Re: [ovirt-devel] [VDSM] Profiling results - 144 cores, 144 vms

Thu, 18 Feb 2016 04:15:59 -0800 

> 1. Why _rtnl_link_alloc_cache is so expensive?
>     called only 389 times, but the most expensive call
Funny enough, we use ctype binding there to be fast. I'll take a look
what can be done there.
_______________________________________________
Devel mailing list
[email protected]
http://lists.ovirt.org/mailman/listinfo/devel

Reply via email to