[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YgAztSuYrKXZBgJx@casper.infradead.org>
Date: Sun, 6 Feb 2022 20:46:45 +0000
From: Matthew Wilcox <willy@...radead.org>
To: Mike Rapoport <rppt@...nel.org>
Cc: linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 41/75] hexagon: Add pmd_pfn()
On Sun, Feb 06, 2022 at 08:13:11PM +0200, Mike Rapoport wrote:
> On Fri, Feb 04, 2022 at 07:58:18PM +0000, Matthew Wilcox (Oracle) wrote:
> > I need to use this function in common code, so define it for hexagon.
> >
> > Signed-off-by: Matthew Wilcox (Oracle) <willy@...radead.org>
> > ---
> > arch/hexagon/include/asm/pgtable.h | 3 ++-
>
> Why hexagon out of all architectures?
> What about m68k, nios2, nds32 etc?
Presumably they don't support CONFIG_TRANSPARENT_HUGEPAGE.
This code isn't compiled when THP are disabled; at least I haven't
had a buildbot complaint for any other architectures.
> > +#define pmd_pfn(pmd) (pmd_val(pmd) >> PAGE_SHIFT)
>
> I'd put it in include/linux/pgtable.h inside #ifndef pmd_pfn
That's completely upside down. pmd_pfn() should be defined by each
architecture (because generic code can't know anything about the format
of PMDs); if anything pmd_page() should be defined in linux/pgtable.h
in terms of pmd_pfn.
I'm not signing up to do that work as part of this series. That seems
like a distraction.
Powered by blists - more mailing lists