[PATCH] serial: core: fix broken console after suspend

[PATCH] serial: core: fix broken console after suspend

[Lukasz Majczak]

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 parameteres
(cflags, ispeed and ospeed) are copied to termios structure and
the orginal one (stored in uport->cons) are cleared, but there is no 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 cleared.
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 = uport->cons->cflag;
 			tty->termios.c_ispeed = uport->cons->ispeed;
 			tty->termios.c_ospeed = uport->cons->ospeed;
-			uport->cons->cflag = 0;
-			uport->cons->ispeed = 0;
-			uport->cons->ospeed = 0;
 		}
 		/*
 		 * Initialise the hardware port settings.
-- 
2.39.2.722.g9855ee24e9-goog

Re: [PATCH] serial: core: fix broken console after suspend

[Greg Kroah-Hartman]

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 parameteres
> (cflags, ispeed and ospeed) are copied to termios structure and
> the orginal one (stored in uport->cons) are cleared, but there is no 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 cleared.
> 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 = uport->cons->cflag;
>  			tty->termios.c_ispeed = uport->cons->ispeed;
>  			tty->termios.c_ospeed = uport->cons->ospeed;
> -			uport->cons->cflag = 0;
> -			uport->cons->ispeed = 0;
> -			uport->cons->ospeed = 0;
>  		}
>  		/*
>  		 * Initialise the hardware port settings.
> -- 
> 2.39.2.722.g9855ee24e9-goog
> 

What commit id does this fix?

thanks,

greg k-h

Re: [PATCH] serial: core: fix broken console after suspend

[Lukasz Majczak]

śr., 1 mar 2023 o 09:39 Greg Kroah-Hartman
<gregkh@linuxfoundation.org> napisał(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 parameteres
> > (cflags, ispeed and ospeed) are copied to termios structure and
> > the orginal one (stored in uport->cons) are cleared, but there is no 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 cleared.
> > 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 = uport->cons->cflag;
> >                       tty->termios.c_ispeed = uport->cons->ispeed;
> >                       tty->termios.c_ospeed = uport->cons->ospeed;
> > -                     uport->cons->cflag = 0;
> > -                     uport->cons->ispeed = 0;
> > -                     uport->cons->ospeed = 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/commit/?h=v6.2&id=c7d7abff40c27f82fe78b1091ab3fad69b2546f9
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v6.2&id=027b57170bf8bb6999a28e4a5f3d78bf1db0f90c
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?

Best regards,
Lukasz

Re: [PATCH] serial: core: fix broken console after suspend

[Ilpo Järvinen]

On Wed, 1 Mar 2023, Lukasz Majczak wrote:

> Re-enable the console device after suspending, causes its cflags,

Re-enabling

> 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()),

Remove "The issue is that" from here and just state:

"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 parameteres

parameters

> (cflags, ispeed and ospeed) are copied to termios structure and
> the orginal one (stored in uport->cons) are cleared, but there is no place

original

> 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 cleared.

missing space after comma.

alread -> already

> The efect is that console is broken.

effect

> This patch address this by preserving the cflags, ispeed and

Too many "this", don't start with "This patch" but go directly to the 
point.


-- 
 i.


> 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 = uport->cons->cflag;
>  			tty->termios.c_ispeed = uport->cons->ispeed;
>  			tty->termios.c_ospeed = uport->cons->ospeed;
> -			uport->cons->cflag = 0;
> -			uport->cons->ispeed = 0;
> -			uport->cons->ospeed = 0;
>  		}
>  		/*
>  		 * Initialise the hardware port settings.
> 

Re: [PATCH] serial: core: fix broken console after suspend

[Greg Kroah-Hartman]

On Wed, Mar 01, 2023 at 10:51:31AM +0100, Lukasz Majczak wrote:
> śr., 1 mar 2023 o 09:39 Greg Kroah-Hartman
> <gregkh@linuxfoundation.org> napisał(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 parameteres
> > > (cflags, ispeed and ospeed) are copied to termios structure and
> > > the orginal one (stored in uport->cons) are cleared, but there is no 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 cleared.
> > > 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 = uport->cons->cflag;
> > >                       tty->termios.c_ispeed = uport->cons->ispeed;
> > >                       tty->termios.c_ospeed = uport->cons->ospeed;
> > > -                     uport->cons->cflag = 0;
> > > -                     uport->cons->ispeed = 0;
> > > -                     uport->cons->ospeed = 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/commit/?h=v6.2&id=c7d7abff40c27f82fe78b1091ab3fad69b2546f9
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v6.2&id=027b57170bf8bb6999a28e4a5f3d78bf1db0f90c
> 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

Re: [PATCH] serial: core: fix broken console after suspend

[Lukasz Majczak]

śr., 1 mar 2023 o 13:22 Greg Kroah-Hartman
<gregkh@linuxfoundation.org> napisał(a):
>
> On Wed, Mar 01, 2023 at 10:51:31AM +0100, Lukasz Majczak wrote:
> > śr., 1 mar 2023 o 09:39 Greg Kroah-Hartman
> > <gregkh@linuxfoundation.org> napisał(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 parameteres
> > > > (cflags, ispeed and ospeed) are copied to termios structure and
> > > > the orginal one (stored in uport->cons) are cleared, but there is no 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 cleared.
> > > > 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 = uport->cons->cflag;
> > > >                       tty->termios.c_ispeed = uport->cons->ispeed;
> > > >                       tty->termios.c_ospeed = uport->cons->ospeed;
> > > > -                     uport->cons->cflag = 0;
> > > > -                     uport->cons->ispeed = 0;
> > > > -                     uport->cons->ospeed = 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/commit/?h=v6.2&id=c7d7abff40c27f82fe78b1091ab3fad69b2546f9
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v6.2&id=027b57170bf8bb6999a28e4a5f3d78bf1db0f90c
> > 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

Re: [PATCH] serial: core: fix broken console after suspend

[Lukasz Majczak]

śr., 1 mar 2023 o 15:09 Lukasz Majczak <lma@semihalf.com> napisał(a):
>
> śr., 1 mar 2023 o 13:22 Greg Kroah-Hartman
> <gregkh@linuxfoundation.org> napisał(a):
> >
> > On Wed, Mar 01, 2023 at 10:51:31AM +0100, Lukasz Majczak wrote:
> > > śr., 1 mar 2023 o 09:39 Greg Kroah-Hartman
> > > <gregkh@linuxfoundation.org> napisał(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 parameteres
> > > > > (cflags, ispeed and ospeed) are copied to termios structure and
> > > > > the orginal one (stored in uport->cons) are cleared, but there is no 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 cleared.
> > > > > 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 = uport->cons->cflag;
> > > > >                       tty->termios.c_ispeed = uport->cons->ispeed;
> > > > >                       tty->termios.c_ospeed = uport->cons->ospeed;
> > > > > -                     uport->cons->cflag = 0;
> > > > > -                     uport->cons->ispeed = 0;
> > > > > -                     uport->cons->ospeed = 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/commit/?h=v6.2&id=c7d7abff40c27f82fe78b1091ab3fad69b2546f9
> > > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v6.2&id=027b57170bf8bb6999a28e4a5f3d78bf1db0f90c
> > > 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

I have tested my patch also with 4.14 and 6.1 (again chromeos tree).
On 4.14 it fixed the issue, but on 6.1 although the console survived
the suspend/resume,
it is printing different characters than requested - I will try to
debug it further.

Best regards
Lukasz