Arch Linux

Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines

Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!
Tasklist

FS#31154 - [linux] enable CONFIG_PARAVIRT_TIME_ACCOUNTING

Attached to Project: Arch Linux
Opened by Christian Hesse (eworm) - Thursday, 16 August 2012, 15:11 GMT
Last edited by Tobias Powalowski (tpowa) - Tuesday, 28 August 2012, 10:08 GMT
Task Type Feature Request
Category Packages: Core
Status Closed
Assigned To Tobias Powalowski (tpowa)
Thomas Bächler (brain0)
Architecture All
Severity Very Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description:
I would like to have CONFIG_PARAVIRT_TIME_ACCOUNTING enabled in default kernel. This brings statistics about virtual machines to /proc/stat which is then used by htop and friends to display the infos.

Additional info:
linux 3.4.8-1
This task depends upon

Closed by  Tobias Powalowski (tpowa)
Tuesday, 28 August 2012, 10:08 GMT
Reason for closing:  Implemented
Additional comments about closing:  3.5.3-1
Comment by Thomas Bächler (brain0) - Thursday, 16 August 2012, 20:58 GMT
I am sceptical to enabling more options that give "a small performance impact", as the help states.
Comment by Christian Hesse (eworm) - Friday, 17 August 2012, 06:19 GMT
I tried to find some numbers, but did not get anything but "a small performance impact" so far.

The code in question lives in kernel/sched/core.c beginning on line 897. This does not look too scary though I am not sure when and how often this is called.

It's up to you whether or not to enable this.
So far I could live without that and if I really need it I can recompile the kernel on my own. ;)
Comment by Thomas Bächler (brain0) - Friday, 17 August 2012, 08:29 GMT
The point is, we enabled many debug options already that give "a small performance impact". I am worried that our kernel is pretty slow right now. I think I'll have to summarize all options with performance impacts and ask an expert about it.

Anyway, as the 3.5 series already started, and such options tend to break external modules a lot, I'd like to wait until the 3.6 series hits testing.

Loading...