Messages in this thread | | | Date | Tue, 2 Jul 2013 17:55:19 -0400 | From | Luiz Capitulino <> | Subject | Re: [PATCH] vmpressure: implement strict mode |
| |
On Tue, 2 Jul 2013 21:47:03 +0200 Pavel Machek <pavel@ucw.cz> wrote:
> On Tue 2013-07-02 11:06:28, Luiz Capitulino wrote: > > On Mon, 1 Jul 2013 10:51:03 +0200 > > Pavel Machek <pavel@ucw.cz> wrote: > > > > > Hi! > > > > > > > diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt > > > > index ddf4f93..3c589cf 100644 > > > > --- a/Documentation/cgroups/memory.txt > > > > +++ b/Documentation/cgroups/memory.txt > > > > @@ -807,12 +807,14 @@ register a notification, an application must: > > > > > > > > - create an eventfd using eventfd(2); > > > > - open memory.pressure_level; > > > > -- write string like "<event_fd> <fd of memory.pressure_level> <level>" > > > > +- write string like "<event_fd> <fd of memory.pressure_level> <level> [strict]" > > > > to cgroup.event_control. > > > > > > > > > > This is.. pretty strange interface. Would it be cleaner to do ioctl()? > > > New syscall? > > > > Are you referring to my new mode or to the whole thing? > > Well. The interface was already very strange and you made it even > worse.
The existing interface is the cgroup's notification mechanism, I think discussing it is a bit out of scope for my extension.
Now, regarding my extension itself and the current vmpressure API, I believe that delivering all events to user-space (ie. w/o any filtering in the kernel) is a better solution.
Point is whether we can do it with the current vmpressure API (which is cgroup based) or whether we should move to something else.
| |