mock cuda runtime api
Project description
The plt hook technology used refers to plthook
mock pytorch cuda runtime interface
-
update submodule
git submodule update --init --recursive
-
build wheel package
pip wheel .
-
direct install
pip install .
collect cuda operator call stack
- find nvcc installed path
which nvcc
- replace nvcc with my nvcc
mv /usr/local/bin/nvcc /usr/local/bin/nvcc_b
chmod 777 tools/nvcc
cp tools/nvcc /usr/local/bin/nvcc
- build and install pytorch
- build and install cuda_mock
- import cuda_mock after import torch
- run your torch train script
- we will dump the stack into console
收集cuda 算子调用堆栈
- 找到nvcc安装路径
which nvcc
- 用我们的nvcc替换系统的nvcc(我们只是在编译选项加了
-g
)
mv /usr/local/bin/nvcc /usr/local/bin/nvcc_b
chmod 777 tools/nvcc
cp tools/nvcc /usr/local/bin/nvcc
- 构建并且安装pytorch
- 构建并且安装cuda_mock
- 注意要在import torch之后import cuda_mock
- 开始跑你的训练脚本
- 我们将会把堆栈打印到控制台
收集统计xpu runtime 内存分配信息/xpu_wait
调用堆栈
-
打印
xpu_malloc
调用序列,统计实时内存使用情况以及历史使用的峰值内存,排查内存碎片问题 -
打印
xpu_wait
调用堆栈,排查流水中断处问题 -
注意要在
import torch
/import paddle
之后import cuda_mock; cuda_mock.xpu_initialize()
-
使用方法:
import paddle import cuda_mock; cuda_mock.xpu_initialize() # 加入这一行
-
关闭打印backtrace(获取backtrace性能下降比较严重)
export HOOK_DISABLE_TRACE='xpuMemcpy=0,xpuSetDevice=0'
example
python test/test_import_mock.py
debug
- export LOG_LEVEL=0
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
cuda-mock-0.0.3.tar.gz
(4.5 kB
view hashes)
Built Distribution
Close
Hashes for cuda_mock-0.0.3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 83b13a7c3150270c20c6bf82615346bd7dd3c0e0bd8d28470672ad7313bac5b1 |
|
MD5 | 0c5319a46bbdf6c7579bc59af0fe44ab |
|
BLAKE2b-256 | 19cf96bfb330306180fa00aa7e6779cf0a4c0a6a9b211db0fa1294761e18ca06 |