linux-kernel - Re: [linux-next PATCH 1/2] mm: khugepaged: add exceed_max_ptes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 30 Apr 2020 18:09:03 -0700 (PDT)
From:   Hugh Dickins <hughd@...gle.com>
To:     Yang Shi <yang.shi@...ux.alibaba.com>
cc:     "Kirill A. Shutemov" <kirill@...temov.name>,
        kirill.shutemov@...ux.intel.com, hughd@...gle.com,
        aarcange@...hat.com, akpm@...ux-foundation.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [linux-next PATCH 1/2] mm: khugepaged: add exceed_max_ptes_*
 helpers

On Thu, 30 Apr 2020, Yang Shi wrote:
> On 4/30/20 2:59 PM, Kirill A. Shutemov wrote:
> > On Thu, Apr 30, 2020 at 06:56:21AM +0800, Yang Shi wrote:
> > > The max_ptes_{swap|none|shared} are defined to tune the behavior of
> > > khugepaged.  The are checked at a couple of places with open coding.
> > > Replace the opencoding to exceed_pax_ptes_{swap|none_shared} helpers to
> > > improve the readability.
> > > 
> > > Cc: Kirill A. Shutemov <kirill.shutemov@...ux.intel.com>
> > > Cc: Hugh Dickins <hughd@...gle.com>
> > > Cc: Andrea Arcangeli <aarcange@...hat.com>
> > > Signed-off-by: Yang Shi <yang.shi@...ux.alibaba.com>
> > > ---
> > >   mm/khugepaged.c | 27 +++++++++++++++++++++------
> > >   1 file changed, 21 insertions(+), 6 deletions(-)
> > > 
> > > diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> > > index a02a4c5..0c8d30b 100644
> > > --- a/mm/khugepaged.c
> > > +++ b/mm/khugepaged.c
> > > @@ -339,6 +339,21 @@ struct attribute_group khugepaged_attr_group = {
> > >   };
> > >   #endif /* CONFIG_SYSFS */
> > >   +static inline bool exceed_max_ptes_none(unsigned int *nr_ptes)
> > > +{
> > > +	return (++(*nr_ptes) > khugepaged_max_ptes_none);
> > > +}
> > > +
> > > +static inline bool exceed_max_ptes_swap(unsigned int *nr_ptes)
> > > +{
> > > +	return (++(*nr_ptes) > khugepaged_max_ptes_swap);
> > > +}
> > > +
> > > +static inline bool exceed_max_ptes_shared(unsigned int *nr_ptes)
> > > +{
> > > +	return (++(*nr_ptes) > khugepaged_max_ptes_shared);
> > > +}
> > > +
> > Frankly, I find this ugly and confusing. Open-coded version is more
> > readable to me.

Wow, yes, I strongly agree with Kirill.

> 
> I'm sorry you feel that way. I tend to agree that dereference looks not good.
> The open-coded version is not hard to understand to me either.
> 
> They are checked at a couple of different places with different variables,
> i.e. unmapped vs swap, and with different comparisons, > vs <=. I just
> thought the helpers with unified name started with "exceed_" may make it more
> self-explained and readable. Anyway this totally depends on taste and I
> really don't insist on this change.