Re: CVS commit: src/sys/arch

[Nick Hudson <skrll%netbsd.org@localhost>]

On 11/07/17 16:09, Ryo Shimizu wrote:
> > On 11/07/17 09:05, Ryo Shimizu wrote:
> > > Module Name:	src
> > > Committed By:	ryo
> > > Date:		Tue Nov  7 09:05:06 UTC 2017
> > >
> > > Modified Files:
> > > 	src/sys/arch/arm/broadcom: bcm2835_intr.c
> > > 	src/sys/arch/evbarm/rpi: rpi_machdep.c
> > >
> > > Log Message:
> > > on RPI2, fix compile failure without options MULTIPROCESSOR.
> > Is this really necessary?
> Yes. As long as it is option, it should be able to select enable or disable.
> And it is useful for problem-isolation and debugging. (at least for me)

OK, I'll try and preserve the behaviour in my work.


> > @@ -626,22 +627,30 @@ rpi_bootparams(void)
> > >    static void
> > >    rpi_bootstrap(void)
> > >    {
> > > -#if defined(BCM2836)
> > > -	arm_cpu_max = 4;
> > > -	extern int cortex_mmuinfo;
> > > +#ifdef BCM2836
> > > +#define RPI_CPU_MAX	4
> > Why #if defined() -> #ifdef?
> >
> > Also please don't mix whitespace and real changes.
> >
> > This causes merge conflicts for me on my FDTisation of RPI which I'm
> > about to commit :(
> sorry. I commited without thinking.
> Is it better to revert the commit once?

Nevermind now. Please try and avoid this in the future.

> --
> ryo shimizu
Nick