From: Mario Limonciello <mario.limonciello@amd.com>
To: Perry Yuan <perry.yuan@amd.com>,
rafael.j.wysocki@intel.com, viresh.kumar@linaro.org,
Ray.Huang@amd.com, gautham.shenoy@amd.com,
Borislav.Petkov@amd.com
Cc: Alexander.Deucher@amd.com, Xinmei.Huang@amd.com,
Xiaojian.Du@amd.com, Li.Meng@amd.com, linux-pm@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/7] cpufreq: amd-pstate: implement cpb_boost sysfs entry for boost control
Date: Fri, 26 Jan 2024 09:56:47 -0600 [thread overview]
Message-ID: <233100b4-e765-48e4-a1c0-88d01c621bd0@amd.com> (raw)
In-Reply-To: <f94bfc67e95fca54a153e24d828ef27e26ee5c47.1706255676.git.perry.yuan@amd.com>
On 1/26/2024 02:08, Perry Yuan wrote:
> From: Perry Yuan <Perry.Yuan@amd.com>
>
> With this new sysfs entry `cpb_boost`created, user can change CPU boost
> state dynamically under `active` and `passive` modes.
What about guided mode?
> And the highest perf and frequency will also be updated as the boost
> state changing.
>
> 0: check current boost state
> cat /sys/devices/system/cpu/amd_pstate/cpb_boost
>
> 1: disable CPU boost
> sudo bash -c "echo 0 > /sys/devices/system/cpu/amd_pstate/cpb_boost"
>
> 2: enable CPU boost
> sudo bash -c "echo 1 > /sys/devices/system/cpu/amd_pstate/cpb_boost"
>
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=217931
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=217618
>
> Signed-off-by: Perry Yuan <Perry.Yuan@amd.com>
> ---
> drivers/cpufreq/amd-pstate.c | 97 ++++++++++++++++++++++++++++++++++++
> 1 file changed, 97 insertions(+)
>
> diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c
> index 0dc9124140d4..b37bea7440b9 100644
> --- a/drivers/cpufreq/amd-pstate.c
> +++ b/drivers/cpufreq/amd-pstate.c
> @@ -1036,6 +1036,101 @@ static ssize_t status_store(struct device *a, struct device_attribute *b,
> return ret < 0 ? ret : count;
> }
>
> +static int amd_cpu_boost_update(struct amd_cpudata *cpudata, u32 on)
> +{
> + struct cpufreq_policy *policy = cpufreq_cpu_acquire(cpudata->cpu);
> + struct cppc_perf_ctrls perf_ctrls;
> + u32 highest_perf, nominal_perf;
> + int ret;
> +
> + if (!policy)
> + return -ENODATA;
> +
> + highest_perf = READ_ONCE(cpudata->highest_perf);
> + nominal_perf = READ_ONCE(cpudata->nominal_perf);
> +
> + if (boot_cpu_has(X86_FEATURE_CPPC)) {
> + u64 value = READ_ONCE(cpudata->cppc_req_cached);
> +
> + value &= ~GENMASK_ULL(7, 0);
I think it's better to use the macro AMD_CPPC_MAX_PERF in this case.
The less magic masks in the code, the easier it is to follow.
> + value |= on ? highest_perf : nominal_perf;
> + WRITE_ONCE(cpudata->cppc_req_cached, value);
> +
> + wrmsrl_on_cpu(cpudata->cpu, MSR_AMD_CPPC_REQ, value);
> +
> + } else {
> + perf_ctrls.max_perf = on ? highest_perf : nominal_perf;
> + ret = cppc_set_epp_perf(cpudata->cpu, &perf_ctrls, 1);
> + if (ret) {
> + pr_debug("failed to set energy perf value (%d)\n", ret);
> + return ret;
> + }
> + }
> +
> + if (on)
> + policy->cpuinfo.max_freq = cpudata->max_freq;
> + else
> + policy->cpuinfo.max_freq = cpudata->nominal_freq;
> +
> + policy->max = policy->cpuinfo.max_freq;
> +
> + if (cppc_state == AMD_PSTATE_PASSIVE) {
> + ret = freq_qos_update_request(&cpudata->req[1],
> + policy->cpuinfo.max_freq);
> + }
> +
> + cpufreq_cpu_release(policy);
> +
> + return ret;
> +}
> +
> +static ssize_t cpb_boost_show(struct device *dev,
> + struct device_attribute *attr, char *buf)
> +{
> + return sysfs_emit(buf, "%u\n", global.cpb_boost);
> +}
> +
> +static ssize_t cpb_boost_store(struct device *dev, struct device_attribute *b,
> + const char *buf, size_t count)
> +{
> + bool new_state;
> + ssize_t ret;
> + int cpu;
> +
> + mutex_lock(&amd_pstate_driver_lock);
> + if (!global.cpb_supported) {
> + pr_err("Boost mode is not supported by this processor or SBIOS\n");
> + return -EINVAL;
> + }
> +
> + ret = kstrtobool(buf, &new_state);
> + if (ret)
> + return -EINVAL;
> +
> + global.cpb_boost = !!new_state;
> +
> + for_each_possible_cpu(cpu) {
> +
> + struct cpufreq_policy *policy = cpufreq_cpu_get(cpu);
> + struct amd_cpudata *cpudata = policy->driver_data;
> +
> + if (!cpudata) {
> + pr_err("cpudata is NULL\n");
> + ret = -ENODATA;
> + cpufreq_cpu_put(policy);
> + goto err_exit;
> + }
> +
> + amd_cpu_boost_update(cpudata, global.cpb_boost);
> + refresh_frequency_limits(policy);
> + cpufreq_cpu_put(policy);
> + }
> +
> +err_exit:
> + mutex_unlock(&amd_pstate_driver_lock);
> + return ret < 0 ? ret : count;
> +}
> +
> cpufreq_freq_attr_ro(amd_pstate_max_freq);
> cpufreq_freq_attr_ro(amd_pstate_lowest_nonlinear_freq);
>
> @@ -1043,6 +1138,7 @@ cpufreq_freq_attr_ro(amd_pstate_highest_perf);
> cpufreq_freq_attr_rw(energy_performance_preference);
> cpufreq_freq_attr_ro(energy_performance_available_preferences);
> static DEVICE_ATTR_RW(status);
> +static DEVICE_ATTR_RW(cpb_boost);
>
> static struct freq_attr *amd_pstate_attr[] = {
> &amd_pstate_max_freq,
> @@ -1062,6 +1158,7 @@ static struct freq_attr *amd_pstate_epp_attr[] = {
>
> static struct attribute *pstate_global_attributes[] = {
> &dev_attr_status.attr,
> + &dev_attr_cpb_boost.attr,
> NULL
> };
>
next prev parent reply other threads:[~2024-01-26 15:56 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-26 8:08 [PATCH 0/7] AMD Pstate Driver Core Performance Boost Perry Yuan
2024-01-26 8:08 ` [PATCH 1/7] cpufreq: amd-pstate: remove set_boost callback for passive mode Perry Yuan
2024-01-26 15:45 ` Mario Limonciello
2024-01-29 4:46 ` Yuan, Perry
2024-01-26 8:08 ` [PATCH 2/7] cpufreq: amd-pstate: initialize new core precision boost state Perry Yuan
2024-01-26 16:01 ` Mario Limonciello
2024-01-27 19:01 ` kernel test robot
2024-02-11 18:19 ` kernel test robot
2024-02-14 6:37 ` Gautham R. Shenoy
2024-01-26 8:08 ` [PATCH 3/7] cpufreq: amd-pstate: implement cpb_boost sysfs entry for boost control Perry Yuan
2024-01-26 15:56 ` Mario Limonciello [this message]
2024-01-29 5:22 ` Yuan, Perry
2024-01-26 8:08 ` [PATCH 4/7] cpufreq: amd-pstate: fix max_perf calculation for amd_get_max_freq() Perry Yuan
2024-01-26 15:54 ` Mario Limonciello
2024-01-29 5:16 ` Yuan, Perry
2024-01-26 8:08 ` [PATCH 5/7] cpufreq: amd-pstate: fix the MSR highest perf will be reset issue while cpb boost off Perry Yuan
2024-01-26 8:08 ` [PATCH 6/7] cpufreq:amd-pstate: add suspend and resume callback for passive mode Perry Yuan
2024-01-26 15:52 ` Mario Limonciello
2024-01-26 8:08 ` [PATCH 7/7] Documentation: cpufreq: amd-pstate: introduce the new cpu boost control method Perry Yuan
2024-01-26 15:47 ` Mario Limonciello
2024-01-29 5:07 ` Yuan, Perry
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=233100b4-e765-48e4-a1c0-88d01c621bd0@amd.com \
--to=mario.limonciello@amd.com \
--cc=Alexander.Deucher@amd.com \
--cc=Borislav.Petkov@amd.com \
--cc=Li.Meng@amd.com \
--cc=Ray.Huang@amd.com \
--cc=Xiaojian.Du@amd.com \
--cc=Xinmei.Huang@amd.com \
--cc=gautham.shenoy@amd.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=perry.yuan@amd.com \
--cc=rafael.j.wysocki@intel.com \
--cc=viresh.kumar@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox