[PATCH 4/5] ARM: S5PV210: Add support CPUFREQ

[JaeCheol Lee]

MyungJoo Ham wrote:
> 
> Hello,
> 
Hi,

> On Wed, Sep 15, 2010 at 4:52 PM, Jaecheol Lee <jc.lee at samsung.com> wrote:
> > This patch adds CPUFREQ driver for supporting DFS(Dynamic Frequency
> Scaling).
> >
> > Signed-off-by: Jaecheol Lee <jc.lee at samsung.com>
> > ---
> >  arch/arm/mach-s5pv210/cpufreq.c |  415
> +++++++++++++++++++++++++++++++++++++++
> >  1 files changed, 415 insertions(+), 0 deletions(-)
> >  create mode 100644 arch/arm/mach-s5pv210/cpufreq.c
> >
> > diff --git a/arch/arm/mach-s5pv210/cpufreq.c b/arch/arm/mach-
> s5pv210/cpufreq.c
> > new file mode 100644
> > index 0000000..aa39c2e
> 
> (snip)
> 
> > +static int s5pv210_target(struct cpufreq_policy *policy,
> > +                         unsigned int target_freq,
> > +                         unsigned int relation)
> > +{
> > +       unsigned long reg;
> > +       unsigned int index, priv_index;
> > +       unsigned int pll_changing = 0;
> > +       unsigned int bus_speed_changing = 0;
> > +
> > +       freqs.old = s5pv210_getspeed ;
> > +
> > +       if (cpufreq_frequency_table_target(policy, s5pv210_freq_table,
> > +                                          target_freq, relation, &index))
> > +               return -EINVAL;
> > +
> > +       freqs.new = s5pv210_freq_table[index].frequency;
> > +       freqs.cpu = 0;
> > +
> > +       if (freqs.new == freqs.old)
> > +               return 0;
> > +
> > +       /* Finding current running level index */
> > +       if (cpufreq_frequency_table_target(policy, s5pv210_freq_table,
> > +                                          freqs.old, relation, &priv_index))
> > +               return -EINVAL;
> > +
> > +       cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
> > +
> > +       if (freqs.new > freqs.old) {
> > +               /* Voltage up: will be implemented */
> 
> This could be crucial to the system stability. It is usually not optional.
> 
I think it does not occur any stability problem because current default voltage is maximum value.

Actually, tested its functionality on the SMDK with max8698 PMIC driver internally.
..but there is no max8698 device driver in the current mainline...

So this feature has been removed in this patch...
Anyway, will be implemented after submitting max8698 later.

> > +       }
> > +
> > +       /* Check if there need to change PLL */
> > +       if ((index == L0) || (priv_index == L0))
> > +               pll_changing = 1;
> > +
> > +       /* Check if there need to change System bus clock */
> > +       if ((index == L4) || (priv_index == L4))
> > +               bus_speed_changing = 1;
> > +
> > +       if (bus_speed_changing) {
> > +               /*
> > +                * Reconfigure DRAM refresh counter value for minimum
> > +                * temporary clock while changing divider.
> > +                * expected clock is 83Mhz : 7.8usec/(1/83Mhz) = 0x287
> > +                */
> > +               if (pll_changing)
> > +                       __raw_writel(0x287, S5P_VA_DMC1 + 0x30);
> > +               else
> > +                       __raw_writel(0x30c, S5P_VA_DMC1 + 0x30);
> > +
> > +               __raw_writel(0x287, S5P_VA_DMC0 + 0x30);
> 
> It'd be better to set values based on the real MPLL, APLL clock
> speeds, not hard-coding them.
> 

Will modify.

(snip)

> > +
> > +               /* Reconfigure DRAM refresh counter value */
> > +               if (index != L4) {
> > +                       /*
> > +                        * DMC0 : 166Mhz
> > +                        * DMC1 : 200Mhz
> > +                        */
> > +                       __raw_writel(0x618, S5P_VA_DMC1 + 0x30);
> > +                       __raw_writel(0x50e, S5P_VA_DMC0 + 0x30);
> > +               } else {
> > +                       /*
> > +                        * DMC0 : 83Mhz
> > +                        * DMC1 : 100Mhz
> > +                        */
> > +                       __raw_writel(0x30c, S5P_VA_DMC1 + 0x30);
> > +                       __raw_writel(0x287, S5P_VA_DMC0 + 0x30);
> > +               }
> 
> Same here. It'd better not be hard coded. We can't sure about the
> clock speeds for different boards.
> 

Will modify.
 
> > +       }
> > +
> > +       if (freqs.new < freqs.old) {
> > +               /* Voltage down: will be implemented */
> > +       }
> > +
> > +       cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
> > +
> > +       printk(KERN_INFO "Perf changed[L%d]\n", index);
> 
> This may incur too many clutters.
> 

Yeah, you're right...should be other print level such as KERN_DEBUG.

(snip)

Thanks.