Re: [PATCH v5 08/13] ARM: sunxi: Add initialization of CNTVOFF

[MylÃne Josserand]

Hello Maxime,

On Wed, 4 Apr 2018 09:45:15 +0200
Maxime Ripard <maxime.ripard@xxxxxxxxxxx> wrote:

> On Tue, Apr 03, 2018 at 10:06:28PM +0200, MylÃne Josserand wrote:
> > Hello,
> > 
> > Thank you for the review.
> > 
> > On Tue, 3 Apr 2018 11:12:18 +0200
> > Maxime Ripard <maxime.ripard@xxxxxxxxxxx> wrote:
> >   
> > > On Tue, Apr 03, 2018 at 08:18:31AM +0200, MylÃne Josserand wrote:  
> > > > Add the initialization of CNTVOFF for sun8i-a83t.
> > > > 
> > > > For boot CPU, Create a new machine that handles this
> > > > function's call in an "init_early" callback.
> > > > For secondary CPUs, add this function into secondary_startup
> > > > assembly entry.
> > > > 
> > > > Signed-off-by: MylÃne Josserand <mylene.josserand@xxxxxxxxxxx>
> > > > ---
> > > >  arch/arm/mach-sunxi/headsmp.S |  1 +
> > > >  arch/arm/mach-sunxi/sunxi.c   | 18 +++++++++++++++++-
> > > >  2 files changed, 18 insertions(+), 1 deletion(-)
> > > > 
> > > > diff --git a/arch/arm/mach-sunxi/headsmp.S b/arch/arm/mach-sunxi/headsmp.S
> > > > index 79890fbe5613..b586b7cf803a 100644
> > > > --- a/arch/arm/mach-sunxi/headsmp.S
> > > > +++ b/arch/arm/mach-sunxi/headsmp.S
> > > > @@ -71,6 +71,7 @@ ENDPROC(sunxi_mc_smp_cluster_cache_enable)
> > > >  
> > > >  ENTRY(sunxi_mc_smp_secondary_startup)
> > > >  	bl	sunxi_mc_smp_cluster_cache_enable
> > > > +	bl	smp_init_cntvoff
> > > >  	b	secondary_startup
> > > >  ENDPROC(sunxi_mc_smp_secondary_startup)
> > > >  
> > > > diff --git a/arch/arm/mach-sunxi/sunxi.c b/arch/arm/mach-sunxi/sunxi.c
> > > > index 5e9602ce1573..090784108c0a 100644
> > > > --- a/arch/arm/mach-sunxi/sunxi.c
> > > > +++ b/arch/arm/mach-sunxi/sunxi.c
> > > > @@ -16,6 +16,7 @@
> > > >  #include <linux/platform_device.h>
> > > >  
> > > >  #include <asm/mach/arch.h>
> > > > +#include <asm/smp_cntvoff.h>
> > > >  
> > > >  static const char * const sunxi_board_dt_compat[] = {
> > > >  	"allwinner,sun4i-a10",
> > > > @@ -62,7 +63,6 @@ MACHINE_END
> > > >  static const char * const sun8i_board_dt_compat[] = {
> > > >  	"allwinner,sun8i-a23",
> > > >  	"allwinner,sun8i-a33",
> > > > -	"allwinner,sun8i-a83t",
> > > >  	"allwinner,sun8i-h2-plus",
> > > >  	"allwinner,sun8i-h3",
> > > >  	"allwinner,sun8i-r40",
> > > > @@ -75,6 +75,22 @@ DT_MACHINE_START(SUN8I_DT, "Allwinner sun8i Family")
> > > >  	.dt_compat	= sun8i_board_dt_compat,
> > > >  MACHINE_END
> > > >  
> > > > +void __init sun8i_cntvoff_init(void)
> > > > +{
> > > > +	smp_init_cntvoff();    
> > > 
> > > Can't this be moved to the SMP setup code?  
> > 
> > I tried to put it in the first lines of "sunxi_mc_smp_init" function
> > but it did not work. I tried to find some callbacks to have an
> > early "init" and I only found the "init_early"'s one. There is probably
> > another way to handle that so do not hesitate to tell me any ideas.  
> 
> It's hard to say without more context about why it doesn't work. Have
> you checked the order between early_initcall, the timer initialization
> function and init_early?
> 

Yes, I tested it. I wanted to test it again to give more context. Here
is the boot log: http://code.bulix.org/n1x864-315948?raw
I added printk in the 3 functions (with "===========>").

The "init_early" and "sun6i_timer" are executed before
"arch_timer_of_init" (which is the function that parses the DT property
that we used previously "arm,cpu-registers-not-fw-configured").
The "sunxi_mc_smp_init" function is called later so I guess that is why
it is not working in that case.

> > > > +}
> > > > +
> > > > +static const char * const sun8i_cntvoff_board_dt_compat[] = {
> > > > +	"allwinner,sun8i-a83t",
> > > > +	NULL,
> > > > +};
> > > > +
> > > > +DT_MACHINE_START(SUN8I_CNTVOFF_DT, "Allwinner sun8i boards needing cntvoff")    
> > > 
> > > All of the SoCs need CNTVOFF, so that doesn't really make sense. Why
> > > not just calling it for what it is: an A83t?  
> > 
> > Sure, I will update it.  
> 
> Looking back at that code, I guess you want to change also the
> smp_init_cntvoff function. It's not really related to SMP either.

Yep, I will update it in next version.

Thanks,

Best regards,

-- 
MylÃne Josserand, Bootlin (formerly Free Electrons)
Embedded Linux and Kernel engineering
http://bootlin.com