Hi Claudiu,
CC Peng
Thanks for your patch!
On Mon, Apr 22, 2024 at 1:11 PM Claudiu <[email protected]> wrote:
> From: Claudiu Beznea <[email protected]>
>
> In case the SCI is used as a UART console, no_console_suspend is
> available in bootargs and SCI is part of a software-controlled power
> domain we need to call device_set_wakeup_path(). This lets the power
> domain core code knows that this domain should not be powered off
know
> durring system suspend. Otherwise, the SCI power domain is turned off,
during
> nothing is printed while suspending and the suspend/resume process is
> blocked. This was detected on the RZ/G3S SoC while adding support
> for power domains.
>
> Signed-off-by: Claudiu Beznea <[email protected]>
> ---
> drivers/tty/serial/sh-sci.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
> index 97031db26ae4..57a7f18e16e4 100644
> --- a/drivers/tty/serial/sh-sci.c
> +++ b/drivers/tty/serial/sh-sci.c
> @@ -3441,8 +3441,12 @@ static __maybe_unused int sci_suspend(struct device *dev)
> {
> struct sci_port *sport = dev_get_drvdata(dev);
>
> - if (sport)
> + if (sport) {
> + if (uart_console(&sport->port) && !console_suspend_enabled)
> + device_set_wakeup_path(dev);
device_set_awake_path(), as of commit 10bb4e4ab7dd3898 ("PM: sleep:
Add helpers to allow a device to remain powered-on") in v6.6
(although I'm still a bit puzzled about the difference).
> +
> uart_suspend_port(&sci_uart_driver, &sport->port);
I think it would be better to make this more general, and move the call
to the existing console_suspend_enabled handling in uart_suspend_port().
> + }
>
> return 0;
> }
If this works, we can remove the console_suspend_enabled handling
from drivers/pmdomain/renesas/rmobile-sysc.c, and revert commit
309864dcf92b76fc ("genpd: imx: scu-pd: do not power off console if
no_console_suspend").
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68korg
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Geert,
> Subject: Re: [PATCH] serial: sh-sci: Call device_set_wakeup_path() for serial
> console
>
> Hi Claudiu,
>
> CC Peng
Thanks for Ccing me, but I am not familiar with sh-sci driver (:
Thanks,
Peng.
>
> Thanks for your patch!
>
> On Mon, Apr 22, 2024 at 1:11 PM Claudiu <[email protected]>
> wrote:
> > From: Claudiu Beznea <[email protected]>
> >
> > In case the SCI is used as a UART console, no_console_suspend is
> > available in bootargs and SCI is part of a software-controlled power
> > domain we need to call device_set_wakeup_path(). This lets the power
> > domain core code knows that this domain should not be powered off
>
> know
>
> > durring system suspend. Otherwise, the SCI power domain is turned off,
>
> during
>
> > nothing is printed while suspending and the suspend/resume process is
> > blocked. This was detected on the RZ/G3S SoC while adding support for
> > power domains.
> >
> > Signed-off-by: Claudiu Beznea <[email protected]>
> > ---
> > drivers/tty/serial/sh-sci.c | 6 +++++-
> > 1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
> > index 97031db26ae4..57a7f18e16e4 100644
> > --- a/drivers/tty/serial/sh-sci.c
> > +++ b/drivers/tty/serial/sh-sci.c
> > @@ -3441,8 +3441,12 @@ static __maybe_unused int sci_suspend(struct
> > device *dev) {
> > struct sci_port *sport = dev_get_drvdata(dev);
> >
> > - if (sport)
> > + if (sport) {
> > + if (uart_console(&sport->port) && !console_suspend_enabled)
> > + device_set_wakeup_path(dev);
>
> device_set_awake_path(), as of commit 10bb4e4ab7dd3898 ("PM: sleep:
> Add helpers to allow a device to remain powered-on") in v6.6 (although I'm
> still a bit puzzled about the difference).
>
> > +
> > uart_suspend_port(&sci_uart_driver, &sport->port);
>
> I think it would be better to make this more general, and move the call to the
> existing console_suspend_enabled handling in uart_suspend_port().
>
> > + }
> >
> > return 0;
> > }
>
> If this works, we can remove the console_suspend_enabled handling from
> drivers/pmdomain/renesas/rmobile-sysc.c, and revert commit
> 309864dcf92b76fc ("genpd: imx: scu-pd: do not power off console if
> no_console_suspend").
>
> Gr{oetje,eeting}s,
>
> Geert
>
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-
> m68k.org
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
> -- Linus Torvalds
Hi Peng,
On Tue, Apr 23, 2024 at 9:35 AM Peng Fan <[email protected]> wrote:
> > Subject: Re: [PATCH] serial: sh-sci: Call device_set_wakeup_path() for serial
> > console
> > CC Peng
>
> Thanks for Ccing me, but I am not familiar with sh-sci driver (:
But you are with the IMX pmdomain driver, so please read on ;-)
> > On Mon, Apr 22, 2024 at 1:11 PM Claudiu <[email protected]>
> > wrote:
> > > From: Claudiu Beznea <[email protected]>
> > >
> > > In case the SCI is used as a UART console, no_console_suspend is
> > > available in bootargs and SCI is part of a software-controlled power
> > > domain we need to call device_set_wakeup_path(). This lets the power
> > > domain core code knows that this domain should not be powered off
> > > durring system suspend. Otherwise, the SCI power domain is turned off,
> > > nothing is printed while suspending and the suspend/resume process is
> > > blocked. This was detected on the RZ/G3S SoC while adding support for
> > > power domains.
> > >
> > > Signed-off-by: Claudiu Beznea <[email protected]>
> > > --- a/drivers/tty/serial/sh-sci.c
> > > +++ b/drivers/tty/serial/sh-sci.c
> > > @@ -3441,8 +3441,12 @@ static __maybe_unused int sci_suspend(struct
> > > device *dev) {
> > > struct sci_port *sport = dev_get_drvdata(dev);
> > >
> > > - if (sport)
> > > + if (sport) {
> > > + if (uart_console(&sport->port) && !console_suspend_enabled)
> > > + device_set_wakeup_path(dev);
> >
> > device_set_awake_path(), as of commit 10bb4e4ab7dd3898 ("PM: sleep:
> > Add helpers to allow a device to remain powered-on") in v6.6 (although I'm
> > still a bit puzzled about the difference).
> >
> > > +
> > > uart_suspend_port(&sci_uart_driver, &sport->port);
> >
> > I think it would be better to make this more general, and move the call to the
> > existing console_suspend_enabled handling in uart_suspend_port().
> >
> > > + }
> > >
> > > return 0;
> > > }
> >
> > If this works, we can remove the console_suspend_enabled handling from
> > drivers/pmdomain/renesas/rmobile-sysc.c, and revert commit
> > 309864dcf92b76fc ("genpd: imx: scu-pd: do not power off console if
> > no_console_suspend").
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68korg
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi, Geert,
On 23.04.2024 10:27, Geert Uytterhoeven wrote:
> Hi Claudiu,
>
> CC Peng
>
> Thanks for your patch!
>
> On Mon, Apr 22, 2024 at 1:11 PM Claudiu <[email protected]> wrote:
>> From: Claudiu Beznea <[email protected]>
>>
>> In case the SCI is used as a UART console, no_console_suspend is
>> available in bootargs and SCI is part of a software-controlled power
>> domain we need to call device_set_wakeup_path(). This lets the power
>> domain core code knows that this domain should not be powered off
>
> know
>
>> durring system suspend. Otherwise, the SCI power domain is turned off,
>
> during
>
>> nothing is printed while suspending and the suspend/resume process is
>> blocked. This was detected on the RZ/G3S SoC while adding support
>> for power domains.
>>
>> Signed-off-by: Claudiu Beznea <[email protected]>
>> ---
>> drivers/tty/serial/sh-sci.c | 6 +++++-
>> 1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
>> index 97031db26ae4..57a7f18e16e4 100644
>> --- a/drivers/tty/serial/sh-sci.c
>> +++ b/drivers/tty/serial/sh-sci.c
>> @@ -3441,8 +3441,12 @@ static __maybe_unused int sci_suspend(struct device *dev)
>> {
>> struct sci_port *sport = dev_get_drvdata(dev);
>>
>> - if (sport)
>> + if (sport) {
>> + if (uart_console(&sport->port) && !console_suspend_enabled)
>> + device_set_wakeup_path(dev);
>
> device_set_awake_path(), as of commit 10bb4e4ab7dd3898 ("PM: sleep:
> Add helpers to allow a device to remain powered-on") in v6.6
> (although I'm still a bit puzzled about the difference).
Ok, I wasn't aware of it. I'll switch to this one.
>
>> +
>> uart_suspend_port(&sci_uart_driver, &sport->port);
>
> I think it would be better to make this more general, and move the call
> to the existing console_suspend_enabled handling in uart_suspend_port().
Ok, I'll try this way.
>
>> + }
>>
>> return 0;
>> }
>
> If this works, we can remove the console_suspend_enabled handling
> from drivers/pmdomain/renesas/rmobile-sysc.c, and revert commit
> 309864dcf92b76fc ("genpd: imx: scu-pd: do not power off console if
> no_console_suspend").
OK, first I'll go with this patch and after things settles down with it
I'll propose changes for rmobile-sysc and imx. Is this ok for you?
Thank you,
Claudiu Beznea
>
> Gr{oetje,eeting}s,
>
> Geert
>
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
> -- Linus Torvalds
Hi Claudiu,
On Tue, Apr 23, 2024 at 2:47 PM claudiu beznea <[email protected]> wrote:
> On 23.04.2024 10:27, Geert Uytterhoeven wrote:
> > On Mon, Apr 22, 2024 at 1:11 PM Claudiu <[email protected]> wrote:
> >> From: Claudiu Beznea <[email protected]>
> >>
> >> In case the SCI is used as a UART console, no_console_suspend is
> >> available in bootargs and SCI is part of a software-controlled power
> >> domain we need to call device_set_wakeup_path(). This lets the power
> >> domain core code knows that this domain should not be powered off
> >
> > know
> >
> >> durring system suspend. Otherwise, the SCI power domain is turned off,
> >
> > during
> >
> >> nothing is printed while suspending and the suspend/resume process is
> >> blocked. This was detected on the RZ/G3S SoC while adding support
> >> for power domains.
> >>
> >> Signed-off-by: Claudiu Beznea <[email protected]>
> > If this works, we can remove the console_suspend_enabled handling
> > from drivers/pmdomain/renesas/rmobile-sysc.c, and revert commit
> > 309864dcf92b76fc ("genpd: imx: scu-pd: do not power off console if
> > no_console_suspend").
>
> OK, first I'll go with this patch and after things settles down with it
> I'll propose changes for rmobile-sysc and imx. Is this ok for you?
I have already made these changes to rmobile-sysc locally to test
your patch ;-)
It works fine on R-Mobile APE6, so
Tested-by: Geert Uytterhoeven <[email protected]>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68korg
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds