Comment 6 for bug 1821023
Revision history for this message
| Alfonso Sanchez-Beato (alfonsosanchezbeato) wrote | #6 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
The problem I have seen is that the first usage from userspace after boot of some HW triggers loading the firmware for that device. It is not the snap explicitly asking the kernel to load FW. The issue is that the kernel tries to get the firmware files looking at the filesystem layout seen by the process calling the kernel.
I have seen this bug in two apparently innocuous cases, from confined processes:
* Trying to connect to a wifi AP
* When using CUDA
If I try any of these things from a non-confined process, things are fine, if then I do the same things from the confined processes, things are fine too. The bug appears only in the first call when the drivers decide to load the FW.
The ideal way to resolve would be to prevent the kernel from using the confined process filesystem layout, but I am not sure if that is possible.