MIME-Version: 1.0
References: <20230301075751.43839-1-lma@semihalf.com> <Y/8PUdEwskXuWZHA@kroah.com>
 <CAFJ_xbp+qD-_MGd3+SgBY=8zruZNy7k3CO3OMMmWhMGhA-tARQ@mail.gmail.com> <Y/9Ddl7c2PKSEpsR@kroah.com>
In-Reply-To: <Y/9Ddl7c2PKSEpsR@kroah.com>
From:   Lukasz Majczak <lma@semihalf.com>
Date:   Wed, 1 Mar 2023 15:09:35 +0100
Message-ID: <CAFJ_xbr_petJ8=wKNLSnJ=97t+cERre07=hNYFeBVHp0nvPtWw@mail.gmail.com>
Subject: Re: [PATCH] serial: core: fix broken console after suspend
To:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc:     Jiri Slaby <jirislaby@kernel.org>,
        Guenter Roeck <groeck@chromium.org>,
        linux-serial@vger.kernel.org, linux-kernel@vger.kernel.org,
        upstream@semihalf.com, stable@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

=C5=9Br., 1 mar 2023 o 13:22 Greg Kroah-Hartman
<gregkh@linuxfoundation.org> napisa=C5=82(a):
>
> On Wed, Mar 01, 2023 at 10:51:31AM +0100, Lukasz Majczak wrote:
> > =C5=9Br., 1 mar 2023 o 09:39 Greg Kroah-Hartman
> > <gregkh@linuxfoundation.org> napisa=C5=82(a):
> > >
> > > On Wed, Mar 01, 2023 at 08:57:51AM +0100, Lukasz Majczak wrote:
> > > > Re-enable the console device after suspending, causes its cflags,
> > > > ispeed and ospeed to be set anew, basing on the values stored in
> > > > uport->cons. The issue is that these values are set only once,
> > > > when parsing console parameters after boot (see uart_set_options())=
,
> > > > next after configuring a port in uart_port_startup() these paramete=
res
> > > > (cflags, ispeed and ospeed) are copied to termios structure and
> > > > the orginal one (stored in uport->cons) are cleared, but there is n=
o place
> > > > in code where those fields are checked against 0.
> > > > When kernel calls uart_resume_port() and setups console, it copies =
cflags,
> > > > ispeed and ospeed values from uart->cons,but those are alread clear=
ed.
> > > > The efect is that console is broken.
> > > > This patch address this by preserving the cflags, ispeed and
> > > > ospeed fields in uart->cons during uart_port_startup().
> > > >
> > > > Signed-off-by: Lukasz Majczak <lma@semihalf.com>
> > > > Cc: stable@vger.kernel.org
> > > > ---
> > > >  drivers/tty/serial/serial_core.c | 3 ---
> > > >  1 file changed, 3 deletions(-)
> > > >
> > > > diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/=
serial_core.c
> > > > index 2bd32c8ece39..394a05c09d87 100644
> > > > --- a/drivers/tty/serial/serial_core.c
> > > > +++ b/drivers/tty/serial/serial_core.c
> > > > @@ -225,9 +225,6 @@ static int uart_port_startup(struct tty_struct =
*tty, struct uart_state *state,
> > > >                       tty->termios.c_cflag =3D uport->cons->cflag;
> > > >                       tty->termios.c_ispeed =3D uport->cons->ispeed=
;
> > > >                       tty->termios.c_ospeed =3D uport->cons->ospeed=
;
> > > > -                     uport->cons->cflag =3D 0;
> > > > -                     uport->cons->ispeed =3D 0;
> > > > -                     uport->cons->ospeed =3D 0;
> > > >               }
> > > >               /*
> > > >                * Initialise the hardware port settings.
> > > > --
> > > > 2.39.2.722.g9855ee24e9-goog
> > > >
> > >
> > > What commit id does this fix?
> > >
> > > thanks,
> > >
> > > greg k-h
> > Hi Greg,
> >
> > There are actually two commits that introduce problematic uport flags
> > clearing in uart_startup (for the sake of simplicity I'd ignore the
> > older history):
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/comm=
it/?h=3Dv6.2&id=3Dc7d7abff40c27f82fe78b1091ab3fad69b2546f9
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/comm=
it/?h=3Dv6.2&id=3D027b57170bf8bb6999a28e4a5f3d78bf1db0f90c
> > It's 10 years between those 2 and to me it was hard to decide about
> > picking a proper one for the `Fixes:` tag.
> > How would you recommend to proceed wrt applying this patch on the
> > stable releases?
>
> Where do you think this needs to go to?  Pick something?
>
> And as you have obviously found this on a device running an older kernel
> version, what kernel tree(s) did you test it on?
>
> thanks,
>
> greg k-h

As this patch applies without conflict on 4.14, I would suggest 4.14+.
I have tested the patch on chromes-5.15 (cannonlake device).

Best regards,
Lukasz