Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

arm源码安装paddle2.6报错void* memcpy(void*, const void*, size_t) #63852

Open
ada1114 opened this issue Apr 25, 2024 · 4 comments
Open

arm源码安装paddle2.6报错void* memcpy(void*, const void*, size_t) #63852

ada1114 opened this issue Apr 25, 2024 · 4 comments
Assignees
Labels
status/new-issue 新建 type/build 编译/安装问题

Comments

@ada1114
Copy link

ada1114 commented Apr 25, 2024

问题描述 Issue Description

按照https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/install/compile/arm-compile.html这篇文档在python3.10.14上安装paddlepaddle。
前面的步骤执行很顺利,执行到步骤9
9.使用以下命令来编译,注意,因为处理器为 ARM 架构,如果不加TARGET=ARMV8则会在编译的时候报错。
make TARGET=ARMV8 -j$(nproc)
报错:error: ‘void* memcpy(void*, const void*, size_t)’ copying an object of non-trivial type ‘Eigen::internal::Packet4c’ {aka ‘struct Eigen::internal::eigen_packet_wrapper<int, 2>’} from an array of ‘const int8_t’ {aka ‘const signed char’} [-Werror=class-memaccess]

企业微信截图_1714008182400

版本&环境信息 Version & Environment Information


Paddle version: N/A
Paddle With CUDA: N/A

OS: debian 12
GCC version: (Debian 12.2.0-14) 12.2.0
Clang version: N/A
CMake version: version 3.16.8
Libc version: glibc 2.36
Python version: 3.10.14

CUDA version: N/A
cuDNN version: N/A
Nvidia driver version: N/A
Nvidia driver List: N/A


@ada1114 ada1114 added status/new-issue 新建 type/build 编译/安装问题 labels Apr 25, 2024
@ada1114
Copy link
Author

ada1114 commented Apr 25, 2024

很多这样的报错
企业微信截图_17140120048722

@lijialin03
Copy link
Contributor

您好,我们正在联系相关开发同学进行复现,复现后会尽快联系您,感谢您的反馈!

@qili93
Copy link
Contributor

qili93 commented Apr 26, 2024

您好,https://www.paddlepaddle.org.cn/documentation/docs/zh/install/compile/arm-compile.html 这篇文章里面对ARM的原吗编译有点老了。您如果需要公版ARM的 PaddlePaddle ARM CPU 的安装包,可以从 https://www.paddlepaddle.org.cn/whl/linux/cpu-mkl/develop.html 这个链接中直接下载。

如果您需要源码编译PaddlePaddle ARM CPU的安装包,可以参考如下步骤进行编译,建议使用飞桨的开发镜像,这样可以避免很多因为环境差异导致的编译问题。

# 启动容器镜像命令参考
docker pull registry.baidubce.com/device/paddle-cpu:ubuntu20-$(uname -m)-gcc84-py310
docker run -dit --name paddle-cpu-dev -v $(pwd):/work \
  --network=host --shm-size=128G --workdir=/work \
  --cap-add=SYS_PTRACE --security-opt seccomp=unconfined \
  registry.baidubce.com/device/paddle-cpu:ubuntu20-$(uname -m)-gcc84-py310 /bin/bash

# 下载源码,默认 develop 分支
git clone https://github.com/PaddlePaddle/Paddle.git
cd Paddle

# 创建编译目录
mkdir build && cd build

# 源码编译cmake命令参考
cmake .. -DCMAKE_EXPORT_COMPILE_COMMANDS=ON \
  -DPY_VERSION=3.10 -DPYTHON_EXECUTABLE=`which python3` \
  -DWITH_TESTING=OFF -DON_INFER=ON -DWITH_DISTRIBUTE=ON -DWITH_ARM=ON

# 编译命令
make TARGET=ARMV8 -j16

@ada1114
Copy link
Author

ada1114 commented Apr 28, 2024

@qili93 您好,我按照您的说明分别试了一下直接执行安装包和源码,仍然都报错。
其中:
直接执行安装包报错信息
直接执行安装包报错

源码安装报错
源码安装报错1_17142732022799
新报错企业微信截图_17142725037631
源码安装报错3_17142732677168
cmake报错完整信息.txt

想问下,这个安装步骤在我上面发的环境下测试过吗?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
status/new-issue 新建 type/build 编译/安装问题
Projects
None yet
Development

No branches or pull requests

3 participants