Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1595852 > unrolled thread

Re: [PATCH] cpufreq: qoriq: enhance bus frequency calculation

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH] cpufreq: qoriq: enhance bus frequency calculation Viresh Kumar <viresh.kumar@linaro.org> - 2017-03-09 10:50 +0100
    RE: [PATCH] cpufreq: qoriq: enhance bus frequency calculation Andy Tang <andy.tang@nxp.com> - 2017-03-10 02:50 +0100
      Re: [PATCH] cpufreq: qoriq: enhance bus frequency calculation Viresh Kumar <viresh.kumar@linaro.org> - 2017-03-10 11:10 +0100
        RE: [PATCH] cpufreq: qoriq: enhance bus frequency calculation Andy Tang <andy.tang@nxp.com> - 2017-03-10 11:20 +0100

#1595852 — Re: [PATCH] cpufreq: qoriq: enhance bus frequency calculation

On 09-03-17, 16:15, YuanTian Tang wrote:
> From: Tang Yuantian <Yuantian.Tang@nxp.com>
> 
> On some platforms, property device-type may be missed in soc node
> in dts which caused the bus-frequency can not be obtained correctly.
> 
> This patch enhanced the bus-frequency calculation. When property
> device-type is missed in dts, bus-frequency will be obtained by
> looking up clock table to get platform clock and hence get its
> frequency.
> 
> Signed-off-by: Tang Yuantian <yuantian.tang@nxp.com>
> ---
>  drivers/cpufreq/qoriq-cpufreq.c | 24 +++++++++++++++++-------
>  1 file changed, 17 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/cpufreq/qoriq-cpufreq.c b/drivers/cpufreq/qoriq-cpufreq.c
> index bfec1bc..0f22e40 100644
> --- a/drivers/cpufreq/qoriq-cpufreq.c
> +++ b/drivers/cpufreq/qoriq-cpufreq.c
> @@ -52,17 +52,27 @@ static u32 get_bus_freq(void)
>  {
>  	struct device_node *soc;
>  	u32 sysfreq;
> +	struct clk *pltclk;
> +	int ret;
>  
> +	/* get platform freq by searching bus-frequency property */
>  	soc = of_find_node_by_type(NULL, "soc");
> -	if (!soc)
> -		return 0;
> -
> -	if (of_property_read_u32(soc, "bus-frequency", &sysfreq))
> -		sysfreq = 0;
> +	if (soc) {
> +		ret = of_property_read_u32(soc, "bus-frequency", &sysfreq);
> +		of_node_put(soc);
> +		if (!ret)
> +			return sysfreq;
> +	}
>  
> -	of_node_put(soc);
> +	/* get platform freq by its clock name */
> +	pltclk = clk_get(NULL, "cg-pll0-div1");

Will this always work? If yes, then what about dropping the code parsing DT
completely ? That is, just rely on clk_get_rate() in all cases.

> +	if (IS_ERR(pltclk)) {
> +		pr_err("%s: can't get bus frequency %ld\n",
> +		__func__, PTR_ERR(pltclk));

You need to properly align this. Try running checkpatch over this patch or:

checkpatch --strict

> +		return PTR_ERR(pltclk);
> +	}
>  
> -	return sysfreq;
> +	return clk_get_rate(pltclk);
>  }
>  
>  static struct clk *cpu_to_clk(int cpu)
> -- 
> 2.1.0.27.g96db324

-- 
viresh

[toc] | [next] | [standalone]

#1596549

Hi Viresh,

> -----Original Message-----
> From: Viresh Kumar [mailto:viresh.kumar@linaro.org]
> Sent: Thursday, March 09, 2017 5:39 PM
> To: Y.T. Tang
> Cc: rjw@rjwysocki.net; linux-pm@vger.kernel.org; linux-
> kernel@vger.kernel.org; linux-arm-kernel@lists.infradead.org; Y.T. Tang
> Subject: Re: [PATCH] cpufreq: qoriq: enhance bus frequency calculation
> 
> On 09-03-17, 16:15, YuanTian Tang wrote:
> > From: Tang Yuantian <Yuantian.Tang@nxp.com>
> >
> > On some platforms, property device-type may be missed in soc node in
> > dts which caused the bus-frequency can not be obtained correctly.
> >
> > This patch enhanced the bus-frequency calculation. When property
> > device-type is missed in dts, bus-frequency will be obtained by
> > looking up clock table to get platform clock and hence get its
> > frequency.
> >
> > Signed-off-by: Tang Yuantian <yuantian.tang@nxp.com>
> > ---
> >  drivers/cpufreq/qoriq-cpufreq.c | 24 +++++++++++++++++-------
> >  1 file changed, 17 insertions(+), 7 deletions(-)
> >
> > diff --git a/drivers/cpufreq/qoriq-cpufreq.c
> > b/drivers/cpufreq/qoriq-cpufreq.c index bfec1bc..0f22e40 100644
> > --- a/drivers/cpufreq/qoriq-cpufreq.c
> > +++ b/drivers/cpufreq/qoriq-cpufreq.c
> > @@ -52,17 +52,27 @@ static u32 get_bus_freq(void)  {
> >  	struct device_node *soc;
> >  	u32 sysfreq;
> > +	struct clk *pltclk;
> > +	int ret;
> >
> > +	/* get platform freq by searching bus-frequency property */
> >  	soc = of_find_node_by_type(NULL, "soc");
> > -	if (!soc)
> > -		return 0;
> > -
> > -	if (of_property_read_u32(soc, "bus-frequency", &sysfreq))
> > -		sysfreq = 0;
> > +	if (soc) {
> > +		ret = of_property_read_u32(soc, "bus-frequency", &sysfreq);
> > +		of_node_put(soc);
> > +		if (!ret)
> > +			return sysfreq;
> > +	}
> >
> > -	of_node_put(soc);
> > +	/* get platform freq by its clock name */
> > +	pltclk = clk_get(NULL, "cg-pll0-div1");
> 
> Will this always work? If yes, then what about dropping the code parsing DT
> completely ? That is, just rely on clk_get_rate() in all cases.
> 
We put all the clock tree configuration in driver, not in dts.  cg-pll0-div1 is hardcoded in driver since we don't depend on dts.  We kind of don't have other choices but use the hardcode clock name here too.

> > +	if (IS_ERR(pltclk)) {
> > +		pr_err("%s: can't get bus frequency %ld\n",
> > +		__func__, PTR_ERR(pltclk));
> 
> You need to properly align this. Try running checkpatch over this patch or:
> 
> checkpatch --strict
> 
I did check it with checkpatch script, but without --strict parameter.
After applying --strict, script tell the alignment issue. :)

Regards,
Andy

> > +		return PTR_ERR(pltclk);
> > +	}
> >
> > -	return sysfreq;
> > +	return clk_get_rate(pltclk);
> >  }
> >
> >  static struct clk *cpu_to_clk(int cpu)
> > --
> > 2.1.0.27.g96db324
> 
> --
> viresh

[toc] | [prev] | [next] | [standalone]

#1596958

On 10-03-17, 01:44, Andy Tang wrote:
> > Will this always work? If yes, then what about dropping the code parsing DT
> > completely ? That is, just rely on clk_get_rate() in all cases.
> > 
> We put all the clock tree configuration in driver, not in dts.
> cg-pll0-div1 is hardcoded in driver since we don't depend on dts.
> We kind of don't have other choices but use the hardcode clock name
> here too.

Looks like you misread my comment. Let me try again. Will it be fine
to write get_bus_freq() this way?

static u32 get_bus_freq(void)
{
	struct clk *pltclk;

	/* get platform freq by its clock name */
	pltclk = clk_get(NULL, "cg-pll0-div1");
	if (IS_ERR(pltclk)) {
		pr_err("%s: can't get bus frequency %ld\n",
		       __func__, PTR_ERR(pltclk));
		return PTR_ERR(pltclk);
	}

	return clk_get_rate(pltclk);
}

-- 
viresh

[toc] | [prev] | [next] | [standalone]

#1596994

Hi Viresh,

> -----Original Message-----
> From: Viresh Kumar [mailto:viresh.kumar@linaro.org]
> Sent: Friday, March 10, 2017 6:05 PM
> To: Andy Tang
> Cc: rjw@rjwysocki.net; linux-pm@vger.kernel.org; linux-
> kernel@vger.kernel.org; linux-arm-kernel@lists.infradead.org
> Subject: Re: [PATCH] cpufreq: qoriq: enhance bus frequency calculation
> 
> On 10-03-17, 01:44, Andy Tang wrote:
> > > Will this always work? If yes, then what about dropping the code
> > > parsing DT completely ? That is, just rely on clk_get_rate() in all cases.
> > >
> > We put all the clock tree configuration in driver, not in dts.
> > cg-pll0-div1 is hardcoded in driver since we don't depend on dts.
> > We kind of don't have other choices but use the hardcode clock name
> > here too.
> 
> Looks like you misread my comment. Let me try again. Will it be fine to write
> get_bus_freq() this way?
> 
> static u32 get_bus_freq(void)
> {
> 	struct clk *pltclk;
> 
> 	/* get platform freq by its clock name */
> 	pltclk = clk_get(NULL, "cg-pll0-div1");
> 	if (IS_ERR(pltclk)) {
> 		pr_err("%s: can't get bus frequency %ld\n",
> 		       __func__, PTR_ERR(pltclk));
> 		return PTR_ERR(pltclk);
> 	}
> 
> 	return clk_get_rate(pltclk);
> }
> 
Yes, we can. But for some legacy powerpc-based socs, this may not work.
powerpc-base socs are still  using legacy clock driver. For compatibility sake, we better be compatible with old ones. It would break any compatibility this way.

Regards,
Yuantian

> --
> viresh

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web