Re: [PATCH v10 3/3] mm: add anonymous vma name refcounting

[Suren Baghdasaryan <surenb@google.com>]

On Fri, Oct 8, 2021 at 7:57 AM Michal Hocko <mhocko@suse.com> wrote:
>
> On Fri 08-10-21 07:14:58, Dave Hansen wrote:
> > On 10/7/21 11:34 PM, Michal Hocko wrote:
> > >> Yes, please. It really seems like the folks that are interested in this
> > >> feature want strings. (I certainly do.)
> > > I am sorry but there were no strong arguments mentioned for strings so
> > > far.
> >
> > The folks who want this have maintained an out-of-tree patch using
> > strings.  They've maintained it for the better part of a decade.  I
> > don't know how widely this shipped in the Android ecosystem, but I
> > suspect we're talking about billions of devices.  Right?

Correct.

> >
> > This is a feature that, if accepted into mainline, will get enabled and
> > used on billions of devices.  If we dumb this down to integers, it's not
> > 100% clear that it _will_ get used.

Not as is and not with some major changes in the userspace, which
relied on a simple interface: set a name to a vma, observe that name
in the /proc/$pid/maps.

> >
> > That's a pretty strong argument in my book, even if the contributors
> > have difficulty articulating exactly why they want strings.
>
> I would agree that if integers would make this unusable then this would
> be a strong argument. But I haven't really heard any arguments like that
> so far. I have heard about IPC overhead and other speculations that do
> not seem really convincing. We shouldn't hand wave concerns regarding
> the implementation complexity and resource handling just by "somebody
> has been using this for decates", right?
>
> Do not get me wrong. This is going to become a user interface and we
> will have to maintain it for ever. As such an extra scrutiny has to be
> applied.

I don't know how to better articulate this. IPC transactions on
Android cannot be scheduled efficiently. We're going to have to stall
after mmap, make binder transaction, schedule a new process, get the
ID, make binder reply, schedule back to the original thread, resume.
Doing this potentially for every mmap is a non-starter. Deferring this
job is possible but we still have to do all this work, so it still
requires cpu cycles and power, not mentioning the additional
complexity in the userspace. I'm adding a rep from the performance
team, maybe Tim can explain this better.

There were a couple suggestions on using filesystem/memfd for naming
purposes which I need to explore but if that works the approach will
likely not involve any IDs. We want human-readable names in the maps
file, not a number.

Thanks for all the feedback and ideas!

> --
> Michal Hocko
> SUSE Labs