netdev - Re: [PATCH net-next v7 1/5] rust: core abstractions for network PHY drivers

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: <ZT6M6WPrCaLb-0QO@Boquns-Mac-mini.home>
Date: Sun, 29 Oct 2023 09:48:41 -0700
From: Boqun Feng <boqun.feng@...il.com>
To: FUJITA Tomonori <fujita.tomonori@...il.com>
Cc: benno.lossin@...ton.me, andrew@...n.ch, netdev@...r.kernel.org,
	rust-for-linux@...r.kernel.org, tmgross@...ch.edu,
	miguel.ojeda.sandonis@...il.com, wedsonaf@...il.com
Subject: Re: [PATCH net-next v7 1/5] rust: core abstractions for network PHY
 drivers

On Sun, Oct 29, 2023 at 01:21:12PM +0900, FUJITA Tomonori wrote:
[...]
> 
> The current code is fine from Rust perspective because the current
> code copies phy_driver on stack and makes a reference to the copy, if
> I undertand correctly.
> 

I had the same thought Benno brought the issue on `&`, but unfortunately
it's not true ;-) In the following code:

	let phydev = unsafe { *self.0.get() };

, semantically the *whole* `bindings::phy_device` is being read, so if
there is any modification (i.e. write) that may happen in the meanwhile,
it's data race, and data races are UB (even in C).

So both implementations have the problem because of the same cause.

> It's not nice to create an 500-bytes object on stack. It turned out
> that it's not so simple to avoid it.

As you can see, copying is not the way to work around this.

Regards,
Boqun