From: Christian Brauner <brauner@kernel.org>
To: Linus Torvalds <torvalds@linux-foundation.org>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>
Cc: Christian Brauner <brauner@kernel.org>,
Mateusz Guzik <mjguzik@gmail.com>,
Al Viro <viro@zeniv.linux.org.uk>, Jan Kara <jack@suse.cz>,
Jeff Layton <jlayton@kernel.org>,
Josef Bacik <josef@toxicpanda.com>
Subject: [PATCH RFC 0/3] mnt_idmapping: avoid pointer chase & inline low-level helpers
Date: Wed, 16 Apr 2025 15:17:21 +0200 [thread overview]
Message-ID: <20250416-work-mnt_idmap-s_user_ns-v1-0-273bef3a61ec@kernel.org> (raw)
In-Reply-To: <20250414-anomalie-abpfiff-9f293dce366b@brauner>
We currently always chase a pointer inode->i_sb->s_user_ns whenever we
need to map a uid/gid. Linus reported in [1] that this is noticable
during path lookup.
In the majority of cases we don't need to bother with that pointer chase
because the inode won't be located on a filesystem that's mounted in a
user namespace. The user namespace of the superblock cannot ever change
once it's mounted. So introduce and raise IOP_USERNS on all inodes and
check for that flag in i_user_ns() when we retrieve the user namespace.
Additionally, we now inline all low-level idmapping helpers.
[1]: https://lore.kernel.org/CAHk-=whJgRDtxTudTQ9HV8BFw5-bBsu+c8Ouwd_PrPqPB6_KEQ@mail.gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
---
Christian Brauner (3):
inode: add fastpath for filesystem user namespace retrieval
mnt_idmapping: add struct mnt_idmap to header
mnt_idmapping: inline all low-level helpers
fs/inode.c | 6 ++
fs/mnt_idmapping.c | 165 ----------------------------------------
include/linux/fs.h | 5 +-
include/linux/mnt_idmapping.h | 168 +++++++++++++++++++++++++++++++++++++++--
include/linux/uidgid.h | 23 +++++-
include/linux/user_namespace.h | 23 +-----
kernel/user_namespace.c | 2 +
7 files changed, 195 insertions(+), 197 deletions(-)
---
base-commit: 0af2f6be1b4281385b618cb86ad946eded089ac8
change-id: 20250416-work-mnt_idmap-s_user_ns-eb57ee83e1d6
next prev parent reply other threads:[~2025-04-16 13:17 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-31 4:16 generic_permission() optimization Linus Torvalds
2024-10-31 6:05 ` Al Viro
2024-10-31 6:42 ` Linus Torvalds
2024-10-31 18:14 ` Linus Torvalds
2024-10-31 22:28 ` Al Viro
2024-10-31 22:34 ` Linus Torvalds
2024-11-01 1:17 ` Linus Torvalds
2024-11-01 1:27 ` Al Viro
2024-11-01 13:15 ` Christian Brauner
2024-10-31 13:02 ` Christian Brauner
2024-10-31 19:04 ` Linus Torvalds
2024-10-31 22:02 ` Linus Torvalds
2024-10-31 22:31 ` Linus Torvalds
2024-11-07 19:54 ` Linus Torvalds
2024-11-07 22:22 ` Mateusz Guzik
2024-11-07 22:49 ` Linus Torvalds
2025-04-12 16:26 ` Mateusz Guzik
2025-04-12 20:22 ` Linus Torvalds
2025-04-14 10:21 ` Christian Brauner
2025-04-16 13:17 ` Christian Brauner [this message]
2025-04-16 13:17 ` [PATCH RFC 1/3] inode: add fastpath for filesystem user namespace retrieval Christian Brauner
2025-04-16 13:49 ` Mateusz Guzik
2025-04-16 14:14 ` Christian Brauner
2025-04-22 10:37 ` Jan Kara
2025-04-22 13:33 ` Mateusz Guzik
2025-04-22 14:05 ` Christian Brauner
2025-04-16 13:17 ` [PATCH RFC 2/3] mnt_idmapping: add struct mnt_idmap to header Christian Brauner
2025-04-16 13:17 ` [PATCH RFC 3/3] mnt_idmapping: inline all low-level helpers Christian Brauner
2025-04-16 15:04 ` Linus Torvalds
2025-04-22 9:28 ` Christian Brauner
2025-04-12 21:52 ` generic_permission() optimization Theodore Ts'o
2025-04-12 22:36 ` Linus Torvalds
2025-04-12 23:12 ` Linus Torvalds
2025-04-12 23:55 ` Theodore Ts'o
2025-04-13 9:41 ` Mateusz Guzik
2025-04-13 12:40 ` Theodore Ts'o
2025-04-13 12:52 ` Mateusz Guzik
2025-04-13 17:29 ` Theodore Ts'o
2025-11-05 11:50 ` Mateusz Guzik
2025-11-05 11:51 ` Mateusz Guzik
2025-11-05 13:37 ` Jan Kara
2025-11-17 11:42 ` Mateusz Guzik
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=20250416-work-mnt_idmap-s_user_ns-v1-0-273bef3a61ec@kernel.org \
--to=brauner@kernel.org \
--cc=jack@suse.cz \
--cc=jlayton@kernel.org \
--cc=josef@toxicpanda.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=mjguzik@gmail.com \
--cc=torvalds@linux-foundation.org \
--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).