linux-kernel - Re: [PATCH v6 3/6] cpufreq:amd-pstate: initialize nominal

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 8 Feb 2024 12:04:34 +0530
From: "Gautham R. Shenoy" <gautham.shenoy@....com>
To: Perry Yuan <perry.yuan@....com>
Cc: rafael.j.wysocki@...el.com, Mario.Limonciello@....com,
	viresh.kumar@...aro.org, Ray.Huang@....com, Borislav.Petkov@....com,
	Alexander.Deucher@....com, Xinmei.Huang@....com,
	Xiaojian.Du@....com, Li.Meng@....com, linux-pm@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 3/6] cpufreq:amd-pstate: initialize nominal_freq of
 each cpudata


On Thu, Feb 08, 2024 at 11:46:30AM +0800, Perry Yuan wrote:
> Optimizes the process of retrieving the nominal frequency by utilizing
> 'cpudata->nominal_freq' instead of repeatedly accessing the cppc_acpi interface.
> 
> To enhance efficiency and reduce the CPU load, shifted to using
> 'cpudata->nominal_freq'. It allows for the nominal frequency to be accessed
> directly from the cached data in 'cpudata' of each CPU.
> It will also slightly reduce the frequency change latency while using pstate
> driver passive mode.
> 
> Reviewed-by: Mario Limonciello <mario.limonciello@....com>
> Signed-off-by: Perry Yuan <perry.yuan@....com>
> ---
>  drivers/cpufreq/amd-pstate.c | 24 ++++++++++++------------
>  1 file changed, 12 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
> index ac7faa98a450..ea8681ea3bad 100644
> --- a/drivers/cpufreq/amd-pstate.c
> +++ b/drivers/cpufreq/amd-pstate.c
> @@ -619,7 +619,7 @@ static int amd_get_max_freq(struct amd_cpudata *cpudata)
>  	if (ret)
>  		return ret;
>  
> -	nominal_freq = cppc_perf.nominal_freq;
> +	nominal_freq = READ_ONCE(cpudata->nominal_freq);

Does it make sense to call cppc_get_perf_caps() in
amd_pstate_cpu_init() and amd_pstate_epp_cpu_init() and cache all the
relevant metrics such as highest_perf, lowest_perf,
lowest_nonlinear_perf, nominal_perf, nominal_freq, lowest_freq so that
subsequent calls to amd_get_max/min/lowest_nonlinear/nominal_freq()
can use these cached values instead of calling cppc_get_perf_caps()
again ?

Because even with this patch, we continue to call cppc_get_perf_caps()
in amd_get_max_freq(), amd_get_min_freq() and
amd_get_lowest_nonlinear_freq().


>  	nominal_perf = READ_ONCE(cpudata->nominal_perf);
>  	max_perf = READ_ONCE(cpudata->highest_perf);
>  
> @@ -654,7 +654,7 @@ static int amd_get_lowest_nonlinear_freq(struct amd_cpudata *cpudata)
>  	if (ret)
>  		return ret;
>  
> -	nominal_freq = cppc_perf.nominal_freq;
> +	nominal_freq = READ_ONCE(cpudata->nominal_freq);
>  	nominal_perf = READ_ONCE(cpudata->nominal_perf);
>  
>  	lowest_nonlinear_perf = cppc_perf.lowest_nonlinear_perf;
> @@ -848,13 +848,14 @@ static int amd_pstate_cpu_init(struct cpufreq_policy *policy)
>  		goto free_cpudata1;
>  
>  	min_freq = amd_get_min_freq(cpudata);
> -	max_freq = amd_get_max_freq(cpudata);
>  	nominal_freq = amd_get_nominal_freq(cpudata);
> +	cpudata->nominal_freq = nominal_freq;
> +	max_freq = amd_get_max_freq(cpudata);
>  	lowest_nonlinear_freq = amd_get_lowest_nonlinear_freq(cpudata);
>  
> -	if (min_freq < 0 || max_freq < 0 || min_freq > max_freq) {
> -		dev_err(dev, "min_freq(%d) or max_freq(%d) value is incorrect\n",
> -			min_freq, max_freq);
> +	if (min_freq < 0 || max_freq < 0 || min_freq > max_freq || nominal_freq == 0) {

+1 for the "nominal_freq == 0" check. On CPPCv2, nominal_freq and
lowest_freq are not advertised. So nominal_freq will be 0. It is
better to fail here than failing later at the time of governor
initialization.


> +		dev_err(dev, "min_freq(%d) or max_freq(%d) or nominal_freq(%d) is incorrect\n",
> +			min_freq, max_freq, nominal_freq);
>  		ret = -EINVAL;
>  		goto free_cpudata1;
>  	}
> @@ -893,7 +894,6 @@ static int amd_pstate_cpu_init(struct cpufreq_policy *policy)
>  	cpudata->min_freq = min_freq;
>  	cpudata->max_limit_freq = max_freq;
>  	cpudata->min_limit_freq = min_freq;
> -	cpudata->nominal_freq = nominal_freq;
>  	cpudata->lowest_nonlinear_freq = lowest_nonlinear_freq;
>  
>  	policy->driver_data = cpudata;
> @@ -1310,12 +1310,13 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy)
>  		goto free_cpudata1;
>  
>  	min_freq = amd_get_min_freq(cpudata);
> -	max_freq = amd_get_max_freq(cpudata);
>  	nominal_freq = amd_get_nominal_freq(cpudata);
> +	cpudata->nominal_freq = nominal_freq;
> +	max_freq = amd_get_max_freq(cpudata);
>  	lowest_nonlinear_freq = amd_get_lowest_nonlinear_freq(cpudata);
> -	if (min_freq < 0 || max_freq < 0 || min_freq > max_freq) {
> -		dev_err(dev, "min_freq(%d) or max_freq(%d) value is incorrect\n",
> -				min_freq, max_freq);
> +	if (min_freq < 0 || max_freq < 0 || min_freq > max_freq || nominal_freq == 0) {


Ditto.

> +		dev_err(dev, "min_freq(%d) or max_freq(%d) or nominal_freq(%d) is incorrect\n",
> +				min_freq, max_freq, nominal_freq);
>  		ret = -EINVAL;
>  		goto free_cpudata1;
>  	}
> @@ -1328,7 +1329,6 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy)
>  	/* Initial processor data capability frequencies */
>  	cpudata->max_freq = max_freq;
>  	cpudata->min_freq = min_freq;
> -	cpudata->nominal_freq = nominal_freq;
>  	cpudata->lowest_nonlinear_freq = lowest_nonlinear_freq;


--
Thanks and Regards
gautham.