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

bad ../physical_package_id value #5

Open
StephaneRenouard opened this issue May 30, 2017 · 11 comments
Open

bad ../physical_package_id value #5

StephaneRenouard opened this issue May 30, 2017 · 11 comments

Comments

@StephaneRenouard
Copy link

StephaneRenouard commented May 30, 2017

When I cat:
cat /sys/devices/system/cpu/cpu0/topology/physical_package_id
I got a return value of -1, which is not correct.

The same for each cpu [0...3].

FYI, I'm using kernel 3.10.102

@avafinger
Copy link
Owner

physical_package_id: physical package id of cpu#.
Typically corresponds to a physical socket number, but the actual value is architecture and platform dependent.

Which values would you expect?
I just typed this in my Intel box and they all return 0

@StephaneRenouard
Copy link
Author

StephaneRenouard commented May 31, 2017

As this processor has 4 cores, I should expected 0,1,2,3 or 0,0,0,0 for cpu0, cpu1, cpu2 and cpu3.
Theses values are used for thread mapping in software.
I'm trying to run Aerospike software on BPI M64, and this param is checked at runtime. "-1" forbid the software to run.
I cheated with a fake physical_package_id file, but this could be a drawback in the futur.

@avafinger
Copy link
Owner

avafinger commented May 31, 2017

Can you please try this and see if still the same? It is kernel 3.10.105.
https://github.com/avafinger/bpi-m64-firmware-v2
PS: try with the SD CARD if you don't want to format your eMMC.

@avafinger
Copy link
Owner

Interesting read: https://patchwork.kernel.org/patch/9265455/

@avafinger
Copy link
Owner

Kernel 4.11
cat /sys/devices/system/cpu/cpu0/topology/physical_package_id
0

Will look into it when possible.

@StephaneRenouard
Copy link
Author

StephaneRenouard commented Jun 3, 2017

I'll take the time to try 3.10.105 this weekend. Did you try Kernel 4.11 on BPIM64 ?

@StephaneRenouard
Copy link
Author

Ok, I updated the board from 3.10.102 to 3.10.105.

Got the same issue.

@avafinger
Copy link
Owner

yeah, 3.10.x seems to get this value as default for ARM on A64. tested also on Armbian on another A64 (Pine64+). Mainline 4.11 (Armbian) was on other A64 board (OPI). Mainline 4.11.x work bare bone on A64, has no HDMI,CSI and Desktop but you get GbE. 4.11 seems the way to go in this case, but to get fully working, you may need to wait several months....

@StephaneRenouard
Copy link
Author

StephaneRenouard commented Jun 11, 2017

I'll take the time to try 3.10.105 this weekend. Did you try Kernel 4.11 on BPIM64 ?

Update: same issue with 3.10.105. I should have to wait for 4.11...

@avafinger
Copy link
Owner

If you care only about command line and connectivity (eth0), not sure it eMMC works in 4.11 but seems it is working (i still in the learning curve with 4.x), maybe i get it running in a few weeks, not a promise....

@StephaneRenouard
Copy link
Author

Take your time, i'll try the UP board next week from my side.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants