* [chanwoo:devfreq-testing-passive-gov 14/15] drivers/devfreq/governor_passive.c:306 cpufreq_passive_register_notifier() warn: possible memory leak of 'cpu_data'
@ 2021-03-31 19:02 ` Dan Carpenter
0 siblings, 0 replies; 3+ messages in thread
From: Dan Carpenter @ 2021-03-31 19:02 UTC (permalink / raw)
To: kbuild
[-- Attachment #1: Type: text/plain, Size: 6887 bytes --]
tree: https://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux.git devfreq-testing-passive-gov
head: 801b60f356489d7a4525d95b615e00544ba34729
commit: 58c52e7878cc305830bc4c784ebdb6cba7a94c87 [14/15] PM / devfreq: Add cpu based scaling support to passive governor
config: x86_64-randconfig-m001-20210330 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
smatch warnings:
drivers/devfreq/governor_passive.c:306 cpufreq_passive_register_notifier() warn: possible memory leak of 'cpu_data'
vim +/cpu_data +306 drivers/devfreq/governor_passive.c
58c52e7878cc30 Saravana Kannan 2021-03-02 242 static int cpufreq_passive_register_notifier(struct devfreq *devfreq)
58c52e7878cc30 Saravana Kannan 2021-03-02 243 {
58c52e7878cc30 Saravana Kannan 2021-03-02 244 struct devfreq_passive_data *p_data
58c52e7878cc30 Saravana Kannan 2021-03-02 245 = (struct devfreq_passive_data *)devfreq->data;
58c52e7878cc30 Saravana Kannan 2021-03-02 246 struct device *dev = devfreq->dev.parent;
58c52e7878cc30 Saravana Kannan 2021-03-02 247 struct opp_table *opp_table = NULL;
58c52e7878cc30 Saravana Kannan 2021-03-02 248 struct devfreq_cpu_data *cpu_data;
58c52e7878cc30 Saravana Kannan 2021-03-02 249 struct cpufreq_policy *policy;
58c52e7878cc30 Saravana Kannan 2021-03-02 250 struct device *cpu_dev;
58c52e7878cc30 Saravana Kannan 2021-03-02 251 unsigned int cpu;
58c52e7878cc30 Saravana Kannan 2021-03-02 252 int ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 253
58c52e7878cc30 Saravana Kannan 2021-03-02 254 get_online_cpus();
58c52e7878cc30 Saravana Kannan 2021-03-02 255
58c52e7878cc30 Saravana Kannan 2021-03-02 256 p_data->nb.notifier_call = cpufreq_passive_notifier_call;
58c52e7878cc30 Saravana Kannan 2021-03-02 257 ret = cpufreq_register_notifier(&p_data->nb, CPUFREQ_TRANSITION_NOTIFIER);
58c52e7878cc30 Saravana Kannan 2021-03-02 258 if (ret) {
58c52e7878cc30 Saravana Kannan 2021-03-02 259 dev_err(dev, "failed to register cpufreq notifier\n");
58c52e7878cc30 Saravana Kannan 2021-03-02 260 p_data->nb.notifier_call = NULL;
58c52e7878cc30 Saravana Kannan 2021-03-02 261 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 262 }
58c52e7878cc30 Saravana Kannan 2021-03-02 263
58c52e7878cc30 Saravana Kannan 2021-03-02 264 for_each_online_cpu(cpu) {
58c52e7878cc30 Saravana Kannan 2021-03-02 265 if (p_data->cpu_data[cpu])
58c52e7878cc30 Saravana Kannan 2021-03-02 266 continue;
58c52e7878cc30 Saravana Kannan 2021-03-02 267
58c52e7878cc30 Saravana Kannan 2021-03-02 268 policy = cpufreq_cpu_get(cpu);
58c52e7878cc30 Saravana Kannan 2021-03-02 269 if (policy) {
Better to flip this condition:
if (!policy) {
ret = -EPROBE_DEFER;
goto out;
}
The problem with goto out is that it doesn't unregister the notifier.
58c52e7878cc30 Saravana Kannan 2021-03-02 270 cpu_data = kzalloc(sizeof(*cpu_data), GFP_KERNEL);
58c52e7878cc30 Saravana Kannan 2021-03-02 271 if (!cpu_data) {
^^^^^^^^
Allocated here.
58c52e7878cc30 Saravana Kannan 2021-03-02 272 ret = -ENOMEM;
58c52e7878cc30 Saravana Kannan 2021-03-02 273 goto out;
The other problem is that these error paths don't call cpufreq_cpu_put(policy);
58c52e7878cc30 Saravana Kannan 2021-03-02 274 }
58c52e7878cc30 Saravana Kannan 2021-03-02 275
58c52e7878cc30 Saravana Kannan 2021-03-02 276 cpu_dev = get_cpu_device(cpu);
58c52e7878cc30 Saravana Kannan 2021-03-02 277 if (!cpu_dev) {
58c52e7878cc30 Saravana Kannan 2021-03-02 278 dev_err(dev, "failed to get cpu device\n");
58c52e7878cc30 Saravana Kannan 2021-03-02 279 ret = -ENODEV;
58c52e7878cc30 Saravana Kannan 2021-03-02 280 goto out;
^^^^^^^^
Leaked on this path. Move the allocation so it happens after
opp_table = dev_pm_opp_get_opp_table(cpu_dev); succeeds.
58c52e7878cc30 Saravana Kannan 2021-03-02 281 }
58c52e7878cc30 Saravana Kannan 2021-03-02 282
58c52e7878cc30 Saravana Kannan 2021-03-02 283 opp_table = dev_pm_opp_get_opp_table(cpu_dev);
58c52e7878cc30 Saravana Kannan 2021-03-02 284 if (IS_ERR(opp_table)) {
58c52e7878cc30 Saravana Kannan 2021-03-02 285 ret = PTR_ERR(opp_table);
58c52e7878cc30 Saravana Kannan 2021-03-02 286 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 287 }
58c52e7878cc30 Saravana Kannan 2021-03-02 288
58c52e7878cc30 Saravana Kannan 2021-03-02 289 cpu_data->dev = cpu_dev;
58c52e7878cc30 Saravana Kannan 2021-03-02 290 cpu_data->opp_table = opp_table;
58c52e7878cc30 Saravana Kannan 2021-03-02 291 cpu_data->first_cpu = cpumask_first(policy->related_cpus);
58c52e7878cc30 Saravana Kannan 2021-03-02 292 cpu_data->cur_freq = policy->cur;
58c52e7878cc30 Saravana Kannan 2021-03-02 293 cpu_data->min_freq = policy->cpuinfo.min_freq;
58c52e7878cc30 Saravana Kannan 2021-03-02 294 cpu_data->max_freq = policy->cpuinfo.max_freq;
58c52e7878cc30 Saravana Kannan 2021-03-02 295
58c52e7878cc30 Saravana Kannan 2021-03-02 296 p_data->cpu_data[cpu] = cpu_data;
58c52e7878cc30 Saravana Kannan 2021-03-02 297 cpufreq_cpu_put(policy);
58c52e7878cc30 Saravana Kannan 2021-03-02 298 } else {
58c52e7878cc30 Saravana Kannan 2021-03-02 299 ret = -EPROBE_DEFER;
58c52e7878cc30 Saravana Kannan 2021-03-02 300 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 301 }
58c52e7878cc30 Saravana Kannan 2021-03-02 302 }
58c52e7878cc30 Saravana Kannan 2021-03-02 303 out:
58c52e7878cc30 Saravana Kannan 2021-03-02 304 put_online_cpus();
58c52e7878cc30 Saravana Kannan 2021-03-02 305 if (ret)
58c52e7878cc30 Saravana Kannan 2021-03-02 @306 return ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 307
58c52e7878cc30 Saravana Kannan 2021-03-02 308 mutex_lock(&devfreq->lock);
58c52e7878cc30 Saravana Kannan 2021-03-02 309 ret = devfreq_update_target(devfreq, 0L);
58c52e7878cc30 Saravana Kannan 2021-03-02 310 mutex_unlock(&devfreq->lock);
58c52e7878cc30 Saravana Kannan 2021-03-02 311 if (ret)
58c52e7878cc30 Saravana Kannan 2021-03-02 312 dev_err(dev, "failed to update the frequency\n");
Do we need to unwind anything if devfreq_update_target() fails?
58c52e7878cc30 Saravana Kannan 2021-03-02 313
58c52e7878cc30 Saravana Kannan 2021-03-02 314 return ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 315 }
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 45343 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread* [chanwoo:devfreq-testing-passive-gov 14/15] drivers/devfreq/governor_passive.c:306 cpufreq_passive_register_notifier() warn: possible memory leak of 'cpu_data'
@ 2021-03-31 19:02 ` Dan Carpenter
0 siblings, 0 replies; 3+ messages in thread
From: Dan Carpenter @ 2021-03-31 19:02 UTC (permalink / raw)
To: kbuild-all
[-- Attachment #1: Type: text/plain, Size: 6887 bytes --]
tree: https://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux.git devfreq-testing-passive-gov
head: 801b60f356489d7a4525d95b615e00544ba34729
commit: 58c52e7878cc305830bc4c784ebdb6cba7a94c87 [14/15] PM / devfreq: Add cpu based scaling support to passive governor
config: x86_64-randconfig-m001-20210330 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
smatch warnings:
drivers/devfreq/governor_passive.c:306 cpufreq_passive_register_notifier() warn: possible memory leak of 'cpu_data'
vim +/cpu_data +306 drivers/devfreq/governor_passive.c
58c52e7878cc30 Saravana Kannan 2021-03-02 242 static int cpufreq_passive_register_notifier(struct devfreq *devfreq)
58c52e7878cc30 Saravana Kannan 2021-03-02 243 {
58c52e7878cc30 Saravana Kannan 2021-03-02 244 struct devfreq_passive_data *p_data
58c52e7878cc30 Saravana Kannan 2021-03-02 245 = (struct devfreq_passive_data *)devfreq->data;
58c52e7878cc30 Saravana Kannan 2021-03-02 246 struct device *dev = devfreq->dev.parent;
58c52e7878cc30 Saravana Kannan 2021-03-02 247 struct opp_table *opp_table = NULL;
58c52e7878cc30 Saravana Kannan 2021-03-02 248 struct devfreq_cpu_data *cpu_data;
58c52e7878cc30 Saravana Kannan 2021-03-02 249 struct cpufreq_policy *policy;
58c52e7878cc30 Saravana Kannan 2021-03-02 250 struct device *cpu_dev;
58c52e7878cc30 Saravana Kannan 2021-03-02 251 unsigned int cpu;
58c52e7878cc30 Saravana Kannan 2021-03-02 252 int ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 253
58c52e7878cc30 Saravana Kannan 2021-03-02 254 get_online_cpus();
58c52e7878cc30 Saravana Kannan 2021-03-02 255
58c52e7878cc30 Saravana Kannan 2021-03-02 256 p_data->nb.notifier_call = cpufreq_passive_notifier_call;
58c52e7878cc30 Saravana Kannan 2021-03-02 257 ret = cpufreq_register_notifier(&p_data->nb, CPUFREQ_TRANSITION_NOTIFIER);
58c52e7878cc30 Saravana Kannan 2021-03-02 258 if (ret) {
58c52e7878cc30 Saravana Kannan 2021-03-02 259 dev_err(dev, "failed to register cpufreq notifier\n");
58c52e7878cc30 Saravana Kannan 2021-03-02 260 p_data->nb.notifier_call = NULL;
58c52e7878cc30 Saravana Kannan 2021-03-02 261 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 262 }
58c52e7878cc30 Saravana Kannan 2021-03-02 263
58c52e7878cc30 Saravana Kannan 2021-03-02 264 for_each_online_cpu(cpu) {
58c52e7878cc30 Saravana Kannan 2021-03-02 265 if (p_data->cpu_data[cpu])
58c52e7878cc30 Saravana Kannan 2021-03-02 266 continue;
58c52e7878cc30 Saravana Kannan 2021-03-02 267
58c52e7878cc30 Saravana Kannan 2021-03-02 268 policy = cpufreq_cpu_get(cpu);
58c52e7878cc30 Saravana Kannan 2021-03-02 269 if (policy) {
Better to flip this condition:
if (!policy) {
ret = -EPROBE_DEFER;
goto out;
}
The problem with goto out is that it doesn't unregister the notifier.
58c52e7878cc30 Saravana Kannan 2021-03-02 270 cpu_data = kzalloc(sizeof(*cpu_data), GFP_KERNEL);
58c52e7878cc30 Saravana Kannan 2021-03-02 271 if (!cpu_data) {
^^^^^^^^
Allocated here.
58c52e7878cc30 Saravana Kannan 2021-03-02 272 ret = -ENOMEM;
58c52e7878cc30 Saravana Kannan 2021-03-02 273 goto out;
The other problem is that these error paths don't call cpufreq_cpu_put(policy);
58c52e7878cc30 Saravana Kannan 2021-03-02 274 }
58c52e7878cc30 Saravana Kannan 2021-03-02 275
58c52e7878cc30 Saravana Kannan 2021-03-02 276 cpu_dev = get_cpu_device(cpu);
58c52e7878cc30 Saravana Kannan 2021-03-02 277 if (!cpu_dev) {
58c52e7878cc30 Saravana Kannan 2021-03-02 278 dev_err(dev, "failed to get cpu device\n");
58c52e7878cc30 Saravana Kannan 2021-03-02 279 ret = -ENODEV;
58c52e7878cc30 Saravana Kannan 2021-03-02 280 goto out;
^^^^^^^^
Leaked on this path. Move the allocation so it happens after
opp_table = dev_pm_opp_get_opp_table(cpu_dev); succeeds.
58c52e7878cc30 Saravana Kannan 2021-03-02 281 }
58c52e7878cc30 Saravana Kannan 2021-03-02 282
58c52e7878cc30 Saravana Kannan 2021-03-02 283 opp_table = dev_pm_opp_get_opp_table(cpu_dev);
58c52e7878cc30 Saravana Kannan 2021-03-02 284 if (IS_ERR(opp_table)) {
58c52e7878cc30 Saravana Kannan 2021-03-02 285 ret = PTR_ERR(opp_table);
58c52e7878cc30 Saravana Kannan 2021-03-02 286 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 287 }
58c52e7878cc30 Saravana Kannan 2021-03-02 288
58c52e7878cc30 Saravana Kannan 2021-03-02 289 cpu_data->dev = cpu_dev;
58c52e7878cc30 Saravana Kannan 2021-03-02 290 cpu_data->opp_table = opp_table;
58c52e7878cc30 Saravana Kannan 2021-03-02 291 cpu_data->first_cpu = cpumask_first(policy->related_cpus);
58c52e7878cc30 Saravana Kannan 2021-03-02 292 cpu_data->cur_freq = policy->cur;
58c52e7878cc30 Saravana Kannan 2021-03-02 293 cpu_data->min_freq = policy->cpuinfo.min_freq;
58c52e7878cc30 Saravana Kannan 2021-03-02 294 cpu_data->max_freq = policy->cpuinfo.max_freq;
58c52e7878cc30 Saravana Kannan 2021-03-02 295
58c52e7878cc30 Saravana Kannan 2021-03-02 296 p_data->cpu_data[cpu] = cpu_data;
58c52e7878cc30 Saravana Kannan 2021-03-02 297 cpufreq_cpu_put(policy);
58c52e7878cc30 Saravana Kannan 2021-03-02 298 } else {
58c52e7878cc30 Saravana Kannan 2021-03-02 299 ret = -EPROBE_DEFER;
58c52e7878cc30 Saravana Kannan 2021-03-02 300 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 301 }
58c52e7878cc30 Saravana Kannan 2021-03-02 302 }
58c52e7878cc30 Saravana Kannan 2021-03-02 303 out:
58c52e7878cc30 Saravana Kannan 2021-03-02 304 put_online_cpus();
58c52e7878cc30 Saravana Kannan 2021-03-02 305 if (ret)
58c52e7878cc30 Saravana Kannan 2021-03-02 @306 return ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 307
58c52e7878cc30 Saravana Kannan 2021-03-02 308 mutex_lock(&devfreq->lock);
58c52e7878cc30 Saravana Kannan 2021-03-02 309 ret = devfreq_update_target(devfreq, 0L);
58c52e7878cc30 Saravana Kannan 2021-03-02 310 mutex_unlock(&devfreq->lock);
58c52e7878cc30 Saravana Kannan 2021-03-02 311 if (ret)
58c52e7878cc30 Saravana Kannan 2021-03-02 312 dev_err(dev, "failed to update the frequency\n");
Do we need to unwind anything if devfreq_update_target() fails?
58c52e7878cc30 Saravana Kannan 2021-03-02 313
58c52e7878cc30 Saravana Kannan 2021-03-02 314 return ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 315 }
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 45343 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
* [chanwoo:devfreq-testing-passive-gov 14/15] drivers/devfreq/governor_passive.c:306 cpufreq_passive_register_notifier() warn: possible memory leak of 'cpu_data'
@ 2021-03-31 17:14 kernel test robot
0 siblings, 0 replies; 3+ messages in thread
From: kernel test robot @ 2021-03-31 17:14 UTC (permalink / raw)
To: kbuild
[-- Attachment #1: Type: text/plain, Size: 6572 bytes --]
CC: kbuild-all(a)lists.01.org
TO: Saravana Kannan <skannan@codeaurora.org>
CC: Chanwoo Choi <cw00.choi@samsung.com>
CC: Sibi Sankar <sibis@codeaurora.org>
tree: https://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux.git devfreq-testing-passive-gov
head: 801b60f356489d7a4525d95b615e00544ba34729
commit: 58c52e7878cc305830bc4c784ebdb6cba7a94c87 [14/15] PM / devfreq: Add cpu based scaling support to passive governor
:::::: branch date: 9 hours ago
:::::: commit date: 9 hours ago
config: x86_64-randconfig-m001-20210330 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
smatch warnings:
drivers/devfreq/governor_passive.c:306 cpufreq_passive_register_notifier() warn: possible memory leak of 'cpu_data'
vim +/cpu_data +306 drivers/devfreq/governor_passive.c
58c52e7878cc30 Saravana Kannan 2021-03-02 241
58c52e7878cc30 Saravana Kannan 2021-03-02 242 static int cpufreq_passive_register_notifier(struct devfreq *devfreq)
58c52e7878cc30 Saravana Kannan 2021-03-02 243 {
58c52e7878cc30 Saravana Kannan 2021-03-02 244 struct devfreq_passive_data *p_data
58c52e7878cc30 Saravana Kannan 2021-03-02 245 = (struct devfreq_passive_data *)devfreq->data;
58c52e7878cc30 Saravana Kannan 2021-03-02 246 struct device *dev = devfreq->dev.parent;
58c52e7878cc30 Saravana Kannan 2021-03-02 247 struct opp_table *opp_table = NULL;
58c52e7878cc30 Saravana Kannan 2021-03-02 248 struct devfreq_cpu_data *cpu_data;
58c52e7878cc30 Saravana Kannan 2021-03-02 249 struct cpufreq_policy *policy;
58c52e7878cc30 Saravana Kannan 2021-03-02 250 struct device *cpu_dev;
58c52e7878cc30 Saravana Kannan 2021-03-02 251 unsigned int cpu;
58c52e7878cc30 Saravana Kannan 2021-03-02 252 int ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 253
58c52e7878cc30 Saravana Kannan 2021-03-02 254 get_online_cpus();
58c52e7878cc30 Saravana Kannan 2021-03-02 255
58c52e7878cc30 Saravana Kannan 2021-03-02 256 p_data->nb.notifier_call = cpufreq_passive_notifier_call;
58c52e7878cc30 Saravana Kannan 2021-03-02 257 ret = cpufreq_register_notifier(&p_data->nb, CPUFREQ_TRANSITION_NOTIFIER);
58c52e7878cc30 Saravana Kannan 2021-03-02 258 if (ret) {
58c52e7878cc30 Saravana Kannan 2021-03-02 259 dev_err(dev, "failed to register cpufreq notifier\n");
58c52e7878cc30 Saravana Kannan 2021-03-02 260 p_data->nb.notifier_call = NULL;
58c52e7878cc30 Saravana Kannan 2021-03-02 261 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 262 }
58c52e7878cc30 Saravana Kannan 2021-03-02 263
58c52e7878cc30 Saravana Kannan 2021-03-02 264 for_each_online_cpu(cpu) {
58c52e7878cc30 Saravana Kannan 2021-03-02 265 if (p_data->cpu_data[cpu])
58c52e7878cc30 Saravana Kannan 2021-03-02 266 continue;
58c52e7878cc30 Saravana Kannan 2021-03-02 267
58c52e7878cc30 Saravana Kannan 2021-03-02 268 policy = cpufreq_cpu_get(cpu);
58c52e7878cc30 Saravana Kannan 2021-03-02 269 if (policy) {
58c52e7878cc30 Saravana Kannan 2021-03-02 270 cpu_data = kzalloc(sizeof(*cpu_data), GFP_KERNEL);
58c52e7878cc30 Saravana Kannan 2021-03-02 271 if (!cpu_data) {
58c52e7878cc30 Saravana Kannan 2021-03-02 272 ret = -ENOMEM;
58c52e7878cc30 Saravana Kannan 2021-03-02 273 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 274 }
58c52e7878cc30 Saravana Kannan 2021-03-02 275
58c52e7878cc30 Saravana Kannan 2021-03-02 276 cpu_dev = get_cpu_device(cpu);
58c52e7878cc30 Saravana Kannan 2021-03-02 277 if (!cpu_dev) {
58c52e7878cc30 Saravana Kannan 2021-03-02 278 dev_err(dev, "failed to get cpu device\n");
58c52e7878cc30 Saravana Kannan 2021-03-02 279 ret = -ENODEV;
58c52e7878cc30 Saravana Kannan 2021-03-02 280 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 281 }
58c52e7878cc30 Saravana Kannan 2021-03-02 282
58c52e7878cc30 Saravana Kannan 2021-03-02 283 opp_table = dev_pm_opp_get_opp_table(cpu_dev);
58c52e7878cc30 Saravana Kannan 2021-03-02 284 if (IS_ERR(opp_table)) {
58c52e7878cc30 Saravana Kannan 2021-03-02 285 ret = PTR_ERR(opp_table);
58c52e7878cc30 Saravana Kannan 2021-03-02 286 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 287 }
58c52e7878cc30 Saravana Kannan 2021-03-02 288
58c52e7878cc30 Saravana Kannan 2021-03-02 289 cpu_data->dev = cpu_dev;
58c52e7878cc30 Saravana Kannan 2021-03-02 290 cpu_data->opp_table = opp_table;
58c52e7878cc30 Saravana Kannan 2021-03-02 291 cpu_data->first_cpu = cpumask_first(policy->related_cpus);
58c52e7878cc30 Saravana Kannan 2021-03-02 292 cpu_data->cur_freq = policy->cur;
58c52e7878cc30 Saravana Kannan 2021-03-02 293 cpu_data->min_freq = policy->cpuinfo.min_freq;
58c52e7878cc30 Saravana Kannan 2021-03-02 294 cpu_data->max_freq = policy->cpuinfo.max_freq;
58c52e7878cc30 Saravana Kannan 2021-03-02 295
58c52e7878cc30 Saravana Kannan 2021-03-02 296 p_data->cpu_data[cpu] = cpu_data;
58c52e7878cc30 Saravana Kannan 2021-03-02 297 cpufreq_cpu_put(policy);
58c52e7878cc30 Saravana Kannan 2021-03-02 298 } else {
58c52e7878cc30 Saravana Kannan 2021-03-02 299 ret = -EPROBE_DEFER;
58c52e7878cc30 Saravana Kannan 2021-03-02 300 goto out;
58c52e7878cc30 Saravana Kannan 2021-03-02 301 }
58c52e7878cc30 Saravana Kannan 2021-03-02 302 }
58c52e7878cc30 Saravana Kannan 2021-03-02 303 out:
58c52e7878cc30 Saravana Kannan 2021-03-02 304 put_online_cpus();
58c52e7878cc30 Saravana Kannan 2021-03-02 305 if (ret)
58c52e7878cc30 Saravana Kannan 2021-03-02 @306 return ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 307
58c52e7878cc30 Saravana Kannan 2021-03-02 308 mutex_lock(&devfreq->lock);
58c52e7878cc30 Saravana Kannan 2021-03-02 309 ret = devfreq_update_target(devfreq, 0L);
58c52e7878cc30 Saravana Kannan 2021-03-02 310 mutex_unlock(&devfreq->lock);
58c52e7878cc30 Saravana Kannan 2021-03-02 311 if (ret)
58c52e7878cc30 Saravana Kannan 2021-03-02 312 dev_err(dev, "failed to update the frequency\n");
58c52e7878cc30 Saravana Kannan 2021-03-02 313
58c52e7878cc30 Saravana Kannan 2021-03-02 314 return ret;
58c52e7878cc30 Saravana Kannan 2021-03-02 315 }
58c52e7878cc30 Saravana Kannan 2021-03-02 316
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 45343 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-03-31 19:02 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2021-03-31 19:02 [chanwoo:devfreq-testing-passive-gov 14/15] drivers/devfreq/governor_passive.c:306 cpufreq_passive_register_notifier() warn: possible memory leak of 'cpu_data' Dan Carpenter
2021-03-31 19:02 ` Dan Carpenter
-- strict thread matches above, loose matches on Subject: below --
2021-03-31 17:14 kernel test robot
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.