Received-SPF: pass (google.com: domain of linux-kernel+bounces-137073-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Date: Tue, 9 Apr 2024 16:15:31 +0200
From: Christoph Hellwig <hch@lst.de>
To: Lennart Poettering <mzxreary@0pointer.de>
Cc: Keith Busch <kbusch@kernel.org>,
	Linux regressions mailing list <regressions@lists.linux.dev>,
	Christoph Hellwig <hch@lst.de>, linux-block@vger.kernel.org,
	LKML <linux-kernel@vger.kernel.org>, Jens Axboe <axboe@kernel.dk>
Subject: Re: API break, sysfs "capability" file
Message-ID: <20240409141531.GB21514@lst.de>
References: <ZhQJf8mzq_wipkBH@gardel-login> <54e3c969-3ee8-40d8-91d9-9b9402001d27@leemhuis.info> <ZhQ6ZBmThBBy_eEX@kbusch-mbp.dhcp.thefacebook.com> <ZhRSVSmNmb_IjCCH@gardel-login> <ZhRyhDCT5cZCMqYj@kbusch-mbp.dhcp.thefacebook.com> <ZhT5_fZ9SrM0053p@gardel-login>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <ZhT5_fZ9SrM0053p@gardel-login>
User-Agent: Mutt/1.5.17 (2007-11-01)

On Tue, Apr 09, 2024 at 10:19:09AM +0200, Lennart Poettering wrote:
> All I am looking for is a very simple test that returns me a boolean:
> is there kernel-level partition scanning enabled on this device or
> not.

And we can add a trivial sysfs attribute for that.

> At this point it's not clear to me if I can write this at all in
> a way that works reasonably correctly on any kernel since let's say
> 4.15 (which is systemd's "recommended baseline" right now).
> 
> I am really not sure how to salvage this mess at all. AFAICS there's
> currently no way to write such a test correctly.

You can't.  Maybe that's a lesson to not depend on undocumented internal
flags exposed by accident by a weirdo interface.  Just talk to people.

> I think it would be nice if the "capabilities" thing would be brought
> back in a limited form. For example, if it would be changed to start
> to return 0x200|0x1000 for part scanning is off, 0x1000 when it is on.
> 
> That would then mean we return to compatibility with Linux <= 5.15,
> but the new 0x1000 bit would tell us that the information is
> reliable. i.e. if userspace sees 0x1000 being set we know that the
> 0x200 bit is definitely correct. That would then just mean that
> kernels >= 5.16 until today are left in the cold...

At this point we're just better off with a clean new interface.
And you can use the old hack for < 5.15 if you care strongly enough
or just talk distros into backporting it to make their lives easier.