue command_queue, + cl_mem image, + cl_bool blocking_write, + const size_t * origin, + const size_t * region, + size_t input_row_pitch, + size_t input_slice_pitch, + const void * ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueFillImage(cl_command_queue command_queue, + cl_mem image, + const void * fill_color, + const size_t * origin, + const size_t * region, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueCopyImage(cl_command_queue command_queue, + cl_mem src_image, + cl_mem dst_image, + const size_t * src_origin, + const size_t * dst_origin, + const size_t * region, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueCopyImageToBuffer(cl_command_queue command_queue, + cl_mem src_image, + cl_mem dst_buffer, + const size_t * src_origin, + const size_t * region, + size_t dst_offset, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueCopyBufferToImage(cl_command_queue command_queue, + cl_mem src_buffer, + cl_mem dst_image, + size_t src_offset, + const size_t * dst_origin, + const size_t * region, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY void * CL_API_CALL +clEnqueueMapBuffer(cl_command_queue command_queue, + cl_mem buffer, + cl_bool blocking_map, + cl_map_flags map_flags, + size_t offset, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY void * CL_API_CALL +clEnqueueMapImage(cl_command_queue command_queue, + cl_mem image, + cl_bool blocking_map, + cl_map_flags map_flags, + const size_t * origin, + const size_t * region, + size_t * image_row_pitch, + size_t * image_slice_pitch, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event, + cl_int * errcode_ret) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueUnmapMemObject(cl_command_queue command_queue, + cl_mem memobj, + void * mapped_ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueMigrateMemObjects(cl_command_queue command_queue, + cl_uint num_mem_objects, + const cl_mem * mem_objects, + cl_mem_migration_flags flags, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueNDRangeKernel(cl_command_queue command_queue, + cl_kernel kernel, + cl_uint work_dim, + const size_t * global_work_offset, + const size_t * global_work_size, + const size_t * local_work_size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueNativeKernel(cl_command_queue command_queue, + void (CL_CALLBACK * user_func)(void *), + void * args, + size_t cb_args, + cl_uint num_mem_objects, + const cl_mem * mem_list, + const void ** args_mem_loc, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueMarkerWithWaitList(cl_command_queue command_queue, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueBarrierWithWaitList(cl_command_queue command_queue, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMFree(cl_command_queue command_queue, + cl_uint num_svm_pointers, + void * svm_pointers[], + void (CL_CALLBACK * pfn_free_func)(cl_command_queue queue, + cl_uint num_svm_pointers, + void * svm_pointers[], + void * user_data), + void * user_data, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMMemcpy(cl_command_queue command_queue, + cl_bool blocking_copy, + void * dst_ptr, + const void * src_ptr, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMMemFill(cl_command_queue command_queue, + void * svm_ptr, + const void * pattern, + size_t pattern_size, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMMap(cl_command_queue command_queue, + cl_bool blocking_map, + cl_map_flags flags, + void * svm_ptr, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMUnmap(cl_command_queue command_queue, + void * svm_ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_2_0; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMMigrateMem(cl_command_queue command_queue, + cl_uint num_svm_pointers, + const void ** svm_pointers, + const size_t * sizes, + cl_mem_migration_flags flags, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_API_SUFFIX__VERSION_2_1; + +/* Extension function access + * + * Returns the extension function address for the given function name, + * or NULL if a valid function can not be found. The client must + * check to make sure the address is not NULL, before using or + * calling the returned function address. + */ +extern CL_API_ENTRY void * CL_API_CALL +clGetExtensionFunctionAddressForPlatform(cl_platform_id platform, + const char * func_name) CL_API_SUFFIX__VERSION_1_2; + +#ifdef CL_USE_DEPRECATED_OPENCL_1_0_APIS + /* + * WARNING: + * This API introduces mutable state into the OpenCL implementation. It has been REMOVED + * to better facilitate thread safety. The 1.0 API is not thread safe. It is not tested by the + * OpenCL 1.1 conformance test, and consequently may not work or may not work dependably. + * It is likely to be non-performant. Use of this API is not advised. Use at your own risk. + * + * Software developers previously relying on this API are instructed to set the command queue + * properties when creating the queue, instead. + */ + extern CL_API_ENTRY cl_int CL_API_CALL + clSetCommandQueueProperty(cl_command_queue command_queue, + cl_command_queue_properties properties, + cl_bool enable, + cl_command_queue_properties * old_properties) CL_EXT_SUFFIX__VERSION_1_0_DEPRECATED; +#endif /* CL_USE_DEPRECATED_OPENCL_1_0_APIS */ + +/* Deprecated OpenCL 1.1 APIs */ +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_1_1_DEPRECATED cl_mem CL_API_CALL +clCreateImage2D(cl_context context, + cl_mem_flags flags, + const cl_image_format * image_format, + size_t image_width, + size_t image_height, + size_t image_row_pitch, + void * host_ptr, + cl_int * errcode_ret) CL_EXT_SUFFIX__VERSION_1_1_DEPRECATED; + +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_1_1_DEPRECATED cl_mem CL_API_CALL +clCreateImage3D(cl_context context, + cl_mem_flags flags, + const cl_image_format * image_format, + size_t image_width, + size_t image_height, + size_t image_depth, + size_t image_row_pitch, + size_t image_slice_pitch, + void * host_ptr, + cl_int * errcode_ret) CL_EXT_SUFFIX__VERSION_1_1_DEPRECATED; + +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_1_1_DEPRECATED cl_int CL_API_CALL +clEnqueueMarker(cl_command_queue command_queue, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_1_DEPRECATED; + +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_1_1_DEPRECATED cl_int CL_API_CALL +clEnqueueWaitForEvents(cl_command_queue command_queue, + cl_uint num_events, + const cl_event * event_list) CL_EXT_SUFFIX__VERSION_1_1_DEPRECATED; + +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_1_1_DEPRECATED cl_int CL_API_CALL +clEnqueueBarrier(cl_command_queue command_queue) CL_EXT_SUFFIX__VERSION_1_1_DEPRECATED; + +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_1_1_DEPRECATED cl_int CL_API_CALL +clUnloadCompiler(void) CL_EXT_SUFFIX__VERSION_1_1_DEPRECATED; + +extern CL_API_ENTRY CL_EXT_PREFIX__VERSION_1_1_DEPRECATED void * CL_API_CALL +clGetExtensionFunctionAddress(const char * func_name) CL_EXT_SUFFIX__VERSION_1_1_DEPRECATED; + +#ifdef __cplusplus +} +#endif + +#endif /* __OPENCL_CL_H */ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/CL/cl_ext.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/CL/cl_ext.h new file mode 100644 index 0000000000000000000000000000000000000000..b8b26159e64945dc54c0457c7d6acd92487bd325 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/CL/cl_ext.h @@ -0,0 +1,1131 @@ +/******************************************************************************* + * Copyright (c) 2008-2020 The Khronos Group Inc. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + ******************************************************************************/ + +/* cl_ext.h contains OpenCL extensions which don't have external */ +/* (OpenGL, D3D) dependencies. */ + +#ifndef __CL_EXT_H +#define __CL_EXT_H + +#ifdef __cplusplus +extern "C" { +#endif + +#ifdef __APPLE__ + #include + #include +#else + #include +#endif + +/* cl_khr_fp64 extension - no extension #define since it has no functions */ +/* CL_DEVICE_DOUBLE_FP_CONFIG is defined in CL.h for OpenCL >= 120 */ +#define CL_DEVICE_DOUBLE_FP_CONFIG 0x1032 + +/* cl_khr_fp16 extension - no extension #define since it has no functions */ +#define CL_DEVICE_HALF_FP_CONFIG 0x1033 + +/* Memory object destruction + * + * Apple extension for use to manage externally allocated buffers used with cl_mem objects with CL_MEM_USE_HOST_PTR + * + * Registers a user callback function that will be called when the memory object is deleted and its resources + * freed. Each call to clSetMemObjectCallbackFn registers the specified user callback function on a callback + * stack associated with memobj. The registered user callback functions are called in the reverse order in + * which they were registered. The user callback functions are called and then the memory object is deleted + * and its resources freed. This provides a mechanism for the application (and libraries) using memobj to be + * notified when the memory referenced by host_ptr, specified when the memory object is created and used as + * the storage bits for the memory object, can be reused or freed. + * + * The application may not call CL api's with the cl_mem object passed to the pfn_notify. + * + * Please check for the "cl_APPLE_SetMemObjectDestructor" extension using clGetDeviceInfo(CL_DEVICE_EXTENSIONS) + * before using. + */ +#define cl_APPLE_SetMemObjectDestructor 1 +cl_int CL_API_ENTRY clSetMemObjectDestructorAPPLE( cl_mem memobj, + void (* pfn_notify)(cl_mem memobj, void * user_data), + void * user_data) CL_EXT_SUFFIX__VERSION_1_0; + + +/* Context Logging Functions + * + * The next three convenience functions are intended to be used as the pfn_notify parameter to clCreateContext(). + * Please check for the "cl_APPLE_ContextLoggingFunctions" extension using clGetDeviceInfo(CL_DEVICE_EXTENSIONS) + * before using. + * + * clLogMessagesToSystemLog forwards on all log messages to the Apple System Logger + */ +#define cl_APPLE_ContextLoggingFunctions 1 +extern void CL_API_ENTRY clLogMessagesToSystemLogAPPLE( const char * errstr, + const void * private_info, + size_t cb, + void * user_data) CL_EXT_SUFFIX__VERSION_1_0; + +/* clLogMessagesToStdout sends all log messages to the file descriptor stdout */ +extern void CL_API_ENTRY clLogMessagesToStdoutAPPLE( const char * errstr, + const void * private_info, + size_t cb, + void * user_data) CL_EXT_SUFFIX__VERSION_1_0; + +/* clLogMessagesToStderr sends all log messages to the file descriptor stderr */ +extern void CL_API_ENTRY clLogMessagesToStderrAPPLE( const char * errstr, + const void * private_info, + size_t cb, + void * user_data) CL_EXT_SUFFIX__VERSION_1_0; + + +/************************ +* cl_khr_icd extension * +************************/ +#define cl_khr_icd 1 + +/* cl_platform_info */ +#define CL_PLATFORM_ICD_SUFFIX_KHR 0x0920 + +/* Additional Error Codes */ +#define CL_PLATFORM_NOT_FOUND_KHR -1001 + +extern CL_API_ENTRY cl_int CL_API_CALL +clIcdGetPlatformIDsKHR(cl_uint num_entries, + cl_platform_id * platforms, + cl_uint * num_platforms); + +typedef CL_API_ENTRY cl_int +(CL_API_CALL *clIcdGetPlatformIDsKHR_fn)(cl_uint num_entries, + cl_platform_id * platforms, + cl_uint * num_platforms); + + +/******************************* + * cl_khr_il_program extension * + *******************************/ +#define cl_khr_il_program 1 + +/* New property to clGetDeviceInfo for retrieving supported intermediate + * languages + */ +#define CL_DEVICE_IL_VERSION_KHR 0x105B + +/* New property to clGetProgramInfo for retrieving for retrieving the IL of a + * program + */ +#define CL_PROGRAM_IL_KHR 0x1169 + +extern CL_API_ENTRY cl_program CL_API_CALL +clCreateProgramWithILKHR(cl_context context, + const void * il, + size_t length, + cl_int * errcode_ret); + +typedef CL_API_ENTRY cl_program +(CL_API_CALL *clCreateProgramWithILKHR_fn)(cl_context context, + const void * il, + size_t length, + cl_int * errcode_ret) CL_EXT_SUFFIX__VERSION_1_2; + +/* Extension: cl_khr_image2d_from_buffer + * + * This extension allows a 2D image to be created from a cl_mem buffer without + * a copy. The type associated with a 2D image created from a buffer in an + * OpenCL program is image2d_t. Both the sampler and sampler-less read_image + * built-in functions are supported for 2D images and 2D images created from + * a buffer. Similarly, the write_image built-ins are also supported for 2D + * images created from a buffer. + * + * When the 2D image from buffer is created, the client must specify the + * width, height, image format (i.e. channel order and channel data type) + * and optionally the row pitch. + * + * The pitch specified must be a multiple of + * CL_DEVICE_IMAGE_PITCH_ALIGNMENT_KHR pixels. + * The base address of the buffer must be aligned to + * CL_DEVICE_IMAGE_BASE_ADDRESS_ALIGNMENT_KHR pixels. + */ + +#define CL_DEVICE_IMAGE_PITCH_ALIGNMENT_KHR 0x104A +#define CL_DEVICE_IMAGE_BASE_ADDRESS_ALIGNMENT_KHR 0x104B + + +/************************************** + * cl_khr_initialize_memory extension * + **************************************/ + +#define CL_CONTEXT_MEMORY_INITIALIZE_KHR 0x2030 + + +/************************************** + * cl_khr_terminate_context extension * + **************************************/ + +#define CL_CONTEXT_TERMINATED_KHR -1121 + +#define CL_DEVICE_TERMINATE_CAPABILITY_KHR 0x2031 +#define CL_CONTEXT_TERMINATE_KHR 0x2032 + +#define cl_khr_terminate_context 1 +extern CL_API_ENTRY cl_int CL_API_CALL +clTerminateContextKHR(cl_context context) CL_EXT_SUFFIX__VERSION_1_2; + +typedef CL_API_ENTRY cl_int +(CL_API_CALL *clTerminateContextKHR_fn)(cl_context context) CL_EXT_SUFFIX__VERSION_1_2; + + +/* + * Extension: cl_khr_spir + * + * This extension adds support to create an OpenCL program object from a + * Standard Portable Intermediate Representation (SPIR) instance + */ + +#define CL_DEVICE_SPIR_VERSIONS 0x40E0 +#define CL_PROGRAM_BINARY_TYPE_INTERMEDIATE 0x40E1 + + +/***************************************** + * cl_khr_create_command_queue extension * + *****************************************/ +#define cl_khr_create_command_queue 1 + +typedef cl_properties cl_queue_properties_khr; + +extern CL_API_ENTRY cl_command_queue CL_API_CALL +clCreateCommandQueueWithPropertiesKHR(cl_context context, + cl_device_id device, + const cl_queue_properties_khr* properties, + cl_int* errcode_ret) CL_EXT_SUFFIX__VERSION_1_2; + +typedef CL_API_ENTRY cl_command_queue +(CL_API_CALL *clCreateCommandQueueWithPropertiesKHR_fn)(cl_context context, + cl_device_id device, + const cl_queue_properties_khr* properties, + cl_int* errcode_ret) CL_EXT_SUFFIX__VERSION_1_2; + + +/****************************************** +* cl_khr_semaphore extension * +******************************************/ + +typedef enum _cl_semaphore_type__enum { + CL_SEMAPHORE_TYPE_BINARY_KHR = 1, +} cl_semaphore_type; + +typedef cl_properties cl_semaphore_properties_khr; + +typedef cl_uint cl_semaphore_info_khr; + +typedef struct _cl_semaphore* cl_semaphore_khr; +typedef cl_ulong cl_semaphore_payload_khr; + +extern CL_API_ENTRY cl_semaphore_khr CL_API_CALL +clCreateSemaphoreWithPropertiesKHR(cl_context context, + cl_semaphore_properties_khr *sema_props, + cl_int *errcode_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueWaitSemaphoresKHR(cl_command_queue command_queue, + cl_uint num_sema_objects, + const cl_semaphore_khr *sema_objects, + const cl_semaphore_payload_khr *sema_payload_list, + cl_uint num_events_in_wait_list, + const cl_event *event_wait_list, + cl_event *event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSignalSemaphoresKHR(cl_command_queue command_queue, + cl_uint num_sema_objects, + const cl_semaphore_khr *sema_objects, + const cl_semaphore_payload_khr *sema_payload_list, + cl_uint num_events_in_wait_list, + const cl_event *event_wait_list, + cl_event *event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetSemaphoreInfoKHR(const cl_semaphore_khr sema_object, + cl_semaphore_info_khr param_name, + size_t param_value_size, + void *param_value, + size_t *param_value_size_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseSemaphoreKHR(cl_semaphore_khr sema_object) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainSemaphoreKHR(cl_semaphore_khr sema_object) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseSemaphoreObjectKHR(cl_semaphore_khr sema_object) CL_API_SUFFIX__VERSION_1_2; + +#define CL_COMMAND_SEMAPHORE_WAIT_KHR 0x2042 +#define CL_COMMAND_SEMAPHORE_SIGNAL_KHR 0x2043 +#define CL_SEMAPHORE_CONTEXT_KHR 0x2039 +#define CL_SEMAPHORE_REFERENCE_COUNT_KHR 0x203A +#define CL_SEMAPHORE_PROPERTIES_KHR 0x203B +#define CL_SEMAPHORE_TYPE_KHR 0x203D +#define CL_PLATFORM_SEMAPHORE_TYPES_KHR 0x2036 +#define CL_SEMAPHORE_PAYLOAD_KHR 0x203C + +/****************************************** +* cl_khr_external_semaphore extension * +******************************************/ + +typedef enum _cl_external_context_type_enum { + CL_EXTERNAL_CONTEXT_TYPE_NONE = 0, + CL_EXTERNAL_CONTEXT_TYPE_CL = 1, + CL_EXTERNAL_CONTEXT_TYPE_VULKAN = 2, +} cl_external_context_type_khr; + +typedef cl_uint cl_external_semaphore_handle_type_khr; +// API-agnostic semaphore handles are defined here in this spec. +#define CL_SEMAPHORE_HANDLE_OPAQUE_FD_KHR 0x2055 +#define CL_SEMAPHORE_HANDLE_OPAQUE_WIN32_KHR 0x2056 +#define CL_SEMAPHORE_HANDLE_OPAQUE_WIN32_KMT_KHR 0x2057 + +typedef struct _cl_semaphore_desc_khr_st { + cl_external_semaphore_handle_type_khr type; + void *handle_ptr; +} cl_semaphore_desc_khr; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetSemaphoreHandleForTypeKHR(const cl_semaphore_khr sema_object, + const cl_device_id device, + cl_external_semaphore_handle_type_khr handle_type, + size_t handle_size, + void *handle_ptr, + size_t *handle_size_ret) CL_API_SUFFIX__VERSION_1_2; + +#define CL_SEMAPHORE_DESC_KHR 0x2460 +#define CL_SEMAPHORE_EXPORT_HANDLE_TYPES_KHR 0x203F + +#define CL_PLATFORM_SEMAPHORE_IMPORT_HANDLE_TYPES_KHR 0x2037 +#define CL_PLATFORM_SEMAPHORE_EXPORT_HANDLE_TYPES_KHR 0x2038 + +#define CL_DEVICE_SEMAPHORE_IMPORT_HANDLE_TYPES_KHR 0x204D +#define CL_DEVICE_SEMAPHORE_EXPORT_HANDLE_TYPES_KHR 0x204E + +// error codes +#define CL_INVALID_SEMAPHORE_KHR -1142 + +/****************************************** +* cl_khr_external_memory extension * +******************************************/ + +typedef cl_uint cl_external_context_info; + +typedef enum _cl_external_context_type_enum cl_external_context_type_khr; + +typedef cl_properties cl_mem_properties_khr; + +typedef cl_uint cl_external_mem_handle_type_khr; +// API-agnostic memory handles are defined here in this spec. +#define CL_EXTERNAL_MEMORY_HANDLE_OPAQUE_FD_KHR 0x2060 +#define CL_EXTERNAL_MEMORY_HANDLE_OPAQUE_WIN32_KHR 0x2061 +#define CL_EXTERNAL_MEMORY_HANDLE_OPAQUE_WIN32_KMT_KHR 0x2062 + +typedef struct _cl_external_mem_desc_khr_st { + cl_external_mem_handle_type_khr type; + void *handle_ptr; + size_t offset; + unsigned long long size; +} cl_external_mem_desc_khr; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetExternalContextInfoKHR(const cl_context_properties *properties, + cl_external_context_info param_name, + size_t param_value_size, + void *param_value, + size_t *param_value_size_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueAcquireExternalMemObjectsKHR(cl_command_queue command_queue, + cl_uint num_mem_objects, + const cl_mem *mem_objects, + cl_uint num_events_in_wait_list, + const cl_event *event_wait_list, + cl_event *event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueReleaseExternalMemObjectsKHR(cl_command_queue command_queue, + cl_uint num_mem_objects, + const cl_mem *mem_objects, + cl_uint num_events_in_wait_list, + const cl_event *event_wait_list, + cl_event *event) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreateBufferFromExternalMemoryKHR(cl_context context, + const cl_mem_properties_khr* properties, + cl_mem_flags flags, + cl_external_mem_desc_khr extMem, + cl_int *errcode_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_mem CL_API_CALL +clCreateImageFromExternalMemoryKHR(cl_context context, + const cl_mem_properties_khr* properties, + cl_mem_flags flags, + cl_external_mem_desc_khr extMem, + const cl_image_format *image_format, + const cl_image_desc *image_desc, + cl_int *errcode_ret) CL_API_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_semaphore_khr CL_API_CALL +clCreateFromExternalSemaphoreKHR(cl_context context, + cl_semaphore_properties_khr *sema_props, + cl_semaphore_desc_khr sema_desc, + cl_int *errcode_ret) + CL_API_SUFFIX__VERSION_1_2; + +#define CL_INVALID_EXTERNAL_DEVICEGROUP_REFERENCE_KHR -1122 +#define CL_INVALID_EXT_MEM_DESC_KHR -1123 +#define CL_INVALID_EXT_MEM_HANDLE_TYPE_KHR -1148 +#define CL_INVALID_EXT_MEM_HANDLE_KHR -1149 +#define CL_INVALID_EXT_MEM_OFFSET_KHR -1150 +#define CL_INVALID_EXT_MEM_SIZE_KHR -1140 + +#define CL_CURRENT_DEVICE_FOR_EXTERNAL_CONTEXT_KHR 0x2036 +#define CL_DEVICES_FOR_EXTERNAL_CONTEXT_KHR 0x2037 +#define CL_EXTERNAL_DEVICE_KHR 0x2038 +#define CL_EXTERNAL_DEVICEGROUP_KHR 0x2039 +#define CL_EXTERNAL_CONTEXT_TYPE_KHR 0x204B +#define CL_DEVICE_HANDLE_LIST_KHR 0x2051 +#define CL_DEVICE_HANDLE_LIST_END_KHR 0x0 + + +#define CL_COMMAND_ACQUIRE_EXTERNAL_MEM_OBJECTS_KHR 0x2047 +#define CL_COMMAND_RELEASE_EXTERNAL_MEM_OBJECTS_KHR 0x2048 +#define CL_EXTERNAL_MEM_DESC_KHR 0x203C +#define CL_EXTERNAL_IMAGE_INFO_KHR 0x203D +#define CL_PLATFORM_EXTERNAL_HANDLE_TYPES_KHR 0x203E +#define CL_PLATFORM_EXTERNAL_MEMORY_IMPORT_HANDLE_TYPES_KHR 0x2044 +#define CL_DEVICE_EXTERNAL_MEMORY_IMPORT_HANDLE_TYPES_KHR 0x204F +#define CL_DEVICE_EXTERNAL_MEMORY_PROPERTIES_KHR 0x2050 + + +/****************************************** +* cl_nv_device_attribute_query extension * +******************************************/ + +/* cl_nv_device_attribute_query extension - no extension #define since it has no functions */ +#define CL_DEVICE_COMPUTE_CAPABILITY_MAJOR_NV 0x4000 +#define CL_DEVICE_COMPUTE_CAPABILITY_MINOR_NV 0x4001 +#define CL_DEVICE_REGISTERS_PER_BLOCK_NV 0x4002 +#define CL_DEVICE_WARP_SIZE_NV 0x4003 +#define CL_DEVICE_GPU_OVERLAP_NV 0x4004 +#define CL_DEVICE_KERNEL_EXEC_TIMEOUT_NV 0x4005 +#define CL_DEVICE_INTEGRATED_MEMORY_NV 0x4006 +#define CL_DEVICE_ATTRIBUTE_ASYNC_ENGINE_COUNT_NV 0x4007 +#define CL_DEVICE_PCI_BUS_ID_NV 0x4008 +#define CL_DEVICE_PCI_SLOT_ID_NV 0x4009 +#define CL_DEVICE_PCI_DOMAIN_ID_NV 0x400A +#define CL_DEVICE_MAX_LOCAL_MEMORY_PER_SM_NV 0x400B +#define CL_DEVICE_UUID_KHR 0x106A +#define CL_DRIVER_UUID_KHR 0x106B +#define CL_DEVICE_LUID_VALID_KHR 0x106C +#define CL_DEVICE_LUID_KHR 0x106D +#define CL_DEVICE_NODE_MASK_KHR 0x106E +#define CL_UUID_SIZE_KHR 16 +#define CL_LUID_SIZE_KHR 8 + +/****************************************** +* cl_nv_create_buffer extension * +******************************************/ + +typedef cl_bitfield cl_mem_flags_NV; +CL_API_ENTRY cl_mem CL_API_CALL +clCreateBufferNV(cl_context context, + cl_mem_flags flags, + cl_mem_flags_NV flags_NV, + size_t size, + void *host_ptr, + cl_int *errcode_ret); + +/****************************************** +* cl_kernel_attribute_nv extension * +*******************************************/ + +typedef enum kernel_attribute_enum { + CL_KERNEL_PREFERRED_LOCAL_MEMORY_SIZE_NV = 0, /* setting preferred shared memory size */ +} cl_kernel_attribute_nv; + +CL_API_ENTRY cl_int CL_API_CALL +clSetKernelAttributeNV(cl_kernel kernel, + cl_device_id device, + cl_kernel_attribute_nv k_attr, + size_t param_value_size, + const void *param_value); + +CL_API_ENTRY cl_int CL_API_CALL +clGetKernelAttributeNV(cl_kernel kernel, + cl_device_id device, + cl_kernel_attribute_nv k_attr, + size_t param_value_size, + void *param_value, + size_t *param_value_size_ret); + +#define CL_MEM_LOCATION_HOST_NV (1 << 0) +#define CL_MEM_PINNED_NV (1 << 1) + +/********************************* +* cl_amd_device_attribute_query * +*********************************/ + +#define CL_DEVICE_PROFILING_TIMER_OFFSET_AMD 0x4036 +#define CL_DEVICE_TOPOLOGY_AMD 0x4037 +#define CL_DEVICE_BOARD_NAME_AMD 0x4038 +#define CL_DEVICE_GLOBAL_FREE_MEMORY_AMD 0x4039 +#define CL_DEVICE_SIMD_PER_COMPUTE_UNIT_AMD 0x4040 +#define CL_DEVICE_SIMD_WIDTH_AMD 0x4041 +#define CL_DEVICE_SIMD_INSTRUCTION_WIDTH_AMD 0x4042 +#define CL_DEVICE_WAVEFRONT_WIDTH_AMD 0x4043 +#define CL_DEVICE_GLOBAL_MEM_CHANNELS_AMD 0x4044 +#define CL_DEVICE_GLOBAL_MEM_CHANNEL_BANKS_AMD 0x4045 +#define CL_DEVICE_GLOBAL_MEM_CHANNEL_BANK_WIDTH_AMD 0x4046 +#define CL_DEVICE_LOCAL_MEM_SIZE_PER_COMPUTE_UNIT_AMD 0x4047 +#define CL_DEVICE_LOCAL_MEM_BANKS_AMD 0x4048 +#define CL_DEVICE_THREAD_TRACE_SUPPORTED_AMD 0x4049 +#define CL_DEVICE_GFXIP_MAJOR_AMD 0x404A +#define CL_DEVICE_GFXIP_MINOR_AMD 0x404B +#define CL_DEVICE_AVAILABLE_ASYNC_QUEUES_AMD 0x404C +#define CL_DEVICE_PREFERRED_WORK_GROUP_SIZE_AMD 0x4030 +#define CL_DEVICE_MAX_WORK_GROUP_SIZE_AMD 0x4031 +#define CL_DEVICE_PREFERRED_CONSTANT_BUFFER_SIZE_AMD 0x4033 +#define CL_DEVICE_PCIE_ID_AMD 0x4034 + + +/********************************* +* cl_arm_printf extension +*********************************/ + +#define CL_PRINTF_CALLBACK_ARM 0x40B0 +#define CL_PRINTF_BUFFERSIZE_ARM 0x40B1 + + +/*********************************** +* cl_ext_device_fission extension +***********************************/ +#define cl_ext_device_fission 1 + +extern CL_API_ENTRY cl_int CL_API_CALL +clReleaseDeviceEXT(cl_device_id device) CL_EXT_SUFFIX__VERSION_1_1; + +typedef CL_API_ENTRY cl_int +(CL_API_CALL *clReleaseDeviceEXT_fn)(cl_device_id device) CL_EXT_SUFFIX__VERSION_1_1; + +extern CL_API_ENTRY cl_int CL_API_CALL +clRetainDeviceEXT(cl_device_id device) CL_EXT_SUFFIX__VERSION_1_1; + +typedef CL_API_ENTRY cl_int +(CL_API_CALL *clRetainDeviceEXT_fn)(cl_device_id device) CL_EXT_SUFFIX__VERSION_1_1; + +typedef cl_ulong cl_device_partition_property_ext; +extern CL_API_ENTRY cl_int CL_API_CALL +clCreateSubDevicesEXT(cl_device_id in_device, + const cl_device_partition_property_ext * properties, + cl_uint num_entries, + cl_device_id * out_devices, + cl_uint * num_devices) CL_EXT_SUFFIX__VERSION_1_1; + +typedef CL_API_ENTRY cl_int +(CL_API_CALL * clCreateSubDevicesEXT_fn)(cl_device_id in_device, + const cl_device_partition_property_ext * properties, + cl_uint num_entries, + cl_device_id * out_devices, + cl_uint * num_devices) CL_EXT_SUFFIX__VERSION_1_1; + +/* cl_device_partition_property_ext */ +#define CL_DEVICE_PARTITION_EQUALLY_EXT 0x4050 +#define CL_DEVICE_PARTITION_BY_COUNTS_EXT 0x4051 +#define CL_DEVICE_PARTITION_BY_NAMES_EXT 0x4052 +#define CL_DEVICE_PARTITION_BY_AFFINITY_DOMAIN_EXT 0x4053 + +/* clDeviceGetInfo selectors */ +#define CL_DEVICE_PARENT_DEVICE_EXT 0x4054 +#define CL_DEVICE_PARTITION_TYPES_EXT 0x4055 +#define CL_DEVICE_AFFINITY_DOMAINS_EXT 0x4056 +#define CL_DEVICE_REFERENCE_COUNT_EXT 0x4057 +#define CL_DEVICE_PARTITION_STYLE_EXT 0x4058 + +/* error codes */ +#define CL_DEVICE_PARTITION_FAILED_EXT -1057 +#define CL_INVALID_PARTITION_COUNT_EXT -1058 +#define CL_INVALID_PARTITION_NAME_EXT -1059 + +/* CL_AFFINITY_DOMAINs */ +#define CL_AFFINITY_DOMAIN_L1_CACHE_EXT 0x1 +#define CL_AFFINITY_DOMAIN_L2_CACHE_EXT 0x2 +#define CL_AFFINITY_DOMAIN_L3_CACHE_EXT 0x3 +#define CL_AFFINITY_DOMAIN_L4_CACHE_EXT 0x4 +#define CL_AFFINITY_DOMAIN_NUMA_EXT 0x10 +#define CL_AFFINITY_DOMAIN_NEXT_FISSIONABLE_EXT 0x100 + +/* cl_device_partition_property_ext list terminators */ +#define CL_PROPERTIES_LIST_END_EXT ((cl_device_partition_property_ext) 0) +#define CL_PARTITION_BY_COUNTS_LIST_END_EXT ((cl_device_partition_property_ext) 0) +#define CL_PARTITION_BY_NAMES_LIST_END_EXT ((cl_device_partition_property_ext) 0 - 1) + + +/*********************************** + * cl_ext_migrate_memobject extension definitions + ***********************************/ +#define cl_ext_migrate_memobject 1 + +typedef cl_bitfield cl_mem_migration_flags_ext; + +#define CL_MIGRATE_MEM_OBJECT_HOST_EXT 0x1 + +#define CL_COMMAND_MIGRATE_MEM_OBJECT_EXT 0x4040 + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueMigrateMemObjectEXT(cl_command_queue command_queue, + cl_uint num_mem_objects, + const cl_mem * mem_objects, + cl_mem_migration_flags_ext flags, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event); + +typedef CL_API_ENTRY cl_int +(CL_API_CALL *clEnqueueMigrateMemObjectEXT_fn)(cl_command_queue command_queue, + cl_uint num_mem_objects, + const cl_mem * mem_objects, + cl_mem_migration_flags_ext flags, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event); + + +/********************************* +* cl_ext_cxx_for_opencl extension +*********************************/ +#define cl_ext_cxx_for_opencl 1 + +#define CL_DEVICE_CXX_FOR_OPENCL_NUMERIC_VERSION_EXT 0x4230 + +/********************************* +* cl_qcom_ext_host_ptr extension +*********************************/ +#define cl_qcom_ext_host_ptr 1 + +#define CL_MEM_EXT_HOST_PTR_QCOM (1 << 29) + +#define CL_DEVICE_EXT_MEM_PADDING_IN_BYTES_QCOM 0x40A0 +#define CL_DEVICE_PAGE_SIZE_QCOM 0x40A1 +#define CL_IMAGE_ROW_ALIGNMENT_QCOM 0x40A2 +#define CL_IMAGE_SLICE_ALIGNMENT_QCOM 0x40A3 +#define CL_MEM_HOST_UNCACHED_QCOM 0x40A4 +#define CL_MEM_HOST_WRITEBACK_QCOM 0x40A5 +#define CL_MEM_HOST_WRITETHROUGH_QCOM 0x40A6 +#define CL_MEM_HOST_WRITE_COMBINING_QCOM 0x40A7 + +typedef cl_uint cl_image_pitch_info_qcom; + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetDeviceImageInfoQCOM(cl_device_id device, + size_t image_width, + size_t image_height, + const cl_image_format *image_format, + cl_image_pitch_info_qcom param_name, + size_t param_value_size, + void *param_value, + size_t *param_value_size_ret); + +typedef struct _cl_mem_ext_host_ptr +{ + /* Type of external memory allocation. */ + /* Legal values will be defined in layered extensions. */ + cl_uint allocation_type; + + /* Host cache policy for this external memory allocation. */ + cl_uint host_cache_policy; + +} cl_mem_ext_host_ptr; + + +/******************************************* +* cl_qcom_ext_host_ptr_iocoherent extension +********************************************/ + +/* Cache policy specifying io-coherence */ +#define CL_MEM_HOST_IOCOHERENT_QCOM 0x40A9 + + +/********************************* +* cl_qcom_ion_host_ptr extension +*********************************/ + +#define CL_MEM_ION_HOST_PTR_QCOM 0x40A8 + +typedef struct _cl_mem_ion_host_ptr +{ + /* Type of external memory allocation. */ + /* Must be CL_MEM_ION_HOST_PTR_QCOM for ION allocations. */ + cl_mem_ext_host_ptr ext_host_ptr; + + /* ION file descriptor */ + int ion_filedesc; + + /* Host pointer to the ION allocated memory */ + void* ion_hostptr; + +} cl_mem_ion_host_ptr; + + +/********************************* +* cl_qcom_android_native_buffer_host_ptr extension +*********************************/ + +#define CL_MEM_ANDROID_NATIVE_BUFFER_HOST_PTR_QCOM 0x40C6 + +typedef struct _cl_mem_android_native_buffer_host_ptr +{ + /* Type of external memory allocation. */ + /* Must be CL_MEM_ANDROID_NATIVE_BUFFER_HOST_PTR_QCOM for Android native buffers. */ + cl_mem_ext_host_ptr ext_host_ptr; + + /* Virtual pointer to the android native buffer */ + void* anb_ptr; + +} cl_mem_android_native_buffer_host_ptr; + + +/****************************************** + * cl_img_yuv_image extension * + ******************************************/ + +/* Image formats used in clCreateImage */ +#define CL_NV21_IMG 0x40D0 +#define CL_YV12_IMG 0x40D1 + + +/****************************************** + * cl_img_cached_allocations extension * + ******************************************/ + +/* Flag values used by clCreateBuffer */ +#define CL_MEM_USE_UNCACHED_CPU_MEMORY_IMG (1 << 26) +#define CL_MEM_USE_CACHED_CPU_MEMORY_IMG (1 << 27) + + +/****************************************** + * cl_img_use_gralloc_ptr extension * + ******************************************/ +#define cl_img_use_gralloc_ptr 1 + +/* Flag values used by clCreateBuffer */ +#define CL_MEM_USE_GRALLOC_PTR_IMG (1 << 28) + +/* To be used by clGetEventInfo: */ +#define CL_COMMAND_ACQUIRE_GRALLOC_OBJECTS_IMG 0x40D2 +#define CL_COMMAND_RELEASE_GRALLOC_OBJECTS_IMG 0x40D3 + +/* Error code from clEnqueueReleaseGrallocObjectsIMG */ +#define CL_GRALLOC_RESOURCE_NOT_ACQUIRED_IMG 0x40D4 + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueAcquireGrallocObjectsIMG(cl_command_queue command_queue, + cl_uint num_objects, + const cl_mem * mem_objects, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueReleaseGrallocObjectsIMG(cl_command_queue command_queue, + cl_uint num_objects, + const cl_mem * mem_objects, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_2; + + +/********************************* +* cl_khr_subgroups extension +*********************************/ +#define cl_khr_subgroups 1 + +#if !defined(CL_VERSION_2_1) +/* For OpenCL 2.1 and newer, cl_kernel_sub_group_info is declared in CL.h. + In hindsight, there should have been a khr suffix on this type for + the extension, but keeping it un-suffixed to maintain backwards + compatibility. */ +typedef cl_uint cl_kernel_sub_group_info; +#endif + +/* cl_kernel_sub_group_info */ +#define CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE_KHR 0x2033 +#define CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE_KHR 0x2034 + +extern CL_API_ENTRY cl_int CL_API_CALL +clGetKernelSubGroupInfoKHR(cl_kernel in_kernel, + cl_device_id in_device, + cl_kernel_sub_group_info param_name, + size_t input_value_size, + const void * input_value, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_EXT_SUFFIX__VERSION_2_0_DEPRECATED; + +typedef CL_API_ENTRY cl_int +(CL_API_CALL * clGetKernelSubGroupInfoKHR_fn)(cl_kernel in_kernel, + cl_device_id in_device, + cl_kernel_sub_group_info param_name, + size_t input_value_size, + const void * input_value, + size_t param_value_size, + void * param_value, + size_t * param_value_size_ret) CL_EXT_SUFFIX__VERSION_2_0_DEPRECATED; + + +/********************************* +* cl_khr_mipmap_image extension +*********************************/ + +/* cl_sampler_properties */ +#define CL_SAMPLER_MIP_FILTER_MODE_KHR 0x1155 +#define CL_SAMPLER_LOD_MIN_KHR 0x1156 +#define CL_SAMPLER_LOD_MAX_KHR 0x1157 + + +/********************************* +* cl_khr_priority_hints extension +*********************************/ +/* This extension define is for backwards compatibility. + It shouldn't be required since this extension has no new functions. */ +#define cl_khr_priority_hints 1 + +typedef cl_uint cl_queue_priority_khr; + +/* cl_command_queue_properties */ +#define CL_QUEUE_PRIORITY_KHR 0x1096 + +/* cl_queue_priority_khr */ +#define CL_QUEUE_PRIORITY_HIGH_KHR (1<<0) +#define CL_QUEUE_PRIORITY_MED_KHR (1<<1) +#define CL_QUEUE_PRIORITY_LOW_KHR (1<<2) + + +/********************************* +* cl_khr_throttle_hints extension +*********************************/ +/* This extension define is for backwards compatibility. + It shouldn't be required since this extension has no new functions. */ +#define cl_khr_throttle_hints 1 + +typedef cl_uint cl_queue_throttle_khr; + +/* cl_command_queue_properties */ +#define CL_QUEUE_THROTTLE_KHR 0x1097 + +/* cl_queue_throttle_khr */ +#define CL_QUEUE_THROTTLE_HIGH_KHR (1<<0) +#define CL_QUEUE_THROTTLE_MED_KHR (1<<1) +#define CL_QUEUE_THROTTLE_LOW_KHR (1<<2) + + +/********************************* +* cl_khr_subgroup_named_barrier +*********************************/ +/* This extension define is for backwards compatibility. + It shouldn't be required since this extension has no new functions. */ +#define cl_khr_subgroup_named_barrier 1 + +/* cl_device_info */ +#define CL_DEVICE_MAX_NAMED_BARRIER_COUNT_KHR 0x2035 + + +/********************************* +* cl_khr_extended_versioning +*********************************/ + +#define cl_khr_extended_versioning 1 + +#define CL_VERSION_MAJOR_BITS_KHR (10) +#define CL_VERSION_MINOR_BITS_KHR (10) +#define CL_VERSION_PATCH_BITS_KHR (12) + +#define CL_VERSION_MAJOR_MASK_KHR ((1 << CL_VERSION_MAJOR_BITS_KHR) - 1) +#define CL_VERSION_MINOR_MASK_KHR ((1 << CL_VERSION_MINOR_BITS_KHR) - 1) +#define CL_VERSION_PATCH_MASK_KHR ((1 << CL_VERSION_PATCH_BITS_KHR) - 1) + +#define CL_VERSION_MAJOR_KHR(version) ((version) >> (CL_VERSION_MINOR_BITS_KHR + CL_VERSION_PATCH_BITS_KHR)) +#define CL_VERSION_MINOR_KHR(version) (((version) >> CL_VERSION_PATCH_BITS_KHR) & CL_VERSION_MINOR_MASK_KHR) +#define CL_VERSION_PATCH_KHR(version) ((version) & CL_VERSION_PATCH_MASK_KHR) + +#define CL_MAKE_VERSION_KHR(major, minor, patch) \ + ((((major) & CL_VERSION_MAJOR_MASK_KHR) << (CL_VERSION_MINOR_BITS_KHR + CL_VERSION_PATCH_BITS_KHR)) | \ + (((minor) & CL_VERSION_MINOR_MASK_KHR) << CL_VERSION_PATCH_BITS_KHR) | \ + ((patch) & CL_VERSION_PATCH_MASK_KHR)) + +typedef cl_uint cl_version_khr; + +#define CL_NAME_VERSION_MAX_NAME_SIZE_KHR 64 + +typedef struct _cl_name_version_khr +{ + cl_version_khr version; + char name[CL_NAME_VERSION_MAX_NAME_SIZE_KHR]; +} cl_name_version_khr; + +/* cl_platform_info */ +#define CL_PLATFORM_NUMERIC_VERSION_KHR 0x0906 +#define CL_PLATFORM_EXTENSIONS_WITH_VERSION_KHR 0x0907 + +/* cl_device_info */ +#define CL_DEVICE_NUMERIC_VERSION_KHR 0x105E +#define CL_DEVICE_OPENCL_C_NUMERIC_VERSION_KHR 0x105F +#define CL_DEVICE_EXTENSIONS_WITH_VERSION_KHR 0x1060 +#define CL_DEVICE_ILS_WITH_VERSION_KHR 0x1061 +#define CL_DEVICE_BUILT_IN_KERNELS_WITH_VERSION_KHR 0x1062 + + +/********************************* +* cl_khr_device_uuid extension +*********************************/ +#define cl_khr_device_uuid 1 + +#define CL_UUID_SIZE_KHR 16 +#define CL_LUID_SIZE_KHR 8 + +#define CL_DEVICE_UUID_KHR 0x106A +#define CL_DRIVER_UUID_KHR 0x106B +#define CL_DEVICE_LUID_VALID_KHR 0x106C +#define CL_DEVICE_LUID_KHR 0x106D +#define CL_DEVICE_NODE_MASK_KHR 0x106E + +/********************************** + * cl_khr_pci_bus_info extension * + **********************************/ +#define cl_khr_pci_bus_info 1 + +#define CL_DEVICE_PCI_BUS_INFO_KHR 0x410F + +typedef struct _cl_device_pci_bus_info_khr { + cl_uint pci_domain; + cl_uint pci_bus; + cl_uint pci_device; + cl_uint pci_function; +} cl_device_pci_bus_info_khr; + +/********************************** + * cl_arm_import_memory extension * + **********************************/ +#define cl_arm_import_memory 1 + +typedef intptr_t cl_import_properties_arm; + +/* Default and valid proporties name for cl_arm_import_memory */ +#define CL_IMPORT_TYPE_ARM 0x40B2 + +/* Host process memory type default value for CL_IMPORT_TYPE_ARM property */ +#define CL_IMPORT_TYPE_HOST_ARM 0x40B3 + +/* DMA BUF memory type value for CL_IMPORT_TYPE_ARM property */ +#define CL_IMPORT_TYPE_DMA_BUF_ARM 0x40B4 + +/* Protected memory property */ +#define CL_IMPORT_TYPE_PROTECTED_ARM 0x40B5 + +/* Android hardware buffer type value for CL_IMPORT_TYPE_ARM property */ +#define CL_IMPORT_TYPE_ANDROID_HARDWARE_BUFFER_ARM 0x41E2 + +/* Data consistency with host property */ +#define CL_IMPORT_DMA_BUF_DATA_CONSISTENCY_WITH_HOST_ARM 0x41E3 + +/* Import memory size value to indicate a size for the whole buffer */ +#define CL_IMPORT_MEMORY_WHOLE_ALLOCATION_ARM SIZE_MAX + +/* This extension adds a new function that allows for direct memory import into + * OpenCL via the clImportMemoryARM function. + * + * Memory imported through this interface will be mapped into the device's page + * tables directly, providing zero copy access. It will never fall back to copy + * operations and aliased buffers. + * + * Types of memory supported for import are specified as additional extension + * strings. + * + * This extension produces cl_mem allocations which are compatible with all other + * users of cl_mem in the standard API. + * + * This extension maps pages with the same properties as the normal buffer creation + * function clCreateBuffer. + */ +extern CL_API_ENTRY cl_mem CL_API_CALL +clImportMemoryARM( cl_context context, + cl_mem_flags flags, + const cl_import_properties_arm *properties, + void *memory, + size_t size, + cl_int *errcode_ret) CL_EXT_SUFFIX__VERSION_1_0; + + +/****************************************** + * cl_arm_shared_virtual_memory extension * + ******************************************/ +#define cl_arm_shared_virtual_memory 1 + +/* Used by clGetDeviceInfo */ +#define CL_DEVICE_SVM_CAPABILITIES_ARM 0x40B6 + +/* Used by clGetMemObjectInfo */ +#define CL_MEM_USES_SVM_POINTER_ARM 0x40B7 + +/* Used by clSetKernelExecInfoARM: */ +#define CL_KERNEL_EXEC_INFO_SVM_PTRS_ARM 0x40B8 +#define CL_KERNEL_EXEC_INFO_SVM_FINE_GRAIN_SYSTEM_ARM 0x40B9 + +/* To be used by clGetEventInfo: */ +#define CL_COMMAND_SVM_FREE_ARM 0x40BA +#define CL_COMMAND_SVM_MEMCPY_ARM 0x40BB +#define CL_COMMAND_SVM_MEMFILL_ARM 0x40BC +#define CL_COMMAND_SVM_MAP_ARM 0x40BD +#define CL_COMMAND_SVM_UNMAP_ARM 0x40BE + +/* Flag values returned by clGetDeviceInfo with CL_DEVICE_SVM_CAPABILITIES_ARM as the param_name. */ +#define CL_DEVICE_SVM_COARSE_GRAIN_BUFFER_ARM (1 << 0) +#define CL_DEVICE_SVM_FINE_GRAIN_BUFFER_ARM (1 << 1) +#define CL_DEVICE_SVM_FINE_GRAIN_SYSTEM_ARM (1 << 2) +#define CL_DEVICE_SVM_ATOMICS_ARM (1 << 3) + +/* Flag values used by clSVMAllocARM: */ +#define CL_MEM_SVM_FINE_GRAIN_BUFFER_ARM (1 << 10) +#define CL_MEM_SVM_ATOMICS_ARM (1 << 11) + +typedef cl_bitfield cl_svm_mem_flags_arm; +typedef cl_uint cl_kernel_exec_info_arm; +typedef cl_bitfield cl_device_svm_capabilities_arm; + +extern CL_API_ENTRY void * CL_API_CALL +clSVMAllocARM(cl_context context, + cl_svm_mem_flags_arm flags, + size_t size, + cl_uint alignment) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY void CL_API_CALL +clSVMFreeARM(cl_context context, + void * svm_pointer) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMFreeARM(cl_command_queue command_queue, + cl_uint num_svm_pointers, + void * svm_pointers[], + void (CL_CALLBACK * pfn_free_func)(cl_command_queue queue, + cl_uint num_svm_pointers, + void * svm_pointers[], + void * user_data), + void * user_data, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMMemcpyARM(cl_command_queue command_queue, + cl_bool blocking_copy, + void * dst_ptr, + const void * src_ptr, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMMemFillARM(cl_command_queue command_queue, + void * svm_ptr, + const void * pattern, + size_t pattern_size, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMMapARM(cl_command_queue command_queue, + cl_bool blocking_map, + cl_map_flags flags, + void * svm_ptr, + size_t size, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clEnqueueSVMUnmapARM(cl_command_queue command_queue, + void * svm_ptr, + cl_uint num_events_in_wait_list, + const cl_event * event_wait_list, + cl_event * event) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetKernelArgSVMPointerARM(cl_kernel kernel, + cl_uint arg_index, + const void * arg_value) CL_EXT_SUFFIX__VERSION_1_2; + +extern CL_API_ENTRY cl_int CL_API_CALL +clSetKernelExecInfoARM(cl_kernel kernel, + cl_kernel_exec_info_arm param_name, + size_t param_value_size, + const void * param_value) CL_EXT_SUFFIX__VERSION_1_2; + +/******************************** + * cl_arm_get_core_id extension * + ********************************/ + +#ifdef CL_VERSION_1_2 + +#define cl_arm_get_core_id 1 + +/* Device info property for bitfield of cores present */ +#define CL_DEVICE_COMPUTE_UNITS_BITFIELD_ARM 0x40BF + +#endif /* CL_VERSION_1_2 */ + +/********************************* +* cl_arm_job_slot_selection +*********************************/ + +#define cl_arm_job_slot_selection 1 + +/* cl_device_info */ +#define CL_DEVICE_JOB_SLOTS_ARM 0x41E0 + +/* cl_command_queue_properties */ +#define CL_QUEUE_JOB_SLOT_ARM 0x41E1 + +/********************************* +* cl_arm_scheduling_controls +*********************************/ + +#define cl_arm_scheduling_controls 1 + +/* cl_device_info */ +#define CL_DEVICE_SCHEDULING_CONTROLS_CAPABILITIES_ARM 0x41E4 + +#define CL_DEVICE_SCHEDULING_KERNEL_BATCHING_ARM (1 << 0) +#define CL_DEVICE_SCHEDULING_WORKGROUP_BATCH_SIZE_ARM (1 << 1) +#define CL_DEVICE_SCHEDULING_WORKGROUP_BATCH_SIZE_MODIFIER_ARM (1 << 2) + +/* cl_kernel_info */ +#define CL_KERNEL_EXEC_INFO_WORKGROUP_BATCH_SIZE_ARM 0x41E5 +#define CL_KERNEL_EXEC_INFO_WORKGROUP_BATCH_SIZE_MODIFIER_ARM 0x41E6 + +/* cl_queue_properties */ +#define CL_QUEUE_KERNEL_BATCHING_ARM 0x41E7 + +#ifdef __cplusplus +} +#endif + +#endif /* __CL_EXT_H */ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/async.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/async.h new file mode 100644 index 0000000000000000000000000000000000000000..1b7dcb2433f2cb7d1ef61290995ac871a901b1e8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/async.h @@ -0,0 +1,452 @@ +/* Copyright 1993-2016 NVIDIA Corporation. All rights reserved. + * + * NOTICE TO LICENSEE: + * + * The source code and/or documentation ("Licensed Deliverables") are + * subject to NVIDIA intellectual property rights under U.S. and + * international Copyright laws. + * + * The Licensed Deliverables contained herein are PROPRIETARY and + * CONFIDENTIAL to NVIDIA and are being provided under the terms and + * conditions of a form of NVIDIA software license agreement by and + * between NVIDIA and Licensee ("License Agreement") or electronically + * accepted by Licensee. Notwithstanding any terms or conditions to + * the contrary in the License Agreement, reproduction or disclosure + * of the Licensed Deliverables to any third party without the express + * written consent of NVIDIA is prohibited. + * + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE + * SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. THEY ARE + * PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND. + * NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED + * DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY, + * NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE. + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY + * SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY + * DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, + * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS + * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE + * OF THESE LICENSED DELIVERABLES. + * + * U.S. Government End Users. These Licensed Deliverables are a + * "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT + * 1995), consisting of "commercial computer software" and "commercial + * computer software documentation" as such terms are used in 48 + * C.F.R. 12.212 (SEPT 1995) and are provided to the U.S. Government + * only as a commercial end item. Consistent with 48 C.F.R.12.212 and + * 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all + * U.S. Government End Users acquire the Licensed Deliverables with + * only those rights set forth herein. + * + * Any use of the Licensed Deliverables in individual and commercial + * software must include, in the user documentation and internal + * comments to the code, the above Disclaimer and U.S. Government End + * Users Notice. + */ + +#ifndef _CG_ASYNC_H +#define _CG_ASYNC_H + +#include "helpers.h" +#include "info.h" + +#include + +_CG_BEGIN_NAMESPACE + +namespace details { +// Groups supported by memcpy_async +template +struct _async_copy_group_supported : public _CG_STL_NAMESPACE::false_type {}; + +template +struct _async_copy_group_supported> + : public _CG_STL_NAMESPACE::true_type {}; +template <> +struct _async_copy_group_supported : public _CG_STL_NAMESPACE::true_type {}; +template <> +struct _async_copy_group_supported : public _CG_STL_NAMESPACE::true_type {}; + +template +using async_copy_group_supported = _async_copy_group_supported>; + +// Groups that require optimization +template +struct _async_copy_optimize_tile : public _CG_STL_NAMESPACE::false_type {}; + +template +struct _async_copy_optimize_tile> + : public _CG_STL_NAMESPACE::false_type {}; + +template +struct _async_copy_optimize_tile> + : public _CG_STL_NAMESPACE::true_type {}; + +template +using async_copy_optimize_tile = _async_copy_optimize_tile>; + +// SFINAE helpers for tile optimizations +template +using enable_tile_optimization = + typename _CG_STL_NAMESPACE::enable_if::value, void *>::type; + +template +using disable_tile_optimization = + typename _CG_STL_NAMESPACE::enable_if::value, void *>::type; + +// Segment for punning to aligned types +template +struct _Segment { + int _seg[N]; +}; + +// Trivial layout guaranteed-aligned copy-async compatible segments +template +struct Segment; +template <> +struct __align__(4) Segment<1> : public _Segment<1>{}; +template <> +struct __align__(8) Segment<2> : public _Segment<2>{}; +template <> +struct __align__(16) Segment<4> : public _Segment<4>{}; + +// Interleaved element by element copies from source to dest +template +_CG_STATIC_QUALIFIER void inline_copy(TyGroup &group, TyElem *__restrict__ dst, const TyElem *__restrict__ src, + size_t count) { + const unsigned int rank = group.thread_rank(); + const unsigned int stride = group.size(); + + for (size_t idx = rank; idx < count; idx += stride) { + dst[idx] = src[idx]; + } +} + +template = nullptr> +_CG_STATIC_QUALIFIER void accelerated_async_copy(TyGroup &group, TyElem *__restrict__ dst, + const TyElem *__restrict__ src, size_t count) { + static_assert(async_copy_group_supported::value, + "Async copy is only supported for groups that represent private shared memory"); + + if (count == 0) { + return; + } + + const bool dstIsNotShared = !__isShared(dst); + const bool srcIsNotGlobal = !__isGlobal(src); + + if (dstIsNotShared || srcIsNotGlobal) { + inline_copy(group, dst, src, count); + return; + } + + const unsigned int stride = group.size(); + const unsigned int rank = group.thread_rank(); + // Efficient copies require warps to operate on the same amount of work at each step. + // remainders are handled in a separate stage to prevent branching + const unsigned int subWarpMask = (stride - 1); + const unsigned int subwarpCopies = (subWarpMask & (unsigned int)count); + const unsigned int maxSubwarpRank = min(rank, subwarpCopies - 1); + + const size_t warpCopies = (count & (~subWarpMask)); + + for (size_t idx = 0; idx < warpCopies; idx += stride) { + size_t _srcIdx = rank + idx; + size_t _dstIdx = rank + idx; + __pipeline_memcpy_async(dst + _dstIdx, src + _srcIdx, sizeof(TyElem)); + } + + if (subwarpCopies) { + size_t _srcIdx = warpCopies + maxSubwarpRank; + size_t _dstIdx = warpCopies + maxSubwarpRank; + __pipeline_memcpy_async(dst + _dstIdx, src + _srcIdx, sizeof(TyElem)); + } +} + +template = nullptr> +_CG_STATIC_QUALIFIER void accelerated_async_copy(TyGroup &group, TyElem *__restrict__ dst, + const TyElem *__restrict__ src, size_t count) { + static_assert(async_copy_group_supported::value, + "Async copy is only supported for groups that represent private shared memory"); + + const bool dstIsNotShared = !__isShared(dst); + const bool srcIsNotGlobal = !__isGlobal(src); + + if (dstIsNotShared || srcIsNotGlobal) { + inline_copy(group, dst, src, count); + return; + } + + unsigned int stride = group.size(); + unsigned int rank = group.thread_rank(); + + for (size_t idx = rank; idx < count; idx += stride) { + size_t _srcIdx = idx; + size_t _dstIdx = idx; + __pipeline_memcpy_async(dst + _dstIdx, src + _srcIdx, sizeof(TyElem)); + } +} + +// Determine best possible alignment given an input and initial conditions +// Attempts to generate as little code as possible, most likely should only be used with 1 and 2 byte alignments +template +_CG_STATIC_QUALIFIER uint32_t find_best_alignment(void *__restrict__ dst, const void *__restrict__ src) { + // Narrowing conversion intentional + uint32_t base1 = (uint32_t) reinterpret_cast(src); + uint32_t base2 = (uint32_t) reinterpret_cast(dst); + + uint32_t diff = ((base1) ^ (base2)) & (MaxAlignment - 1); + + // range [MaxAlignment, alignof(elem)], step: x >> 1 + // over range of possible alignments, choose best available out of range + uint32_t out = MaxAlignment; +#pragma unroll + for (uint32_t alignment = (MaxAlignment >> 1); alignment >= MinAlignment; alignment >>= 1) { + if (alignment & diff) + out = alignment; + } + + return out; +} + +// Determine best possible alignment given an input and initial conditions +// Attempts to generate as little code as possible, most likely should only be used with 1 and 2 byte alignments +template +_CG_STATIC_QUALIFIER void copy_like(const TyGroup &group, void *__restrict__ _dst, const void *__restrict__ _src, + size_t count) { + const char *src = reinterpret_cast(_src); + char *dst = reinterpret_cast(_dst); + + constexpr uint32_t targetAlignment = (uint32_t)alignof(TyType); + + uint32_t base = (uint32_t) reinterpret_cast(src); + uint32_t alignOffset = ((~base) + 1) & (targetAlignment - 1); + + inline_copy(group, dst, src, alignOffset); + count -= alignOffset; + src += alignOffset; + dst += alignOffset; + + // Copy using the best available alignment, async_copy expects n-datums, not bytes + size_t asyncCount = count / sizeof(TyType); + accelerated_async_copy(group, reinterpret_cast(dst), reinterpret_cast(src), asyncCount); + asyncCount *= sizeof(TyType); + + count -= asyncCount; + src += asyncCount; + dst += asyncCount; + inline_copy(group, dst, src, count); +} + +// We must determine alignment and manually align src/dst ourselves +template +struct _memcpy_async_align_dispatch { + template + _CG_STATIC_QUALIFIER void copy(TyGroup &group, void *__restrict__ dst, const void *__restrict__ src, size_t count) { + uint32_t alignment = find_best_alignment(dst, src); + + // Avoid copying the extra bytes if desired copy count is smaller + alignment = count < alignment ? AlignHint : alignment; + + switch (alignment) { + default: + case 1: + inline_copy(group, reinterpret_cast(dst), reinterpret_cast(src), count); + break; + case 2: + inline_copy(group, reinterpret_cast(dst), reinterpret_cast(src), count >> 1); + break; + case 4: + copy_like>(group, dst, src, count); + break; + case 8: + copy_like>(group, dst, src, count); + break; + case 16: + copy_like>(group, dst, src, count); + break; + } + } +}; + +// Specialization for 4 byte alignments +template <> +struct _memcpy_async_align_dispatch<4> { + template + _CG_STATIC_QUALIFIER void copy(TyGroup &group, void *__restrict__ _dst, const void *__restrict__ _src, + size_t count) { + const Segment<1> *src = reinterpret_cast *>(_src); + Segment<1> *dst = reinterpret_cast *>(_dst); + + // Dispatch straight to aligned LDGSTS calls + accelerated_async_copy(group, dst, src, count / sizeof(*dst)); + } +}; + +// Specialization for 8 byte alignments +template <> +struct _memcpy_async_align_dispatch<8> { + template + _CG_STATIC_QUALIFIER void copy(TyGroup &group, void *__restrict__ _dst, const void *__restrict__ _src, + size_t count) { + const Segment<2> *src = reinterpret_cast *>(_src); + Segment<2> *dst = reinterpret_cast *>(_dst); + + // Dispatch straight to aligned LDGSTS calls + accelerated_async_copy(group, dst, src, count / sizeof(*dst)); + } +}; + +// Alignments over 16 are truncated to 16 and bypass alignment +// This is the highest performing memcpy available +template <> +struct _memcpy_async_align_dispatch<16> { + template + _CG_STATIC_QUALIFIER void copy(TyGroup &group, void *__restrict__ _dst, const void *__restrict__ _src, + size_t count) { + const Segment<4> *src = reinterpret_cast *>(_src); + Segment<4> *dst = reinterpret_cast *>(_dst); + + // Dispatch straight to aligned LDGSTS calls + accelerated_async_copy(group, dst, src, count / sizeof(*dst)); + } +}; + +// byte-wide API +template +_CG_STATIC_QUALIFIER void _memcpy_async_dispatch_to_aligned_copy(const TyGroup &group, void *__restrict__ _dst, + const void *__restrict__ _src, size_t count) { + static_assert(!(Alignment & (Alignment - 1)), "Known static alignment dispatch must be a power of 2"); + details::_memcpy_async_align_dispatch::copy(group, _dst, _src, count); +} + +// Internal dispatch APIs +// These deduce the alignments and sizes necessary to invoke the underlying copy engine +template +using is_void = _CG_STL_NAMESPACE::is_same; + +template +using enable_if_not_void = typename _CG_STL_NAMESPACE::enable_if::value, void *>::type; + +template +using enable_if_void = typename _CG_STL_NAMESPACE::enable_if::value, void *>::type; + +template +using enable_if_integral = + typename _CG_STL_NAMESPACE::enable_if<_CG_STL_NAMESPACE::is_integral::value, void *>::type; + +// byte-wide API using aligned_sized_t +template typename Alignment, size_t Hint> +_CG_STATIC_QUALIFIER void _memcpy_async_bytes(const TyGroup &group, void *__restrict__ _dst, + const void *__restrict__ _src, const Alignment &count) { + constexpr size_t _align = (Hint > 16) ? 16 : Hint; + + details::_memcpy_async_dispatch_to_aligned_copy<_align>(group, _dst, _src, (size_t)count); +} + +// byte-wide API using type for aligment +template = nullptr, enable_if_integral = nullptr> +_CG_STATIC_QUALIFIER void _memcpy_async_bytes(const TyGroup &group, TyElem *__restrict__ _dst, + const TyElem *__restrict__ _src, const TySize& count) { + constexpr size_t _align = (Hint > 16) ? 16 : Hint; + + details::_memcpy_async_dispatch_to_aligned_copy<_align>(group, _dst, _src, count); +} + +// byte-wide API with full alignment deduction required +template = nullptr, + enable_if_integral = nullptr> +_CG_STATIC_QUALIFIER void _memcpy_async_bytes(const TyGroup &group, TyElem *__restrict__ _dst, + const TyElem *__restrict__ _src, const TySize& count) { + details::_memcpy_async_dispatch_to_aligned_copy<1>(group, _dst, _src, count); +} + +// 1d-datum API +template +_CG_STATIC_QUALIFIER void _memcpy_async_datum(const TyGroup &group, TyElem *__restrict__ dst, const size_t dstCount, + const TyElem *__restrict__ src, const size_t srcCount) { + constexpr unsigned int _align = Hint; + const size_t totalCount = min(dstCount, srcCount) * sizeof(TyElem); + + details::_memcpy_async_dispatch_to_aligned_copy<_align>(group, dst, src, totalCount); +} + +// 1d-datum API using aligned_size_t +template typename Alignment, size_t Hint> +_CG_STATIC_QUALIFIER void _memcpy_async_datum(const TyGroup &group, TyElem *__restrict__ dst, const Alignment &dstCount, + const TyElem *__restrict__ src, const Alignment &srcCount) { + constexpr unsigned int _align = Hint; + const size_t totalCount = min((size_t)dstCount, (size_t)srcCount) * sizeof(TyElem); + + details::_memcpy_async_dispatch_to_aligned_copy<_align>(group, dst, src, totalCount); +} + +} // namespace details + +/* + * Group submit batch of async-copy to cover contiguous 1D array + * and commit that batch to eventually wait for completion. + */ +template +_CG_STATIC_QUALIFIER void memcpy_async(const TyGroup &group, TyElem *__restrict__ _dst, const TyElem *__restrict__ _src, + const TySizeT &count) { + details::_memcpy_async_bytes(group, _dst, _src, count); + __pipeline_commit(); +} + +/* + * Group submit batch of async-copy to cover contiguous 1D array + * and commit that batch to eventually wait for completion. + * Object counts are in datum sized chunks, not bytes. + */ +template +_CG_STATIC_QUALIFIER void memcpy_async(const TyGroup &group, TyElem *__restrict__ dst, const DstLayout &dstLayout, + const TyElem *__restrict__ src, const SrcLayout &srcLayout) { + details::_memcpy_async_datum(group, dst, dstLayout, src, srcLayout); + __pipeline_commit(); +} + +/* Group wait for prior Nth stage of memcpy_async to complete. */ +template +_CG_STATIC_QUALIFIER void wait_prior(const TyGroup &group) { + __pipeline_wait_prior(Stage); + group.sync(); +} + +/* Group wait all previously submitted memcpy_async to complete. */ +template +_CG_STATIC_QUALIFIER void wait(const TyGroup &group) { + __pipeline_wait_prior(0); + group.sync(); +} + +/***************** CG APIs including pipeline are deprecated *****************/ + +/* Group submit batch of async-copy to cover of contiguous 1D array + to a pipeline and commit the batch*/ +template +_CG_DEPRECATED _CG_STATIC_QUALIFIER void memcpy_async(TyGroup &group, TyElem *dst, size_t dstCount, const TyElem *src, size_t srcCount, + nvcuda::experimental::pipeline &pipe) { + details::_memcpy_async_datum(group, dst, dstCount, src, srcCount); + pipe.commit(); +} + +/* Group wait for prior Nth stage of memcpy_async to complete. */ +template +_CG_DEPRECATED _CG_STATIC_QUALIFIER void wait_prior(TyGroup &group, nvcuda::experimental::pipeline &pipe) { + pipe.wait_prior(); + group.sync(); +} + +/* Group wait for stage-S of memcpy_async to complete. */ +template +_CG_DEPRECATED _CG_STATIC_QUALIFIER void wait(TyGroup &group, nvcuda::experimental::pipeline &pipe, size_t stage) { + pipe.wait(stage); + group.sync(); +} +_CG_END_NAMESPACE + +#endif // _CG_ASYNC_H diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/coalesced_scan.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/coalesced_scan.h new file mode 100644 index 0000000000000000000000000000000000000000..383f4bde059dd8daad7d1c56e99152ea7ee28a08 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/coalesced_scan.h @@ -0,0 +1,174 @@ +/* Copyright 1993-2016 NVIDIA Corporation. All rights reserved. + * + * NOTICE TO LICENSEE: + * + * The source code and/or documentation ("Licensed Deliverables") are + * subject to NVIDIA intellectual property rights under U.S. and + * international Copyright laws. + * + * The Licensed Deliverables contained herein are PROPRIETARY and + * CONFIDENTIAL to NVIDIA and are being provided under the terms and + * conditions of a form of NVIDIA software license agreement by and + * between NVIDIA and Licensee ("License Agreement") or electronically + * accepted by Licensee. Notwithstanding any terms or conditions to + * the contrary in the License Agreement, reproduction or disclosure + * of the Licensed Deliverables to any third party without the express + * written consent of NVIDIA is prohibited. + * + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE + * SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. THEY ARE + * PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND. + * NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED + * DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY, + * NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE. + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY + * SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY + * DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, + * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS + * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE + * OF THESE LICENSED DELIVERABLES. + * + * U.S. Government End Users. These Licensed Deliverables are a + * "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT + * 1995), consisting of "commercial computer software" and "commercial + * computer software documentation" as such terms are used in 48 + * C.F.R. 12.212 (SEPT 1995) and are provided to the U.S. Government + * only as a commercial end item. Consistent with 48 C.F.R.12.212 and + * 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all + * U.S. Government End Users acquire the Licensed Deliverables with + * only those rights set forth herein. + * + * Any use of the Licensed Deliverables in individual and commercial + * software must include, in the user documentation and internal + * comments to the code, the above Disclaimer and U.S. Government End + * Users Notice. + */ + +#ifndef _CG_COALESCED_SCAN_H_ +#define _CG_COALESCED_SCAN_H_ + +#include "info.h" +#include "helpers.h" +#include "cooperative_groups.h" +#include "partitioning.h" +#include "functional.h" + +_CG_BEGIN_NAMESPACE + +namespace details { + +template +_CG_QUALIFIER auto inclusive_scan_contiguous(const TyGroup& group, TyVal&& val, TyOp&& op) -> decltype(op(val, val)) { + auto out = val; + for (int mask = 1; mask < group.size(); mask <<= 1) { + auto tmp = group.shfl_up(out, mask); + if (mask <= group.thread_rank()) { + out = op(out, tmp); + } + } + + return out; +} + +template +_CG_QUALIFIER auto inclusive_scan_non_contiguous(const TyGroup& group, TyVal&& val, TyOp&& op) -> decltype(op(val, val)) { + const unsigned int groupSize = group.size(); + auto out = val; + + const unsigned int mask = details::_coalesced_group_data_access::get_mask(group); + unsigned int lanemask = details::lanemask32_lt() & mask; + unsigned int srcLane = details::laneid(); + + const unsigned int base = __ffs(mask)-1; /* lane with rank == 0 */ + const unsigned int rank = __popc(lanemask); + + for (unsigned int i = 1, j = 1; i < groupSize; i <<= 1) { + if (i <= rank) { + srcLane -= j; + j = i; /* maximum possible lane */ + + unsigned int begLane = base + rank - i; /* minimum possible lane */ + + /* Next source lane is in the range [ begLane .. srcLane ] + * If begLane < srcLane then do a binary search. + */ + while (begLane < srcLane) { + const unsigned int halfLane = (begLane + srcLane) >> 1; + const unsigned int halfMask = lanemask >> halfLane; + const unsigned int d = __popc(halfMask); + if (d < i) { + srcLane = halfLane - 1; /* halfLane too large */ + } + else if ((i < d) || !(halfMask & 0x01)) { + begLane = halfLane + 1; /* halfLane too small */ + } + else { + begLane = srcLane = halfLane; /* happen to hit */ + } + } + } + + auto tmp = details::tile::shuffle_dispatch::shfl(out, mask, srcLane, 32); + if (i <= rank) { + out = op(out, tmp); + } + } + return out; +} + +template +_CG_QUALIFIER auto coalesced_inclusive_scan(const __single_warp_thread_block_tile& group, + TyVal&& val, + TyOp&& op) -> decltype(op(val, val)) { + return inclusive_scan_contiguous(group, _CG_STL_NAMESPACE::forward(val), _CG_STL_NAMESPACE::forward(op)); +} + +template +_CG_QUALIFIER auto coalesced_inclusive_scan(const coalesced_group& group, TyVal&& val, TyOp&& op) -> decltype(op(val, val)) { + if (group.size() == 32) { + return inclusive_scan_contiguous(group, _CG_STL_NAMESPACE::forward(val), _CG_STL_NAMESPACE::forward(op)); + } + else { + return inclusive_scan_non_contiguous(group, _CG_STL_NAMESPACE::forward(val), _CG_STL_NAMESPACE::forward(op)); + } +} + +template +struct scan_choose_convertion; + +template<> +struct scan_choose_convertion { + template + _CG_STATIC_QUALIFIER details::remove_qual convert_inclusive_to_exclusive(const TyGroup& group, TyRes& result, TyVal&& val) { + return result - val; + } +}; + +template<> +struct scan_choose_convertion { + template + _CG_STATIC_QUALIFIER details::remove_qual convert_inclusive_to_exclusive(const TyGroup& group, TyRes& result, TyVal&& val) { + auto ret = group.shfl_up(result, 1); + if (group.thread_rank() == 0) { + return {}; + } + else { + return ret; + } + } +}; + +template +_CG_QUALIFIER auto convert_inclusive_to_exclusive(const TyGroup& group, TyRes& result, TyVal&& val, TyFn&& op) -> decltype(op(val, val)) { + using conversion = scan_choose_convertion<_CG_STL_NAMESPACE::is_same, cooperative_groups::plus>>::value + && _CG_STL_NAMESPACE::is_integral>::value>; + return conversion::convert_inclusive_to_exclusive(group, result, _CG_STL_NAMESPACE::forward(val)); +} + +} // details + +_CG_END_NAMESPACE + +#endif // _CG_COALESCED_SCAN_H_ \ No newline at end of file diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/driver_abi.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/driver_abi.h new file mode 100644 index 0000000000000000000000000000000000000000..9c866fcf740beb709a106057d28e8a2a1ac37924 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/driver_abi.h @@ -0,0 +1,99 @@ + /* Copyright 1993-2016 NVIDIA Corporation. All rights reserved. + * + * NOTICE TO LICENSEE: + * + * The source code and/or documentation ("Licensed Deliverables") are + * subject to NVIDIA intellectual property rights under U.S. and + * international Copyright laws. + * + * The Licensed Deliverables contained herein are PROPRIETARY and + * CONFIDENTIAL to NVIDIA and are being provided under the terms and + * conditions of a form of NVIDIA software license agreement by and + * between NVIDIA and Licensee ("License Agreement") or electronically + * accepted by Licensee. Notwithstanding any terms or conditions to + * the contrary in the License Agreement, reproduction or disclosure + * of the Licensed Deliverables to any third party without the express + * written consent of NVIDIA is prohibited. + * + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE + * SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. THEY ARE + * PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND. + * NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED + * DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY, + * NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE. + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY + * SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY + * DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, + * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS + * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE + * OF THESE LICENSED DELIVERABLES. + * + * U.S. Government End Users. These Licensed Deliverables are a + * "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT + * 1995), consisting of "commercial computer software" and "commercial + * computer software documentation" as such terms are used in 48 + * C.F.R. 12.212 (SEPT 1995) and are provided to the U.S. Government + * only as a commercial end item. Consistent with 48 C.F.R.12.212 and + * 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all + * U.S. Government End Users acquire the Licensed Deliverables with + * only those rights set forth herein. + * + * Any use of the Licensed Deliverables in individual and commercial + * software must include, in the user documentation and internal + * comments to the code, the above Disclaimer and U.S. Government End + * Users Notice. + */ + +#ifndef _CG_DRIVER_API_H +#define _CG_DRIVER_API_H + +#include "info.h" + +_CG_BEGIN_NAMESPACE + +namespace details { + template + _CG_QUALIFIER unsigned int load_env_reg() { + // Abort by default + _CG_ABORT(); + return 0; + } + + template + _CG_QUALIFIER unsigned long long load_env_reg64() { + unsigned long long registerLo = load_env_reg(); + unsigned long long registerHi = load_env_reg(); + + return (registerHi << 32) | registerLo; + } + +// inline PTX for accessing registers requires an immediate for the special reg +# define LOAD_ENVREG(NUMBER) \ + template <> _CG_QUALIFIER unsigned int load_env_reg() { \ + unsigned int r; \ + asm ("mov.u32 %0, %%envreg" #NUMBER ";" : "=r"(r)); \ + return r; \ + } + + // Instantiate loaders for registers used + LOAD_ENVREG(0); + LOAD_ENVREG(1); + LOAD_ENVREG(2); +# undef LOAD_ENVREG + + struct grid_workspace { + unsigned int wsSize; + unsigned int barrier; + }; + + _CG_QUALIFIER grid_workspace* get_grid_workspace() { + unsigned long long gridWsAbiAddress = load_env_reg64<1, 2>(); + // Interpret the address from envreg 1 and 2 as the driver's grid workspace + return (reinterpret_cast(gridWsAbiAddress)); + } +} +_CG_END_NAMESPACE + +#endif // _CG_DRIVER_API_H diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/sync.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/sync.h new file mode 100644 index 0000000000000000000000000000000000000000..af144a68a877a00f662e2f6c36417f284e47d337 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/details/sync.h @@ -0,0 +1,276 @@ + /* Copyright 1993-2016 NVIDIA Corporation. All rights reserved. + * + * NOTICE TO LICENSEE: + * + * The source code and/or documentation ("Licensed Deliverables") are + * subject to NVIDIA intellectual property rights under U.S. and + * international Copyright laws. + * + * The Licensed Deliverables contained herein are PROPRIETARY and + * CONFIDENTIAL to NVIDIA and are being provided under the terms and + * conditions of a form of NVIDIA software license agreement by and + * between NVIDIA and Licensee ("License Agreement") or electronically + * accepted by Licensee. Notwithstanding any terms or conditions to + * the contrary in the License Agreement, reproduction or disclosure + * of the Licensed Deliverables to any third party without the express + * written consent of NVIDIA is prohibited. + * + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE + * SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. THEY ARE + * PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND. + * NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED + * DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY, + * NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE. + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY + * SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY + * DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, + * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS + * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE + * OF THESE LICENSED DELIVERABLES. + * + * U.S. Government End Users. These Licensed Deliverables are a + * "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT + * 1995), consisting of "commercial computer software" and "commercial + * computer software documentation" as such terms are used in 48 + * C.F.R. 12.212 (SEPT 1995) and are provided to the U.S. Government + * only as a commercial end item. Consistent with 48 C.F.R.12.212 and + * 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all + * U.S. Government End Users acquire the Licensed Deliverables with + * only those rights set forth herein. + * + * Any use of the Licensed Deliverables in individual and commercial + * software must include, in the user documentation and internal + * comments to the code, the above Disclaimer and U.S. Government End + * Users Notice. + */ + +#ifndef _CG_GRID_H +#define _CG_GRID_H + +#include "info.h" + +_CG_BEGIN_NAMESPACE + +namespace details +{ + +typedef unsigned int barrier_t; + +_CG_STATIC_QUALIFIER bool bar_has_flipped(unsigned int old_arrive, unsigned int current_arrive) { + return (((old_arrive ^ current_arrive) & 0x80000000) != 0); +} + +_CG_STATIC_QUALIFIER void bar_flush(volatile unsigned int *addr) { +#if __CUDA_ARCH__ < 700 + __threadfence(); +#else + unsigned int val; + asm volatile("ld.acquire.gpu.u32 %0,[%1];" : "=r"(val) : _CG_ASM_PTR_CONSTRAINT((unsigned int*)addr) : "memory"); + // Avoids compiler warnings from unused variable val + (void)(val = val); +#endif +} + +_CG_STATIC_QUALIFIER unsigned int atomic_add(volatile unsigned int *addr, unsigned int val) { + unsigned int old; +#if __CUDA_ARCH__ < 700 + old = atomicAdd((unsigned int*)addr, val); +#else + asm volatile("atom.add.release.gpu.u32 %0,[%1],%2;" : "=r"(old) : _CG_ASM_PTR_CONSTRAINT((unsigned int*)addr), "r"(val) : "memory"); +#endif + return old; +} + +_CG_STATIC_QUALIFIER void sync_grids(unsigned int expected, volatile barrier_t *arrived) { + bool cta_master = (threadIdx.x + threadIdx.y + threadIdx.z == 0); + bool gpu_master = (blockIdx.x + blockIdx.y + blockIdx.z == 0); + + __syncthreads(); + + if (cta_master) { + unsigned int nb = 1; + if (gpu_master) { + nb = 0x80000000 - (expected - 1); + } + + __threadfence(); + + unsigned int oldArrive; + oldArrive = atomic_add(arrived, nb); + + while (!bar_has_flipped(oldArrive, *arrived)); + + //flush barrier upon leaving + bar_flush((unsigned int*)arrived); + } + + __syncthreads(); +} + +/* - Multi warp groups synchronization routines - */ + +// Get synchronization bit mask of my thread_block_tile of size num_warps. Thread ranks 0..31 have the first bit assigned to them, +// thread ranks 32..63 second etc +// Bit masks are unique for each group, groups of the same size will have the same number of bits set, but on different positions +_CG_STATIC_QUALIFIER unsigned int get_group_mask(unsigned int thread_rank, unsigned int num_warps) { + return num_warps == 32 ? ~0 : ((1 << num_warps) - 1) << (num_warps * (thread_rank / (num_warps * 32))); +} + +// Default blocking sync. +_CG_STATIC_QUALIFIER void sync_warps(volatile barrier_t *arrived, unsigned int thread_rank, unsigned int num_warps) { + unsigned int warp_id = thread_rank / 32; + bool warp_master = (thread_rank % 32 == 0); + unsigned int warp_bit = 1 << warp_id; + unsigned int group_mask = get_group_mask(thread_rank, num_warps); + + __syncwarp(0xFFFFFFFF); + + if (warp_master) { + unsigned int old = atomicOr((unsigned int *)arrived, warp_bit); + if (((old | warp_bit) & group_mask) == group_mask) { + atomicAnd((unsigned int *)arrived, ~group_mask); + } + else { + while(*arrived & warp_bit); + } + } + + __syncwarp(0xFFFFFFFF); +} + +// Blocking sync, except the last arriving warp, that releases other warps, returns to do other stuff first. +// Warp returning true from this function needs to call sync_warps_release. +_CG_STATIC_QUALIFIER bool sync_warps_last_releases(volatile barrier_t *arrived, unsigned int thread_rank, unsigned int num_warps) { + unsigned int warp_id = thread_rank / 32; + bool warp_master = (thread_rank % 32 == 0); + unsigned int warp_bit = 1 << warp_id; + unsigned int group_mask = get_group_mask(thread_rank, num_warps); + + __syncwarp(0xFFFFFFFF); + + unsigned int old = 0; + if (warp_master) { + old = atomicOr((unsigned int *)arrived, warp_bit); + } + old = __shfl_sync(0xFFFFFFFF, old, 0); + if (((old | warp_bit) & group_mask) == group_mask) { + return true; + } + while(*arrived & warp_bit); + + return false; +} + +// Release my group from the barrier. +_CG_STATIC_QUALIFIER void sync_warps_release(volatile barrier_t *arrived, bool is_master, unsigned int thread_rank, unsigned int num_warps) { + unsigned int group_mask = get_group_mask(thread_rank, num_warps); + if (is_master) { + atomicAnd((unsigned int *)arrived, ~group_mask); + } +} + +// Arrive at my group barrier, but don't block or release the barrier, even if every one arrives. +// sync_warps_release needs to be called by some warp after this one to reset the barrier. +_CG_STATIC_QUALIFIER void sync_warps_arrive(volatile barrier_t *arrived, unsigned int thread_rank, unsigned int num_warps) { + unsigned int warp_id = thread_rank / 32; + bool warp_master = (thread_rank % 32 == 0); + unsigned int warp_bit = 1 << warp_id; + unsigned int group_mask = get_group_mask(thread_rank, num_warps); + + __syncwarp(0xFFFFFFFF); + + if (warp_master) { + unsigned int old = atomicOr((unsigned int *)arrived, warp_bit); + } + __syncwarp(0xFFFFFFFF); +} + +// Arrive at my group barrier, but don't block. Last arriving warp immediately releases the barrier. +_CG_STATIC_QUALIFIER void sync_warps_arrive_release(volatile barrier_t *arrived, unsigned int thread_rank, unsigned int num_warps) { + unsigned int warp_id = thread_rank / 32; + bool warp_master = (thread_rank % 32 == 0); + unsigned int warp_bit = 1 << warp_id; + unsigned int group_mask = get_group_mask(thread_rank, num_warps); + + __syncwarp(0xFFFFFFFF); + + if (warp_master) { + unsigned int old = atomicOr((unsigned int *)arrived, warp_bit); + if (((old | warp_bit) & group_mask) == group_mask) { + atomicAnd((unsigned int *)arrived, ~group_mask); + } + } + __syncwarp(0xFFFFFFFF); +} + +// Wait for my warp to be released from the barrier. Warp must have arrived first. +_CG_STATIC_QUALIFIER void sync_warps_wait(volatile barrier_t *arrived, unsigned int thread_rank) { + unsigned int warp_id = thread_rank / 32; + unsigned int warp_bit = 1 << warp_id; + + while(*arrived & warp_bit); + __syncwarp(0xFFFFFFFF); +} + +// Arrive at my group barrier and block. Barrier is not released, even if every warp arrives. +// sync_warps_release needs to be called by some warp after this one. +_CG_STATIC_QUALIFIER void sync_warps_wait_for_release( + volatile barrier_t *arrived, + bool is_master, + unsigned int thread_rank, + unsigned int num_warps) { + + unsigned int warp_id = thread_rank / 32; + unsigned int warp_bit = 1 << warp_id; + + __syncwarp(0xFFFFFFFF); + + if (is_master) { + atomicOr((unsigned int *)arrived, warp_bit); + while(*arrived & warp_bit); + } + + __syncwarp(0xFFFFFFFF); +} + +enum wait_for_warps_kind { + wait_for_all_other_warps, + wait_for_specific_warp +}; + +// Wait for a combinantion of warps specified by Kind parameter to arrive at the group barrier. +// This function does not arrive at the barrier. +template +_CG_QUALIFIER void sync_warps_wait_for_warps( + unsigned int wait_warp_id, volatile barrier_t *arrived, unsigned int thread_rank, unsigned int num_warps); + +template <> +_CG_QUALIFIER void sync_warps_wait_for_warps( + unsigned int wait_warp_id, + volatile barrier_t *arrived, + unsigned int thread_rank, + unsigned int num_warps) { + + unsigned int wait_mask = get_group_mask(thread_rank, num_warps); + wait_mask &= ~(1 << wait_warp_id); + while((*arrived & wait_mask) != wait_mask); +} + +template <> +_CG_QUALIFIER void sync_warps_wait_for_warps( + unsigned int wait_warp_id, + volatile barrier_t *arrived, + unsigned int thread_rank, + unsigned int num_warps) { + + unsigned int wait_mask = 1 << wait_warp_id; + while((*arrived & wait_mask) != wait_mask); +} + +} // details + +_CG_END_NAMESPACE + +#endif // _CG_GRID_H diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/scan.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/scan.h new file mode 100644 index 0000000000000000000000000000000000000000..9bc27078028318ada00cbcccd052e0d6cc930cfe --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/include/cooperative_groups/scan.h @@ -0,0 +1,63 @@ +/* Copyright 1993-2016 NVIDIA Corporation. All rights reserved. + * + * NOTICE TO LICENSEE: + * + * The source code and/or documentation ("Licensed Deliverables") are + * subject to NVIDIA intellectual property rights under U.S. and + * international Copyright laws. + * + * The Licensed Deliverables contained herein are PROPRIETARY and + * CONFIDENTIAL to NVIDIA and are being provided under the terms and + * conditions of a form of NVIDIA software license agreement by and + * between NVIDIA and Licensee ("License Agreement") or electronically + * accepted by Licensee. Notwithstanding any terms or conditions to + * the contrary in the License Agreement, reproduction or disclosure + * of the Licensed Deliverables to any third party without the express + * written consent of NVIDIA is prohibited. + * + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE + * SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. THEY ARE + * PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND. + * NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED + * DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY, + * NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE. + * NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE + * LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY + * SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY + * DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, + * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS + * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE + * OF THESE LICENSED DELIVERABLES. + * + * U.S. Government End Users. These Licensed Deliverables are a + * "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT + * 1995), consisting of "commercial computer software" and "commercial + * computer software documentation" as such terms are used in 48 + * C.F.R. 12.212 (SEPT 1995) and are provided to the U.S. Government + * only as a commercial end item. Consistent with 48 C.F.R.12.212 and + * 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all + * U.S. Government End Users acquire the Licensed Deliverables with + * only those rights set forth herein. + * + * Any use of the Licensed Deliverables in individual and commercial + * software must include, in the user documentation and internal + * comments to the code, the above Disclaimer and U.S. Government End + * Users Notice. + */ + +#ifndef _COOPERATIVE_GROUPS_SCAN_H +#define _COOPERATIVE_GROUPS_SCAN_H + +#include "../cooperative_groups.h" +#include "details/info.h" + +#ifdef _CG_CPP11_FEATURES +# include "details/scan.h" +#else +# error This file requires compiler support for the ISO C++ 2011 standard. This support must be enabled with the \ + -std=c++11 compiler option. +#endif + + +#endif //_COOPERATIVE_GROUPS_SCAN_H diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/lib/__init__.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/lib/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/lib/libOpenCL.so.1 b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/lib/libOpenCL.so.1 new file mode 100644 index 0000000000000000000000000000000000000000..22f3e2204a9c22de4a4e27014b29a53131487ae1 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/nvidia/cuda_runtime/lib/libOpenCL.so.1 differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/idna/compat.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/idna/compat.py new file mode 100644 index 0000000000000000000000000000000000000000..786e6bda63699b72d588ba91dd73df017570aee5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/idna/compat.py @@ -0,0 +1,13 @@ +from .core import * +from .codec import * +from typing import Any, Union + +def ToASCII(label: str) -> bytes: + return encode(label) + +def ToUnicode(label: Union[bytes, bytearray]) -> str: + return decode(label) + +def nameprep(s: Any) -> None: + raise NotImplementedError('IDNA 2008 does not utilise nameprep protocol') + diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/adapters.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/adapters.py new file mode 100644 index 0000000000000000000000000000000000000000..7030777465fb2823ebb72e89b11d5a36824330da --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/adapters.py @@ -0,0 +1,719 @@ +""" +requests.adapters +~~~~~~~~~~~~~~~~~ + +This module contains the transport adapters that Requests uses to define +and maintain connections. +""" + +import os.path +import socket # noqa: F401 +import typing +import warnings + +from pip._vendor.urllib3.exceptions import ClosedPoolError, ConnectTimeoutError +from pip._vendor.urllib3.exceptions import HTTPError as _HTTPError +from pip._vendor.urllib3.exceptions import InvalidHeader as _InvalidHeader +from pip._vendor.urllib3.exceptions import ( + LocationValueError, + MaxRetryError, + NewConnectionError, + ProtocolError, +) +from pip._vendor.urllib3.exceptions import ProxyError as _ProxyError +from pip._vendor.urllib3.exceptions import ReadTimeoutError, ResponseError +from pip._vendor.urllib3.exceptions import SSLError as _SSLError +from pip._vendor.urllib3.poolmanager import PoolManager, proxy_from_url +from pip._vendor.urllib3.util import Timeout as TimeoutSauce +from pip._vendor.urllib3.util import parse_url +from pip._vendor.urllib3.util.retry import Retry +from pip._vendor.urllib3.util.ssl_ import create_urllib3_context + +from .auth import _basic_auth_str +from .compat import basestring, urlparse +from .cookies import extract_cookies_to_jar +from .exceptions import ( + ConnectionError, + ConnectTimeout, + InvalidHeader, + InvalidProxyURL, + InvalidSchema, + InvalidURL, + ProxyError, + ReadTimeout, + RetryError, + SSLError, +) +from .models import Response +from .structures import CaseInsensitiveDict +from .utils import ( + DEFAULT_CA_BUNDLE_PATH, + extract_zipped_paths, + get_auth_from_url, + get_encoding_from_headers, + prepend_scheme_if_needed, + select_proxy, + urldefragauth, +) + +try: + from pip._vendor.urllib3.contrib.socks import SOCKSProxyManager +except ImportError: + + def SOCKSProxyManager(*args, **kwargs): + raise InvalidSchema("Missing dependencies for SOCKS support.") + + +if typing.TYPE_CHECKING: + from .models import PreparedRequest + + +DEFAULT_POOLBLOCK = False +DEFAULT_POOLSIZE = 10 +DEFAULT_RETRIES = 0 +DEFAULT_POOL_TIMEOUT = None + + +try: + import ssl # noqa: F401 + + _preloaded_ssl_context = create_urllib3_context() + _preloaded_ssl_context.load_verify_locations( + extract_zipped_paths(DEFAULT_CA_BUNDLE_PATH) + ) +except ImportError: + # Bypass default SSLContext creation when Python + # interpreter isn't built with the ssl module. + _preloaded_ssl_context = None + + +def _urllib3_request_context( + request: "PreparedRequest", + verify: "bool | str | None", + client_cert: "typing.Tuple[str, str] | str | None", + poolmanager: "PoolManager", +) -> "(typing.Dict[str, typing.Any], typing.Dict[str, typing.Any])": + host_params = {} + pool_kwargs = {} + parsed_request_url = urlparse(request.url) + scheme = parsed_request_url.scheme.lower() + port = parsed_request_url.port + + # Determine if we have and should use our default SSLContext + # to optimize performance on standard requests. + poolmanager_kwargs = getattr(poolmanager, "connection_pool_kw", {}) + has_poolmanager_ssl_context = poolmanager_kwargs.get("ssl_context") + should_use_default_ssl_context = ( + _preloaded_ssl_context is not None and not has_poolmanager_ssl_context + ) + + cert_reqs = "CERT_REQUIRED" + if verify is False: + cert_reqs = "CERT_NONE" + elif verify is True and should_use_default_ssl_context: + pool_kwargs["ssl_context"] = _preloaded_ssl_context + elif isinstance(verify, str): + if not os.path.isdir(verify): + pool_kwargs["ca_certs"] = verify + else: + pool_kwargs["ca_cert_dir"] = verify + pool_kwargs["cert_reqs"] = cert_reqs + if client_cert is not None: + if isinstance(client_cert, tuple) and len(client_cert) == 2: + pool_kwargs["cert_file"] = client_cert[0] + pool_kwargs["key_file"] = client_cert[1] + else: + # According to our docs, we allow users to specify just the client + # cert path + pool_kwargs["cert_file"] = client_cert + host_params = { + "scheme": scheme, + "host": parsed_request_url.hostname, + "port": port, + } + return host_params, pool_kwargs + + +class BaseAdapter: + """The Base Transport Adapter""" + + def __init__(self): + super().__init__() + + def send( + self, request, stream=False, timeout=None, verify=True, cert=None, proxies=None + ): + """Sends PreparedRequest object. Returns Response object. + + :param request: The :class:`PreparedRequest ` being sent. + :param stream: (optional) Whether to stream the request content. + :param timeout: (optional) How long to wait for the server to send + data before giving up, as a float, or a :ref:`(connect timeout, + read timeout) ` tuple. + :type timeout: float or tuple + :param verify: (optional) Either a boolean, in which case it controls whether we verify + the server's TLS certificate, or a string, in which case it must be a path + to a CA bundle to use + :param cert: (optional) Any user-provided SSL certificate to be trusted. + :param proxies: (optional) The proxies dictionary to apply to the request. + """ + raise NotImplementedError + + def close(self): + """Cleans up adapter specific items.""" + raise NotImplementedError + + +class HTTPAdapter(BaseAdapter): + """The built-in HTTP Adapter for urllib3. + + Provides a general-case interface for Requests sessions to contact HTTP and + HTTPS urls by implementing the Transport Adapter interface. This class will + usually be created by the :class:`Session ` class under the + covers. + + :param pool_connections: The number of urllib3 connection pools to cache. + :param pool_maxsize: The maximum number of connections to save in the pool. + :param max_retries: The maximum number of retries each connection + should attempt. Note, this applies only to failed DNS lookups, socket + connections and connection timeouts, never to requests where data has + made it to the server. By default, Requests does not retry failed + connections. If you need granular control over the conditions under + which we retry a request, import urllib3's ``Retry`` class and pass + that instead. + :param pool_block: Whether the connection pool should block for connections. + + Usage:: + + >>> import requests + >>> s = requests.Session() + >>> a = requests.adapters.HTTPAdapter(max_retries=3) + >>> s.mount('http://', a) + """ + + __attrs__ = [ + "max_retries", + "config", + "_pool_connections", + "_pool_maxsize", + "_pool_block", + ] + + def __init__( + self, + pool_connections=DEFAULT_POOLSIZE, + pool_maxsize=DEFAULT_POOLSIZE, + max_retries=DEFAULT_RETRIES, + pool_block=DEFAULT_POOLBLOCK, + ): + if max_retries == DEFAULT_RETRIES: + self.max_retries = Retry(0, read=False) + else: + self.max_retries = Retry.from_int(max_retries) + self.config = {} + self.proxy_manager = {} + + super().__init__() + + self._pool_connections = pool_connections + self._pool_maxsize = pool_maxsize + self._pool_block = pool_block + + self.init_poolmanager(pool_connections, pool_maxsize, block=pool_block) + + def __getstate__(self): + return {attr: getattr(self, attr, None) for attr in self.__attrs__} + + def __setstate__(self, state): + # Can't handle by adding 'proxy_manager' to self.__attrs__ because + # self.poolmanager uses a lambda function, which isn't pickleable. + self.proxy_manager = {} + self.config = {} + + for attr, value in state.items(): + setattr(self, attr, value) + + self.init_poolmanager( + self._pool_connections, self._pool_maxsize, block=self._pool_block + ) + + def init_poolmanager( + self, connections, maxsize, block=DEFAULT_POOLBLOCK, **pool_kwargs + ): + """Initializes a urllib3 PoolManager. + + This method should not be called from user code, and is only + exposed for use when subclassing the + :class:`HTTPAdapter `. + + :param connections: The number of urllib3 connection pools to cache. + :param maxsize: The maximum number of connections to save in the pool. + :param block: Block when no free connections are available. + :param pool_kwargs: Extra keyword arguments used to initialize the Pool Manager. + """ + # save these values for pickling + self._pool_connections = connections + self._pool_maxsize = maxsize + self._pool_block = block + + self.poolmanager = PoolManager( + num_pools=connections, + maxsize=maxsize, + block=block, + **pool_kwargs, + ) + + def proxy_manager_for(self, proxy, **proxy_kwargs): + """Return urllib3 ProxyManager for the given proxy. + + This method should not be called from user code, and is only + exposed for use when subclassing the + :class:`HTTPAdapter `. + + :param proxy: The proxy to return a urllib3 ProxyManager for. + :param proxy_kwargs: Extra keyword arguments used to configure the Proxy Manager. + :returns: ProxyManager + :rtype: urllib3.ProxyManager + """ + if proxy in self.proxy_manager: + manager = self.proxy_manager[proxy] + elif proxy.lower().startswith("socks"): + username, password = get_auth_from_url(proxy) + manager = self.proxy_manager[proxy] = SOCKSProxyManager( + proxy, + username=username, + password=password, + num_pools=self._pool_connections, + maxsize=self._pool_maxsize, + block=self._pool_block, + **proxy_kwargs, + ) + else: + proxy_headers = self.proxy_headers(proxy) + manager = self.proxy_manager[proxy] = proxy_from_url( + proxy, + proxy_headers=proxy_headers, + num_pools=self._pool_connections, + maxsize=self._pool_maxsize, + block=self._pool_block, + **proxy_kwargs, + ) + + return manager + + def cert_verify(self, conn, url, verify, cert): + """Verify a SSL certificate. This method should not be called from user + code, and is only exposed for use when subclassing the + :class:`HTTPAdapter `. + + :param conn: The urllib3 connection object associated with the cert. + :param url: The requested URL. + :param verify: Either a boolean, in which case it controls whether we verify + the server's TLS certificate, or a string, in which case it must be a path + to a CA bundle to use + :param cert: The SSL certificate to verify. + """ + if url.lower().startswith("https") and verify: + conn.cert_reqs = "CERT_REQUIRED" + + # Only load the CA certificates if 'verify' is a string indicating the CA bundle to use. + # Otherwise, if verify is a boolean, we don't load anything since + # the connection will be using a context with the default certificates already loaded, + # and this avoids a call to the slow load_verify_locations() + if verify is not True: + # `verify` must be a str with a path then + cert_loc = verify + + if not os.path.exists(cert_loc): + raise OSError( + f"Could not find a suitable TLS CA certificate bundle, " + f"invalid path: {cert_loc}" + ) + + if not os.path.isdir(cert_loc): + conn.ca_certs = cert_loc + else: + conn.ca_cert_dir = cert_loc + else: + conn.cert_reqs = "CERT_NONE" + conn.ca_certs = None + conn.ca_cert_dir = None + + if cert: + if not isinstance(cert, basestring): + conn.cert_file = cert[0] + conn.key_file = cert[1] + else: + conn.cert_file = cert + conn.key_file = None + if conn.cert_file and not os.path.exists(conn.cert_file): + raise OSError( + f"Could not find the TLS certificate file, " + f"invalid path: {conn.cert_file}" + ) + if conn.key_file and not os.path.exists(conn.key_file): + raise OSError( + f"Could not find the TLS key file, invalid path: {conn.key_file}" + ) + + def build_response(self, req, resp): + """Builds a :class:`Response ` object from a urllib3 + response. This should not be called from user code, and is only exposed + for use when subclassing the + :class:`HTTPAdapter ` + + :param req: The :class:`PreparedRequest ` used to generate the response. + :param resp: The urllib3 response object. + :rtype: requests.Response + """ + response = Response() + + # Fallback to None if there's no status_code, for whatever reason. + response.status_code = getattr(resp, "status", None) + + # Make headers case-insensitive. + response.headers = CaseInsensitiveDict(getattr(resp, "headers", {})) + + # Set encoding. + response.encoding = get_encoding_from_headers(response.headers) + response.raw = resp + response.reason = response.raw.reason + + if isinstance(req.url, bytes): + response.url = req.url.decode("utf-8") + else: + response.url = req.url + + # Add new cookies from the server. + extract_cookies_to_jar(response.cookies, req, resp) + + # Give the Response some context. + response.request = req + response.connection = self + + return response + + def build_connection_pool_key_attributes(self, request, verify, cert=None): + """Build the PoolKey attributes used by urllib3 to return a connection. + + This looks at the PreparedRequest, the user-specified verify value, + and the value of the cert parameter to determine what PoolKey values + to use to select a connection from a given urllib3 Connection Pool. + + The SSL related pool key arguments are not consistently set. As of + this writing, use the following to determine what keys may be in that + dictionary: + + * If ``verify`` is ``True``, ``"ssl_context"`` will be set and will be the + default Requests SSL Context + * If ``verify`` is ``False``, ``"ssl_context"`` will not be set but + ``"cert_reqs"`` will be set + * If ``verify`` is a string, (i.e., it is a user-specified trust bundle) + ``"ca_certs"`` will be set if the string is not a directory recognized + by :py:func:`os.path.isdir`, otherwise ``"ca_certs_dir"`` will be + set. + * If ``"cert"`` is specified, ``"cert_file"`` will always be set. If + ``"cert"`` is a tuple with a second item, ``"key_file"`` will also + be present + + To override these settings, one may subclass this class, call this + method and use the above logic to change parameters as desired. For + example, if one wishes to use a custom :py:class:`ssl.SSLContext` one + must both set ``"ssl_context"`` and based on what else they require, + alter the other keys to ensure the desired behaviour. + + :param request: + The PreparedReqest being sent over the connection. + :type request: + :class:`~requests.models.PreparedRequest` + :param verify: + Either a boolean, in which case it controls whether + we verify the server's TLS certificate, or a string, in which case it + must be a path to a CA bundle to use. + :param cert: + (optional) Any user-provided SSL certificate for client + authentication (a.k.a., mTLS). This may be a string (i.e., just + the path to a file which holds both certificate and key) or a + tuple of length 2 with the certificate file path and key file + path. + :returns: + A tuple of two dictionaries. The first is the "host parameters" + portion of the Pool Key including scheme, hostname, and port. The + second is a dictionary of SSLContext related parameters. + """ + return _urllib3_request_context(request, verify, cert, self.poolmanager) + + def get_connection_with_tls_context(self, request, verify, proxies=None, cert=None): + """Returns a urllib3 connection for the given request and TLS settings. + This should not be called from user code, and is only exposed for use + when subclassing the :class:`HTTPAdapter `. + + :param request: + The :class:`PreparedRequest ` object to be sent + over the connection. + :param verify: + Either a boolean, in which case it controls whether we verify the + server's TLS certificate, or a string, in which case it must be a + path to a CA bundle to use. + :param proxies: + (optional) The proxies dictionary to apply to the request. + :param cert: + (optional) Any user-provided SSL certificate to be used for client + authentication (a.k.a., mTLS). + :rtype: + urllib3.ConnectionPool + """ + proxy = select_proxy(request.url, proxies) + try: + host_params, pool_kwargs = self.build_connection_pool_key_attributes( + request, + verify, + cert, + ) + except ValueError as e: + raise InvalidURL(e, request=request) + if proxy: + proxy = prepend_scheme_if_needed(proxy, "http") + proxy_url = parse_url(proxy) + if not proxy_url.host: + raise InvalidProxyURL( + "Please check proxy URL. It is malformed " + "and could be missing the host." + ) + proxy_manager = self.proxy_manager_for(proxy) + conn = proxy_manager.connection_from_host( + **host_params, pool_kwargs=pool_kwargs + ) + else: + # Only scheme should be lower case + conn = self.poolmanager.connection_from_host( + **host_params, pool_kwargs=pool_kwargs + ) + + return conn + + def get_connection(self, url, proxies=None): + """DEPRECATED: Users should move to `get_connection_with_tls_context` + for all subclasses of HTTPAdapter using Requests>=2.32.2. + + Returns a urllib3 connection for the given URL. This should not be + called from user code, and is only exposed for use when subclassing the + :class:`HTTPAdapter `. + + :param url: The URL to connect to. + :param proxies: (optional) A Requests-style dictionary of proxies used on this request. + :rtype: urllib3.ConnectionPool + """ + warnings.warn( + ( + "`get_connection` has been deprecated in favor of " + "`get_connection_with_tls_context`. Custom HTTPAdapter subclasses " + "will need to migrate for Requests>=2.32.2. Please see " + "https://github.com/psf/requests/pull/6710 for more details." + ), + DeprecationWarning, + ) + proxy = select_proxy(url, proxies) + + if proxy: + proxy = prepend_scheme_if_needed(proxy, "http") + proxy_url = parse_url(proxy) + if not proxy_url.host: + raise InvalidProxyURL( + "Please check proxy URL. It is malformed " + "and could be missing the host." + ) + proxy_manager = self.proxy_manager_for(proxy) + conn = proxy_manager.connection_from_url(url) + else: + # Only scheme should be lower case + parsed = urlparse(url) + url = parsed.geturl() + conn = self.poolmanager.connection_from_url(url) + + return conn + + def close(self): + """Disposes of any internal state. + + Currently, this closes the PoolManager and any active ProxyManager, + which closes any pooled connections. + """ + self.poolmanager.clear() + for proxy in self.proxy_manager.values(): + proxy.clear() + + def request_url(self, request, proxies): + """Obtain the url to use when making the final request. + + If the message is being sent through a HTTP proxy, the full URL has to + be used. Otherwise, we should only use the path portion of the URL. + + This should not be called from user code, and is only exposed for use + when subclassing the + :class:`HTTPAdapter `. + + :param request: The :class:`PreparedRequest ` being sent. + :param proxies: A dictionary of schemes or schemes and hosts to proxy URLs. + :rtype: str + """ + proxy = select_proxy(request.url, proxies) + scheme = urlparse(request.url).scheme + + is_proxied_http_request = proxy and scheme != "https" + using_socks_proxy = False + if proxy: + proxy_scheme = urlparse(proxy).scheme.lower() + using_socks_proxy = proxy_scheme.startswith("socks") + + url = request.path_url + if url.startswith("//"): # Don't confuse urllib3 + url = f"/{url.lstrip('/')}" + + if is_proxied_http_request and not using_socks_proxy: + url = urldefragauth(request.url) + + return url + + def add_headers(self, request, **kwargs): + """Add any headers needed by the connection. As of v2.0 this does + nothing by default, but is left for overriding by users that subclass + the :class:`HTTPAdapter `. + + This should not be called from user code, and is only exposed for use + when subclassing the + :class:`HTTPAdapter `. + + :param request: The :class:`PreparedRequest ` to add headers to. + :param kwargs: The keyword arguments from the call to send(). + """ + pass + + def proxy_headers(self, proxy): + """Returns a dictionary of the headers to add to any request sent + through a proxy. This works with urllib3 magic to ensure that they are + correctly sent to the proxy, rather than in a tunnelled request if + CONNECT is being used. + + This should not be called from user code, and is only exposed for use + when subclassing the + :class:`HTTPAdapter `. + + :param proxy: The url of the proxy being used for this request. + :rtype: dict + """ + headers = {} + username, password = get_auth_from_url(proxy) + + if username: + headers["Proxy-Authorization"] = _basic_auth_str(username, password) + + return headers + + def send( + self, request, stream=False, timeout=None, verify=True, cert=None, proxies=None + ): + """Sends PreparedRequest object. Returns Response object. + + :param request: The :class:`PreparedRequest ` being sent. + :param stream: (optional) Whether to stream the request content. + :param timeout: (optional) How long to wait for the server to send + data before giving up, as a float, or a :ref:`(connect timeout, + read timeout) ` tuple. + :type timeout: float or tuple or urllib3 Timeout object + :param verify: (optional) Either a boolean, in which case it controls whether + we verify the server's TLS certificate, or a string, in which case it + must be a path to a CA bundle to use + :param cert: (optional) Any user-provided SSL certificate to be trusted. + :param proxies: (optional) The proxies dictionary to apply to the request. + :rtype: requests.Response + """ + + try: + conn = self.get_connection_with_tls_context( + request, verify, proxies=proxies, cert=cert + ) + except LocationValueError as e: + raise InvalidURL(e, request=request) + + self.cert_verify(conn, request.url, verify, cert) + url = self.request_url(request, proxies) + self.add_headers( + request, + stream=stream, + timeout=timeout, + verify=verify, + cert=cert, + proxies=proxies, + ) + + chunked = not (request.body is None or "Content-Length" in request.headers) + + if isinstance(timeout, tuple): + try: + connect, read = timeout + timeout = TimeoutSauce(connect=connect, read=read) + except ValueError: + raise ValueError( + f"Invalid timeout {timeout}. Pass a (connect, read) timeout tuple, " + f"or a single float to set both timeouts to the same value." + ) + elif isinstance(timeout, TimeoutSauce): + pass + else: + timeout = TimeoutSauce(connect=timeout, read=timeout) + + try: + resp = conn.urlopen( + method=request.method, + url=url, + body=request.body, + headers=request.headers, + redirect=False, + assert_same_host=False, + preload_content=False, + decode_content=False, + retries=self.max_retries, + timeout=timeout, + chunked=chunked, + ) + + except (ProtocolError, OSError) as err: + raise ConnectionError(err, request=request) + + except MaxRetryError as e: + if isinstance(e.reason, ConnectTimeoutError): + # TODO: Remove this in 3.0.0: see #2811 + if not isinstance(e.reason, NewConnectionError): + raise ConnectTimeout(e, request=request) + + if isinstance(e.reason, ResponseError): + raise RetryError(e, request=request) + + if isinstance(e.reason, _ProxyError): + raise ProxyError(e, request=request) + + if isinstance(e.reason, _SSLError): + # This branch is for urllib3 v1.22 and later. + raise SSLError(e, request=request) + + raise ConnectionError(e, request=request) + + except ClosedPoolError as e: + raise ConnectionError(e, request=request) + + except _ProxyError as e: + raise ProxyError(e) + + except (_SSLError, _HTTPError) as e: + if isinstance(e, _SSLError): + # This branch is for urllib3 versions earlier than v1.22 + raise SSLError(e, request=request) + elif isinstance(e, ReadTimeoutError): + raise ReadTimeout(e, request=request) + elif isinstance(e, _InvalidHeader): + raise InvalidHeader(e, request=request) + else: + raise + + return self.build_response(request, resp) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/auth.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/auth.py new file mode 100644 index 0000000000000000000000000000000000000000..4a7ce6dc1460e0de8aa0c38ea9123faa69bd5110 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/auth.py @@ -0,0 +1,314 @@ +""" +requests.auth +~~~~~~~~~~~~~ + +This module contains the authentication handlers for Requests. +""" + +import hashlib +import os +import re +import threading +import time +import warnings +from base64 import b64encode + +from ._internal_utils import to_native_string +from .compat import basestring, str, urlparse +from .cookies import extract_cookies_to_jar +from .utils import parse_dict_header + +CONTENT_TYPE_FORM_URLENCODED = "application/x-www-form-urlencoded" +CONTENT_TYPE_MULTI_PART = "multipart/form-data" + + +def _basic_auth_str(username, password): + """Returns a Basic Auth string.""" + + # "I want us to put a big-ol' comment on top of it that + # says that this behaviour is dumb but we need to preserve + # it because people are relying on it." + # - Lukasa + # + # These are here solely to maintain backwards compatibility + # for things like ints. This will be removed in 3.0.0. + if not isinstance(username, basestring): + warnings.warn( + "Non-string usernames will no longer be supported in Requests " + "3.0.0. Please convert the object you've passed in ({!r}) to " + "a string or bytes object in the near future to avoid " + "problems.".format(username), + category=DeprecationWarning, + ) + username = str(username) + + if not isinstance(password, basestring): + warnings.warn( + "Non-string passwords will no longer be supported in Requests " + "3.0.0. Please convert the object you've passed in ({!r}) to " + "a string or bytes object in the near future to avoid " + "problems.".format(type(password)), + category=DeprecationWarning, + ) + password = str(password) + # -- End Removal -- + + if isinstance(username, str): + username = username.encode("latin1") + + if isinstance(password, str): + password = password.encode("latin1") + + authstr = "Basic " + to_native_string( + b64encode(b":".join((username, password))).strip() + ) + + return authstr + + +class AuthBase: + """Base class that all auth implementations derive from""" + + def __call__(self, r): + raise NotImplementedError("Auth hooks must be callable.") + + +class HTTPBasicAuth(AuthBase): + """Attaches HTTP Basic Authentication to the given Request object.""" + + def __init__(self, username, password): + self.username = username + self.password = password + + def __eq__(self, other): + return all( + [ + self.username == getattr(other, "username", None), + self.password == getattr(other, "password", None), + ] + ) + + def __ne__(self, other): + return not self == other + + def __call__(self, r): + r.headers["Authorization"] = _basic_auth_str(self.username, self.password) + return r + + +class HTTPProxyAuth(HTTPBasicAuth): + """Attaches HTTP Proxy Authentication to a given Request object.""" + + def __call__(self, r): + r.headers["Proxy-Authorization"] = _basic_auth_str(self.username, self.password) + return r + + +class HTTPDigestAuth(AuthBase): + """Attaches HTTP Digest Authentication to the given Request object.""" + + def __init__(self, username, password): + self.username = username + self.password = password + # Keep state in per-thread local storage + self._thread_local = threading.local() + + def init_per_thread_state(self): + # Ensure state is initialized just once per-thread + if not hasattr(self._thread_local, "init"): + self._thread_local.init = True + self._thread_local.last_nonce = "" + self._thread_local.nonce_count = 0 + self._thread_local.chal = {} + self._thread_local.pos = None + self._thread_local.num_401_calls = None + + def build_digest_header(self, method, url): + """ + :rtype: str + """ + + realm = self._thread_local.chal["realm"] + nonce = self._thread_local.chal["nonce"] + qop = self._thread_local.chal.get("qop") + algorithm = self._thread_local.chal.get("algorithm") + opaque = self._thread_local.chal.get("opaque") + hash_utf8 = None + + if algorithm is None: + _algorithm = "MD5" + else: + _algorithm = algorithm.upper() + # lambdas assume digest modules are imported at the top level + if _algorithm == "MD5" or _algorithm == "MD5-SESS": + + def md5_utf8(x): + if isinstance(x, str): + x = x.encode("utf-8") + return hashlib.md5(x).hexdigest() + + hash_utf8 = md5_utf8 + elif _algorithm == "SHA": + + def sha_utf8(x): + if isinstance(x, str): + x = x.encode("utf-8") + return hashlib.sha1(x).hexdigest() + + hash_utf8 = sha_utf8 + elif _algorithm == "SHA-256": + + def sha256_utf8(x): + if isinstance(x, str): + x = x.encode("utf-8") + return hashlib.sha256(x).hexdigest() + + hash_utf8 = sha256_utf8 + elif _algorithm == "SHA-512": + + def sha512_utf8(x): + if isinstance(x, str): + x = x.encode("utf-8") + return hashlib.sha512(x).hexdigest() + + hash_utf8 = sha512_utf8 + + KD = lambda s, d: hash_utf8(f"{s}:{d}") # noqa:E731 + + if hash_utf8 is None: + return None + + # XXX not implemented yet + entdig = None + p_parsed = urlparse(url) + #: path is request-uri defined in RFC 2616 which should not be empty + path = p_parsed.path or "/" + if p_parsed.query: + path += f"?{p_parsed.query}" + + A1 = f"{self.username}:{realm}:{self.password}" + A2 = f"{method}:{path}" + + HA1 = hash_utf8(A1) + HA2 = hash_utf8(A2) + + if nonce == self._thread_local.last_nonce: + self._thread_local.nonce_count += 1 + else: + self._thread_local.nonce_count = 1 + ncvalue = f"{self._thread_local.nonce_count:08x}" + s = str(self._thread_local.nonce_count).encode("utf-8") + s += nonce.encode("utf-8") + s += time.ctime().encode("utf-8") + s += os.urandom(8) + + cnonce = hashlib.sha1(s).hexdigest()[:16] + if _algorithm == "MD5-SESS": + HA1 = hash_utf8(f"{HA1}:{nonce}:{cnonce}") + + if not qop: + respdig = KD(HA1, f"{nonce}:{HA2}") + elif qop == "auth" or "auth" in qop.split(","): + noncebit = f"{nonce}:{ncvalue}:{cnonce}:auth:{HA2}" + respdig = KD(HA1, noncebit) + else: + # XXX handle auth-int. + return None + + self._thread_local.last_nonce = nonce + + # XXX should the partial digests be encoded too? + base = ( + f'username="{self.username}", realm="{realm}", nonce="{nonce}", ' + f'uri="{path}", response="{respdig}"' + ) + if opaque: + base += f', opaque="{opaque}"' + if algorithm: + base += f', algorithm="{algorithm}"' + if entdig: + base += f', digest="{entdig}"' + if qop: + base += f', qop="auth", nc={ncvalue}, cnonce="{cnonce}"' + + return f"Digest {base}" + + def handle_redirect(self, r, **kwargs): + """Reset num_401_calls counter on redirects.""" + if r.is_redirect: + self._thread_local.num_401_calls = 1 + + def handle_401(self, r, **kwargs): + """ + Takes the given response and tries digest-auth, if needed. + + :rtype: requests.Response + """ + + # If response is not 4xx, do not auth + # See https://github.com/psf/requests/issues/3772 + if not 400 <= r.status_code < 500: + self._thread_local.num_401_calls = 1 + return r + + if self._thread_local.pos is not None: + # Rewind the file position indicator of the body to where + # it was to resend the request. + r.request.body.seek(self._thread_local.pos) + s_auth = r.headers.get("www-authenticate", "") + + if "digest" in s_auth.lower() and self._thread_local.num_401_calls < 2: + self._thread_local.num_401_calls += 1 + pat = re.compile(r"digest ", flags=re.IGNORECASE) + self._thread_local.chal = parse_dict_header(pat.sub("", s_auth, count=1)) + + # Consume content and release the original connection + # to allow our new request to reuse the same one. + r.content + r.close() + prep = r.request.copy() + extract_cookies_to_jar(prep._cookies, r.request, r.raw) + prep.prepare_cookies(prep._cookies) + + prep.headers["Authorization"] = self.build_digest_header( + prep.method, prep.url + ) + _r = r.connection.send(prep, **kwargs) + _r.history.append(r) + _r.request = prep + + return _r + + self._thread_local.num_401_calls = 1 + return r + + def __call__(self, r): + # Initialize per-thread state, if needed + self.init_per_thread_state() + # If we have a saved nonce, skip the 401 + if self._thread_local.last_nonce: + r.headers["Authorization"] = self.build_digest_header(r.method, r.url) + try: + self._thread_local.pos = r.body.tell() + except AttributeError: + # In the case of HTTPDigestAuth being reused and the body of + # the previous request was a file-like object, pos has the + # file position of the previous body. Ensure it's set to + # None. + self._thread_local.pos = None + r.register_hook("response", self.handle_401) + r.register_hook("response", self.handle_redirect) + self._thread_local.num_401_calls = 1 + + return r + + def __eq__(self, other): + return all( + [ + self.username == getattr(other, "username", None), + self.password == getattr(other, "password", None), + ] + ) + + def __ne__(self, other): + return not self == other diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/certs.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/certs.py new file mode 100644 index 0000000000000000000000000000000000000000..38696a1fb3419dd810004d5aec9654e5224042ed --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/certs.py @@ -0,0 +1,24 @@ +#!/usr/bin/env python + +""" +requests.certs +~~~~~~~~~~~~~~ + +This module returns the preferred default CA certificate bundle. There is +only one — the one from the certifi package. + +If you are packaging Requests, e.g., for a Linux distribution or a managed +environment, you can change the definition of where() to return a separately +packaged CA bundle. +""" + +import os + +if "_PIP_STANDALONE_CERT" not in os.environ: + from pip._vendor.certifi import where +else: + def where(): + return os.environ["_PIP_STANDALONE_CERT"] + +if __name__ == "__main__": + print(where()) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/cookies.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/cookies.py new file mode 100644 index 0000000000000000000000000000000000000000..f69d0cda9e1c893401015a09f2db2de5a5960fd2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/cookies.py @@ -0,0 +1,561 @@ +""" +requests.cookies +~~~~~~~~~~~~~~~~ + +Compatibility code to be able to use `http.cookiejar.CookieJar` with requests. + +requests.utils imports from here, so be careful with imports. +""" + +import calendar +import copy +import time + +from ._internal_utils import to_native_string +from .compat import Morsel, MutableMapping, cookielib, urlparse, urlunparse + +try: + import threading +except ImportError: + import dummy_threading as threading + + +class MockRequest: + """Wraps a `requests.Request` to mimic a `urllib2.Request`. + + The code in `http.cookiejar.CookieJar` expects this interface in order to correctly + manage cookie policies, i.e., determine whether a cookie can be set, given the + domains of the request and the cookie. + + The original request object is read-only. The client is responsible for collecting + the new headers via `get_new_headers()` and interpreting them appropriately. You + probably want `get_cookie_header`, defined below. + """ + + def __init__(self, request): + self._r = request + self._new_headers = {} + self.type = urlparse(self._r.url).scheme + + def get_type(self): + return self.type + + def get_host(self): + return urlparse(self._r.url).netloc + + def get_origin_req_host(self): + return self.get_host() + + def get_full_url(self): + # Only return the response's URL if the user hadn't set the Host + # header + if not self._r.headers.get("Host"): + return self._r.url + # If they did set it, retrieve it and reconstruct the expected domain + host = to_native_string(self._r.headers["Host"], encoding="utf-8") + parsed = urlparse(self._r.url) + # Reconstruct the URL as we expect it + return urlunparse( + [ + parsed.scheme, + host, + parsed.path, + parsed.params, + parsed.query, + parsed.fragment, + ] + ) + + def is_unverifiable(self): + return True + + def has_header(self, name): + return name in self._r.headers or name in self._new_headers + + def get_header(self, name, default=None): + return self._r.headers.get(name, self._new_headers.get(name, default)) + + def add_header(self, key, val): + """cookiejar has no legitimate use for this method; add it back if you find one.""" + raise NotImplementedError( + "Cookie headers should be added with add_unredirected_header()" + ) + + def add_unredirected_header(self, name, value): + self._new_headers[name] = value + + def get_new_headers(self): + return self._new_headers + + @property + def unverifiable(self): + return self.is_unverifiable() + + @property + def origin_req_host(self): + return self.get_origin_req_host() + + @property + def host(self): + return self.get_host() + + +class MockResponse: + """Wraps a `httplib.HTTPMessage` to mimic a `urllib.addinfourl`. + + ...what? Basically, expose the parsed HTTP headers from the server response + the way `http.cookiejar` expects to see them. + """ + + def __init__(self, headers): + """Make a MockResponse for `cookiejar` to read. + + :param headers: a httplib.HTTPMessage or analogous carrying the headers + """ + self._headers = headers + + def info(self): + return self._headers + + def getheaders(self, name): + self._headers.getheaders(name) + + +def extract_cookies_to_jar(jar, request, response): + """Extract the cookies from the response into a CookieJar. + + :param jar: http.cookiejar.CookieJar (not necessarily a RequestsCookieJar) + :param request: our own requests.Request object + :param response: urllib3.HTTPResponse object + """ + if not (hasattr(response, "_original_response") and response._original_response): + return + # the _original_response field is the wrapped httplib.HTTPResponse object, + req = MockRequest(request) + # pull out the HTTPMessage with the headers and put it in the mock: + res = MockResponse(response._original_response.msg) + jar.extract_cookies(res, req) + + +def get_cookie_header(jar, request): + """ + Produce an appropriate Cookie header string to be sent with `request`, or None. + + :rtype: str + """ + r = MockRequest(request) + jar.add_cookie_header(r) + return r.get_new_headers().get("Cookie") + + +def remove_cookie_by_name(cookiejar, name, domain=None, path=None): + """Unsets a cookie by name, by default over all domains and paths. + + Wraps CookieJar.clear(), is O(n). + """ + clearables = [] + for cookie in cookiejar: + if cookie.name != name: + continue + if domain is not None and domain != cookie.domain: + continue + if path is not None and path != cookie.path: + continue + clearables.append((cookie.domain, cookie.path, cookie.name)) + + for domain, path, name in clearables: + cookiejar.clear(domain, path, name) + + +class CookieConflictError(RuntimeError): + """There are two cookies that meet the criteria specified in the cookie jar. + Use .get and .set and include domain and path args in order to be more specific. + """ + + +class RequestsCookieJar(cookielib.CookieJar, MutableMapping): + """Compatibility class; is a http.cookiejar.CookieJar, but exposes a dict + interface. + + This is the CookieJar we create by default for requests and sessions that + don't specify one, since some clients may expect response.cookies and + session.cookies to support dict operations. + + Requests does not use the dict interface internally; it's just for + compatibility with external client code. All requests code should work + out of the box with externally provided instances of ``CookieJar``, e.g. + ``LWPCookieJar`` and ``FileCookieJar``. + + Unlike a regular CookieJar, this class is pickleable. + + .. warning:: dictionary operations that are normally O(1) may be O(n). + """ + + def get(self, name, default=None, domain=None, path=None): + """Dict-like get() that also supports optional domain and path args in + order to resolve naming collisions from using one cookie jar over + multiple domains. + + .. warning:: operation is O(n), not O(1). + """ + try: + return self._find_no_duplicates(name, domain, path) + except KeyError: + return default + + def set(self, name, value, **kwargs): + """Dict-like set() that also supports optional domain and path args in + order to resolve naming collisions from using one cookie jar over + multiple domains. + """ + # support client code that unsets cookies by assignment of a None value: + if value is None: + remove_cookie_by_name( + self, name, domain=kwargs.get("domain"), path=kwargs.get("path") + ) + return + + if isinstance(value, Morsel): + c = morsel_to_cookie(value) + else: + c = create_cookie(name, value, **kwargs) + self.set_cookie(c) + return c + + def iterkeys(self): + """Dict-like iterkeys() that returns an iterator of names of cookies + from the jar. + + .. seealso:: itervalues() and iteritems(). + """ + for cookie in iter(self): + yield cookie.name + + def keys(self): + """Dict-like keys() that returns a list of names of cookies from the + jar. + + .. seealso:: values() and items(). + """ + return list(self.iterkeys()) + + def itervalues(self): + """Dict-like itervalues() that returns an iterator of values of cookies + from the jar. + + .. seealso:: iterkeys() and iteritems(). + """ + for cookie in iter(self): + yield cookie.value + + def values(self): + """Dict-like values() that returns a list of values of cookies from the + jar. + + .. seealso:: keys() and items(). + """ + return list(self.itervalues()) + + def iteritems(self): + """Dict-like iteritems() that returns an iterator of name-value tuples + from the jar. + + .. seealso:: iterkeys() and itervalues(). + """ + for cookie in iter(self): + yield cookie.name, cookie.value + + def items(self): + """Dict-like items() that returns a list of name-value tuples from the + jar. Allows client-code to call ``dict(RequestsCookieJar)`` and get a + vanilla python dict of key value pairs. + + .. seealso:: keys() and values(). + """ + return list(self.iteritems()) + + def list_domains(self): + """Utility method to list all the domains in the jar.""" + domains = [] + for cookie in iter(self): + if cookie.domain not in domains: + domains.append(cookie.domain) + return domains + + def list_paths(self): + """Utility method to list all the paths in the jar.""" + paths = [] + for cookie in iter(self): + if cookie.path not in paths: + paths.append(cookie.path) + return paths + + def multiple_domains(self): + """Returns True if there are multiple domains in the jar. + Returns False otherwise. + + :rtype: bool + """ + domains = [] + for cookie in iter(self): + if cookie.domain is not None and cookie.domain in domains: + return True + domains.append(cookie.domain) + return False # there is only one domain in jar + + def get_dict(self, domain=None, path=None): + """Takes as an argument an optional domain and path and returns a plain + old Python dict of name-value pairs of cookies that meet the + requirements. + + :rtype: dict + """ + dictionary = {} + for cookie in iter(self): + if (domain is None or cookie.domain == domain) and ( + path is None or cookie.path == path + ): + dictionary[cookie.name] = cookie.value + return dictionary + + def __contains__(self, name): + try: + return super().__contains__(name) + except CookieConflictError: + return True + + def __getitem__(self, name): + """Dict-like __getitem__() for compatibility with client code. Throws + exception if there are more than one cookie with name. In that case, + use the more explicit get() method instead. + + .. warning:: operation is O(n), not O(1). + """ + return self._find_no_duplicates(name) + + def __setitem__(self, name, value): + """Dict-like __setitem__ for compatibility with client code. Throws + exception if there is already a cookie of that name in the jar. In that + case, use the more explicit set() method instead. + """ + self.set(name, value) + + def __delitem__(self, name): + """Deletes a cookie given a name. Wraps ``http.cookiejar.CookieJar``'s + ``remove_cookie_by_name()``. + """ + remove_cookie_by_name(self, name) + + def set_cookie(self, cookie, *args, **kwargs): + if ( + hasattr(cookie.value, "startswith") + and cookie.value.startswith('"') + and cookie.value.endswith('"') + ): + cookie.value = cookie.value.replace('\\"', "") + return super().set_cookie(cookie, *args, **kwargs) + + def update(self, other): + """Updates this jar with cookies from another CookieJar or dict-like""" + if isinstance(other, cookielib.CookieJar): + for cookie in other: + self.set_cookie(copy.copy(cookie)) + else: + super().update(other) + + def _find(self, name, domain=None, path=None): + """Requests uses this method internally to get cookie values. + + If there are conflicting cookies, _find arbitrarily chooses one. + See _find_no_duplicates if you want an exception thrown if there are + conflicting cookies. + + :param name: a string containing name of cookie + :param domain: (optional) string containing domain of cookie + :param path: (optional) string containing path of cookie + :return: cookie.value + """ + for cookie in iter(self): + if cookie.name == name: + if domain is None or cookie.domain == domain: + if path is None or cookie.path == path: + return cookie.value + + raise KeyError(f"name={name!r}, domain={domain!r}, path={path!r}") + + def _find_no_duplicates(self, name, domain=None, path=None): + """Both ``__get_item__`` and ``get`` call this function: it's never + used elsewhere in Requests. + + :param name: a string containing name of cookie + :param domain: (optional) string containing domain of cookie + :param path: (optional) string containing path of cookie + :raises KeyError: if cookie is not found + :raises CookieConflictError: if there are multiple cookies + that match name and optionally domain and path + :return: cookie.value + """ + toReturn = None + for cookie in iter(self): + if cookie.name == name: + if domain is None or cookie.domain == domain: + if path is None or cookie.path == path: + if toReturn is not None: + # if there are multiple cookies that meet passed in criteria + raise CookieConflictError( + f"There are multiple cookies with name, {name!r}" + ) + # we will eventually return this as long as no cookie conflict + toReturn = cookie.value + + if toReturn: + return toReturn + raise KeyError(f"name={name!r}, domain={domain!r}, path={path!r}") + + def __getstate__(self): + """Unlike a normal CookieJar, this class is pickleable.""" + state = self.__dict__.copy() + # remove the unpickleable RLock object + state.pop("_cookies_lock") + return state + + def __setstate__(self, state): + """Unlike a normal CookieJar, this class is pickleable.""" + self.__dict__.update(state) + if "_cookies_lock" not in self.__dict__: + self._cookies_lock = threading.RLock() + + def copy(self): + """Return a copy of this RequestsCookieJar.""" + new_cj = RequestsCookieJar() + new_cj.set_policy(self.get_policy()) + new_cj.update(self) + return new_cj + + def get_policy(self): + """Return the CookiePolicy instance used.""" + return self._policy + + +def _copy_cookie_jar(jar): + if jar is None: + return None + + if hasattr(jar, "copy"): + # We're dealing with an instance of RequestsCookieJar + return jar.copy() + # We're dealing with a generic CookieJar instance + new_jar = copy.copy(jar) + new_jar.clear() + for cookie in jar: + new_jar.set_cookie(copy.copy(cookie)) + return new_jar + + +def create_cookie(name, value, **kwargs): + """Make a cookie from underspecified parameters. + + By default, the pair of `name` and `value` will be set for the domain '' + and sent on every request (this is sometimes called a "supercookie"). + """ + result = { + "version": 0, + "name": name, + "value": value, + "port": None, + "domain": "", + "path": "/", + "secure": False, + "expires": None, + "discard": True, + "comment": None, + "comment_url": None, + "rest": {"HttpOnly": None}, + "rfc2109": False, + } + + badargs = set(kwargs) - set(result) + if badargs: + raise TypeError( + f"create_cookie() got unexpected keyword arguments: {list(badargs)}" + ) + + result.update(kwargs) + result["port_specified"] = bool(result["port"]) + result["domain_specified"] = bool(result["domain"]) + result["domain_initial_dot"] = result["domain"].startswith(".") + result["path_specified"] = bool(result["path"]) + + return cookielib.Cookie(**result) + + +def morsel_to_cookie(morsel): + """Convert a Morsel object into a Cookie containing the one k/v pair.""" + + expires = None + if morsel["max-age"]: + try: + expires = int(time.time() + int(morsel["max-age"])) + except ValueError: + raise TypeError(f"max-age: {morsel['max-age']} must be integer") + elif morsel["expires"]: + time_template = "%a, %d-%b-%Y %H:%M:%S GMT" + expires = calendar.timegm(time.strptime(morsel["expires"], time_template)) + return create_cookie( + comment=morsel["comment"], + comment_url=bool(morsel["comment"]), + discard=False, + domain=morsel["domain"], + expires=expires, + name=morsel.key, + path=morsel["path"], + port=None, + rest={"HttpOnly": morsel["httponly"]}, + rfc2109=False, + secure=bool(morsel["secure"]), + value=morsel.value, + version=morsel["version"] or 0, + ) + + +def cookiejar_from_dict(cookie_dict, cookiejar=None, overwrite=True): + """Returns a CookieJar from a key/value dictionary. + + :param cookie_dict: Dict of key/values to insert into CookieJar. + :param cookiejar: (optional) A cookiejar to add the cookies to. + :param overwrite: (optional) If False, will not replace cookies + already in the jar with new ones. + :rtype: CookieJar + """ + if cookiejar is None: + cookiejar = RequestsCookieJar() + + if cookie_dict is not None: + names_from_jar = [cookie.name for cookie in cookiejar] + for name in cookie_dict: + if overwrite or (name not in names_from_jar): + cookiejar.set_cookie(create_cookie(name, cookie_dict[name])) + + return cookiejar + + +def merge_cookies(cookiejar, cookies): + """Add cookies to cookiejar and returns a merged CookieJar. + + :param cookiejar: CookieJar object to add the cookies to. + :param cookies: Dictionary or CookieJar object to be added. + :rtype: CookieJar + """ + if not isinstance(cookiejar, cookielib.CookieJar): + raise ValueError("You can only merge into CookieJar") + + if isinstance(cookies, dict): + cookiejar = cookiejar_from_dict(cookies, cookiejar=cookiejar, overwrite=False) + elif isinstance(cookies, cookielib.CookieJar): + try: + cookiejar.update(cookies) + except AttributeError: + for cookie_in_jar in cookies: + cookiejar.set_cookie(cookie_in_jar) + + return cookiejar diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/exceptions.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/exceptions.py new file mode 100644 index 0000000000000000000000000000000000000000..7f3660f00d9bd4525490d411af471d0333654ea0 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/exceptions.py @@ -0,0 +1,151 @@ +""" +requests.exceptions +~~~~~~~~~~~~~~~~~~~ + +This module contains the set of Requests' exceptions. +""" +from pip._vendor.urllib3.exceptions import HTTPError as BaseHTTPError + +from .compat import JSONDecodeError as CompatJSONDecodeError + + +class RequestException(IOError): + """There was an ambiguous exception that occurred while handling your + request. + """ + + def __init__(self, *args, **kwargs): + """Initialize RequestException with `request` and `response` objects.""" + response = kwargs.pop("response", None) + self.response = response + self.request = kwargs.pop("request", None) + if response is not None and not self.request and hasattr(response, "request"): + self.request = self.response.request + super().__init__(*args, **kwargs) + + +class InvalidJSONError(RequestException): + """A JSON error occurred.""" + + +class JSONDecodeError(InvalidJSONError, CompatJSONDecodeError): + """Couldn't decode the text into json""" + + def __init__(self, *args, **kwargs): + """ + Construct the JSONDecodeError instance first with all + args. Then use it's args to construct the IOError so that + the json specific args aren't used as IOError specific args + and the error message from JSONDecodeError is preserved. + """ + CompatJSONDecodeError.__init__(self, *args) + InvalidJSONError.__init__(self, *self.args, **kwargs) + + def __reduce__(self): + """ + The __reduce__ method called when pickling the object must + be the one from the JSONDecodeError (be it json/simplejson) + as it expects all the arguments for instantiation, not just + one like the IOError, and the MRO would by default call the + __reduce__ method from the IOError due to the inheritance order. + """ + return CompatJSONDecodeError.__reduce__(self) + + +class HTTPError(RequestException): + """An HTTP error occurred.""" + + +class ConnectionError(RequestException): + """A Connection error occurred.""" + + +class ProxyError(ConnectionError): + """A proxy error occurred.""" + + +class SSLError(ConnectionError): + """An SSL error occurred.""" + + +class Timeout(RequestException): + """The request timed out. + + Catching this error will catch both + :exc:`~requests.exceptions.ConnectTimeout` and + :exc:`~requests.exceptions.ReadTimeout` errors. + """ + + +class ConnectTimeout(ConnectionError, Timeout): + """The request timed out while trying to connect to the remote server. + + Requests that produced this error are safe to retry. + """ + + +class ReadTimeout(Timeout): + """The server did not send any data in the allotted amount of time.""" + + +class URLRequired(RequestException): + """A valid URL is required to make a request.""" + + +class TooManyRedirects(RequestException): + """Too many redirects.""" + + +class MissingSchema(RequestException, ValueError): + """The URL scheme (e.g. http or https) is missing.""" + + +class InvalidSchema(RequestException, ValueError): + """The URL scheme provided is either invalid or unsupported.""" + + +class InvalidURL(RequestException, ValueError): + """The URL provided was somehow invalid.""" + + +class InvalidHeader(RequestException, ValueError): + """The header value provided was somehow invalid.""" + + +class InvalidProxyURL(InvalidURL): + """The proxy URL provided is invalid.""" + + +class ChunkedEncodingError(RequestException): + """The server declared chunked encoding but sent an invalid chunk.""" + + +class ContentDecodingError(RequestException, BaseHTTPError): + """Failed to decode response content.""" + + +class StreamConsumedError(RequestException, TypeError): + """The content for this response was already consumed.""" + + +class RetryError(RequestException): + """Custom retries logic failed""" + + +class UnrewindableBodyError(RequestException): + """Requests encountered an error when trying to rewind a body.""" + + +# Warnings + + +class RequestsWarning(Warning): + """Base warning for Requests.""" + + +class FileModeWarning(RequestsWarning, DeprecationWarning): + """A file was opened in text mode, but Requests determined its binary length.""" + + +class RequestsDependencyWarning(RequestsWarning): + """An imported dependency doesn't match the expected version range.""" diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/help.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/help.py new file mode 100644 index 0000000000000000000000000000000000000000..ddbb6150d644cf13ffd365f40c4e01d73a9cab32 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/help.py @@ -0,0 +1,127 @@ +"""Module containing bug report helper(s).""" + +import json +import platform +import ssl +import sys + +from pip._vendor import idna +from pip._vendor import urllib3 + +from . import __version__ as requests_version + +charset_normalizer = None +chardet = None + +try: + from pip._vendor.urllib3.contrib import pyopenssl +except ImportError: + pyopenssl = None + OpenSSL = None + cryptography = None +else: + import cryptography + import OpenSSL + + +def _implementation(): + """Return a dict with the Python implementation and version. + + Provide both the name and the version of the Python implementation + currently running. For example, on CPython 3.10.3 it will return + {'name': 'CPython', 'version': '3.10.3'}. + + This function works best on CPython and PyPy: in particular, it probably + doesn't work for Jython or IronPython. Future investigation should be done + to work out the correct shape of the code for those platforms. + """ + implementation = platform.python_implementation() + + if implementation == "CPython": + implementation_version = platform.python_version() + elif implementation == "PyPy": + implementation_version = "{}.{}.{}".format( + sys.pypy_version_info.major, + sys.pypy_version_info.minor, + sys.pypy_version_info.micro, + ) + if sys.pypy_version_info.releaselevel != "final": + implementation_version = "".join( + [implementation_version, sys.pypy_version_info.releaselevel] + ) + elif implementation == "Jython": + implementation_version = platform.python_version() # Complete Guess + elif implementation == "IronPython": + implementation_version = platform.python_version() # Complete Guess + else: + implementation_version = "Unknown" + + return {"name": implementation, "version": implementation_version} + + +def info(): + """Generate information for a bug report.""" + try: + platform_info = { + "system": platform.system(), + "release": platform.release(), + } + except OSError: + platform_info = { + "system": "Unknown", + "release": "Unknown", + } + + implementation_info = _implementation() + urllib3_info = {"version": urllib3.__version__} + charset_normalizer_info = {"version": None} + chardet_info = {"version": None} + if charset_normalizer: + charset_normalizer_info = {"version": charset_normalizer.__version__} + if chardet: + chardet_info = {"version": chardet.__version__} + + pyopenssl_info = { + "version": None, + "openssl_version": "", + } + if OpenSSL: + pyopenssl_info = { + "version": OpenSSL.__version__, + "openssl_version": f"{OpenSSL.SSL.OPENSSL_VERSION_NUMBER:x}", + } + cryptography_info = { + "version": getattr(cryptography, "__version__", ""), + } + idna_info = { + "version": getattr(idna, "__version__", ""), + } + + system_ssl = ssl.OPENSSL_VERSION_NUMBER + system_ssl_info = {"version": f"{system_ssl:x}" if system_ssl is not None else ""} + + return { + "platform": platform_info, + "implementation": implementation_info, + "system_ssl": system_ssl_info, + "using_pyopenssl": pyopenssl is not None, + "using_charset_normalizer": chardet is None, + "pyOpenSSL": pyopenssl_info, + "urllib3": urllib3_info, + "chardet": chardet_info, + "charset_normalizer": charset_normalizer_info, + "cryptography": cryptography_info, + "idna": idna_info, + "requests": { + "version": requests_version, + }, + } + + +def main(): + """Pretty-print the bug information as JSON.""" + print(json.dumps(info(), sort_keys=True, indent=2)) + + +if __name__ == "__main__": + main() diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/models.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/models.py new file mode 100644 index 0000000000000000000000000000000000000000..85a008cfb56d42f99d3fe277368d954619b6bdc9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/models.py @@ -0,0 +1,1037 @@ +""" +requests.models +~~~~~~~~~~~~~~~ + +This module contains the primary objects that power Requests. +""" + +import datetime + +# Import encoding now, to avoid implicit import later. +# Implicit import within threads may cause LookupError when standard library is in a ZIP, +# such as in Embedded Python. See https://github.com/psf/requests/issues/3578. +import encodings.idna # noqa: F401 +from io import UnsupportedOperation + +from pip._vendor.urllib3.exceptions import ( + DecodeError, + LocationParseError, + ProtocolError, + ReadTimeoutError, + SSLError, +) +from pip._vendor.urllib3.fields import RequestField +from pip._vendor.urllib3.filepost import encode_multipart_formdata +from pip._vendor.urllib3.util import parse_url + +from ._internal_utils import to_native_string, unicode_is_ascii +from .auth import HTTPBasicAuth +from .compat import ( + Callable, + JSONDecodeError, + Mapping, + basestring, + builtin_str, + chardet, + cookielib, +) +from .compat import json as complexjson +from .compat import urlencode, urlsplit, urlunparse +from .cookies import _copy_cookie_jar, cookiejar_from_dict, get_cookie_header +from .exceptions import ( + ChunkedEncodingError, + ConnectionError, + ContentDecodingError, + HTTPError, + InvalidJSONError, + InvalidURL, +) +from .exceptions import JSONDecodeError as RequestsJSONDecodeError +from .exceptions import MissingSchema +from .exceptions import SSLError as RequestsSSLError +from .exceptions import StreamConsumedError +from .hooks import default_hooks +from .status_codes import codes +from .structures import CaseInsensitiveDict +from .utils import ( + check_header_validity, + get_auth_from_url, + guess_filename, + guess_json_utf, + iter_slices, + parse_header_links, + requote_uri, + stream_decode_response_unicode, + super_len, + to_key_val_list, +) + +#: The set of HTTP status codes that indicate an automatically +#: processable redirect. +REDIRECT_STATI = ( + codes.moved, # 301 + codes.found, # 302 + codes.other, # 303 + codes.temporary_redirect, # 307 + codes.permanent_redirect, # 308 +) + +DEFAULT_REDIRECT_LIMIT = 30 +CONTENT_CHUNK_SIZE = 10 * 1024 +ITER_CHUNK_SIZE = 512 + + +class RequestEncodingMixin: + @property + def path_url(self): + """Build the path URL to use.""" + + url = [] + + p = urlsplit(self.url) + + path = p.path + if not path: + path = "/" + + url.append(path) + + query = p.query + if query: + url.append("?") + url.append(query) + + return "".join(url) + + @staticmethod + def _encode_params(data): + """Encode parameters in a piece of data. + + Will successfully encode parameters when passed as a dict or a list of + 2-tuples. Order is retained if data is a list of 2-tuples but arbitrary + if parameters are supplied as a dict. + """ + + if isinstance(data, (str, bytes)): + return data + elif hasattr(data, "read"): + return data + elif hasattr(data, "__iter__"): + result = [] + for k, vs in to_key_val_list(data): + if isinstance(vs, basestring) or not hasattr(vs, "__iter__"): + vs = [vs] + for v in vs: + if v is not None: + result.append( + ( + k.encode("utf-8") if isinstance(k, str) else k, + v.encode("utf-8") if isinstance(v, str) else v, + ) + ) + return urlencode(result, doseq=True) + else: + return data + + @staticmethod + def _encode_files(files, data): + """Build the body for a multipart/form-data request. + + Will successfully encode files when passed as a dict or a list of + tuples. Order is retained if data is a list of tuples but arbitrary + if parameters are supplied as a dict. + The tuples may be 2-tuples (filename, fileobj), 3-tuples (filename, fileobj, contentype) + or 4-tuples (filename, fileobj, contentype, custom_headers). + """ + if not files: + raise ValueError("Files must be provided.") + elif isinstance(data, basestring): + raise ValueError("Data must not be a string.") + + new_fields = [] + fields = to_key_val_list(data or {}) + files = to_key_val_list(files or {}) + + for field, val in fields: + if isinstance(val, basestring) or not hasattr(val, "__iter__"): + val = [val] + for v in val: + if v is not None: + # Don't call str() on bytestrings: in Py3 it all goes wrong. + if not isinstance(v, bytes): + v = str(v) + + new_fields.append( + ( + field.decode("utf-8") + if isinstance(field, bytes) + else field, + v.encode("utf-8") if isinstance(v, str) else v, + ) + ) + + for k, v in files: + # support for explicit filename + ft = None + fh = None + if isinstance(v, (tuple, list)): + if len(v) == 2: + fn, fp = v + elif len(v) == 3: + fn, fp, ft = v + else: + fn, fp, ft, fh = v + else: + fn = guess_filename(v) or k + fp = v + + if isinstance(fp, (str, bytes, bytearray)): + fdata = fp + elif hasattr(fp, "read"): + fdata = fp.read() + elif fp is None: + continue + else: + fdata = fp + + rf = RequestField(name=k, data=fdata, filename=fn, headers=fh) + rf.make_multipart(content_type=ft) + new_fields.append(rf) + + body, content_type = encode_multipart_formdata(new_fields) + + return body, content_type + + +class RequestHooksMixin: + def register_hook(self, event, hook): + """Properly register a hook.""" + + if event not in self.hooks: + raise ValueError(f'Unsupported event specified, with event name "{event}"') + + if isinstance(hook, Callable): + self.hooks[event].append(hook) + elif hasattr(hook, "__iter__"): + self.hooks[event].extend(h for h in hook if isinstance(h, Callable)) + + def deregister_hook(self, event, hook): + """Deregister a previously registered hook. + Returns True if the hook existed, False if not. + """ + + try: + self.hooks[event].remove(hook) + return True + except ValueError: + return False + + +class Request(RequestHooksMixin): + """A user-created :class:`Request ` object. + + Used to prepare a :class:`PreparedRequest `, which is sent to the server. + + :param method: HTTP method to use. + :param url: URL to send. + :param headers: dictionary of headers to send. + :param files: dictionary of {filename: fileobject} files to multipart upload. + :param data: the body to attach to the request. If a dictionary or + list of tuples ``[(key, value)]`` is provided, form-encoding will + take place. + :param json: json for the body to attach to the request (if files or data is not specified). + :param params: URL parameters to append to the URL. If a dictionary or + list of tuples ``[(key, value)]`` is provided, form-encoding will + take place. + :param auth: Auth handler or (user, pass) tuple. + :param cookies: dictionary or CookieJar of cookies to attach to this request. + :param hooks: dictionary of callback hooks, for internal usage. + + Usage:: + + >>> import requests + >>> req = requests.Request('GET', 'https://httpbin.org/get') + >>> req.prepare() + + """ + + def __init__( + self, + method=None, + url=None, + headers=None, + files=None, + data=None, + params=None, + auth=None, + cookies=None, + hooks=None, + json=None, + ): + # Default empty dicts for dict params. + data = [] if data is None else data + files = [] if files is None else files + headers = {} if headers is None else headers + params = {} if params is None else params + hooks = {} if hooks is None else hooks + + self.hooks = default_hooks() + for k, v in list(hooks.items()): + self.register_hook(event=k, hook=v) + + self.method = method + self.url = url + self.headers = headers + self.files = files + self.data = data + self.json = json + self.params = params + self.auth = auth + self.cookies = cookies + + def __repr__(self): + return f"" + + def prepare(self): + """Constructs a :class:`PreparedRequest ` for transmission and returns it.""" + p = PreparedRequest() + p.prepare( + method=self.method, + url=self.url, + headers=self.headers, + files=self.files, + data=self.data, + json=self.json, + params=self.params, + auth=self.auth, + cookies=self.cookies, + hooks=self.hooks, + ) + return p + + +class PreparedRequest(RequestEncodingMixin, RequestHooksMixin): + """The fully mutable :class:`PreparedRequest ` object, + containing the exact bytes that will be sent to the server. + + Instances are generated from a :class:`Request ` object, and + should not be instantiated manually; doing so may produce undesirable + effects. + + Usage:: + + >>> import requests + >>> req = requests.Request('GET', 'https://httpbin.org/get') + >>> r = req.prepare() + >>> r + + + >>> s = requests.Session() + >>> s.send(r) + + """ + + def __init__(self): + #: HTTP verb to send to the server. + self.method = None + #: HTTP URL to send the request to. + self.url = None + #: dictionary of HTTP headers. + self.headers = None + # The `CookieJar` used to create the Cookie header will be stored here + # after prepare_cookies is called + self._cookies = None + #: request body to send to the server. + self.body = None + #: dictionary of callback hooks, for internal usage. + self.hooks = default_hooks() + #: integer denoting starting position of a readable file-like body. + self._body_position = None + + def prepare( + self, + method=None, + url=None, + headers=None, + files=None, + data=None, + params=None, + auth=None, + cookies=None, + hooks=None, + json=None, + ): + """Prepares the entire request with the given parameters.""" + + self.prepare_method(method) + self.prepare_url(url, params) + self.prepare_headers(headers) + self.prepare_cookies(cookies) + self.prepare_body(data, files, json) + self.prepare_auth(auth, url) + + # Note that prepare_auth must be last to enable authentication schemes + # such as OAuth to work on a fully prepared request. + + # This MUST go after prepare_auth. Authenticators could add a hook + self.prepare_hooks(hooks) + + def __repr__(self): + return f"" + + def copy(self): + p = PreparedRequest() + p.method = self.method + p.url = self.url + p.headers = self.headers.copy() if self.headers is not None else None + p._cookies = _copy_cookie_jar(self._cookies) + p.body = self.body + p.hooks = self.hooks + p._body_position = self._body_position + return p + + def prepare_method(self, method): + """Prepares the given HTTP method.""" + self.method = method + if self.method is not None: + self.method = to_native_string(self.method.upper()) + + @staticmethod + def _get_idna_encoded_host(host): + from pip._vendor import idna + + try: + host = idna.encode(host, uts46=True).decode("utf-8") + except idna.IDNAError: + raise UnicodeError + return host + + def prepare_url(self, url, params): + """Prepares the given HTTP URL.""" + #: Accept objects that have string representations. + #: We're unable to blindly call unicode/str functions + #: as this will include the bytestring indicator (b'') + #: on python 3.x. + #: https://github.com/psf/requests/pull/2238 + if isinstance(url, bytes): + url = url.decode("utf8") + else: + url = str(url) + + # Remove leading whitespaces from url + url = url.lstrip() + + # Don't do any URL preparation for non-HTTP schemes like `mailto`, + # `data` etc to work around exceptions from `url_parse`, which + # handles RFC 3986 only. + if ":" in url and not url.lower().startswith("http"): + self.url = url + return + + # Support for unicode domain names and paths. + try: + scheme, auth, host, port, path, query, fragment = parse_url(url) + except LocationParseError as e: + raise InvalidURL(*e.args) + + if not scheme: + raise MissingSchema( + f"Invalid URL {url!r}: No scheme supplied. " + f"Perhaps you meant https://{url}?" + ) + + if not host: + raise InvalidURL(f"Invalid URL {url!r}: No host supplied") + + # In general, we want to try IDNA encoding the hostname if the string contains + # non-ASCII characters. This allows users to automatically get the correct IDNA + # behaviour. For strings containing only ASCII characters, we need to also verify + # it doesn't start with a wildcard (*), before allowing the unencoded hostname. + if not unicode_is_ascii(host): + try: + host = self._get_idna_encoded_host(host) + except UnicodeError: + raise InvalidURL("URL has an invalid label.") + elif host.startswith(("*", ".")): + raise InvalidURL("URL has an invalid label.") + + # Carefully reconstruct the network location + netloc = auth or "" + if netloc: + netloc += "@" + netloc += host + if port: + netloc += f":{port}" + + # Bare domains aren't valid URLs. + if not path: + path = "/" + + if isinstance(params, (str, bytes)): + params = to_native_string(params) + + enc_params = self._encode_params(params) + if enc_params: + if query: + query = f"{query}&{enc_params}" + else: + query = enc_params + + url = requote_uri(urlunparse([scheme, netloc, path, None, query, fragment])) + self.url = url + + def prepare_headers(self, headers): + """Prepares the given HTTP headers.""" + + self.headers = CaseInsensitiveDict() + if headers: + for header in headers.items(): + # Raise exception on invalid header value. + check_header_validity(header) + name, value = header + self.headers[to_native_string(name)] = value + + def prepare_body(self, data, files, json=None): + """Prepares the given HTTP body data.""" + + # Check if file, fo, generator, iterator. + # If not, run through normal process. + + # Nottin' on you. + body = None + content_type = None + + if not data and json is not None: + # urllib3 requires a bytes-like body. Python 2's json.dumps + # provides this natively, but Python 3 gives a Unicode string. + content_type = "application/json" + + try: + body = complexjson.dumps(json, allow_nan=False) + except ValueError as ve: + raise InvalidJSONError(ve, request=self) + + if not isinstance(body, bytes): + body = body.encode("utf-8") + + is_stream = all( + [ + hasattr(data, "__iter__"), + not isinstance(data, (basestring, list, tuple, Mapping)), + ] + ) + + if is_stream: + try: + length = super_len(data) + except (TypeError, AttributeError, UnsupportedOperation): + length = None + + body = data + + if getattr(body, "tell", None) is not None: + # Record the current file position before reading. + # This will allow us to rewind a file in the event + # of a redirect. + try: + self._body_position = body.tell() + except OSError: + # This differentiates from None, allowing us to catch + # a failed `tell()` later when trying to rewind the body + self._body_position = object() + + if files: + raise NotImplementedError( + "Streamed bodies and files are mutually exclusive." + ) + + if length: + self.headers["Content-Length"] = builtin_str(length) + else: + self.headers["Transfer-Encoding"] = "chunked" + else: + # Multi-part file uploads. + if files: + (body, content_type) = self._encode_files(files, data) + else: + if data: + body = self._encode_params(data) + if isinstance(data, basestring) or hasattr(data, "read"): + content_type = None + else: + content_type = "application/x-www-form-urlencoded" + + self.prepare_content_length(body) + + # Add content-type if it wasn't explicitly provided. + if content_type and ("content-type" not in self.headers): + self.headers["Content-Type"] = content_type + + self.body = body + + def prepare_content_length(self, body): + """Prepare Content-Length header based on request method and body""" + if body is not None: + length = super_len(body) + if length: + # If length exists, set it. Otherwise, we fallback + # to Transfer-Encoding: chunked. + self.headers["Content-Length"] = builtin_str(length) + elif ( + self.method not in ("GET", "HEAD") + and self.headers.get("Content-Length") is None + ): + # Set Content-Length to 0 for methods that can have a body + # but don't provide one. (i.e. not GET or HEAD) + self.headers["Content-Length"] = "0" + + def prepare_auth(self, auth, url=""): + """Prepares the given HTTP auth data.""" + + # If no Auth is explicitly provided, extract it from the URL first. + if auth is None: + url_auth = get_auth_from_url(self.url) + auth = url_auth if any(url_auth) else None + + if auth: + if isinstance(auth, tuple) and len(auth) == 2: + # special-case basic HTTP auth + auth = HTTPBasicAuth(*auth) + + # Allow auth to make its changes. + r = auth(self) + + # Update self to reflect the auth changes. + self.__dict__.update(r.__dict__) + + # Recompute Content-Length + self.prepare_content_length(self.body) + + def prepare_cookies(self, cookies): + """Prepares the given HTTP cookie data. + + This function eventually generates a ``Cookie`` header from the + given cookies using cookielib. Due to cookielib's design, the header + will not be regenerated if it already exists, meaning this function + can only be called once for the life of the + :class:`PreparedRequest ` object. Any subsequent calls + to ``prepare_cookies`` will have no actual effect, unless the "Cookie" + header is removed beforehand. + """ + if isinstance(cookies, cookielib.CookieJar): + self._cookies = cookies + else: + self._cookies = cookiejar_from_dict(cookies) + + cookie_header = get_cookie_header(self._cookies, self) + if cookie_header is not None: + self.headers["Cookie"] = cookie_header + + def prepare_hooks(self, hooks): + """Prepares the given hooks.""" + # hooks can be passed as None to the prepare method and to this + # method. To prevent iterating over None, simply use an empty list + # if hooks is False-y + hooks = hooks or [] + for event in hooks: + self.register_hook(event, hooks[event]) + + +class Response: + """The :class:`Response ` object, which contains a + server's response to an HTTP request. + """ + + __attrs__ = [ + "_content", + "status_code", + "headers", + "url", + "history", + "encoding", + "reason", + "cookies", + "elapsed", + "request", + ] + + def __init__(self): + self._content = False + self._content_consumed = False + self._next = None + + #: Integer Code of responded HTTP Status, e.g. 404 or 200. + self.status_code = None + + #: Case-insensitive Dictionary of Response Headers. + #: For example, ``headers['content-encoding']`` will return the + #: value of a ``'Content-Encoding'`` response header. + self.headers = CaseInsensitiveDict() + + #: File-like object representation of response (for advanced usage). + #: Use of ``raw`` requires that ``stream=True`` be set on the request. + #: This requirement does not apply for use internally to Requests. + self.raw = None + + #: Final URL location of Response. + self.url = None + + #: Encoding to decode with when accessing r.text. + self.encoding = None + + #: A list of :class:`Response ` objects from + #: the history of the Request. Any redirect responses will end + #: up here. The list is sorted from the oldest to the most recent request. + self.history = [] + + #: Textual reason of responded HTTP Status, e.g. "Not Found" or "OK". + self.reason = None + + #: A CookieJar of Cookies the server sent back. + self.cookies = cookiejar_from_dict({}) + + #: The amount of time elapsed between sending the request + #: and the arrival of the response (as a timedelta). + #: This property specifically measures the time taken between sending + #: the first byte of the request and finishing parsing the headers. It + #: is therefore unaffected by consuming the response content or the + #: value of the ``stream`` keyword argument. + self.elapsed = datetime.timedelta(0) + + #: The :class:`PreparedRequest ` object to which this + #: is a response. + self.request = None + + def __enter__(self): + return self + + def __exit__(self, *args): + self.close() + + def __getstate__(self): + # Consume everything; accessing the content attribute makes + # sure the content has been fully read. + if not self._content_consumed: + self.content + + return {attr: getattr(self, attr, None) for attr in self.__attrs__} + + def __setstate__(self, state): + for name, value in state.items(): + setattr(self, name, value) + + # pickled objects do not have .raw + setattr(self, "_content_consumed", True) + setattr(self, "raw", None) + + def __repr__(self): + return f"" + + def __bool__(self): + """Returns True if :attr:`status_code` is less than 400. + + This attribute checks if the status code of the response is between + 400 and 600 to see if there was a client error or a server error. If + the status code, is between 200 and 400, this will return True. This + is **not** a check to see if the response code is ``200 OK``. + """ + return self.ok + + def __nonzero__(self): + """Returns True if :attr:`status_code` is less than 400. + + This attribute checks if the status code of the response is between + 400 and 600 to see if there was a client error or a server error. If + the status code, is between 200 and 400, this will return True. This + is **not** a check to see if the response code is ``200 OK``. + """ + return self.ok + + def __iter__(self): + """Allows you to use a response as an iterator.""" + return self.iter_content(128) + + @property + def ok(self): + """Returns True if :attr:`status_code` is less than 400, False if not. + + This attribute checks if the status code of the response is between + 400 and 600 to see if there was a client error or a server error. If + the status code is between 200 and 400, this will return True. This + is **not** a check to see if the response code is ``200 OK``. + """ + try: + self.raise_for_status() + except HTTPError: + return False + return True + + @property + def is_redirect(self): + """True if this Response is a well-formed HTTP redirect that could have + been processed automatically (by :meth:`Session.resolve_redirects`). + """ + return "location" in self.headers and self.status_code in REDIRECT_STATI + + @property + def is_permanent_redirect(self): + """True if this Response one of the permanent versions of redirect.""" + return "location" in self.headers and self.status_code in ( + codes.moved_permanently, + codes.permanent_redirect, + ) + + @property + def next(self): + """Returns a PreparedRequest for the next request in a redirect chain, if there is one.""" + return self._next + + @property + def apparent_encoding(self): + """The apparent encoding, provided by the charset_normalizer or chardet libraries.""" + if chardet is not None: + return chardet.detect(self.content)["encoding"] + else: + # If no character detection library is available, we'll fall back + # to a standard Python utf-8 str. + return "utf-8" + + def iter_content(self, chunk_size=1, decode_unicode=False): + """Iterates over the response data. When stream=True is set on the + request, this avoids reading the content at once into memory for + large responses. The chunk size is the number of bytes it should + read into memory. This is not necessarily the length of each item + returned as decoding can take place. + + chunk_size must be of type int or None. A value of None will + function differently depending on the value of `stream`. + stream=True will read data as it arrives in whatever size the + chunks are received. If stream=False, data is returned as + a single chunk. + + If decode_unicode is True, content will be decoded using the best + available encoding based on the response. + """ + + def generate(): + # Special case for urllib3. + if hasattr(self.raw, "stream"): + try: + yield from self.raw.stream(chunk_size, decode_content=True) + except ProtocolError as e: + raise ChunkedEncodingError(e) + except DecodeError as e: + raise ContentDecodingError(e) + except ReadTimeoutError as e: + raise ConnectionError(e) + except SSLError as e: + raise RequestsSSLError(e) + else: + # Standard file-like object. + while True: + chunk = self.raw.read(chunk_size) + if not chunk: + break + yield chunk + + self._content_consumed = True + + if self._content_consumed and isinstance(self._content, bool): + raise StreamConsumedError() + elif chunk_size is not None and not isinstance(chunk_size, int): + raise TypeError( + f"chunk_size must be an int, it is instead a {type(chunk_size)}." + ) + # simulate reading small chunks of the content + reused_chunks = iter_slices(self._content, chunk_size) + + stream_chunks = generate() + + chunks = reused_chunks if self._content_consumed else stream_chunks + + if decode_unicode: + chunks = stream_decode_response_unicode(chunks, self) + + return chunks + + def iter_lines( + self, chunk_size=ITER_CHUNK_SIZE, decode_unicode=False, delimiter=None + ): + """Iterates over the response data, one line at a time. When + stream=True is set on the request, this avoids reading the + content at once into memory for large responses. + + .. note:: This method is not reentrant safe. + """ + + pending = None + + for chunk in self.iter_content( + chunk_size=chunk_size, decode_unicode=decode_unicode + ): + if pending is not None: + chunk = pending + chunk + + if delimiter: + lines = chunk.split(delimiter) + else: + lines = chunk.splitlines() + + if lines and lines[-1] and chunk and lines[-1][-1] == chunk[-1]: + pending = lines.pop() + else: + pending = None + + yield from lines + + if pending is not None: + yield pending + + @property + def content(self): + """Content of the response, in bytes.""" + + if self._content is False: + # Read the contents. + if self._content_consumed: + raise RuntimeError("The content for this response was already consumed") + + if self.status_code == 0 or self.raw is None: + self._content = None + else: + self._content = b"".join(self.iter_content(CONTENT_CHUNK_SIZE)) or b"" + + self._content_consumed = True + # don't need to release the connection; that's been handled by urllib3 + # since we exhausted the data. + return self._content + + @property + def text(self): + """Content of the response, in unicode. + + If Response.encoding is None, encoding will be guessed using + ``charset_normalizer`` or ``chardet``. + + The encoding of the response content is determined based solely on HTTP + headers, following RFC 2616 to the letter. If you can take advantage of + non-HTTP knowledge to make a better guess at the encoding, you should + set ``r.encoding`` appropriately before accessing this property. + """ + + # Try charset from content-type + content = None + encoding = self.encoding + + if not self.content: + return "" + + # Fallback to auto-detected encoding. + if self.encoding is None: + encoding = self.apparent_encoding + + # Decode unicode from given encoding. + try: + content = str(self.content, encoding, errors="replace") + except (LookupError, TypeError): + # A LookupError is raised if the encoding was not found which could + # indicate a misspelling or similar mistake. + # + # A TypeError can be raised if encoding is None + # + # So we try blindly encoding. + content = str(self.content, errors="replace") + + return content + + def json(self, **kwargs): + r"""Returns the json-encoded content of a response, if any. + + :param \*\*kwargs: Optional arguments that ``json.loads`` takes. + :raises requests.exceptions.JSONDecodeError: If the response body does not + contain valid json. + """ + + if not self.encoding and self.content and len(self.content) > 3: + # No encoding set. JSON RFC 4627 section 3 states we should expect + # UTF-8, -16 or -32. Detect which one to use; If the detection or + # decoding fails, fall back to `self.text` (using charset_normalizer to make + # a best guess). + encoding = guess_json_utf(self.content) + if encoding is not None: + try: + return complexjson.loads(self.content.decode(encoding), **kwargs) + except UnicodeDecodeError: + # Wrong UTF codec detected; usually because it's not UTF-8 + # but some other 8-bit codec. This is an RFC violation, + # and the server didn't bother to tell us what codec *was* + # used. + pass + except JSONDecodeError as e: + raise RequestsJSONDecodeError(e.msg, e.doc, e.pos) + + try: + return complexjson.loads(self.text, **kwargs) + except JSONDecodeError as e: + # Catch JSON-related errors and raise as requests.JSONDecodeError + # This aliases json.JSONDecodeError and simplejson.JSONDecodeError + raise RequestsJSONDecodeError(e.msg, e.doc, e.pos) + + @property + def links(self): + """Returns the parsed header links of the response, if any.""" + + header = self.headers.get("link") + + resolved_links = {} + + if header: + links = parse_header_links(header) + + for link in links: + key = link.get("rel") or link.get("url") + resolved_links[key] = link + + return resolved_links + + def raise_for_status(self): + """Raises :class:`HTTPError`, if one occurred.""" + + http_error_msg = "" + if isinstance(self.reason, bytes): + # We attempt to decode utf-8 first because some servers + # choose to localize their reason strings. If the string + # isn't utf-8, we fall back to iso-8859-1 for all other + # encodings. (See PR #3538) + try: + reason = self.reason.decode("utf-8") + except UnicodeDecodeError: + reason = self.reason.decode("iso-8859-1") + else: + reason = self.reason + + if 400 <= self.status_code < 500: + http_error_msg = ( + f"{self.status_code} Client Error: {reason} for url: {self.url}" + ) + + elif 500 <= self.status_code < 600: + http_error_msg = ( + f"{self.status_code} Server Error: {reason} for url: {self.url}" + ) + + if http_error_msg: + raise HTTPError(http_error_msg, response=self) + + def close(self): + """Releases the connection back to the pool. Once this method has been + called the underlying ``raw`` object must not be accessed again. + + *Note: Should not normally need to be called explicitly.* + """ + if not self._content_consumed: + self.raw.close() + + release_conn = getattr(self.raw, "release_conn", None) + if release_conn is not None: + release_conn() diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/sessions.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/sessions.py new file mode 100644 index 0000000000000000000000000000000000000000..b387bc36df7bc064b502adcb3c1a4527dd401fda --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/sessions.py @@ -0,0 +1,831 @@ +""" +requests.sessions +~~~~~~~~~~~~~~~~~ + +This module provides a Session object to manage and persist settings across +requests (cookies, auth, proxies). +""" +import os +import sys +import time +from collections import OrderedDict +from datetime import timedelta + +from ._internal_utils import to_native_string +from .adapters import HTTPAdapter +from .auth import _basic_auth_str +from .compat import Mapping, cookielib, urljoin, urlparse +from .cookies import ( + RequestsCookieJar, + cookiejar_from_dict, + extract_cookies_to_jar, + merge_cookies, +) +from .exceptions import ( + ChunkedEncodingError, + ContentDecodingError, + InvalidSchema, + TooManyRedirects, +) +from .hooks import default_hooks, dispatch_hook + +# formerly defined here, reexposed here for backward compatibility +from .models import ( # noqa: F401 + DEFAULT_REDIRECT_LIMIT, + REDIRECT_STATI, + PreparedRequest, + Request, +) +from .status_codes import codes +from .structures import CaseInsensitiveDict +from .utils import ( # noqa: F401 + DEFAULT_PORTS, + default_headers, + get_auth_from_url, + get_environ_proxies, + get_netrc_auth, + requote_uri, + resolve_proxies, + rewind_body, + should_bypass_proxies, + to_key_val_list, +) + +# Preferred clock, based on which one is more accurate on a given system. +if sys.platform == "win32": + preferred_clock = time.perf_counter +else: + preferred_clock = time.time + + +def merge_setting(request_setting, session_setting, dict_class=OrderedDict): + """Determines appropriate setting for a given request, taking into account + the explicit setting on that request, and the setting in the session. If a + setting is a dictionary, they will be merged together using `dict_class` + """ + + if session_setting is None: + return request_setting + + if request_setting is None: + return session_setting + + # Bypass if not a dictionary (e.g. verify) + if not ( + isinstance(session_setting, Mapping) and isinstance(request_setting, Mapping) + ): + return request_setting + + merged_setting = dict_class(to_key_val_list(session_setting)) + merged_setting.update(to_key_val_list(request_setting)) + + # Remove keys that are set to None. Extract keys first to avoid altering + # the dictionary during iteration. + none_keys = [k for (k, v) in merged_setting.items() if v is None] + for key in none_keys: + del merged_setting[key] + + return merged_setting + + +def merge_hooks(request_hooks, session_hooks, dict_class=OrderedDict): + """Properly merges both requests and session hooks. + + This is necessary because when request_hooks == {'response': []}, the + merge breaks Session hooks entirely. + """ + if session_hooks is None or session_hooks.get("response") == []: + return request_hooks + + if request_hooks is None or request_hooks.get("response") == []: + return session_hooks + + return merge_setting(request_hooks, session_hooks, dict_class) + + +class SessionRedirectMixin: + def get_redirect_target(self, resp): + """Receives a Response. Returns a redirect URI or ``None``""" + # Due to the nature of how requests processes redirects this method will + # be called at least once upon the original response and at least twice + # on each subsequent redirect response (if any). + # If a custom mixin is used to handle this logic, it may be advantageous + # to cache the redirect location onto the response object as a private + # attribute. + if resp.is_redirect: + location = resp.headers["location"] + # Currently the underlying http module on py3 decode headers + # in latin1, but empirical evidence suggests that latin1 is very + # rarely used with non-ASCII characters in HTTP headers. + # It is more likely to get UTF8 header rather than latin1. + # This causes incorrect handling of UTF8 encoded location headers. + # To solve this, we re-encode the location in latin1. + location = location.encode("latin1") + return to_native_string(location, "utf8") + return None + + def should_strip_auth(self, old_url, new_url): + """Decide whether Authorization header should be removed when redirecting""" + old_parsed = urlparse(old_url) + new_parsed = urlparse(new_url) + if old_parsed.hostname != new_parsed.hostname: + return True + # Special case: allow http -> https redirect when using the standard + # ports. This isn't specified by RFC 7235, but is kept to avoid + # breaking backwards compatibility with older versions of requests + # that allowed any redirects on the same host. + if ( + old_parsed.scheme == "http" + and old_parsed.port in (80, None) + and new_parsed.scheme == "https" + and new_parsed.port in (443, None) + ): + return False + + # Handle default port usage corresponding to scheme. + changed_port = old_parsed.port != new_parsed.port + changed_scheme = old_parsed.scheme != new_parsed.scheme + default_port = (DEFAULT_PORTS.get(old_parsed.scheme, None), None) + if ( + not changed_scheme + and old_parsed.port in default_port + and new_parsed.port in default_port + ): + return False + + # Standard case: root URI must match + return changed_port or changed_scheme + + def resolve_redirects( + self, + resp, + req, + stream=False, + timeout=None, + verify=True, + cert=None, + proxies=None, + yield_requests=False, + **adapter_kwargs, + ): + """Receives a Response. Returns a generator of Responses or Requests.""" + + hist = [] # keep track of history + + url = self.get_redirect_target(resp) + previous_fragment = urlparse(req.url).fragment + while url: + prepared_request = req.copy() + + # Update history and keep track of redirects. + # resp.history must ignore the original request in this loop + hist.append(resp) + resp.history = hist[1:] + + try: + resp.content # Consume socket so it can be released + except (ChunkedEncodingError, ContentDecodingError, RuntimeError): + resp.raw.read(decode_content=False) + + if len(resp.history) >= self.max_redirects: + raise TooManyRedirects( + f"Exceeded {self.max_redirects} redirects.", response=resp + ) + + # Release the connection back into the pool. + resp.close() + + # Handle redirection without scheme (see: RFC 1808 Section 4) + if url.startswith("//"): + parsed_rurl = urlparse(resp.url) + url = ":".join([to_native_string(parsed_rurl.scheme), url]) + + # Normalize url case and attach previous fragment if needed (RFC 7231 7.1.2) + parsed = urlparse(url) + if parsed.fragment == "" and previous_fragment: + parsed = parsed._replace(fragment=previous_fragment) + elif parsed.fragment: + previous_fragment = parsed.fragment + url = parsed.geturl() + + # Facilitate relative 'location' headers, as allowed by RFC 7231. + # (e.g. '/path/to/resource' instead of 'http://domain.tld/path/to/resource') + # Compliant with RFC3986, we percent encode the url. + if not parsed.netloc: + url = urljoin(resp.url, requote_uri(url)) + else: + url = requote_uri(url) + + prepared_request.url = to_native_string(url) + + self.rebuild_method(prepared_request, resp) + + # https://github.com/psf/requests/issues/1084 + if resp.status_code not in ( + codes.temporary_redirect, + codes.permanent_redirect, + ): + # https://github.com/psf/requests/issues/3490 + purged_headers = ("Content-Length", "Content-Type", "Transfer-Encoding") + for header in purged_headers: + prepared_request.headers.pop(header, None) + prepared_request.body = None + + headers = prepared_request.headers + headers.pop("Cookie", None) + + # Extract any cookies sent on the response to the cookiejar + # in the new request. Because we've mutated our copied prepared + # request, use the old one that we haven't yet touched. + extract_cookies_to_jar(prepared_request._cookies, req, resp.raw) + merge_cookies(prepared_request._cookies, self.cookies) + prepared_request.prepare_cookies(prepared_request._cookies) + + # Rebuild auth and proxy information. + proxies = self.rebuild_proxies(prepared_request, proxies) + self.rebuild_auth(prepared_request, resp) + + # A failed tell() sets `_body_position` to `object()`. This non-None + # value ensures `rewindable` will be True, allowing us to raise an + # UnrewindableBodyError, instead of hanging the connection. + rewindable = prepared_request._body_position is not None and ( + "Content-Length" in headers or "Transfer-Encoding" in headers + ) + + # Attempt to rewind consumed file-like object. + if rewindable: + rewind_body(prepared_request) + + # Override the original request. + req = prepared_request + + if yield_requests: + yield req + else: + resp = self.send( + req, + stream=stream, + timeout=timeout, + verify=verify, + cert=cert, + proxies=proxies, + allow_redirects=False, + **adapter_kwargs, + ) + + extract_cookies_to_jar(self.cookies, prepared_request, resp.raw) + + # extract redirect url, if any, for the next loop + url = self.get_redirect_target(resp) + yield resp + + def rebuild_auth(self, prepared_request, response): + """When being redirected we may want to strip authentication from the + request to avoid leaking credentials. This method intelligently removes + and reapplies authentication where possible to avoid credential loss. + """ + headers = prepared_request.headers + url = prepared_request.url + + if "Authorization" in headers and self.should_strip_auth( + response.request.url, url + ): + # If we get redirected to a new host, we should strip out any + # authentication headers. + del headers["Authorization"] + + # .netrc might have more auth for us on our new host. + new_auth = get_netrc_auth(url) if self.trust_env else None + if new_auth is not None: + prepared_request.prepare_auth(new_auth) + + def rebuild_proxies(self, prepared_request, proxies): + """This method re-evaluates the proxy configuration by considering the + environment variables. If we are redirected to a URL covered by + NO_PROXY, we strip the proxy configuration. Otherwise, we set missing + proxy keys for this URL (in case they were stripped by a previous + redirect). + + This method also replaces the Proxy-Authorization header where + necessary. + + :rtype: dict + """ + headers = prepared_request.headers + scheme = urlparse(prepared_request.url).scheme + new_proxies = resolve_proxies(prepared_request, proxies, self.trust_env) + + if "Proxy-Authorization" in headers: + del headers["Proxy-Authorization"] + + try: + username, password = get_auth_from_url(new_proxies[scheme]) + except KeyError: + username, password = None, None + + # urllib3 handles proxy authorization for us in the standard adapter. + # Avoid appending this to TLS tunneled requests where it may be leaked. + if not scheme.startswith("https") and username and password: + headers["Proxy-Authorization"] = _basic_auth_str(username, password) + + return new_proxies + + def rebuild_method(self, prepared_request, response): + """When being redirected we may want to change the method of the request + based on certain specs or browser behavior. + """ + method = prepared_request.method + + # https://tools.ietf.org/html/rfc7231#section-6.4.4 + if response.status_code == codes.see_other and method != "HEAD": + method = "GET" + + # Do what the browsers do, despite standards... + # First, turn 302s into GETs. + if response.status_code == codes.found and method != "HEAD": + method = "GET" + + # Second, if a POST is responded to with a 301, turn it into a GET. + # This bizarre behaviour is explained in Issue 1704. + if response.status_code == codes.moved and method == "POST": + method = "GET" + + prepared_request.method = method + + +class Session(SessionRedirectMixin): + """A Requests session. + + Provides cookie persistence, connection-pooling, and configuration. + + Basic Usage:: + + >>> import requests + >>> s = requests.Session() + >>> s.get('https://httpbin.org/get') + + + Or as a context manager:: + + >>> with requests.Session() as s: + ... s.get('https://httpbin.org/get') + + """ + + __attrs__ = [ + "headers", + "cookies", + "auth", + "proxies", + "hooks", + "params", + "verify", + "cert", + "adapters", + "stream", + "trust_env", + "max_redirects", + ] + + def __init__(self): + #: A case-insensitive dictionary of headers to be sent on each + #: :class:`Request ` sent from this + #: :class:`Session `. + self.headers = default_headers() + + #: Default Authentication tuple or object to attach to + #: :class:`Request `. + self.auth = None + + #: Dictionary mapping protocol or protocol and host to the URL of the proxy + #: (e.g. {'http': 'foo.bar:3128', 'http://host.name': 'foo.bar:4012'}) to + #: be used on each :class:`Request `. + self.proxies = {} + + #: Event-handling hooks. + self.hooks = default_hooks() + + #: Dictionary of querystring data to attach to each + #: :class:`Request `. The dictionary values may be lists for + #: representing multivalued query parameters. + self.params = {} + + #: Stream response content default. + self.stream = False + + #: SSL Verification default. + #: Defaults to `True`, requiring requests to verify the TLS certificate at the + #: remote end. + #: If verify is set to `False`, requests will accept any TLS certificate + #: presented by the server, and will ignore hostname mismatches and/or + #: expired certificates, which will make your application vulnerable to + #: man-in-the-middle (MitM) attacks. + #: Only set this to `False` for testing. + self.verify = True + + #: SSL client certificate default, if String, path to ssl client + #: cert file (.pem). If Tuple, ('cert', 'key') pair. + self.cert = None + + #: Maximum number of redirects allowed. If the request exceeds this + #: limit, a :class:`TooManyRedirects` exception is raised. + #: This defaults to requests.models.DEFAULT_REDIRECT_LIMIT, which is + #: 30. + self.max_redirects = DEFAULT_REDIRECT_LIMIT + + #: Trust environment settings for proxy configuration, default + #: authentication and similar. + self.trust_env = True + + #: A CookieJar containing all currently outstanding cookies set on this + #: session. By default it is a + #: :class:`RequestsCookieJar `, but + #: may be any other ``cookielib.CookieJar`` compatible object. + self.cookies = cookiejar_from_dict({}) + + # Default connection adapters. + self.adapters = OrderedDict() + self.mount("https://", HTTPAdapter()) + self.mount("http://", HTTPAdapter()) + + def __enter__(self): + return self + + def __exit__(self, *args): + self.close() + + def prepare_request(self, request): + """Constructs a :class:`PreparedRequest ` for + transmission and returns it. The :class:`PreparedRequest` has settings + merged from the :class:`Request ` instance and those of the + :class:`Session`. + + :param request: :class:`Request` instance to prepare with this + session's settings. + :rtype: requests.PreparedRequest + """ + cookies = request.cookies or {} + + # Bootstrap CookieJar. + if not isinstance(cookies, cookielib.CookieJar): + cookies = cookiejar_from_dict(cookies) + + # Merge with session cookies + merged_cookies = merge_cookies( + merge_cookies(RequestsCookieJar(), self.cookies), cookies + ) + + # Set environment's basic authentication if not explicitly set. + auth = request.auth + if self.trust_env and not auth and not self.auth: + auth = get_netrc_auth(request.url) + + p = PreparedRequest() + p.prepare( + method=request.method.upper(), + url=request.url, + files=request.files, + data=request.data, + json=request.json, + headers=merge_setting( + request.headers, self.headers, dict_class=CaseInsensitiveDict + ), + params=merge_setting(request.params, self.params), + auth=merge_setting(auth, self.auth), + cookies=merged_cookies, + hooks=merge_hooks(request.hooks, self.hooks), + ) + return p + + def request( + self, + method, + url, + params=None, + data=None, + headers=None, + cookies=None, + files=None, + auth=None, + timeout=None, + allow_redirects=True, + proxies=None, + hooks=None, + stream=None, + verify=None, + cert=None, + json=None, + ): + """Constructs a :class:`Request `, prepares it and sends it. + Returns :class:`Response ` object. + + :param method: method for the new :class:`Request` object. + :param url: URL for the new :class:`Request` object. + :param params: (optional) Dictionary or bytes to be sent in the query + string for the :class:`Request`. + :param data: (optional) Dictionary, list of tuples, bytes, or file-like + object to send in the body of the :class:`Request`. + :param json: (optional) json to send in the body of the + :class:`Request`. + :param headers: (optional) Dictionary of HTTP Headers to send with the + :class:`Request`. + :param cookies: (optional) Dict or CookieJar object to send with the + :class:`Request`. + :param files: (optional) Dictionary of ``'filename': file-like-objects`` + for multipart encoding upload. + :param auth: (optional) Auth tuple or callable to enable + Basic/Digest/Custom HTTP Auth. + :param timeout: (optional) How long to wait for the server to send + data before giving up, as a float, or a :ref:`(connect timeout, + read timeout) ` tuple. + :type timeout: float or tuple + :param allow_redirects: (optional) Set to True by default. + :type allow_redirects: bool + :param proxies: (optional) Dictionary mapping protocol or protocol and + hostname to the URL of the proxy. + :param hooks: (optional) Dictionary mapping hook name to one event or + list of events, event must be callable. + :param stream: (optional) whether to immediately download the response + content. Defaults to ``False``. + :param verify: (optional) Either a boolean, in which case it controls whether we verify + the server's TLS certificate, or a string, in which case it must be a path + to a CA bundle to use. Defaults to ``True``. When set to + ``False``, requests will accept any TLS certificate presented by + the server, and will ignore hostname mismatches and/or expired + certificates, which will make your application vulnerable to + man-in-the-middle (MitM) attacks. Setting verify to ``False`` + may be useful during local development or testing. + :param cert: (optional) if String, path to ssl client cert file (.pem). + If Tuple, ('cert', 'key') pair. + :rtype: requests.Response + """ + # Create the Request. + req = Request( + method=method.upper(), + url=url, + headers=headers, + files=files, + data=data or {}, + json=json, + params=params or {}, + auth=auth, + cookies=cookies, + hooks=hooks, + ) + prep = self.prepare_request(req) + + proxies = proxies or {} + + settings = self.merge_environment_settings( + prep.url, proxies, stream, verify, cert + ) + + # Send the request. + send_kwargs = { + "timeout": timeout, + "allow_redirects": allow_redirects, + } + send_kwargs.update(settings) + resp = self.send(prep, **send_kwargs) + + return resp + + def get(self, url, **kwargs): + r"""Sends a GET request. Returns :class:`Response` object. + + :param url: URL for the new :class:`Request` object. + :param \*\*kwargs: Optional arguments that ``request`` takes. + :rtype: requests.Response + """ + + kwargs.setdefault("allow_redirects", True) + return self.request("GET", url, **kwargs) + + def options(self, url, **kwargs): + r"""Sends a OPTIONS request. Returns :class:`Response` object. + + :param url: URL for the new :class:`Request` object. + :param \*\*kwargs: Optional arguments that ``request`` takes. + :rtype: requests.Response + """ + + kwargs.setdefault("allow_redirects", True) + return self.request("OPTIONS", url, **kwargs) + + def head(self, url, **kwargs): + r"""Sends a HEAD request. Returns :class:`Response` object. + + :param url: URL for the new :class:`Request` object. + :param \*\*kwargs: Optional arguments that ``request`` takes. + :rtype: requests.Response + """ + + kwargs.setdefault("allow_redirects", False) + return self.request("HEAD", url, **kwargs) + + def post(self, url, data=None, json=None, **kwargs): + r"""Sends a POST request. Returns :class:`Response` object. + + :param url: URL for the new :class:`Request` object. + :param data: (optional) Dictionary, list of tuples, bytes, or file-like + object to send in the body of the :class:`Request`. + :param json: (optional) json to send in the body of the :class:`Request`. + :param \*\*kwargs: Optional arguments that ``request`` takes. + :rtype: requests.Response + """ + + return self.request("POST", url, data=data, json=json, **kwargs) + + def put(self, url, data=None, **kwargs): + r"""Sends a PUT request. Returns :class:`Response` object. + + :param url: URL for the new :class:`Request` object. + :param data: (optional) Dictionary, list of tuples, bytes, or file-like + object to send in the body of the :class:`Request`. + :param \*\*kwargs: Optional arguments that ``request`` takes. + :rtype: requests.Response + """ + + return self.request("PUT", url, data=data, **kwargs) + + def patch(self, url, data=None, **kwargs): + r"""Sends a PATCH request. Returns :class:`Response` object. + + :param url: URL for the new :class:`Request` object. + :param data: (optional) Dictionary, list of tuples, bytes, or file-like + object to send in the body of the :class:`Request`. + :param \*\*kwargs: Optional arguments that ``request`` takes. + :rtype: requests.Response + """ + + return self.request("PATCH", url, data=data, **kwargs) + + def delete(self, url, **kwargs): + r"""Sends a DELETE request. Returns :class:`Response` object. + + :param url: URL for the new :class:`Request` object. + :param \*\*kwargs: Optional arguments that ``request`` takes. + :rtype: requests.Response + """ + + return self.request("DELETE", url, **kwargs) + + def send(self, request, **kwargs): + """Send a given PreparedRequest. + + :rtype: requests.Response + """ + # Set defaults that the hooks can utilize to ensure they always have + # the correct parameters to reproduce the previous request. + kwargs.setdefault("stream", self.stream) + kwargs.setdefault("verify", self.verify) + kwargs.setdefault("cert", self.cert) + if "proxies" not in kwargs: + kwargs["proxies"] = resolve_proxies(request, self.proxies, self.trust_env) + + # It's possible that users might accidentally send a Request object. + # Guard against that specific failure case. + if isinstance(request, Request): + raise ValueError("You can only send PreparedRequests.") + + # Set up variables needed for resolve_redirects and dispatching of hooks + allow_redirects = kwargs.pop("allow_redirects", True) + stream = kwargs.get("stream") + hooks = request.hooks + + # Get the appropriate adapter to use + adapter = self.get_adapter(url=request.url) + + # Start time (approximately) of the request + start = preferred_clock() + + # Send the request + r = adapter.send(request, **kwargs) + + # Total elapsed time of the request (approximately) + elapsed = preferred_clock() - start + r.elapsed = timedelta(seconds=elapsed) + + # Response manipulation hooks + r = dispatch_hook("response", hooks, r, **kwargs) + + # Persist cookies + if r.history: + # If the hooks create history then we want those cookies too + for resp in r.history: + extract_cookies_to_jar(self.cookies, resp.request, resp.raw) + + extract_cookies_to_jar(self.cookies, request, r.raw) + + # Resolve redirects if allowed. + if allow_redirects: + # Redirect resolving generator. + gen = self.resolve_redirects(r, request, **kwargs) + history = [resp for resp in gen] + else: + history = [] + + # Shuffle things around if there's history. + if history: + # Insert the first (original) request at the start + history.insert(0, r) + # Get the last request made + r = history.pop() + r.history = history + + # If redirects aren't being followed, store the response on the Request for Response.next(). + if not allow_redirects: + try: + r._next = next( + self.resolve_redirects(r, request, yield_requests=True, **kwargs) + ) + except StopIteration: + pass + + if not stream: + r.content + + return r + + def merge_environment_settings(self, url, proxies, stream, verify, cert): + """ + Check the environment and merge it with some settings. + + :rtype: dict + """ + # Gather clues from the surrounding environment. + if self.trust_env: + # Set environment's proxies. + no_proxy = proxies.get("no_proxy") if proxies is not None else None + env_proxies = get_environ_proxies(url, no_proxy=no_proxy) + for k, v in env_proxies.items(): + proxies.setdefault(k, v) + + # Look for requests environment configuration + # and be compatible with cURL. + if verify is True or verify is None: + verify = ( + os.environ.get("REQUESTS_CA_BUNDLE") + or os.environ.get("CURL_CA_BUNDLE") + or verify + ) + + # Merge all the kwargs. + proxies = merge_setting(proxies, self.proxies) + stream = merge_setting(stream, self.stream) + verify = merge_setting(verify, self.verify) + cert = merge_setting(cert, self.cert) + + return {"proxies": proxies, "stream": stream, "verify": verify, "cert": cert} + + def get_adapter(self, url): + """ + Returns the appropriate connection adapter for the given URL. + + :rtype: requests.adapters.BaseAdapter + """ + for prefix, adapter in self.adapters.items(): + if url.lower().startswith(prefix.lower()): + return adapter + + # Nothing matches :-/ + raise InvalidSchema(f"No connection adapters were found for {url!r}") + + def close(self): + """Closes all adapters and as such the session""" + for v in self.adapters.values(): + v.close() + + def mount(self, prefix, adapter): + """Registers a connection adapter to a prefix. + + Adapters are sorted in descending order by prefix length. + """ + self.adapters[prefix] = adapter + keys_to_move = [k for k in self.adapters if len(k) < len(prefix)] + + for key in keys_to_move: + self.adapters[key] = self.adapters.pop(key) + + def __getstate__(self): + state = {attr: getattr(self, attr, None) for attr in self.__attrs__} + return state + + def __setstate__(self, state): + for attr, value in state.items(): + setattr(self, attr, value) + + +def session(): + """ + Returns a :class:`Session` for context-management. + + .. deprecated:: 1.0.0 + + This method has been deprecated since version 1.0.0 and is only kept for + backwards compatibility. New code should use :class:`~requests.sessions.Session` + to create a session. This may be removed at a future date. + + :rtype: Session + """ + return Session() diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/structures.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/structures.py new file mode 100644 index 0000000000000000000000000000000000000000..188e13e4829591facb23ae0e2eda84b9807cb818 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/requests/structures.py @@ -0,0 +1,99 @@ +""" +requests.structures +~~~~~~~~~~~~~~~~~~~ + +Data structures that power Requests. +""" + +from collections import OrderedDict + +from .compat import Mapping, MutableMapping + + +class CaseInsensitiveDict(MutableMapping): + """A case-insensitive ``dict``-like object. + + Implements all methods and operations of + ``MutableMapping`` as well as dict's ``copy``. Also + provides ``lower_items``. + + All keys are expected to be strings. The structure remembers the + case of the last key to be set, and ``iter(instance)``, + ``keys()``, ``items()``, ``iterkeys()``, and ``iteritems()`` + will contain case-sensitive keys. However, querying and contains + testing is case insensitive:: + + cid = CaseInsensitiveDict() + cid['Accept'] = 'application/json' + cid['aCCEPT'] == 'application/json' # True + list(cid) == ['Accept'] # True + + For example, ``headers['content-encoding']`` will return the + value of a ``'Content-Encoding'`` response header, regardless + of how the header name was originally stored. + + If the constructor, ``.update``, or equality comparison + operations are given keys that have equal ``.lower()``s, the + behavior is undefined. + """ + + def __init__(self, data=None, **kwargs): + self._store = OrderedDict() + if data is None: + data = {} + self.update(data, **kwargs) + + def __setitem__(self, key, value): + # Use the lowercased key for lookups, but store the actual + # key alongside the value. + self._store[key.lower()] = (key, value) + + def __getitem__(self, key): + return self._store[key.lower()][1] + + def __delitem__(self, key): + del self._store[key.lower()] + + def __iter__(self): + return (casedkey for casedkey, mappedvalue in self._store.values()) + + def __len__(self): + return len(self._store) + + def lower_items(self): + """Like iteritems(), but with all lowercase keys.""" + return ((lowerkey, keyval[1]) for (lowerkey, keyval) in self._store.items()) + + def __eq__(self, other): + if isinstance(other, Mapping): + other = CaseInsensitiveDict(other) + else: + return NotImplemented + # Compare insensitively + return dict(self.lower_items()) == dict(other.lower_items()) + + # Copy is required + def copy(self): + return CaseInsensitiveDict(self._store.values()) + + def __repr__(self): + return str(dict(self.items())) + + +class LookupDict(dict): + """Dictionary lookup object.""" + + def __init__(self, name=None): + self.name = name + super().__init__() + + def __repr__(self): + return f"" + + def __getitem__(self, key): + # We allow fall-through here, so values default to None + + return self.__dict__.get(key, None) + + def get(self, key, default=None): + return self.__dict__.get(key, default) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/resolvelib/__pycache__/reporters.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/resolvelib/__pycache__/reporters.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..89d22492836c9dde55a79316d0c484ef959f1dd9 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/resolvelib/__pycache__/reporters.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/resolvelib/compat/__pycache__/__init__.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/resolvelib/compat/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..e5085b4fb87b2225b00f2a891732c36330155f98 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/resolvelib/compat/__pycache__/__init__.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/_stack.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/_stack.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..b1021dda13cd2912ef6b3102f5fabf672e710566 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/_stack.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/_wrap.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/_wrap.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..98b8bd053c3db77a8c966cb9f177e98dd4b6f0eb Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/_wrap.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/constrain.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/constrain.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..05d4ed19c00ceb23287de81958a374731c1df0e5 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/constrain.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/file_proxy.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/file_proxy.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..7fb149bf2c6e9ddcb3f28def021235a49e2b18a1 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/file_proxy.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/filesize.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/filesize.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..5c14d3ccd8904e4d6412ed18ca505c68fc9d1e84 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/filesize.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/layout.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/layout.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..42f94c92b8c16a235b03252b77612048c6128445 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/layout.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/palette.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/palette.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..79c4e9c60c1f6119a8c8e011c83b1f4a048e8b68 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/palette.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/status.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/status.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..10b37588ba5f7717e3bc4f8c27b1ad14121ea04e Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/status.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/table.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/table.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..ccc2d90285460cf886df4ba082aca37cbd1a85c1 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/table.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/traceback.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/traceback.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..1d025b2442d4ac2e1f87f9e12d0879a1a8152595 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/traceback.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/tree.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/tree.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..29e10af498bc572ab8ec31c7dec4220efeb9754a Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/rich/__pycache__/tree.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/_appengine_environ.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/_appengine_environ.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..ece5c2ee53aacf94e202466762a68982198e596b Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/_appengine_environ.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/ntlmpool.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/ntlmpool.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..e5397a037ecc26d4acba79017239e46ec9dbcc6b Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/ntlmpool.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/pyopenssl.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/pyopenssl.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..57fb25acdfe8f699ac05267ff57b39e59f810f86 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/pyopenssl.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/socks.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/socks.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..f241f7592cde11671babb06a8664bf1f680ab6e0 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/contrib/__pycache__/socks.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/__init__.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..4547fc522b690ba2697843edd044f2039a4123a9 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/__init__.py @@ -0,0 +1,49 @@ +from __future__ import absolute_import + +# For backwards compatibility, provide imports that used to be here. +from .connection import is_connection_dropped +from .request import SKIP_HEADER, SKIPPABLE_HEADERS, make_headers +from .response import is_fp_closed +from .retry import Retry +from .ssl_ import ( + ALPN_PROTOCOLS, + HAS_SNI, + IS_PYOPENSSL, + IS_SECURETRANSPORT, + PROTOCOL_TLS, + SSLContext, + assert_fingerprint, + resolve_cert_reqs, + resolve_ssl_version, + ssl_wrap_socket, +) +from .timeout import Timeout, current_time +from .url import Url, get_host, parse_url, split_first +from .wait import wait_for_read, wait_for_write + +__all__ = ( + "HAS_SNI", + "IS_PYOPENSSL", + "IS_SECURETRANSPORT", + "SSLContext", + "PROTOCOL_TLS", + "ALPN_PROTOCOLS", + "Retry", + "Timeout", + "Url", + "assert_fingerprint", + "current_time", + "is_connection_dropped", + "is_fp_closed", + "get_host", + "parse_url", + "make_headers", + "resolve_cert_reqs", + "resolve_ssl_version", + "split_first", + "ssl_wrap_socket", + "wait_for_read", + "wait_for_write", + "SKIP_HEADER", + "SKIPPABLE_HEADERS", +) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/connection.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/connection.py new file mode 100644 index 0000000000000000000000000000000000000000..6af1138f260e4eaaa0aa242f7f50b918a283b49f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/connection.py @@ -0,0 +1,149 @@ +from __future__ import absolute_import + +import socket + +from ..contrib import _appengine_environ +from ..exceptions import LocationParseError +from ..packages import six +from .wait import NoWayToWaitForSocketError, wait_for_read + + +def is_connection_dropped(conn): # Platform-specific + """ + Returns True if the connection is dropped and should be closed. + + :param conn: + :class:`http.client.HTTPConnection` object. + + Note: For platforms like AppEngine, this will always return ``False`` to + let the platform handle connection recycling transparently for us. + """ + sock = getattr(conn, "sock", False) + if sock is False: # Platform-specific: AppEngine + return False + if sock is None: # Connection already closed (such as by httplib). + return True + try: + # Returns True if readable, which here means it's been dropped + return wait_for_read(sock, timeout=0.0) + except NoWayToWaitForSocketError: # Platform-specific: AppEngine + return False + + +# This function is copied from socket.py in the Python 2.7 standard +# library test suite. Added to its signature is only `socket_options`. +# One additional modification is that we avoid binding to IPv6 servers +# discovered in DNS if the system doesn't have IPv6 functionality. +def create_connection( + address, + timeout=socket._GLOBAL_DEFAULT_TIMEOUT, + source_address=None, + socket_options=None, +): + """Connect to *address* and return the socket object. + + Convenience function. Connect to *address* (a 2-tuple ``(host, + port)``) and return the socket object. Passing the optional + *timeout* parameter will set the timeout on the socket instance + before attempting to connect. If no *timeout* is supplied, the + global default timeout setting returned by :func:`socket.getdefaulttimeout` + is used. If *source_address* is set it must be a tuple of (host, port) + for the socket to bind as a source address before making the connection. + An host of '' or port 0 tells the OS to use the default. + """ + + host, port = address + if host.startswith("["): + host = host.strip("[]") + err = None + + # Using the value from allowed_gai_family() in the context of getaddrinfo lets + # us select whether to work with IPv4 DNS records, IPv6 records, or both. + # The original create_connection function always returns all records. + family = allowed_gai_family() + + try: + host.encode("idna") + except UnicodeError: + return six.raise_from( + LocationParseError(u"'%s', label empty or too long" % host), None + ) + + for res in socket.getaddrinfo(host, port, family, socket.SOCK_STREAM): + af, socktype, proto, canonname, sa = res + sock = None + try: + sock = socket.socket(af, socktype, proto) + + # If provided, set socket level options before connecting. + _set_socket_options(sock, socket_options) + + if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT: + sock.settimeout(timeout) + if source_address: + sock.bind(source_address) + sock.connect(sa) + return sock + + except socket.error as e: + err = e + if sock is not None: + sock.close() + sock = None + + if err is not None: + raise err + + raise socket.error("getaddrinfo returns an empty list") + + +def _set_socket_options(sock, options): + if options is None: + return + + for opt in options: + sock.setsockopt(*opt) + + +def allowed_gai_family(): + """This function is designed to work in the context of + getaddrinfo, where family=socket.AF_UNSPEC is the default and + will perform a DNS search for both IPv6 and IPv4 records.""" + + family = socket.AF_INET + if HAS_IPV6: + family = socket.AF_UNSPEC + return family + + +def _has_ipv6(host): + """Returns True if the system can bind an IPv6 address.""" + sock = None + has_ipv6 = False + + # App Engine doesn't support IPV6 sockets and actually has a quota on the + # number of sockets that can be used, so just early out here instead of + # creating a socket needlessly. + # See https://github.com/urllib3/urllib3/issues/1446 + if _appengine_environ.is_appengine_sandbox(): + return False + + if socket.has_ipv6: + # has_ipv6 returns true if cPython was compiled with IPv6 support. + # It does not tell us if the system has IPv6 support enabled. To + # determine that we must bind to an IPv6 address. + # https://github.com/urllib3/urllib3/pull/611 + # https://bugs.python.org/issue658327 + try: + sock = socket.socket(socket.AF_INET6) + sock.bind((host, 0)) + has_ipv6 = True + except Exception: + pass + + if sock: + sock.close() + return has_ipv6 + + +HAS_IPV6 = _has_ipv6("::1") diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/proxy.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/proxy.py new file mode 100644 index 0000000000000000000000000000000000000000..2199cc7b7f004009493d032720c36d6568f9d89e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/proxy.py @@ -0,0 +1,57 @@ +from .ssl_ import create_urllib3_context, resolve_cert_reqs, resolve_ssl_version + + +def connection_requires_http_tunnel( + proxy_url=None, proxy_config=None, destination_scheme=None +): + """ + Returns True if the connection requires an HTTP CONNECT through the proxy. + + :param URL proxy_url: + URL of the proxy. + :param ProxyConfig proxy_config: + Proxy configuration from poolmanager.py + :param str destination_scheme: + The scheme of the destination. (i.e https, http, etc) + """ + # If we're not using a proxy, no way to use a tunnel. + if proxy_url is None: + return False + + # HTTP destinations never require tunneling, we always forward. + if destination_scheme == "http": + return False + + # Support for forwarding with HTTPS proxies and HTTPS destinations. + if ( + proxy_url.scheme == "https" + and proxy_config + and proxy_config.use_forwarding_for_https + ): + return False + + # Otherwise always use a tunnel. + return True + + +def create_proxy_ssl_context( + ssl_version, cert_reqs, ca_certs=None, ca_cert_dir=None, ca_cert_data=None +): + """ + Generates a default proxy ssl context if one hasn't been provided by the + user. + """ + ssl_context = create_urllib3_context( + ssl_version=resolve_ssl_version(ssl_version), + cert_reqs=resolve_cert_reqs(cert_reqs), + ) + + if ( + not ca_certs + and not ca_cert_dir + and not ca_cert_data + and hasattr(ssl_context, "load_default_certs") + ): + ssl_context.load_default_certs() + + return ssl_context diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/queue.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/queue.py new file mode 100644 index 0000000000000000000000000000000000000000..41784104ee4bd5796006d1052536325d52db1e8c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/queue.py @@ -0,0 +1,22 @@ +import collections + +from ..packages import six +from ..packages.six.moves import queue + +if six.PY2: + # Queue is imported for side effects on MS Windows. See issue #229. + import Queue as _unused_module_Queue # noqa: F401 + + +class LifoQueue(queue.Queue): + def _init(self, _): + self.queue = collections.deque() + + def _qsize(self, len=len): + return len(self.queue) + + def _put(self, item): + self.queue.append(item) + + def _get(self): + return self.queue.pop() diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/response.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/response.py new file mode 100644 index 0000000000000000000000000000000000000000..5ea609ccedf18eb4ab70f8fc6990448eb6407237 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/response.py @@ -0,0 +1,107 @@ +from __future__ import absolute_import + +from email.errors import MultipartInvariantViolationDefect, StartBoundaryNotFoundDefect + +from ..exceptions import HeaderParsingError +from ..packages.six.moves import http_client as httplib + + +def is_fp_closed(obj): + """ + Checks whether a given file-like object is closed. + + :param obj: + The file-like object to check. + """ + + try: + # Check `isclosed()` first, in case Python3 doesn't set `closed`. + # GH Issue #928 + return obj.isclosed() + except AttributeError: + pass + + try: + # Check via the official file-like-object way. + return obj.closed + except AttributeError: + pass + + try: + # Check if the object is a container for another file-like object that + # gets released on exhaustion (e.g. HTTPResponse). + return obj.fp is None + except AttributeError: + pass + + raise ValueError("Unable to determine whether fp is closed.") + + +def assert_header_parsing(headers): + """ + Asserts whether all headers have been successfully parsed. + Extracts encountered errors from the result of parsing headers. + + Only works on Python 3. + + :param http.client.HTTPMessage headers: Headers to verify. + + :raises urllib3.exceptions.HeaderParsingError: + If parsing errors are found. + """ + + # This will fail silently if we pass in the wrong kind of parameter. + # To make debugging easier add an explicit check. + if not isinstance(headers, httplib.HTTPMessage): + raise TypeError("expected httplib.Message, got {0}.".format(type(headers))) + + defects = getattr(headers, "defects", None) + get_payload = getattr(headers, "get_payload", None) + + unparsed_data = None + if get_payload: + # get_payload is actually email.message.Message.get_payload; + # we're only interested in the result if it's not a multipart message + if not headers.is_multipart(): + payload = get_payload() + + if isinstance(payload, (bytes, str)): + unparsed_data = payload + if defects: + # httplib is assuming a response body is available + # when parsing headers even when httplib only sends + # header data to parse_headers() This results in + # defects on multipart responses in particular. + # See: https://github.com/urllib3/urllib3/issues/800 + + # So we ignore the following defects: + # - StartBoundaryNotFoundDefect: + # The claimed start boundary was never found. + # - MultipartInvariantViolationDefect: + # A message claimed to be a multipart but no subparts were found. + defects = [ + defect + for defect in defects + if not isinstance( + defect, (StartBoundaryNotFoundDefect, MultipartInvariantViolationDefect) + ) + ] + + if defects or unparsed_data: + raise HeaderParsingError(defects=defects, unparsed_data=unparsed_data) + + +def is_response_to_head(response): + """ + Checks whether the request of a response has been a HEAD-request. + Handles the quirks of AppEngine. + + :param http.client.HTTPResponse response: + Response to check if the originating request + used 'HEAD' as a method. + """ + # FIXME: Can we do this somehow without accessing private httplib _method? + method = response._method + if isinstance(method, int): # Platform-specific: Appengine + return method == 3 + return method.upper() == "HEAD" diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/retry.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/retry.py new file mode 100644 index 0000000000000000000000000000000000000000..9a1e90d0b236420d7f8b4c5c0325a7c17a1f3703 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/retry.py @@ -0,0 +1,622 @@ +from __future__ import absolute_import + +import email +import logging +import re +import time +import warnings +from collections import namedtuple +from itertools import takewhile + +from ..exceptions import ( + ConnectTimeoutError, + InvalidHeader, + MaxRetryError, + ProtocolError, + ProxyError, + ReadTimeoutError, + ResponseError, +) +from ..packages import six + +log = logging.getLogger(__name__) + + +# Data structure for representing the metadata of requests that result in a retry. +RequestHistory = namedtuple( + "RequestHistory", ["method", "url", "error", "status", "redirect_location"] +) + + +# TODO: In v2 we can remove this sentinel and metaclass with deprecated options. +_Default = object() + + +class _RetryMeta(type): + @property + def DEFAULT_METHOD_WHITELIST(cls): + warnings.warn( + "Using 'Retry.DEFAULT_METHOD_WHITELIST' is deprecated and " + "will be removed in v2.0. Use 'Retry.DEFAULT_ALLOWED_METHODS' instead", + DeprecationWarning, + ) + return cls.DEFAULT_ALLOWED_METHODS + + @DEFAULT_METHOD_WHITELIST.setter + def DEFAULT_METHOD_WHITELIST(cls, value): + warnings.warn( + "Using 'Retry.DEFAULT_METHOD_WHITELIST' is deprecated and " + "will be removed in v2.0. Use 'Retry.DEFAULT_ALLOWED_METHODS' instead", + DeprecationWarning, + ) + cls.DEFAULT_ALLOWED_METHODS = value + + @property + def DEFAULT_REDIRECT_HEADERS_BLACKLIST(cls): + warnings.warn( + "Using 'Retry.DEFAULT_REDIRECT_HEADERS_BLACKLIST' is deprecated and " + "will be removed in v2.0. Use 'Retry.DEFAULT_REMOVE_HEADERS_ON_REDIRECT' instead", + DeprecationWarning, + ) + return cls.DEFAULT_REMOVE_HEADERS_ON_REDIRECT + + @DEFAULT_REDIRECT_HEADERS_BLACKLIST.setter + def DEFAULT_REDIRECT_HEADERS_BLACKLIST(cls, value): + warnings.warn( + "Using 'Retry.DEFAULT_REDIRECT_HEADERS_BLACKLIST' is deprecated and " + "will be removed in v2.0. Use 'Retry.DEFAULT_REMOVE_HEADERS_ON_REDIRECT' instead", + DeprecationWarning, + ) + cls.DEFAULT_REMOVE_HEADERS_ON_REDIRECT = value + + @property + def BACKOFF_MAX(cls): + warnings.warn( + "Using 'Retry.BACKOFF_MAX' is deprecated and " + "will be removed in v2.0. Use 'Retry.DEFAULT_BACKOFF_MAX' instead", + DeprecationWarning, + ) + return cls.DEFAULT_BACKOFF_MAX + + @BACKOFF_MAX.setter + def BACKOFF_MAX(cls, value): + warnings.warn( + "Using 'Retry.BACKOFF_MAX' is deprecated and " + "will be removed in v2.0. Use 'Retry.DEFAULT_BACKOFF_MAX' instead", + DeprecationWarning, + ) + cls.DEFAULT_BACKOFF_MAX = value + + +@six.add_metaclass(_RetryMeta) +class Retry(object): + """Retry configuration. + + Each retry attempt will create a new Retry object with updated values, so + they can be safely reused. + + Retries can be defined as a default for a pool:: + + retries = Retry(connect=5, read=2, redirect=5) + http = PoolManager(retries=retries) + response = http.request('GET', 'http://example.com/') + + Or per-request (which overrides the default for the pool):: + + response = http.request('GET', 'http://example.com/', retries=Retry(10)) + + Retries can be disabled by passing ``False``:: + + response = http.request('GET', 'http://example.com/', retries=False) + + Errors will be wrapped in :class:`~urllib3.exceptions.MaxRetryError` unless + retries are disabled, in which case the causing exception will be raised. + + :param int total: + Total number of retries to allow. Takes precedence over other counts. + + Set to ``None`` to remove this constraint and fall back on other + counts. + + Set to ``0`` to fail on the first retry. + + Set to ``False`` to disable and imply ``raise_on_redirect=False``. + + :param int connect: + How many connection-related errors to retry on. + + These are errors raised before the request is sent to the remote server, + which we assume has not triggered the server to process the request. + + Set to ``0`` to fail on the first retry of this type. + + :param int read: + How many times to retry on read errors. + + These errors are raised after the request was sent to the server, so the + request may have side-effects. + + Set to ``0`` to fail on the first retry of this type. + + :param int redirect: + How many redirects to perform. Limit this to avoid infinite redirect + loops. + + A redirect is a HTTP response with a status code 301, 302, 303, 307 or + 308. + + Set to ``0`` to fail on the first retry of this type. + + Set to ``False`` to disable and imply ``raise_on_redirect=False``. + + :param int status: + How many times to retry on bad status codes. + + These are retries made on responses, where status code matches + ``status_forcelist``. + + Set to ``0`` to fail on the first retry of this type. + + :param int other: + How many times to retry on other errors. + + Other errors are errors that are not connect, read, redirect or status errors. + These errors might be raised after the request was sent to the server, so the + request might have side-effects. + + Set to ``0`` to fail on the first retry of this type. + + If ``total`` is not set, it's a good idea to set this to 0 to account + for unexpected edge cases and avoid infinite retry loops. + + :param iterable allowed_methods: + Set of uppercased HTTP method verbs that we should retry on. + + By default, we only retry on methods which are considered to be + idempotent (multiple requests with the same parameters end with the + same state). See :attr:`Retry.DEFAULT_ALLOWED_METHODS`. + + Set to a ``False`` value to retry on any verb. + + .. warning:: + + Previously this parameter was named ``method_whitelist``, that + usage is deprecated in v1.26.0 and will be removed in v2.0. + + :param iterable status_forcelist: + A set of integer HTTP status codes that we should force a retry on. + A retry is initiated if the request method is in ``allowed_methods`` + and the response status code is in ``status_forcelist``. + + By default, this is disabled with ``None``. + + :param float backoff_factor: + A backoff factor to apply between attempts after the second try + (most errors are resolved immediately by a second try without a + delay). urllib3 will sleep for:: + + {backoff factor} * (2 ** ({number of total retries} - 1)) + + seconds. If the backoff_factor is 0.1, then :func:`.sleep` will sleep + for [0.0s, 0.2s, 0.4s, ...] between retries. It will never be longer + than :attr:`Retry.DEFAULT_BACKOFF_MAX`. + + By default, backoff is disabled (set to 0). + + :param bool raise_on_redirect: Whether, if the number of redirects is + exhausted, to raise a MaxRetryError, or to return a response with a + response code in the 3xx range. + + :param bool raise_on_status: Similar meaning to ``raise_on_redirect``: + whether we should raise an exception, or return a response, + if status falls in ``status_forcelist`` range and retries have + been exhausted. + + :param tuple history: The history of the request encountered during + each call to :meth:`~Retry.increment`. The list is in the order + the requests occurred. Each list item is of class :class:`RequestHistory`. + + :param bool respect_retry_after_header: + Whether to respect Retry-After header on status codes defined as + :attr:`Retry.RETRY_AFTER_STATUS_CODES` or not. + + :param iterable remove_headers_on_redirect: + Sequence of headers to remove from the request when a response + indicating a redirect is returned before firing off the redirected + request. + """ + + #: Default methods to be used for ``allowed_methods`` + DEFAULT_ALLOWED_METHODS = frozenset( + ["HEAD", "GET", "PUT", "DELETE", "OPTIONS", "TRACE"] + ) + + #: Default status codes to be used for ``status_forcelist`` + RETRY_AFTER_STATUS_CODES = frozenset([413, 429, 503]) + + #: Default headers to be used for ``remove_headers_on_redirect`` + DEFAULT_REMOVE_HEADERS_ON_REDIRECT = frozenset( + ["Cookie", "Authorization", "Proxy-Authorization"] + ) + + #: Maximum backoff time. + DEFAULT_BACKOFF_MAX = 120 + + def __init__( + self, + total=10, + connect=None, + read=None, + redirect=None, + status=None, + other=None, + allowed_methods=_Default, + status_forcelist=None, + backoff_factor=0, + raise_on_redirect=True, + raise_on_status=True, + history=None, + respect_retry_after_header=True, + remove_headers_on_redirect=_Default, + # TODO: Deprecated, remove in v2.0 + method_whitelist=_Default, + ): + + if method_whitelist is not _Default: + if allowed_methods is not _Default: + raise ValueError( + "Using both 'allowed_methods' and " + "'method_whitelist' together is not allowed. " + "Instead only use 'allowed_methods'" + ) + warnings.warn( + "Using 'method_whitelist' with Retry is deprecated and " + "will be removed in v2.0. Use 'allowed_methods' instead", + DeprecationWarning, + stacklevel=2, + ) + allowed_methods = method_whitelist + if allowed_methods is _Default: + allowed_methods = self.DEFAULT_ALLOWED_METHODS + if remove_headers_on_redirect is _Default: + remove_headers_on_redirect = self.DEFAULT_REMOVE_HEADERS_ON_REDIRECT + + self.total = total + self.connect = connect + self.read = read + self.status = status + self.other = other + + if redirect is False or total is False: + redirect = 0 + raise_on_redirect = False + + self.redirect = redirect + self.status_forcelist = status_forcelist or set() + self.allowed_methods = allowed_methods + self.backoff_factor = backoff_factor + self.raise_on_redirect = raise_on_redirect + self.raise_on_status = raise_on_status + self.history = history or tuple() + self.respect_retry_after_header = respect_retry_after_header + self.remove_headers_on_redirect = frozenset( + [h.lower() for h in remove_headers_on_redirect] + ) + + def new(self, **kw): + params = dict( + total=self.total, + connect=self.connect, + read=self.read, + redirect=self.redirect, + status=self.status, + other=self.other, + status_forcelist=self.status_forcelist, + backoff_factor=self.backoff_factor, + raise_on_redirect=self.raise_on_redirect, + raise_on_status=self.raise_on_status, + history=self.history, + remove_headers_on_redirect=self.remove_headers_on_redirect, + respect_retry_after_header=self.respect_retry_after_header, + ) + + # TODO: If already given in **kw we use what's given to us + # If not given we need to figure out what to pass. We decide + # based on whether our class has the 'method_whitelist' property + # and if so we pass the deprecated 'method_whitelist' otherwise + # we use 'allowed_methods'. Remove in v2.0 + if "method_whitelist" not in kw and "allowed_methods" not in kw: + if "method_whitelist" in self.__dict__: + warnings.warn( + "Using 'method_whitelist' with Retry is deprecated and " + "will be removed in v2.0. Use 'allowed_methods' instead", + DeprecationWarning, + ) + params["method_whitelist"] = self.allowed_methods + else: + params["allowed_methods"] = self.allowed_methods + + params.update(kw) + return type(self)(**params) + + @classmethod + def from_int(cls, retries, redirect=True, default=None): + """Backwards-compatibility for the old retries format.""" + if retries is None: + retries = default if default is not None else cls.DEFAULT + + if isinstance(retries, Retry): + return retries + + redirect = bool(redirect) and None + new_retries = cls(retries, redirect=redirect) + log.debug("Converted retries value: %r -> %r", retries, new_retries) + return new_retries + + def get_backoff_time(self): + """Formula for computing the current backoff + + :rtype: float + """ + # We want to consider only the last consecutive errors sequence (Ignore redirects). + consecutive_errors_len = len( + list( + takewhile(lambda x: x.redirect_location is None, reversed(self.history)) + ) + ) + if consecutive_errors_len <= 1: + return 0 + + backoff_value = self.backoff_factor * (2 ** (consecutive_errors_len - 1)) + return min(self.DEFAULT_BACKOFF_MAX, backoff_value) + + def parse_retry_after(self, retry_after): + # Whitespace: https://tools.ietf.org/html/rfc7230#section-3.2.4 + if re.match(r"^\s*[0-9]+\s*$", retry_after): + seconds = int(retry_after) + else: + retry_date_tuple = email.utils.parsedate_tz(retry_after) + if retry_date_tuple is None: + raise InvalidHeader("Invalid Retry-After header: %s" % retry_after) + if retry_date_tuple[9] is None: # Python 2 + # Assume UTC if no timezone was specified + # On Python2.7, parsedate_tz returns None for a timezone offset + # instead of 0 if no timezone is given, where mktime_tz treats + # a None timezone offset as local time. + retry_date_tuple = retry_date_tuple[:9] + (0,) + retry_date_tuple[10:] + + retry_date = email.utils.mktime_tz(retry_date_tuple) + seconds = retry_date - time.time() + + if seconds < 0: + seconds = 0 + + return seconds + + def get_retry_after(self, response): + """Get the value of Retry-After in seconds.""" + + retry_after = response.headers.get("Retry-After") + + if retry_after is None: + return None + + return self.parse_retry_after(retry_after) + + def sleep_for_retry(self, response=None): + retry_after = self.get_retry_after(response) + if retry_after: + time.sleep(retry_after) + return True + + return False + + def _sleep_backoff(self): + backoff = self.get_backoff_time() + if backoff <= 0: + return + time.sleep(backoff) + + def sleep(self, response=None): + """Sleep between retry attempts. + + This method will respect a server's ``Retry-After`` response header + and sleep the duration of the time requested. If that is not present, it + will use an exponential backoff. By default, the backoff factor is 0 and + this method will return immediately. + """ + + if self.respect_retry_after_header and response: + slept = self.sleep_for_retry(response) + if slept: + return + + self._sleep_backoff() + + def _is_connection_error(self, err): + """Errors when we're fairly sure that the server did not receive the + request, so it should be safe to retry. + """ + if isinstance(err, ProxyError): + err = err.original_error + return isinstance(err, ConnectTimeoutError) + + def _is_read_error(self, err): + """Errors that occur after the request has been started, so we should + assume that the server began processing it. + """ + return isinstance(err, (ReadTimeoutError, ProtocolError)) + + def _is_method_retryable(self, method): + """Checks if a given HTTP method should be retried upon, depending if + it is included in the allowed_methods + """ + # TODO: For now favor if the Retry implementation sets its own method_whitelist + # property outside of our constructor to avoid breaking custom implementations. + if "method_whitelist" in self.__dict__: + warnings.warn( + "Using 'method_whitelist' with Retry is deprecated and " + "will be removed in v2.0. Use 'allowed_methods' instead", + DeprecationWarning, + ) + allowed_methods = self.method_whitelist + else: + allowed_methods = self.allowed_methods + + if allowed_methods and method.upper() not in allowed_methods: + return False + return True + + def is_retry(self, method, status_code, has_retry_after=False): + """Is this method/status code retryable? (Based on allowlists and control + variables such as the number of total retries to allow, whether to + respect the Retry-After header, whether this header is present, and + whether the returned status code is on the list of status codes to + be retried upon on the presence of the aforementioned header) + """ + if not self._is_method_retryable(method): + return False + + if self.status_forcelist and status_code in self.status_forcelist: + return True + + return ( + self.total + and self.respect_retry_after_header + and has_retry_after + and (status_code in self.RETRY_AFTER_STATUS_CODES) + ) + + def is_exhausted(self): + """Are we out of retries?""" + retry_counts = ( + self.total, + self.connect, + self.read, + self.redirect, + self.status, + self.other, + ) + retry_counts = list(filter(None, retry_counts)) + if not retry_counts: + return False + + return min(retry_counts) < 0 + + def increment( + self, + method=None, + url=None, + response=None, + error=None, + _pool=None, + _stacktrace=None, + ): + """Return a new Retry object with incremented retry counters. + + :param response: A response object, or None, if the server did not + return a response. + :type response: :class:`~urllib3.response.HTTPResponse` + :param Exception error: An error encountered during the request, or + None if the response was received successfully. + + :return: A new ``Retry`` object. + """ + if self.total is False and error: + # Disabled, indicate to re-raise the error. + raise six.reraise(type(error), error, _stacktrace) + + total = self.total + if total is not None: + total -= 1 + + connect = self.connect + read = self.read + redirect = self.redirect + status_count = self.status + other = self.other + cause = "unknown" + status = None + redirect_location = None + + if error and self._is_connection_error(error): + # Connect retry? + if connect is False: + raise six.reraise(type(error), error, _stacktrace) + elif connect is not None: + connect -= 1 + + elif error and self._is_read_error(error): + # Read retry? + if read is False or not self._is_method_retryable(method): + raise six.reraise(type(error), error, _stacktrace) + elif read is not None: + read -= 1 + + elif error: + # Other retry? + if other is not None: + other -= 1 + + elif response and response.get_redirect_location(): + # Redirect retry? + if redirect is not None: + redirect -= 1 + cause = "too many redirects" + redirect_location = response.get_redirect_location() + status = response.status + + else: + # Incrementing because of a server error like a 500 in + # status_forcelist and the given method is in the allowed_methods + cause = ResponseError.GENERIC_ERROR + if response and response.status: + if status_count is not None: + status_count -= 1 + cause = ResponseError.SPECIFIC_ERROR.format(status_code=response.status) + status = response.status + + history = self.history + ( + RequestHistory(method, url, error, status, redirect_location), + ) + + new_retry = self.new( + total=total, + connect=connect, + read=read, + redirect=redirect, + status=status_count, + other=other, + history=history, + ) + + if new_retry.is_exhausted(): + raise MaxRetryError(_pool, url, error or ResponseError(cause)) + + log.debug("Incremented Retry for (url='%s'): %r", url, new_retry) + + return new_retry + + def __repr__(self): + return ( + "{cls.__name__}(total={self.total}, connect={self.connect}, " + "read={self.read}, redirect={self.redirect}, status={self.status})" + ).format(cls=type(self), self=self) + + def __getattr__(self, item): + if item == "method_whitelist": + # TODO: Remove this deprecated alias in v2.0 + warnings.warn( + "Using 'method_whitelist' with Retry is deprecated and " + "will be removed in v2.0. Use 'allowed_methods' instead", + DeprecationWarning, + ) + return self.allowed_methods + try: + return getattr(super(Retry, self), item) + except AttributeError: + return getattr(Retry, item) + + +# For backwards compatibility (equivalent to pre-v1.9): +Retry.DEFAULT = Retry(3) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/ssl_.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/ssl_.py new file mode 100644 index 0000000000000000000000000000000000000000..0a6a0e06a0d4dbd2c918782f8eda310ba3feca12 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/ssl_.py @@ -0,0 +1,504 @@ +from __future__ import absolute_import + +import hashlib +import hmac +import os +import sys +import warnings +from binascii import hexlify, unhexlify + +from ..exceptions import ( + InsecurePlatformWarning, + ProxySchemeUnsupported, + SNIMissingWarning, + SSLError, +) +from ..packages import six +from .url import BRACELESS_IPV6_ADDRZ_RE, IPV4_RE + +SSLContext = None +SSLTransport = None +HAS_SNI = False +IS_PYOPENSSL = False +IS_SECURETRANSPORT = False +ALPN_PROTOCOLS = ["http/1.1"] + +# Maps the length of a digest to a possible hash function producing this digest +HASHFUNC_MAP = { + length: getattr(hashlib, algorithm, None) + for length, algorithm in ((32, "md5"), (40, "sha1"), (64, "sha256")) +} + + +def _const_compare_digest_backport(a, b): + """ + Compare two digests of equal length in constant time. + + The digests must be of type str/bytes. + Returns True if the digests match, and False otherwise. + """ + result = abs(len(a) - len(b)) + for left, right in zip(bytearray(a), bytearray(b)): + result |= left ^ right + return result == 0 + + +_const_compare_digest = getattr(hmac, "compare_digest", _const_compare_digest_backport) + +try: # Test for SSL features + import ssl + from ssl import CERT_REQUIRED, wrap_socket +except ImportError: + pass + +try: + from ssl import HAS_SNI # Has SNI? +except ImportError: + pass + +try: + from .ssltransport import SSLTransport +except ImportError: + pass + + +try: # Platform-specific: Python 3.6 + from ssl import PROTOCOL_TLS + + PROTOCOL_SSLv23 = PROTOCOL_TLS +except ImportError: + try: + from ssl import PROTOCOL_SSLv23 as PROTOCOL_TLS + + PROTOCOL_SSLv23 = PROTOCOL_TLS + except ImportError: + PROTOCOL_SSLv23 = PROTOCOL_TLS = 2 + +try: + from ssl import PROTOCOL_TLS_CLIENT +except ImportError: + PROTOCOL_TLS_CLIENT = PROTOCOL_TLS + + +try: + from ssl import OP_NO_COMPRESSION, OP_NO_SSLv2, OP_NO_SSLv3 +except ImportError: + OP_NO_SSLv2, OP_NO_SSLv3 = 0x1000000, 0x2000000 + OP_NO_COMPRESSION = 0x20000 + + +try: # OP_NO_TICKET was added in Python 3.6 + from ssl import OP_NO_TICKET +except ImportError: + OP_NO_TICKET = 0x4000 + + +# A secure default. +# Sources for more information on TLS ciphers: +# +# - https://wiki.mozilla.org/Security/Server_Side_TLS +# - https://www.ssllabs.com/projects/best-practices/index.html +# - https://hynek.me/articles/hardening-your-web-servers-ssl-ciphers/ +# +# The general intent is: +# - prefer cipher suites that offer perfect forward secrecy (DHE/ECDHE), +# - prefer ECDHE over DHE for better performance, +# - prefer any AES-GCM and ChaCha20 over any AES-CBC for better performance and +# security, +# - prefer AES-GCM over ChaCha20 because hardware-accelerated AES is common, +# - disable NULL authentication, MD5 MACs, DSS, and other +# insecure ciphers for security reasons. +# - NOTE: TLS 1.3 cipher suites are managed through a different interface +# not exposed by CPython (yet!) and are enabled by default if they're available. +DEFAULT_CIPHERS = ":".join( + [ + "ECDHE+AESGCM", + "ECDHE+CHACHA20", + "DHE+AESGCM", + "DHE+CHACHA20", + "ECDH+AESGCM", + "DH+AESGCM", + "ECDH+AES", + "DH+AES", + "RSA+AESGCM", + "RSA+AES", + "!aNULL", + "!eNULL", + "!MD5", + "!DSS", + ] +) + +try: + from ssl import SSLContext # Modern SSL? +except ImportError: + + class SSLContext(object): # Platform-specific: Python 2 + def __init__(self, protocol_version): + self.protocol = protocol_version + # Use default values from a real SSLContext + self.check_hostname = False + self.verify_mode = ssl.CERT_NONE + self.ca_certs = None + self.options = 0 + self.certfile = None + self.keyfile = None + self.ciphers = None + + def load_cert_chain(self, certfile, keyfile): + self.certfile = certfile + self.keyfile = keyfile + + def load_verify_locations(self, cafile=None, capath=None, cadata=None): + self.ca_certs = cafile + + if capath is not None: + raise SSLError("CA directories not supported in older Pythons") + + if cadata is not None: + raise SSLError("CA data not supported in older Pythons") + + def set_ciphers(self, cipher_suite): + self.ciphers = cipher_suite + + def wrap_socket(self, socket, server_hostname=None, server_side=False): + warnings.warn( + "A true SSLContext object is not available. This prevents " + "urllib3 from configuring SSL appropriately and may cause " + "certain SSL connections to fail. You can upgrade to a newer " + "version of Python to solve this. For more information, see " + "https://urllib3.readthedocs.io/en/1.26.x/advanced-usage.html" + "#ssl-warnings", + InsecurePlatformWarning, + ) + kwargs = { + "keyfile": self.keyfile, + "certfile": self.certfile, + "ca_certs": self.ca_certs, + "cert_reqs": self.verify_mode, + "ssl_version": self.protocol, + "server_side": server_side, + } + return wrap_socket(socket, ciphers=self.ciphers, **kwargs) + + +def assert_fingerprint(cert, fingerprint): + """ + Checks if given fingerprint matches the supplied certificate. + + :param cert: + Certificate as bytes object. + :param fingerprint: + Fingerprint as string of hexdigits, can be interspersed by colons. + """ + + fingerprint = fingerprint.replace(":", "").lower() + digest_length = len(fingerprint) + if digest_length not in HASHFUNC_MAP: + raise SSLError("Fingerprint of invalid length: {0}".format(fingerprint)) + hashfunc = HASHFUNC_MAP.get(digest_length) + if hashfunc is None: + raise SSLError( + "Hash function implementation unavailable for fingerprint length: {0}".format( + digest_length + ) + ) + + # We need encode() here for py32; works on py2 and p33. + fingerprint_bytes = unhexlify(fingerprint.encode()) + + cert_digest = hashfunc(cert).digest() + + if not _const_compare_digest(cert_digest, fingerprint_bytes): + raise SSLError( + 'Fingerprints did not match. Expected "{0}", got "{1}".'.format( + fingerprint, hexlify(cert_digest) + ) + ) + + +def resolve_cert_reqs(candidate): + """ + Resolves the argument to a numeric constant, which can be passed to + the wrap_socket function/method from the ssl module. + Defaults to :data:`ssl.CERT_REQUIRED`. + If given a string it is assumed to be the name of the constant in the + :mod:`ssl` module or its abbreviation. + (So you can specify `REQUIRED` instead of `CERT_REQUIRED`. + If it's neither `None` nor a string we assume it is already the numeric + constant which can directly be passed to wrap_socket. + """ + if candidate is None: + return CERT_REQUIRED + + if isinstance(candidate, str): + res = getattr(ssl, candidate, None) + if res is None: + res = getattr(ssl, "CERT_" + candidate) + return res + + return candidate + + +def resolve_ssl_version(candidate): + """ + like resolve_cert_reqs + """ + if candidate is None: + return PROTOCOL_TLS + + if isinstance(candidate, str): + res = getattr(ssl, candidate, None) + if res is None: + res = getattr(ssl, "PROTOCOL_" + candidate) + return res + + return candidate + + +def create_urllib3_context( + ssl_version=None, cert_reqs=None, options=None, ciphers=None +): + """All arguments have the same meaning as ``ssl_wrap_socket``. + + By default, this function does a lot of the same work that + ``ssl.create_default_context`` does on Python 3.4+. It: + + - Disables SSLv2, SSLv3, and compression + - Sets a restricted set of server ciphers + + If you wish to enable SSLv3, you can do:: + + from pip._vendor.urllib3.util import ssl_ + context = ssl_.create_urllib3_context() + context.options &= ~ssl_.OP_NO_SSLv3 + + You can do the same to enable compression (substituting ``COMPRESSION`` + for ``SSLv3`` in the last line above). + + :param ssl_version: + The desired protocol version to use. This will default to + PROTOCOL_SSLv23 which will negotiate the highest protocol that both + the server and your installation of OpenSSL support. + :param cert_reqs: + Whether to require the certificate verification. This defaults to + ``ssl.CERT_REQUIRED``. + :param options: + Specific OpenSSL options. These default to ``ssl.OP_NO_SSLv2``, + ``ssl.OP_NO_SSLv3``, ``ssl.OP_NO_COMPRESSION``, and ``ssl.OP_NO_TICKET``. + :param ciphers: + Which cipher suites to allow the server to select. + :returns: + Constructed SSLContext object with specified options + :rtype: SSLContext + """ + # PROTOCOL_TLS is deprecated in Python 3.10 + if not ssl_version or ssl_version == PROTOCOL_TLS: + ssl_version = PROTOCOL_TLS_CLIENT + + context = SSLContext(ssl_version) + + context.set_ciphers(ciphers or DEFAULT_CIPHERS) + + # Setting the default here, as we may have no ssl module on import + cert_reqs = ssl.CERT_REQUIRED if cert_reqs is None else cert_reqs + + if options is None: + options = 0 + # SSLv2 is easily broken and is considered harmful and dangerous + options |= OP_NO_SSLv2 + # SSLv3 has several problems and is now dangerous + options |= OP_NO_SSLv3 + # Disable compression to prevent CRIME attacks for OpenSSL 1.0+ + # (issue #309) + options |= OP_NO_COMPRESSION + # TLSv1.2 only. Unless set explicitly, do not request tickets. + # This may save some bandwidth on wire, and although the ticket is encrypted, + # there is a risk associated with it being on wire, + # if the server is not rotating its ticketing keys properly. + options |= OP_NO_TICKET + + context.options |= options + + # Enable post-handshake authentication for TLS 1.3, see GH #1634. PHA is + # necessary for conditional client cert authentication with TLS 1.3. + # The attribute is None for OpenSSL <= 1.1.0 or does not exist in older + # versions of Python. We only enable on Python 3.7.4+ or if certificate + # verification is enabled to work around Python issue #37428 + # See: https://bugs.python.org/issue37428 + if (cert_reqs == ssl.CERT_REQUIRED or sys.version_info >= (3, 7, 4)) and getattr( + context, "post_handshake_auth", None + ) is not None: + context.post_handshake_auth = True + + def disable_check_hostname(): + if ( + getattr(context, "check_hostname", None) is not None + ): # Platform-specific: Python 3.2 + # We do our own verification, including fingerprints and alternative + # hostnames. So disable it here + context.check_hostname = False + + # The order of the below lines setting verify_mode and check_hostname + # matter due to safe-guards SSLContext has to prevent an SSLContext with + # check_hostname=True, verify_mode=NONE/OPTIONAL. This is made even more + # complex because we don't know whether PROTOCOL_TLS_CLIENT will be used + # or not so we don't know the initial state of the freshly created SSLContext. + if cert_reqs == ssl.CERT_REQUIRED: + context.verify_mode = cert_reqs + disable_check_hostname() + else: + disable_check_hostname() + context.verify_mode = cert_reqs + + # Enable logging of TLS session keys via defacto standard environment variable + # 'SSLKEYLOGFILE', if the feature is available (Python 3.8+). Skip empty values. + if hasattr(context, "keylog_filename"): + sslkeylogfile = os.environ.get("SSLKEYLOGFILE") + if sslkeylogfile: + context.keylog_filename = sslkeylogfile + + return context + + +def ssl_wrap_socket( + sock, + keyfile=None, + certfile=None, + cert_reqs=None, + ca_certs=None, + server_hostname=None, + ssl_version=None, + ciphers=None, + ssl_context=None, + ca_cert_dir=None, + key_password=None, + ca_cert_data=None, + tls_in_tls=False, +): + """ + All arguments except for server_hostname, ssl_context, and ca_cert_dir have + the same meaning as they do when using :func:`ssl.wrap_socket`. + + :param server_hostname: + When SNI is supported, the expected hostname of the certificate + :param ssl_context: + A pre-made :class:`SSLContext` object. If none is provided, one will + be created using :func:`create_urllib3_context`. + :param ciphers: + A string of ciphers we wish the client to support. + :param ca_cert_dir: + A directory containing CA certificates in multiple separate files, as + supported by OpenSSL's -CApath flag or the capath argument to + SSLContext.load_verify_locations(). + :param key_password: + Optional password if the keyfile is encrypted. + :param ca_cert_data: + Optional string containing CA certificates in PEM format suitable for + passing as the cadata parameter to SSLContext.load_verify_locations() + :param tls_in_tls: + Use SSLTransport to wrap the existing socket. + """ + context = ssl_context + if context is None: + # Note: This branch of code and all the variables in it are no longer + # used by urllib3 itself. We should consider deprecating and removing + # this code. + context = create_urllib3_context(ssl_version, cert_reqs, ciphers=ciphers) + + if ca_certs or ca_cert_dir or ca_cert_data: + try: + context.load_verify_locations(ca_certs, ca_cert_dir, ca_cert_data) + except (IOError, OSError) as e: + raise SSLError(e) + + elif ssl_context is None and hasattr(context, "load_default_certs"): + # try to load OS default certs; works well on Windows (require Python3.4+) + context.load_default_certs() + + # Attempt to detect if we get the goofy behavior of the + # keyfile being encrypted and OpenSSL asking for the + # passphrase via the terminal and instead error out. + if keyfile and key_password is None and _is_key_file_encrypted(keyfile): + raise SSLError("Client private key is encrypted, password is required") + + if certfile: + if key_password is None: + context.load_cert_chain(certfile, keyfile) + else: + context.load_cert_chain(certfile, keyfile, key_password) + + try: + if hasattr(context, "set_alpn_protocols"): + context.set_alpn_protocols(ALPN_PROTOCOLS) + except NotImplementedError: # Defensive: in CI, we always have set_alpn_protocols + pass + + # If we detect server_hostname is an IP address then the SNI + # extension should not be used according to RFC3546 Section 3.1 + use_sni_hostname = server_hostname and not is_ipaddress(server_hostname) + # SecureTransport uses server_hostname in certificate verification. + send_sni = (use_sni_hostname and HAS_SNI) or ( + IS_SECURETRANSPORT and server_hostname + ) + # Do not warn the user if server_hostname is an invalid SNI hostname. + if not HAS_SNI and use_sni_hostname: + warnings.warn( + "An HTTPS request has been made, but the SNI (Server Name " + "Indication) extension to TLS is not available on this platform. " + "This may cause the server to present an incorrect TLS " + "certificate, which can cause validation failures. You can upgrade to " + "a newer version of Python to solve this. For more information, see " + "https://urllib3.readthedocs.io/en/1.26.x/advanced-usage.html" + "#ssl-warnings", + SNIMissingWarning, + ) + + if send_sni: + ssl_sock = _ssl_wrap_socket_impl( + sock, context, tls_in_tls, server_hostname=server_hostname + ) + else: + ssl_sock = _ssl_wrap_socket_impl(sock, context, tls_in_tls) + return ssl_sock + + +def is_ipaddress(hostname): + """Detects whether the hostname given is an IPv4 or IPv6 address. + Also detects IPv6 addresses with Zone IDs. + + :param str hostname: Hostname to examine. + :return: True if the hostname is an IP address, False otherwise. + """ + if not six.PY2 and isinstance(hostname, bytes): + # IDN A-label bytes are ASCII compatible. + hostname = hostname.decode("ascii") + return bool(IPV4_RE.match(hostname) or BRACELESS_IPV6_ADDRZ_RE.match(hostname)) + + +def _is_key_file_encrypted(key_file): + """Detects if a key file is encrypted or not.""" + with open(key_file, "r") as f: + for line in f: + # Look for Proc-Type: 4,ENCRYPTED + if "ENCRYPTED" in line: + return True + + return False + + +def _ssl_wrap_socket_impl(sock, ssl_context, tls_in_tls, server_hostname=None): + if tls_in_tls: + if not SSLTransport: + # Import error, ssl is not available. + raise ProxySchemeUnsupported( + "TLS in TLS requires support for the 'ssl' module" + ) + + SSLTransport._validate_ssl_context_for_tls_in_tls(ssl_context) + return SSLTransport(sock, ssl_context, server_hostname) + + if server_hostname: + return ssl_context.wrap_socket(sock, server_hostname=server_hostname) + else: + return ssl_context.wrap_socket(sock) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/url.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/url.py new file mode 100644 index 0000000000000000000000000000000000000000..a960b2f3c5f3d11fc9ae43638da9877d635e8d91 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pip/_vendor/urllib3/util/url.py @@ -0,0 +1,435 @@ +from __future__ import absolute_import + +import re +from collections import namedtuple + +from ..exceptions import LocationParseError +from ..packages import six + +url_attrs = ["scheme", "auth", "host", "port", "path", "query", "fragment"] + +# We only want to normalize urls with an HTTP(S) scheme. +# urllib3 infers URLs without a scheme (None) to be http. +NORMALIZABLE_SCHEMES = ("http", "https", None) + +# Almost all of these patterns were derived from the +# 'rfc3986' module: https://github.com/python-hyper/rfc3986 +PERCENT_RE = re.compile(r"%[a-fA-F0-9]{2}") +SCHEME_RE = re.compile(r"^(?:[a-zA-Z][a-zA-Z0-9+-]*:|/)") +URI_RE = re.compile( + r"^(?:([a-zA-Z][a-zA-Z0-9+.-]*):)?" + r"(?://([^\\/?#]*))?" + r"([^?#]*)" + r"(?:\?([^#]*))?" + r"(?:#(.*))?$", + re.UNICODE | re.DOTALL, +) + +IPV4_PAT = r"(?:[0-9]{1,3}\.){3}[0-9]{1,3}" +HEX_PAT = "[0-9A-Fa-f]{1,4}" +LS32_PAT = "(?:{hex}:{hex}|{ipv4})".format(hex=HEX_PAT, ipv4=IPV4_PAT) +_subs = {"hex": HEX_PAT, "ls32": LS32_PAT} +_variations = [ + # 6( h16 ":" ) ls32 + "(?:%(hex)s:){6}%(ls32)s", + # "::" 5( h16 ":" ) ls32 + "::(?:%(hex)s:){5}%(ls32)s", + # [ h16 ] "::" 4( h16 ":" ) ls32 + "(?:%(hex)s)?::(?:%(hex)s:){4}%(ls32)s", + # [ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32 + "(?:(?:%(hex)s:)?%(hex)s)?::(?:%(hex)s:){3}%(ls32)s", + # [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32 + "(?:(?:%(hex)s:){0,2}%(hex)s)?::(?:%(hex)s:){2}%(ls32)s", + # [ *3( h16 ":" ) h16 ] "::" h16 ":" ls32 + "(?:(?:%(hex)s:){0,3}%(hex)s)?::%(hex)s:%(ls32)s", + # [ *4( h16 ":" ) h16 ] "::" ls32 + "(?:(?:%(hex)s:){0,4}%(hex)s)?::%(ls32)s", + # [ *5( h16 ":" ) h16 ] "::" h16 + "(?:(?:%(hex)s:){0,5}%(hex)s)?::%(hex)s", + # [ *6( h16 ":" ) h16 ] "::" + "(?:(?:%(hex)s:){0,6}%(hex)s)?::", +] + +UNRESERVED_PAT = r"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789._\-~" +IPV6_PAT = "(?:" + "|".join([x % _subs for x in _variations]) + ")" +ZONE_ID_PAT = "(?:%25|%)(?:[" + UNRESERVED_PAT + "]|%[a-fA-F0-9]{2})+" +IPV6_ADDRZ_PAT = r"\[" + IPV6_PAT + r"(?:" + ZONE_ID_PAT + r")?\]" +REG_NAME_PAT = r"(?:[^\[\]%:/?#]|%[a-fA-F0-9]{2})*" +TARGET_RE = re.compile(r"^(/[^?#]*)(?:\?([^#]*))?(?:#.*)?$") + +IPV4_RE = re.compile("^" + IPV4_PAT + "$") +IPV6_RE = re.compile("^" + IPV6_PAT + "$") +IPV6_ADDRZ_RE = re.compile("^" + IPV6_ADDRZ_PAT + "$") +BRACELESS_IPV6_ADDRZ_RE = re.compile("^" + IPV6_ADDRZ_PAT[2:-2] + "$") +ZONE_ID_RE = re.compile("(" + ZONE_ID_PAT + r")\]$") + +_HOST_PORT_PAT = ("^(%s|%s|%s)(?::0*?(|0|[1-9][0-9]{0,4}))?$") % ( + REG_NAME_PAT, + IPV4_PAT, + IPV6_ADDRZ_PAT, +) +_HOST_PORT_RE = re.compile(_HOST_PORT_PAT, re.UNICODE | re.DOTALL) + +UNRESERVED_CHARS = set( + "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789._-~" +) +SUB_DELIM_CHARS = set("!$&'()*+,;=") +USERINFO_CHARS = UNRESERVED_CHARS | SUB_DELIM_CHARS | {":"} +PATH_CHARS = USERINFO_CHARS | {"@", "/"} +QUERY_CHARS = FRAGMENT_CHARS = PATH_CHARS | {"?"} + + +class Url(namedtuple("Url", url_attrs)): + """ + Data structure for representing an HTTP URL. Used as a return value for + :func:`parse_url`. Both the scheme and host are normalized as they are + both case-insensitive according to RFC 3986. + """ + + __slots__ = () + + def __new__( + cls, + scheme=None, + auth=None, + host=None, + port=None, + path=None, + query=None, + fragment=None, + ): + if path and not path.startswith("/"): + path = "/" + path + if scheme is not None: + scheme = scheme.lower() + return super(Url, cls).__new__( + cls, scheme, auth, host, port, path, query, fragment + ) + + @property + def hostname(self): + """For backwards-compatibility with urlparse. We're nice like that.""" + return self.host + + @property + def request_uri(self): + """Absolute path including the query string.""" + uri = self.path or "/" + + if self.query is not None: + uri += "?" + self.query + + return uri + + @property + def netloc(self): + """Network location including host and port""" + if self.port: + return "%s:%d" % (self.host, self.port) + return self.host + + @property + def url(self): + """ + Convert self into a url + + This function should more or less round-trip with :func:`.parse_url`. The + returned url may not be exactly the same as the url inputted to + :func:`.parse_url`, but it should be equivalent by the RFC (e.g., urls + with a blank port will have : removed). + + Example: :: + + >>> U = parse_url('http://google.com/mail/') + >>> U.url + 'http://google.com/mail/' + >>> Url('http', 'username:password', 'host.com', 80, + ... '/path', 'query', 'fragment').url + 'http://username:password@host.com:80/path?query#fragment' + """ + scheme, auth, host, port, path, query, fragment = self + url = u"" + + # We use "is not None" we want things to happen with empty strings (or 0 port) + if scheme is not None: + url += scheme + u"://" + if auth is not None: + url += auth + u"@" + if host is not None: + url += host + if port is not None: + url += u":" + str(port) + if path is not None: + url += path + if query is not None: + url += u"?" + query + if fragment is not None: + url += u"#" + fragment + + return url + + def __str__(self): + return self.url + + +def split_first(s, delims): + """ + .. deprecated:: 1.25 + + Given a string and an iterable of delimiters, split on the first found + delimiter. Return two split parts and the matched delimiter. + + If not found, then the first part is the full input string. + + Example:: + + >>> split_first('foo/bar?baz', '?/=') + ('foo', 'bar?baz', '/') + >>> split_first('foo/bar?baz', '123') + ('foo/bar?baz', '', None) + + Scales linearly with number of delims. Not ideal for large number of delims. + """ + min_idx = None + min_delim = None + for d in delims: + idx = s.find(d) + if idx < 0: + continue + + if min_idx is None or idx < min_idx: + min_idx = idx + min_delim = d + + if min_idx is None or min_idx < 0: + return s, "", None + + return s[:min_idx], s[min_idx + 1 :], min_delim + + +def _encode_invalid_chars(component, allowed_chars, encoding="utf-8"): + """Percent-encodes a URI component without reapplying + onto an already percent-encoded component. + """ + if component is None: + return component + + component = six.ensure_text(component) + + # Normalize existing percent-encoded bytes. + # Try to see if the component we're encoding is already percent-encoded + # so we can skip all '%' characters but still encode all others. + component, percent_encodings = PERCENT_RE.subn( + lambda match: match.group(0).upper(), component + ) + + uri_bytes = component.encode("utf-8", "surrogatepass") + is_percent_encoded = percent_encodings == uri_bytes.count(b"%") + encoded_component = bytearray() + + for i in range(0, len(uri_bytes)): + # Will return a single character bytestring on both Python 2 & 3 + byte = uri_bytes[i : i + 1] + byte_ord = ord(byte) + if (is_percent_encoded and byte == b"%") or ( + byte_ord < 128 and byte.decode() in allowed_chars + ): + encoded_component += byte + continue + encoded_component.extend(b"%" + (hex(byte_ord)[2:].encode().zfill(2).upper())) + + return encoded_component.decode(encoding) + + +def _remove_path_dot_segments(path): + # See http://tools.ietf.org/html/rfc3986#section-5.2.4 for pseudo-code + segments = path.split("/") # Turn the path into a list of segments + output = [] # Initialize the variable to use to store output + + for segment in segments: + # '.' is the current directory, so ignore it, it is superfluous + if segment == ".": + continue + # Anything other than '..', should be appended to the output + elif segment != "..": + output.append(segment) + # In this case segment == '..', if we can, we should pop the last + # element + elif output: + output.pop() + + # If the path starts with '/' and the output is empty or the first string + # is non-empty + if path.startswith("/") and (not output or output[0]): + output.insert(0, "") + + # If the path starts with '/.' or '/..' ensure we add one more empty + # string to add a trailing '/' + if path.endswith(("/.", "/..")): + output.append("") + + return "/".join(output) + + +def _normalize_host(host, scheme): + if host: + if isinstance(host, six.binary_type): + host = six.ensure_str(host) + + if scheme in NORMALIZABLE_SCHEMES: + is_ipv6 = IPV6_ADDRZ_RE.match(host) + if is_ipv6: + # IPv6 hosts of the form 'a::b%zone' are encoded in a URL as + # such per RFC 6874: 'a::b%25zone'. Unquote the ZoneID + # separator as necessary to return a valid RFC 4007 scoped IP. + match = ZONE_ID_RE.search(host) + if match: + start, end = match.span(1) + zone_id = host[start:end] + + if zone_id.startswith("%25") and zone_id != "%25": + zone_id = zone_id[3:] + else: + zone_id = zone_id[1:] + zone_id = "%" + _encode_invalid_chars(zone_id, UNRESERVED_CHARS) + return host[:start].lower() + zone_id + host[end:] + else: + return host.lower() + elif not IPV4_RE.match(host): + return six.ensure_str( + b".".join([_idna_encode(label) for label in host.split(".")]) + ) + return host + + +def _idna_encode(name): + if name and any(ord(x) >= 128 for x in name): + try: + from pip._vendor import idna + except ImportError: + six.raise_from( + LocationParseError("Unable to parse URL without the 'idna' module"), + None, + ) + try: + return idna.encode(name.lower(), strict=True, std3_rules=True) + except idna.IDNAError: + six.raise_from( + LocationParseError(u"Name '%s' is not a valid IDNA label" % name), None + ) + return name.lower().encode("ascii") + + +def _encode_target(target): + """Percent-encodes a request target so that there are no invalid characters""" + path, query = TARGET_RE.match(target).groups() + target = _encode_invalid_chars(path, PATH_CHARS) + query = _encode_invalid_chars(query, QUERY_CHARS) + if query is not None: + target += "?" + query + return target + + +def parse_url(url): + """ + Given a url, return a parsed :class:`.Url` namedtuple. Best-effort is + performed to parse incomplete urls. Fields not provided will be None. + This parser is RFC 3986 and RFC 6874 compliant. + + The parser logic and helper functions are based heavily on + work done in the ``rfc3986`` module. + + :param str url: URL to parse into a :class:`.Url` namedtuple. + + Partly backwards-compatible with :mod:`urlparse`. + + Example:: + + >>> parse_url('http://google.com/mail/') + Url(scheme='http', host='google.com', port=None, path='/mail/', ...) + >>> parse_url('google.com:80') + Url(scheme=None, host='google.com', port=80, path=None, ...) + >>> parse_url('/foo?bar') + Url(scheme=None, host=None, port=None, path='/foo', query='bar', ...) + """ + if not url: + # Empty + return Url() + + source_url = url + if not SCHEME_RE.search(url): + url = "//" + url + + try: + scheme, authority, path, query, fragment = URI_RE.match(url).groups() + normalize_uri = scheme is None or scheme.lower() in NORMALIZABLE_SCHEMES + + if scheme: + scheme = scheme.lower() + + if authority: + auth, _, host_port = authority.rpartition("@") + auth = auth or None + host, port = _HOST_PORT_RE.match(host_port).groups() + if auth and normalize_uri: + auth = _encode_invalid_chars(auth, USERINFO_CHARS) + if port == "": + port = None + else: + auth, host, port = None, None, None + + if port is not None: + port = int(port) + if not (0 <= port <= 65535): + raise LocationParseError(url) + + host = _normalize_host(host, scheme) + + if normalize_uri and path: + path = _remove_path_dot_segments(path) + path = _encode_invalid_chars(path, PATH_CHARS) + if normalize_uri and query: + query = _encode_invalid_chars(query, QUERY_CHARS) + if normalize_uri and fragment: + fragment = _encode_invalid_chars(fragment, FRAGMENT_CHARS) + + except (ValueError, AttributeError): + return six.raise_from(LocationParseError(source_url), None) + + # For the sake of backwards compatibility we put empty + # string values for path if there are any defined values + # beyond the path in the URL. + # TODO: Remove this when we break backwards compatibility. + if not path: + if query is not None or fragment is not None: + path = "" + else: + path = None + + # Ensure that each part of the URL is a `str` for + # backwards compatibility. + if isinstance(url, six.text_type): + ensure_func = six.ensure_text + else: + ensure_func = six.ensure_str + + def ensure_type(x): + return x if x is None else ensure_func(x) + + return Url( + scheme=ensure_type(scheme), + auth=ensure_type(auth), + host=ensure_type(host), + port=port, + path=ensure_type(path), + query=ensure_type(query), + fragment=ensure_type(fragment), + ) + + +def get_host(url): + """ + Deprecated. Use :func:`parse_url` instead. + """ + p = parse_url(url) + return p.scheme or "http", p.hostname, p.port diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/_version.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/_version.py new file mode 100644 index 0000000000000000000000000000000000000000..c298836bc33beb10a672a07bd027869c6a93ba1c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/_version.py @@ -0,0 +1,12 @@ +from __future__ import annotations + + +def _to_int(s: str) -> int | str: + try: + return int(s) + except ValueError: + return s + + +__version__ = "2.13.6" +version_info = tuple(_to_int(s) for s in __version__.split(".")) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/attr.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/attr.h new file mode 100644 index 0000000000000000000000000000000000000000..1044db94d906ac5fcf6faab6ac7668187314598f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/attr.h @@ -0,0 +1,690 @@ +/* + pybind11/attr.h: Infrastructure for processing custom + type and function attributes + + Copyright (c) 2016 Wenzel Jakob + + All rights reserved. Use of this source code is governed by a + BSD-style license that can be found in the LICENSE file. +*/ + +#pragma once + +#include "detail/common.h" +#include "cast.h" + +#include + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) + +/// \addtogroup annotations +/// @{ + +/// Annotation for methods +struct is_method { + handle class_; + explicit is_method(const handle &c) : class_(c) {} +}; + +/// Annotation for setters +struct is_setter {}; + +/// Annotation for operators +struct is_operator {}; + +/// Annotation for classes that cannot be subclassed +struct is_final {}; + +/// Annotation for parent scope +struct scope { + handle value; + explicit scope(const handle &s) : value(s) {} +}; + +/// Annotation for documentation +struct doc { + const char *value; + explicit doc(const char *value) : value(value) {} +}; + +/// Annotation for function names +struct name { + const char *value; + explicit name(const char *value) : value(value) {} +}; + +/// Annotation indicating that a function is an overload associated with a given "sibling" +struct sibling { + handle value; + explicit sibling(const handle &value) : value(value.ptr()) {} +}; + +/// Annotation indicating that a class derives from another given type +template +struct base { + + PYBIND11_DEPRECATED( + "base() was deprecated in favor of specifying 'T' as a template argument to class_") + base() = default; +}; + +/// Keep patient alive while nurse lives +template +struct keep_alive {}; + +/// Annotation indicating that a class is involved in a multiple inheritance relationship +struct multiple_inheritance {}; + +/// Annotation which enables dynamic attributes, i.e. adds `__dict__` to a class +struct dynamic_attr {}; + +/// Annotation which enables the buffer protocol for a type +struct buffer_protocol {}; + +/// Annotation which requests that a special metaclass is created for a type +struct metaclass { + handle value; + + PYBIND11_DEPRECATED("py::metaclass() is no longer required. It's turned on by default now.") + metaclass() = default; + + /// Override pybind11's default metaclass + explicit metaclass(handle value) : value(value) {} +}; + +/// Specifies a custom callback with signature `void (PyHeapTypeObject*)` that +/// may be used to customize the Python type. +/// +/// The callback is invoked immediately before `PyType_Ready`. +/// +/// Note: This is an advanced interface, and uses of it may require changes to +/// work with later versions of pybind11. You may wish to consult the +/// implementation of `make_new_python_type` in `detail/classes.h` to understand +/// the context in which the callback will be run. +struct custom_type_setup { + using callback = std::function; + + explicit custom_type_setup(callback value) : value(std::move(value)) {} + + callback value; +}; + +/// Annotation that marks a class as local to the module: +struct module_local { + const bool value; + constexpr explicit module_local(bool v = true) : value(v) {} +}; + +/// Annotation to mark enums as an arithmetic type +struct arithmetic {}; + +/// Mark a function for addition at the beginning of the existing overload chain instead of the end +struct prepend {}; + +/** \rst + A call policy which places one or more guard variables (``Ts...``) around the function call. + + For example, this definition: + + .. code-block:: cpp + + m.def("foo", foo, py::call_guard()); + + is equivalent to the following pseudocode: + + .. code-block:: cpp + + m.def("foo", [](args...) { + T scope_guard; + return foo(args...); // forwarded arguments + }); + \endrst */ +template +struct call_guard; + +template <> +struct call_guard<> { + using type = detail::void_type; +}; + +template +struct call_guard { + static_assert(std::is_default_constructible::value, + "The guard type must be default constructible"); + + using type = T; +}; + +template +struct call_guard { + struct type { + T guard{}; // Compose multiple guard types with left-to-right default-constructor order + typename call_guard::type next{}; + }; +}; + +/// @} annotations + +PYBIND11_NAMESPACE_BEGIN(detail) +/* Forward declarations */ +enum op_id : int; +enum op_type : int; +struct undefined_t; +template +struct op_; +void keep_alive_impl(size_t Nurse, size_t Patient, function_call &call, handle ret); + +/// Internal data structure which holds metadata about a keyword argument +struct argument_record { + const char *name; ///< Argument name + const char *descr; ///< Human-readable version of the argument value + handle value; ///< Associated Python object + bool convert : 1; ///< True if the argument is allowed to convert when loading + bool none : 1; ///< True if None is allowed when loading + + argument_record(const char *name, const char *descr, handle value, bool convert, bool none) + : name(name), descr(descr), value(value), convert(convert), none(none) {} +}; + +/// Internal data structure which holds metadata about a bound function (signature, overloads, +/// etc.) +struct function_record { + function_record() + : is_constructor(false), is_new_style_constructor(false), is_stateless(false), + is_operator(false), is_method(false), is_setter(false), has_args(false), + has_kwargs(false), prepend(false) {} + + /// Function name + char *name = nullptr; /* why no C++ strings? They generate heavier code.. */ + + // User-specified documentation string + char *doc = nullptr; + + /// Human-readable version of the function signature + char *signature = nullptr; + + /// List of registered keyword arguments + std::vector args; + + /// Pointer to lambda function which converts arguments and performs the actual call + handle (*impl)(function_call &) = nullptr; + + /// Storage for the wrapped function pointer and captured data, if any + void *data[3] = {}; + + /// Pointer to custom destructor for 'data' (if needed) + void (*free_data)(function_record *ptr) = nullptr; + + /// Return value policy associated with this function + return_value_policy policy = return_value_policy::automatic; + + /// True if name == '__init__' + bool is_constructor : 1; + + /// True if this is a new-style `__init__` defined in `detail/init.h` + bool is_new_style_constructor : 1; + + /// True if this is a stateless function pointer + bool is_stateless : 1; + + /// True if this is an operator (__add__), etc. + bool is_operator : 1; + + /// True if this is a method + bool is_method : 1; + + /// True if this is a setter + bool is_setter : 1; + + /// True if the function has a '*args' argument + bool has_args : 1; + + /// True if the function has a '**kwargs' argument + bool has_kwargs : 1; + + /// True if this function is to be inserted at the beginning of the overload resolution chain + bool prepend : 1; + + /// Number of arguments (including py::args and/or py::kwargs, if present) + std::uint16_t nargs; + + /// Number of leading positional arguments, which are terminated by a py::args or py::kwargs + /// argument or by a py::kw_only annotation. + std::uint16_t nargs_pos = 0; + + /// Number of leading arguments (counted in `nargs`) that are positional-only + std::uint16_t nargs_pos_only = 0; + + /// Python method object + PyMethodDef *def = nullptr; + + /// Python handle to the parent scope (a class or a module) + handle scope; + + /// Python handle to the sibling function representing an overload chain + handle sibling; + + /// Pointer to next overload + function_record *next = nullptr; +}; + +/// Special data structure which (temporarily) holds metadata about a bound class +struct type_record { + PYBIND11_NOINLINE type_record() + : multiple_inheritance(false), dynamic_attr(false), buffer_protocol(false), + default_holder(true), module_local(false), is_final(false) {} + + /// Handle to the parent scope + handle scope; + + /// Name of the class + const char *name = nullptr; + + // Pointer to RTTI type_info data structure + const std::type_info *type = nullptr; + + /// How large is the underlying C++ type? + size_t type_size = 0; + + /// What is the alignment of the underlying C++ type? + size_t type_align = 0; + + /// How large is the type's holder? + size_t holder_size = 0; + + /// The global operator new can be overridden with a class-specific variant + void *(*operator_new)(size_t) = nullptr; + + /// Function pointer to class_<..>::init_instance + void (*init_instance)(instance *, const void *) = nullptr; + + /// Function pointer to class_<..>::dealloc + void (*dealloc)(detail::value_and_holder &) = nullptr; + + /// List of base classes of the newly created type + list bases; + + /// Optional docstring + const char *doc = nullptr; + + /// Custom metaclass (optional) + handle metaclass; + + /// Custom type setup. + custom_type_setup::callback custom_type_setup_callback; + + /// Multiple inheritance marker + bool multiple_inheritance : 1; + + /// Does the class manage a __dict__? + bool dynamic_attr : 1; + + /// Does the class implement the buffer protocol? + bool buffer_protocol : 1; + + /// Is the default (unique_ptr) holder type used? + bool default_holder : 1; + + /// Is the class definition local to the module shared object? + bool module_local : 1; + + /// Is the class inheritable from python classes? + bool is_final : 1; + + PYBIND11_NOINLINE void add_base(const std::type_info &base, void *(*caster)(void *) ) { + auto *base_info = detail::get_type_info(base, false); + if (!base_info) { + std::string tname(base.name()); + detail::clean_type_id(tname); + pybind11_fail("generic_type: type \"" + std::string(name) + + "\" referenced unknown base type \"" + tname + "\""); + } + + if (default_holder != base_info->default_holder) { + std::string tname(base.name()); + detail::clean_type_id(tname); + pybind11_fail("generic_type: type \"" + std::string(name) + "\" " + + (default_holder ? "does not have" : "has") + + " a non-default holder type while its base \"" + tname + "\" " + + (base_info->default_holder ? "does not" : "does")); + } + + bases.append((PyObject *) base_info->type); + +#if PY_VERSION_HEX < 0x030B0000 + dynamic_attr |= base_info->type->tp_dictoffset != 0; +#else + dynamic_attr |= (base_info->type->tp_flags & Py_TPFLAGS_MANAGED_DICT) != 0; +#endif + + if (caster) { + base_info->implicit_casts.emplace_back(type, caster); + } + } +}; + +inline function_call::function_call(const function_record &f, handle p) : func(f), parent(p) { + args.reserve(f.nargs); + args_convert.reserve(f.nargs); +} + +/// Tag for a new-style `__init__` defined in `detail/init.h` +struct is_new_style_constructor {}; + +/** + * Partial template specializations to process custom attributes provided to + * cpp_function_ and class_. These are either used to initialize the respective + * fields in the type_record and function_record data structures or executed at + * runtime to deal with custom call policies (e.g. keep_alive). + */ +template +struct process_attribute; + +template +struct process_attribute_default { + /// Default implementation: do nothing + static void init(const T &, function_record *) {} + static void init(const T &, type_record *) {} + static void precall(function_call &) {} + static void postcall(function_call &, handle) {} +}; + +/// Process an attribute specifying the function's name +template <> +struct process_attribute : process_attribute_default { + static void init(const name &n, function_record *r) { r->name = const_cast(n.value); } +}; + +/// Process an attribute specifying the function's docstring +template <> +struct process_attribute : process_attribute_default { + static void init(const doc &n, function_record *r) { r->doc = const_cast(n.value); } +}; + +/// Process an attribute specifying the function's docstring (provided as a C-style string) +template <> +struct process_attribute : process_attribute_default { + static void init(const char *d, function_record *r) { r->doc = const_cast(d); } + static void init(const char *d, type_record *r) { r->doc = d; } +}; +template <> +struct process_attribute : process_attribute {}; + +/// Process an attribute indicating the function's return value policy +template <> +struct process_attribute : process_attribute_default { + static void init(const return_value_policy &p, function_record *r) { r->policy = p; } +}; + +/// Process an attribute which indicates that this is an overloaded function associated with a +/// given sibling +template <> +struct process_attribute : process_attribute_default { + static void init(const sibling &s, function_record *r) { r->sibling = s.value; } +}; + +/// Process an attribute which indicates that this function is a method +template <> +struct process_attribute : process_attribute_default { + static void init(const is_method &s, function_record *r) { + r->is_method = true; + r->scope = s.class_; + } +}; + +/// Process an attribute which indicates that this function is a setter +template <> +struct process_attribute : process_attribute_default { + static void init(const is_setter &, function_record *r) { r->is_setter = true; } +}; + +/// Process an attribute which indicates the parent scope of a method +template <> +struct process_attribute : process_attribute_default { + static void init(const scope &s, function_record *r) { r->scope = s.value; } +}; + +/// Process an attribute which indicates that this function is an operator +template <> +struct process_attribute : process_attribute_default { + static void init(const is_operator &, function_record *r) { r->is_operator = true; } +}; + +template <> +struct process_attribute + : process_attribute_default { + static void init(const is_new_style_constructor &, function_record *r) { + r->is_new_style_constructor = true; + } +}; + +inline void check_kw_only_arg(const arg &a, function_record *r) { + if (r->args.size() > r->nargs_pos && (!a.name || a.name[0] == '\0')) { + pybind11_fail("arg(): cannot specify an unnamed argument after a kw_only() annotation or " + "args() argument"); + } +} + +inline void append_self_arg_if_needed(function_record *r) { + if (r->is_method && r->args.empty()) { + r->args.emplace_back("self", nullptr, handle(), /*convert=*/true, /*none=*/false); + } +} + +/// Process a keyword argument attribute (*without* a default value) +template <> +struct process_attribute : process_attribute_default { + static void init(const arg &a, function_record *r) { + append_self_arg_if_needed(r); + r->args.emplace_back(a.name, nullptr, handle(), !a.flag_noconvert, a.flag_none); + + check_kw_only_arg(a, r); + } +}; + +/// Process a keyword argument attribute (*with* a default value) +template <> +struct process_attribute : process_attribute_default { + static void init(const arg_v &a, function_record *r) { + if (r->is_method && r->args.empty()) { + r->args.emplace_back( + "self", /*descr=*/nullptr, /*parent=*/handle(), /*convert=*/true, /*none=*/false); + } + + if (!a.value) { +#if defined(PYBIND11_DETAILED_ERROR_MESSAGES) + std::string descr("'"); + if (a.name) { + descr += std::string(a.name) + ": "; + } + descr += a.type + "'"; + if (r->is_method) { + if (r->name) { + descr += " in method '" + (std::string) str(r->scope) + "." + + (std::string) r->name + "'"; + } else { + descr += " in method of '" + (std::string) str(r->scope) + "'"; + } + } else if (r->name) { + descr += " in function '" + (std::string) r->name + "'"; + } + pybind11_fail("arg(): could not convert default argument " + descr + + " into a Python object (type not registered yet?)"); +#else + pybind11_fail("arg(): could not convert default argument " + "into a Python object (type not registered yet?). " + "#define PYBIND11_DETAILED_ERROR_MESSAGES or compile in debug mode for " + "more information."); +#endif + } + r->args.emplace_back(a.name, a.descr, a.value.inc_ref(), !a.flag_noconvert, a.flag_none); + + check_kw_only_arg(a, r); + } +}; + +/// Process a keyword-only-arguments-follow pseudo argument +template <> +struct process_attribute : process_attribute_default { + static void init(const kw_only &, function_record *r) { + append_self_arg_if_needed(r); + if (r->has_args && r->nargs_pos != static_cast(r->args.size())) { + pybind11_fail("Mismatched args() and kw_only(): they must occur at the same relative " + "argument location (or omit kw_only() entirely)"); + } + r->nargs_pos = static_cast(r->args.size()); + } +}; + +/// Process a positional-only-argument maker +template <> +struct process_attribute : process_attribute_default { + static void init(const pos_only &, function_record *r) { + append_self_arg_if_needed(r); + r->nargs_pos_only = static_cast(r->args.size()); + if (r->nargs_pos_only > r->nargs_pos) { + pybind11_fail("pos_only(): cannot follow a py::args() argument"); + } + // It also can't follow a kw_only, but a static_assert in pybind11.h checks that + } +}; + +/// Process a parent class attribute. Single inheritance only (class_ itself already guarantees +/// that) +template +struct process_attribute::value>> + : process_attribute_default { + static void init(const handle &h, type_record *r) { r->bases.append(h); } +}; + +/// Process a parent class attribute (deprecated, does not support multiple inheritance) +template +struct process_attribute> : process_attribute_default> { + static void init(const base &, type_record *r) { r->add_base(typeid(T), nullptr); } +}; + +/// Process a multiple inheritance attribute +template <> +struct process_attribute : process_attribute_default { + static void init(const multiple_inheritance &, type_record *r) { + r->multiple_inheritance = true; + } +}; + +template <> +struct process_attribute : process_attribute_default { + static void init(const dynamic_attr &, type_record *r) { r->dynamic_attr = true; } +}; + +template <> +struct process_attribute { + static void init(const custom_type_setup &value, type_record *r) { + r->custom_type_setup_callback = value.value; + } +}; + +template <> +struct process_attribute : process_attribute_default { + static void init(const is_final &, type_record *r) { r->is_final = true; } +}; + +template <> +struct process_attribute : process_attribute_default { + static void init(const buffer_protocol &, type_record *r) { r->buffer_protocol = true; } +}; + +template <> +struct process_attribute : process_attribute_default { + static void init(const metaclass &m, type_record *r) { r->metaclass = m.value; } +}; + +template <> +struct process_attribute : process_attribute_default { + static void init(const module_local &l, type_record *r) { r->module_local = l.value; } +}; + +/// Process a 'prepend' attribute, putting this at the beginning of the overload chain +template <> +struct process_attribute : process_attribute_default { + static void init(const prepend &, function_record *r) { r->prepend = true; } +}; + +/// Process an 'arithmetic' attribute for enums (does nothing here) +template <> +struct process_attribute : process_attribute_default {}; + +template +struct process_attribute> : process_attribute_default> {}; + +/** + * Process a keep_alive call policy -- invokes keep_alive_impl during the + * pre-call handler if both Nurse, Patient != 0 and use the post-call handler + * otherwise + */ +template +struct process_attribute> + : public process_attribute_default> { + template = 0> + static void precall(function_call &call) { + keep_alive_impl(Nurse, Patient, call, handle()); + } + template = 0> + static void postcall(function_call &, handle) {} + template = 0> + static void precall(function_call &) {} + template = 0> + static void postcall(function_call &call, handle ret) { + keep_alive_impl(Nurse, Patient, call, ret); + } +}; + +/// Recursively iterate over variadic template arguments +template +struct process_attributes { + static void init(const Args &...args, function_record *r) { + PYBIND11_WORKAROUND_INCORRECT_MSVC_C4100(r); + PYBIND11_WORKAROUND_INCORRECT_GCC_UNUSED_BUT_SET_PARAMETER(r); + using expander = int[]; + (void) expander{ + 0, ((void) process_attribute::type>::init(args, r), 0)...}; + } + static void init(const Args &...args, type_record *r) { + PYBIND11_WORKAROUND_INCORRECT_MSVC_C4100(r); + PYBIND11_WORKAROUND_INCORRECT_GCC_UNUSED_BUT_SET_PARAMETER(r); + using expander = int[]; + (void) expander{0, + (process_attribute::type>::init(args, r), 0)...}; + } + static void precall(function_call &call) { + PYBIND11_WORKAROUND_INCORRECT_MSVC_C4100(call); + using expander = int[]; + (void) expander{0, + (process_attribute::type>::precall(call), 0)...}; + } + static void postcall(function_call &call, handle fn_ret) { + PYBIND11_WORKAROUND_INCORRECT_MSVC_C4100(call, fn_ret); + PYBIND11_WORKAROUND_INCORRECT_GCC_UNUSED_BUT_SET_PARAMETER(fn_ret); + using expander = int[]; + (void) expander{ + 0, (process_attribute::type>::postcall(call, fn_ret), 0)...}; + } +}; + +template +using is_call_guard = is_instantiation; + +/// Extract the ``type`` from the first `call_guard` in `Extras...` (or `void_type` if none found) +template +using extract_guard_t = typename exactly_one_t, Extra...>::type; + +/// Check the number of named arguments at compile time +template ::value...), + size_t self = constexpr_sum(std::is_same::value...)> +constexpr bool expected_num_args(size_t nargs, bool has_args, bool has_kwargs) { + PYBIND11_WORKAROUND_INCORRECT_MSVC_C4100(nargs, has_args, has_kwargs); + return named == 0 || (self + named + size_t(has_args) + size_t(has_kwargs)) == nargs; +} + +PYBIND11_NAMESPACE_END(detail) +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/chrono.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/chrono.h new file mode 100644 index 0000000000000000000000000000000000000000..167ea0e3d1057c54a469c8ecd656d49eec495fe7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/chrono.h @@ -0,0 +1,225 @@ +/* + pybind11/chrono.h: Transparent conversion between std::chrono and python's datetime + + Copyright (c) 2016 Trent Houliston and + Wenzel Jakob + + All rights reserved. Use of this source code is governed by a + BSD-style license that can be found in the LICENSE file. +*/ + +#pragma once + +#include "pybind11.h" + +#include +#include +#include +#include +#include + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) +PYBIND11_NAMESPACE_BEGIN(detail) + +template +class duration_caster { +public: + using rep = typename type::rep; + using period = typename type::period; + + // signed 25 bits required by the standard. + using days = std::chrono::duration>; + + bool load(handle src, bool) { + using namespace std::chrono; + + // Lazy initialise the PyDateTime import + if (!PyDateTimeAPI) { + PyDateTime_IMPORT; + } + + if (!src) { + return false; + } + // If invoked with datetime.delta object + if (PyDelta_Check(src.ptr())) { + value = type(duration_cast>( + days(PyDateTime_DELTA_GET_DAYS(src.ptr())) + + seconds(PyDateTime_DELTA_GET_SECONDS(src.ptr())) + + microseconds(PyDateTime_DELTA_GET_MICROSECONDS(src.ptr())))); + return true; + } + // If invoked with a float we assume it is seconds and convert + if (PyFloat_Check(src.ptr())) { + value = type(duration_cast>( + duration(PyFloat_AsDouble(src.ptr())))); + return true; + } + return false; + } + + // If this is a duration just return it back + static const std::chrono::duration & + get_duration(const std::chrono::duration &src) { + return src; + } + + // If this is a time_point get the time_since_epoch + template + static std::chrono::duration + get_duration(const std::chrono::time_point> &src) { + return src.time_since_epoch(); + } + + static handle cast(const type &src, return_value_policy /* policy */, handle /* parent */) { + using namespace std::chrono; + + // Use overloaded function to get our duration from our source + // Works out if it is a duration or time_point and get the duration + auto d = get_duration(src); + + // Lazy initialise the PyDateTime import + if (!PyDateTimeAPI) { + PyDateTime_IMPORT; + } + + // Declare these special duration types so the conversions happen with the correct + // primitive types (int) + using dd_t = duration>; + using ss_t = duration>; + using us_t = duration; + + auto dd = duration_cast(d); + auto subd = d - dd; + auto ss = duration_cast(subd); + auto us = duration_cast(subd - ss); + return PyDelta_FromDSU(dd.count(), ss.count(), us.count()); + } + + PYBIND11_TYPE_CASTER(type, const_name("datetime.timedelta")); +}; + +inline std::tm *localtime_thread_safe(const std::time_t *time, std::tm *buf) { +#if (defined(__STDC_LIB_EXT1__) && defined(__STDC_WANT_LIB_EXT1__)) || defined(_MSC_VER) + if (localtime_s(buf, time)) + return nullptr; + return buf; +#else + static std::mutex mtx; + std::lock_guard lock(mtx); + std::tm *tm_ptr = std::localtime(time); + if (tm_ptr != nullptr) { + *buf = *tm_ptr; + } + return tm_ptr; +#endif +} + +// This is for casting times on the system clock into datetime.datetime instances +template +class type_caster> { +public: + using type = std::chrono::time_point; + bool load(handle src, bool) { + using namespace std::chrono; + + // Lazy initialise the PyDateTime import + if (!PyDateTimeAPI) { + PyDateTime_IMPORT; + } + + if (!src) { + return false; + } + + std::tm cal; + microseconds msecs; + + if (PyDateTime_Check(src.ptr())) { + cal.tm_sec = PyDateTime_DATE_GET_SECOND(src.ptr()); + cal.tm_min = PyDateTime_DATE_GET_MINUTE(src.ptr()); + cal.tm_hour = PyDateTime_DATE_GET_HOUR(src.ptr()); + cal.tm_mday = PyDateTime_GET_DAY(src.ptr()); + cal.tm_mon = PyDateTime_GET_MONTH(src.ptr()) - 1; + cal.tm_year = PyDateTime_GET_YEAR(src.ptr()) - 1900; + cal.tm_isdst = -1; + msecs = microseconds(PyDateTime_DATE_GET_MICROSECOND(src.ptr())); + } else if (PyDate_Check(src.ptr())) { + cal.tm_sec = 0; + cal.tm_min = 0; + cal.tm_hour = 0; + cal.tm_mday = PyDateTime_GET_DAY(src.ptr()); + cal.tm_mon = PyDateTime_GET_MONTH(src.ptr()) - 1; + cal.tm_year = PyDateTime_GET_YEAR(src.ptr()) - 1900; + cal.tm_isdst = -1; + msecs = microseconds(0); + } else if (PyTime_Check(src.ptr())) { + cal.tm_sec = PyDateTime_TIME_GET_SECOND(src.ptr()); + cal.tm_min = PyDateTime_TIME_GET_MINUTE(src.ptr()); + cal.tm_hour = PyDateTime_TIME_GET_HOUR(src.ptr()); + cal.tm_mday = 1; // This date (day, month, year) = (1, 0, 70) + cal.tm_mon = 0; // represents 1-Jan-1970, which is the first + cal.tm_year = 70; // earliest available date for Python's datetime + cal.tm_isdst = -1; + msecs = microseconds(PyDateTime_TIME_GET_MICROSECOND(src.ptr())); + } else { + return false; + } + + value = time_point_cast(system_clock::from_time_t(std::mktime(&cal)) + msecs); + return true; + } + + static handle cast(const std::chrono::time_point &src, + return_value_policy /* policy */, + handle /* parent */) { + using namespace std::chrono; + + // Lazy initialise the PyDateTime import + if (!PyDateTimeAPI) { + PyDateTime_IMPORT; + } + + // Get out microseconds, and make sure they are positive, to avoid bug in eastern + // hemisphere time zones (cfr. https://github.com/pybind/pybind11/issues/2417) + using us_t = duration; + auto us = duration_cast(src.time_since_epoch() % seconds(1)); + if (us.count() < 0) { + us += seconds(1); + } + + // Subtract microseconds BEFORE `system_clock::to_time_t`, because: + // > If std::time_t has lower precision, it is implementation-defined whether the value is + // rounded or truncated. (https://en.cppreference.com/w/cpp/chrono/system_clock/to_time_t) + std::time_t tt + = system_clock::to_time_t(time_point_cast(src - us)); + + std::tm localtime; + std::tm *localtime_ptr = localtime_thread_safe(&tt, &localtime); + if (!localtime_ptr) { + throw cast_error("Unable to represent system_clock in local time"); + } + return PyDateTime_FromDateAndTime(localtime.tm_year + 1900, + localtime.tm_mon + 1, + localtime.tm_mday, + localtime.tm_hour, + localtime.tm_min, + localtime.tm_sec, + us.count()); + } + PYBIND11_TYPE_CASTER(type, const_name("datetime.datetime")); +}; + +// Other clocks that are not the system clock are not measured as datetime.datetime objects +// since they are not measured on calendar time. So instead we just make them timedeltas +// Or if they have passed us a time as a float we convert that +template +class type_caster> + : public duration_caster> {}; + +template +class type_caster> + : public duration_caster> {}; + +PYBIND11_NAMESPACE_END(detail) +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/eval.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/eval.h new file mode 100644 index 0000000000000000000000000000000000000000..bd5f981f53b87dd8c378a20dcd8109a6b3172cfe --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/eval.h @@ -0,0 +1,156 @@ +/* + pybind11/eval.h: Support for evaluating Python expressions and statements + from strings and files + + Copyright (c) 2016 Klemens Morgenstern and + Wenzel Jakob + + All rights reserved. Use of this source code is governed by a + BSD-style license that can be found in the LICENSE file. +*/ + +#pragma once + +#include "pybind11.h" + +#include + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) +PYBIND11_NAMESPACE_BEGIN(detail) + +inline void ensure_builtins_in_globals(object &global) { +#if defined(PYPY_VERSION) || PY_VERSION_HEX < 0x03080000 + // Running exec and eval adds `builtins` module under `__builtins__` key to + // globals if not yet present. Python 3.8 made PyRun_String behave + // similarly. Let's also do that for older versions, for consistency. This + // was missing from PyPy3.8 7.3.7. + if (!global.contains("__builtins__")) + global["__builtins__"] = module_::import(PYBIND11_BUILTINS_MODULE); +#else + (void) global; +#endif +} + +PYBIND11_NAMESPACE_END(detail) + +enum eval_mode { + /// Evaluate a string containing an isolated expression + eval_expr, + + /// Evaluate a string containing a single statement. Returns \c none + eval_single_statement, + + /// Evaluate a string containing a sequence of statement. Returns \c none + eval_statements +}; + +template +object eval(const str &expr, object global = globals(), object local = object()) { + if (!local) { + local = global; + } + + detail::ensure_builtins_in_globals(global); + + /* PyRun_String does not accept a PyObject / encoding specifier, + this seems to be the only alternative */ + std::string buffer = "# -*- coding: utf-8 -*-\n" + (std::string) expr; + + int start = 0; + switch (mode) { + case eval_expr: + start = Py_eval_input; + break; + case eval_single_statement: + start = Py_single_input; + break; + case eval_statements: + start = Py_file_input; + break; + default: + pybind11_fail("invalid evaluation mode"); + } + + PyObject *result = PyRun_String(buffer.c_str(), start, global.ptr(), local.ptr()); + if (!result) { + throw error_already_set(); + } + return reinterpret_steal(result); +} + +template +object eval(const char (&s)[N], object global = globals(), object local = object()) { + /* Support raw string literals by removing common leading whitespace */ + auto expr = (s[0] == '\n') ? str(module_::import("textwrap").attr("dedent")(s)) : str(s); + return eval(expr, std::move(global), std::move(local)); +} + +inline void exec(const str &expr, object global = globals(), object local = object()) { + eval(expr, std::move(global), std::move(local)); +} + +template +void exec(const char (&s)[N], object global = globals(), object local = object()) { + eval(s, std::move(global), std::move(local)); +} + +#if defined(PYPY_VERSION) +template +object eval_file(str, object, object) { + pybind11_fail("eval_file not supported in PyPy3. Use eval"); +} +template +object eval_file(str, object) { + pybind11_fail("eval_file not supported in PyPy3. Use eval"); +} +template +object eval_file(str) { + pybind11_fail("eval_file not supported in PyPy3. Use eval"); +} +#else +template +object eval_file(str fname, object global = globals(), object local = object()) { + if (!local) { + local = global; + } + + detail::ensure_builtins_in_globals(global); + + int start = 0; + switch (mode) { + case eval_expr: + start = Py_eval_input; + break; + case eval_single_statement: + start = Py_single_input; + break; + case eval_statements: + start = Py_file_input; + break; + default: + pybind11_fail("invalid evaluation mode"); + } + + int closeFile = 1; + std::string fname_str = (std::string) fname; + FILE *f = _Py_fopen_obj(fname.ptr(), "r"); + if (!f) { + PyErr_Clear(); + pybind11_fail("File \"" + fname_str + "\" could not be opened!"); + } + + if (!global.contains("__file__")) { + global["__file__"] = std::move(fname); + } + + PyObject *result + = PyRun_FileEx(f, fname_str.c_str(), start, global.ptr(), local.ptr(), closeFile); + + if (!result) { + throw error_already_set(); + } + return reinterpret_steal(result); +} +#endif + +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/operators.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/operators.h new file mode 100644 index 0000000000000000000000000000000000000000..16a88ae171846844ddac28839e2150f11d287839 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/operators.h @@ -0,0 +1,202 @@ +/* + pybind11/operator.h: Metatemplates for operator overloading + + Copyright (c) 2016 Wenzel Jakob + + All rights reserved. Use of this source code is governed by a + BSD-style license that can be found in the LICENSE file. +*/ + +#pragma once + +#include "pybind11.h" + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) +PYBIND11_NAMESPACE_BEGIN(detail) + +/// Enumeration with all supported operator types +enum op_id : int { + op_add, + op_sub, + op_mul, + op_div, + op_mod, + op_divmod, + op_pow, + op_lshift, + op_rshift, + op_and, + op_xor, + op_or, + op_neg, + op_pos, + op_abs, + op_invert, + op_int, + op_long, + op_float, + op_str, + op_cmp, + op_gt, + op_ge, + op_lt, + op_le, + op_eq, + op_ne, + op_iadd, + op_isub, + op_imul, + op_idiv, + op_imod, + op_ilshift, + op_irshift, + op_iand, + op_ixor, + op_ior, + op_complex, + op_bool, + op_nonzero, + op_repr, + op_truediv, + op_itruediv, + op_hash +}; + +enum op_type : int { + op_l, /* base type on left */ + op_r, /* base type on right */ + op_u /* unary operator */ +}; + +struct self_t {}; +static const self_t self = self_t(); + +/// Type for an unused type slot +struct undefined_t {}; + +/// Don't warn about an unused variable +inline self_t __self() { return self; } + +/// base template of operator implementations +template +struct op_impl {}; + +/// Operator implementation generator +template +struct op_ { + static constexpr bool op_enable_if_hook = true; + template + void execute(Class &cl, const Extra &...extra) const { + using Base = typename Class::type; + using L_type = conditional_t::value, Base, L>; + using R_type = conditional_t::value, Base, R>; + using op = op_impl; + cl.def(op::name(), &op::execute, is_operator(), extra...); + } + template + void execute_cast(Class &cl, const Extra &...extra) const { + using Base = typename Class::type; + using L_type = conditional_t::value, Base, L>; + using R_type = conditional_t::value, Base, R>; + using op = op_impl; + cl.def(op::name(), &op::execute_cast, is_operator(), extra...); + } +}; + +#define PYBIND11_BINARY_OPERATOR(id, rid, op, expr) \ + template \ + struct op_impl { \ + static char const *name() { return "__" #id "__"; } \ + static auto execute(const L &l, const R &r) -> decltype(expr) { return (expr); } \ + static B execute_cast(const L &l, const R &r) { return B(expr); } \ + }; \ + template \ + struct op_impl { \ + static char const *name() { return "__" #rid "__"; } \ + static auto execute(const R &r, const L &l) -> decltype(expr) { return (expr); } \ + static B execute_cast(const R &r, const L &l) { return B(expr); } \ + }; \ + inline op_ op(const self_t &, const self_t &) { \ + return op_(); \ + } \ + template \ + op_ op(const self_t &, const T &) { \ + return op_(); \ + } \ + template \ + op_ op(const T &, const self_t &) { \ + return op_(); \ + } + +#define PYBIND11_INPLACE_OPERATOR(id, op, expr) \ + template \ + struct op_impl { \ + static char const *name() { return "__" #id "__"; } \ + static auto execute(L &l, const R &r) -> decltype(expr) { return expr; } \ + static B execute_cast(L &l, const R &r) { return B(expr); } \ + }; \ + template \ + op_ op(const self_t &, const T &) { \ + return op_(); \ + } + +#define PYBIND11_UNARY_OPERATOR(id, op, expr) \ + template \ + struct op_impl { \ + static char const *name() { return "__" #id "__"; } \ + static auto execute(const L &l) -> decltype(expr) { return expr; } \ + static B execute_cast(const L &l) { return B(expr); } \ + }; \ + inline op_ op(const self_t &) { \ + return op_(); \ + } + +PYBIND11_BINARY_OPERATOR(sub, rsub, operator-, l - r) +PYBIND11_BINARY_OPERATOR(add, radd, operator+, l + r) +PYBIND11_BINARY_OPERATOR(mul, rmul, operator*, l *r) +PYBIND11_BINARY_OPERATOR(truediv, rtruediv, operator/, l / r) +PYBIND11_BINARY_OPERATOR(mod, rmod, operator%, l % r) +PYBIND11_BINARY_OPERATOR(lshift, rlshift, operator<<, l << r) +PYBIND11_BINARY_OPERATOR(rshift, rrshift, operator>>, l >> r) +PYBIND11_BINARY_OPERATOR(and, rand, operator&, l &r) +PYBIND11_BINARY_OPERATOR(xor, rxor, operator^, l ^ r) +PYBIND11_BINARY_OPERATOR(eq, eq, operator==, l == r) +PYBIND11_BINARY_OPERATOR(ne, ne, operator!=, l != r) +PYBIND11_BINARY_OPERATOR(or, ror, operator|, l | r) +PYBIND11_BINARY_OPERATOR(gt, lt, operator>, l > r) +PYBIND11_BINARY_OPERATOR(ge, le, operator>=, l >= r) +PYBIND11_BINARY_OPERATOR(lt, gt, operator<, l < r) +PYBIND11_BINARY_OPERATOR(le, ge, operator<=, l <= r) +// PYBIND11_BINARY_OPERATOR(pow, rpow, pow, std::pow(l, r)) +PYBIND11_INPLACE_OPERATOR(iadd, operator+=, l += r) +PYBIND11_INPLACE_OPERATOR(isub, operator-=, l -= r) +PYBIND11_INPLACE_OPERATOR(imul, operator*=, l *= r) +PYBIND11_INPLACE_OPERATOR(itruediv, operator/=, l /= r) +PYBIND11_INPLACE_OPERATOR(imod, operator%=, l %= r) +PYBIND11_INPLACE_OPERATOR(ilshift, operator<<=, l <<= r) +PYBIND11_INPLACE_OPERATOR(irshift, operator>>=, l >>= r) +PYBIND11_INPLACE_OPERATOR(iand, operator&=, l &= r) +PYBIND11_INPLACE_OPERATOR(ixor, operator^=, l ^= r) +PYBIND11_INPLACE_OPERATOR(ior, operator|=, l |= r) +PYBIND11_UNARY_OPERATOR(neg, operator-, -l) +PYBIND11_UNARY_OPERATOR(pos, operator+, +l) +// WARNING: This usage of `abs` should only be done for existing STL overloads. +// Adding overloads directly in to the `std::` namespace is advised against: +// https://en.cppreference.com/w/cpp/language/extending_std +PYBIND11_UNARY_OPERATOR(abs, abs, std::abs(l)) +PYBIND11_UNARY_OPERATOR(hash, hash, std::hash()(l)) +PYBIND11_UNARY_OPERATOR(invert, operator~, (~l)) +PYBIND11_UNARY_OPERATOR(bool, operator!, !!l) +PYBIND11_UNARY_OPERATOR(int, int_, (int) l) +PYBIND11_UNARY_OPERATOR(float, float_, (double) l) + +#undef PYBIND11_BINARY_OPERATOR +#undef PYBIND11_INPLACE_OPERATOR +#undef PYBIND11_UNARY_OPERATOR +PYBIND11_NAMESPACE_END(detail) + +using detail::self; +// Add named operators so that they are accessible via `py::`. +using detail::hash; + +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/options.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/options.h new file mode 100644 index 0000000000000000000000000000000000000000..1b2122522d5b4043c9df57f2beb943682eb4b208 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/options.h @@ -0,0 +1,92 @@ +/* + pybind11/options.h: global settings that are configurable at runtime. + + Copyright (c) 2016 Wenzel Jakob + + All rights reserved. Use of this source code is governed by a + BSD-style license that can be found in the LICENSE file. +*/ + +#pragma once + +#include "detail/common.h" + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) + +class options { +public: + // Default RAII constructor, which leaves settings as they currently are. + options() : previous_state(global_state()) {} + + // Class is non-copyable. + options(const options &) = delete; + options &operator=(const options &) = delete; + + // Destructor, which restores settings that were in effect before. + ~options() { global_state() = previous_state; } + + // Setter methods (affect the global state): + + options &disable_user_defined_docstrings() & { + global_state().show_user_defined_docstrings = false; + return *this; + } + + options &enable_user_defined_docstrings() & { + global_state().show_user_defined_docstrings = true; + return *this; + } + + options &disable_function_signatures() & { + global_state().show_function_signatures = false; + return *this; + } + + options &enable_function_signatures() & { + global_state().show_function_signatures = true; + return *this; + } + + options &disable_enum_members_docstring() & { + global_state().show_enum_members_docstring = false; + return *this; + } + + options &enable_enum_members_docstring() & { + global_state().show_enum_members_docstring = true; + return *this; + } + + // Getter methods (return the global state): + + static bool show_user_defined_docstrings() { + return global_state().show_user_defined_docstrings; + } + + static bool show_function_signatures() { return global_state().show_function_signatures; } + + static bool show_enum_members_docstring() { + return global_state().show_enum_members_docstring; + } + + // This type is not meant to be allocated on the heap. + void *operator new(size_t) = delete; + +private: + struct state { + bool show_user_defined_docstrings = true; //< Include user-supplied texts in docstrings. + bool show_function_signatures = true; //< Include auto-generated function signatures + // in docstrings. + bool show_enum_members_docstring = true; //< Include auto-generated member list in enum + // docstrings. + }; + + static state &global_state() { + static state instance; + return instance; + } + + state previous_state; +}; + +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/stl_bind.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/stl_bind.h new file mode 100644 index 0000000000000000000000000000000000000000..fcb48dea3396435fb6321a376d4cbfc099f8542a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/stl_bind.h @@ -0,0 +1,822 @@ +/* + pybind11/std_bind.h: Binding generators for STL data types + + Copyright (c) 2016 Sergey Lyskov and Wenzel Jakob + + All rights reserved. Use of this source code is governed by a + BSD-style license that can be found in the LICENSE file. +*/ + +#pragma once + +#include "detail/common.h" +#include "detail/type_caster_base.h" +#include "cast.h" +#include "operators.h" + +#include +#include +#include + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) +PYBIND11_NAMESPACE_BEGIN(detail) + +/* SFINAE helper class used by 'is_comparable */ +template +struct container_traits { + template + static std::true_type + test_comparable(decltype(std::declval() == std::declval()) *); + template + static std::false_type test_comparable(...); + template + static std::true_type test_value(typename T2::value_type *); + template + static std::false_type test_value(...); + template + static std::true_type test_pair(typename T2::first_type *, typename T2::second_type *); + template + static std::false_type test_pair(...); + + static constexpr const bool is_comparable + = std::is_same(nullptr))>::value; + static constexpr const bool is_pair + = std::is_same(nullptr, nullptr))>::value; + static constexpr const bool is_vector + = std::is_same(nullptr))>::value; + static constexpr const bool is_element = !is_pair && !is_vector; +}; + +/* Default: is_comparable -> std::false_type */ +template +struct is_comparable : std::false_type {}; + +/* For non-map data structures, check whether operator== can be instantiated */ +template +struct is_comparable< + T, + enable_if_t::is_element && container_traits::is_comparable>> + : std::true_type {}; + +/* For a vector/map data structure, recursively check the value type + (which is std::pair for maps) */ +template +struct is_comparable::is_vector>> + : is_comparable::type_to_check_recursively> {}; + +template <> +struct is_comparable : std::true_type {}; + +/* For pairs, recursively check the two data types */ +template +struct is_comparable::is_pair>> { + static constexpr const bool value = is_comparable::value + && is_comparable::value; +}; + +/* Fallback functions */ +template +void vector_if_copy_constructible(const Args &...) {} +template +void vector_if_equal_operator(const Args &...) {} +template +void vector_if_insertion_operator(const Args &...) {} +template +void vector_modifiers(const Args &...) {} + +template +void vector_if_copy_constructible(enable_if_t::value, Class_> &cl) { + cl.def(init(), "Copy constructor"); +} + +template +void vector_if_equal_operator(enable_if_t::value, Class_> &cl) { + using T = typename Vector::value_type; + + cl.def(self == self); + cl.def(self != self); + + cl.def( + "count", + [](const Vector &v, const T &x) { return std::count(v.begin(), v.end(), x); }, + arg("x"), + "Return the number of times ``x`` appears in the list"); + + cl.def( + "remove", + [](Vector &v, const T &x) { + auto p = std::find(v.begin(), v.end(), x); + if (p != v.end()) { + v.erase(p); + } else { + throw value_error(); + } + }, + arg("x"), + "Remove the first item from the list whose value is x. " + "It is an error if there is no such item."); + + cl.def( + "__contains__", + [](const Vector &v, const T &x) { return std::find(v.begin(), v.end(), x) != v.end(); }, + arg("x"), + "Return true the container contains ``x``"); +} + +// Vector modifiers -- requires a copyable vector_type: +// (Technically, some of these (pop and __delitem__) don't actually require copyability, but it +// seems silly to allow deletion but not insertion, so include them here too.) +template +void vector_modifiers( + enable_if_t::value, Class_> &cl) { + using T = typename Vector::value_type; + using SizeType = typename Vector::size_type; + using DiffType = typename Vector::difference_type; + + auto wrap_i = [](DiffType i, SizeType n) { + if (i < 0) { + i += n; + } + if (i < 0 || (SizeType) i >= n) { + throw index_error(); + } + return i; + }; + + cl.def( + "append", + [](Vector &v, const T &value) { v.push_back(value); }, + arg("x"), + "Add an item to the end of the list"); + + cl.def(init([](const iterable &it) { + auto v = std::unique_ptr(new Vector()); + v->reserve(len_hint(it)); + for (handle h : it) { + v->push_back(h.cast()); + } + return v.release(); + })); + + cl.def("clear", [](Vector &v) { v.clear(); }, "Clear the contents"); + + cl.def( + "extend", + [](Vector &v, const Vector &src) { v.insert(v.end(), src.begin(), src.end()); }, + arg("L"), + "Extend the list by appending all the items in the given list"); + + cl.def( + "extend", + [](Vector &v, const iterable &it) { + const size_t old_size = v.size(); + v.reserve(old_size + len_hint(it)); + try { + for (handle h : it) { + v.push_back(h.cast()); + } + } catch (const cast_error &) { + v.erase(v.begin() + static_cast(old_size), + v.end()); + try { + v.shrink_to_fit(); + } catch (const std::exception &) { // NOLINT(bugprone-empty-catch) + // Do nothing + } + throw; + } + }, + arg("L"), + "Extend the list by appending all the items in the given list"); + + cl.def( + "insert", + [](Vector &v, DiffType i, const T &x) { + // Can't use wrap_i; i == v.size() is OK + if (i < 0) { + i += v.size(); + } + if (i < 0 || (SizeType) i > v.size()) { + throw index_error(); + } + v.insert(v.begin() + i, x); + }, + arg("i"), + arg("x"), + "Insert an item at a given position."); + + cl.def( + "pop", + [](Vector &v) { + if (v.empty()) { + throw index_error(); + } + T t = std::move(v.back()); + v.pop_back(); + return t; + }, + "Remove and return the last item"); + + cl.def( + "pop", + [wrap_i](Vector &v, DiffType i) { + i = wrap_i(i, v.size()); + T t = std::move(v[(SizeType) i]); + v.erase(std::next(v.begin(), i)); + return t; + }, + arg("i"), + "Remove and return the item at index ``i``"); + + cl.def("__setitem__", [wrap_i](Vector &v, DiffType i, const T &t) { + i = wrap_i(i, v.size()); + v[(SizeType) i] = t; + }); + + /// Slicing protocol + cl.def( + "__getitem__", + [](const Vector &v, const slice &slice) -> Vector * { + size_t start = 0, stop = 0, step = 0, slicelength = 0; + + if (!slice.compute(v.size(), &start, &stop, &step, &slicelength)) { + throw error_already_set(); + } + + auto *seq = new Vector(); + seq->reserve((size_t) slicelength); + + for (size_t i = 0; i < slicelength; ++i) { + seq->push_back(v[start]); + start += step; + } + return seq; + }, + arg("s"), + "Retrieve list elements using a slice object"); + + cl.def( + "__setitem__", + [](Vector &v, const slice &slice, const Vector &value) { + size_t start = 0, stop = 0, step = 0, slicelength = 0; + if (!slice.compute(v.size(), &start, &stop, &step, &slicelength)) { + throw error_already_set(); + } + + if (slicelength != value.size()) { + throw std::runtime_error( + "Left and right hand size of slice assignment have different sizes!"); + } + + for (size_t i = 0; i < slicelength; ++i) { + v[start] = value[i]; + start += step; + } + }, + "Assign list elements using a slice object"); + + cl.def( + "__delitem__", + [wrap_i](Vector &v, DiffType i) { + i = wrap_i(i, v.size()); + v.erase(v.begin() + i); + }, + "Delete the list elements at index ``i``"); + + cl.def( + "__delitem__", + [](Vector &v, const slice &slice) { + size_t start = 0, stop = 0, step = 0, slicelength = 0; + + if (!slice.compute(v.size(), &start, &stop, &step, &slicelength)) { + throw error_already_set(); + } + + if (step == 1 && false) { + v.erase(v.begin() + (DiffType) start, v.begin() + DiffType(start + slicelength)); + } else { + for (size_t i = 0; i < slicelength; ++i) { + v.erase(v.begin() + DiffType(start)); + start += step - 1; + } + } + }, + "Delete list elements using a slice object"); +} + +// If the type has an operator[] that doesn't return a reference (most notably std::vector), +// we have to access by copying; otherwise we return by reference. +template +using vector_needs_copy + = negation()[typename Vector::size_type()]), + typename Vector::value_type &>>; + +// The usual case: access and iterate by reference +template +void vector_accessor(enable_if_t::value, Class_> &cl) { + using T = typename Vector::value_type; + using SizeType = typename Vector::size_type; + using DiffType = typename Vector::difference_type; + using ItType = typename Vector::iterator; + + auto wrap_i = [](DiffType i, SizeType n) { + if (i < 0) { + i += n; + } + if (i < 0 || (SizeType) i >= n) { + throw index_error(); + } + return i; + }; + + cl.def( + "__getitem__", + [wrap_i](Vector &v, DiffType i) -> T & { + i = wrap_i(i, v.size()); + return v[(SizeType) i]; + }, + return_value_policy::reference_internal // ref + keepalive + ); + + cl.def( + "__iter__", + [](Vector &v) { + return make_iterator( + v.begin(), v.end()); + }, + keep_alive<0, 1>() /* Essential: keep list alive while iterator exists */ + ); +} + +// The case for special objects, like std::vector, that have to be returned-by-copy: +template +void vector_accessor(enable_if_t::value, Class_> &cl) { + using T = typename Vector::value_type; + using SizeType = typename Vector::size_type; + using DiffType = typename Vector::difference_type; + using ItType = typename Vector::iterator; + cl.def("__getitem__", [](const Vector &v, DiffType i) -> T { + if (i < 0) { + i += v.size(); + if (i < 0) { + throw index_error(); + } + } + auto i_st = static_cast(i); + if (i_st >= v.size()) { + throw index_error(); + } + return v[i_st]; + }); + + cl.def( + "__iter__", + [](Vector &v) { + return make_iterator(v.begin(), v.end()); + }, + keep_alive<0, 1>() /* Essential: keep list alive while iterator exists */ + ); +} + +template +auto vector_if_insertion_operator(Class_ &cl, std::string const &name) + -> decltype(std::declval() << std::declval(), + void()) { + using size_type = typename Vector::size_type; + + cl.def( + "__repr__", + [name](Vector &v) { + std::ostringstream s; + s << name << '['; + for (size_type i = 0; i < v.size(); ++i) { + s << v[i]; + if (i != v.size() - 1) { + s << ", "; + } + } + s << ']'; + return s.str(); + }, + "Return the canonical string representation of this list."); +} + +// Provide the buffer interface for vectors if we have data() and we have a format for it +// GCC seems to have "void std::vector::data()" - doing SFINAE on the existence of data() +// is insufficient, we need to check it returns an appropriate pointer +template +struct vector_has_data_and_format : std::false_type {}; +template +struct vector_has_data_and_format< + Vector, + enable_if_t::format(), + std::declval().data()), + typename Vector::value_type *>::value>> : std::true_type {}; + +// [workaround(intel)] Separate function required here +// Workaround as the Intel compiler does not compile the enable_if_t part below +// (tested with icc (ICC) 2021.1 Beta 20200827) +template +constexpr bool args_any_are_buffer() { + return detail::any_of...>::value; +} + +// [workaround(intel)] Separate function required here +// [workaround(msvc)] Can't use constexpr bool in return type + +// Add the buffer interface to a vector +template +void vector_buffer_impl(Class_ &cl, std::true_type) { + using T = typename Vector::value_type; + + static_assert(vector_has_data_and_format::value, + "There is not an appropriate format descriptor for this vector"); + + // numpy.h declares this for arbitrary types, but it may raise an exception and crash hard + // at runtime if PYBIND11_NUMPY_DTYPE hasn't been called, so check here + format_descriptor::format(); + + cl.def_buffer([](Vector &v) -> buffer_info { + return buffer_info(v.data(), + static_cast(sizeof(T)), + format_descriptor::format(), + 1, + {v.size()}, + {sizeof(T)}); + }); + + cl.def(init([](const buffer &buf) { + auto info = buf.request(); + if (info.ndim != 1 || info.strides[0] % static_cast(sizeof(T))) { + throw type_error("Only valid 1D buffers can be copied to a vector"); + } + if (!detail::compare_buffer_info::compare(info) + || (ssize_t) sizeof(T) != info.itemsize) { + throw type_error("Format mismatch (Python: " + info.format + + " C++: " + format_descriptor::format() + ")"); + } + + T *p = static_cast(info.ptr); + ssize_t step = info.strides[0] / static_cast(sizeof(T)); + T *end = p + info.shape[0] * step; + if (step == 1) { + return Vector(p, end); + } + Vector vec; + vec.reserve((size_t) info.shape[0]); + for (; p != end; p += step) { + vec.push_back(*p); + } + return vec; + })); + + return; +} + +template +void vector_buffer_impl(Class_ &, std::false_type) {} + +template +void vector_buffer(Class_ &cl) { + vector_buffer_impl( + cl, detail::any_of...>{}); +} + +PYBIND11_NAMESPACE_END(detail) + +// +// std::vector +// +template , typename... Args> +class_ bind_vector(handle scope, std::string const &name, Args &&...args) { + using Class_ = class_; + + // If the value_type is unregistered (e.g. a converting type) or is itself registered + // module-local then make the vector binding module-local as well: + using vtype = typename Vector::value_type; + auto *vtype_info = detail::get_type_info(typeid(vtype)); + bool local = !vtype_info || vtype_info->module_local; + + Class_ cl(scope, name.c_str(), pybind11::module_local(local), std::forward(args)...); + + // Declare the buffer interface if a buffer_protocol() is passed in + detail::vector_buffer(cl); + + cl.def(init<>()); + + // Register copy constructor (if possible) + detail::vector_if_copy_constructible(cl); + + // Register comparison-related operators and functions (if possible) + detail::vector_if_equal_operator(cl); + + // Register stream insertion operator (if possible) + detail::vector_if_insertion_operator(cl, name); + + // Modifiers require copyable vector value type + detail::vector_modifiers(cl); + + // Accessor and iterator; return by value if copyable, otherwise we return by ref + keep-alive + detail::vector_accessor(cl); + + cl.def( + "__bool__", + [](const Vector &v) -> bool { return !v.empty(); }, + "Check whether the list is nonempty"); + + cl.def("__len__", [](const Vector &vec) { return vec.size(); }); + +#if 0 + // C++ style functions deprecated, leaving it here as an example + cl.def(init()); + + cl.def("resize", + (void (Vector::*) (size_type count)) & Vector::resize, + "changes the number of elements stored"); + + cl.def("erase", + [](Vector &v, SizeType i) { + if (i >= v.size()) + throw index_error(); + v.erase(v.begin() + i); + }, "erases element at index ``i``"); + + cl.def("empty", &Vector::empty, "checks whether the container is empty"); + cl.def("size", &Vector::size, "returns the number of elements"); + cl.def("push_back", (void (Vector::*)(const T&)) &Vector::push_back, "adds an element to the end"); + cl.def("pop_back", &Vector::pop_back, "removes the last element"); + + cl.def("max_size", &Vector::max_size, "returns the maximum possible number of elements"); + cl.def("reserve", &Vector::reserve, "reserves storage"); + cl.def("capacity", &Vector::capacity, "returns the number of elements that can be held in currently allocated storage"); + cl.def("shrink_to_fit", &Vector::shrink_to_fit, "reduces memory usage by freeing unused memory"); + + cl.def("clear", &Vector::clear, "clears the contents"); + cl.def("swap", &Vector::swap, "swaps the contents"); + + cl.def("front", [](Vector &v) { + if (v.size()) return v.front(); + else throw index_error(); + }, "access the first element"); + + cl.def("back", [](Vector &v) { + if (v.size()) return v.back(); + else throw index_error(); + }, "access the last element "); + +#endif + + return cl; +} + +// +// std::map, std::unordered_map +// + +PYBIND11_NAMESPACE_BEGIN(detail) + +/* Fallback functions */ +template +void map_if_insertion_operator(const Args &...) {} +template +void map_assignment(const Args &...) {} + +// Map assignment when copy-assignable: just copy the value +template +void map_assignment( + enable_if_t::value, Class_> &cl) { + using KeyType = typename Map::key_type; + using MappedType = typename Map::mapped_type; + + cl.def("__setitem__", [](Map &m, const KeyType &k, const MappedType &v) { + auto it = m.find(k); + if (it != m.end()) { + it->second = v; + } else { + m.emplace(k, v); + } + }); +} + +// Not copy-assignable, but still copy-constructible: we can update the value by erasing and +// reinserting +template +void map_assignment(enable_if_t::value + && is_copy_constructible::value, + Class_> &cl) { + using KeyType = typename Map::key_type; + using MappedType = typename Map::mapped_type; + + cl.def("__setitem__", [](Map &m, const KeyType &k, const MappedType &v) { + // We can't use m[k] = v; because value type might not be default constructable + auto r = m.emplace(k, v); + if (!r.second) { + // value type is not copy assignable so the only way to insert it is to erase it + // first... + m.erase(r.first); + m.emplace(k, v); + } + }); +} + +template +auto map_if_insertion_operator(Class_ &cl, std::string const &name) + -> decltype(std::declval() << std::declval() + << std::declval(), + void()) { + + cl.def( + "__repr__", + [name](Map &m) { + std::ostringstream s; + s << name << '{'; + bool f = false; + for (auto const &kv : m) { + if (f) { + s << ", "; + } + s << kv.first << ": " << kv.second; + f = true; + } + s << '}'; + return s.str(); + }, + "Return the canonical string representation of this map."); +} + +struct keys_view { + virtual size_t len() = 0; + virtual iterator iter() = 0; + virtual bool contains(const handle &k) = 0; + virtual ~keys_view() = default; +}; + +struct values_view { + virtual size_t len() = 0; + virtual iterator iter() = 0; + virtual ~values_view() = default; +}; + +struct items_view { + virtual size_t len() = 0; + virtual iterator iter() = 0; + virtual ~items_view() = default; +}; + +template +struct KeysViewImpl : public detail::keys_view { + explicit KeysViewImpl(Map &map) : map(map) {} + size_t len() override { return map.size(); } + iterator iter() override { return make_key_iterator(map.begin(), map.end()); } + bool contains(const handle &k) override { + try { + return map.find(k.template cast()) != map.end(); + } catch (const cast_error &) { + return false; + } + } + Map ↦ +}; + +template +struct ValuesViewImpl : public detail::values_view { + explicit ValuesViewImpl(Map &map) : map(map) {} + size_t len() override { return map.size(); } + iterator iter() override { return make_value_iterator(map.begin(), map.end()); } + Map ↦ +}; + +template +struct ItemsViewImpl : public detail::items_view { + explicit ItemsViewImpl(Map &map) : map(map) {} + size_t len() override { return map.size(); } + iterator iter() override { return make_iterator(map.begin(), map.end()); } + Map ↦ +}; + +PYBIND11_NAMESPACE_END(detail) + +template , typename... Args> +class_ bind_map(handle scope, const std::string &name, Args &&...args) { + using KeyType = typename Map::key_type; + using MappedType = typename Map::mapped_type; + using KeysView = detail::keys_view; + using ValuesView = detail::values_view; + using ItemsView = detail::items_view; + using Class_ = class_; + + // If either type is a non-module-local bound type then make the map binding non-local as well; + // otherwise (e.g. both types are either module-local or converting) the map will be + // module-local. + auto *tinfo = detail::get_type_info(typeid(MappedType)); + bool local = !tinfo || tinfo->module_local; + if (local) { + tinfo = detail::get_type_info(typeid(KeyType)); + local = !tinfo || tinfo->module_local; + } + + Class_ cl(scope, name.c_str(), pybind11::module_local(local), std::forward(args)...); + + // Wrap KeysView if it wasn't already wrapped + if (!detail::get_type_info(typeid(KeysView))) { + class_ keys_view(scope, "KeysView", pybind11::module_local(local)); + keys_view.def("__len__", &KeysView::len); + keys_view.def("__iter__", + &KeysView::iter, + keep_alive<0, 1>() /* Essential: keep view alive while iterator exists */ + ); + keys_view.def("__contains__", &KeysView::contains); + } + // Similarly for ValuesView: + if (!detail::get_type_info(typeid(ValuesView))) { + class_ values_view(scope, "ValuesView", pybind11::module_local(local)); + values_view.def("__len__", &ValuesView::len); + values_view.def("__iter__", + &ValuesView::iter, + keep_alive<0, 1>() /* Essential: keep view alive while iterator exists */ + ); + } + // Similarly for ItemsView: + if (!detail::get_type_info(typeid(ItemsView))) { + class_ items_view(scope, "ItemsView", pybind11::module_local(local)); + items_view.def("__len__", &ItemsView::len); + items_view.def("__iter__", + &ItemsView::iter, + keep_alive<0, 1>() /* Essential: keep view alive while iterator exists */ + ); + } + + cl.def(init<>()); + + // Register stream insertion operator (if possible) + detail::map_if_insertion_operator(cl, name); + + cl.def( + "__bool__", + [](const Map &m) -> bool { return !m.empty(); }, + "Check whether the map is nonempty"); + + cl.def( + "__iter__", + [](Map &m) { return make_key_iterator(m.begin(), m.end()); }, + keep_alive<0, 1>() /* Essential: keep map alive while iterator exists */ + ); + + cl.def( + "keys", + [](Map &m) { return std::unique_ptr(new detail::KeysViewImpl(m)); }, + keep_alive<0, 1>() /* Essential: keep map alive while view exists */ + ); + + cl.def( + "values", + [](Map &m) { return std::unique_ptr(new detail::ValuesViewImpl(m)); }, + keep_alive<0, 1>() /* Essential: keep map alive while view exists */ + ); + + cl.def( + "items", + [](Map &m) { return std::unique_ptr(new detail::ItemsViewImpl(m)); }, + keep_alive<0, 1>() /* Essential: keep map alive while view exists */ + ); + + cl.def( + "__getitem__", + [](Map &m, const KeyType &k) -> MappedType & { + auto it = m.find(k); + if (it == m.end()) { + throw key_error(); + } + return it->second; + }, + return_value_policy::reference_internal // ref + keepalive + ); + + cl.def("__contains__", [](Map &m, const KeyType &k) -> bool { + auto it = m.find(k); + if (it == m.end()) { + return false; + } + return true; + }); + // Fallback for when the object is not of the key type + cl.def("__contains__", [](Map &, const object &) -> bool { return false; }); + + // Assignment provided only if the type is copyable + detail::map_assignment(cl); + + cl.def("__delitem__", [](Map &m, const KeyType &k) { + auto it = m.find(k); + if (it == m.end()) { + throw key_error(); + } + m.erase(it); + }); + + // Always use a lambda in case of `using` declaration + cl.def("__len__", [](const Map &m) { return m.size(); }); + + return cl; +} + +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/type_caster_pyobject_ptr.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/type_caster_pyobject_ptr.h new file mode 100644 index 0000000000000000000000000000000000000000..aa914f9e1563b3480481865f50adb7358f1a7134 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/type_caster_pyobject_ptr.h @@ -0,0 +1,61 @@ +// Copyright (c) 2023 The pybind Community. + +#pragma once + +#include "detail/common.h" +#include "detail/descr.h" +#include "cast.h" +#include "pytypes.h" + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) +PYBIND11_NAMESPACE_BEGIN(detail) + +template <> +class type_caster { +public: + static constexpr auto name = const_name("object"); // See discussion under PR #4601. + + // This overload is purely to guard against accidents. + template ::value, int> = 0> + static handle cast(T &&, return_value_policy, handle /*parent*/) { + static_assert(is_same_ignoring_cvref::value, + "Invalid C++ type T for to-Python conversion (type_caster)."); + return nullptr; // Unreachable. + } + + static handle cast(PyObject *src, return_value_policy policy, handle /*parent*/) { + if (src == nullptr) { + throw error_already_set(); + } + if (PyErr_Occurred()) { + raise_from(PyExc_SystemError, "src != nullptr but PyErr_Occurred()"); + throw error_already_set(); + } + if (policy == return_value_policy::take_ownership) { + return src; + } + if (policy == return_value_policy::reference + || policy == return_value_policy::automatic_reference) { + return handle(src).inc_ref(); + } + pybind11_fail("type_caster::cast(): unsupported return_value_policy: " + + std::to_string(static_cast(policy))); + } + + bool load(handle src, bool) { + value = reinterpret_borrow(src); + return true; + } + + template + using cast_op_type = PyObject *; + + explicit operator PyObject *() { return value.ptr(); } + +private: + object value; +}; + +PYBIND11_NAMESPACE_END(detail) +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/typing.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/typing.h new file mode 100644 index 0000000000000000000000000000000000000000..84aaf9f702d780b1415b806f4efc7ebfd3d2675d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/include/pybind11/typing.h @@ -0,0 +1,242 @@ +/* + pybind11/typing.h: Convenience wrapper classes for basic Python types + with more explicit annotations. + + Copyright (c) 2023 Dustin Spicuzza + + All rights reserved. Use of this source code is governed by a + BSD-style license that can be found in the LICENSE file. +*/ + +#pragma once + +#include "detail/common.h" +#include "cast.h" +#include "pytypes.h" + +#include + +PYBIND11_NAMESPACE_BEGIN(PYBIND11_NAMESPACE) +PYBIND11_NAMESPACE_BEGIN(typing) + +/* + The following types can be used to direct pybind11-generated docstrings + to have have more explicit types (e.g., `list[str]` instead of `list`). + Just use these in place of existing types. + + There is no additional enforcement of types at runtime. +*/ + +template +class Tuple : public tuple { + using tuple::tuple; +}; + +template +class Dict : public dict { + using dict::dict; +}; + +template +class List : public list { + using list::list; +}; + +template +class Set : public set { + using set::set; +}; + +template +class Iterable : public iterable { + using iterable::iterable; +}; + +template +class Iterator : public iterator { + using iterator::iterator; +}; + +template +class Callable; + +template +class Callable : public function { + using function::function; +}; + +template +class Type : public type { + using type::type; +}; + +template +class Union : public object { + PYBIND11_OBJECT_DEFAULT(Union, object, PyObject_Type) + using object::object; +}; + +template +class Optional : public object { + PYBIND11_OBJECT_DEFAULT(Optional, object, PyObject_Type) + using object::object; +}; + +template +class TypeGuard : public bool_ { + using bool_::bool_; +}; + +template +class TypeIs : public bool_ { + using bool_::bool_; +}; + +class NoReturn : public none { + using none::none; +}; + +class Never : public none { + using none::none; +}; + +#if defined(__cpp_nontype_template_args) && __cpp_nontype_template_args >= 201911L +# define PYBIND11_TYPING_H_HAS_STRING_LITERAL +template +struct StringLiteral { + constexpr StringLiteral(const char (&str)[N]) { std::copy_n(str, N, name); } + char name[N]; +}; + +template +class Literal : public object { + PYBIND11_OBJECT_DEFAULT(Literal, object, PyObject_Type) +}; + +// Example syntax for creating a TypeVar. +// typedef typing::TypeVar<"T"> TypeVarT; +template +class TypeVar : public object { + PYBIND11_OBJECT_DEFAULT(TypeVar, object, PyObject_Type) + using object::object; +}; +#endif + +PYBIND11_NAMESPACE_END(typing) + +PYBIND11_NAMESPACE_BEGIN(detail) + +template +struct handle_type_name> { + static constexpr auto name = const_name("tuple[") + + ::pybind11::detail::concat(make_caster::name...) + + const_name("]"); +}; + +template <> +struct handle_type_name> { + // PEP 484 specifies this syntax for an empty tuple + static constexpr auto name = const_name("tuple[()]"); +}; + +template +struct handle_type_name> { + // PEP 484 specifies this syntax for a variable-length tuple + static constexpr auto name + = const_name("tuple[") + make_caster::name + const_name(", ...]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("dict[") + make_caster::name + const_name(", ") + + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("list[") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("set[") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("Iterable[") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("Iterator[") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + using retval_type = conditional_t::value, void_type, Return>; + static constexpr auto name + = const_name("Callable[[") + ::pybind11::detail::concat(make_caster::name...) + + const_name("], ") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + // PEP 484 specifies this syntax for defining only return types of callables + using retval_type = conditional_t::value, void_type, Return>; + static constexpr auto name + = const_name("Callable[..., ") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("type[") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("Union[") + + ::pybind11::detail::concat(make_caster::name...) + + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("Optional[") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("TypeGuard[") + make_caster::name + const_name("]"); +}; + +template +struct handle_type_name> { + static constexpr auto name = const_name("TypeIs[") + make_caster::name + const_name("]"); +}; + +template <> +struct handle_type_name { + static constexpr auto name = const_name("NoReturn"); +}; + +template <> +struct handle_type_name { + static constexpr auto name = const_name("Never"); +}; + +#if defined(PYBIND11_TYPING_H_HAS_STRING_LITERAL) +template +struct handle_type_name> { + static constexpr auto name = const_name("Literal[") + + pybind11::detail::concat(const_name(Literals.name)...) + + const_name("]"); +}; +template +struct handle_type_name> { + static constexpr auto name = const_name(StrLit.name); +}; +#endif + +PYBIND11_NAMESPACE_END(detail) +PYBIND11_NAMESPACE_END(PYBIND11_NAMESPACE) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/setup_helpers.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/setup_helpers.py new file mode 100644 index 0000000000000000000000000000000000000000..ced506f8c25fddd76258df380648c89f36fc6d59 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/pybind11/setup_helpers.py @@ -0,0 +1,500 @@ +""" +This module provides helpers for C++11+ projects using pybind11. + +LICENSE: + +Copyright (c) 2016 Wenzel Jakob , All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +1. Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. + +2. Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +3. Neither the name of the copyright holder nor the names of its contributors + may be used to endorse or promote products derived from this software + without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND +ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED +WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE +FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR +SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER +CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, +OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +""" + +# IMPORTANT: If you change this file in the pybind11 repo, also review +# setup_helpers.pyi for matching changes. +# +# If you copy this file in, you don't +# need the .pyi file; it's just an interface file for static type checkers. +from __future__ import annotations + +import contextlib +import os +import platform +import shlex +import shutil +import sys +import sysconfig +import tempfile +import threading +import warnings +from functools import lru_cache +from pathlib import Path +from typing import ( + Any, + Callable, + Iterable, + Iterator, + List, + Optional, + Tuple, + TypeVar, + Union, +) + +try: + from setuptools import Extension as _Extension + from setuptools.command.build_ext import build_ext as _build_ext +except ImportError: + from distutils.command.build_ext import ( # type: ignore[assignment] + build_ext as _build_ext, + ) + from distutils.extension import Extension as _Extension # type: ignore[assignment] + +import distutils.ccompiler +import distutils.errors + +WIN = sys.platform.startswith("win32") and "mingw" not in sysconfig.get_platform() +MACOS = sys.platform.startswith("darwin") +STD_TMPL = "/std:c++{}" if WIN else "-std=c++{}" + + +# It is recommended to use PEP 518 builds if using this module. However, this +# file explicitly supports being copied into a user's project directory +# standalone, and pulling pybind11 with the deprecated setup_requires feature. +# If you copy the file, remember to add it to your MANIFEST.in, and add the current +# directory into your path if it sits beside your setup.py. + + +class Pybind11Extension(_Extension): + """ + Build a C++11+ Extension module with pybind11. This automatically adds the + recommended flags when you init the extension and assumes C++ sources - you + can further modify the options yourself. + + The customizations are: + + * ``/EHsc`` and ``/bigobj`` on Windows + * ``stdlib=libc++`` on macOS + * ``visibility=hidden`` and ``-g0`` on Unix + + Finally, you can set ``cxx_std`` via constructor or afterwards to enable + flags for C++ std, and a few extra helper flags related to the C++ standard + level. It is _highly_ recommended you either set this, or use the provided + ``build_ext``, which will search for the highest supported extension for + you if the ``cxx_std`` property is not set. Do not set the ``cxx_std`` + property more than once, as flags are added when you set it. Set the + property to None to disable the addition of C++ standard flags. + + If you want to add pybind11 headers manually, for example for an exact + git checkout, then set ``include_pybind11=False``. + """ + + # flags are prepended, so that they can be further overridden, e.g. by + # ``extra_compile_args=["-g"]``. + + def _add_cflags(self, flags: list[str]) -> None: + self.extra_compile_args[:0] = flags + + def _add_ldflags(self, flags: list[str]) -> None: + self.extra_link_args[:0] = flags + + def __init__(self, *args: Any, **kwargs: Any) -> None: + self._cxx_level = 0 + cxx_std = kwargs.pop("cxx_std", 0) + + if "language" not in kwargs: + kwargs["language"] = "c++" + + include_pybind11 = kwargs.pop("include_pybind11", True) + + super().__init__(*args, **kwargs) + + # Include the installed package pybind11 headers + if include_pybind11: + # If using setup_requires, this fails the first time - that's okay + try: + import pybind11 + + pyinc = pybind11.get_include() + + if pyinc not in self.include_dirs: + self.include_dirs.append(pyinc) + except ModuleNotFoundError: + pass + + self.cxx_std = cxx_std + + cflags = [] + if WIN: + cflags += ["/EHsc", "/bigobj"] + else: + cflags += ["-fvisibility=hidden"] + env_cflags = os.environ.get("CFLAGS", "") + env_cppflags = os.environ.get("CPPFLAGS", "") + c_cpp_flags = shlex.split(env_cflags) + shlex.split(env_cppflags) + if not any(opt.startswith("-g") for opt in c_cpp_flags): + cflags += ["-g0"] + self._add_cflags(cflags) + + @property + def cxx_std(self) -> int: + """ + The CXX standard level. If set, will add the required flags. If left at + 0, it will trigger an automatic search when pybind11's build_ext is + used. If None, will have no effect. Besides just the flags, this may + add a macos-min 10.9 or 10.14 flag if MACOSX_DEPLOYMENT_TARGET is + unset. + """ + return self._cxx_level + + @cxx_std.setter + def cxx_std(self, level: int) -> None: + if self._cxx_level: + warnings.warn( + "You cannot safely change the cxx_level after setting it!", stacklevel=2 + ) + + # MSVC 2015 Update 3 and later only have 14 (and later 17) modes, so + # force a valid flag here. + if WIN and level == 11: + level = 14 + + self._cxx_level = level + + if not level: + return + + cflags = [STD_TMPL.format(level)] + ldflags = [] + + if MACOS and "MACOSX_DEPLOYMENT_TARGET" not in os.environ: + # C++17 requires a higher min version of macOS. An earlier version + # (10.12 or 10.13) can be set manually via environment variable if + # you are careful in your feature usage, but 10.14 is the safest + # setting for general use. However, never set higher than the + # current macOS version! + current_macos = tuple(int(x) for x in platform.mac_ver()[0].split(".")[:2]) + desired_macos = (10, 9) if level < 17 else (10, 14) + macos_string = ".".join(str(x) for x in min(current_macos, desired_macos)) + macosx_min = f"-mmacosx-version-min={macos_string}" + cflags += [macosx_min] + ldflags += [macosx_min] + + self._add_cflags(cflags) + self._add_ldflags(ldflags) + + +# Just in case someone clever tries to multithread +tmp_chdir_lock = threading.Lock() + + +@contextlib.contextmanager +def tmp_chdir() -> Iterator[str]: + "Prepare and enter a temporary directory, cleanup when done" + + # Threadsafe + with tmp_chdir_lock: + olddir = os.getcwd() + try: + tmpdir = tempfile.mkdtemp() + os.chdir(tmpdir) + yield tmpdir + finally: + os.chdir(olddir) + shutil.rmtree(tmpdir) + + +# cf http://bugs.python.org/issue26689 +def has_flag(compiler: Any, flag: str) -> bool: + """ + Return the flag if a flag name is supported on the + specified compiler, otherwise None (can be used as a boolean). + If multiple flags are passed, return the first that matches. + """ + + with tmp_chdir(): + fname = Path("flagcheck.cpp") + # Don't trigger -Wunused-parameter. + fname.write_text("int main (int, char **) { return 0; }", encoding="utf-8") + + try: + compiler.compile([str(fname)], extra_postargs=[flag]) + except distutils.errors.CompileError: + return False + return True + + +# Every call will cache the result +cpp_flag_cache = None + + +@lru_cache() +def auto_cpp_level(compiler: Any) -> str | int: + """ + Return the max supported C++ std level (17, 14, or 11). Returns latest on Windows. + """ + + if WIN: + return "latest" + + levels = [17, 14, 11] + + for level in levels: + if has_flag(compiler, STD_TMPL.format(level)): + return level + + msg = "Unsupported compiler -- at least C++11 support is needed!" + raise RuntimeError(msg) + + +class build_ext(_build_ext): # noqa: N801 + """ + Customized build_ext that allows an auto-search for the highest supported + C++ level for Pybind11Extension. This is only needed for the auto-search + for now, and is completely optional otherwise. + """ + + def build_extensions(self) -> None: + """ + Build extensions, injecting C++ std for Pybind11Extension if needed. + """ + + for ext in self.extensions: + if hasattr(ext, "_cxx_level") and ext._cxx_level == 0: + ext.cxx_std = auto_cpp_level(self.compiler) + + super().build_extensions() + + +def intree_extensions( + paths: Iterable[str], package_dir: dict[str, str] | None = None +) -> list[Pybind11Extension]: + """ + Generate Pybind11Extensions from source files directly located in a Python + source tree. + + ``package_dir`` behaves as in ``setuptools.setup``. If unset, the Python + package root parent is determined as the first parent directory that does + not contain an ``__init__.py`` file. + """ + exts = [] + + if package_dir is None: + for path in paths: + parent, _ = os.path.split(path) + while os.path.exists(os.path.join(parent, "__init__.py")): + parent, _ = os.path.split(parent) + relname, _ = os.path.splitext(os.path.relpath(path, parent)) + qualified_name = relname.replace(os.path.sep, ".") + exts.append(Pybind11Extension(qualified_name, [path])) + return exts + + for path in paths: + for prefix, parent in package_dir.items(): + if path.startswith(parent): + relname, _ = os.path.splitext(os.path.relpath(path, parent)) + qualified_name = relname.replace(os.path.sep, ".") + if prefix: + qualified_name = prefix + "." + qualified_name + exts.append(Pybind11Extension(qualified_name, [path])) + break + else: + msg = ( + f"path {path} is not a child of any of the directories listed " + f"in 'package_dir' ({package_dir})" + ) + raise ValueError(msg) + + return exts + + +def naive_recompile(obj: str, src: str) -> bool: + """ + This will recompile only if the source file changes. It does not check + header files, so a more advanced function or Ccache is better if you have + editable header files in your package. + """ + return os.stat(obj).st_mtime < os.stat(src).st_mtime + + +def no_recompile(obg: str, src: str) -> bool: # noqa: ARG001 + """ + This is the safest but slowest choice (and is the default) - will always + recompile sources. + """ + return True + + +S = TypeVar("S", bound="ParallelCompile") + +CCompilerMethod = Callable[ + [ + distutils.ccompiler.CCompiler, + List[str], + Optional[str], + Optional[List[Union[Tuple[str], Tuple[str, Optional[str]]]]], + Optional[List[str]], + bool, + Optional[List[str]], + Optional[List[str]], + Optional[List[str]], + ], + List[str], +] + + +# Optional parallel compile utility +# inspired by: http://stackoverflow.com/questions/11013851/speeding-up-build-process-with-distutils +# and: https://github.com/tbenthompson/cppimport/blob/stable/cppimport/build_module.py +# and NumPy's parallel distutils module: +# https://github.com/numpy/numpy/blob/master/numpy/distutils/ccompiler.py +class ParallelCompile: + """ + Make a parallel compile function. Inspired by + numpy.distutils.ccompiler.CCompiler.compile and cppimport. + + This takes several arguments that allow you to customize the compile + function created: + + envvar: + Set an environment variable to control the compilation threads, like + NPY_NUM_BUILD_JOBS + default: + 0 will automatically multithread, or 1 will only multithread if the + envvar is set. + max: + The limit for automatic multithreading if non-zero + needs_recompile: + A function of (obj, src) that returns True when recompile is needed. No + effect in isolated mode; use ccache instead, see + https://github.com/matplotlib/matplotlib/issues/1507/ + + To use:: + + ParallelCompile("NPY_NUM_BUILD_JOBS").install() + + or:: + + with ParallelCompile("NPY_NUM_BUILD_JOBS"): + setup(...) + + By default, this assumes all files need to be recompiled. A smarter + function can be provided via needs_recompile. If the output has not yet + been generated, the compile will always run, and this function is not + called. + """ + + __slots__ = ("envvar", "default", "max", "_old", "needs_recompile") + + def __init__( + self, + envvar: str | None = None, + default: int = 0, + max: int = 0, # pylint: disable=redefined-builtin + needs_recompile: Callable[[str, str], bool] = no_recompile, + ) -> None: + self.envvar = envvar + self.default = default + self.max = max + self.needs_recompile = needs_recompile + self._old: list[CCompilerMethod] = [] + + def function(self) -> CCompilerMethod: + """ + Builds a function object usable as distutils.ccompiler.CCompiler.compile. + """ + + def compile_function( + compiler: distutils.ccompiler.CCompiler, + sources: list[str], + output_dir: str | None = None, + macros: list[tuple[str] | tuple[str, str | None]] | None = None, + include_dirs: list[str] | None = None, + debug: bool = False, + extra_preargs: list[str] | None = None, + extra_postargs: list[str] | None = None, + depends: list[str] | None = None, + ) -> Any: + # These lines are directly from distutils.ccompiler.CCompiler + macros, objects, extra_postargs, pp_opts, build = compiler._setup_compile( # type: ignore[attr-defined] + output_dir, macros, include_dirs, sources, depends, extra_postargs + ) + cc_args = compiler._get_cc_args(pp_opts, debug, extra_preargs) # type: ignore[attr-defined] + + # The number of threads; start with default. + threads = self.default + + # Determine the number of compilation threads, unless set by an environment variable. + if self.envvar is not None: + threads = int(os.environ.get(self.envvar, self.default)) + + def _single_compile(obj: Any) -> None: + try: + src, ext = build[obj] + except KeyError: + return + + if not os.path.exists(obj) or self.needs_recompile(obj, src): + compiler._compile(obj, src, ext, cc_args, extra_postargs, pp_opts) # type: ignore[attr-defined] + + try: + # Importing .synchronize checks for platforms that have some multiprocessing + # capabilities but lack semaphores, such as AWS Lambda and Android Termux. + import multiprocessing.synchronize + from multiprocessing.pool import ThreadPool + except ImportError: + threads = 1 + + if threads == 0: + try: + threads = multiprocessing.cpu_count() + threads = self.max if self.max and self.max < threads else threads + except NotImplementedError: + threads = 1 + + if threads > 1: + with ThreadPool(threads) as pool: + for _ in pool.imap_unordered(_single_compile, objects): + pass + else: + for ob in objects: + _single_compile(ob) + + return objects + + return compile_function + + def install(self: S) -> S: + """ + Installs the compile function into distutils.ccompiler.CCompiler.compile. + """ + distutils.ccompiler.CCompiler.compile = self.function() # type: ignore[assignment] + return self + + def __enter__(self: S) -> S: + self._old.append(distutils.ccompiler.CCompiler.compile) + return self.install() + + def __exit__(self, *args: Any) -> None: + distutils.ccompiler.CCompiler.compile = self._old.pop() # type: ignore[assignment] diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/METADATA b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/METADATA new file mode 100644 index 0000000000000000000000000000000000000000..a9fe0954d7e4eb70d4805d8f0d8df08036eb63ef --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/METADATA @@ -0,0 +1,520 @@ +Metadata-Version: 2.1 +Name: torch +Version: 2.3.0+cu118 +Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration +Home-page: https://pytorch.org/ +Download-URL: https://github.com/pytorch/pytorch/tags +Author: PyTorch Team +Author-email: packages@pytorch.org +License: BSD-3 +Keywords: pytorch,machine learning +Classifier: Development Status :: 5 - Production/Stable +Classifier: Intended Audience :: Developers +Classifier: Intended Audience :: Education +Classifier: Intended Audience :: Science/Research +Classifier: License :: OSI Approved :: BSD License +Classifier: Topic :: Scientific/Engineering +Classifier: Topic :: Scientific/Engineering :: Mathematics +Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence +Classifier: Topic :: Software Development +Classifier: Topic :: Software Development :: Libraries +Classifier: Topic :: Software Development :: Libraries :: Python Modules +Classifier: Programming Language :: C++ +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Programming Language :: Python :: 3.12 +Requires-Python: >=3.8.0 +Description-Content-Type: text/markdown +License-File: LICENSE +License-File: NOTICE +Requires-Dist: filelock +Requires-Dist: typing-extensions (>=4.8.0) +Requires-Dist: sympy +Requires-Dist: networkx +Requires-Dist: jinja2 +Requires-Dist: fsspec +Requires-Dist: nvidia-cuda-nvrtc-cu11 (==11.8.89) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-cuda-runtime-cu11 (==11.8.89) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-cuda-cupti-cu11 (==11.8.87) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-cudnn-cu11 (==8.7.0.84) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-cublas-cu11 (==11.11.3.6) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-cufft-cu11 (==10.9.0.58) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-curand-cu11 (==10.3.0.86) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-cusolver-cu11 (==11.4.1.48) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-cusparse-cu11 (==11.7.5.86) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-nccl-cu11 (==2.20.5) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: nvidia-nvtx-cu11 (==11.8.86) ; platform_system == "Linux" and platform_machine == "x86_64" +Requires-Dist: triton (==2.3.0) ; platform_system == "Linux" and platform_machine == "x86_64" and python_version < "3.12" +Requires-Dist: mkl (<=2021.4.0,>=2021.1.1) ; platform_system == "Windows" +Provides-Extra: opt-einsum +Requires-Dist: opt-einsum (>=3.3) ; extra == 'opt-einsum' +Provides-Extra: optree +Requires-Dist: optree (>=0.9.1) ; extra == 'optree' + +![PyTorch Logo](https://github.com/pytorch/pytorch/blob/main/docs/source/_static/img/pytorch-logo-dark.png) + +-------------------------------------------------------------------------------- + +PyTorch is a Python package that provides two high-level features: +- Tensor computation (like NumPy) with strong GPU acceleration +- Deep neural networks built on a tape-based autograd system + +You can reuse your favorite Python packages such as NumPy, SciPy, and Cython to extend PyTorch when needed. + +Our trunk health (Continuous Integration signals) can be found at [hud.pytorch.org](https://hud.pytorch.org/ci/pytorch/pytorch/main). + + + +- [More About PyTorch](#more-about-pytorch) + - [A GPU-Ready Tensor Library](#a-gpu-ready-tensor-library) + - [Dynamic Neural Networks: Tape-Based Autograd](#dynamic-neural-networks-tape-based-autograd) + - [Python First](#python-first) + - [Imperative Experiences](#imperative-experiences) + - [Fast and Lean](#fast-and-lean) + - [Extensions Without Pain](#extensions-without-pain) +- [Installation](#installation) + - [Binaries](#binaries) + - [NVIDIA Jetson Platforms](#nvidia-jetson-platforms) + - [From Source](#from-source) + - [Prerequisites](#prerequisites) + - [Install Dependencies](#install-dependencies) + - [Get the PyTorch Source](#get-the-pytorch-source) + - [Install PyTorch](#install-pytorch) + - [Adjust Build Options (Optional)](#adjust-build-options-optional) + - [Docker Image](#docker-image) + - [Using pre-built images](#using-pre-built-images) + - [Building the image yourself](#building-the-image-yourself) + - [Building the Documentation](#building-the-documentation) + - [Previous Versions](#previous-versions) +- [Getting Started](#getting-started) +- [Resources](#resources) +- [Communication](#communication) +- [Releases and Contributing](#releases-and-contributing) +- [The Team](#the-team) +- [License](#license) + + + +## More About PyTorch + +[Learn the basics of PyTorch](https://pytorch.org/tutorials/beginner/basics/intro.html) + +At a granular level, PyTorch is a library that consists of the following components: + +| Component | Description | +| ---- | --- | +| [**torch**](https://pytorch.org/docs/stable/torch.html) | A Tensor library like NumPy, with strong GPU support | +| [**torch.autograd**](https://pytorch.org/docs/stable/autograd.html) | A tape-based automatic differentiation library that supports all differentiable Tensor operations in torch | +| [**torch.jit**](https://pytorch.org/docs/stable/jit.html) | A compilation stack (TorchScript) to create serializable and optimizable models from PyTorch code | +| [**torch.nn**](https://pytorch.org/docs/stable/nn.html) | A neural networks library deeply integrated with autograd designed for maximum flexibility | +| [**torch.multiprocessing**](https://pytorch.org/docs/stable/multiprocessing.html) | Python multiprocessing, but with magical memory sharing of torch Tensors across processes. Useful for data loading and Hogwild training | +| [**torch.utils**](https://pytorch.org/docs/stable/data.html) | DataLoader and other utility functions for convenience | + +Usually, PyTorch is used either as: + +- A replacement for NumPy to use the power of GPUs. +- A deep learning research platform that provides maximum flexibility and speed. + +Elaborating Further: + +### A GPU-Ready Tensor Library + +If you use NumPy, then you have used Tensors (a.k.a. ndarray). + +![Tensor illustration](./docs/source/_static/img/tensor_illustration.png) + +PyTorch provides Tensors that can live either on the CPU or the GPU and accelerates the +computation by a huge amount. + +We provide a wide variety of tensor routines to accelerate and fit your scientific computation needs +such as slicing, indexing, mathematical operations, linear algebra, reductions. +And they are fast! + +### Dynamic Neural Networks: Tape-Based Autograd + +PyTorch has a unique way of building neural networks: using and replaying a tape recorder. + +Most frameworks such as TensorFlow, Theano, Caffe, and CNTK have a static view of the world. +One has to build a neural network and reuse the same structure again and again. +Changing the way the network behaves means that one has to start from scratch. + +With PyTorch, we use a technique called reverse-mode auto-differentiation, which allows you to +change the way your network behaves arbitrarily with zero lag or overhead. Our inspiration comes +from several research papers on this topic, as well as current and past work such as +[torch-autograd](https://github.com/twitter/torch-autograd), +[autograd](https://github.com/HIPS/autograd), +[Chainer](https://chainer.org), etc. + +While this technique is not unique to PyTorch, it's one of the fastest implementations of it to date. +You get the best of speed and flexibility for your crazy research. + +![Dynamic graph](https://github.com/pytorch/pytorch/blob/main/docs/source/_static/img/dynamic_graph.gif) + +### Python First + +PyTorch is not a Python binding into a monolithic C++ framework. +It is built to be deeply integrated into Python. +You can use it naturally like you would use [NumPy](https://www.numpy.org/) / [SciPy](https://www.scipy.org/) / [scikit-learn](https://scikit-learn.org) etc. +You can write your new neural network layers in Python itself, using your favorite libraries +and use packages such as [Cython](https://cython.org/) and [Numba](http://numba.pydata.org/). +Our goal is to not reinvent the wheel where appropriate. + +### Imperative Experiences + +PyTorch is designed to be intuitive, linear in thought, and easy to use. +When you execute a line of code, it gets executed. There isn't an asynchronous view of the world. +When you drop into a debugger or receive error messages and stack traces, understanding them is straightforward. +The stack trace points to exactly where your code was defined. +We hope you never spend hours debugging your code because of bad stack traces or asynchronous and opaque execution engines. + +### Fast and Lean + +PyTorch has minimal framework overhead. We integrate acceleration libraries +such as [Intel MKL](https://software.intel.com/mkl) and NVIDIA ([cuDNN](https://developer.nvidia.com/cudnn), [NCCL](https://developer.nvidia.com/nccl)) to maximize speed. +At the core, its CPU and GPU Tensor and neural network backends +are mature and have been tested for years. + +Hence, PyTorch is quite fast — whether you run small or large neural networks. + +The memory usage in PyTorch is extremely efficient compared to Torch or some of the alternatives. +We've written custom memory allocators for the GPU to make sure that +your deep learning models are maximally memory efficient. +This enables you to train bigger deep learning models than before. + +### Extensions Without Pain + +Writing new neural network modules, or interfacing with PyTorch's Tensor API was designed to be straightforward +and with minimal abstractions. + +You can write new neural network layers in Python using the torch API +[or your favorite NumPy-based libraries such as SciPy](https://pytorch.org/tutorials/advanced/numpy_extensions_tutorial.html). + +If you want to write your layers in C/C++, we provide a convenient extension API that is efficient and with minimal boilerplate. +No wrapper code needs to be written. You can see [a tutorial here](https://pytorch.org/tutorials/advanced/cpp_extension.html) and [an example here](https://github.com/pytorch/extension-cpp). + + +## Installation + +### Binaries +Commands to install binaries via Conda or pip wheels are on our website: [https://pytorch.org/get-started/locally/](https://pytorch.org/get-started/locally/) + + +#### NVIDIA Jetson Platforms + +Python wheels for NVIDIA's Jetson Nano, Jetson TX1/TX2, Jetson Xavier NX/AGX, and Jetson AGX Orin are provided [here](https://forums.developer.nvidia.com/t/pytorch-for-jetson-version-1-10-now-available/72048) and the L4T container is published [here](https://catalog.ngc.nvidia.com/orgs/nvidia/containers/l4t-pytorch) + +They require JetPack 4.2 and above, and [@dusty-nv](https://github.com/dusty-nv) and [@ptrblck](https://github.com/ptrblck) are maintaining them. + + +### From Source + +#### Prerequisites +If you are installing from source, you will need: +- Python 3.8 or later (for Linux, Python 3.8.1+ is needed) +- A compiler that fully supports C++17, such as clang or gcc (gcc 9.4.0 or newer is required) + +We highly recommend installing an [Anaconda](https://www.anaconda.com/download) environment. You will get a high-quality BLAS library (MKL) and you get controlled dependency versions regardless of your Linux distro. + +If you want to compile with CUDA support, [select a supported version of CUDA from our support matrix](https://pytorch.org/get-started/locally/), then install the following: +- [NVIDIA CUDA](https://developer.nvidia.com/cuda-downloads) +- [NVIDIA cuDNN](https://developer.nvidia.com/cudnn) v8.5 or above +- [Compiler](https://gist.github.com/ax3l/9489132) compatible with CUDA + +Note: You could refer to the [cuDNN Support Matrix](https://docs.nvidia.com/deeplearning/cudnn/reference/support-matrix.html) for cuDNN versions with the various supported CUDA, CUDA driver and NVIDIA hardware + +If you want to disable CUDA support, export the environment variable `USE_CUDA=0`. +Other potentially useful environment variables may be found in `setup.py`. + +If you are building for NVIDIA's Jetson platforms (Jetson Nano, TX1, TX2, AGX Xavier), Instructions to install PyTorch for Jetson Nano are [available here](https://devtalk.nvidia.com/default/topic/1049071/jetson-nano/pytorch-for-jetson-nano/) + +If you want to compile with ROCm support, install +- [AMD ROCm](https://rocm.docs.amd.com/en/latest/deploy/linux/quick_start.html) 4.0 and above installation +- ROCm is currently supported only for Linux systems. + +If you want to disable ROCm support, export the environment variable `USE_ROCM=0`. +Other potentially useful environment variables may be found in `setup.py`. + +#### Install Dependencies + +**Common** + +```bash +conda install cmake ninja +# Run this command from the PyTorch directory after cloning the source code using the “Get the PyTorch Source“ section below +pip install -r requirements.txt +``` + +**On Linux** + +```bash +conda install intel::mkl-static intel::mkl-include +# CUDA only: Add LAPACK support for the GPU if needed +conda install -c pytorch magma-cuda110 # or the magma-cuda* that matches your CUDA version from https://anaconda.org/pytorch/repo + +# (optional) If using torch.compile with inductor/triton, install the matching version of triton +# Run from the pytorch directory after cloning +make triton +``` + +**On MacOS** + +```bash +# Add this package on intel x86 processor machines only +conda install intel::mkl-static intel::mkl-include +# Add these packages if torch.distributed is needed +conda install pkg-config libuv +``` + +**On Windows** + +```bash +conda install intel::mkl-static intel::mkl-include +# Add these packages if torch.distributed is needed. +# Distributed package support on Windows is a prototype feature and is subject to changes. +conda install -c conda-forge libuv=1.39 +``` + +#### Get the PyTorch Source +```bash +git clone --recursive https://github.com/pytorch/pytorch +cd pytorch +# if you are updating an existing checkout +git submodule sync +git submodule update --init --recursive +``` + +#### Install PyTorch +**On Linux** + +If you would like to compile PyTorch with [new C++ ABI](https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dual_abi.html) enabled, then first run this command: +```bash +export _GLIBCXX_USE_CXX11_ABI=1 +``` + +If you're compiling for AMD ROCm then first run this command: +```bash +# Only run this if you're compiling for ROCm +python tools/amd_build/build_amd.py +``` + +Install PyTorch +```bash +export CMAKE_PREFIX_PATH=${CONDA_PREFIX:-"$(dirname $(which conda))/../"} +python setup.py develop +``` + +> _Aside:_ If you are using [Anaconda](https://www.anaconda.com/distribution/#download-section), you may experience an error caused by the linker: +> +> ```plaintext +> build/temp.linux-x86_64-3.7/torch/csrc/stub.o: file not recognized: file format not recognized +> collect2: error: ld returned 1 exit status +> error: command 'g++' failed with exit status 1 +> ``` +> +> This is caused by `ld` from the Conda environment shadowing the system `ld`. You should use a newer version of Python that fixes this issue. The recommended Python version is 3.8.1+. + +**On macOS** + +```bash +python3 setup.py develop +``` + +**On Windows** + +Choose Correct Visual Studio Version. + +PyTorch CI uses Visual C++ BuildTools, which come with Visual Studio Enterprise, +Professional, or Community Editions. You can also install the build tools from +https://visualstudio.microsoft.com/visual-cpp-build-tools/. The build tools *do not* +come with Visual Studio Code by default. + +If you want to build legacy python code, please refer to [Building on legacy code and CUDA](https://github.com/pytorch/pytorch/blob/main/CONTRIBUTING.md#building-on-legacy-code-and-cuda) + +**CPU-only builds** + +In this mode PyTorch computations will run on your CPU, not your GPU + +```cmd +conda activate +python setup.py develop +``` + +Note on OpenMP: The desired OpenMP implementation is Intel OpenMP (iomp). In order to link against iomp, you'll need to manually download the library and set up the building environment by tweaking `CMAKE_INCLUDE_PATH` and `LIB`. The instruction [here](https://github.com/pytorch/pytorch/blob/main/docs/source/notes/windows.rst#building-from-source) is an example for setting up both MKL and Intel OpenMP. Without these configurations for CMake, Microsoft Visual C OpenMP runtime (vcomp) will be used. + +**CUDA based build** + +In this mode PyTorch computations will leverage your GPU via CUDA for faster number crunching + +[NVTX](https://docs.nvidia.com/gameworks/content/gameworkslibrary/nvtx/nvidia_tools_extension_library_nvtx.htm) is needed to build Pytorch with CUDA. +NVTX is a part of CUDA distributive, where it is called "Nsight Compute". To install it onto an already installed CUDA run CUDA installation once again and check the corresponding checkbox. +Make sure that CUDA with Nsight Compute is installed after Visual Studio. + +Currently, VS 2017 / 2019, and Ninja are supported as the generator of CMake. If `ninja.exe` is detected in `PATH`, then Ninja will be used as the default generator, otherwise, it will use VS 2017 / 2019. +
If Ninja is selected as the generator, the latest MSVC will get selected as the underlying toolchain. + +Additional libraries such as +[Magma](https://developer.nvidia.com/magma), [oneDNN, a.k.a. MKLDNN or DNNL](https://github.com/oneapi-src/oneDNN), and [Sccache](https://github.com/mozilla/sccache) are often needed. Please refer to the [installation-helper](https://github.com/pytorch/pytorch/tree/main/.ci/pytorch/win-test-helpers/installation-helpers) to install them. + +You can refer to the [build_pytorch.bat](https://github.com/pytorch/pytorch/blob/main/.ci/pytorch/win-test-helpers/build_pytorch.bat) script for some other environment variables configurations + + +```cmd +cmd + +:: Set the environment variables after you have downloaded and unzipped the mkl package, +:: else CMake would throw an error as `Could NOT find OpenMP`. +set CMAKE_INCLUDE_PATH={Your directory}\mkl\include +set LIB={Your directory}\mkl\lib;%LIB% + +:: Read the content in the previous section carefully before you proceed. +:: [Optional] If you want to override the underlying toolset used by Ninja and Visual Studio with CUDA, please run the following script block. +:: "Visual Studio 2019 Developer Command Prompt" will be run automatically. +:: Make sure you have CMake >= 3.12 before you do this when you use the Visual Studio generator. +set CMAKE_GENERATOR_TOOLSET_VERSION=14.27 +set DISTUTILS_USE_SDK=1 +for /f "usebackq tokens=*" %i in (`"%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" -version [15^,17^) -products * -latest -property installationPath`) do call "%i\VC\Auxiliary\Build\vcvarsall.bat" x64 -vcvars_ver=%CMAKE_GENERATOR_TOOLSET_VERSION% + +:: [Optional] If you want to override the CUDA host compiler +set CUDAHOSTCXX=C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\bin\HostX64\x64\cl.exe + +python setup.py develop + +``` + +##### Adjust Build Options (Optional) + +You can adjust the configuration of cmake variables optionally (without building first), by doing +the following. For example, adjusting the pre-detected directories for CuDNN or BLAS can be done +with such a step. + +On Linux +```bash +export CMAKE_PREFIX_PATH=${CONDA_PREFIX:-"$(dirname $(which conda))/../"} +python setup.py build --cmake-only +ccmake build # or cmake-gui build +``` + +On macOS +```bash +export CMAKE_PREFIX_PATH=${CONDA_PREFIX:-"$(dirname $(which conda))/../"} +MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py build --cmake-only +ccmake build # or cmake-gui build +``` + +### Docker Image + +#### Using pre-built images + +You can also pull a pre-built docker image from Docker Hub and run with docker v19.03+ + +```bash +docker run --gpus all --rm -ti --ipc=host pytorch/pytorch:latest +``` + +Please note that PyTorch uses shared memory to share data between processes, so if torch multiprocessing is used (e.g. +for multithreaded data loaders) the default shared memory segment size that container runs with is not enough, and you +should increase shared memory size either with `--ipc=host` or `--shm-size` command line options to `nvidia-docker run`. + +#### Building the image yourself + +**NOTE:** Must be built with a docker version > 18.06 + +The `Dockerfile` is supplied to build images with CUDA 11.1 support and cuDNN v8. +You can pass `PYTHON_VERSION=x.y` make variable to specify which Python version is to be used by Miniconda, or leave it +unset to use the default. + +```bash +make -f docker.Makefile +# images are tagged as docker.io/${your_docker_username}/pytorch +``` + +You can also pass the `CMAKE_VARS="..."` environment variable to specify additional CMake variables to be passed to CMake during the build. +See [setup.py](./setup.py) for the list of available variables. + +```bash +CMAKE_VARS="BUILD_CAFFE2=ON BUILD_CAFFE2_OPS=ON" make -f docker.Makefile +``` + +### Building the Documentation + +To build documentation in various formats, you will need [Sphinx](http://www.sphinx-doc.org) and the +readthedocs theme. + +```bash +cd docs/ +pip install -r requirements.txt +``` +You can then build the documentation by running `make ` from the +`docs/` folder. Run `make` to get a list of all available output formats. + +If you get a katex error run `npm install katex`. If it persists, try +`npm install -g katex` + +> Note: if you installed `nodejs` with a different package manager (e.g., +`conda`) then `npm` will probably install a version of `katex` that is not +compatible with your version of `nodejs` and doc builds will fail. +A combination of versions that is known to work is `node@6.13.1` and +`katex@0.13.18`. To install the latter with `npm` you can run +```npm install -g katex@0.13.18``` + +### Previous Versions + +Installation instructions and binaries for previous PyTorch versions may be found +on [our website](https://pytorch.org/previous-versions). + + +## Getting Started + +Three-pointers to get you started: +- [Tutorials: get you started with understanding and using PyTorch](https://pytorch.org/tutorials/) +- [Examples: easy to understand PyTorch code across all domains](https://github.com/pytorch/examples) +- [The API Reference](https://pytorch.org/docs/) +- [Glossary](https://github.com/pytorch/pytorch/blob/main/GLOSSARY.md) + +## Resources + +* [PyTorch.org](https://pytorch.org/) +* [PyTorch Tutorials](https://pytorch.org/tutorials/) +* [PyTorch Examples](https://github.com/pytorch/examples) +* [PyTorch Models](https://pytorch.org/hub/) +* [Intro to Deep Learning with PyTorch from Udacity](https://www.udacity.com/course/deep-learning-pytorch--ud188) +* [Intro to Machine Learning with PyTorch from Udacity](https://www.udacity.com/course/intro-to-machine-learning-nanodegree--nd229) +* [Deep Neural Networks with PyTorch from Coursera](https://www.coursera.org/learn/deep-neural-networks-with-pytorch) +* [PyTorch Twitter](https://twitter.com/PyTorch) +* [PyTorch Blog](https://pytorch.org/blog/) +* [PyTorch YouTube](https://www.youtube.com/channel/UCWXI5YeOsh03QvJ59PMaXFw) + +## Communication +* Forums: Discuss implementations, research, etc. https://discuss.pytorch.org +* GitHub Issues: Bug reports, feature requests, install issues, RFCs, thoughts, etc. +* Slack: The [PyTorch Slack](https://pytorch.slack.com/) hosts a primary audience of moderate to experienced PyTorch users and developers for general chat, online discussions, collaboration, etc. If you are a beginner looking for help, the primary medium is [PyTorch Forums](https://discuss.pytorch.org). If you need a slack invite, please fill this form: https://goo.gl/forms/PP1AGvNHpSaJP8to1 +* Newsletter: No-noise, a one-way email newsletter with important announcements about PyTorch. You can sign-up here: https://eepurl.com/cbG0rv +* Facebook Page: Important announcements about PyTorch. https://www.facebook.com/pytorch +* For brand guidelines, please visit our website at [pytorch.org](https://pytorch.org/) + +## Releases and Contributing + +Typically, PyTorch has three minor releases a year. Please let us know if you encounter a bug by [filing an issue](https://github.com/pytorch/pytorch/issues). + +We appreciate all contributions. If you are planning to contribute back bug-fixes, please do so without any further discussion. + +If you plan to contribute new features, utility functions, or extensions to the core, please first open an issue and discuss the feature with us. +Sending a PR without discussion might end up resulting in a rejected PR because we might be taking the core in a different direction than you might be aware of. + +To learn more about making a contribution to Pytorch, please see our [Contribution page](CONTRIBUTING.md). For more information about PyTorch releases, see [Release page](RELEASE.md). + +## The Team + +PyTorch is a community-driven project with several skillful engineers and researchers contributing to it. + +PyTorch is currently maintained by [Soumith Chintala](http://soumith.ch), [Gregory Chanan](https://github.com/gchanan), [Dmytro Dzhulgakov](https://github.com/dzhulgakov), [Edward Yang](https://github.com/ezyang), and [Nikita Shulga](https://github.com/malfet) with major contributions coming from hundreds of talented individuals in various forms and means. +A non-exhaustive but growing list needs to mention: Trevor Killeen, Sasank Chilamkurthy, Sergey Zagoruyko, Adam Lerer, Francisco Massa, Alykhan Tejani, Luca Antiga, Alban Desmaison, Andreas Koepf, James Bradbury, Zeming Lin, Yuandong Tian, Guillaume Lample, Marat Dukhan, Natalia Gimelshein, Christian Sarofeen, Martin Raison, Edward Yang, Zachary Devito. + +Note: This project is unrelated to [hughperkins/pytorch](https://github.com/hughperkins/pytorch) with the same name. Hugh is a valuable contributor to the Torch community and has helped with many things Torch and PyTorch. + +## License + +PyTorch has a BSD-style license, as found in the [LICENSE](LICENSE) file. diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/NOTICE b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/NOTICE new file mode 100644 index 0000000000000000000000000000000000000000..6effb8b5d70709f90835f2f5d646352fd77b6943 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/NOTICE @@ -0,0 +1,456 @@ +======================================================================= +Software under third_party +======================================================================= +Software libraries under third_party are provided as github submodule +links, and their content is not part of the Caffe2 codebase. Their +licences can be found under the respective software repositories. + +======================================================================= +Earlier BSD License +======================================================================= +Early development of Caffe2 in 2015 and early 2016 is licensed under the +BSD license. The license is attached below: + +All contributions by Facebook: +Copyright (c) 2016 Facebook Inc. + +All contributions by Google: +Copyright (c) 2015 Google Inc. +All rights reserved. + +All contributions by Yangqing Jia: +Copyright (c) 2015 Yangqing Jia +All rights reserved. + +All contributions by Kakao Brain: +Copyright 2019-2020 Kakao Brain + +All other contributions: +Copyright(c) 2015, 2016 the respective contributors +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +1. Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. +2. Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND +ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED +WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR +ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES +(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; +LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND +ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + +======================================================================= +Caffe's BSD License +======================================================================= +Some parts of the caffe2 code is derived from the original Caffe code, which is +created by Yangqing Jia and is now a BSD-licensed open-source project. The Caffe +license is as follows: + +COPYRIGHT + +All contributions by the University of California: +Copyright (c) 2014, The Regents of the University of California (Regents) +All rights reserved. + +All other contributions: +Copyright (c) 2014, the respective contributors +All rights reserved. + +Caffe uses a shared copyright model: each contributor holds copyright over +their contributions to Caffe. The project versioning records all such +contribution and copyright details. If a contributor wants to further mark +their specific copyright on a particular contribution, they should indicate +their copyright solely in the commit message of the change when it is +committed. + +LICENSE + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are met: + +1. Redistributions of source code must retain the above copyright notice, this + list of conditions and the following disclaimer. +2. Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND +ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED +WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE +DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR +ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES +(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; +LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND +ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +CONTRIBUTION AGREEMENT + +By contributing to the BVLC/caffe repository through pull-request, comment, +or otherwise, the contributor releases their content to the +license and copyright terms herein. + +======================================================================= +Caffe2's Apache License +======================================================================= + +This repo contains Caffe2 code, which was previously licensed under +Apache License Version 2.0: + + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + +======================================================================= +Cephes's 3-Clause BSD License +======================================================================= + +Code derived from implementations in the Cephes Math Library should mention +its derivation and reference the following license: + + 3-Clause BSD License for the Cephes Math Library + Copyright (c) 2018, Steven Moshier + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + * Neither the name of the nor the + names of its contributors may be used to endorse or promote products + derived from this software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + DISCLAIMED. IN NO EVENT SHALL Steven Moshier BE LIABLE FOR ANY + DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + +======================================================================= +SciPy's 3-Clause BSD License +======================================================================= + +Code derived from implementations in SciPy should mention its derivation +and reference the following license: + + Copyright (c) 2001-2002 Enthought, Inc. 2003-2019, SciPy Developers. + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + 2. Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + + 3. Neither the name of the copyright holder nor the names of its + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +======================================================================= +Boost's 1.0 Software License +======================================================================= + +Code derived from implementations in Boost 1.0 should mention its +derivation and reference the following license: + + Boost Software License - Version 1.0 - August 17th, 2003 + + Permission is hereby granted, free of charge, to any person or organization + obtaining a copy of the software and accompanying documentation covered by + this license (the "Software") to use, reproduce, display, distribute, + execute, and transmit the Software, and to prepare derivative works of the + Software, and to permit third-parties to whom the Software is furnished to + do so, all subject to the following: + + The copyright notices in the Software and this entire statement, including + the above license grant, this restriction and the following disclaimer, + must be included in all copies of the Software, in whole or in part, and + all derivative works of the Software, unless such copies or derivative + works are solely in the form of machine-executable object code generated by + a source language processor. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT + SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE + FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, + ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + DEALINGS IN THE SOFTWARE. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. + +======================================================================= +PILLOW-SIMD Software License +======================================================================= + +Code derived from implementations in PILLOW-SIMD should mention its derivation +and reference the following license: + + The Python Imaging Library (PIL) is + + Copyright © 1997-2011 by Secret Labs AB + Copyright © 1995-2011 by Fredrik Lundh + + Pillow is the friendly PIL fork. It is + + Copyright © 2010-2022 by Alex Clark and contributors + + Like PIL, Pillow is licensed under the open source HPND License: + + By obtaining, using, and/or copying this software and/or its associated + documentation, you agree that you have read, understood, and will comply + with the following terms and conditions: + + Permission to use, copy, modify, and distribute this software and its + associated documentation for any purpose and without fee is hereby granted, + provided that the above copyright notice appears in all copies, and that + both that copyright notice and this permission notice appear in supporting + documentation, and that the name of Secret Labs AB or the author not be + used in advertising or publicity pertaining to distribution of the software + without specific, written prior permission. + + SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS + SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. + IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR BE LIABLE FOR ANY SPECIAL, + INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM + LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE + OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR + PERFORMANCE OF THIS SOFTWARE. diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/entry_points.txt b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/entry_points.txt new file mode 100644 index 0000000000000000000000000000000000000000..430fb81fbc380b225a3895356ccb54ecaf9e2494 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/entry_points.txt @@ -0,0 +1,7 @@ +[console_scripts] +convert-caffe2-to-onnx = caffe2.python.onnx.bin.conversion:caffe2_to_onnx +convert-onnx-to-caffe2 = caffe2.python.onnx.bin.conversion:onnx_to_caffe2 +torchrun = torch.distributed.run:main + +[torchrun.logs_specs] +default = torch.distributed.elastic.multiprocessing:DefaultLogsSpecs diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/top_level.txt b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/top_level.txt new file mode 100644 index 0000000000000000000000000000000000000000..90d81bec1d35eb3996334268974885a5398b6c6c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch-2.3.0+cu118.dist-info/top_level.txt @@ -0,0 +1,3 @@ +functorch +torch +torchgen diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/__init__.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/__init__.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..47ed28ac53f5ab1fe9342bdf4ed0f2a46f5d95a6 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/__init__.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/lazy_ir.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/lazy_ir.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..0288104cbb9c1be7f6eda0f164bf1f90e08f29aa Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/lazy_ir.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/native_functions.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/native_functions.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..749173fe0693dc9b4738592b8a925ec8d45b80cc Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/native_functions.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/register_dispatch_key.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/register_dispatch_key.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..ed0d76dcba2454cc9c48b154b678fa34132f38dd Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/dest/__pycache__/register_dispatch_key.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/__pycache__/model.cpython-311.pyc b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/__pycache__/model.cpython-311.pyc new file mode 100644 index 0000000000000000000000000000000000000000..05a70c5f6b46a550c892d54353555893d70b44e6 Binary files /dev/null and b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/__pycache__/model.cpython-311.pyc differ diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/__init__.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/types/signatures.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/types/signatures.py new file mode 100644 index 0000000000000000000000000000000000000000..a53d15c036a9106e865f4665945ab3b9cf0de6e6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/types/signatures.py @@ -0,0 +1,73 @@ +from dataclasses import dataclass +from typing import List, Optional, Set + +import torchgen.api.cpp as aten_cpp + +from torchgen.api.types import Binding, CType +from torchgen.model import FunctionSchema, NativeFunction + +from .types import contextArg + + +@dataclass(frozen=True) +class ExecutorchCppSignature: + """ + This signature is merely a CppSignature with Executorch types (optionally + contains KernelRuntimeContext as well). The inline definition of + CppSignature is generated in Functions.h and it's used by unboxing + functions. + """ + + # The schema this signature is derived from + func: FunctionSchema + + # The set of C++ arguments which should not have defaults applied to them + cpp_no_default_args: Set[str] + + # Allows you to prepend an arbitrary prefix to the signature name. + # This is useful for parts of the codegen that generate wrappers around kernels, + # and need to avoid naming collisions. + prefix: str = "" + + def arguments(self, *, include_context: bool = True) -> List[Binding]: + return ([contextArg] if include_context else []) + et_cpp.arguments( + self.func.arguments, + faithful=True, # always faithful, out argument at the end + method=False, # method not supported + cpp_no_default_args=self.cpp_no_default_args, + ) + + def name(self) -> str: + return self.prefix + aten_cpp.name( + self.func, + faithful_name_for_out_overloads=True, + ) + + def decl(self, name: Optional[str] = None, *, include_context: bool = True) -> str: + args_str = ", ".join( + a.decl() for a in self.arguments(include_context=include_context) + ) + if name is None: + name = self.name() + return f"{self.returns_type().cpp_type()} {name}({args_str})" + + def defn(self, name: Optional[str] = None) -> str: + args = [a.defn() for a in self.arguments()] + args_str = ", ".join(args) + if name is None: + name = self.name() + return f"{self.returns_type().cpp_type()} {name}({args_str})" + + def returns_type(self) -> CType: + return et_cpp.returns_type(self.func.returns) + + @staticmethod + def from_native_function( + f: NativeFunction, *, prefix: str = "" + ) -> "ExecutorchCppSignature": + return ExecutorchCppSignature( + func=f.func, prefix=prefix, cpp_no_default_args=f.cpp_no_default_args + ) + + +from torchgen.executorch.api import et_cpp diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/types/types.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/types/types.py new file mode 100644 index 0000000000000000000000000000000000000000..c9db1baa245fa2375896930febeddcd98ae2d4e7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/types/types.py @@ -0,0 +1,81 @@ +from dataclasses import dataclass +from typing import Dict + +from torchgen.api.types import ( + BaseCppType, + BaseCType, + Binding, + boolT, + CType, + doubleT, + Expr, + longT, + MutRefCType, + NamedCType, +) +from torchgen.model import BaseTy + +halfT = BaseCppType("torch::executor", "Half") +bfloat16T = BaseCppType("torch::executor", "BFloat16") +stringT = BaseCppType("torch::executor", "string_view") +scalarTypeT = BaseCppType("torch::executor", "ScalarType") +tensorT = BaseCppType("torch::executor", "Tensor") +tensorListT = BaseCppType("torch::executor", "TensorList") +scalarT = BaseCppType("torch::executor", "Scalar") +memoryFormatT = BaseCppType("torch::executor", "MemoryFormat") +intArrayRefT = BaseCppType("torch::executor", "IntArrayRef") +optionalT = BaseCppType("torch::executor", "optional") +contextT = BaseCppType("torch::executor", "KernelRuntimeContext") + +contextExpr = Expr( + expr="context", + type=NamedCType(name="context", type=MutRefCType(BaseCType(contextT))), +) + +contextArg = Binding( + name="context", + nctype=contextExpr.type, + argument=None, # type: ignore[arg-type] + default=None, +) + +BaseTypeToCppMapping: Dict[BaseTy, BaseCppType] = { + BaseTy.int: longT, + BaseTy.float: doubleT, + BaseTy.bool: boolT, + BaseTy.str: stringT, + BaseTy.ScalarType: scalarTypeT, + BaseTy.Tensor: tensorT, + BaseTy.Scalar: scalarT, + BaseTy.MemoryFormat: memoryFormatT, +} + + +@dataclass(frozen=True) +class OptionalCType(CType): + elem: "CType" + + def cpp_type(self, *, strip_ref: bool = False) -> str: + # Do not pass `strip_ref` recursively. + return f"torch::executor::optional<{self.elem.cpp_type()}>" + + def cpp_type_registration_declarations(self) -> str: + return f"torch::executor::optional<{self.elem.cpp_type_registration_declarations()}>" + + def remove_const_ref(self) -> "CType": + return OptionalCType(self.elem.remove_const_ref()) + + +@dataclass(frozen=True) +class ArrayRefCType(CType): + elem: "CType" + + def cpp_type(self, *, strip_ref: bool = False) -> str: + # Do not pass `strip_ref` recursively. + return f"torch::executor::ArrayRef<{self.elem.cpp_type()}>" + + def cpp_type_registration_declarations(self) -> str: + return f"torch::executor::ArrayRef<{self.elem.cpp_type_registration_declarations()}>" + + def remove_const_ref(self) -> "CType": + return ArrayRefCType(self.elem.remove_const_ref()) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/unboxing.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/unboxing.py new file mode 100644 index 0000000000000000000000000000000000000000..9a8f717ddbb28d970779d2247d84c58450c5de45 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/api/unboxing.py @@ -0,0 +1,213 @@ +from dataclasses import dataclass +from typing import Callable, List, Sequence, Tuple + +from torchgen.api.types import Binding, CType, NamedCType +from torchgen.model import ( + Argument, + BaseTy, + BaseType, + ListType, + NativeFunction, + OptionalType, + Type, +) + +connector = "\n\t" + + +# Return unboxing function name for a NativeFunction +def name(f: NativeFunction) -> str: + return f.func.name.unambiguous_name() + + +@dataclass(frozen=True) +class Unboxing: + """ + Takes a sequence of Bindings and unbox EValues to these Bindings. Return generated code that performs correct unboxing. + A sample generated code: + // aten::mul.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!) + void mul_out(EValue** stack) { + EValue& self = *stack[0]; + EValue& other = *stack[1]; + EValue& out = *stack[2]; + const torch::executor::Tensor & self_base = self.to(); + const torch::executor::Tensor & other_base = other.to(); + torch::executor::Tensor & out_base = out.to(); + + EXECUTORCH_SCOPE_PROF("native_call_mul.out"); + torch::executor::mul_outf(self_base, other_base, out_base); + + + } + """ + + # this is a callable that converts a JIT argument, into its C++ type. + # Translates (type, mutability, binds) to NamedCType. E.g., torchgen.api.cpp.argumenttype_type. + argument_type_gen: Callable[ + ..., + NamedCType, + ] + + # Convert all the arguments in a NativeFunction to C++ code + def convert_arguments( + self, args: Sequence[Binding] + ) -> Tuple[List[Binding], List[str]]: + code_list = [f"EValue& {args[i].name} = *stack[{i}];" for i in range(len(args))] + binding_list = [] + for arg in args: + # expecting only Argument + if not isinstance(arg.argument, Argument): + raise Exception( + f"Unexpected argument type, expecting `Argument` but got {arg}" + ) + argument: Argument = arg.argument + unboxed_name, _, code, decl = self.argumenttype_evalue_convert( + argument.type, argument.name, mutable=argument.is_write + ) + code_list.extend(decl) + code_list.extend(code) + binding_list.append(arg.with_name(unboxed_name)) + return binding_list, code_list + + def argumenttype_evalue_convert( + self, t: Type, arg_name: str, *, mutable: bool = False + ) -> Tuple[str, CType, List[str], List[str]]: + """ + Takes in the type, name and mutability corresponding to an argument, and generates a tuple of: + (1) the C++ code necessary to unbox the argument + (2) A Binding corresponding to the newly created unboxed variable, including variable name and its CType + :param t: a `Type` of an argument + :param arg_name: argument name + :param mutable: boolean for whether this argument type is mutable + :return: unboxed result + """ + ctype = self.argument_type_gen(t, mutable=mutable, binds=arg_name).type + + if isinstance(t, BaseType): + out_name = f"{arg_name}_base" + code, decl = self._gen_code_base_type( + arg_name=arg_name, out_name=out_name, ctype=ctype + ) + elif isinstance(t, OptionalType): + out_name = f"{arg_name}_opt_out" + code, decl = self._gen_code_optional_type( + arg_name=arg_name, out_name=out_name, t=t, ctype=ctype + ) + elif isinstance(t, ListType): + out_name = f"{arg_name}_list_out" + code, decl = self._gen_code_list_type( + arg_name=arg_name, out_name=out_name, t=t, ctype=ctype + ) + else: + raise Exception(f"Cannot handle type {t}. arg_name: {arg_name}") + return out_name, ctype, code, decl + + def _gen_code_base_type( + self, arg_name: str, out_name: str, ctype: CType + ) -> Tuple[List[str], List[str]]: + return [ + f"{ctype.cpp_type()} {out_name} = {arg_name}.to<{ctype.cpp_type(strip_ref=True)}>();" + ], [] + + def _gen_code_optional_type( + self, arg_name: str, out_name: str, t: OptionalType, ctype: CType + ) -> Tuple[List[str], List[str]]: + in_name = f"{arg_name}_opt_in" + res_name, base_type, res_code, decl = self.argumenttype_evalue_convert( + t.elem, in_name + ) + return ( + f""" + {ctype.cpp_type(strip_ref=True)} {out_name} = {arg_name}.toOptional<{base_type.cpp_type(strip_ref=True)}>(); + """.split( + "\n" + ), + decl, + ) + + def _gen_code_list_type( + self, arg_name: str, out_name: str, t: ListType, ctype: CType + ) -> Tuple[List[str], List[str]]: + in_name = f"{arg_name}_list_in" + elem_name = f"{arg_name}_elem" + code = [] + res_name, res_ctype, res_code, decl = self.argumenttype_evalue_convert( + t.elem, elem_name + ) + + if isinstance(t.elem, BaseType) and t.elem.name == BaseTy.Tensor: + code.extend( + f""" + {ctype.cpp_type(strip_ref=True)} {out_name} = {arg_name}.toTensorList(); + """.split( + "\n" + ) + ) + elif isinstance(t.elem, BaseType) and ( + t.elem.name == BaseTy.int or t.elem.name == BaseTy.SymInt + ): + code.extend( + f""" + {ctype.cpp_type(strip_ref=True)} {out_name} = {arg_name}.toIntList(); + """.split( + "\n" + ) + ) + elif isinstance(t.elem, BaseType) and t.elem.name == BaseTy.float: + code.extend( + f""" + {ctype.cpp_type(strip_ref=True)} {out_name} = {arg_name}.toDoubleList(); + """.split( + "\n" + ) + ) + elif isinstance(t.elem, BaseType) and t.elem.name == BaseTy.bool: + # handle list type with size, e.g., bool[4] + code.extend( + f""" + {ctype.cpp_type(strip_ref=True)} {out_name} = {arg_name}.toBoolList(); + """.split( + "\n" + ) + ) + # pytorch codegen: + # we have to use c10::List for optional element. e.g., Tensor?[] -> c10::List> + elif ( + isinstance(t.elem, OptionalType) + and isinstance(t.elem.elem, BaseType) + and t.elem.elem.name == BaseTy.Tensor + ): + code.extend( + f""" +#ifdef USE_ATEN_LIB +at::ArrayRef> {in_name} = {arg_name}.toListOptionalTensor(); +c10::List> {out_name}; +for (auto {elem_name}: {in_name}) {{ + {out_name}.push_back({elem_name}); +}} +#else +torch::executor::ArrayRef> {out_name} = {arg_name}.toListOptionalTensor(); +#endif + """.split( + "\n" + ) + ) + else: + # use ArrayRef as default. + vec_name = arg_name + "_vec" + # need to bring vector instantiation out of scope so that ArrayRef has valid data + decl.append( + f"std::vector<{res_ctype.cpp_type(strip_ref=True)}> {vec_name};" + ) + code.extend( + f""" + for (EValue {elem_name}: {in_name}) {{ + {connector.join(res_code)} + {vec_name}.push_back({res_name}); + }} + {ctype.cpp_type(strip_ref=True)} {out_name}({vec_name}); + """.split( + "\n" + ) + ) + return code, decl diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/model.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/model.py new file mode 100644 index 0000000000000000000000000000000000000000..cec9251a3187cfe0a1a3e84744f49760331761f2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/executorch/model.py @@ -0,0 +1,220 @@ +# Represents all kernels used by an Executorch model. +# It maintains a Dict[OperatorName, Dict[ETKernelKey, BackendMetadata]] structure. + +import itertools +from collections import defaultdict, namedtuple +from dataclasses import dataclass +from enum import IntEnum +from typing import Dict, List, Tuple, Union + +from torchgen.model import ( + BackendIndex, + BackendMetadata, + DispatchKey, + NativeFunction, + NativeFunctionsGroup, + OperatorName, +) +from torchgen.utils import assert_never + +KERNEL_KEY_VERSION = 1 + + +# TODO: Duplicated Subset from codegen.tool.gen_oplist, remove declaration in codegen +class ScalarType(IntEnum): + Byte = 0 + Char = 1 + Short = 2 + Int = 3 + Long = 4 + Float = 6 + Double = 7 + Bool = 11 + + +ETParsedYaml = namedtuple("ETParsedYaml", ["native_functions", "kernel_index"]) + + +@dataclass(frozen=True) +class ETKernelKeyOpArgMeta: + arg_name: str + dtype: str + # The order of the dimensions if entry is a Tensor + dim_order: Tuple[int, ...] + + def to_native_string(self) -> str: + dtype_str = ScalarType[self.dtype].value + dim_str = str(self.dim_order)[1:-1].replace(" ", "") + return f"{dtype_str};{dim_str}" + + +@dataclass(frozen=True) +class ETKernelKey: + # Field undefined is default = True + arg_meta: Tuple[ETKernelKeyOpArgMeta, ...] = () + + # Indicator for this kernel being used as a catch all + default: bool = False + + version: int = KERNEL_KEY_VERSION + + @staticmethod + def gen_from_yaml( + args: Dict[str, Tuple[str, str]], + type_alias_map: Dict[str, List[str]], # TODO: Support unwrapped str val + dim_order_alias_map: Dict[str, List[int]], + ) -> List["ETKernelKey"]: + """Generate ETKernelKeys from arg kernel specs + Multiple ETKernelKeys are returned due to dtype permutations from utilizing + type_alias_map (actualizing each potential type permutation as a KernelKey) + + Args: + args: Mapping from argument name to kernel specs + Kernel specs are a tuple of (dtype, dim_order). + Currently tuple entries must be aliased via the alias map arguments + type_alias_map: Mapping from type alias to potential type enums + i.e { T0 : [Double, Int] } means T0 can be either Double or Int + Used for lookup by args + dim_order_alias_map: Mapping from alias to a list of dimension orders + Used for lookup by args + """ + # Cast to dim order to int + dim_order_alias_map = { + k: [int(alias) for alias in v] for k, v in dim_order_alias_map.items() + } + kernel_keys = [] + + # Get all used Dtype Alias + dtype_alias_used = set() + for type_alias, dim_order in args.values(): + # Enforce usage of alias initially + # TODO: Support inlined arguments + assert type_alias in type_alias_map, "Undefined type alias: " + str( + type_alias + ) + assert ( + dim_order in dim_order_alias_map + ), "Undefined dim_order alias: " + str(dim_order) + dtype_alias_used.add(type_alias) + + # Generate all permutations of dtype alias values + alias_dtypes = [ + [(alias, dtype) for dtype in type_alias_map[alias]] + for alias in dtype_alias_used + ] + alias_permutations = [ + dict(permutation) for permutation in list(itertools.product(*alias_dtypes)) + ] + + # Using each alias value permutation, generate kernel keys + op_arg_cache = {} + for permutation in alias_permutations: + arg_list = [] + for arg_name, arg_spec in args.items(): + dtype = permutation[arg_spec[0]] + dim_order = dim_order_alias_map[arg_spec[1]] # type: ignore[assignment] + if ( + cache_key := (arg_name, dtype, tuple(dim_order)) + ) not in op_arg_cache: + op_arg_cache[cache_key] = ETKernelKeyOpArgMeta(*cache_key) # type: ignore[arg-type] + + arg_list.append(op_arg_cache[cache_key]) + kernel_keys.append(ETKernelKey(tuple(arg_list))) + + return kernel_keys + + def to_native_string(self) -> str: + if self.default: + return "default" + return ( + "v" + + str(KERNEL_KEY_VERSION) + + "/" + + "|".join([arg.to_native_string() for arg in self.arg_meta]) + ) + + +@dataclass(frozen=True) +class ETKernelIndex: + index: Dict[OperatorName, Dict[ETKernelKey, BackendMetadata]] + + def has_kernels(self, g: Union[NativeFunction, NativeFunctionsGroup]) -> bool: + m = self.get_kernels(g) + return m is not None + + def get_kernels( + self, g: Union[NativeFunction, NativeFunctionsGroup] + ) -> Dict[ETKernelKey, BackendMetadata]: + if isinstance(g, NativeFunction): + f = g + elif isinstance(g, NativeFunctionsGroup): + f = g.functional + else: + assert_never(g) + if f.func.name not in self.index: + return {} + return self.index[f.func.name] + + @staticmethod + def grow_from_backend_indices( + kernel_index: Dict[OperatorName, Dict[ETKernelKey, BackendMetadata]], + backend_indices: Dict[DispatchKey, Dict[OperatorName, BackendMetadata]], + ) -> None: + for dk in backend_indices: + index = backend_indices[dk] + for op, backend_metadata in index.items(): + if op in kernel_index: + kernel_index[op][ETKernelKey(default=True)] = backend_metadata + else: + kernel_index[op] = {ETKernelKey(default=True): backend_metadata} + + @staticmethod + def from_backend_indices( + backend_indices: Dict[DispatchKey, Dict[OperatorName, BackendMetadata]] + ) -> "ETKernelIndex": + kernel_index: Dict[ + OperatorName, Dict[ETKernelKey, BackendMetadata] + ] = defaultdict(dict) + ETKernelIndex.grow_from_backend_indices(kernel_index, backend_indices) + return ETKernelIndex(kernel_index) + + def grow( + self, backend_indices: Dict[DispatchKey, Dict[OperatorName, BackendMetadata]] + ) -> "ETKernelIndex": + ETKernelIndex.grow_from_backend_indices(self.index, backend_indices) + return self + + def _to_backend_index(self) -> BackendIndex: + """ + WARNING: this will be deprecated once all the codegen places know how to handle ETKernelIndex. + """ + index: Dict[OperatorName, BackendMetadata] = {} + for op in self.index: + kernel_dict = self.index[op] + assert ( + len(kernel_dict.values()) == 1 + ), f"Can't convert ETKernelIndex to BackendIndex because {op} has more than one kernels. Got {kernel_dict}" + index[op] = kernel_dict.get( + ETKernelKey(default=True), + BackendMetadata(kernel="", structured=False, cpp_namespace=""), + ) + return BackendIndex( + dispatch_key=DispatchKey.CPU, + use_out_as_primary=False, + device_guard=False, + external=False, + index=index, + ) + + # Note duplicate ETKernelKey from index_b will clobber the metadata from index_a + @staticmethod + def merge_indices( + index_a: "ETKernelIndex", index_b: "ETKernelIndex" + ) -> "ETKernelIndex": + combined = defaultdict(dict, index_a.index.copy()) + + for op, entry in index_b.index.items(): + for key, metadata in entry.items(): + combined[op][key] = metadata + + return ETKernelIndex(combined) diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/gen_aoti_c_shim.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/gen_aoti_c_shim.py new file mode 100644 index 0000000000000000000000000000000000000000..af16258b39605c7d7ae6bf6ee542f0be18588dc8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/gen_aoti_c_shim.py @@ -0,0 +1,431 @@ +import textwrap +from dataclasses import dataclass +from typing import Dict, List, Optional, Sequence, Tuple, Union + +from torchgen.api.types import DispatcherSignature +from torchgen.api.types.signatures import CppSignature, CppSignatureGroup + +from torchgen.context import method_with_native_function +from torchgen.model import ( + Argument, + BackendIndex, + BaseTy, + BaseType, + DispatchKey, + FunctionSchema, + ListType, + NativeFunction, + OptionalType, + Type, +) +from torchgen.utils import mapMaybe + + +def returns_are_all_tensor(schema: FunctionSchema) -> bool: + return len(schema.returns) != 0 and all( + ret.type.is_tensor_like() for ret in schema.returns + ) + + +base_type_to_c_type = { + BaseTy.Tensor: "AtenTensorHandle", + BaseTy.bool: "int32_t", # Use int to pass bool + BaseTy.int: "int64_t", + BaseTy.SymInt: "int64_t", # Inductor-generated code won't see a SymInt + BaseTy.Scalar: "double", # Use double to pass both integer and floating point + BaseTy.float: "double", # TODO: how about other floating point types? + BaseTy.str: "const char*", + BaseTy.DeviceIndex: "int32_t", + BaseTy.Layout: "int32_t", # Represent enum as int + BaseTy.MemoryFormat: "int32_t", # Represent enum as int + BaseTy.ScalarType: "int32_t", # Represent enum as int +} + +base_type_to_aten_type = { + BaseTy.Tensor: "at::Tensor", + BaseTy.bool: "bool", + BaseTy.int: "int64_t", + BaseTy.SymInt: "c10::SymInt", + BaseTy.Scalar: "c10::Scalar", + BaseTy.float: "double", + BaseTy.str: "c10::string_view", + BaseTy.DeviceIndex: "c10::DeviceIndex", + BaseTy.Layout: "c10::Layout", + BaseTy.MemoryFormat: "c10::MemoryFormat", + BaseTy.ScalarType: "c10::ScalarType", +} + +base_type_to_callsite_expr = { + BaseTy.Tensor: "*tensor_handle_to_tensor_pointer", + BaseTy.bool: "", + BaseTy.int: "", + BaseTy.SymInt: "", + BaseTy.Scalar: "", + BaseTy.float: "", + BaseTy.str: "", + BaseTy.DeviceIndex: "static_cast", + BaseTy.Layout: "static_cast", + BaseTy.MemoryFormat: "static_cast", + BaseTy.ScalarType: "static_cast", +} + + +# convert args to C types, names in declarations, and expressions in function bodies +def convert_arg_type_and_name(typ: Type, name: str) -> Tuple[List[str], List[str], List[str], List[str]]: # type: ignore[return] + if isinstance(typ, BaseType): + if typ.name in base_type_to_c_type: + return ( + [base_type_to_c_type[typ.name]], + [name], + [base_type_to_aten_type[typ.name]], + [ + f"{base_type_to_callsite_expr[typ.name]}({name})" + if base_type_to_callsite_expr[typ.name] + else name + ], + ) + elif typ.name == BaseTy.Device: + return ( + ["int32_t", "int32_t"], + [name, name + "_index_"], + ["c10::Device"], + [ + f"c10::Device(static_cast({name}), static_cast({name}_index_))" + ], + ) + else: + # TODO: BaseTy.Dimname, BaseTy.Generator, etc. + raise NotImplementedError(f"TODO: add support for arg type {repr(typ)}") + elif isinstance(typ, OptionalType): + c_types, names, aten_types, callsite_exprs = convert_arg_type_and_name( + typ.elem, name + ) + j = 0 # index for names + new_aten_types = [] + new_callsite_exprs = [] + for i, aten_type in enumerate(aten_types): + # Use pointer to denote optional type + c_types[j] = c_types[j] + "*" + if aten_type.startswith("c10::ArrayRef<"): + # ArrayRef is passed as pointer + size, but no need to add "*" to the size argument + new_aten_types.append(f"c10::optional<{aten_type}>") + base_type = aten_type[len("c10::ArrayRef<") : -1] + new_callsite_exprs.append( + f"pointer_to_optional_list<{base_type}>({names[j]}, {names[j+1]})" + ) + j += 2 + elif aten_type == "c10::Device": + # Device is passed as device_type + device_index + new_aten_types.append("c10::optional") + new_callsite_exprs.append( + f"pointer_to_optional_device({names[j]}, {names[j+1]})" + ) + j += 2 + else: + new_aten_types.append(f"c10::optional<{aten_type}>") + new_callsite_exprs.append( + f"pointer_to_optional<{aten_type}>({names[j]})" + ) + j += 1 + + return ( + c_types, + names, + new_aten_types, + new_callsite_exprs, + ) + elif isinstance(typ, ListType): + # Need to explictly pass the list as pointer + length + c_types, names, aten_types, _ = convert_arg_type_and_name(typ.elem, name) + assert len(c_types) == 1, "ListType with unsupported element type " + repr(typ) + + # The list content should never be modified + c_types[0] = f"const {c_types[0]}*" + c_types.append("int64_t") + name = names[0] + names.append(name + "_len_") + + atype = aten_types[0] + callsite_exprs = [] + if atype == "bool": + # no converter from std::vector to c10::ArrayRef + # construct std::array instead + assert typ.size is not None + callsite_exprs.append(f"pointer_to_list<{typ.size}>({name})") + elif atype == "c10::optional": + # convert from std::vector> to c10::List> + callsite_exprs.append( + f"c10::List<{atype}>(c10::ArrayRef<{atype}>(pointer_to_list<{atype}>({name}, {name}_len_)))" + ) + else: + callsite_exprs.append(f"pointer_to_list<{atype}>({name}, {name}_len_)") + + aten_types = [f"c10::ArrayRef<{t}>" for t in aten_types] + return ( + c_types, + names, + aten_types, + callsite_exprs, + ) + + +def zip_type_and_name(types: List[str], names: List[str]) -> List[str]: + return [typ + " " + name for typ, name in zip(types, names)] + + +# Generate argument declarations and callsite expressions +def gen_arguments(flat_arguments: Sequence[Argument]) -> Tuple[List[str], List[str]]: + types = [] + new_names = [] + callsite_exprs = [] + for arg in flat_arguments: + new_types, names, _, new_callsite_exprs = convert_arg_type_and_name( + arg.type, arg.name + ) + types.extend(new_types) + new_names.extend(names) + callsite_exprs.extend(new_callsite_exprs) + return zip_type_and_name(types, new_names), callsite_exprs + + +# Return values are passed out as pointer arguments because all the C shim functions +# are expected to return AOTITorchError. +# Generate returns as declarations and callsite expressions +def gen_returns(schema: FunctionSchema) -> Tuple[List[str], List[str]]: + types = [] + names = [] + for idx, ret in enumerate(schema.returns): + names.append(f"ret{idx}") + if isinstance(ret.type, BaseType) and ret.type.name in base_type_to_c_type: + types.append(base_type_to_c_type[ret.type.name] + "*") + else: + raise NotImplementedError( + f"TODO: add support for return type {repr(ret.type)}" + ) + + def convert_return(typ: BaseType, val: str) -> str: + if typ.name == BaseTy.Tensor: + return f"new_tensor_handle(std::move({val}));" + elif typ.name == BaseTy.SymInt: + return f"{val}.expect_int()" + elif typ.name == BaseTy.Scalar: + return f"{val}.toDouble()" + else: + return val + + ret_pointer_can_be_null = False + unambiguous_name = schema.name.unambiguous_name() + for name in ["_scaled_dot_product_flash_attention"]: + if name in unambiguous_name: + ret_pointer_can_be_null = True + break + + callsite_exprs: List[str] = [] + for idx, ret in enumerate(schema.returns): + tmp = "tmp_result" if len(names) == 1 else f"std::get<{idx}>(tmp_result)" + assert isinstance(ret.type, BaseType) + rval = convert_return(ret.type, tmp) + if ret_pointer_can_be_null: + callsite_exprs.append(f"if ({names[idx]}) {{ *{names[idx]} = {rval}; }}") + else: + callsite_exprs.append(f"*{names[idx]} = {rval};") + + return zip_type_and_name(types, names), callsite_exprs + + +# gen.py generates header first and then src, so caching the result here to avoid duplicate work +declaration_definition_cache: Dict[Tuple[str, str, str], Tuple[str, str]] = {} + + +def gen_declaration_and_definition( + schema: FunctionSchema, device: str, backend_call: str +) -> Tuple[str, str]: + func_name = schema.name.unambiguous_name() + + global declaration_definition_cache + if (func_name, device, backend_call) in declaration_definition_cache: + return declaration_definition_cache[(func_name, device, backend_call)] + + if schema.is_out_fn(): + # out_variant has out arguments in the front, and it's ok to ignore return value + # because C shim functions only return AOTITorchError + # Somehow at::native out-variant functions have out arguments in the back + args, callsite_exprs = gen_arguments( + [*schema.arguments.flat_non_out, *schema.arguments.out] + if "at::native" in backend_call + else [*schema.arguments.out, *schema.arguments.flat_non_out], + ) + ret_assignments: List[str] = [] + else: + args, callsite_exprs = gen_arguments(schema.arguments.flat_all) + ret_declarations, ret_assignments = gen_returns(schema) + args.extend(ret_declarations) + + declaration = f"AOTITorchError aoti_torch_{device}_{func_name}({', '.join(args)})" + + tmp_result = "auto tmp_result = " if ret_assignments else "" + ret_assignments_str = "\n" + "\n".join(ret_assignments) if ret_assignments else "" + definition = f""" +{declaration} {{ + AOTI_TORCH_CONVERT_EXCEPTION_TO_ERROR_CODE({{ + {tmp_result}{backend_call}( +{textwrap.indent(', '.join(callsite_exprs), " ")} + );{textwrap.indent(ret_assignments_str, " ")} + }}); +}} +""" + declaration_definition_cache[(func_name, device, backend_call)] = ( + declaration, + definition, + ) + return declaration, definition + + +def gen_static_dispatch_backend_call_signature( + sig: Union[CppSignature, DispatcherSignature], + f: NativeFunction, +) -> CppSignature: + sig = DispatcherSignature.from_schema(f.func) + cpp_sigs = CppSignatureGroup.from_native_function( + f, method=False, fallback_binding=False + ) + if sig.symint and f.func.has_symint(): + cpp_sig = cpp_sigs.symint_signature + else: + cpp_sig = cpp_sigs.signature + assert cpp_sig is not None + return cpp_sig + + +def gen_static_dispatch_backend_call( + f: NativeFunction, + backend_index: BackendIndex, +) -> str: + assert backend_index.has_kernel(f) + sig = DispatcherSignature.from_schema(f.func) + cpp_sig = gen_static_dispatch_backend_call_signature(sig, f) + return f"at::{backend_index.dispatch_key.lower()}::{cpp_sig.name()}" + + +def get_backend_index_for_aoti( + f: NativeFunction, + dispatch_key: DispatchKey, + backend_indices: Dict[DispatchKey, BackendIndex], +) -> Optional[BackendIndex]: + if "pointwise" in f.tags: + # TODO: No need to generate C shim for Inductor lowered ops. + # Only skip pointwise kernels for now, and we can add more tags later. + return None + + backend_index = None + if backend_indices[dispatch_key].has_kernel(f): + backend_index = backend_indices[dispatch_key] + elif backend_indices[DispatchKey.CompositeExplicitAutograd].has_kernel(f): + # We need to create C shim wrappers for CompositeExplicitAutograd kernels + backend_index = backend_indices[DispatchKey.CompositeExplicitAutograd] + elif backend_indices[DispatchKey.CompositeExplicitAutogradNonFunctional].has_kernel( + f + ): + # We need to create C shim wrappers for CompositeExplicitAutogradNonFunctional kernels + backend_index = backend_indices[ + DispatchKey.CompositeExplicitAutogradNonFunctional + ] + return backend_index + + +def gen_c_shim( + f: NativeFunction, + dispatch_key: DispatchKey, + backend_indices: Dict[DispatchKey, BackendIndex], + header: bool, +) -> Optional[str]: + backend_index = get_backend_index_for_aoti(f, dispatch_key, backend_indices) + if backend_index is None: + return None + + schema = f.func + device = dispatch_key.lower() + backend_call = gen_static_dispatch_backend_call( + f, + backend_index, + ) + + try: + if header: + declaration, _ = gen_declaration_and_definition( + schema, device, backend_call + ) + return f"AOTI_TORCH_EXPORT {declaration};" + else: + _, definition = gen_declaration_and_definition(schema, device, backend_call) + return definition + + except NotImplementedError: + return None + + +@dataclass(frozen=True) +class ShimGenerator: + dispatch_key: DispatchKey + backend_indices: Dict[DispatchKey, BackendIndex] + header: bool # True to generate .h and False to generate .cpp + + @method_with_native_function + def __call__(self, f: NativeFunction) -> Optional[str]: + result = gen_c_shim(f, self.dispatch_key, self.backend_indices, self.header) + return result + + +def gen_aoti_c_shim( + native_functions: Sequence[NativeFunction], + dispatch_key: DispatchKey, + backend_indices: Dict[DispatchKey, BackendIndex], + header: bool, + includes: str = "", +) -> str: + body = "\n".join( + list( + mapMaybe( + ShimGenerator(dispatch_key, backend_indices, header), + native_functions, + ) + ) + ) + + if header: + return f""" +#pragma once + +#include + +#ifdef __cplusplus +extern "C" {{ +#endif + +{body} + +#ifdef __cplusplus +}} // extern "C" +#endif + +""" + else: + device = dispatch_key.lower() + return f""" +#include +#include +#include + +#ifndef AT_PER_OPERATOR_HEADERS +#include +#include +#include +#else +{includes} +#endif + +using namespace torch::aot_inductor; + +{body} + +""" diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/gen_vmap_plumbing.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/gen_vmap_plumbing.py new file mode 100644 index 0000000000000000000000000000000000000000..0876f3e343453afcd42f146495384a40cf5fecd1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/gen_vmap_plumbing.py @@ -0,0 +1,265 @@ +import textwrap +from dataclasses import dataclass +from typing import List, Optional, Sequence, Tuple + +from torchgen.api.translate import translate +from torchgen.api.types import DispatcherSignature +from torchgen.context import method_with_native_function +from torchgen.model import ( + Argument, + BaseTy, + BaseType, + FunctionSchema, + ListType, + NativeFunction, + OptionalType, + Return, + SchemaKind, + Type, +) +from torchgen.utils import mapMaybe + + +def is_tensor(typ: Type) -> bool: + return isinstance(typ, BaseType) and typ.name == BaseTy.Tensor + + +def is_optional_tensor(typ: Type) -> bool: + return isinstance(typ, OptionalType) and is_tensor(typ.elem) + + +def is_tensor_list(typ: Type) -> bool: + return isinstance(typ, ListType) and is_tensor(typ.elem) + + +def unwrap_tensor(name: str, cur_level_var: str) -> List[str]: + result = f"""\ + Tensor {name}_value; + optional {name}_bdim; + std::tie({name}_value, {name}_bdim) = unwrapTensorAtLevel({name}, {cur_level_var});""" + return textwrap.dedent(result).split("\n") + + +def unwrap_optional_tensor(name: str, cur_level_var: str) -> List[str]: + result = f"""\ + optional {name}_value; + optional {name}_bdim; + if ({name}) {{ + std::tie({name}_value, {name}_bdim) = unwrapTensorAtLevel({name}.value(), {cur_level_var}); + }}""" + return textwrap.dedent(result).split("\n") + + +def gen_unwraps( + flat_arguments: Sequence[Argument], cur_level_var: str +) -> Tuple[str, List[str]]: + arg_names = [a.name for a in flat_arguments] + arg_types = [a.type for a in flat_arguments] + + tensors = [name for typ, name in zip(arg_types, arg_names) if is_tensor(typ)] + optional_tensors = [ + name for typ, name in zip(arg_types, arg_names) if is_optional_tensor(typ) + ] + + unwraps = [] + for tensor in tensors: + unwraps += unwrap_tensor(tensor, cur_level_var) + + for opt_tensor in optional_tensors: + unwraps += unwrap_optional_tensor(opt_tensor, cur_level_var) + unwrap_code = "\n".join(unwraps) + + unwrapped_arg_list = [] + for arg in arg_names: + if arg in tensors or arg in optional_tensors: + unwrapped_arg_list += [f"{arg}_value", f"{arg}_bdim"] + else: + unwrapped_arg_list.append(arg) + return unwrap_code, unwrapped_arg_list + + +def gen_case_where_all_bdims_are_none( + outer_sig: DispatcherSignature, schema: FunctionSchema, cur_level_var: str +) -> str: + conditions = [] + flat_args = schema.arguments.flat_all + for arg in flat_args: + if not arg.type.is_tensor_like(): + continue + conditions.append(f"!isBatchedAtLevel({arg.name}, {cur_level_var})") + + sig = DispatcherSignature.from_schema(schema) + translated_args = ", ".join( + e.expr for e in translate(outer_sig.arguments(), sig.arguments()) + ) + return f"""\ +if ({' && '.join(conditions)}) {{ + return at::_ops::{sig.func.name.unambiguous_name()}::call({translated_args}); +}}""" + + +def gen_returns( + returns: Tuple[Return, ...], cur_level_var: str, results_var: str +) -> str: + idx = 0 + wrapped_returns = [] + for ret in returns: + if is_tensor(ret.type): + wrapped_returns.append( + f"makeBatched(std::get<{idx}>({results_var}), std::get<{idx + 1}>({results_var}), {cur_level_var})" + ) + idx += 2 + elif is_tensor_list(ret.type): + wrapped_returns.append( + f"makeBatchedVector(std::get<{idx}>({results_var}), std::get<{idx+1}>({results_var}), {cur_level_var})" + ) + idx += 2 + else: + wrapped_returns.append(f"std::get<{idx}>({results_var})") + idx += 1 + if len(wrapped_returns) == 1: + result = f"return {wrapped_returns[0]};" + else: + result = f'return std::make_tuple({", ".join(wrapped_returns)});' + return result + + +def accepts_at_least_one_tensor_input(schema: FunctionSchema) -> bool: + return any(a.type.is_tensor_like() for a in schema.arguments.flat_all) + + +def is_mutated_arg(argument: Argument) -> bool: + return argument.annotation is not None and argument.annotation.is_write + + +def gen_vmap_inplace_plumbing(native_function: NativeFunction) -> Optional[str]: + # Assumptions: + # - only one argument is being modified in-place + # - the argument that is being modified in-place is the first argument + # - all returns are either Tensor, tuple of Tensor, or TensorList + schema = native_function.func + sig = DispatcherSignature.from_schema(schema) + returns = schema.returns + + # Check assumptions. If these are invalid we return None + # and punt the work to handle them to the future. + assert schema.kind() == SchemaKind.inplace + if not is_mutated_arg(schema.arguments.flat_all[0]): + return None + if not len([arg for arg in schema.arguments.flat_all if is_mutated_arg(arg)]) == 1: + return None + + # Only support cases where all returns are Tensors or vector + if len(returns) == 0: + return None + if not all(is_tensor(ret.type) or is_tensor_list(ret.type) for ret in returns): + return None + if not accepts_at_least_one_tensor_input(schema): + return None + + cur_level_var = "cur_level" + + unwraps, unwrapped_arg_list = gen_unwraps(schema.arguments.flat_all, cur_level_var) + bdims_all_none_case = gen_case_where_all_bdims_are_none(sig, schema, cur_level_var) + + return f"""\ +template +{sig.decl(name=schema.name.unambiguous_name() + '_generated_plumbing')} {{ + c10::impl::ExcludeDispatchKeyGuard guard(DispatchKey::FuncTorchBatched); + auto maybe_layer = maybeCurrentDynamicLayer(); + vmap_check_escaped(maybe_layer, "gen_vmap_inplace_plumbing"); + int64_t {cur_level_var} = maybe_layer->layerId(); +{textwrap.indent(bdims_all_none_case, " ")} +{textwrap.indent(unwraps, " ")} + batch_rule({', '.join(unwrapped_arg_list)}); + return {schema.arguments.flat_all[0].name}; +}}""" + + +def gen_vmap_plumbing_no_returns(native_function: NativeFunction) -> str: + schema = native_function.func + sig = DispatcherSignature.from_schema(schema) + cur_level_var = "cur_level" + + unwraps, unwrapped_arg_list = gen_unwraps(schema.arguments.flat_all, cur_level_var) + bdims_all_none_case = gen_case_where_all_bdims_are_none(sig, schema, cur_level_var) + + return f"""\ +template +{sig.decl(name=schema.name.unambiguous_name() + '_generated_plumbing')} {{ + c10::impl::ExcludeDispatchKeyGuard guard(DispatchKey::FuncTorchBatched); + auto maybe_layer = maybeCurrentDynamicLayer(); + vmap_check_escaped(maybe_layer, "gen_vmap_plumbing_no_returns"); + int64_t {cur_level_var} = maybe_layer->layerId(); +{textwrap.indent(bdims_all_none_case, " ")} +{textwrap.indent(unwraps, " ")} + batch_rule({', '.join(unwrapped_arg_list)}); +}}""" + + +def gen_vmap_plumbing(native_function: NativeFunction) -> Optional[str]: + schema = native_function.func + sig = DispatcherSignature.from_schema(schema) + returns = schema.returns + + # Only support cases where all returns are Tensors or vector + if not accepts_at_least_one_tensor_input(schema): + return None + if len(returns) == 0: + return gen_vmap_plumbing_no_returns(native_function) + if not all(ret.type.is_tensor_like() for ret in returns): + return None + # in-place views need special handling + if "inplace_view" in native_function.tags: + return None + + if schema.kind() == SchemaKind.inplace: + return gen_vmap_inplace_plumbing(native_function) + + # Don't support these (mutable, out, scratch) + if schema.kind() != SchemaKind.functional: + return None + + results_var = "results" + cur_level_var = "cur_level" + + unwraps, unwrapped_arg_list = gen_unwraps(schema.arguments.flat_all, cur_level_var) + bdims_all_none_case = gen_case_where_all_bdims_are_none(sig, schema, cur_level_var) + + wrapped_returns = gen_returns(returns, cur_level_var, results_var) + return f"""\ +template +{sig.decl(name=schema.name.unambiguous_name() + '_generated_plumbing')} {{ + c10::impl::ExcludeDispatchKeyGuard guard(DispatchKey::FuncTorchBatched); + auto maybe_layer = maybeCurrentDynamicLayer(); + vmap_check_escaped(maybe_layer, "gen_vmap_plumbing"); + int64_t {cur_level_var} = maybe_layer->layerId(); +{textwrap.indent(bdims_all_none_case, " ")} +{textwrap.indent(unwraps, " ")} + auto {results_var} = batch_rule({', '.join(unwrapped_arg_list)}); + {wrapped_returns} +}}""" + + +@dataclass(frozen=True) +class ComputeBatchRulePlumbing: + @method_with_native_function + def __call__(self, f: NativeFunction) -> Optional[str]: + opname = str(f.func.name) + result = gen_vmap_plumbing(f) + return result + + +def gen_all_vmap_plumbing(native_functions: Sequence[NativeFunction]) -> str: + body = "\n".join(list(mapMaybe(ComputeBatchRulePlumbing(), native_functions))) + return f""" +#pragma once +#include +#include + +namespace at {{ namespace functorch {{ + +{body} + +}}}} // namespace at::functorch +""" diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/native_function_generation.py b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/native_function_generation.py new file mode 100644 index 0000000000000000000000000000000000000000..a09cce312258537cff8c7516b3a8f495406d927d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torchgen/native_function_generation.py @@ -0,0 +1,643 @@ +from collections import defaultdict + +from typing import Dict, List, Optional, Sequence, Tuple, Union + +import torchgen.api.dispatcher as dispatcher +from torchgen.api.translate import translate +from torchgen.api.types import Binding, DispatcherSignature, Expr +from torchgen.context import with_native_function +from torchgen.model import ( + Annotation, + Argument, + BackendIndex, + BackendMetadata, + BaseOperatorName, + BaseTy, + BaseType, + DEFAULT_KERNEL_NAMESPACE, + DeviceCheckType, + DispatchKey, + FunctionSchema, + NativeFunction, + NativeFunctionsGroup, + OperatorName, + Return, + SchemaKind, + Variant, +) +from torchgen.utils import concatMap + +# See Note: [Out ops with functional variants that don't get grouped properly] +OUT_OPS_THAT_DONT_GET_GROUPED_PROPERLY = [ + # This has a functional variant, but it's currently marked private. + # This function should be marked private as well (*_backward ops aren't exposed to python anyway). + "adaptive_avg_pool3d_backward.grad_input", + # There's a functional variant, _slow_conv2d_backward.output_mask, that isn't grouped properly. + # Maybe we can kill this operator in favor of convolution_backward? + "_slow_conv2d_backward.grad_input", +] + + +# See Note: [Mutable ops that cannot get an out variant] +MUTABLE_OPS_THAT_CANNOT_GET_AN_OUT_VARIANT = [ + # should be out=? + "_cummax_helper", + # should be out=? + "_cummin_helper", +] + +# All of these operators don't have any tensor like returns +FUNCTIONAL_OPS_THAT_CANNOT_GET_AN_OUT_VARIANT = [ + "_assert_async", # no return + "_assert_async.msg", # no return + "_cslt_sparse_mm_search", # returns an int + "_assert_scalar", # no return + "_dimI", # returns an int + "_dimV", # returns an int + "_has_same_storage_numel", # returns a boolean + "_linalg_check_errors", # no return + "_local_scalar_dense", # returns a Scalar + "_nested_tensor_from_mask_left_aligned", # returns a boolean + "_nnz", # returns an int + "_use_cudnn_ctc_loss", # returns a boolean + "_use_cudnn_ctc_loss.Tensor", # returns a boolean + "_validate_compressed_sparse_indices", # no return + "allclose", # returns a boolean + "dense_dim", # returns an int + "equal", # returns a boolean + "is_coalesced", # returns an boolean + "is_pinned", # returns a boolean + "is_same_size", # returns a boolean + "is_set_to", # returns a boolean + "q_per_channel_axis", # returns an int + "q_scale", # returns a float + "q_zero_point", # returns an int + "qscheme", # returns a QScheme + "record_stream", # no return + "sparse_dim", # returns an int + "sym_constrain_range", # no return + "sym_constrain_range_for_size", # no return + "_nested_tensor_storage_offsets", # returns a vector of ints + "_chunk_grad_outputs_efficient_attention", # returns a bool + "_fused_sdp_choice", # returns an int + "_print", # no return + "_nested_get_ragged_idx", # returns an int +] + +INPLACE_OPS_THAT_DONT_GET_GROUPED_PROPERLY = [ + # polygamma and polygamma.out both exist, but have a + # pre-self arg (while polygamma_ does not) + # We should either fix this schema so it can be grouped properly, + # or allow the codegen to generate new functional/out= NativeFunctions for this op + # (which would require changing its overload name to prevent overload ambiguity). + "polygamma_" +] + + +# Groups "similar" NativeFunctions together +# example add.Tensor, add_.Tensor, add.out +# "similar" NativeFunctions are all expected to have an identical `signature()`, +# But have differing SchemaKinds. +def pre_group_native_functions( + native_functions: Sequence[NativeFunction], +) -> Dict[FunctionSchema, Dict[SchemaKind, NativeFunction]]: + pre_grouped_native_functions: Dict[ + FunctionSchema, Dict[SchemaKind, NativeFunction] + ] = defaultdict(dict) + for f in native_functions: + d = pre_grouped_native_functions[f.func.signature()] + assert f.func.kind() not in d + d[f.func.kind()] = f + return pre_grouped_native_functions + + +# Returns the out variant overload name given a base function overload name +def get_expected_out_variant_overload_name(overload_name: Optional[str]) -> str: + return "out" if not overload_name else f"{overload_name}_out" + + +# Helper function: given an inplace FunctionSchema, generate its corresponding out= variant +# Example before: +# _add_relu_.Scalar(Tensor(a!) self, Scalar other, Scalar alpha=1) -> Tensor(a!) +# Example after: +# _add_relu.Scalar_out(Tensor self, Scalar other, Scalar alpha=1, *, Tensor(a!) out) +def self_to_out_signature(func: FunctionSchema) -> FunctionSchema: + # Generating an out= schema from an inplace schema. + assert func.kind() == SchemaKind.inplace + assert func.arguments.self_arg is not None + # The new out= schema has: + # - a new out argument with the same type as "func" (but with a mutable annotation) + # - The returns (if any) now alias the out= argument instead of "func" + # - an "out" overload name + return FunctionSchema( + name=func.name.remove_inplace().with_overload( + get_expected_out_variant_overload_name(func.name.overload_name) + ), + arguments=func.arguments.remove_self_annotation().with_out_args( + [ + Argument( + name="out", + type=func.arguments.self_arg.argument.type, + default=None, + annotation=func.arguments.self_arg.argument.annotation, + ) + ] + ), + returns=func.returns, + ) + + +# Helper function: given a functional FunctionSchema, generate its corresponding out= variant +# Example before: +# _to_copy(Tensor self, *, ScalarType? dtype=None, Layout? layout=None, Device? device=None, +# bool? pin_memory=None, bool non_blocking=False, MemoryFormat? memory_format=None) -> Tensor +# Example after: +# _to_copy._out(Tensor self, *, bool non_blocking=False, MemoryFormat? memory_format=None, +# Tensor(a!) out) -> Tensor(a!) +def functional_to_out_signature(func: FunctionSchema) -> FunctionSchema: + # Generating an out= schema from a functional schema. + assert func.kind() == SchemaKind.functional + + new_returns, new_out_args = generate_out_args_from_schema(func) + # The new out= schema has: + # - one or more new out argument(s) with the same type as returns (but with a mutable annotation) + # - The returns now alias the out= arguments + # - an "_out" overload name + return FunctionSchema( + name=func.name.with_overload( + get_expected_out_variant_overload_name(func.name.overload_name) + ), + arguments=func.arguments.signature().with_out_args( + new_out_args, + ), + returns=tuple(new_returns), + ) + + +# Helper function: given a function schema, generate corresponding out arguments, also the updated return annotations. +def generate_out_args_from_schema( + func: FunctionSchema, +) -> Tuple[List[Return], List[Argument]]: + # More of a sanity check - our existing restrictions on schemas should enforce that + # mutable schema kinds never return their mutable arguments. + assert not any( + r.annotation is not None and r.annotation.is_write for r in func.returns + ) + + tensorlike_rets = [r for r in func.returns if r.type.is_tensor_like()] + assert len(tensorlike_rets) > 0 + + used_annotations = concatMap( + lambda a: [] if a.annotation is None else a.annotation.alias_set, + func.arguments.flat_all, + ) + valid_annotations = [ + x for x in "abcdefghijklmnopqrstuvwxyz" if x not in used_annotations + ] + + all_rets_are_tensors = all(r.type == BaseType(BaseTy.Tensor) for r in func.returns) + + new_out_args: List[Argument] = [] + # The end result of new_returns is that: + # - If every return is a plain tensor, then the new returns == the old returns, but with the out= alias annotations added. + # - Otherwise, none of the out arguments show up in the returns (and we're only left with non-tensor-like returns, if any). + new_returns: List[Return] = [] + for i, r in enumerate(func.returns): + if r.type.is_tensor_like(): + new_out = Argument( + name="out" if len(func.returns) == 1 else f"out{i}", + type=r.type, + default=None, + annotation=Annotation.parse(f"{valid_annotations[i]}!"), + ) + new_out_args.append(new_out) + if all_rets_are_tensors: + # The convention for out= schemas is that they only return their out arguments + # if the return is a plain Tensor (or if it's a tuple of plain Tensors) + new_ret = Return( + name=None, type=new_out.type, annotation=new_out.annotation + ) + new_returns.append(new_ret) + else: + new_returns.append(r) + return new_returns, new_out_args + + +# Helper function: given a mutable FunctionSchema, generate its corresponding out= variant +# Example before: +# _fused_moving_avg_obs_fq_helper(Tensor self, Tensor observer_on, Tensor fake_quant_on, Tensor(a!) running_min, Tensor(b!) running_max, Tensor(c!) scale, Tensor(d!) zero_point, float averaging_const, int quant_min, int quant_max, int ch_axis, bool per_row_fake_quant=False, bool symmetric_quant=False) -> (Tensor output, Tensor mask) # noqa: B950 +# Example after: +# _fused_moving_avg_obs_fq_helper._out(Tensor self, Tensor observer_on, Tensor fake_quant_on, Tensor(a!) running_min, Tensor(b!) running_max, Tensor(c!) scale, Tensor(d!) zero_point, float averaging_const, int quant_min, int quant_max, int ch_axis, bool per_row_fake_quant=False, bool symmetric_quant=False, *, Tensor(e!) out0, Tensor(f!) out1) -> (Tensor(e!), Tensor(f!)) # noqa: B950 +def mutable_to_out_signature(func: FunctionSchema) -> FunctionSchema: + # Generating an out= schema from a mutable schema. + assert func.kind() == SchemaKind.mutable + # The new out= schema has: + # - Any non-aliased tensor-like returns are converted to mutable, aliased out= arguments + # (if the argument is a tensor then we also return it for method chaining, + # otherwise we return nothing) + # - an "out" overload name + # + # Note that: + # (1) This also means that we can *only* generate an out= variant from a mutable schema + # if the mutable schema has at least one tensor-like non-aliasing return. + # (2) The generated out= variant still has mutable positional arguments, + # but if necessary we could probably add another out= variant that also + # functionalizes the mutable arguments (a functional_out variant) + + new_returns, new_out_args = generate_out_args_from_schema(func) + + return FunctionSchema( + name=func.name.remove_inplace().with_overload( + get_expected_out_variant_overload_name(func.name.overload_name) + ), + arguments=func.arguments.with_out_args(new_out_args), + returns=tuple(new_returns), + ) + + +# This function, given function of one SchemaKind, as well as a target SchemaKind, +# generates a new NativeFunction with the same properties, but using the target SchemaKind. +# We only actually generate functions for either functional or out= SchemaKinds. +# This function returns a tuple, with: +# - The generated NativeFunction +# - a dictionary of `BackendIndex` objects, describing which dispatch keys +# we will generate kernels for, for the new NativeFunction. +# Details are in the function, but we only generate composite kernels (in some cases) today. +def generate_function( + f: NativeFunction, k: SchemaKind +) -> Tuple[NativeFunction, Dict[DispatchKey, Dict["OperatorName", "BackendMetadata"]]]: + from torchgen.api import cpp + + if k == SchemaKind.functional: + assert f.func.kind() != SchemaKind.functional + # The new "functional" NativeFunction has: + # - any mutable arguments have been converted into (immutable) returns. + # (if a mutable argument was not also a return, it gets converted to one) + # - "_functional" appended to the base name, ONLY IF this op has a mutable variant. + # See Note [Overload Ambiguity With Functional Variants] + # The default grouping logic in signature() actually already does this, + # so we can piggy-back off it (but we still want return names) + func = f.func.signature(keep_return_names=True).with_name( + OperatorName( + name=BaseOperatorName( + base=f.func.name.name.base, + inplace=False, + dunder_method=f.func.name.name.dunder_method, + # See Note [Overload Ambiguity With Functional Variants] + functional_overload=f.func.kind() == SchemaKind.mutable, + ), + overload_name=f.func.name.overload_name, + ) + ) + elif k == SchemaKind.out: + # We generate out= ops mostly just so that we can pair up NativeFunctions into groups easily, + # but at least today, there is no good reason to actually use them. + # we'll generate a dispatcher entry for them, but won't actually register any kernels for them. + if f.func.kind() == SchemaKind.inplace: + func = self_to_out_signature(f.func) + elif f.func.kind() == SchemaKind.mutable: + func = mutable_to_out_signature(f.func) + elif f.func.kind() == SchemaKind.functional: + func = functional_to_out_signature(f.func) + else: + raise AssertionError( + "We only bother generating out= functions from either inplace or mutable or functional variants" + ) + else: + raise AssertionError( + "We currently only generate either functional or out= NativeFunctions" + ) + + # Generated kernel naming convention for out: _. The reason for this is to + # disambiguate operator with the same name but different overload name, e.g., `randn.names_out` and + # `randn.generator_with_names_out`. + kernel_name = ( + func.name.unambiguous_name() + if func.kind() == SchemaKind.out + else cpp.name(func) + ) + if f.func.has_symint(): + kernel_name += "_symint" + backend_metadata = { + DispatchKey.CompositeExplicitAutograd: { + func.name: BackendMetadata( + kernel=kernel_name, + structured=False, + cpp_namespace=DEFAULT_KERNEL_NAMESPACE, + ) + } + } + tags = {"generated"} | set( + f.tags & {"nondeterministic_seeded", "view_copy", "pt2_compliant_tag"} + ) + + return ( + NativeFunction( + func=func, + use_const_ref_for_mutable_tensors=f.use_const_ref_for_mutable_tensors, + # These generated fn's aren't meant to be user friendly- don't generate methods. + variants={Variant.function}, + structured=False, + structured_delegate=None, + structured_inherits=None, + precomputed=None, + autogen=[], + ufunc_inner_loop={}, + manual_kernel_registration=False, + manual_cpp_binding=False, + python_module=None, + category_override=None, + device_guard=False, + device_check=DeviceCheckType.NoCheck, + loc=f.loc, + cpp_no_default_args=set(), + is_abstract=f.is_abstract, + has_composite_implicit_autograd_kernel=False, + has_composite_implicit_autograd_nested_tensor_kernel=False, + has_composite_explicit_autograd_kernel=True, + has_composite_explicit_autograd_non_functional_kernel=False, + # Every generated NativeFunction gets a "generated" tag, so it's easy to tell + # which NativeFunction objects did not come directly from native_functions.yaml. + tags=tags, + namespace=f.namespace, + ), + backend_metadata, + ) + + +# This function is responsible for adding generated NativeFunctions which don't appear +# explicitly in the codegen. +# You can inspect the full list of NativeFunctions yourself with the torchgen package, by running +# torchgen.parse_native_yaml("aten/src/ATen/native/native_functions.yaml", "aten/src/ATen/native/tags.yaml") +# (Maybe we should make a friendly API for this) +# +# Note: this function *mutates* its two inputs, +# adding the new NativeFunctions / BackendMetadata to them +def add_generated_native_functions( + rs: List[NativeFunction], + indices: Dict[DispatchKey, Dict[OperatorName, BackendMetadata]], +) -> None: + # The main code for generating new NativeFunctions + # First we group of NativeFunctions by schema kind, + # then we detect which ones are missing and generate them. + pre_grouped_native_functions = pre_group_native_functions(rs) + for d in pre_grouped_native_functions.values(): + has_functional = SchemaKind.functional in d + has_inplace = SchemaKind.inplace in d + has_mutable = SchemaKind.mutable in d + has_out = SchemaKind.out in d + + # We automatically generate a few native functions that don't exist in the yaml, for a few reasons: + # (1) If an operator has an inplace/out= variant but no functional variant, we can generate + # a simple functional variant that the functionalization pass can consume. + # (2) If an operator has an inplace or functional but no out= variant, we generate an out= + # variant, mostly so we can easily pair up functions into NativeFunctionsGroup, + # while maintaining the constraint that the out= variant is "required". + if has_mutable or has_inplace or has_out or has_functional: + # Don't bother generating functions trio's for native functions that bypass the dispatcher. + are_manual = all(f.manual_cpp_binding for f in d.values()) + # Don't bother generating functional + out= variants for view operators + # set_ is technically an inplace_view, but for now it is treated + # as a normal inplace op in the codegen + has_view_ops = any( + f.is_view_op and str(f.func.name.name) != "set_" for f in d.values() + ) + # Don't generate the other variants for CompositeImplicitAutograd operators. + # We could probably do this, but the main benefit of generating the function triplets + # is for transforms that need them, and transforms don't need to act directly + # on CompositeImplicitAutograd operators (since we let them decompose). + are_composite_implicit = all( + f.has_composite_implicit_autograd_kernel for f in d.values() + ) + if are_manual or has_view_ops or are_composite_implicit: + continue + if has_out and len(d.values()) == 1: + # Note: [Out ops with functional variants that don't get grouped properly] + # In theory we could validly have an out= operator in native_functions.yaml + # that has no other variants. + # But today, all of the operators where that's the case actually do have + # functional variants, that we are just unable to pair up properly. + # I think banning this all together is probably safer + # (you can always add a functional variant yourself if you want to add a new out= operator). + # + # We should probably fix the existing cases; this check is to prevent us from adding more over time. + if ( + str(d[SchemaKind.out].func.name) + not in OUT_OPS_THAT_DONT_GET_GROUPED_PROPERLY + ): + raise AssertionError( + f"Found an out= operator that we could not find any other variants of: {str(d[SchemaKind.out].func)}" + ) + continue + + # Some inplace ops that have problematic schemas (that we should fix), which prevent us + # from generating out= and functional variants + if ( + has_inplace + and str(d[SchemaKind.inplace].func.name) + in INPLACE_OPS_THAT_DONT_GET_GROUPED_PROPERLY + ): + continue + + base_fn = ( + d[SchemaKind.inplace] + if has_inplace + else d[SchemaKind.mutable] + if has_mutable + else d[SchemaKind.out] + if has_out + else d[SchemaKind.functional] + ) + + # Note: [Mutable ops that cannot get an out variant] + # We can only generate an out= variant if either: + # - the original function has tensor-like returns (since we can convert them to out kwargs) + # - or it's inplace (since we can convert `self` to an out kwarg) + # There are only two functions that don't fit this criteria today though, + # and they both look like they should be fixed to be out= variants, + # so if feels safer to ban this schema all-together + base_fn_valid = base_fn.func.kind() == SchemaKind.inplace or any( + r.type.is_tensor_like() for r in base_fn.func.returns + ) + # Note: [Loosen the assertion that all functional should have out variant] + # By design all functional operators should have our variants. The needs_out check + # is loosening this requirement, changing it to only generate out variant if there's + # an `autogen` block in the native function, in the long run it should be removed. + # FIXME: Remove this after figuring out CI job failures related to min, max, mean + needs_out = any("out" in str(op_name) for op_name in base_fn.autogen) + gets_out_variant = not has_out and base_fn_valid and needs_out + if not has_out and not base_fn_valid: + if ( + str(base_fn.func.name) + not in MUTABLE_OPS_THAT_CANNOT_GET_AN_OUT_VARIANT + and str(base_fn.func.name) + not in FUNCTIONAL_OPS_THAT_CANNOT_GET_AN_OUT_VARIANT + ): + raise AssertionError( + f"""Found an operator that we could not generate an out= variant for: {str(base_fn.func)}. +This type of operators don't have tensor-like return, making it difficult to generate a proper out= variant. If +out= variant is not needed, please add the function name into FUNCTIONAL_OPS_THAT_CANNOT_GET_AN_OUT_VARIANT list.""" + ) + + # Generate an out= variant + if gets_out_variant: + fn, metadata = generate_function(base_fn, SchemaKind.out) + d[SchemaKind.out] = fn + BackendIndex.grow_index(indices, metadata) + rs.append(fn) + + # Generate a functional variant, but only do it if the operator got an out= variant + # (Functional variants are only useful if we can group up the variants, + # which we can only do if they have an out= variant) + if not has_functional and (has_out or gets_out_variant): + fn, metadata = generate_function(base_fn, SchemaKind.functional) + d[SchemaKind.functional] = fn + BackendIndex.grow_index(indices, metadata) + rs.append(fn) + + +def return_str(rets: Tuple[Return, ...], names: List[str]) -> str: + assert len(rets) == len(names) + if len(rets) == 0: + return "" + elif len(rets) == 1: + return f"return {names[0]};" + else: + return f"return {dispatcher.returns_type(rets).cpp_type()}({', '.join(names)});" + + +# Given a function, and the name of a variable corresponding to the output of that function, +# gather up all of the individual returns that are not aliased +def gather_nonaliased_inner_rets(func: FunctionSchema, out_var: str) -> List[str]: + aliased_rets = func.aliased_return_names() + non_aliased_names = [] + is_out_var_a_tuple = len(func.returns) > 1 + for i, r in enumerate(aliased_rets): + if r is None: + non_aliased_names.append( + f"std::get<{i}>({out_var})" if is_out_var_a_tuple else out_var + ) + return non_aliased_names + + +# Generates functional kernels in terms of their inplace.mutable counterparts. +# We only do this for "generated" NativeFunctions +@with_native_function +def gen_composite_functional_kernel(g: NativeFunctionsGroup) -> Optional[str]: + # We should only be generating these for code-generated NativeFunctions + if "generated" not in g.functional.tags: + return None + # And we always write the kernel for a generated op in terms of a non-generated op. + if g.inplace is not None and "generated" not in g.inplace.tags: + target_f = g.inplace + elif g.mutable is not None and "generated" not in g.mutable.tags: + target_f = g.mutable + else: + # We should be guaranteed to have a valid inplace/mutable variant to call into. + # See Note: [Mutable Ops Not Using Functionalization] + raise AssertionError(str(g.functional.func)) + + sig = DispatcherSignature(g.functional.func) + target_sig = DispatcherSignature(target_f.func) + + context: List[Union[Binding, Expr]] = [] + clone_mutable_inputs = [] + cloned_return_names = [] + # We can't just directly pass all of the arguments from the functional op into the mutating op. + # We need to check for which inputs to the mutating operator are mutable, + # and clone those inputs first. + for a_curr, a_tgt in zip( + dispatcher.jit_arguments(g.functional.func), + dispatcher.jit_arguments(target_f.func), + ): + if a_tgt.annotation is not None and a_tgt.annotation.is_write: + clone_mutable_inputs.append( + f"auto {a_curr.name}_clone = clone_arg({a_curr.name});" + ) + context.append( + Expr( + expr=f"{a_curr.name}_clone", + type=dispatcher.argument_type(a_curr, binds=a_curr.name), + ) + ) + # Invariant: mutable arguments on the inner mutable op are always returns on the functional op. + cloned_return_names.append(f"{a_curr.name}_clone") + else: + context.append(dispatcher.argument(a_curr)) + exprs = ", ".join([e.expr for e in translate(context, target_sig.arguments())]) + + out_name = "output" + maybe_assign = f"auto {out_name} = " if len(target_f.func.returns) > 0 else "" + inner_return_names = gather_nonaliased_inner_rets(target_f.func, out_name) + ret_str = return_str( + g.functional.func.returns, inner_return_names + cloned_return_names + ) + + clone_mutable_inputs_str = "\n".join(clone_mutable_inputs) + return f""" +{sig.defn(name=sig.name() + ("_symint" if g.out.func.has_symint() else ""))} {{ + {clone_mutable_inputs_str} + {maybe_assign}at::_ops::{target_f.func.name.unambiguous_name()}::call({exprs}); + {ret_str} +}} +""" + + +# Generates out= kernels in terms of their functional counterparts. +# We only do this for "generated" NativeFunctions +@with_native_function +def gen_composite_out_kernel(g: NativeFunctionsGroup) -> Optional[str]: + # We should only be generating these for code-generated NativeFunctions + if "generated" not in g.out.tags: + return None + # And we always write the kernel for the out= op in terms of the functional. + # Note that the functional op might have also been generated, but we don't have to + # worry about cycles, because the generated functional kernels are always implemented + # in terms of non-generated kernels (see gen_composite_functional_kernel). + + sig = DispatcherSignature(g.out.func) + target_sig = DispatcherSignature(g.functional.func) + + exprs = ", ".join( + [e.expr for e in translate(sig.arguments(), target_sig.arguments())] + ) + + copy_outs = [] + out_name = "tmp_output" + for i, out_arg in enumerate(g.out.func.arguments.out): + functional_return_name = ( + out_name + if len(g.functional.func.returns) == 1 + else f"std::get<{i}>({out_name})" + ) + copy_outs.append( + f"""\ + resize_out_helper({out_arg.name}, {functional_return_name}); + copy_arg({out_arg.name}, {functional_return_name});""" + ) + + rets = [] + # For each return arg in the calling (out=) operator, + # If it corresponds to an aliased input, return the input. + # Otherwise, return the corresponding output from calling the functional operator. + for i, ret_name in enumerate(g.out.func.aliased_return_names()): + if ret_name is not None: + rets.append(ret_name) + else: + functional_return_name = ( + out_name + if len(g.functional.func.returns) == 1 + else f"std::get<{i}>({out_name})" + ) + rets.append(functional_return_name) + + copy_outs_str = "\n".join(copy_outs) + + # Kernel name needs to follow the naming convention defined in `generate_function()` + return f""" +{sig.defn(name=g.out.func.name.unambiguous_name() + ("_symint" if g.out.func.has_symint() else ""))} {{ + auto {out_name} = at::_ops::{g.functional.func.name.unambiguous_name()}::call({exprs}); + {copy_outs_str} + {return_str(g.out.func.returns, rets)} +}} +""" diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/LICENSE.txt b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/LICENSE.txt new file mode 100644 index 0000000000000000000000000000000000000000..a31470f14c5978d5fcc3bc173b8399b6c9a6443f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/LICENSE.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2012 Daniel Holth and contributors + +Permission is hereby granted, free of charge, to any person obtaining a +copy of this software and associated documentation files (the "Software"), +to deal in the Software without restriction, including without limitation +the rights to use, copy, modify, merge, publish, distribute, sublicense, +and/or sell copies of the Software, and to permit persons to whom the +Software is furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR +OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +OTHER DEALINGS IN THE SOFTWARE. diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/RECORD b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/RECORD new file mode 100644 index 0000000000000000000000000000000000000000..af89647b17e5ea7d8bc667b7b6499ae22ac58d85 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/RECORD @@ -0,0 +1,68 @@ +../../../bin/wheel,sha256=8jeGSkM6837i3Wir2Il3ba-2_OOzLoHdnaz9GDVotaI,257 +wheel-0.45.1.dist-info/INSTALLER,sha256=zuuue4knoyJ-UwPPXg8fezS7VCrXJQrAP7zeNuwvFQg,4 +wheel-0.45.1.dist-info/LICENSE.txt,sha256=MMI2GGeRCPPo6h0qZYx8pBe9_IkcmO8aifpP8MmChlQ,1107 +wheel-0.45.1.dist-info/METADATA,sha256=mKz84H7m7jsxJyzeIcTVORiTb0NPMV39KvOIYhGgmjA,2313 +wheel-0.45.1.dist-info/RECORD,, +wheel-0.45.1.dist-info/REQUESTED,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +wheel-0.45.1.dist-info/WHEEL,sha256=CpUCUxeHQbRN5UGRQHYRJorO5Af-Qy_fHMctcQ8DSGI,82 +wheel-0.45.1.dist-info/entry_points.txt,sha256=rTY1BbkPHhkGMm4Q3F0pIzJBzW2kMxoG1oriffvGdA0,104 +wheel/__init__.py,sha256=mrxMnvdXACur_LWegbUfh5g5ysWZrd63UJn890wvGNk,59 +wheel/__main__.py,sha256=NkMUnuTCGcOkgY0IBLgBCVC_BGGcWORx2K8jYGS12UE,455 +wheel/__pycache__/__init__.cpython-311.pyc,, +wheel/__pycache__/__main__.cpython-311.pyc,, +wheel/__pycache__/_bdist_wheel.cpython-311.pyc,, +wheel/__pycache__/_setuptools_logging.cpython-311.pyc,, +wheel/__pycache__/bdist_wheel.cpython-311.pyc,, +wheel/__pycache__/macosx_libfile.cpython-311.pyc,, +wheel/__pycache__/metadata.cpython-311.pyc,, +wheel/__pycache__/util.cpython-311.pyc,, +wheel/__pycache__/wheelfile.cpython-311.pyc,, +wheel/_bdist_wheel.py,sha256=UghCQjSH_pVfcZh6oRjzSw_TQhcf3anSx1OkiLSL82M,21694 +wheel/_setuptools_logging.py,sha256=-5KC-lne0ilOUWIDfOkqapUWGMFZhuKYDIavIZiB5kM,781 +wheel/bdist_wheel.py,sha256=tpf9WufiSO1RuEMg5oPhIfSG8DMziCZ_4muCKF69Cqo,1107 +wheel/cli/__init__.py,sha256=Npq6_jKi03dhIcRnmbuFhwviVJxwO0tYEnEhWMv9cJo,4402 +wheel/cli/__pycache__/__init__.cpython-311.pyc,, +wheel/cli/__pycache__/convert.cpython-311.pyc,, +wheel/cli/__pycache__/pack.cpython-311.pyc,, +wheel/cli/__pycache__/tags.cpython-311.pyc,, +wheel/cli/__pycache__/unpack.cpython-311.pyc,, +wheel/cli/convert.py,sha256=Bi0ntEXb9nTllCxWeTRQ4j-nPs3szWSEKipG_GgnMkQ,12634 +wheel/cli/pack.py,sha256=CAFcHdBVulvsHYJlndKVO7KMI9JqBTZz5ii0PKxxCOs,3103 +wheel/cli/tags.py,sha256=lHw-LaWrkS5Jy_qWcw-6pSjeNM6yAjDnqKI3E5JTTCU,4760 +wheel/cli/unpack.py,sha256=Y_J7ynxPSoFFTT7H0fMgbBlVErwyDGcObgme5MBuz58,1021 +wheel/macosx_libfile.py,sha256=k1x7CE3LPtOVGqj6NXQ1nTGYVPaeRrhVzUG_KPq3zDs,16572 +wheel/metadata.py,sha256=JC4p7jlQZu2bUTAQ2fevkqLjg_X6gnNyRhLn6OUO1tc,6171 +wheel/util.py,sha256=aL7aibHwYUgfc8WlolL5tXdkV4DatbJxZHb1kwHFJAU,423 +wheel/vendored/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +wheel/vendored/__pycache__/__init__.cpython-311.pyc,, +wheel/vendored/packaging/LICENSE,sha256=ytHvW9NA1z4HS6YU0m996spceUDD2MNIUuZcSQlobEg,197 +wheel/vendored/packaging/LICENSE.APACHE,sha256=DVQuDIgE45qn836wDaWnYhSdxoLXgpRRKH4RuTjpRZQ,10174 +wheel/vendored/packaging/LICENSE.BSD,sha256=tw5-m3QvHMb5SLNMFqo5_-zpQZY2S8iP8NIYDwAo-sU,1344 +wheel/vendored/packaging/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0 +wheel/vendored/packaging/__pycache__/__init__.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/_elffile.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/_manylinux.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/_musllinux.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/_parser.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/_structures.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/_tokenizer.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/markers.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/requirements.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/specifiers.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/tags.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/utils.cpython-311.pyc,, +wheel/vendored/packaging/__pycache__/version.cpython-311.pyc,, +wheel/vendored/packaging/_elffile.py,sha256=hbmK8OD6Z7fY6hwinHEUcD1by7czkGiNYu7ShnFEk2k,3266 +wheel/vendored/packaging/_manylinux.py,sha256=P7sdR5_7XBY09LVYYPhHmydMJIIwPXWsh4olk74Uuj4,9588 +wheel/vendored/packaging/_musllinux.py,sha256=z1s8To2hQ0vpn_d-O2i5qxGwEK8WmGlLt3d_26V7NeY,2674 +wheel/vendored/packaging/_parser.py,sha256=4tT4emSl2qTaU7VTQE1Xa9o1jMPCsBezsYBxyNMUN-s,10347 +wheel/vendored/packaging/_structures.py,sha256=q3eVNmbWJGG_S0Dit_S3Ao8qQqz_5PYTXFAKBZe5yr4,1431 +wheel/vendored/packaging/_tokenizer.py,sha256=alCtbwXhOFAmFGZ6BQ-wCTSFoRAJ2z-ysIf7__MTJ_k,5292 +wheel/vendored/packaging/markers.py,sha256=_TSPI1BhJYO7Bp9AzTmHQxIqHEVXaTjmDh9G-w8qzPA,8232 +wheel/vendored/packaging/requirements.py,sha256=dgoBeVprPu2YE6Q8nGfwOPTjATHbRa_ZGLyXhFEln6Q,2933 +wheel/vendored/packaging/specifiers.py,sha256=IWSt0SrLSP72heWhAC8UL0eGvas7XIQHjqiViVfmPKE,39778 +wheel/vendored/packaging/tags.py,sha256=fedHXiOHkBxNZTXotXv8uXPmMFU9ae-TKBujgYHigcA,18950 +wheel/vendored/packaging/utils.py,sha256=XgdmP3yx9-wQEFjO7OvMj9RjEf5JlR5HFFR69v7SQ9E,5268 +wheel/vendored/packaging/version.py,sha256=PFJaYZDxBgyxkfYhH3SQw4qfE9ICCWrTmitvq14y3bs,16234 +wheel/vendored/vendor.txt,sha256=Z2ENjB1i5prfez8CdM1Sdr3c6Zxv2rRRolMpLmBncAE,16 +wheel/wheelfile.py,sha256=USCttNlJwafxt51YYFFKG7jnxz8dfhbyqAZL6jMTA9s,8411 diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/WHEEL b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/WHEEL new file mode 100644 index 0000000000000000000000000000000000000000..e3c6feefa22927866e3fd5575379ea972b432aaf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/wheel-0.45.1.dist-info/WHEEL @@ -0,0 +1,4 @@ +Wheel-Version: 1.0 +Generator: flit 3.10.1 +Root-Is-Purelib: true +Tag: py3-none-any