提交 3e087f76 编写于 作者: Y Yi Wang

Add buddy_allocator.cc and system_allocator.cc

上级 b8f5922d
develop 1.8.5 2.0.1-rocm-post 2.4.1 Ligoml-patch-1 OliverLPH-patch-1 OliverLPH-patch-2 PaddlePM-patch-1 PaddlePM-patch-2 ZHUI-patch-1 add_default_att add_kylinv10 add_model_benchmark_ci add_some_yaml_config addfile all_new_design_exec ascendrc ascendrelease bugfix-eval-frame-leakgae cherry-pick-fix-customOP-random-fail cherry_undefined_var compile_windows cp_2.4_fix_numpy delete_2.0.1-rocm-post delete_add_default_att delete_all_new_design_exec delete_ascendrc delete_compile_windows delete_delete_addfile delete_disable_iterable_dataset_unittest delete_fix_dataloader_memory_leak delete_fix_imperative_dygraph_error delete_fix_retry_ci delete_fix_undefined_var delete_improve_sccache delete_incubate/lite delete_paddle_tiny_install delete_paralleltest delete_prv-disable-more-cache delete_revert-31068-fix_conv3d_windows delete_revert-31562-mean delete_revert-33630-bug-fix delete_revert-34159-add_npu_bce_logical_dev delete_revert-34910-spinlocks_for_allocator delete_revert-35069-revert-34910-spinlocks_for_allocator delete_revert-36057-dev/read_flags_in_ut dingjiaweiww-patch-1 disable_iterable_dataset_unittest dy2static enable_eager_model_test final_state_gen_python_c final_state_intermediate fix-numpy-issue fix-run-program-grad-node-mem fix_check fix_concat_slice fix_custom_device_copy_sync fix_dataloader_memory_leak fix_dlpack_for fix_imperative_dygraph_error fix_newexe_gc fix_npu_ci fix_op_flops fix_retry_ci fix_rnn_docs fix_tensor_type fix_undefined_var fix_var_stop_gradient_error fixiscan fixiscan1 fixiscan2 fixiscan3 github/fork/123malin/netifaces github/fork/123malin/tdm_abacus github/fork/AshburnLee/dev_unique github/fork/ForFishes/fix_memory_matmul github/fork/ForFishes/rm_fluid github/fork/LielinJiang/move-2.0-api github/fork/LielinJiang/visual-dl-cb github/fork/LiuChiachi/add-transformer-generate-square-subsequent-mask-api github/fork/LiuChiachi/fix-example-code-for-hapi-Model github/fork/LiuChiachi/remove-input-requirment-in-dygraph-Model github/fork/MrChengmo/fix_ps_profiler github/fork/MrChengmo/update_ps_heter github/fork/PWhiddy/patch-1 github/fork/Shixiaowei02/dev/save_load_upgrade github/fork/TCChenlong/fix_hapi github/fork/TCChenlong/fix_inden github/fork/Thunderbrook/xpu_slice github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var_2 github/fork/XieYunshen/disable_ut_test_parallel_executor_fetch_isolated_var_3 github/fork/XieYunshen/timeout_20S_ut github/fork/ZeyuChen/remove-nltk github/fork/arlesniak/arlesniak/selective__mkldnn_flags github/fork/baiyfbupt/code_doc_mig github/fork/chalsliu/set_timeout github/fork/chen-zhiyu/develop github/fork/chenwhql/ci/try_to_find_test_buffer_shared_memory_reuse_pass_error github/fork/chenwhql/dygraph/remove_scale_loss_and_apply_collective_grads github/fork/chenwhql/saveload/add_get_inference_program github/fork/chenwhql/saveload/remove_save_load_config github/fork/cryoco/pass-compatibility-trt github/fork/danleifeng/isempty_api2.0 github/fork/frankwhzhang/api_transfer github/fork/hbwx24/error_msg/cuda_kernel_error_msg github/fork/heavengate/cherry_yolo_box github/fork/heavengate/update_yolo_box github/fork/iclementine/rnn_fix github/fork/iducn/testestse github/fork/jczaja/prv-25537-fix github/fork/jeff41404/release/1.8 github/fork/jiweibo/api_2.0 github/fork/jiweibo/fix_lite_resnet50_test github/fork/juncaipeng/fix_doc_1 github/fork/lfchener/sample_code github/fork/littletomatodonkey/fix_reg_doc github/fork/liym27/dy2stat_update_assign_to_rc20 github/fork/luotao1/profiler_ut github/fork/mapingshuo/add_wait github/fork/mapingshuo/doc_2.0 github/fork/mapingshuo/zero-0.5 github/fork/miraiwk/dev github/fork/pangyoki/add-Categorical-class-branch github/fork/pangyoki/add-multinomial-op-branch github/fork/pangyoki/fix-test_distritbution-CI github/fork/qjing666/doublegrad github/fork/qjing666/fix_hdfs_download github/fork/sandyhouse/add_gather_etc github/fork/sandyhouse/add_send_recv_alltoall_etc github/fork/sandyhouse/pipeline_exe_run github/fork/seiriosPlus/feature/large_scale_kv_save_delta github/fork/seiriosPlus/fix/paddle_errors_fix github/fork/seiriosPlus/fix/paddle_op_errors github/fork/shangzhizhou/fix_test_activation_op_random_bug github/fork/smallv0221/yxp0924 github/fork/smallv0221/yxp0925 github/fork/swtkiwi/del-matplotlib github/fork/tianshuo78520a/kunlun_test github/fork/tianshuo78520a/update_dockerfile github/fork/wanghaoshuang/bert_fuse github/fork/wanghaoshuang/label_smooth github/fork/wanghuancoder/develop_CUDASynchronize github/fork/wanghuancoder/develop_Layer_doc github/fork/wanghuancoder/develop_ParameterList_doc github/fork/wanghuancoder/develop_Sequential_doc github/fork/wanghuancoder/develop_bilinear_tensor_product github/fork/wanghuancoder/develop_coverage_build_sh github/fork/wanghuancoder/develop_in_dynamic_mode_doc github/fork/wanghuancoder/develop_unique_name_doc github/fork/wangxicoding/fleet_meta_combine github/fork/wawltor/error_message_fix_5 github/fork/willthefrog/remove_l2_norm github/fork/windstamp/momentum_op github/fork/windstamp/mv_op_5 github/fork/windstamp/normal_api github/fork/wojtuss/wojtuss/fusion_gru_quantization github/fork/wojtuss/wojtuss/quantization-with-shift github/fork/wzzju/fix_err_info github/fork/wzzju/pure_fp16 github/fork/xiemoyuan/op_error_message github/fork/xiemoyuan/optimize_error_message github/fork/yaoxuefeng6/fix_doc github/fork/yaoxuefeng6/mod_dataset_v2 github/fork/yongqiangma/lod github/fork/ysh329/fix-clip-by-norm-error github/fork/ysh329/fix-error-clip-by-value github/fork/yukavio/error_info github/fork/zhangting2020/conv_filter_grad github/fork/zhangting2020/is_compile_with_cuda github/fork/zhangting2020/place_doc github/fork/zhangting2020/program github/fork/zhhsplendid/fix_any github/fork/zhhsplendid/refine_api2 github/fork/zhhsplendid/refine_api2_test github/fork/zhhsplendid/refine_api_test_ptb_lm github/fork/zhhsplendid/refine_api_test_resnet github/fork/zhhsplendid/refine_api_test_simnet github/fork/zhiqiu/dev/refine_initializer github/fork/zhiqiu/dev/remove_inplace_argument github/fork/zlsh80826/nvinfer_plugin_var_len_cuda11 hack_event improve_sccache incuabte/new_frl incubate/frl_train_eval incubate/infrt incubate/lite incubate/new_frl incubate/new_frl_rc incubate/stride inplace_addto layer_norm make_flag_adding_easier master matmul_double_grad move_embedding_to_phi move_histogram_to_pten move_sgd_to_phi move_slice_to_pten move_temporal_shift_to_phi move_yolo_box_to_phi npu_fix_alloc numel operator_opt paddle_tiny_install paralleltest pass-compile-eval-frame preln_ernie prv-disable-more-cache prv-md-even-more prv-onednn-2.5 prv-reshape-mkldnn-ut2 pten_tensor_refactor release-deleted/2.5 release-rc/2.5 release/0.11.0 release/0.12.0 release/0.13.0 release/0.14.0 release/0.15.0 release/1.0.0 release/1.1 release/1.2 release/1.3 release/1.4 release/1.5 release/1.6 release/1.7 release/1.8 release/2.0 release/2.0-alpha release/2.0-beta release/2.0-rc release/2.0-rc1 release/2.1 release/2.2 release/2.3 release/2.3-fc-ernie-fix release/2.4 release/2.5 release/lite-0.1 release/llm_2.5 revert-24981-add_device_attr_for_regulization revert-26856-strategy_example2 revert-27520-disable_pr revert-31068-fix_conv3d_windows revert-31562-mean revert-32290-develop-hardlabel revert-33037-forci revert-33475-fix_cifar_label_dimension revert-33630-bug-fix revert-34159-add_npu_bce_logical_dev revert-34406-add_copy_from_tensor revert-34910-spinlocks_for_allocator revert-35069-revert-34910-spinlocks_for_allocator revert-36057-dev/read_flags_in_ut revert-36201-refine_fast_threaded_ssa_graph_executor revert-36985-add_license revert-37318-refactor_dygraph_to_eager revert-37926-eager_coreops_500 revert-37956-revert-37727-pylayer_support_tuple revert-38100-mingdong revert-38301-allocation_rearrange_pr revert-38703-numpy_bf16_package_reupload revert-38732-remove_useless_header_in_elementwise_mul_grad revert-38959-Reduce_Grad revert-39143-adjust_empty revert-39227-move_trace_op_to_pten revert-39268-dev/remove_concat_fluid_kernel revert-40170-support_partial_grad revert-41056-revert-40727-move_some_activaion_to_phi revert-41065-revert-40993-mv_ele_floordiv_pow revert-41068-revert-40790-phi_new revert-41944-smaller_inference_api_test revert-42149-do-not-reset-default-stream-for-stream-safe-cuda-allocator revert-43155-fix_ut_tempfile revert-43882-revert-41944-smaller_inference_api_test revert-45808-phi/simplify_size_op revert-46827-deform_comment revert-47325-remove_cudnn_hardcode revert-47645-add_npu_storage_dims revert-48815-set_free_when_no_cache_hit_default_value_true revert-49499-test_ninja_on_ci revert-49654-prim_api_gen revert-49673-modify_get_single_cov revert-49763-fix_static_composite_gen revert-50158-fix_found_inf_bug_for_custom_optimizer revert-50188-refine_optimizer_create_accumulators revert-50335-fix_optminizer_set_auxiliary_var_bug revert-51676-flag_delete revert-51850-fix_softmaxce_dev revert-52175-dev_peak_memory revert-52186-deve revert-52523-test_py38 revert-52912-develop revert-53248-set_cmake_policy revert-54029-fix_windows_compile_bug revert-54068-support_translating_op_attribute revert-54214-modify_cmake_dependencies revert-54370-offline_pslib revert-54391-fix_cmake_md5error revert-54411-fix_cpp17_compile revert-54466-offline_pslib revert-54480-cmake-rocksdb revert-55568-fix_BF16_bug1 revert-56328-new_ir_support_vector_type_place_transfer revert-56366-fix_openssl_bug revert-56545-revert-56366-fix_openssl_bug revert-56620-fix_new_ir_ocr_bug revert-56925-check_inputs_grad_semantic revert-57005-refine_stride_flag rocm_dev_0217 sd_conv_linear_autocast semi-auto/rule-base support-0D-sort support_weight_transpose test_benchmark_ci test_feature_precision_test_c test_for_Filtetfiles test_model_benchmark test_model_benchmark_ci zhiqiu-patch-1 v2.5.1 v2.5.0 v2.5.0-rc1 v2.5.0-rc0 v2.4.2 v2.4.1 v2.4.0 v2.4.0-rc0 v2.3.2 v2.3.1 v2.3.0 v2.3.0-rc0 v2.2.2 v2.2.1 v2.2.0 v2.2.0-rc0 v2.2.0-bak0 v2.1.3 v2.1.2 v2.1.1 v2.1.0 v2.1.0-rc0 v2.0.2 v2.0.1 v2.0.0 v2.0.0-rc1 v2.0.0-rc0 v2.0.0-beta0 v2.0.0-alpha0 v1.8.5 v1.8.4 v1.8.3 v1.8.2 v1.8.1 v1.8.0 v1.7.2 v1.7.1 v1.7.0 v1.6.3 v1.6.2 v1.6.1 v1.6.0 v1.6.0-rc0 v1.5.2 v1.5.1 v1.5.0 v1.4.1 v1.4.0 v1.3.2 v1.3.1 v1.3.0 v1.2.1 v1.2.0 v1.1.0 v1.0.2 v1.0.1 v1.0.0 v1.0.0-rc0 v0.15.0 v0.15.0-rc0 v0.14.0 v0.13.0 v0.12.0 v0.11.1a2 v0.11.1a1 v0.11.0 lite-v0.1
7 合并请求!11636[IMPORTANT] MKLDNN layout: Support for sum operator,!8482Release/0.11.0,!8190Release/0.11.0,!8189Release/0.11.0,!6633给线性回归的get-started代码加上了预测的示例~~,!4615Feature/tensor array add python binding,!2596Memory CPU allocator
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
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. */
#pragma once
#include "paddle/memory/detail/system_allocator.h"
namespace paddle {
namespace memory {
namespace detail {
BuddyAllocator::BuddyAllocator(size_t pool_size, size_t max_pools,
SystemAllocator* system_allocator)
: pool_size_(pool_size),
max_pools_(max_pools),
system_allocator_(system_allocator) {
PADDLE_ASSERT(pool_size > 0);
PADDLE_ASSERT(max_pools > 0);
PADDLE_ASSERT(system_allocator != nullptr);
}
} // namespace detail
} // namespace memory
} // namespace paddle
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
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
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
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. */
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. */
#pragma once
......@@ -20,34 +20,38 @@ namespace paddle {
namespace memory {
namespace detail {
template<typename Allocator>
class BuddyAllocator {
public:
// TODO(gangliao): This is a draft, add Buddy Allocator Algorithm soon
BuddyAllocator() {}
~BuddyAllocator() {}
public:
void* Alloc(size_t size) {
return Allocator::Alloc(size);
}
void Free(void*) {
// Because all info like size are stored in meta data,
// thus it's duplicate if add the parameter `size` in
// `Free(void*)` interface.
}
size_t Used();
public:
BuddyAllocator(size_t pool_size, size_t max_pools,
SystemAllocator* system_allocator);
~BuddyAllocator();
void* Alloc(size_t size);
void Free(void*);
size_t Used();
private:
struct Block {
size_t size_;
Block* left_; // left buddy
Block* right_; // right buddy
};
// Initially, there is only one pool. If a Alloc founds not enough
// memory from that pool, and there has not been max_num_pools_,
// create a new pool by calling system_allocator_.Alloc(pool_size_).
std::vector<void*> pools_;
size_t pool_size_; // the size of each pool;
size_t max_num_pools_; // the size of all pools;
public:
BuddyAllocator(const BuddyAllocator&) = delete;
BuddyAllocator& operator=(const BuddyAllocator&) = delete;
SystemAllocator* system_allocator_;
private:
size_t min_alloc_size_;
size_t max_alloc_size_;
std::mutex mutex_;
private:
std::mutex mutex_;
// Disable copy and assignment.
BuddyAllocator(const BuddyAllocator&) = delete;
BuddyAllocator& operator=(const BuddyAllocator&) = delete;
};
BuddyAllocator<CPUAllocator>* GetCPUBuddyAllocator() {
......@@ -63,16 +67,16 @@ BuddyAllocator<CPUAllocator>* GetCPUBuddyAllocator() {
BuddyAllocator<GPUAllocator>* GetGPUBuddyAllocator(int gpu_id) {
static BuddyAllocator<GPUAllocator>** as = NULL;
if (as == NULL) {
int gpu_num = platform::GetDeviceCount();
int gpu_num = platform::GetDeviceCount();
as = new BuddyAllocator<GPUAllocator>*[gpu_num];
for (int gpu = 0; gpu < gpu_num; gpu++) {
as[gpu] = new BuddyAllocator<GPUAllocator>();
as[gpu] = new BuddyAllocator<GPUAllocator>();
}
}
return as[gpu_id];
}
#endif // PADDLE_ONLY_CPU
#endif // PADDLE_ONLY_CPU
} // namespace detail
} // namespace memory
......
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
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. */
#include "paddle/memory/detail/system_allocator.h"
#include <stdlib.h> // for malloc and free
#include <sys/mman.h> // for mlock and munlock
#include "gflags/gflags.h"
#include "paddle/platform/assert.h"
#include "paddle/platform/cuda.h"
// If use_pinned_memory is true, CPUAllocator calls mlock, which
// returns pinned and locked memory as staging areas for data exchange
// between host and device. Allocates too much would reduce the amount
// of memory available to the system for paging. So, by default, we
// should set false to use_pinned_memory.
DEFINE_bool(use_pinned_memory, false,
"If set, allocate cpu/gpu pinned memory.");
namespace paddle {
namespace memory {
namespace detail {
void* CPUAllocator::Alloc(size_t size) {
// According to http://www.cplusplus.com/reference/cstdlib/malloc/,
// malloc might not return nullptr if size is zero, but the returned
// pointer shall not be dereferenced -- so we make it nullptr.
if (size <= 0) return nullptr;
void* p = malloc(size);
if (p != nullptr && FLAGS_use_pinned_memory) {
mlock(p, size);
}
return p;
}
void CPUAllocator::Free(void* p, size_t size) {
if (p != nullptr && FLAGS_use_pinned_memory) {
munlock(p, size);
}
free(p);
}
#ifndef PADDLE_ONLY_CPU
void* GPUAllocator::Alloc(size_t size) {
// CUDA documentation doesn't explain if cudaMalloc returns nullptr
// if size is 0. We just make sure it does.
if (size <= 0) {
return nullptr;
}
void* p = 0;
cudaError_t result =
FLAGS_use_pinned_memory ? cudaMallocHost(&p, size) : cudaMalloc(&p, size);
if (result != cudaSuccess) {
cudaGetLastError(); // clear error if there is any.
}
return result == cudaSuccess ? p : nullptr;
}
void GPUAllocator::Free(void* p, size_t size) {
// Purposefully allow cudaErrorCudartUnloading, because
// that is returned if you ever call cudaFree after the
// driver has already shutdown. This happens only if the
// process is terminating, in which case we don't care if
// cudaFree succeeds.
cudaError_t err = FLAGS_use_pinned_memory ? cudaFreeHost(p) : cudaFree(p);
if (err != cudaErrorCudartUnloading) {
platform::throw_on_error(err, "cudaFree{Host} failed");
}
}
#endif // PADDLE_ONLY_CPU
} // namespace detail
} // namespace memory
} // namespace paddle
......@@ -22,10 +22,10 @@ limitations under the License. */
DECLARE_bool(use_pinned_memory);
void TestAllocator(paddle::memory::detail::SystemAllocator* a, size_t size) {
void TestAllocator(paddle::memory::detail::SystemAllocator& a, size_t size) {
bool freed = false;
{
void* p = a->Alloc(size);
void* p = a.Alloc(size);
if (size > 0) {
EXPECT_NE(p, nullptr);
} else {
......@@ -33,9 +33,9 @@ void TestAllocator(paddle::memory::detail::SystemAllocator* a, size_t size) {
}
int* i = static_cast<int*>(p);
std::shared_ptr<int> ptr(i, [&freed, a, size](void* p) {
std::shared_ptr<int> ptr(i, [&](void* p) {
freed = true;
a->Free(p, size);
a.Free(p, size);
});
}
EXPECT_TRUE(freed);
......@@ -44,28 +44,28 @@ void TestAllocator(paddle::memory::detail::SystemAllocator* a, size_t size) {
TEST(CPUAllocator, NoLockMem) {
FLAGS_use_pinned_memory = false;
paddle::memory::detail::CPUAllocator a;
TestAllocator(&a, 2048);
TestAllocator(&a, 0);
TestAllocator(a, 2048);
TestAllocator(a, 0);
}
TEST(CPUAllocator, LockMem) {
FLAGS_use_pinned_memory = true;
paddle::memory::detail::CPUAllocator a;
TestAllocator(&a, 2048);
TestAllocator(&a, 0);
TestAllocator(a, 2048);
TestAllocator(a, 0);
}
#ifndef PADDLE_ONLY_CPU
TEST(GPUAllocator, NoStaging) {
FLAGS_use_pinned_memory = false;
paddle::memory::detail::GPUAllocator a;
TestAllocator(&a, 2048);
TestAllocator(&a, 0);
TestAllocator(a, 2048);
TestAllocator(a, 0);
}
TEST(GPUAllocator, Staging) {
FLAGS_use_pinned_memory = true;
paddle::memory::detail::GPUAllocator a;
TestAllocator(&a, 2048);
TestAllocator(&a, 0);
TestAllocator(a, 2048);
TestAllocator(a, 0);
}
#endif // PADDLE_ONLY_CPU
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册
反馈
建议
客服 返回
顶部