linux-kernel - Re: [PATCH] MAINTAINERS: add mm GUP section

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

Message-ID: <7deb77f8-6ee6-48c9-a579-4d76393c0532@lucifer.local>
Date: Mon, 12 May 2025 14:10:41 +0100
From: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
To: David Hildenbrand <david@...hat.com>
Cc: Zi Yan <ziy@...dia.com>, Vlastimil Babka <vbabka@...e.cz>,
        Uladzislau Rezki <urezki@...il.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Jason Gunthorpe <jgg@...dia.com>, John Hubbard <jhubbard@...dia.com>,
        Peter Xu <peterx@...hat.com>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] MAINTAINERS: add mm GUP section

On Mon, May 12, 2025 at 03:01:30PM +0200, David Hildenbrand wrote:
> On 12.05.25 14:54, Zi Yan wrote:
> > On 12 May 2025, at 3:38, Vlastimil Babka wrote:
> >
> > > On 5/8/25 14:23, Lorenzo Stoakes wrote:
> > > > >
> > > > > M for "PAGE ALLOCATOR", hmmm ..., I was hoping that Vlastimil might have
> > > > > capacity for that? :)
> > > >
> > > > Vlastimil? ;)
> > > >
> > > > I'd certainly support this.
> > >
> > > OK, can do, thanks.
> > >
> > > > >
> > > > >
> > > > >
> > > > > Not 100% sure what to do with
> > > > >
> > > > > * include/linux/page_isolation.h
> > > > > * mm/page_isolation.c
> > > > >
> > > > > (I hate the word "page isolation")
> > > > >
> > > > > They are mostly about page migration (either for alloc_contig... or memory
> > > > > hotunplug). Likely they should either go to the MIGRATION section or to the
> > > > > PAGE ALLOCATOR? Maybe MIGRATION makes more sense. Thoughts?
> > > >
> > > > I mean it explicitly relates to migrate type and migration so seems to me
> > > > it ought to be in migration.
> > > >
> > > > Though migrate type + the machinary around it is a product of the physical
> > > > page allocator (I even cover it in the 'physical memory' section of the
> > > > book).
> > > >
> > > > I wonder if our soon-to-be page allocator maintainer Vlastimil has
> > > > thoughts? ;)
> > > >
> > > > I'd vote for migration though to be honest.
> > >
> > > I checked the code briefly and although migratetypes are related to
> > > migration, it seems rather page allocator code to me.
> > >
> > > In fact if I didn't miss these files, I would have included them when
> > > proposing the PAGE ALLOCATOR section.
> > > Zi Yan has a series on that topic now and is one of the R: in PAGE
> > > ALLOCATOR. What do you think?
> >
> > I agree with Vlastimil that these two files belong to PAGE ALLOCATOR
> > section. Page isolation (actually should be pageblock isolation) is
> > doing work on pageblock migratetype, which IMHO is an important part
> > of anti-fragmentation mechanism for page allocation.
>
> IIRC, it's a bit confusing, because pageblock isolation as in
> mm/page_isolation.c does not have a lot to do with anti-fragmentation in
> reality.
>
> All of these functions should primarily be used for memory offlining +
> alloc_contig. (where we try page migration afterwards)
>
> Anyhow, I am fine as long as these files live somewhere related :)

Yeah, I think key thing is to get them in _somewhere_ vaguely sensible, and
we can figure out fixing up this kind of thing after :) which I think
aligns with what you're saying here!

>
> --
> Cheers,
>
> David / dhildenb
>