Received-SPF: pass (google.com: domain of linux-kernel+bounces-179281-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Precedence: bulk
MIME-Version: 1.0
References: <20240415163527.626541-1-jeffxu@chromium.org> <20240514104646.e6af4292f19b834777ec1e32@linux-foundation.org>
 <871q646rea.fsf@meer.lwn.net> <ZkPXcT_JuQeZCAv0@casper.infradead.org> <56001.1715726927@cvs.openbsd.org>
In-Reply-To: <56001.1715726927@cvs.openbsd.org>
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Tue, 14 May 2024 17:43:47 -0700
Message-ID: <CAHk-=wgsGCKvN0Db6ZRZqJwXQrmhZyWB6RmABaOp4DiZbXgNew@mail.gmail.com>
Subject: Re: [PATCH v10 0/5] Introduce mseal
To: Theo de Raadt <deraadt@openbsd.org>
Cc: Matthew Wilcox <willy@infradead.org>, Jonathan Corbet <corbet@lwn.net>, 
	Andrew Morton <akpm@linux-foundation.org>, jeffxu@chromium.org, keescook@chromium.org, 
	jannh@google.com, sroettger@google.com, gregkh@linuxfoundation.org, 
	usama.anjum@collabora.com, Liam.Howlett@oracle.com, surenb@google.com, 
	merimus@google.com, rdunlap@infradead.org, jeffxu@google.com, 
	jorgelo@chromium.org, groeck@chromium.org, linux-kernel@vger.kernel.org, 
	linux-kselftest@vger.kernel.org, linux-mm@kvack.org, pedro.falcato@gmail.com, 
	dave.hansen@intel.com, linux-hardening@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"

On Tue, 14 May 2024 at 15:48, Theo de Raadt <deraadt@openbsd.org> wrote:
>
> and can't imagine it affecting a single application

Honestly, that's the reason for not caring.

You have to do actively wrong things for this to matter AT ALL.

So no, we're not making gratuitous changes for stupid reasons.

> I worry that the non-atomicity will one day be used by an attacker.

Blah blah blah. That's a made-up scare tactic if I ever heard one.
It's unworthy of you.

Anybody who does mprotect/mmap/munmap/whatever over multiple
independent memory mappings had better know exactly what mappings they
are touching. Otherwise they are *already* just doing random crap.

In other words: nobody actually does that. Yes, you have people who
first carve out one big area with an mmap(), and then do their own
memory management within that area. But the point is, they are very
much in control and if they do something inconsistent, they absolutely
only have themselves to blame.

And if you have some app that randomly does mprotect etc over multipl
memory mappings that it doesn't know what the f*^% they are, then
there is no saving such a piece of unbelievable garbahe.

So stop the pointless fear-mongering. Linux does the smart thing,
which is to not waste a single cycle on something that cannot possibly
be relevant.

             Linus