[PATCH v6 1/4] scmi-cpufreq: Remove deferred probe
From: Nicola Mazzucato <hidden>
Date: 2021-01-11 15:44:16
Also in:
linux-arm-kernel, lkml
Subsystem:
cpu frequency scaling framework, system control & power/management interface (scpi/scmi) message protocol drivers, the rest · Maintainers:
"Rafael J. Wysocki", Viresh Kumar, Sudeep Holla, Linus Torvalds
The current implementation still carries a case for a deferred probe, but in practise this should not happen anymore. Since the energy model expects to pass the number of OPPs, let us just move the call dev_pm_opp_get_opp_count closer to EM registration instead. Signed-off-by: Nicola Mazzucato <redacted> --- drivers/cpufreq/scmi-cpufreq.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c
index 491a0a24fb1e..15b213ed78fa 100644
--- a/drivers/cpufreq/scmi-cpufreq.c
+++ b/drivers/cpufreq/scmi-cpufreq.c@@ -153,13 +153,6 @@ static int scmi_cpufreq_init(struct cpufreq_policy *policy) return ret; } - nr_opp = dev_pm_opp_get_opp_count(cpu_dev); - if (nr_opp <= 0) { - dev_dbg(cpu_dev, "OPP table is not ready, deferring probe\n"); - ret = -EPROBE_DEFER; - goto out_free_opp; - } - priv = kzalloc(sizeof(*priv), GFP_KERNEL); if (!priv) { ret = -ENOMEM;
@@ -190,6 +183,15 @@ static int scmi_cpufreq_init(struct cpufreq_policy *policy) policy->fast_switch_possible = handle->perf_ops->fast_switch_possible(handle, cpu_dev); + nr_opp = dev_pm_opp_get_opp_count(cpu_dev); + if (nr_opp <= 0) { + dev_err(cpu_dev, "%s: No OPPs for this device: %d\n", + __func__, ret); + + ret = -ENODEV; + goto out_free_priv; + } + power_scale_mw = handle->perf_ops->power_scale_mw_get(handle); em_dev_register_perf_domain(cpu_dev, nr_opp, &em_cb, policy->cpus, power_scale_mw);
--
2.27.0