linux-kernel - Re: [PATCH 30/33] serial: 8250: invert serial8250_register_8250

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: <aFj_dax6Gb9HrGyV@smile.fi.intel.com>
Date: Mon, 23 Jun 2025 10:17:09 +0300
From: Andy Shevchenko <andriy.shevchenko@...el.com>
To: Jiri Slaby <jirislaby@...nel.org>
Cc: Andy Shevchenko <andy.shevchenko@...il.com>,
	"Maciej S. Szmigiero" <mail@...iej.szmigiero.name>,
	Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>,
	gregkh@...uxfoundation.org, linux-serial@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 30/33] serial: 8250: invert
 serial8250_register_8250_port() CIR condition

On Mon, Jun 23, 2025 at 09:05:46AM +0200, Jiri Slaby wrote:
> On 21. 06. 25, 21:08, Andy Shevchenko wrote:
> > Fri, Jun 20, 2025 at 11:48:09PM +0200, Maciej S. Szmigiero kirjoitti:
> > > On 18.06.2025 07:48, Andy Shevchenko wrote:
> > > > On Wed, Jun 11, 2025 at 12:03:16PM +0200, Jiri Slaby (SUSE) wrote:

...

> > > > > +	if (uart->port.type == PORT_8250_CIR) {
> > > > > +		ret = -ENODEV;
> > > > > +		goto unlock;
> > > > > +	}
> > > > 
> > > > > +	if (up->port.flags & UPF_FIXED_TYPE)
> > > > > +		uart->port.type = up->port.type;
> > > > 
> > > > > +	if (uart->port.type != PORT_8250_CIR) {
> > > > 
> > > > I admit that there tons of mysterious ways of UART initialisation, but can you
> > > > elaborate how this is not a always-true conditional?
> > > 
> > > Careful here, someone had an idea in the past that this is indeed
> > > a dead code/branch and ended causing a regression [1].
> 
> Right, I was confused too, but then I noticed there is:
>   uart->port.type = up->port.type;
> in between the tests.
> 
> > > It would definitely make sense to add a comment describing the code
> > > flow there though as it proven to bewilder people.
> > 
> > Yes, this is my point between the lines. I left the code that may affect the
> > type change and the second check needs a comment explaining these cases, if any.
> > "If any" defines "always-true" or not conditional. W//o a comment this code
> > tends to be updated again and lead to a regression.
> 
> ACK, I will.

Thanks!

Looking at the code again, I think it deserves actually two comments, on top of
each of the checks against PORT_8250_CIR.

-- 
With Best Regards,
Andy Shevchenko