[PATCH 7/9] lib/cpumask: add num_{possible,present,active}_cpus_{eq,gt,le}

Yury Norov yury.norov at gmail.com
Sun Nov 28 10:47:12 PST 2021


On Sun, Nov 28, 2021 at 12:54:00PM -0500, Dennis Zhou wrote:
> Hello,
> 
> On Sun, Nov 28, 2021 at 09:43:20AM -0800, Yury Norov wrote:
> > On Sun, Nov 28, 2021 at 09:07:52AM -0800, Joe Perches wrote:
> > > On Sat, 2021-11-27 at 19:57 -0800, Yury Norov wrote:
> > > > Add num_{possible,present,active}_cpus_{eq,gt,le} and replace num_*_cpus()
> > > > with one of new functions where appropriate. This allows num_*_cpus_*()
> > > > to return earlier depending on the condition.
> > > []
> > > > diff --git a/arch/arc/kernel/smp.c b/arch/arc/kernel/smp.c
> > > []
> > > > @@ -103,7 +103,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
> > > >  	 * if platform didn't set the present map already, do it now
> > > >  	 * boot cpu is set to present already by init/main.c
> > > >  	 */
> > > > -	if (num_present_cpus() <= 1)
> > > > +	if (num_present_cpus_le(2))
> > > >  		init_cpu_present(cpu_possible_mask);
> > > 
> > > ?  is this supposed to be 2 or 1
> > 
> > X <= 1 is the equivalent of X < 2.
> > 
> > > > diff --git a/drivers/cpufreq/pcc-cpufreq.c b/drivers/cpufreq/pcc-cpufreq.c
> > > []
> > > > @@ -593,7 +593,7 @@ static int __init pcc_cpufreq_init(void)
> > > >  		return ret;
> > > >  	}
> > > >  
> > > > -	if (num_present_cpus() > 4) {
> > > > +	if (num_present_cpus_gt(4)) {
> > > >  		pcc_cpufreq_driver.flags |= CPUFREQ_NO_AUTO_DYNAMIC_SWITCHING;
> > > >  		pr_err("%s: Too many CPUs, dynamic performance scaling disabled\n",
> > > >  		       __func__);
> > > 
> > > It looks as if the present variants should be using the same values
> > > so the _le test above with 1 changed to 2 looks odd.
> >  
> 
> I think the confusion comes from le meaning less than rather than lt.
> Given the general convention of: lt (<), le (<=), eg (=), ge (>=),
> gt (>), I'd consider renaming your le to lt.

Ok, makes sense. I'll rename in v2 and add <= and >= versions.



More information about the linux-riscv mailing list