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

FatalError: Erroneous arithmetic operation is detected by the operating system. #908

Closed
bottleofwater11 opened this issue Nov 8, 2024 · 9 comments
Labels
bug Something isn't working

Comments

@bottleofwater11
Copy link

Description of the bug | 错误描述


C++ Traceback (most recent call last):

0 at::_ops::linear::call(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&)
1 at::native::linear(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&)
2 at::_ops::addmm::call(at::Tensor const&, at::Tensor const&, at::Tensor const&, c10::Scalar const&, c10::Scalar const&)
3 at::_ops::addmm::redispatch(c10::DispatchKeySet, at::Tensor const&, at::Tensor const&, at::Tensor const&, c10::Scalar const&, c10::Scalar const&)


Error Message Summary:

FatalError: Erroneous arithmetic operation is detected by the operating system.
[TimeInfo: *** Aborted at 1731057575 (unix time) try "date -d @1731057575" if you are using GNU date ***]
[SignalInfo: *** SIGFPE (@0x7f6db8afd914) received by PID 146315 (TID 0x7f6efa521480) from PID 18446744072513116436 ***]

How to reproduce the bug | 如何复现

magic-pdf -p kaiti.pdf -o . -m ocr 指令运行报错如上

Operating system | 操作系统

Linux

Python version | Python 版本

3.10

Software version | 软件版本 (magic-pdf --version)

0.9.x

Device mode | 设备模式

cuda

@bottleofwater11 bottleofwater11 added the bug Something isn't working label Nov 8, 2024
@myhloli
Copy link
Collaborator

myhloli commented Nov 8, 2024

看看是不是显存不足

@bottleofwater11
Copy link
Author

看看是不是显存不足

image
我指定的这张卡跑的

@bottleofwater11
Copy link
Author

看看是不是显存不足

image
这是跑的时候的状态,会不会是冲突了,上面还有部署的一些小模型

@myhloli
Copy link
Collaborator

myhloli commented Nov 8, 2024

H系列可能太新了,和cuda11.8不兼容

@bottleofwater11
Copy link
Author

H系列可能太新了,和cuda11.8不兼容

啊,之前就有过这个问题当时是0.8.1的版本,当时在V100可以跑,那后续会有优化么请问

@myhloli
Copy link
Collaborator

myhloli commented Nov 8, 2024

我看有人装了cu12的paddlegpu,说是能跑,你可以在issue里找下,我这边自测cu12的paddle好像和torch有冲突,实在不行就用cpu的paddle一样跑

@bottleofwater11
Copy link
Author

我看有人装了cu12的paddlegpu,说是能跑,你可以在issue里找下,我这边自测cu12的paddle好像和torch有冲突,实在不行就用cpu的paddle一样跑

行,cpu处理太慢了,当时测得一个表格处理要两分钟

@myhloli
Copy link
Collaborator

myhloli commented Nov 8, 2024

表格我们后面会加rapidtable,用cpu会比tablemaster快很多

@bottleofwater11
Copy link
Author

表格我们后面会加rapidtable,用cpu会比tablemaster快很多

好的好的

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants