Re: [PATCH v2] vmpressure: implement strict mode

[Luiz Capitulino]

On Tue, 2 Jul 2013 10:24:09 -0700
Anton Vorontsov <anton@xxxxxxxxxx> wrote:

> > Honestly, what Andrew suggested is the best design for me: apps
> > are notified on all events but the event name is sent to the application.
> 
> I am fine with this approach (or any other, I'm really indifferent to the
> API itself -- read/netlink/notification per file/whatever for the
> payload),

That's a very good thing because we've managed to agree on something :)

I'm also indifferent to the API, as long as we have 100% of the policy
in user-space. To me this means we do absolutely no filtering in the
kernel, which in turn means user-space gets all the events. Of course,
we need the event name as a payload.

Do we agree this solves all use-cases we have discussed so far?

> except that you still have the similar problem:
> 
>   read() old    read() new
>   --------------------------
>        "low"           "low"
>        "low"           "foo" -- the app does not know what does this mean
>        "med"           "bar" -- ditto

It can just ignore it, have a special handling, log it, fail or whatever.
That's the good of having the policy in user-space.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/