Message ID | 20210802204550.12647-1-lukasz.luba@arm.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | cpufreq: arm_scmi: Fix error path when allocation failed | expand |
On 02-08-21, 21:45, Lukasz Luba wrote: > Print warning and return an error which would stop the initialization > when cpumask allocation failed. > > Fixes: 80a064dbd556 ("scmi-cpufreq: Get opp_shared_cpus from opp-v2 for EM") > Signed-off-by: Lukasz Luba <lukasz.luba@arm.com> > --- > drivers/cpufreq/scmi-cpufreq.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c > index ec9a87ca2dbb..b159123e68fd 100644 > --- a/drivers/cpufreq/scmi-cpufreq.c > +++ b/drivers/cpufreq/scmi-cpufreq.c > @@ -133,8 +133,10 @@ static int scmi_cpufreq_init(struct cpufreq_policy *policy) > return -ENODEV; > } > > - if (!zalloc_cpumask_var(&opp_shared_cpus, GFP_KERNEL)) > - ret = -ENOMEM; > + if (!zalloc_cpumask_var(&opp_shared_cpus, GFP_KERNEL)) { > + dev_warn(cpu_dev, "failed to allocate cpumask\n"); We shouldn't be printing here anything I believe as the allocation core does it for us. That's why you won't see a print message anywhere for failed allocations. > + return -ENOMEM; > + } > > /* Obtain CPUs that share SCMI performance controls */ > ret = scmi_get_sharing_cpus(cpu_dev, policy->cpus); > -- > 2.17.1 >
Hi Viresh, On 8/3/21 5:34 AM, Viresh Kumar wrote: > On 02-08-21, 21:45, Lukasz Luba wrote: >> Print warning and return an error which would stop the initialization >> when cpumask allocation failed. >> >> Fixes: 80a064dbd556 ("scmi-cpufreq: Get opp_shared_cpus from opp-v2 for EM") >> Signed-off-by: Lukasz Luba <lukasz.luba@arm.com> >> --- >> drivers/cpufreq/scmi-cpufreq.c | 6 ++++-- >> 1 file changed, 4 insertions(+), 2 deletions(-) >> >> diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c >> index ec9a87ca2dbb..b159123e68fd 100644 >> --- a/drivers/cpufreq/scmi-cpufreq.c >> +++ b/drivers/cpufreq/scmi-cpufreq.c >> @@ -133,8 +133,10 @@ static int scmi_cpufreq_init(struct cpufreq_policy *policy) >> return -ENODEV; >> } >> >> - if (!zalloc_cpumask_var(&opp_shared_cpus, GFP_KERNEL)) >> - ret = -ENOMEM; >> + if (!zalloc_cpumask_var(&opp_shared_cpus, GFP_KERNEL)) { >> + dev_warn(cpu_dev, "failed to allocate cpumask\n"); > > We shouldn't be printing here anything I believe as the allocation > core does it for us. That's why you won't see a print message anywhere > for failed allocations. Thanks for the comment. Let me send the v2.
diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c index ec9a87ca2dbb..b159123e68fd 100644 --- a/drivers/cpufreq/scmi-cpufreq.c +++ b/drivers/cpufreq/scmi-cpufreq.c @@ -133,8 +133,10 @@ static int scmi_cpufreq_init(struct cpufreq_policy *policy) return -ENODEV; } - if (!zalloc_cpumask_var(&opp_shared_cpus, GFP_KERNEL)) - ret = -ENOMEM; + if (!zalloc_cpumask_var(&opp_shared_cpus, GFP_KERNEL)) { + dev_warn(cpu_dev, "failed to allocate cpumask\n"); + return -ENOMEM; + } /* Obtain CPUs that share SCMI performance controls */ ret = scmi_get_sharing_cpus(cpu_dev, policy->cpus);
Print warning and return an error which would stop the initialization when cpumask allocation failed. Fixes: 80a064dbd556 ("scmi-cpufreq: Get opp_shared_cpus from opp-v2 for EM") Signed-off-by: Lukasz Luba <lukasz.luba@arm.com> --- drivers/cpufreq/scmi-cpufreq.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)