From nobody Sun Feb 8 19:58:42 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id DCD1B32F761; Mon, 26 Jan 2026 10:18:53 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422735; cv=none; b=GHmC3WZ+cCMaCbhV/PoNoed4ocnxzf8NR2EpJtcLPYLZUkZ8ByklY7HX+WDVkWvvU6Us3VenDK3AShWwxWRgBHzhxMNNzKc+jyN5ci6dK1xPVYoa7fAHdayVYZko0JQMTxYMk8zx68doBPrhTSPTqwHP7y4ijyjjgXSUxicp+44= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422735; c=relaxed/simple; bh=5SmnI7xKFFWoHVZZ/JELbovu9NDevXGbdihnZfFH59I=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=bdlSRQ92aLfa29iu6e/sX6p1QL+OqE/wibYvyKOrXdJmUfuRQMoO9/HJqh+H2JLb44o3N1pKknwADKTMeo06vvpHN2UnnbhcTonf81MOsCelXK60xuSVDtprYOkudLr1KIVQzOnpW2p/nGx4iOdjmJdED7hzrPa9+ceC70vpmOw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id DFCDF497; Mon, 26 Jan 2026 02:18:46 -0800 (PST) Received: from e135073.nice.arm.com (e135073.arm.com [10.34.125.23]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 8D06F3F632; Mon, 26 Jan 2026 02:18:49 -0800 (PST) From: Pierre Gondois To: linux-kernel@vger.kernel.org Cc: Jie Zhan , zhenglifeng1@huawei.com, Ionela Voinescu , Christian Loehle , sumitg@nvidia.com, Pierre Gondois , "Rafael J. Wysocki" , Viresh Kumar , Huang Rui , "Gautham R. Shenoy" , Mario Limonciello , Perry Yuan , Srinivas Pandruvada , Len Brown , Saravana Kannan , linux-pm@vger.kernel.org Subject: [PATCH 1/6] cpufreq: Remove per-CPU QoS constraint Date: Mon, 26 Jan 2026 11:18:10 +0100 Message-ID: <20260126101826.94030-2-pierre.gondois@arm.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260126101826.94030-1-pierre.gondois@arm.com> References: <20260126101826.94030-1-pierre.gondois@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" policy->max_freq_req represents the maximum allowed frequency as requested by the policyX/scaling_max_freq sysfs file. This request applies to all CPUs of the policy. It is not possible to request a per-CPU maximum frequency. Thus, the interaction between the policy boost and scaling_max_freq settings should be handled by adding a boost specific QoS constraint. This will be handled in the following patches. This patch reverts of: commit 1608f0230510 ("cpufreq: Fix re-boost issue after hotplugging a CPU") Signed-off-by: Pierre Gondois Reviewed-by: Lifeng Zheng --- drivers/cpufreq/cpufreq.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 4472bb1ec83c7..db414c052658b 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1481,10 +1481,6 @@ static int cpufreq_policy_online(struct cpufreq_poli= cy *policy, =20 blocking_notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_CREATE_POLICY, policy); - } else { - ret =3D freq_qos_update_request(policy->max_freq_req, policy->max); - if (ret < 0) - goto out_destroy_policy; } =20 if (cpufreq_driver->get && has_target()) { --=20 2.43.0 From nobody Sun Feb 8 19:58:42 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 105B732F761; Mon, 26 Jan 2026 10:18:58 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422740; cv=none; b=t2wG5o+QjYhW7PbBr+d3RyZzN+0Pnadf00bi+ZCtu8G880icF+5g3oVFiuxy18/qA2sICkfsfLIG8YyuiLcrQ/EKNCH4aA6Pkys3L0KPMwPqCzI4jNCjhCS9iU+sfe03QwTlvj1LgzEAc0HjokSA+hnRPwBOIZLY+YmHlMdhTOo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422740; c=relaxed/simple; bh=dGp34TShUS6ECyHuxDXtDkIqXYeyziBTGA6ETXCT6Zo=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=k34QFpaHT8gHrYZLeutrK8a48GTHDMqfrrO4iipmJB2pDh3EJr9xmYjOm/axcvC15oTLAYurqMBjTyKQVo7b31KS8lKang0gUk+SM4CXokzfhZ5nmDeDQ/nClG86zGqxuW18InB4JXXHL4s1VXYT/FMhsnv7JOmPXbYEANgYWKA= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 0BCB5339; Mon, 26 Jan 2026 02:18:52 -0800 (PST) Received: from e135073.nice.arm.com (e135073.arm.com [10.34.125.23]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 1A40B3F632; Mon, 26 Jan 2026 02:18:54 -0800 (PST) From: Pierre Gondois To: linux-kernel@vger.kernel.org Cc: Jie Zhan , zhenglifeng1@huawei.com, Ionela Voinescu , Christian Loehle , sumitg@nvidia.com, Pierre Gondois , "Rafael J. Wysocki" , Viresh Kumar , Huang Rui , "Gautham R. Shenoy" , Mario Limonciello , Perry Yuan , Srinivas Pandruvada , Len Brown , Saravana Kannan , linux-pm@vger.kernel.org Subject: [PATCH 2/6] cpufreq: Add boost_freq_req QoS request Date: Mon, 26 Jan 2026 11:18:11 +0100 Message-ID: <20260126101826.94030-3-pierre.gondois@arm.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260126101826.94030-1-pierre.gondois@arm.com> References: <20260126101826.94030-1-pierre.gondois@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" The Power Management Quality of Service (PM QoS) allows to aggregate constraints from multiple entities. It is currently used to manage the min/max frequency of a given policy. Frequency constraints can come for instance from: - Thermal framework: acpi_thermal_cpufreq_init() - Firmware: _PPC objects: acpi_processor_ppc_init() - User: by setting policyX/scaling_[min|max]_freq The minimum of the max frequency constraints is used to compute the resulting maximum allowed frequency. When enabling boost frequencies, the same frequency request object (policy->max_freq_req) as to handle requests from users is used. As a result, when setting: - scaling_max_freq - boost The last sysfs file used overwrites the request from the other sysfs file. To avoid this, create a per-policy boost_freq_req to save the boost constraints instead of overwriting the last scaling_max_freq constraint. Signed-off-by: Pierre Gondois --- drivers/cpufreq/cpufreq.c | 40 ++++++++++++++++++++++++++++++++++----- include/linux/cpufreq.h | 1 + 2 files changed, 36 insertions(+), 5 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index db414c052658b..c8fb4c6656e94 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1359,17 +1359,24 @@ static void cpufreq_policy_free(struct cpufreq_poli= cy *policy) /* Cancel any pending policy->update work before freeing the policy. */ cancel_work_sync(&policy->update); =20 - if (policy->max_freq_req) { + if (policy->max_freq_req || policy->boost_freq_req) { /* - * Remove max_freq_req after sending CPUFREQ_REMOVE_POLICY - * notification, since CPUFREQ_CREATE_POLICY notification was - * sent after adding max_freq_req earlier. + * Remove max/boost _freq_req after sending CPUFREQ_REMOVE_POLICY + * notification, since CPUFREQ_CREATE_POLICY notification was sent + * after adding max/boost _freq_req earlier. */ blocking_notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_REMOVE_POLICY, policy); - freq_qos_remove_request(policy->max_freq_req); } =20 + if (policy->boost_freq_req) { + freq_qos_remove_request(policy->boost_freq_req); + kfree(policy->boost_freq_req); + } + + if (policy->max_freq_req) + freq_qos_remove_request(policy->max_freq_req); + freq_qos_remove_request(policy->min_freq_req); kfree(policy->min_freq_req); =20 @@ -1479,6 +1486,29 @@ static int cpufreq_policy_online(struct cpufreq_poli= cy *policy, goto out_destroy_policy; } =20 + if (policy->boost_supported) { + policy->boost_freq_req =3D kzalloc(sizeof(*policy->boost_freq_req), + GFP_KERNEL); + if (!policy->boost_freq_req) { + ret =3D -ENOMEM; + goto out_destroy_policy; + } + + ret =3D freq_qos_add_request(&policy->constraints, + policy->boost_freq_req, + FREQ_QOS_MAX, + FREQ_QOS_MAX_DEFAULT_VALUE); + if (ret < 0) { + /* + * So we don't call freq_qos_remove_request() for an + * uninitialized request. + */ + kfree(policy->boost_freq_req); + policy->boost_freq_req =3D NULL; + goto out_destroy_policy; + } + } + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_CREATE_POLICY, policy); } diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 0465d1e6f72ac..c292a6a19e4f5 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -81,6 +81,7 @@ struct cpufreq_policy { struct freq_constraints constraints; struct freq_qos_request *min_freq_req; struct freq_qos_request *max_freq_req; + struct freq_qos_request *boost_freq_req; =20 struct cpufreq_frequency_table *freq_table; enum cpufreq_table_sorting freq_table_sorted; --=20 2.43.0 From nobody Sun Feb 8 19:58:42 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 6619932E729; Mon, 26 Jan 2026 10:19:04 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422746; cv=none; b=rJCdM94sNnUr/fpy0P/tn3fZX/SwD1ABrdKY/z2kBl8VwCeU9aNbOi90qQZPXqp3R0AkJYlIJbX5U94g5R13PORD4x6+IGUE0Ky6Ezt8mH5oekoWypprRaJ5joua6DwYq60qBSr37S77nBxu/bmAP0wwIy5VUR+cld89kkgj2J0= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422746; c=relaxed/simple; bh=qNgwY11334MhqPnATW7mp92jGX7npLU+UmtkVBoewhQ=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=uqxh7zbIZehr9I10MgOusXtEXMfxBLS0svNARW0ayPxazcQUW4XT6YexV50MA+OIRBh8Mh6a4ETeNoXEfHAX9eAd0hhndgtg+Dqb35ZKuBQxtKg+YPaPu92+sSNOMJ24RWjrrPuOEmqXBcrL1QkihXOyC5own1e3ee3CgQfR01E= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 58C9E497; Mon, 26 Jan 2026 02:18:57 -0800 (PST) Received: from e135073.nice.arm.com (e135073.arm.com [10.34.125.23]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 1E4D43F632; Mon, 26 Jan 2026 02:18:59 -0800 (PST) From: Pierre Gondois To: linux-kernel@vger.kernel.org Cc: Jie Zhan , zhenglifeng1@huawei.com, Ionela Voinescu , Christian Loehle , sumitg@nvidia.com, Pierre Gondois , "Rafael J. Wysocki" , Viresh Kumar , Huang Rui , "Gautham R. Shenoy" , Mario Limonciello , Perry Yuan , Srinivas Pandruvada , Len Brown , Saravana Kannan , linux-pm@vger.kernel.org Subject: [PATCH 3/6] cpufreq: Centralize boost freq QoS requests Date: Mon, 26 Jan 2026 11:18:12 +0100 Message-ID: <20260126101826.94030-4-pierre.gondois@arm.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260126101826.94030-1-pierre.gondois@arm.com> References: <20260126101826.94030-1-pierre.gondois@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" policy_set_boost() calls the cpufreq set_boost callback. Update the newly added boost_freq_req request from there: - whenever boost is toggled - to cover all possible paths Signed-off-by: Pierre Gondois --- drivers/cpufreq/cpufreq.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index c8fb4c6656e94..505da5b00e575 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -603,10 +603,19 @@ static int policy_set_boost(struct cpufreq_policy *po= licy, bool enable) policy->boost_enabled =3D enable; =20 ret =3D cpufreq_driver->set_boost(policy, enable); - if (ret) + if (ret) { policy->boost_enabled =3D !policy->boost_enabled; + return ret; + } =20 - return ret; + ret =3D freq_qos_update_request(policy->boost_freq_req, policy->cpuinfo.m= ax_freq); + if (ret < 0) { + policy->boost_enabled =3D !policy->boost_enabled; + cpufreq_driver->set_boost(policy, !policy->boost_enabled); + return ret; + } + + return 0; } =20 static ssize_t store_local_boost(struct cpufreq_policy *policy, --=20 2.43.0 From nobody Sun Feb 8 19:58:42 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 93365330B12; Mon, 26 Jan 2026 10:19:09 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422751; cv=none; b=N6Bk8ddMwteVg0J/b0FpMhdLC4tWhsO1GEvdvt0+jvBoim7imzR1Evj5dZ1YpVYKLxxgL2yUSIIxhoh5KT9Ip15VZ7CU4VPzmkZe7pTW7YZ+de7NsBaT1amQ9vyGjyPJJ4ge17IP+HiwADodeAEsvoOfHXbRXyl591YpPgbtSro= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422751; c=relaxed/simple; bh=bOKPuomEMnmOPegpG/zRmP8QZ+QTrtrDMU5RfPkZqpU=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=M9BpODdM34uyQQmUdl8WUr9yvdZRQHZN/pqaBjOyC7UgUdfIRpjenynoLikC9ze3FJcMyNxPdsVKjrvgyiyf7UKtUV+TAym6Q9GILkFIVNamGzNmslz5+MsmO0njhNPkos2V1Q8EVmAOkfY65zHvrEeegMlf6PdxnuKmD8OGEug= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 75A6F339; Mon, 26 Jan 2026 02:19:02 -0800 (PST) Received: from e135073.nice.arm.com (e135073.arm.com [10.34.125.23]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 86CB83F632; Mon, 26 Jan 2026 02:19:05 -0800 (PST) From: Pierre Gondois To: linux-kernel@vger.kernel.org Cc: Jie Zhan , zhenglifeng1@huawei.com, Ionela Voinescu , Christian Loehle , sumitg@nvidia.com, Pierre Gondois , "Rafael J. Wysocki" , Viresh Kumar , Huang Rui , "Gautham R. Shenoy" , Mario Limonciello , Perry Yuan , Srinivas Pandruvada , Len Brown , Saravana Kannan , linux-pm@vger.kernel.org Subject: [PATCH 4/6] cpufreq: Update .set_boost() callbacks to rely on boost_freq_req Date: Mon, 26 Jan 2026 11:18:13 +0100 Message-ID: <20260126101826.94030-5-pierre.gondois@arm.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260126101826.94030-1-pierre.gondois@arm.com> References: <20260126101826.94030-1-pierre.gondois@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" In the existing .set_boost() callbacks: - Don't update policy->max as this is done through the qos notifier cpufreq_notifier_max() which calls cpufreq_set_policy(). - Remove freq_qos_update_request() calls as the qos request is now done in policy_set_boost() and updates the new boost_freq_req Note: cpufreq_frequency_table_cpuinfo() is also called through: cpufreq_policy_online() \-cpufreq_table_validate_and_sort() \-cpufreq_frequency_table_cpuinfo() which relies on cpufreq_frequency_table_cpuinfo() to set policy->min and max initizalization at driver init. This regression is solved in the next patch. Note2: acpi-cpufreq.c seems to be the only cpufreq driver not setting cpuinfo.max_freq. Populate it the nominal frequency at driver init. Signed-off-by: Pierre Gondois --- drivers/cpufreq/acpi-cpufreq.c | 1 + drivers/cpufreq/amd-pstate.c | 2 -- drivers/cpufreq/cppc_cpufreq.c | 10 ++-------- drivers/cpufreq/cpufreq.c | 16 +++++++--------- drivers/cpufreq/freq_table.c | 7 +++---- 5 files changed, 13 insertions(+), 23 deletions(-) diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index e73a66785d69d..6a6e26e1be14a 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -857,6 +857,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy = *policy) * governor from selecting inadequate CPU frequencies. */ arch_set_max_freq_ratio(true); + policy->cpuinfo.max_freq =3D nominal_freq; } =20 policy->freq_table =3D freq_table; diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index c45bc98721d24..310d5938cbdf6 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -756,8 +756,6 @@ static int amd_pstate_cpu_boost_update(struct cpufreq_p= olicy *policy, bool on) else if (policy->cpuinfo.max_freq > nominal_freq) policy->cpuinfo.max_freq =3D nominal_freq; =20 - policy->max =3D policy->cpuinfo.max_freq; - if (cppc_state =3D=3D AMD_PSTATE_PASSIVE) { ret =3D freq_qos_update_request(&cpudata->req[1], policy->cpuinfo.max_fr= eq); if (ret < 0) diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c index 9eac77c4f2944..4c46c7ea318eb 100644 --- a/drivers/cpufreq/cppc_cpufreq.c +++ b/drivers/cpufreq/cppc_cpufreq.c @@ -775,17 +775,11 @@ static int cppc_cpufreq_set_boost(struct cpufreq_poli= cy *policy, int state) { struct cppc_cpudata *cpu_data =3D policy->driver_data; struct cppc_perf_caps *caps =3D &cpu_data->perf_caps; - int ret; =20 if (state) - policy->max =3D cppc_perf_to_khz(caps, caps->highest_perf); + policy->cpuinfo.max_freq =3D cppc_perf_to_khz(caps, caps->highest_perf); else - policy->max =3D cppc_perf_to_khz(caps, caps->nominal_perf); - policy->cpuinfo.max_freq =3D policy->max; - - ret =3D freq_qos_update_request(policy->max_freq_req, policy->max); - if (ret < 0) - return ret; + policy->cpuinfo.max_freq =3D cppc_perf_to_khz(caps, caps->nominal_perf); =20 return 0; } diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 505da5b00e575..43bf6aed90e49 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1503,10 +1503,14 @@ static int cpufreq_policy_online(struct cpufreq_pol= icy *policy, goto out_destroy_policy; } =20 + /* + * If boost is supported, + * init the constraint with cpuinfo.max_freq. + */ ret =3D freq_qos_add_request(&policy->constraints, policy->boost_freq_req, FREQ_QOS_MAX, - FREQ_QOS_MAX_DEFAULT_VALUE); + policy->cpuinfo.max_freq); if (ret < 0) { /* * So we don't call freq_qos_remove_request() for an @@ -2821,16 +2825,10 @@ int cpufreq_boost_set_sw(struct cpufreq_policy *pol= icy, int state) return -ENXIO; =20 ret =3D cpufreq_frequency_table_cpuinfo(policy); - if (ret) { + if (ret) pr_err("%s: Policy frequency update failed\n", __func__); - return ret; - } =20 - ret =3D freq_qos_update_request(policy->max_freq_req, policy->max); - if (ret < 0) - return ret; - - return 0; + return ret; } EXPORT_SYMBOL_GPL(cpufreq_boost_set_sw); =20 diff --git a/drivers/cpufreq/freq_table.c b/drivers/cpufreq/freq_table.c index 7f251daf03ce3..9b37f37c36389 100644 --- a/drivers/cpufreq/freq_table.c +++ b/drivers/cpufreq/freq_table.c @@ -49,16 +49,15 @@ int cpufreq_frequency_table_cpuinfo(struct cpufreq_poli= cy *policy) max_freq =3D freq; } =20 - policy->min =3D policy->cpuinfo.min_freq =3D min_freq; - policy->max =3D max_freq; + policy->cpuinfo.min_freq =3D min_freq; /* * If the driver has set its own cpuinfo.max_freq above max_freq, leave * it as is. */ if (policy->cpuinfo.max_freq < max_freq) - policy->max =3D policy->cpuinfo.max_freq =3D max_freq; + policy->cpuinfo.max_freq =3D max_freq; =20 - if (policy->min =3D=3D ~0) + if (min_freq =3D=3D ~0) return -EINVAL; else return 0; --=20 2.43.0 From nobody Sun Feb 8 19:58:42 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id D7DE1330640; Mon, 26 Jan 2026 10:19:14 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422756; cv=none; b=cld0+kyv2Pr5uHN0GWPGK2e0DHUcHp+S/GqBf01vVkhJh3qyOKVe8gKZLdOlFvf0S8OOo3bThdJgjD2YnoFrjzh9yP7E6W1NhaVR6V+fD+6GELXIkqqSVn8VYgp2PUV+FrUzk6QhRTN6QBx76C91VMGTsQy5b9HF566CQofSm+M= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422756; c=relaxed/simple; bh=c1eWi+UqnTIEqEHycUHgp/VD+sNLi0VbrxP4lpcjTEc=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=j+KgOVYm1zmrXW1Rbw2QdAPzH+zpVKW44HoppQ3qdJ9l964IjhdG+cmwziOdkDLf6bfaWvyfHLvs1w5b3c4pTuAJNO1el0nJzbSAPpKBthbDVpFOzYAo7uWCkJ49QjyqCkjNIpqloFBObJ0ehUhiIv3Gth+BoF7YjTr+NxcTzlU= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 762AF497; Mon, 26 Jan 2026 02:19:07 -0800 (PST) Received: from e135073.nice.arm.com (e135073.arm.com [10.34.125.23]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 796D53F632; Mon, 26 Jan 2026 02:19:10 -0800 (PST) From: Pierre Gondois To: linux-kernel@vger.kernel.org Cc: Jie Zhan , zhenglifeng1@huawei.com, Ionela Voinescu , Christian Loehle , sumitg@nvidia.com, Pierre Gondois , "Rafael J. Wysocki" , Viresh Kumar , Huang Rui , "Gautham R. Shenoy" , Mario Limonciello , Perry Yuan , Srinivas Pandruvada , Len Brown , Saravana Kannan , linux-pm@vger.kernel.org Subject: [PATCH 5/6] cpufreq: Set policy->min and max as real QoS constraints Date: Mon, 26 Jan 2026 11:18:14 +0100 Message-ID: <20260126101826.94030-6-pierre.gondois@arm.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260126101826.94030-1-pierre.gondois@arm.com> References: <20260126101826.94030-1-pierre.gondois@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" cpufreq_set_policy() will ultimately override the policy min/max values written in the .init() callback through: cpufreq_policy_online() \-cpufreq_init_policy() \-cpufreq_set_policy() \-/* Set policy->min/max */ Thus the policy min/max values provided are only temporary. There is an exception if CPUFREQ_NEED_INITIAL_FREQ_CHECK is set and: cpufreq_policy_online() \-cpufreq_init_policy() \-__cpufreq_driver_target() \-cpufreq_driver->target() is called. In this case, some drivers use the policy min/max values in their .target() callback before they are overridden. Check cpufreq drivers: - if their .target() callback doesn't use policy->min or max, remove the initialization - assuming policy->min or max values were populated as constraints, set them as QoS real constraints in cpufreq_policy_online() Signed-off-by: Pierre Gondois --- drivers/cpufreq/amd-pstate.c | 24 ++++++++++++------------ drivers/cpufreq/cpufreq-nforce2.c | 4 ++-- drivers/cpufreq/cpufreq.c | 16 ++++++++++++++-- drivers/cpufreq/gx-suspmod.c | 9 ++++----- drivers/cpufreq/intel_pstate.c | 3 --- drivers/cpufreq/pcc-cpufreq.c | 8 ++++---- drivers/cpufreq/pxa3xx-cpufreq.c | 4 ++-- drivers/cpufreq/virtual-cpufreq.c | 6 +++--- 8 files changed, 41 insertions(+), 33 deletions(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 310d5938cbdf6..aaafbe9b26cae 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -1003,12 +1003,12 @@ static int amd_pstate_cpu_init(struct cpufreq_polic= y *policy) =20 perf =3D READ_ONCE(cpudata->perf); =20 - policy->cpuinfo.min_freq =3D policy->min =3D perf_to_freq(perf, - cpudata->nominal_freq, - perf.lowest_perf); - policy->cpuinfo.max_freq =3D policy->max =3D perf_to_freq(perf, - cpudata->nominal_freq, - perf.highest_perf); + policy->cpuinfo.min_freq =3D perf_to_freq(perf, + cpudata->nominal_freq, + perf.lowest_perf); + policy->cpuinfo.max_freq =3D perf_to_freq(perf, + cpudata->nominal_freq, + perf.highest_perf); =20 ret =3D amd_pstate_cppc_enable(policy); if (ret) @@ -1485,12 +1485,12 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_p= olicy *policy) =20 perf =3D READ_ONCE(cpudata->perf); =20 - policy->cpuinfo.min_freq =3D policy->min =3D perf_to_freq(perf, - cpudata->nominal_freq, - perf.lowest_perf); - policy->cpuinfo.max_freq =3D policy->max =3D perf_to_freq(perf, - cpudata->nominal_freq, - perf.highest_perf); + policy->cpuinfo.min_freq =3D perf_to_freq(perf, + cpudata->nominal_freq, + perf.lowest_perf); + policy->cpuinfo.max_freq =3D perf_to_freq(perf, + cpudata->nominal_freq, + perf.highest_perf); policy->driver_data =3D cpudata; =20 ret =3D amd_pstate_cppc_enable(policy); diff --git a/drivers/cpufreq/cpufreq-nforce2.c b/drivers/cpufreq/cpufreq-nf= orce2.c index fbbbe501cf2dc..831102522ad64 100644 --- a/drivers/cpufreq/cpufreq-nforce2.c +++ b/drivers/cpufreq/cpufreq-nforce2.c @@ -355,8 +355,8 @@ static int nforce2_cpu_init(struct cpufreq_policy *poli= cy) min_fsb =3D NFORCE2_MIN_FSB; =20 /* cpuinfo and default policy values */ - policy->min =3D policy->cpuinfo.min_freq =3D min_fsb * fid * 100; - policy->max =3D policy->cpuinfo.max_freq =3D max_fsb * fid * 100; + policy->cpuinfo.min_freq =3D min_fsb * fid * 100; + policy->cpuinfo.max_freq =3D max_fsb * fid * 100; =20 return 0; } diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 43bf6aed90e49..209e2673ca50b 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1455,6 +1455,18 @@ static int cpufreq_policy_online(struct cpufreq_poli= cy *policy, cpumask_and(policy->cpus, policy->cpus, cpu_online_mask); =20 if (new_policy) { + unsigned int min, max; + + /* + * If the driver has set policy->min or max, + * use the value as a QoS request. + */ + min =3D max(FREQ_QOS_MIN_DEFAULT_VALUE, policy->min); + if (policy->max) + max =3D min(FREQ_QOS_MAX_DEFAULT_VALUE, policy->max); + else + max =3D FREQ_QOS_MAX_DEFAULT_VALUE; + for_each_cpu(j, policy->related_cpus) { per_cpu(cpufreq_cpu_data, j) =3D policy; add_cpu_dev_symlink(policy, j, get_cpu_device(j)); @@ -1469,7 +1481,7 @@ static int cpufreq_policy_online(struct cpufreq_polic= y *policy, =20 ret =3D freq_qos_add_request(&policy->constraints, policy->min_freq_req, FREQ_QOS_MIN, - FREQ_QOS_MIN_DEFAULT_VALUE); + min); if (ret < 0) { /* * So we don't call freq_qos_remove_request() for an @@ -1489,7 +1501,7 @@ static int cpufreq_policy_online(struct cpufreq_polic= y *policy, =20 ret =3D freq_qos_add_request(&policy->constraints, policy->max_freq_req, FREQ_QOS_MAX, - FREQ_QOS_MAX_DEFAULT_VALUE); + max); if (ret < 0) { policy->max_freq_req =3D NULL; goto out_destroy_policy; diff --git a/drivers/cpufreq/gx-suspmod.c b/drivers/cpufreq/gx-suspmod.c index 75b3ef7ec6796..57999b8d51fa2 100644 --- a/drivers/cpufreq/gx-suspmod.c +++ b/drivers/cpufreq/gx-suspmod.c @@ -397,7 +397,7 @@ static int cpufreq_gx_target(struct cpufreq_policy *pol= icy, =20 static int cpufreq_gx_cpu_init(struct cpufreq_policy *policy) { - unsigned int maxfreq; + unsigned int minfreq, maxfreq; =20 if (!policy || policy->cpu !=3D 0) return -ENODEV; @@ -418,11 +418,10 @@ static int cpufreq_gx_cpu_init(struct cpufreq_policy = *policy) policy->cpu =3D 0; =20 if (max_duration < POLICY_MIN_DIV) - policy->min =3D maxfreq / max_duration; + minfreq =3D maxfreq / max_duration; else - policy->min =3D maxfreq / POLICY_MIN_DIV; - policy->max =3D maxfreq; - policy->cpuinfo.min_freq =3D maxfreq / max_duration; + minfreq =3D maxfreq / POLICY_MIN_DIV; + policy->cpuinfo.min_freq =3D minfreq; policy->cpuinfo.max_freq =3D maxfreq; =20 return 0; diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index ec4abe3745736..bf2f7524d04a9 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -3047,9 +3047,6 @@ static int __intel_pstate_cpu_init(struct cpufreq_pol= icy *policy) policy->cpuinfo.max_freq =3D READ_ONCE(global.no_turbo) ? cpu->pstate.max_freq : cpu->pstate.turbo_freq; =20 - policy->min =3D policy->cpuinfo.min_freq; - policy->max =3D policy->cpuinfo.max_freq; - intel_pstate_init_acpi_perf_limits(policy); =20 policy->fast_switch_possible =3D true; diff --git a/drivers/cpufreq/pcc-cpufreq.c b/drivers/cpufreq/pcc-cpufreq.c index ac2e90a65f0c4..231edfe8cabaa 100644 --- a/drivers/cpufreq/pcc-cpufreq.c +++ b/drivers/cpufreq/pcc-cpufreq.c @@ -551,13 +551,13 @@ static int pcc_cpufreq_cpu_init(struct cpufreq_policy= *policy) goto out; } =20 - policy->max =3D policy->cpuinfo.max_freq =3D + policy->cpuinfo.max_freq =3D ioread32(&pcch_hdr->nominal) * 1000; - policy->min =3D policy->cpuinfo.min_freq =3D + policy->cpuinfo.min_freq =3D ioread32(&pcch_hdr->minimum_frequency) * 1000; =20 - pr_debug("init: policy->max is %d, policy->min is %d\n", - policy->max, policy->min); + pr_debug("init: max_freq is %d, min_freq is %d\n", + policy->cpuinfo.max_freq, policy->cpuinfo.min_freq); out: return result; } diff --git a/drivers/cpufreq/pxa3xx-cpufreq.c b/drivers/cpufreq/pxa3xx-cpuf= req.c index 4afa48d172dbe..f53b9d7edc76a 100644 --- a/drivers/cpufreq/pxa3xx-cpufreq.c +++ b/drivers/cpufreq/pxa3xx-cpufreq.c @@ -185,8 +185,8 @@ static int pxa3xx_cpufreq_init(struct cpufreq_policy *p= olicy) int ret =3D -EINVAL; =20 /* set default policy and cpuinfo */ - policy->min =3D policy->cpuinfo.min_freq =3D 104000; - policy->max =3D policy->cpuinfo.max_freq =3D + policy->cpuinfo.min_freq =3D 104000; + policy->cpuinfo.max_freq =3D (cpu_is_pxa320()) ? 806000 : 624000; policy->cpuinfo.transition_latency =3D 1000; /* FIXME: 1 ms, assumed */ =20 diff --git a/drivers/cpufreq/virtual-cpufreq.c b/drivers/cpufreq/virtual-cp= ufreq.c index 6ffa16d239b2b..6f7dbef7fda96 100644 --- a/drivers/cpufreq/virtual-cpufreq.c +++ b/drivers/cpufreq/virtual-cpufreq.c @@ -164,10 +164,10 @@ static int virt_cpufreq_get_freq_info(struct cpufreq_= policy *policy) policy->cpuinfo.min_freq =3D 1; policy->cpuinfo.max_freq =3D virt_cpufreq_get_perftbl_entry(policy->cpu,= 0); =20 - policy->min =3D policy->cpuinfo.min_freq; - policy->max =3D policy->cpuinfo.max_freq; + policy->cpuinfo.min_freq; + policy->cpuinfo.max_freq; =20 - policy->cur =3D policy->max; + policy->cur =3D policy->cpuinfo.max_freq; return 0; } =20 --=20 2.43.0 From nobody Sun Feb 8 19:58:42 2026 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 6AD99330D47; Mon, 26 Jan 2026 10:19:19 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422761; cv=none; b=PzvMnEN9lAkrUPbJsGoHTxzRnPiqXviZ5zQJLS589cDbcv7SFIFvoZOpe45Tcd+ktDKwo/+P4YHsLkb5R+MIOp/dChp995R7XSyyM8g+XzptSxN4PSzTf2JlzcvKB0hdsUJ4xhuq4cRhIMWHd9CzdtUyuTMdxi8dGYx709tlN5M= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1769422761; c=relaxed/simple; bh=dsB++htWglnJo1qJa8OcgIx7Ri+H3o9ijvfmCEjgPvw=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=QJo6gJZtfl5boxgDoee4DoZIhu7svQrhp59CuXbxscxpLKnqOCuonsNwym5uoyOQPV7LRoMIJIDoImg7Br2GXTxPLmI62tbI6DqpMjgLehPYPa/j9IiaJ3aGcuTBFYhaIbQ2w+MLqTqFXJqL8UXz8vtms9YvoxiF8COheYD9qNw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 975BE339; Mon, 26 Jan 2026 02:19:12 -0800 (PST) Received: from e135073.nice.arm.com (e135073.arm.com [10.34.125.23]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 8DE993F632; Mon, 26 Jan 2026 02:19:15 -0800 (PST) From: Pierre Gondois To: linux-kernel@vger.kernel.org Cc: Jie Zhan , zhenglifeng1@huawei.com, Ionela Voinescu , Christian Loehle , sumitg@nvidia.com, Pierre Gondois , "Rafael J. Wysocki" , Viresh Kumar , Huang Rui , "Gautham R. Shenoy" , Mario Limonciello , Perry Yuan , Srinivas Pandruvada , Len Brown , Saravana Kannan , linux-pm@vger.kernel.org Subject: [RFC PATCH 6/6] cpufreq/freq_table: Allow decreasing cpuinfo.max_freq Date: Mon, 26 Jan 2026 11:18:15 +0100 Message-ID: <20260126101826.94030-7-pierre.gondois@arm.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260126101826.94030-1-pierre.gondois@arm.com> References: <20260126101826.94030-1-pierre.gondois@arm.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Drivers not using freq. tables update cpuinfo.max_freq in their .set_boost() callback. E.g. amd-pstate, cppc_cpufreq. Drivers relying on freq. tables and supporting boost frequencies rely on cpufreq_frequency_table_cpuinfo(). cpuinfo.max_freq is only updated if the new maximal value is higher than the previous one. Using the scmi-cpufreq driver which relies on freq. tables, enabling boost will permanently increases the cpuinfo.max_freq value. This patch allows to lower cpuinfo.max_freq. Note: commit 538b0188da46 ("cpufreq: ACPI: Set cpuinfo.max_freq directly if max boost is known") favored having cpuinfo.max_freq reporting the maximal boosted frequency of a CPU instead of the maximal reachable frequency due to regressions in the frequency reported by cpuinfo.max and scaling_cur_freq. As stated above, this is not what most of the other cpufreq driver do. I assume that the following patch: commit 3c55e94c0ade ("cpufreq: ACPI: Extend frequency tables to cover boost frequencies") was correct, but might not have tagged the boosted frequency with the CPUFREQ_BOOST_FREQ flag in the freq. table. Signed-off-by: Pierre Gondois --- drivers/cpufreq/freq_table.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/cpufreq/freq_table.c b/drivers/cpufreq/freq_table.c index 9b37f37c36389..bd08cbe9e9ba3 100644 --- a/drivers/cpufreq/freq_table.c +++ b/drivers/cpufreq/freq_table.c @@ -50,12 +50,7 @@ int cpufreq_frequency_table_cpuinfo(struct cpufreq_polic= y *policy) } policy->cpuinfo.min_freq =3D min_freq; - /* - * If the driver has set its own cpuinfo.max_freq above max_freq, leave - * it as is. - */ - if (policy->cpuinfo.max_freq < max_freq) - policy->cpuinfo.max_freq =3D max_freq; + policy->cpuinfo.max_freq =3D max_freq; if (min_freq =3D=3D ~0) return -EINVAL; -- 2.43.0