From: Jan Kara <jack@suse.cz>
To: Christoph Hellwig <hch@infradead.org>
Cc: John Hubbard <jhubbard@nvidia.com>,
Andrew Morton <akpm@linux-foundation.org>,
Jens Axboe <axboe@kernel.dk>,
Alexander Viro <viro@zeniv.linux.org.uk>,
Miklos Szeredi <miklos@szeredi.hu>,
"Darrick J . Wong" <djwong@kernel.org>,
Trond Myklebust <trond.myklebust@hammerspace.com>,
Anna Schumaker <anna@kernel.org>, Jan Kara <jack@suse.cz>,
David Hildenbrand <david@redhat.com>,
Logan Gunthorpe <logang@deltatee.com>,
linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-xfs@vger.kernel.org, linux-nfs@vger.kernel.org,
linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 4/7] iov_iter: new iov_iter_pin_pages*() routines
Date: Tue, 6 Sep 2022 12:21:06 +0200 [thread overview]
Message-ID: <20220906102106.q23ovgyjyrsnbhkp@quack3> (raw)
In-Reply-To: <Yxb7YQWgjHkZet4u@infradead.org>
On Tue 06-09-22 00:48:49, Christoph Hellwig wrote:
> On Tue, Sep 06, 2022 at 12:44:28AM -0700, John Hubbard wrote:
> > OK, that part is clear.
> >
> > > - for the pin case don't use the existing bvec helper at all, but
> > > copy the logic for the block layer for not pinning.
> >
> > I'm almost, but not quite sure I get the idea above. Overall, what
> > happens to bvec pages? Leave the get_page() pin in place for FOLL_GET
> > (or USE_FOLL_GET), I suppose, but do...what, for FOLL_PIN callers?
>
> Do not change anyhing for FOLL_GET callers, as they are on the way out
> anyway.
>
> For FOLL_PIN callers, never pin bvec and kvec pages: For file systems
> not acquiring a reference is obviously safe, and the other callers will
> need an audit, but I can't think of why it woul ever be unsafe.
Are you sure about "For file systems not acquiring a reference is obviously
safe"? I can see places e.g. in orangefs, afs, etc. which create bvec iters
from pagecache pages. And then we have iter_file_splice_write() which
creates bvec from pipe pages (which can also be pagecache pages if
vmsplice() is used). So perhaps there are no lifetime issues even without
acquiring a reference (but looking at the code I would not say it is
obvious) but I definitely don't see how it would be safe to not get a pin
to signal to filesystem backing the pagecache page that there is DMA
happening to/from the page.
Honza
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
next prev parent reply other threads:[~2022-09-06 10:21 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-31 4:18 [PATCH v2 0/7] convert most filesystems to pin_user_pages_fast() John Hubbard
2022-08-31 4:18 ` [PATCH v2 1/7] mm: change release_pages() to use unsigned long for npages John Hubbard
2022-08-31 4:18 ` [PATCH v2 2/7] mm/gup: introduce pin_user_page() John Hubbard
2022-09-06 6:37 ` Christoph Hellwig
2022-09-06 7:12 ` John Hubbard
2022-08-31 4:18 ` [PATCH v2 3/7] block: add dio_w_*() wrappers for pin, unpin user pages John Hubbard
2022-08-31 4:18 ` [PATCH v2 4/7] iov_iter: new iov_iter_pin_pages*() routines John Hubbard
2022-09-01 0:42 ` Al Viro
2022-09-01 1:48 ` John Hubbard
2022-09-06 6:47 ` Christoph Hellwig
2022-09-06 7:44 ` John Hubbard
2022-09-06 7:48 ` Christoph Hellwig
2022-09-06 7:58 ` John Hubbard
2022-09-07 8:50 ` Christoph Hellwig
2022-09-06 10:21 ` Jan Kara [this message]
2022-09-07 8:45 ` Christoph Hellwig
2022-09-14 3:51 ` Al Viro
2022-09-14 14:52 ` Jan Kara
2022-09-14 16:42 ` Al Viro
2022-09-15 8:16 ` Jan Kara
2022-09-16 1:55 ` Al Viro
2022-09-20 5:02 ` Al Viro
2022-09-22 14:36 ` Christoph Hellwig
2022-09-22 14:43 ` David Hildenbrand
2022-09-22 14:45 ` Christoph Hellwig
2022-09-22 2:22 ` Al Viro
2022-09-22 6:09 ` John Hubbard
2022-09-22 11:29 ` Jan Kara
2022-09-23 3:19 ` Al Viro
2022-09-23 4:05 ` John Hubbard
2022-09-23 8:39 ` Christoph Hellwig
2022-09-23 12:22 ` Jan Kara
2022-09-23 4:34 ` John Hubbard
2022-09-22 14:38 ` Christoph Hellwig
2022-09-23 4:22 ` Al Viro
2022-09-23 8:44 ` Christoph Hellwig
2022-09-23 16:13 ` Al Viro
2022-09-26 15:53 ` Christoph Hellwig
2022-09-26 19:55 ` Al Viro
2022-09-22 14:31 ` Christoph Hellwig
2022-09-22 14:36 ` Al Viro
2022-08-31 4:18 ` [PATCH v2 5/7] block, bio, fs: convert most filesystems to pin_user_pages_fast() John Hubbard
2022-09-06 6:48 ` Christoph Hellwig
2022-09-06 7:15 ` John Hubbard
2022-08-31 4:18 ` [PATCH v2 6/7] NFS: direct-io: convert to FOLL_PIN pages John Hubbard
2022-09-06 6:49 ` Christoph Hellwig
2022-09-06 7:16 ` John Hubbard
2022-08-31 4:18 ` [PATCH v2 7/7] fuse: convert direct IO paths to use FOLL_PIN John Hubbard
2022-08-31 10:37 ` Miklos Szeredi
2022-09-01 1:33 ` John Hubbard
2022-09-06 6:36 ` [PATCH v2 0/7] convert most filesystems to pin_user_pages_fast() Christoph Hellwig
2022-09-06 7:10 ` John Hubbard
2022-09-06 7:22 ` Christoph Hellwig
2022-09-06 7:37 ` John Hubbard
2022-09-06 7:46 ` Christoph Hellwig
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220906102106.q23ovgyjyrsnbhkp@quack3 \
--to=jack@suse.cz \
--cc=akpm@linux-foundation.org \
--cc=anna@kernel.org \
--cc=axboe@kernel.dk \
--cc=david@redhat.com \
--cc=djwong@kernel.org \
--cc=hch@infradead.org \
--cc=jhubbard@nvidia.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=miklos@szeredi.hu \
--cc=trond.myklebust@hammerspace.com \
--cc=viro@zeniv.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).