[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHk-=wj6aTsqq6BAUci-NYJ3b-EkDwVgz_NvW_kW8KBqGocouQ@mail.gmail.com>
Date: Mon, 28 Sep 2020 09:17:09 -0700
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Jason Gunthorpe <jgg@...pe.ca>
Cc: Leon Romanovsky <leonro@...dia.com>, Peter Xu <peterx@...hat.com>,
John Hubbard <jhubbard@...dia.com>,
Linux-MM <linux-mm@...ck.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Jan Kara <jack@...e.cz>, Michal Hocko <mhocko@...e.com>,
Kirill Tkhai <ktkhai@...tuozzo.com>,
Kirill Shutemov <kirill@...temov.name>,
Hugh Dickins <hughd@...gle.com>,
Christoph Hellwig <hch@....de>,
Andrea Arcangeli <aarcange@...hat.com>,
Oleg Nesterov <oleg@...hat.com>, Jann Horn <jannh@...gle.com>
Subject: Re: [PATCH 1/5] mm: Introduce mm_struct.has_pinned
On Mon, Sep 28, 2020 at 5:49 AM Jason Gunthorpe <jgg@...pe.ca> wrote:
>
> Not seeing an obvious option besides adding a smp_mb() before
> page_maybe_dma_pinned() as Peter once suggested.
That is going to be prohibitively expensive - needing it for each pte
whether it's pinned or not.
I really think the better option is a "don't do that then". This has
_never_ worked before either except by pure luck.
I also doubt anybody does it. forking with threads is a bad idea to
begin with. Doing so while pinning pages even more so.
Linus
Powered by blists - more mailing lists