Re: [git pull] device mapper changes for 6.1

From: "Jason A. Donenfeld" <Jason@zx2c4.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Christoph Hellwig <hch@infradead.org>,
	Mike Snitzer <snitzer@kernel.org>,
	dm-devel@redhat.com, linux-block@vger.kernel.org,
	Alasdair G Kergon <agk@redhat.com>,
	Genjian Zhang <zhanggenjian@kylinos.cn>,
	Jiangshan Yi <yijiangshan@kylinos.cn>,
	Jilin Yuan <yuanjilin@cdjrlc.com>,
	Mikulas Patocka <mpatocka@redhat.com>,
	Milan Broz <gmazyland@gmail.com>,
	Nathan Huckleberry <nhuck@google.com>,
	Nikos Tsironis <ntsironis@arrikto.com>,
	Shaomin Deng <dengshaomin@cdjrlc.com>
Subject: Re: [git pull] device mapper changes for 6.1
Date: Wed, 19 Oct 2022 00:18:52 -0600	[thread overview]
Message-ID: <Y0+WzAZMiQAhQdgj@zx2c4.com> (raw)
In-Reply-To: <CAHk-=wg3cpPyoO8u+8Fu1uk86bgTUYanfKhmxMsZzvY_mV=Cxw@mail.gmail.com>

On Tue, Oct 18, 2022 at 11:54:49AM -0700, Linus Torvalds wrote:
> I think we have one major interface that is string-based (apart from
> the obvious pathname ones and the strings passed to 'execve()').
> 
> It's 'mount()' (and now fsconfig() etc), and it's string-based mainly
> because it has that nasty "arbitrary things that different filesystem
> may need for configuration"). And it has some nasty logging model
> associated with it too for output.
> 
> But no, we absolutely do *not* want to emulate that particular horror
> anywhere else.

This might ruin your day, but FYI, Netlink

   [...did you already run off screaming at the mention of Netlink?...]

Netlink has the whole "extack" extended acknowledgement mechanism, in
which netlink replies can and do contain error strings. Grep the kernel
for NL_SET_ERR_MSG and you'll see a bunch of these. (And as you
suggested, I wouldn't be surprised if some bad userspaces parse them.)

Jason