Re: [PATCH v2] serial: 8250: Apply FSL workarounds also without SERIAL_8250_CONSOLE

["Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>]

[-- Attachment #1: Type: text/plain, Size: 2763 bytes --]

On Wed, 31 May 2023, Uwe Kleine-König wrote:

> On Wed, May 31, 2023 at 12:47:54PM +0300, Ilpo Järvinen wrote:
> > On Wed, 31 May 2023, Uwe Kleine-König wrote:
> > 
> > > The need to handle the FSL variant of 8250 in a special way is also
> > > present without console support. So soften the dependency for
> > > SERIAL_8250_FSL accordingly.
> > > 
> > > This issue was identified by Dominik Andreas Schorpp.
> > > 
> > > To cope for CONFIG_SERIAL_8250=m + CONFIG_SERIAL_8250_FSL=y, 8250_fsl.o
> > > must be put in the same compilation unit as 8250_port.o because the
> > > latter defines some functions needed in the former and so 8250_fsl.o
> > > must not be built-in if 8250_port.o is available in a module.
> > > 
> > > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > > ---
> > > Hello,
> > > 
> > > (implicit) v1 was already applied by Greg (a0807ca158e0 in tty-testing)
> > > but that didn't handle CONFIG_SERIAL_8250=m + CONFIG_SERIAL_8250_FSL=y
> > > correctly which was pointed out by the 0-day bot. (Thanks!)
> > 
> > That would warrant Reported-by (0-day's reports give you the tag).
> 
> I'd add this tag if I created a commit that fixes the broken commit.
> However I understood that if a v2 patch fixes a v1 that was broken, the
> tag is not to be added?! I don't feel strong here however, so if people
> agree that the tag should be there, I can add it.
> 
> > > That wasn't a problem before because SERIAL_8250_CONSOLE depends on
> > > SERIAL_8250=y.
> > > 
> > > Having said that I wonder if there are a few more .o files that should
> > > better be used with 8250_base-$(CONFIG_SERIAL_8250_XXX) instead of
> > > obj-$(CONFIG_SERIAL_8250_XXX).
> > > 
> > > Best regards
> > > Uwe
> > > 
> > >  drivers/tty/serial/8250/Kconfig  | 2 +-
> > >  drivers/tty/serial/8250/Makefile | 2 +-
> > >  2 files changed, 2 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/drivers/tty/serial/8250/Kconfig b/drivers/tty/serial/8250/Kconfig
> > > index 5313aa31930f..10c09b19c871 100644
> > > --- a/drivers/tty/serial/8250/Kconfig
> > > +++ b/drivers/tty/serial/8250/Kconfig
> > > @@ -378,7 +378,7 @@ config SERIAL_8250_BCM2835AUX
> > >  
> > >  config SERIAL_8250_FSL
> > >  	bool "Freescale 16550 UART support" if COMPILE_TEST && !(PPC || ARM || ARM64)
> > > -	depends on SERIAL_8250_CONSOLE
> > > +	depends on SERIAL_8250
> > 
> > Why this cannot simply be:
> > 	depends on SERIAL_8250=y
> 
> This doesn't work, because then the FSL-workarounds are missing if the
> 8250 driver is compiled as a module.

How can 8250 driver be a module and fsl still get enabled?

What I think (not a Kconfig expert for sure) would happen is that 8250_fsl 
won't be enabled at all if CONFIG_SERIAL_8250=m because it depends on 
SERIAL_8250=y.

-- 
 i.