linux-kernel - Re: [RFC PATCH] tty: serial: kgdboc: Fix 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: <CAD=FV=WrBg9PuDW__pZbo5YNuWct17gcK4FF-xKeyxEOsw6Qag@mail.gmail.com>
Date:   Mon, 11 Dec 2023 14:00:12 -0800
From:   Doug Anderson <dianders@...omium.org>
To:     Michael Nazzareno Trimarchi <michael@...rulasolutions.com>
Cc:     Jason Wessel <jason.wessel@...driver.com>,
        Daniel Thompson <daniel.thompson@...aro.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jiri Slaby <jirislaby@...nel.org>,
        kgdb-bugreport@...ts.sourceforge.net, linux-kernel@...r.kernel.org,
        linux-serial@...r.kernel.org
Subject: Re: [RFC PATCH] tty: serial: kgdboc: Fix 8250_* kgd over serial

Hi,

On Mon, Dec 11, 2023 at 1:42 PM Michael Nazzareno Trimarchi
<michael@...rulasolutions.com> wrote:
>
> > 1. init_kgdboc() runs and registers the singleton kgdb "platform driver".
> >
> > 2. The platform driver's probe function, kgdboc_probe(), runs and
> > checks to see if the console is ready by looking at the return value
> > of configure_kgdboc(). If it's ready then we're good to go. If it's
> > not ready then we defer.
> >
> > So I think the bug here is that somehow the console looks "ready"
> > (because tty_find_polling_driver() can find it) but it isn't actually
> > ready yet (because it crashes). That's what you need to fix.
> >
>
> The polling driver look for uart and uart8250_core is registered and 4 fake uart
> are there but there are not still replaced by platform driver that can
> come later.
> The try_polling find it but it's the isa-8250 driver. It means that
> add_uart 8250 is
> not still happen

The 8250 driver is always a maze, so you might need to do a bunch of
digging. ...but it sure sounds like the console shouldn't be
registered until the correct ops are in place. That either means
getting the ops put in place earlier or deferring when the console is
registered...

-Doug