[PATCH] powerpc: Support CMDLINE_EXTEND

[Chris Packham]

On Wed, 2019-07-31 at 09:23 +0200, Christophe Leroy wrote:
> 
> Le 30/07/2019 à 23:10, Chris Packham a écrit :
> > 
> > Hi Christophe,
> > 
> > On Tue, 2019-07-30 at 09:02 +0200, Christophe Leroy wrote:
> > > 
> > > 
> > > Le 24/07/2019 à 07:33, Chris Packham a écrit :
> > > > 
> > > > 
> > > > Device tree aware platforms can make use of CMDLINE_EXTEND to
> > > > extend the
> > > > kernel command line provided by the bootloader. This is
> > > > particularly
> > > > useful to set parameters for built-in modules that would
> > > > otherwise
> > > > be
> > > > done at module insertion. Add support for this in the powerpc
> > > > architecture.
> > > > 
> > > > Signed-off-by: Chris Packham <chris.packham at alliedtelesis.co.nz
> > > > >
> > > > ---
> > > >    arch/powerpc/Kconfig | 12 ++++++++++++
> > > I think you also have to implement some stuff in
> > > early_cmdline_parse()
> > > in arch/powerpc/kernel/prom_init.c
> > I my case I didn't need to since the generic code
> > in drivers/of/fdt.c
> > did what I need. For early options or platforms that don't use a
> > device
> > tree then I can see why I'd need the update to update to prom_init.
> > 
> > > 
> > > 
> > > Maybe look at https://patchwork.ozlabs.org/patch/1074126/
> > > 
> > Do you mind if I take this and fold it into a v2 of my patch? Any
> > particular reason it didn't get picked up in April?
> Sure, take it, I don't mind.
> 
> Two reasons it was not picked up in April I believe:
> - It was part of a larger series 
> (https://patchwork.ozlabs.org/project/linuxppc-dev/list/?series=10051
> 8) 
> and was intended to challenge the series proposed by Daniel 
> (https://patchwork.ozlabs.org/project/linuxppc-dev/list/?series=98106
> ) 
> but nothing happened.
> - It was conflicting with the ongoing changes for implementing KASAN.
> 
> What you will have to do is to define prom_strlcat() in the same
> spirit 
> as https://patchwork.ozlabs.org/patch/1091621/ by copying it from 
> lib/string.c and 

Is it OK to use BUG_ON in prom_init? If I copy it verbatim then the
code from lib/string.c has a BUG_ON. I could probably change that to
if(x) return -1 if BUG_ON is not appropriate.

> I think you'll be able to drop prom_strlcpy() as that 
> function what only used there.

I think I need to keep prom_strlcpy to handle the CMDLINE_FORCE case.

> 
> Christophe
> 
> > 
> > 
> > > 
> > > Christophe
> > > 
> > > > 
> > > > 
> > > >    1 file changed, 12 insertions(+)
> > > > 
> > > > diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
> > > > index d8dcd8820369..cd9b3974aa36 100644
> > > > --- a/arch/powerpc/Kconfig
> > > > +++ b/arch/powerpc/Kconfig
> > > > @@ -851,6 +851,11 @@ config CMDLINE
> > > >    	  some command-line options at build time by
> > > > entering
> > > > them here.  In
> > > >    	  most cases you will need to specify the root
> > > > device
> > > > here.
> > > >    
> > > > +choice
> > > > +	prompt "Kernel command line type" if CMDLINE != ""
> > > > +	default CMDLINE_FORCE
> > > > +	depends on CMDLINE_BOOL
> > > > +
> > > >    config CMDLINE_FORCE
> > > >    	bool "Always use the default kernel command string"
> > > >    	depends on CMDLINE_BOOL
> > > > @@ -860,6 +865,13 @@ config CMDLINE_FORCE
> > > >    	  This is useful if you cannot or don't want to
> > > > change
> > > > the
> > > >    	  command-line options your boot loader passes to
> > > > the
> > > > kernel.
> > > >    
> > > > +config CMDLINE_EXTEND
> > > > +	bool "Extend bootloader kernel arguments"
> > > > +	help
> > > > +	  The command-line arguments provided by the boot
> > > > loader
> > > > will be
> > > > +	  appended to the default kernel command string.
> > > > +endchoice
> > > > +
> > > >    config EXTRA_TARGETS
> > > >    	string "Additional default image types"
> > > >    	help