Re: [PATCH] mm/ksm: introduce ksm_enabled for each process

[Michal Hocko <mhocko@suse.com>]

On Thu 19-05-22 08:02:10, CGEL wrote:
> On Thu, May 19, 2022 at 09:35:30AM +0200, Michal Hocko wrote:
> > On Thu 19-05-22 06:23:30, CGEL wrote:
> > > On Wed, May 18, 2022 at 02:12:26PM +0200, Michal Hocko wrote:
> > > > On Wed 18-05-22 02:47:06, CGEL wrote:
> > > > > On Tue, May 17, 2022 at 04:04:38PM +0200, Michal Hocko wrote:
> > > > > > [CCing Hugh and linux-api]
> > > > > > 
> > > > > > On Tue 17-05-22 09:27:01, cgel.zte@gmail.com wrote:
> > > > > > per mm but the actual implementation currently relies on the per-vma
> > > > > > flags. That means that one can explicitly disallow merging by madvise
> > > > > > for a range. Is it wise to override that by a per-process knob? I mean
> > > > > > there might be a very good reason why a particular memory ranges should
> > > > > > never be merged but a per-process knob could easily ignore that hint
> > > > > > from the application. Or am I just confuse?
> > > > > For now, there is no any hints for letting KSM never merge some memory
> > > > > ranges.
> > > > 
> > > > I am not sure I understand. Could you be more specific?
> > > 
> > > Not like THP, KSM doesn't have anything like VM_NOHUGEPAGE, so apps
> > > cann't explicitly disallow merging by madvise. If it is really necessary for
> > > a particular meory ranges of a process to be never merged, we have to submit
> > > one more patch to achieve that.
> > 
> > What about MADV_UNMERGEABLE?
> 
> MADV_UNMERGEABLE and MADV_MERGEABLE usually appear in pairs, MADV_UNMERGEABLE cannot
> appear alone.

That might be the case currently because KSM is an opt-in feature that
has to be explicitly enabled. The existing interface only allows to
enable it by MADV_MERGEABLE but now you are proposing an extension when
there would be other way to achieve the same (with a wider scope but
that is not really all that important). MADV_UNMERGEABLE has a well
defined behavior even on VMAs which are not marked for merging.

Let's say that somebody would like to use a process wide setup except
for few special mappings because merging is not really desirable for
whatever reason. How do you achieve that?

> I mean MADV_UNMERGEABLE is used to unmerges whatever it merged in the
> specifed range, not to disallow merging.

I disagree. It clearly drops the mergeable flag so it effectivelly disallow
merging.
-- 
Michal Hocko
SUSE Labs