Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1739792

Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings

From Michal Hocko <mhocko@kernel.org>
Newsgroups linux.kernel
Subject Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings
Date 2017-09-26 13:30 +0200
Message-ID <utWdR-46A-21@gated-at.bofh.it> (permalink)
References <ursae-6K5-21@gated-at.bofh.it> <utVhM-3uw-27@gated-at.bofh.it> <utVUu-3YA-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Tue 26-09-17 19:06:37, Yafang Shao wrote:
> 2017-09-26 18:25 GMT+08:00 Michal Hocko <mhocko@kernel.org>:
> > On Wed 20-09-17 06:43:35, Yafang Shao wrote:
> >> we can find the logic in domain_dirty_limits() that
> >> when dirty bg_thresh is bigger than dirty thresh,
> >> bg_thresh will be set as thresh * 1 / 2.
> >>       if (bg_thresh >= thresh)
> >>               bg_thresh = thresh / 2;
> >>
> >> But actually we can set vm background dirtiness bigger than
> >> vm dirtiness successfully. This behavior may mislead us.
> >> We'd better do this validity check at the beginning.
> >
> > This is an admin only interface. You can screw setting this up even
> > when you keep consistency between the background and direct limits. In
> > general we do not try to be clever for these knobs because we _expect_
> > admins to do sane things. Why is this any different and why do we need
> > to add quite some code to handle one particular corner case?
> >
> 
> Of course we expect admins to do the sane things, but not all admins
> are expert or faimilar with linux kernel source code.
> If we have to read the source code to know what is the right thing to
> do, I don't think this is a good interface, even for the admin.

Well, it is kind of natural to setup background below the direct limit
in general so I am not sure what is so surprising here. Moreover setting
a non default drity limits already requires some expertise. It is not
like an arbitrary value will work just fine...

> Anyway, there's no document on that direct limits should not less than
> background limits.

Then improve the documentation.

> > To be honest I am not entirely sure this is worth the code and the
> > future maintenance burden.
> I'm not sure if this code is a burden for the future maintenance, but
> I think that if we don't introduce this code it is a burden to the
> admins.

anytime we might need to tweak background vs direct limit we would have
to change these checks as well and that sounds like a maint. burden to
me.
-- 
Michal Hocko
SUSE Labs

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH v3] mm: introduce validity check on vm dirtiness settings Yafang Shao <laoar.shao@gmail.com> - 2017-09-19 17:00 +0200
  Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings Michal Hocko <mhocko@kernel.org> - 2017-09-26 12:30 +0200
    Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings Yafang Shao <laoar.shao@gmail.com> - 2017-09-26 13:10 +0200
      Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings Michal Hocko <mhocko@kernel.org> - 2017-09-26 13:30 +0200
        Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings Yafang Shao <laoar.shao@gmail.com> - 2017-09-26 13:50 +0200
          Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings Michal Hocko <mhocko@kernel.org> - 2017-09-26 14:00 +0200
            Re: [PATCH v3] mm: introduce validity check on vm dirtiness settings Jan Kara <jack@suse.cz> - 2017-09-26 15:40 +0200

csiph-web