[ISSUE] Reproducible but not universal breakage with openblas 0.3.19 and numpy/scipy

[ISSUE] Reproducible but not universal breakage with openblas 0.3.19 and numpy/scipy

[ahesford]

[-- Attachment #1: Type: text/plain, Size: 1020 bytes --]

New issue by ahesford on void-packages repository

https://github.com/void-linux/void-packages/issues/34673

Description:
This issue tracks the issue described in 722054111e50649f3381906c9a646f44adb5cc02:

>Python code built around python3-numpy and python3-scipy (both linked with openblas) began emitting warnings after the 0.3.19 update:
>```python
>    /usr/lib/python3.10/concurrent/futures/thread.py:58:
>        RuntimeWarning: overflow encountered in multiply
>            result = self.fn(*self.args, **self.kwargs)
>```
>
>that broke program behavior in an Azure VM environment using Intel Xeon Platinum 8272CL CPUs; the problem was not seen on an AMD Ryzen 5 3600 or an Intel Core i7-9750H. Downgrading openblas while leaving the rest of the stack untouched eliminates the problem.
>
>Unfortunately, I do not yet have a proof of concept that can be shared publicly or more detail about the specific failure, but let's revert to the last known good version until a specific cause can be identified.

Re: Reproducible but not universal breakage with openblas 0.3.19 and numpy/scipy

[ahesford]

[-- Attachment #1: Type: text/plain, Size: 191 bytes --]

New comment by ahesford on void-packages repository

https://github.com/void-linux/void-packages/issues/34673#issuecomment-1000343893

Comment:
Worth looking into, thanks for digging this up

Re: Reproducible but not universal breakage with openblas 0.3.19 and numpy/scipy

[github-actions]

[-- Attachment #1: Type: text/plain, Size: 293 bytes --]

New comment by github-actions[bot] on void-packages repository

https://github.com/void-linux/void-packages/issues/34673#issuecomment-1159891949

Comment:
Issues become stale 90 days after last activity and are closed 14 days after that.  If this issue is still relevant bump it or assign it.

Re: [ISSUE] [CLOSED] Reproducible but not universal breakage with openblas 0.3.19 and numpy/scipy

[ahesford]

[-- Attachment #1: Type: text/plain, Size: 1023 bytes --]

Closed issue by ahesford on void-packages repository

https://github.com/void-linux/void-packages/issues/34673

Description:
This issue tracks the issue described in 722054111e50649f3381906c9a646f44adb5cc02:

>Python code built around python3-numpy and python3-scipy (both linked with openblas) began emitting warnings after the 0.3.19 update:
>```python
>    /usr/lib/python3.10/concurrent/futures/thread.py:58:
>        RuntimeWarning: overflow encountered in multiply
>            result = self.fn(*self.args, **self.kwargs)
>```
>
>that broke program behavior in an Azure VM environment using Intel Xeon Platinum 8272CL CPUs; the problem was not seen on an AMD Ryzen 5 3600 or an Intel Core i7-9750H. Downgrading openblas while leaving the rest of the stack untouched eliminates the problem.
>
>Unfortunately, I do not yet have a proof of concept that can be shared publicly or more detail about the specific failure, but let's revert to the last known good version until a specific cause can be identified.

Re: Reproducible but not universal breakage with openblas 0.3.19 and numpy/scipy

[ahesford]

[-- Attachment #1: Type: text/plain, Size: 188 bytes --]

New comment by ahesford on void-packages repository

https://github.com/void-linux/void-packages/issues/34673#issuecomment-1610175122

Comment:
This seems resolved by a subsequent update.