From: Al Viro <viro@zeniv.linux.org.uk>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Christian Brauner <brauner@kernel.org>,
linux-fsdevel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>,
Jann Horn <jannh@google.com>
Subject: Re: [PATCH RFC 0/4] fs: port files to rcuref_long_t
Date: Sat, 5 Oct 2024 23:01:00 +0100 [thread overview]
Message-ID: <20241005220100.GA4017910@ZenIV> (raw)
In-Reply-To: <CAHk-=wj7=Ynmk9+Fm860NqHu5q119AiN4YNXNJPt=6Q=Y=w3HA@mail.gmail.com>
On Sat, Oct 05, 2024 at 02:42:25PM -0700, Linus Torvalds wrote:
> Also, honestly, the only reason the file counting is using a "long" is
> because the code does *NOT* do overflow checking. But once you start
> looking at the sign and do conditional increments, you can actually
> just make the whole refcount be a "int" instead, and make "struct
> file" potentially smaller.
I wouldn't bet on that. You *can* get over 2G references on 64bit box
with arseloads of memory, and we have no way to make fget() et.al.
fail when refcount gets that high - no valid error to return and
a serious DoS potential if we start doing that.
Overflow on leaks is one thing, a huge pile of real references is a
different story, and yes, we can get that. And boxen with 1Tb RAM
are not as exotic these days as they used to be...
next prev parent reply other threads:[~2024-10-05 22:01 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-05 19:16 [PATCH RFC 0/4] fs: port files to rcuref_long_t Christian Brauner
2024-10-05 19:16 ` [PATCH RFC 1/4] fs: protect backing files with rcu Christian Brauner
2024-10-05 19:16 ` [PATCH RFC 2/4] types: add rcuref_long_t Christian Brauner
2024-10-05 19:16 ` [PATCH RFC 3/4] rcuref: add rcuref_long_*() helpers Christian Brauner
2024-10-05 19:16 ` [PATCH RFC 4/4] fs: port files to rcuref_long_t Christian Brauner
2024-10-05 21:42 ` [PATCH RFC 0/4] " Linus Torvalds
2024-10-05 22:01 ` Al Viro [this message]
2024-10-05 22:14 ` Linus Torvalds
2024-10-05 22:28 ` Al Viro
2024-10-05 22:43 ` Linus Torvalds
2024-10-05 22:51 ` Al Viro
2024-10-06 9:55 ` Christian Brauner
2024-10-05 22:01 ` Linus Torvalds
2024-10-06 10:21 ` Christian Brauner
2024-10-06 18:09 ` Linus Torvalds
2024-10-07 7:37 ` Christian Brauner
2024-10-06 9:48 ` Christian Brauner
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=20241005220100.GA4017910@ZenIV \
--to=viro@zeniv.linux.org.uk \
--cc=brauner@kernel.org \
--cc=jannh@google.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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;
as well as URLs for NNTP newsgroup(s).