linux-ext4 - Re: [PATCHES] RCU pathwalk race fixes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c33d3939f30df5fb087c9dfd16ae6734eca1708f.camel@kernel.org>
Date: Mon, 05 Feb 2024 07:48:17 -0500
From: Jeff Layton <jlayton@...nel.org>
To: Al Viro <viro@...iv.linux.org.uk>, linux-fsdevel@...r.kernel.org
Cc: Linus Torvalds <torvalds@...ux-foundation.org>, Christian Brauner
 <brauner@...nel.org>, linux-ext4@...r.kernel.org,
 linux-nfs@...r.kernel.org,  Miklos Szeredi <miklos@...redi.hu>,
 linux-cifs@...r.kernel.org
Subject: Re: [PATCHES] RCU pathwalk race fixes

On Sun, 2024-02-04 at 02:14 +0000, Al Viro wrote:
> 	We still have some races in filesystem methods when exposed
> to RCU pathwalk.  The series below is a result of code audit (the
> second round of it) and it should deal with most of that stuff.
> Exceptions: ntfs3 ->d_hash()/->d_compare() and ceph_d_revalidate().
> Up to maintainers (a note for NTFS folks - when documentation says
> that a method may not block, it *does* imply that blocking allocations
> are to be avoided.  Really).
> 
> 	Branch is 6.8-rc1-based; it lives in
> git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git fixes.pathwalk-rcu
> 
> Individual patches are in followups; ditto for code audit notes.  Beginning
> of the latter should probably be converted into docs; if anyone is willing
> to help with such conversion, please say so - I'll be glad to answer any
> questions, etc.
> 
> If somebody wants to grab bits and pieces of that series into individual
> filesystem git trees, please say so.  Same for any problems spotted in
> the patches, obviously.  If nothing shows up, that goes into #fixes and
> into mainline.
> 
> Shortlog:
>       fs/super.c: don't drop ->s_user_ns until we free struct super_block itself
>       rcu pathwalk: prevent bogus hard errors from may_lookup()
>       affs: free affs_sb_info with kfree_rcu()
>       exfat: move freeing sbi, upcase table and dropping nls into rcu-delayed helper
>       hfsplus: switch to rcu-delayed unloading of nls and freeing ->s_fs_info
>       afs: fix __afs_break_callback() / afs_drop_open_mmap() race
>       nfs: make nfs_set_verifier() safe for use in RCU pathwalk
>       nfs: fix UAF on pathwalk running into umount
>       procfs: move dropping pde and pid from ->evict_inode() to ->free_inode()
>       procfs: make freeing proc_fs_info rcu-delayed
>       fuse: fix UAF in rcu pathwalks
>       cifs_get_link(): bail out in unsafe case
>       ext4_get_link(): fix breakage in RCU mode
> 
> Diffstat:
>  fs/affs/affs.h            |  1 +
>  fs/affs/super.c           |  2 +-
>  fs/afs/file.c             |  8 ++++++--
>  fs/exfat/exfat_fs.h       |  1 +
>  fs/exfat/nls.c            | 14 ++++----------
>  fs/exfat/super.c          | 20 +++++++++++---------
>  fs/ext4/symlink.c         |  8 +++++---
>  fs/fuse/cuse.c            |  3 +--
>  fs/fuse/fuse_i.h          |  1 +
>  fs/fuse/inode.c           | 15 +++++++++++----
>  fs/hfsplus/hfsplus_fs.h   |  1 +
>  fs/hfsplus/super.c        | 12 +++++++++---
>  fs/namei.c                |  6 +++++-
>  fs/nfs/client.c           | 13 ++++++++++---
>  fs/nfs/dir.c              |  4 ++--
>  fs/proc/base.c            |  2 --
>  fs/proc/inode.c           | 19 ++++++++-----------
>  fs/proc/root.c            |  2 +-
>  fs/smb/client/cifsfs.c    |  3 +++
>  fs/super.c                | 13 ++++---------
>  include/linux/nfs_fs_sb.h |  2 ++
>  include/linux/proc_fs.h   |  1 +
>  22 files changed, 88 insertions(+), 63 deletions(-)
> 

Reviewed-by: Jeff Layton <jlayton@...nel.org>