search

PaddlePaddle / Paddle-Lite

PaddlePaddle High Performance Deep Learning Inference Engine for Mobile and Edge (飞桨高性能深度学习端侧推理引擎）
https://www.paddlepaddle.org.cn/lite
Apache License 2.0
6.93k stars 1.61k forks source link

[OpenCL] fix opencl multi-thread CL_INVALID_CONTEXT #10529

Closed wasupandceacar closed 2 months ago

wasupandceacar commented 2 months ago

PR devices

OpenCL

PR types

Bug fixes

PR changes

Backends

Description

Fix CL_INVALID_CONTEXT error when the model does predictions in multiple threads.

Previous issues: https://github.com/PaddlePaddle/Paddle-Lite/issues/9931 https://github.com/PaddlePaddle/Paddle-Lite/issues/10267

Caused by https://github.com/PaddlePaddle/Paddle-Lite/blob/bd60e696cc8233b0ebef3b4db932923d668e8c50/lite/backends/opencl/cl_runtime.cc#L27

Use local_thread here will lead to different opencl contexts in different threads, as mentioned in https://github.com/PaddlePaddle/Paddle-Lite/issues/9931#issuecomment-1407544996. Also, this is not a good solution, as discussed in https://github.com/PaddlePaddle/Paddle-Lite/issues/7888#issuecomment-990679917, which will cause other bugs.

New solution: Use single instance, all threads share a global CLRuntime.

paddle-bot[bot] commented 2 months ago

Thanks for your contribution!

wasupandceacar commented 2 months ago

@zhupengyang @hong19860320

wasupandceacar commented 2 months ago

@zhupengyang @hong19860320

wasupandceacar commented 2 months ago

@zhupengyang @hong19860320 一直没人啊老铁

wasupandceacar commented 2 months ago

@zhupengyang @hong19860320 ？

wasupandceacar commented 2 months ago

@zhupengyang @hong19860320 你们这还有没有人维护

wasupandceacar commented 2 months ago

@hong19860320 @zhupengyang 两周了还没人回复？

wasupandceacar commented 2 months ago

@hong19860320 @zhupengyang 是没人 review 得了还是怎么了，吱个声，我看比我新的 pr 都 merge 了

xiebaiyuan commented 2 months ago

建议测试一下两个线程同时跑两个OpenCL模型的情况。 隐约记得之前 @zhaoyang-star 添加这个是为了解决这个问题。