linux-kernel - Re: [PATCH v1 31/31] serial: 8250_mxpcie: add RS485-2W auto-adjust sysfs control

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] [day] [month] [year] [list]

Message-ID: <CAHp75VdnC5FmXzf3oTqo0ZGZzwkd_+GAioTAss8EBCcVhC5Kww@mail.gmail.com>
Date: Fri, 5 Dec 2025 10:35:31 +0200
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Crescent Hsieh <crescentcy.hsieh@...a.com>
Cc: gregkh@...uxfoundation.org, jirislaby@...nel.org, 
	ilpo.jarvinen@...ux.intel.com, linux-kernel@...r.kernel.org, 
	linux-serial@...r.kernel.org
Subject: Re: [PATCH v1 31/31] serial: 8250_mxpcie: add RS485-2W auto-adjust
 sysfs control

On Fri, Dec 5, 2025 at 7:26 AM Crescent Hsieh <crescentcy.hsieh@...a.com> wrote:
> On Wed, Dec 03, 2025 at 11:24:58AM +0200, Andy Shevchenko wrote:
> > On Wed, Dec 3, 2025 at 4:56 AM Crescent Hsieh <crescentcy.hsieh@...a.com> wrote:
> > > On Mon, Dec 01, 2025 at 04:45:58AM +0200, Andy Shevchenko wrote:
> > > > On Sun, Nov 30, 2025 at 12:46 PM Crescent Hsieh
> > > > <crescentcy.hsieh@...a.com> wrote:

...

> > > > I'm not going to review this, the patch series is already quite big. I
> > > > suggest you to start from the small things in a different series E.g.,
> > > > the first series is just converting MOXA from custom to 8250-based
> > > > (assuming all features are kept working while ABI is being broken,
> >
> > In case you are wondering what I was talking about in the above, I
> > meant move from /dev/ttyMIxx to /dev/ttySxx. This will break all the
> > current kernel command lines and hence setups with the explicit
> > mention of the /dev/ttyMIxx, such as console=.  There might be other
> > breakages, but I leave it up to you to research and come up with a
> > solution.
>
> Just to clarify my intention: the in-tree UPCI serial driver (mxser) has
> been unmaintained for years, and my goal is to replace it with a clean
> 8250-based implementation that preserves reasonable user expectations
> while following the upstream serial framework. This will require some
> analysis to reconcile the legacy behavior with what upstream expects.

Right, so it's a good justification to break the above on the
expectations that the users most likely don't use the in-kernel driver
as it's unmaintained and has received no bug fixes, etc.

> I’d also like to ask about patch ordering. Since the PCIe serial driver
> is much simpler to migrate and has minimal user impact, would it be
> acceptable to upstream the PCIe conversion first, before the more
> complex UPCI transition? I’m happy to follow whichever order makes
> review easiest for you.

I don't see the downsides of this order. I believe the order of these
big parts is not so important per se.

-- 
With Best Regards,
Andy Shevchenko