linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jianglei Nie <niejianglei2021@163.com>
To: daniel.lezcano@kernel.org, rafael@kernel.org
Cc: linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org,
	Jianglei Nie <niejianglei@gmail.com>
Subject: [PATCH] powercap: DTPM: Fix reference leak in cpuhp_dtpm_cpu_offline()
Date: Thu, 25 Nov 2021 00:46:57 +0800	[thread overview]
Message-ID: <20211124164657.20519-1-niejianglei2021@163.com> (raw)

From: Jianglei Nie <niejianglei@gmail.com>

In line 153 (#1), cpufreq_cpu_get() increments the kobject reference
counter of the policy it returned on success. According to the
document, the policy returned by cpufreq_cpu_get() has to be
released with the help of cpufreq_cpu_put() to balance its kobject
reference counter properly. Forgetting the cpufreq_cpu_put()
operation will result in reference leak.

We can fix it by calling cpufreq_cpu_put() before the function
returns (#2, #3 and #4).

147 static int cpuhp_dtpm_cpu_offline(unsigned int cpu)
148 {
153     policy = cpufreq_cpu_get(cpu);
        // #1: reference increment

155     if (!policy)
156         return 0;

158     pd = em_cpu_get(cpu);
159     if (!pd)
160         return -EINVAL; // #2: missing reference decrement

166     if (cpumask_weight(policy->cpus) != 1)
167         return 0; // #3: missing reference decrement

174     return 0; // #4: missing reference decrement
175 }

Signed-off-by: Jianglei Nie <niejianglei@gmail.com>
---
 drivers/powercap/dtpm_cpu.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/powercap/dtpm_cpu.c b/drivers/powercap/dtpm_cpu.c
index 51c366938acd..182a07ee14b6 100644
--- a/drivers/powercap/dtpm_cpu.c
+++ b/drivers/powercap/dtpm_cpu.c
@@ -156,21 +156,26 @@ static int cpuhp_dtpm_cpu_offline(unsigned int cpu)
 		return 0;
 
 	pd = em_cpu_get(cpu);
-	if (!pd)
+	if (!pd) {
+		cpufreq_cpu_put(policy);
 		return -EINVAL;
+	}
 
 	dtpm = per_cpu(dtpm_per_cpu, cpu);
 
 	power_sub(dtpm, pd);
 
-	if (cpumask_weight(policy->cpus) != 1)
+	if (cpumask_weight(policy->cpus) != 1) {
+		cpufreq_cpu_put(policy);
 		return 0;
+	}
 
 	for_each_cpu(cpu, policy->related_cpus)
 		per_cpu(dtpm_per_cpu, cpu) = NULL;
 
 	dtpm_unregister(dtpm);
 
+	cpufreq_cpu_put(policy);
 	return 0;
 }
 
-- 
2.25.1


             reply	other threads:[~2021-11-24 17:02 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-24 16:46 Jianglei Nie [this message]
2021-12-10  7:28 [PATCH] powercap: DTPM: Fix reference leak in cpuhp_dtpm_cpu_offline() Jianglei Nie

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=20211124164657.20519-1-niejianglei2021@163.com \
    --to=niejianglei2021@163.com \
    --cc=daniel.lezcano@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=niejianglei@gmail.com \
    --cc=rafael@kernel.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;
as well as URLs for NNTP newsgroup(s).