From: Nikolay Borisov <nborisov@suse.com>
To: linux-btrfs@vger.kernel.org
Cc: Nikolay Borisov <nborisov@suse.com>
Subject: [PATCH 00/10] Cleanup metadata page reading path
Date: Wed, 9 Sep 2020 12:49:04 +0300 [thread overview]
Message-ID: <20200909094914.29721-1-nborisov@suse.com> (raw)
This series streamlines the page read path for metadata pages. Currently
__do_readpage is shared between ordinary data inodes as well as the btree_inode.
This is unnecenssary and brings clutter to the interface of __do_readpage et al
in the form of needlessly passed parameters and unnecessary actions performed
in btree_get_extent just to appease the interface/logic of __do_readpage.
To simplify the code metadata read is switched from calling __do_readpage to
directly calling submit_extent_page in read_extent_buffer_pages which is sufficient
to grab metadata pages. This in turn paves the way for further cleanups by
removing a lot of arguments from top level functions and sinking them in
lowe-level worker functions.
Patch 1 and 2 remove unused function and a parameter from btrfs_get_extent.
Patch 3 switches read_extent_buffer_pages (sole metadata read path) to using
submit_extent_page which enables further simplifications. Following patches
now-unused functions and removes certain indirection now that __do_readpage
is used only for data read.
Every patch has survived a -g quick and the overall series survived a full
xfstest run.
Nikolay Borisov (10):
btrfs: Remove btree_readpage
btrfs: Remove pg_offset from btrfs_get_extent
btrfs: Simplify metadata pages reading
btrfs: Remove btree_get_extent
btrfs: Remove btrfs_get_extent indirection from __do_readpage
btrfs: Remove mirror_num argument from extent_read_full_page
btrfs: Promote extent_read_full_page to btrfs_readpage
btrfs: Sink mirror_num argument in extent_read_full_page
btrfs: Sink read_flags argument into extent_read_full_page
btrfs: Sink mirror_num argument in __do_readpage
fs/btrfs/ctree.h | 3 +-
fs/btrfs/disk-io.c | 53 --------------------------
fs/btrfs/disk-io.h | 3 --
fs/btrfs/extent_io.c | 74 ++++++++++++------------------------
fs/btrfs/extent_io.h | 10 +++--
fs/btrfs/file.c | 12 +++---
fs/btrfs/inode.c | 37 +++++++++---------
fs/btrfs/ioctl.c | 2 +-
fs/btrfs/tests/inode-tests.c | 42 ++++++++++----------
9 files changed, 79 insertions(+), 157 deletions(-)
--
2.17.1
next reply other threads:[~2020-09-09 9:49 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-09 9:49 Nikolay Borisov [this message]
2020-09-09 9:49 ` [PATCH 01/10] btrfs: Remove btree_readpage Nikolay Borisov
2020-09-09 10:37 ` Johannes Thumshirn
2020-09-09 11:13 ` Qu Wenruo
2020-09-09 9:49 ` [PATCH 02/10] btrfs: Remove pg_offset from btrfs_get_extent Nikolay Borisov
2020-09-09 10:40 ` Johannes Thumshirn
2020-09-09 11:15 ` Qu Wenruo
2020-09-09 9:49 ` [PATCH 03/10] btrfs: Simplify metadata pages reading Nikolay Borisov
2020-09-09 11:20 ` Qu Wenruo
2020-09-14 8:08 ` Nikolay Borisov
2020-09-14 8:22 ` Qu Wenruo
2020-09-10 14:56 ` Josef Bacik
2020-09-09 9:49 ` [PATCH 04/10] btrfs: Remove btree_get_extent Nikolay Borisov
2020-09-10 14:57 ` Josef Bacik
2020-09-09 9:49 ` [PATCH 05/10] btrfs: Remove btrfs_get_extent indirection from __do_readpage Nikolay Borisov
2020-09-09 11:24 ` Qu Wenruo
2020-09-09 11:56 ` Nikolay Borisov
2020-09-09 9:49 ` [PATCH 06/10] btrfs: Remove mirror_num argument from extent_read_full_page Nikolay Borisov
2020-09-10 14:58 ` Josef Bacik
2020-09-09 9:49 ` [PATCH 07/10] btrfs: Promote extent_read_full_page to btrfs_readpage Nikolay Borisov
2020-09-10 15:01 ` Josef Bacik
2020-09-09 9:49 ` [PATCH 08/10] btrfs: Sink mirror_num argument in extent_read_full_page Nikolay Borisov
2020-09-10 15:02 ` Josef Bacik
2020-09-09 9:49 ` [PATCH 09/10] btrfs: Sink read_flags argument into extent_read_full_page Nikolay Borisov
2020-09-10 15:03 ` Josef Bacik
2020-09-09 9:49 ` [PATCH 10/10] btrfs: Sink mirror_num argument in __do_readpage Nikolay Borisov
2020-09-10 15:04 ` Josef Bacik
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=20200909094914.29721-1-nborisov@suse.com \
--to=nborisov@suse.com \
--cc=linux-btrfs@vger.kernel.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