From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 170F830C16E; Wed, 17 Jun 2026 16:14:34 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1781712882; cv=none; b=CAAuQNwfIQ0P/CpfnNKqGTtFf/HnPKxEx8HTiz3K596d5XXcp/xujFFA4b2ZQpMem1o66rG0G4WU4qA9+7tcKik0SEWce0XeGZMz2dGk9xfD5r2y5qzvd6oHZ4TgkKHCS7Xfxi2cuN7dljoIR7K8Oe1Cc/xUKsGrWqCisI5/4Jw= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1781712882; c=relaxed/simple; bh=TS1rzuSkSYvUZa0kKqZA4F19mYo09LbKUz7kLc2R7c4=; h=From:To:Cc:Subject:Date:Message-ID:MIME-Version:Content-Type; b=dx0abR7FXGXa/mDlJCFbuvpRGWIYJ0J3n7JXaTIAzFHNiK++gAnTxXxV8ZRKm37gzLH3wLYiZkH+Y3igNk31+KQhpnUyz2LUevTXXDVWjdJAUMM3J6iEw2wZ0kN7LYuoAmuRkEuvM5UvyfF9iERy55rWmdt7EgwVrqD60hCIxgc= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=NCaWe3/X; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="NCaWe3/X" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 3F16F1F000E9; Wed, 17 Jun 2026 16:14:32 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1781712873; bh=lEeXcVaibcbRS/Vz+saNgFaKALdonox1Eju25A6Eq2c=; h=From:To:Cc:Subject:Date; b=NCaWe3/XEf3xiLjb81l0msxUArWukc0y/lkEik4MCixXBzYaCg90ZPVKBJcpQTFuW dixanZ1ogS/hGH8He4npoUsnt/YP19YvNwveERCUmVklgNS9sv7nvx7SHiK/HtcZKZ 6HB7Ad9vnQOrGmgFQCVO3pbV7RwCdb1YhvDaa0vh/uk2K0s4skD+MZhaRPr1NEttv/ IoZvSAO3/rn3q+iaOAAj3dkxfXtL7D9LzoZ01Jn15edKMGygTV4jvVrqjLI6W/Dz+J 2uf4t0NwQVSJuVkGigyMxLQkmW+uaW651kn7DLCBYbojJbfzeZuuqK+QxgEoMHAlNs hnAKROmVm90Zg== From: "Rafael J. Wysocki" To: Linux PM Cc: LKML , Viresh Kumar , Srinivas Pandruvada , Christian Loehle , Peter Zijlstra , Vincent Guittot Subject: [PATCH v1] cpufreq: intel_pstate: Adjust the .adjust_perf() driver callback Date: Wed, 17 Jun 2026 18:14:30 +0200 Message-ID: <14060154.uLZWGnKmhe@rafael.j.wysocki> Organization: Linux Kernel Development Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="UTF-8" From: Rafael J. Wysocki In some cases, the processor may not actually stick to the "desired" performance level programmed through the driver's .adjust_perf() callback and may go above it, which may not be desirable (for instance, there may be a UCLAMP_MAX limit set for the task currently running on the given CPU which should be respected). Address that by adjusting the .adjust_perf() callback to take an additional argument, max_perf, representing the maximum allowed performance level of the CPU and update the intel_pstate driver to take that argument into account as appropriate. Accordingly, adjust cpufreq_driver_adjust_perf() and the other existing user of .adjust_perf(), which is the amd-pstate driver (but the behavior of that driver is not changed). While at it, also update the cpufreq_driver_adjust_perf() documentation to reflect this change and some previous code changes that have not been taken into account in it. Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/amd-pstate.c | 1 + drivers/cpufreq/cpufreq.c | 14 +++++++++----- drivers/cpufreq/intel_pstate.c | 9 ++++++++- include/linux/cpufreq.h | 2 ++ kernel/sched/cpufreq_schedutil.c | 4 +++- 5 files changed, 23 insertions(+), 7 deletions(-) --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -781,6 +781,7 @@ static unsigned int amd_pstate_fast_swit static void amd_pstate_adjust_perf(struct cpufreq_policy *policy, unsigned long _min_perf, unsigned long target_perf, + unsigned long _max_perf, unsigned long capacity) { u8 max_perf, min_perf, des_perf, cap_perf; --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -2252,14 +2252,17 @@ EXPORT_SYMBOL_GPL(cpufreq_driver_fast_sw * @policy: cpufreq policy object of the target CPU. * @min_perf: Minimum (required) performance level (units of @capacity). * @target_perf: Target (desired) performance level (units of @capacity). + * @max_perf: Maximum (allowed) performance level (units of @capacity). * @capacity: Capacity of the target CPU. * - * Carry out a fast performance level switch of @cpu without sleeping. + * Carry out a fast performance level adjustment for the CPU represeted by + * @policy without sleeping. * * The driver's ->adjust_perf() callback invoked by this function must be - * suitable for being called from within RCU-sched read-side critical sections - * and it is expected to select a suitable performance level equal to or above - * @min_perf and preferably equal to or below @target_perf. + * suitable for calling from within RCU-sched read-side critical sections and + * it is expected to program the proocessor to select suitable performance + * levels betweem @min_perf and @max_perf inclusive and preferably close to + * @target_perf going forward for the CPU represented by @policy. * * This function must not be called if policy->fast_switch_enabled is unset. * @@ -2271,9 +2274,10 @@ EXPORT_SYMBOL_GPL(cpufreq_driver_fast_sw void cpufreq_driver_adjust_perf(struct cpufreq_policy *policy, unsigned long min_perf, unsigned long target_perf, + unsigned long max_perf, unsigned long capacity) { - cpufreq_driver->adjust_perf(policy, min_perf, target_perf, capacity); + cpufreq_driver->adjust_perf(policy, min_perf, target_perf, max_perf, capacity); } /** --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -3241,6 +3241,7 @@ static unsigned int intel_cpufreq_fast_s static void intel_cpufreq_adjust_perf(struct cpufreq_policy *policy, unsigned long min_perf, unsigned long target_perf, + unsigned long max_perf, unsigned long capacity) { struct cpudata *cpu = all_cpu_data[policy->cpu]; @@ -3271,7 +3272,13 @@ static void intel_cpufreq_adjust_perf(st if (min_pstate > cpu->max_perf_ratio) min_pstate = cpu->max_perf_ratio; - max_pstate = min(cap_pstate, cpu->max_perf_ratio); + max_pstate = cap_pstate; + if (max_perf < capacity) + max_pstate = DIV_ROUND_UP(cap_pstate * max_perf, capacity); + + if (max_pstate > cpu->max_perf_ratio) + max_pstate = cpu->max_perf_ratio; + if (max_pstate < min_pstate) max_pstate = min_pstate; --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -379,6 +379,7 @@ struct cpufreq_driver { void (*adjust_perf)(struct cpufreq_policy *policy, unsigned long min_perf, unsigned long target_perf, + unsigned long max_perf, unsigned long capacity); /* @@ -624,6 +625,7 @@ unsigned int cpufreq_driver_fast_switch( void cpufreq_driver_adjust_perf(struct cpufreq_policy *policy, unsigned long min_perf, unsigned long target_perf, + unsigned long max_perf, unsigned long capacity); bool cpufreq_driver_has_adjust_perf(void); int cpufreq_driver_target(struct cpufreq_policy *policy, --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -50,6 +50,7 @@ struct sugov_cpu { unsigned long util; unsigned long bw_min; + unsigned long bw_max; /* The field below is for single-CPU policies only: */ #ifdef CONFIG_NO_HZ_COMMON @@ -232,6 +233,7 @@ static void sugov_get_util(struct sugov_ util = effective_cpu_util(sg_cpu->cpu, util, &min, &max); util = max(util, boost); sg_cpu->bw_min = min; + sg_cpu->bw_max = max; sg_cpu->util = sugov_effective_cpu_perf(sg_cpu->cpu, util, min, max); } @@ -484,7 +486,7 @@ static void sugov_update_single_perf(str sg_cpu->util = prev_util; cpufreq_driver_adjust_perf(sg_policy->policy, sg_cpu->bw_min, - sg_cpu->util, max_cap); + sg_cpu->util, sg_cpu->bw_max, max_cap); sg_policy->last_freq_update_time = time; }