From: Matthew Wilcox <willy@infradead.org>
To: Chao Yu <chao@kernel.org>
Cc: "Vishal Moola (Oracle)" <vishal.moola@gmail.com>,
linux-kernel@vger.kernel.org,
linux-f2fs-devel@lists.sourceforge.net, linux-mm@kvack.org,
fengnan chang <fengnanchang@gmail.com>,
linux-fsdevel@vger.kernel.org
Subject: Re: [f2fs-dev] [PATCH v3 14/23] f2fs: Convert f2fs_write_cache_pages() to use filemap_get_folios_tag()
Date: Tue, 29 Nov 2022 19:14:13 +0000 [thread overview]
Message-ID: <Y4ZaBd1r45waieQs@casper.infradead.org> (raw)
In-Reply-To: <9c01bb74-97b3-d1c0-6a5f-dc8b11113e1a@kernel.org>
On Mon, Nov 14, 2022 at 03:02:34PM +0800, Chao Yu wrote:
> On 2022/10/18 4:24, Vishal Moola (Oracle) wrote:
> > Converted the function to use a folio_batch instead of pagevec. This is in
> > preparation for the removal of find_get_pages_range_tag().
> >
> > Also modified f2fs_all_cluster_page_ready to take in a folio_batch instead
> > of pagevec. This does NOT support large folios. The function currently
>
> Vishal,
>
> It looks this patch tries to revert Fengnan's change:
>
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=01fc4b9a6ed8eacb64e5609bab7ac963e1c7e486
>
> How about doing some tests to evaluate its performance effect?
>
> +Cc Fengnan Chang
Thanks for reviewing this. I think the real solution to this is
that f2fs should be using large folios. That way, the page cache
will keep track of dirtiness on a per-folio basis, and if your folios
are at least as large as your cluster size, you won't need to do the
f2fs_prepare_compress_overwrite() dance. And you'll get at least fifteen
dirty folios per call instead of fifteen dirty pages, so your costs will
be much lower.
Is anyone interested in doing the work to convert f2fs to support
large folios? I can help, or you can look at the work done for XFS,
AFS and a few other filesystems.
next prev parent reply other threads:[~2022-11-29 19:14 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-17 20:24 [PATCH v3 00/23] Convert to filemap_get_folios_tag() Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 01/23] pagemap: Add filemap_grab_folio() Vishal Moola (Oracle)
2022-10-24 19:36 ` Vishal Moola
2022-10-24 19:38 ` Matthew Wilcox
2022-10-17 20:24 ` [PATCH v3 02/23] filemap: Added filemap_get_folios_tag() Vishal Moola (Oracle)
2022-10-24 19:42 ` Matthew Wilcox
2022-10-17 20:24 ` [PATCH v3 03/23] filemap: Convert __filemap_fdatawait_range() to use filemap_get_folios_tag() Vishal Moola (Oracle)
2022-10-24 20:06 ` Matthew Wilcox
2022-10-17 20:24 ` [PATCH v3 04/23] page-writeback: Convert write_cache_pages() " Vishal Moola (Oracle)
2022-10-24 20:12 ` Matthew Wilcox
2022-10-17 20:24 ` [PATCH v3 05/23] afs: Convert afs_writepages_region() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 06/23] btrfs: Convert btree_write_cache_pages() to use filemap_get_folio_tag() Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 07/23] btrfs: Convert extent_write_cache_pages() to use filemap_get_folios_tag() Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 08/23] ceph: Convert ceph_writepages_start() " Vishal Moola (Oracle)
2022-10-28 17:20 ` Jeff Layton
2022-10-17 20:24 ` [PATCH v3 09/23] cifs: Convert wdata_alloc_and_fillpages() " Vishal Moola (Oracle)
2022-10-23 14:01 ` kernel test robot
2022-10-17 20:24 ` [PATCH v3 10/23] ext4: Convert mpage_prepare_extent_to_map() " Vishal Moola (Oracle)
2022-10-24 19:26 ` Vishal Moola
2022-10-17 20:24 ` [PATCH v3 11/23] f2fs: Convert f2fs_fsync_node_pages() " Vishal Moola (Oracle)
2022-10-24 19:31 ` Vishal Moola
2022-11-10 18:51 ` Vishal Moola
2022-10-29 4:46 ` [f2fs-dev] " Chao Yu
2022-10-17 20:24 ` [PATCH v3 12/23] f2fs: Convert f2fs_flush_inline_data() " Vishal Moola (Oracle)
2022-10-29 4:47 ` [f2fs-dev] " Chao Yu
2022-10-17 20:24 ` [PATCH v3 13/23] f2fs: Convert f2fs_sync_node_pages() " Vishal Moola (Oracle)
2022-10-29 4:47 ` [f2fs-dev] " Chao Yu
2022-10-17 20:24 ` [PATCH v3 14/23] f2fs: Convert f2fs_write_cache_pages() " Vishal Moola (Oracle)
2022-11-14 7:02 ` [f2fs-dev] " Chao Yu
2022-11-14 21:38 ` Vishal Moola
2022-11-23 2:26 ` Vishal Moola
2022-11-23 7:51 ` Vishal Moola
2022-12-05 20:34 ` Vishal Moola
2022-12-12 14:41 ` Chao Yu
2022-12-12 19:13 ` [RFC PATCH] " Vishal Moola (Oracle)
2022-12-15 1:48 ` Chao Yu
2022-12-15 18:45 ` Matthew Wilcox
2022-12-21 17:17 ` Vishal Moola
2022-12-23 8:07 ` Christoph Hellwig
2022-12-15 19:02 ` Jaegeuk Kim
2023-01-03 20:53 ` Matthew Wilcox
2022-11-29 19:14 ` Matthew Wilcox [this message]
2022-11-30 12:48 ` [PATCH] f2fs: Support enhanced hot/cold data separation for f2fs Yangtao Li
2022-11-30 15:18 ` Matthew Wilcox
2022-12-07 20:51 ` Luis Chamberlain
2024-01-25 20:47 ` Matthew Wilcox
2024-01-25 20:54 ` Luis Chamberlain
2024-01-26 21:01 ` Matthew Wilcox
2024-01-26 21:32 ` Luis Chamberlain
2024-01-27 7:05 ` Eric Biggers
2022-11-30 12:51 ` [PATCH]f2fs: Convert f2fs_write_cache_pages() to use filemap_get_folios_tag() Yangtao Li
2022-10-17 20:24 ` [PATCH v3 15/23] f2fs: Convert last_fsync_dnode() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 16/23] f2fs: Convert f2fs_sync_meta_pages() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 17/23] gfs2: Convert gfs2_write_cache_jdata() " Vishal Moola (Oracle)
2022-10-24 19:23 ` Vishal Moola
2022-10-17 20:24 ` [PATCH v3 18/23] nilfs2: Convert nilfs_lookup_dirty_data_buffers() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 19/23] nilfs2: Convert nilfs_lookup_dirty_node_buffers() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 20/23] nilfs2: Convert nilfs_btree_lookup_dirty_buffers() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 21/23] nilfs2: Convert nilfs_copy_dirty_pages() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 22/23] nilfs2: Convert nilfs_clear_dirty_pages() " Vishal Moola (Oracle)
2022-10-17 20:24 ` [PATCH v3 23/23] filemap: Remove find_get_pages_range_tag() Vishal Moola (Oracle)
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=Y4ZaBd1r45waieQs@casper.infradead.org \
--to=willy@infradead.org \
--cc=chao@kernel.org \
--cc=fengnanchang@gmail.com \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=vishal.moola@gmail.com \
/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