Messages in this thread | | | Subject | Re: [PATCH] jump_label: jump_label for boot options. | From | Mike Galbraith <> | Date | Fri, 02 Dec 2011 06:34:05 +0100 |
| |
On Fri, 2011-12-02 at 09:28 +0900, KAMEZAWA Hiroyuki wrote: > On Thu, 1 Dec 2011 10:40:36 -0500 > Jason Baron <jbaron@redhat.com> wrote: > > > On Thu, Dec 01, 2011 at 11:53:53AM +0900, KAMEZAWA Hiroyuki wrote: > > > I tried to use jump_label for handling memcg's boot options which sets > > > global variable true/false and never changes after boot. And found jump_table > > > is larger than expected. This patch is a trial to allow to place jump_table > > > in .init section. How do you think ? > > > > > > > Remeber too, that 'static_branch()' is inherently biased. That is, the > > 'false' path is assumed to be the the most likely path. Thus, the 'true' > > path is move out-of-line. Thus, if the 'true' branch is potentially > > used all the time, we would want to make sure that the savings of not > > having to check a variable is still worth it. I should probably rename > > static_branch() -> 'static_branch_default_false()' to make that clear. > > > Thank you for pointing out. > > My assumption is that it's disabled at boot by few people because sane > people will not config memcg if they don't want. Most of distro users > will not turn off it...
I think that's a bad assumption. I see a lot of cgroup_disable=memory command lines, and _always_ add it to grub when installing "everything plus the kitchen sink" distro kernels on my boxen.
-Mike
| |