From: Josef Bacik <josef@toxicpanda.com>
To: Joanne Koong <joannelkoong@gmail.com>
Cc: miklos@szeredi.hu, linux-fsdevel@vger.kernel.org,
bernd.schubert@fastmail.fm, jefflexu@linux.alibaba.com,
willy@infradead.org, shakeel.butt@linux.dev,
kernel-team@meta.com
Subject: Re: [PATCH 04/12] fuse: support large folios for non-writeback writes
Date: Thu, 21 Nov 2024 17:32:13 -0500 [thread overview]
Message-ID: <20241121223213.GE1974911@perftesting> (raw)
In-Reply-To: <20241109001258.2216604-5-joannelkoong@gmail.com>
On Fri, Nov 08, 2024 at 04:12:50PM -0800, Joanne Koong wrote:
> Add support for folios larger than one page size for non-writeback
> writes.
>
> Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
> ---
> fs/fuse/file.c | 29 ++++++++++++++++++-----------
> 1 file changed, 18 insertions(+), 11 deletions(-)
>
> diff --git a/fs/fuse/file.c b/fs/fuse/file.c
> index a89fdc55a40b..6ee23ab9b7f2 100644
> --- a/fs/fuse/file.c
> +++ b/fs/fuse/file.c
> @@ -1146,19 +1146,15 @@ static ssize_t fuse_fill_write_pages(struct fuse_io_args *ia,
> num = min(num, max_pages << PAGE_SHIFT);
>
> ap->args.in_pages = true;
> - ap->descs[0].offset = offset;
>
> while (num) {
> size_t tmp;
> struct folio *folio;
> pgoff_t index = pos >> PAGE_SHIFT;
> - unsigned int bytes = min(PAGE_SIZE - offset, num);
> -
> - again:
> - err = -EFAULT;
> - if (fault_in_iov_iter_readable(ii, bytes))
> - break;
> + unsigned int bytes;
> + unsigned int folio_offset;
>
> + again:
> folio = __filemap_get_folio(mapping, index, FGP_WRITEBEGIN,
> mapping_gfp_mask(mapping));
> if (IS_ERR(folio)) {
> @@ -1166,10 +1162,20 @@ static ssize_t fuse_fill_write_pages(struct fuse_io_args *ia,
> break;
> }
>
> + folio_offset = ((index - folio->index) << PAGE_SHIFT) + offset;
> + bytes = min(folio_size(folio) - folio_offset, num);
> +
> + err = -EFAULT;
> + if (fault_in_iov_iter_readable(ii, bytes)) {
> + folio_unlock(folio);
> + folio_put(folio);
> + break;
> + }
This is a deadlock potentially. mmap the file, then buffered write to it with
the mmaped region and you will deadlock because fault_in_iov_iter_readable()
will try to lock the folio to submit the read, and you're already holding the
lock from the __filemap_get_folio() call.
If you look at iomap_write_iter it does a the fault_in_iov_iter_readable() call
before getting the folio, and it just uses the mapping_max_folio_size() value to
fault in larger chunks for whatever folio we manage to grab. Thanks,
Josef
next prev parent reply other threads:[~2024-11-21 22:32 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-09 0:12 [PATCH 00/12] fuse: support large folios Joanne Koong
2024-11-09 0:12 ` [PATCH 01/12] fuse: support copying " Joanne Koong
2024-11-21 22:27 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 02/12] fuse: support large folios for retrieves Joanne Koong
2024-11-21 22:28 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 03/12] fuse: refactor fuse_fill_write_pages() Joanne Koong
2024-11-21 22:28 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 04/12] fuse: support large folios for non-writeback writes Joanne Koong
2024-11-12 17:32 ` Joanne Koong
2024-11-13 18:41 ` Joanne Koong
2024-11-21 22:32 ` Josef Bacik [this message]
2024-11-09 0:12 ` [PATCH 05/12] fuse: support large folios for folio reads Joanne Koong
2024-11-22 14:42 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 06/12] fuse: support large folios for symlinks Joanne Koong
2024-11-22 14:43 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 07/12] fuse: support large folios for stores Joanne Koong
2024-11-22 14:45 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 08/12] fuse: support large folios for queued writes Joanne Koong
2024-11-22 14:45 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 09/12] fuse: support large folios for readahead Joanne Koong
2024-11-22 14:47 ` Josef Bacik
2024-11-25 19:23 ` Joanne Koong
2024-11-09 0:12 ` [PATCH 10/12] fuse: support large folios for direct io Joanne Koong
2024-11-22 14:49 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 11/12] fuse: support large folios for writeback Joanne Koong
2024-11-22 14:50 ` Josef Bacik
2024-11-09 0:12 ` [PATCH 12/12] fuse: enable large folios Joanne Koong
2024-11-22 14:51 ` Josef Bacik
2024-11-09 0:22 ` [PATCH 00/12] fuse: support " Joanne Koong
2024-11-09 0:32 ` Bernd Schubert
2024-11-11 17:44 ` Joanne Koong
2024-11-13 18:58 ` Joanne Koong
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=20241121223213.GE1974911@perftesting \
--to=josef@toxicpanda.com \
--cc=bernd.schubert@fastmail.fm \
--cc=jefflexu@linux.alibaba.com \
--cc=joannelkoong@gmail.com \
--cc=kernel-team@meta.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=miklos@szeredi.hu \
--cc=shakeel.butt@linux.dev \
--cc=willy@infradead.org \
/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