linux-kernel - Re: [linus:master] [eventfs] 852e46e239: BUG:unable_to_handle_page_fault_for

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: <CAHk-=wi+WbXZcc2Sx1i-MGV2DfG4eS4Ci+mrqi-PBSLSnww6qA@mail.gmail.com>
Date: Tue, 30 Jan 2024 01:12:05 -0800
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Steven Rostedt <rostedt@...dmis.org>
Cc: kernel test robot <oliver.sang@...el.com>, oe-lkp@...ts.linux.dev, lkp@...el.com, 
	linux-kernel@...r.kernel.org, Masami Hiramatsu <mhiramat@...nel.org>, 
	Mark Rutland <mark.rutland@....com>, Mathieu Desnoyers <mathieu.desnoyers@...icios.com>, 
	Christian Brauner <brauner@...nel.org>, Al Viro <viro@...iv.linux.org.uk>, 
	Ajay Kaher <ajay.kaher@...adcom.com>, linux-trace-kernel@...r.kernel.org
Subject: Re: [linus:master] [eventfs] 852e46e239: BUG:unable_to_handle_page_fault_for_address

On Tue, 30 Jan 2024 at 00:43, Linus Torvalds
<torvalds@...ux-foundation.org> wrote:
>
> I'll go back to bed, but I think the fix is something trivial like this:

Almost.

>   +     result = ERR_PTR(ENOENT);

That needs a '-' in front of the ENOENT, otherwise you have a positive
error number and things go wrong very quickly.

And that does indeed fix the lookup problem, but you end up with the
same problem later when you do the eventfs_remove_dir(). Again the
eventfs data structure changes, but we don't have a reliable dentry
that we can invalidate.

The dentry cache is just very good at caching those old dentries, and
the interface for eventfs_create_dir() and eventfs_remove_dir() is
just not great.

If those did an actual path lookup (like eventfs_create_events_dir()
does), we'd have the dentry, and it's trivial to get from dentry to
eventfs_inode.

But going the other way is the broken thing because of how the
dentries are just temporary caches.

I suspect the solution is to make eventfs_create_dir() do the same as
the events directory case does, and actually pin the directory dentry
and save it off.

Oh well. At least I understand what the problem is. Now I'm going to
try to go back to sleep.

              Linus