linux-kernel - [PATCH] cpufreq: intel_pstate: fix inconsistency in setting policy limits

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-next>] [day] [month] [year] [list]

Date:	Sun,  3 Apr 2016 13:06:12 -0700
From:	Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
To:	rjw@...ysocki.net
Cc:	linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org,
	Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
Subject: [PATCH] cpufreq: intel_pstate: fix inconsistency in setting policy limits

When user sets performance policy using cpufreq interface, it is possible
that because of policy->max limits, the actual performance is still
limited. But the current implementation will silently switch the
policy to powersave and start using powersave limits. If user modifies
any limits using intel_pstate sysfs, this is actually changing powersave
limits.

The current implementation tracks limits under powersave and performance
policy using two different variables. When policy->max is less than
policy->cpuinfo.max_freq, only powersave limit variable is used.

This fix involves uses performance limits variable always when policy
is performance.

Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>
---
 drivers/cpufreq/intel_pstate.c | 40 ++++++++++++++++++++++++++++++++--------
 1 file changed, 32 insertions(+), 8 deletions(-)

diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c
index 4b64452..776cea7 100644
--- a/drivers/cpufreq/intel_pstate.c
+++ b/drivers/cpufreq/intel_pstate.c
@@ -1122,22 +1122,46 @@ static unsigned int intel_pstate_get(unsigned int cpu_num)
 	return get_avg_frequency(cpu);
 }
 
+static void intel_pstate_set_performance_limits(struct perf_limits *limits)
+{
+	limits->no_turbo = 0;
+	limits->turbo_disabled = 0;
+	limits->max_perf_pct = 100;
+	limits->max_perf = int_tofp(1);
+	limits->min_perf_pct = 100;
+	limits->min_perf = int_tofp(1);
+	limits->max_policy_pct = 100;
+	limits->max_sysfs_pct = 100;
+	limits->min_policy_pct = 0;
+	limits->min_sysfs_pct = 0;
+}
+
 static int intel_pstate_set_policy(struct cpufreq_policy *policy)
 {
 	if (!policy->cpuinfo.max_freq)
 		return -ENODEV;
 
-	if (policy->policy == CPUFREQ_POLICY_PERFORMANCE &&
-	    policy->max >= policy->cpuinfo.max_freq) {
-		pr_debug("intel_pstate: set performance\n");
+	if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) {
 		limits = &performance_limits;
-		if (hwp_active)
-			intel_pstate_hwp_set(policy->cpus);
-		return 0;
+		/*
+		 * policy->cpuinfo.max_freq is the max frequency supported,
+		 * which is set during cpufreq init() callback.
+		 * policy->max is the current max frequency, which can less
+		 * than policy->cpuinfo.max_freq, because of limits placed
+		 * by cpufreq thermal interface.
+		 */
+		if (policy->max >= policy->cpuinfo.max_freq) {
+			pr_debug("intel_pstate: set performance\n");
+			intel_pstate_set_performance_limits(limits);
+			if (hwp_active)
+				intel_pstate_hwp_set(policy->cpus);
+			return 0;
+		}
+	} else {
+		pr_debug("intel_pstate: set powersave\n");
+		limits = &powersave_limits;
 	}
 
-	pr_debug("intel_pstate: set powersave\n");
-	limits = &powersave_limits;
 	limits->min_policy_pct = (policy->min * 100) / policy->cpuinfo.max_freq;
 	limits->min_policy_pct = clamp_t(int, limits->min_policy_pct, 0 , 100);
 	limits->max_policy_pct = DIV_ROUND_UP(policy->max * 100,
-- 
2.5.0