[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200110175519.myahx3tnxajt4eam@box>
Date: Fri, 10 Jan 2020 20:55:19 +0300
From: "Kirill A. Shutemov" <kirill@...temov.name>
To: David Hildenbrand <david@...hat.com>
Cc: Michal Hocko <mhocko@...nel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Dan Williams <dan.j.williams@...el.com>,
Vlastimil Babka <vbabka@...e.cz>, Mel Gorman <mgorman@...e.de>,
"Jin, Zhi" <zhi.jin@...el.com>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org,
"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>
Subject: Re: [PATCH] mm/page_alloc: Skip non present sections on zone
initialization
On Fri, Jan 10, 2020 at 03:56:14PM +0100, David Hildenbrand wrote:
> On 10.01.20 15:54, Kirill A. Shutemov wrote:
> > On Fri, Jan 10, 2020 at 03:48:39PM +0100, David Hildenbrand wrote:
> >>>> + if (!present_section_nr(section_nr))
> >>>> + return section_nr_to_pfn(next_present_section_nr(section_nr));
> >>>
> >>> This won't compile. next_present_section_nr() is static to mm/sparse.c.
> >>
> >> We should then move that to the header IMHO.
> >
> > It looks like too much for a trivial cleanup.
> >
>
> Cleanup? This is a performance improvement ("fix the issue."). We should
> avoid duplicating code where it can be avoided.
My original patch is in -mm tree and fixes the issue. The thread is about
tiding it up.
--
Kirill A. Shutemov
Powered by blists - more mailing lists