[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <3598255.1674307847@warthog.procyon.org.uk>
Date: Sat, 21 Jan 2023 13:30:47 +0000
From: David Howells <dhowells@...hat.com>
To: Christoph Hellwig <hch@...radead.org>
Cc: dhowells@...hat.com, Al Viro <viro@...iv.linux.org.uk>,
Matthew Wilcox <willy@...radead.org>,
Jens Axboe <axboe@...nel.dk>, Jan Kara <jack@...e.cz>,
Jeff Layton <jlayton@...nel.org>,
Logan Gunthorpe <logang@...tatee.com>,
linux-fsdevel@...r.kernel.org, linux-block@...r.kernel.org,
linux-kernel@...r.kernel.org, Christoph Hellwig <hch@....de>,
John Hubbard <jhubbard@...dia.com>, linux-mm@...ck.org
Subject: Re: [PATCH v7 2/8] iov_iter: Add a function to extract a page list from an iterator
Christoph Hellwig <hch@...radead.org> wrote:
> On Fri, Jan 20, 2023 at 05:55:50PM +0000, David Howells wrote:
> > (3) Any other sort of iterator.
> >
> > No refs or pins are obtained on the page, the assumption is made that
> > the caller will manage page retention. ITER_ALLOW_P2PDMA is not
> > permitted.
>
> The ITER_ALLOW_P2PDMA check here is fundamentally wrong and will
> break things like io_uring to fixed buffers on NVMe. ITER_ALLOW_P2PDMA
> should strictly increase the group of acceptable iters, it does must
> not restrict anything.
So just drop the check? Or do I actually need to do something to the pages to
make it work? If I understand the code correctly, it's not actually operable
right now on any page that doesn't have an appropriately marked PTE.
David
Powered by blists - more mailing lists