Message ID | 40b676a1e261d1cfd12788d88e0d2026dbbe226b.1599817789.git.viresh.kumar@linaro.org |
---|---|
State | Accepted |
Commit | a5663c9b1e31c00e0bdfaf4d92eb51358fc3950f |
Headers | show |
Series | opp: Allow opp-level to be set to 0 | expand |
On Fri, Sep 11, 2020 at 03:19:53PM +0530, Viresh Kumar wrote: > The DT bindings don't put such a constraint, nor should the kernel. It > is perfectly fine for opp-level to be set to 0, if we need to put the > performance state votes for a domain for a particular OPP. > > Reported-by: Stephan Gerhold <stephan@gerhold.net> > Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> I tried adding an opp-level = <0> and voting for it using required-opps for some of the OPPs in the CPU OPP table and it seems to work fine now. Tested-by: Stephan Gerhold <stephan@gerhold.net> Just need to solve the "enable power domains" problem now :) Thanks! Stephan > --- > drivers/opp/core.c | 3 --- > drivers/opp/of.c | 20 +++++++------------- > 2 files changed, 7 insertions(+), 16 deletions(-) > > diff --git a/drivers/opp/core.c b/drivers/opp/core.c > index 214c1619b445..2483e765318a 100644 > --- a/drivers/opp/core.c > +++ b/drivers/opp/core.c > @@ -2117,9 +2117,6 @@ int dev_pm_opp_xlate_performance_state(struct opp_table *src_table, > int dest_pstate = -EINVAL; > int i; > > - if (!pstate) > - return 0; > - > /* > * Normally the src_table will have the "required_opps" property set to > * point to one of the OPPs in the dst_table, but in some cases the > diff --git a/drivers/opp/of.c b/drivers/opp/of.c > index e72753be7dc7..87f839a1899a 100644 > --- a/drivers/opp/of.c > +++ b/drivers/opp/of.c > @@ -842,7 +842,7 @@ static struct dev_pm_opp *_opp_add_static_v2(struct opp_table *opp_table, > static int _of_add_opp_table_v2(struct device *dev, struct opp_table *opp_table) > { > struct device_node *np; > - int ret, count = 0, pstate_count = 0; > + int ret, count = 0; > struct dev_pm_opp *opp; > > /* OPP table is already initialized for the device */ > @@ -876,20 +876,14 @@ static int _of_add_opp_table_v2(struct device *dev, struct opp_table *opp_table) > goto remove_static_opp; > } > > - list_for_each_entry(opp, &opp_table->opp_list, node) > - pstate_count += !!opp->pstate; > - > - /* Either all or none of the nodes shall have performance state set */ > - if (pstate_count && pstate_count != count) { > - dev_err(dev, "Not all nodes have performance state set (%d: %d)\n", > - count, pstate_count); > - ret = -ENOENT; > - goto remove_static_opp; > + list_for_each_entry(opp, &opp_table->opp_list, node) { > + /* Any non-zero performance state would enable the feature */ > + if (opp->pstate) { > + opp_table->genpd_performance_state = true; > + break; > + } > } > > - if (pstate_count) > - opp_table->genpd_performance_state = true; > - > return 0; > > remove_static_opp: > -- > 2.25.0.rc1.19.g042ed3e048af >
diff --git a/drivers/opp/core.c b/drivers/opp/core.c index 214c1619b445..2483e765318a 100644 --- a/drivers/opp/core.c +++ b/drivers/opp/core.c @@ -2117,9 +2117,6 @@ int dev_pm_opp_xlate_performance_state(struct opp_table *src_table, int dest_pstate = -EINVAL; int i; - if (!pstate) - return 0; - /* * Normally the src_table will have the "required_opps" property set to * point to one of the OPPs in the dst_table, but in some cases the diff --git a/drivers/opp/of.c b/drivers/opp/of.c index e72753be7dc7..87f839a1899a 100644 --- a/drivers/opp/of.c +++ b/drivers/opp/of.c @@ -842,7 +842,7 @@ static struct dev_pm_opp *_opp_add_static_v2(struct opp_table *opp_table, static int _of_add_opp_table_v2(struct device *dev, struct opp_table *opp_table) { struct device_node *np; - int ret, count = 0, pstate_count = 0; + int ret, count = 0; struct dev_pm_opp *opp; /* OPP table is already initialized for the device */ @@ -876,20 +876,14 @@ static int _of_add_opp_table_v2(struct device *dev, struct opp_table *opp_table) goto remove_static_opp; } - list_for_each_entry(opp, &opp_table->opp_list, node) - pstate_count += !!opp->pstate; - - /* Either all or none of the nodes shall have performance state set */ - if (pstate_count && pstate_count != count) { - dev_err(dev, "Not all nodes have performance state set (%d: %d)\n", - count, pstate_count); - ret = -ENOENT; - goto remove_static_opp; + list_for_each_entry(opp, &opp_table->opp_list, node) { + /* Any non-zero performance state would enable the feature */ + if (opp->pstate) { + opp_table->genpd_performance_state = true; + break; + } } - if (pstate_count) - opp_table->genpd_performance_state = true; - return 0; remove_static_opp:
The DT bindings don't put such a constraint, nor should the kernel. It is perfectly fine for opp-level to be set to 0, if we need to put the performance state votes for a domain for a particular OPP. Reported-by: Stephan Gerhold <stephan@gerhold.net> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> --- drivers/opp/core.c | 3 --- drivers/opp/of.c | 20 +++++++------------- 2 files changed, 7 insertions(+), 16 deletions(-) -- 2.25.0.rc1.19.g042ed3e048af