* [PATCH 0/2] clk: add assigned-clock-rates-u64
@ 2024-06-21 12:36 Peng Fan (OSS)
2024-06-21 12:36 ` [PATCH 1/2] of: property: add of_property_for_each_u64 Peng Fan (OSS)
2024-06-21 12:36 ` [PATCH 2/2] clk: clk-conf: support assigned-clock-rates-u64 Peng Fan (OSS)
0 siblings, 2 replies; 8+ messages in thread
From: Peng Fan (OSS) @ 2024-06-21 12:36 UTC (permalink / raw)
To: Rob Herring, Saravana Kannan, Michael Turquette, Stephen Boyd
Cc: devicetree, linux-kernel, linux-clk, Peng Fan
i.MX95 PLL VCO supports rates that exceeds UINT32_MAX, and the
i.MX95 System Controller Management Firmware(SCMI) server exports
PLL VCO for SCMI Agents to configure. So introduce
assigned-clock-rates-u64 to support rates that exceeds UINT32_MAX.
And introduce of_property_for_each_u64 to iterate each u64 rate.
The PR to add assigned-clock-rates-u64 to dt-schema:
https://github.com/devicetree-org/dt-schema/pull/140
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
Peng Fan (2):
of: property: add of_property_for_each_u64
clk: clk-conf: support assigned-clock-rates-u64
drivers/clk/clk-conf.c | 106 ++++++++++++++++++++++++++++++++++---------------
drivers/of/property.c | 23 +++++++++++
include/linux/of.h | 24 +++++++++++
3 files changed, 121 insertions(+), 32 deletions(-)
---
base-commit: 76db4c64526c5e8ba0f56ad3d890dce8f9b00bbc
change-id: 20240621-clk-u64-70c4333f0f80
Best regards,
--
Peng Fan <peng.fan@nxp.com>
^ permalink raw reply [flat|nested] 8+ messages in thread
* [PATCH 1/2] of: property: add of_property_for_each_u64
2024-06-21 12:36 [PATCH 0/2] clk: add assigned-clock-rates-u64 Peng Fan (OSS)
@ 2024-06-21 12:36 ` Peng Fan (OSS)
2024-06-27 21:43 ` Rob Herring
2024-06-21 12:36 ` [PATCH 2/2] clk: clk-conf: support assigned-clock-rates-u64 Peng Fan (OSS)
1 sibling, 1 reply; 8+ messages in thread
From: Peng Fan (OSS) @ 2024-06-21 12:36 UTC (permalink / raw)
To: Rob Herring, Saravana Kannan, Michael Turquette, Stephen Boyd
Cc: devicetree, linux-kernel, linux-clk, Peng Fan
From: Peng Fan <peng.fan@nxp.com>
Preparing for assigned-clock-rates-u64 support, add function
of_property_for_each_u64 to iterate each u64 value
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
drivers/of/property.c | 23 +++++++++++++++++++++++
include/linux/of.h | 24 ++++++++++++++++++++++++
2 files changed, 47 insertions(+)
diff --git a/drivers/of/property.c b/drivers/of/property.c
index 164d77cb9445..b89c3ab01d44 100644
--- a/drivers/of/property.c
+++ b/drivers/of/property.c
@@ -548,6 +548,29 @@ const __be32 *of_prop_next_u32(struct property *prop, const __be32 *cur,
}
EXPORT_SYMBOL_GPL(of_prop_next_u32);
+const __be32 *of_prop_next_u64(struct property *prop, const __be32 *cur,
+ u64 *pu)
+{
+ const void *curv = cur;
+
+ if (!prop)
+ return NULL;
+
+ if (!cur) {
+ curv = prop->value;
+ goto out_val;
+ }
+
+ curv += sizeof(*cur) * 2;
+ if (curv >= prop->value + prop->length)
+ return NULL;
+
+out_val:
+ *pu = of_read_number(curv, 2);
+ return curv;
+}
+EXPORT_SYMBOL_GPL(of_prop_next_u64);
+
const char *of_prop_next_string(struct property *prop, const char *cur)
{
const void *curv = cur;
diff --git a/include/linux/of.h b/include/linux/of.h
index 13cf7a43b473..464eca6a4636 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -439,6 +439,18 @@ extern int of_detach_node(struct device_node *);
*/
const __be32 *of_prop_next_u32(struct property *prop, const __be32 *cur,
u32 *pu);
+
+/*
+ * struct property *prop;
+ * const __be32 *p;
+ * u64 u;
+ *
+ * of_property_for_each_u64(np, "propname", prop, p, u)
+ * printk("U64 value: %llx\n", u);
+ */
+const __be32 *of_prop_next_u64(struct property *prop, const __be32 *cur,
+ u64 *pu);
+
/*
* struct property *prop;
* const char *s;
@@ -834,6 +846,12 @@ static inline const __be32 *of_prop_next_u32(struct property *prop,
return NULL;
}
+static inline const __be32 *of_prop_next_u64(struct property *prop,
+ const __be32 *cur, u64 *pu)
+{
+ return NULL;
+}
+
static inline const char *of_prop_next_string(struct property *prop,
const char *cur)
{
@@ -1437,6 +1455,12 @@ static inline int of_property_read_s32(const struct device_node *np,
p; \
p = of_prop_next_u32(prop, p, &u))
+#define of_property_for_each_u64(np, propname, prop, p, u) \
+ for (prop = of_find_property(np, propname, NULL), \
+ p = of_prop_next_u64(prop, NULL, &u); \
+ p; \
+ p = of_prop_next_u64(prop, p, &u))
+
#define of_property_for_each_string(np, propname, prop, s) \
for (prop = of_find_property(np, propname, NULL), \
s = of_prop_next_string(prop, NULL); \
--
2.37.1
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH 2/2] clk: clk-conf: support assigned-clock-rates-u64
2024-06-21 12:36 [PATCH 0/2] clk: add assigned-clock-rates-u64 Peng Fan (OSS)
2024-06-21 12:36 ` [PATCH 1/2] of: property: add of_property_for_each_u64 Peng Fan (OSS)
@ 2024-06-21 12:36 ` Peng Fan (OSS)
1 sibling, 0 replies; 8+ messages in thread
From: Peng Fan (OSS) @ 2024-06-21 12:36 UTC (permalink / raw)
To: Rob Herring, Saravana Kannan, Michael Turquette, Stephen Boyd
Cc: devicetree, linux-kernel, linux-clk, Peng Fan
From: Peng Fan <peng.fan@nxp.com>
i.MX95 System Management Control Firmware(SCMI) manages the clock
function, it exposes PLL VCO which could support up to 5GHz rate that
exceeds UINT32_MAX. So add assigned-clock-rates-u64 support
to set rate that exceeds UINT32_MAX.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
drivers/clk/clk-conf.c | 106 ++++++++++++++++++++++++++++++++++---------------
1 file changed, 74 insertions(+), 32 deletions(-)
diff --git a/drivers/clk/clk-conf.c b/drivers/clk/clk-conf.c
index 1a4e6340f95c..2bb00c859eb4 100644
--- a/drivers/clk/clk-conf.c
+++ b/drivers/clk/clk-conf.c
@@ -78,49 +78,91 @@ static int __set_clk_parents(struct device_node *node, bool clk_supplier)
return rc;
}
-static int __set_clk_rates(struct device_node *node, bool clk_supplier)
+static int __set_clk_rate(struct device_node *node, bool clk_supplier, int index,
+ unsigned long rate)
{
struct of_phandle_args clkspec;
+ struct clk *clk;
+ int rc;
+
+ rc = of_parse_phandle_with_args(node, "assigned-clocks",
+ "#clock-cells", index, &clkspec);
+ if (rc < 0)
+ return rc;
+
+ if (clkspec.np == node && !clk_supplier) {
+ of_node_put(clkspec.np);
+ return 1;
+ }
+
+ clk = of_clk_get_from_provider(&clkspec);
+ of_node_put(clkspec.np);
+ if (IS_ERR(clk)) {
+ if (PTR_ERR(clk) != -EPROBE_DEFER)
+ pr_warn("clk: couldn't get clock %d for %pOF\n",
+ index, node);
+ return PTR_ERR(clk);
+ }
+
+ rc = clk_set_rate(clk, rate);
+ if (rc < 0)
+ pr_err("clk: couldn't set %s clk rate to %lu (%d), current rate: %lu\n",
+ __clk_get_name(clk), rate, rc, clk_get_rate(clk));
+ clk_put(clk);
+
+ return 0;
+}
+
+static int __set_clk_rates(struct device_node *node, bool clk_supplier)
+{
struct property *prop;
const __be32 *cur;
int rc, index = 0;
- struct clk *clk;
- u32 rate;
+ u64 rate;
+ u32 rate_32;
+ bool is_rate_32 = false;
- of_property_for_each_u32(node, "assigned-clock-rates", prop, cur, rate) {
- if (rate) {
- rc = of_parse_phandle_with_args(node, "assigned-clocks",
- "#clock-cells", index, &clkspec);
- if (rc < 0) {
- /* skip empty (null) phandles */
- if (rc == -ENOENT)
- continue;
- else
- return rc;
- }
- if (clkspec.np == node && !clk_supplier) {
- of_node_put(clkspec.np);
- return 0;
- }
+ if (!of_find_property(node, "assigned-clock-rates-u64", NULL))
+ is_rate_32 = true;
- clk = of_clk_get_from_provider(&clkspec);
- of_node_put(clkspec.np);
- if (IS_ERR(clk)) {
- if (PTR_ERR(clk) != -EPROBE_DEFER)
- pr_warn("clk: couldn't get clock %d for %pOF\n",
- index, node);
- return PTR_ERR(clk);
+ if (is_rate_32) {
+ of_property_for_each_u32(node, "assigned-clock-rates", prop, cur, rate_32) {
+ if (rate_32) {
+ rc = __set_clk_rate(node, clk_supplier, index, rate_32);
+
+ if (rc == 1 && !clk_supplier)
+ return 0;
+
+ if (rc < 0) {
+ /* skip empty (null) phandles */
+ if (rc == -ENOENT)
+ continue;
+ else
+ return rc;
+ }
}
+ index++;
+ }
+ } else {
+ of_property_for_each_u64(node, "assigned-clock-rates-u64", prop, cur, rate) {
+ if (rate) {
+ rc = __set_clk_rate(node, clk_supplier, index, rate);
- rc = clk_set_rate(clk, rate);
- if (rc < 0)
- pr_err("clk: couldn't set %s clk rate to %u (%d), current rate: %lu\n",
- __clk_get_name(clk), rate, rc,
- clk_get_rate(clk));
- clk_put(clk);
+ if (rc == 1 && !clk_supplier)
+ return 0;
+
+ if (rc < 0) {
+ /* skip empty (null) phandles */
+ if (rc == -ENOENT)
+ continue;
+ else
+ return rc;
+ }
+ }
+ index++;
}
- index++;
}
+
return 0;
}
--
2.37.1
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH 1/2] of: property: add of_property_for_each_u64
2024-06-21 12:36 ` [PATCH 1/2] of: property: add of_property_for_each_u64 Peng Fan (OSS)
@ 2024-06-27 21:43 ` Rob Herring
2024-06-28 12:33 ` Peng Fan
0 siblings, 1 reply; 8+ messages in thread
From: Rob Herring @ 2024-06-27 21:43 UTC (permalink / raw)
To: Peng Fan (OSS), Luca Ceresoli
Cc: Saravana Kannan, Michael Turquette, Stephen Boyd, devicetree,
linux-kernel, linux-clk, Peng Fan
+Luca
On Fri, Jun 21, 2024 at 08:36:39PM +0800, Peng Fan (OSS) wrote:
> From: Peng Fan <peng.fan@nxp.com>
>
> Preparing for assigned-clock-rates-u64 support, add function
> of_property_for_each_u64 to iterate each u64 value
>
> Signed-off-by: Peng Fan <peng.fan@nxp.com>
> ---
> drivers/of/property.c | 23 +++++++++++++++++++++++
> include/linux/of.h | 24 ++++++++++++++++++++++++
> 2 files changed, 47 insertions(+)
>
> diff --git a/drivers/of/property.c b/drivers/of/property.c
> index 164d77cb9445..b89c3ab01d44 100644
> --- a/drivers/of/property.c
> +++ b/drivers/of/property.c
> @@ -548,6 +548,29 @@ const __be32 *of_prop_next_u32(struct property *prop, const __be32 *cur,
> }
> EXPORT_SYMBOL_GPL(of_prop_next_u32);
>
> +const __be32 *of_prop_next_u64(struct property *prop, const __be32 *cur,
> + u64 *pu)
struct property can be const
> +{
> + const void *curv = cur;
> +
> + if (!prop)
> + return NULL;
> +
> + if (!cur) {
> + curv = prop->value;
> + goto out_val;
> + }
> +
> + curv += sizeof(*cur) * 2;
> + if (curv >= prop->value + prop->length)
> + return NULL;
> +
> +out_val:
> + *pu = of_read_number(curv, 2);
> + return curv;
> +}
> +EXPORT_SYMBOL_GPL(of_prop_next_u64);
> +
> const char *of_prop_next_string(struct property *prop, const char *cur)
> {
> const void *curv = cur;
> diff --git a/include/linux/of.h b/include/linux/of.h
> index 13cf7a43b473..464eca6a4636 100644
> --- a/include/linux/of.h
> +++ b/include/linux/of.h
> @@ -439,6 +439,18 @@ extern int of_detach_node(struct device_node *);
> */
> const __be32 *of_prop_next_u32(struct property *prop, const __be32 *cur,
> u32 *pu);
> +
> +/*
> + * struct property *prop;
> + * const __be32 *p;
> + * u64 u;
> + *
> + * of_property_for_each_u64(np, "propname", prop, p, u)
> + * printk("U64 value: %llx\n", u);
> + */
> +const __be32 *of_prop_next_u64(struct property *prop, const __be32 *cur,
> + u64 *pu);
> +
> /*
> * struct property *prop;
> * const char *s;
> @@ -834,6 +846,12 @@ static inline const __be32 *of_prop_next_u32(struct property *prop,
> return NULL;
> }
>
> +static inline const __be32 *of_prop_next_u64(struct property *prop,
> + const __be32 *cur, u64 *pu)
> +{
> + return NULL;
> +}
> +
> static inline const char *of_prop_next_string(struct property *prop,
> const char *cur)
> {
> @@ -1437,6 +1455,12 @@ static inline int of_property_read_s32(const struct device_node *np,
> p; \
> p = of_prop_next_u32(prop, p, &u))
>
> +#define of_property_for_each_u64(np, propname, prop, p, u) \
> + for (prop = of_find_property(np, propname, NULL), \
> + p = of_prop_next_u64(prop, NULL, &u); \
> + p; \
> + p = of_prop_next_u64(prop, p, &u))
I think we want to define this differently to avoid exposing struct
property and the property data directly. Like this:
#define of_property_for_each_u64(np, propname, u) \
for (struct property *_prop = of_find_property(np, propname, NULL),
const __be32 *_p = of_prop_next_u64(_prop, NULL, &u);
_p;
_p = of_prop_next_u64(_prop, _p, &u))
See this discussion for context[1].
Rob
[1] https://lore.kernel.org/all/20240624232122.3cfe03f8@booty/
^ permalink raw reply [flat|nested] 8+ messages in thread
* RE: [PATCH 1/2] of: property: add of_property_for_each_u64
2024-06-27 21:43 ` Rob Herring
@ 2024-06-28 12:33 ` Peng Fan
2024-06-28 13:10 ` Peng Fan
0 siblings, 1 reply; 8+ messages in thread
From: Peng Fan @ 2024-06-28 12:33 UTC (permalink / raw)
To: Rob Herring, Peng Fan (OSS), Luca Ceresoli
Cc: Saravana Kannan, Michael Turquette, Stephen Boyd,
devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-clk@vger.kernel.org
> Subject: Re: [PATCH 1/2] of: property: add of_property_for_each_u64
>
> +Luca
>
> On Fri, Jun 21, 2024 at 08:36:39PM +0800, Peng Fan (OSS) wrote:
> > From: Peng Fan <peng.fan@nxp.com>
> >
> > Preparing for assigned-clock-rates-u64 support, add function
> > of_property_for_each_u64 to iterate each u64 value
> >
> > Signed-off-by: Peng Fan <peng.fan@nxp.com>
> > ---
> > drivers/of/property.c | 23 +++++++++++++++++++++++
> > include/linux/of.h | 24 ++++++++++++++++++++++++
> > 2 files changed, 47 insertions(+)
> >
> > diff --git a/drivers/of/property.c b/drivers/of/property.c index
> > 164d77cb9445..b89c3ab01d44 100644
> > --- a/drivers/of/property.c
> > +++ b/drivers/of/property.c
> > @@ -548,6 +548,29 @@ const __be32 *of_prop_next_u32(struct
> property
> > *prop, const __be32 *cur, }
> EXPORT_SYMBOL_GPL(of_prop_next_u32);
> >
> > +const __be32 *of_prop_next_u64(struct property *prop, const
> __be32 *cur,
> > + u64 *pu)
>
> struct property can be const
Fix in v2. BTW, I am thinking something as below:
const __be64 *of_prop_next_u64(const struct property *prop, const __be64 *cur,
u64 *pu)
{
const void *curv = cur;
if (!prop)
return NULL;
if (!cur) {
curv = prop->value;
goto out_val;
}
curv += sizeof(*cur);
if (curv >= prop->value + prop->length)
return NULL;
out_val:
*pu = be64_to_cpup(curv);
return curv;
}
EXPORT_SYMBOL_GPL(of_prop_next_u64);
>
> > +{
> > + const void *curv = cur;
> > +
> > + if (!prop)
> > + return NULL;
> > +
> > + if (!cur) {
> > + curv = prop->value;
> > + goto out_val;
> > + }
> > +
> > + curv += sizeof(*cur) * 2;
> > + if (curv >= prop->value + prop->length)
> > + return NULL;
> > +
> > +out_val:
> > + *pu = of_read_number(curv, 2);
> > + return curv;
> > +}
> > +EXPORT_SYMBOL_GPL(of_prop_next_u64);
> > +
> > const char *of_prop_next_string(struct property *prop, const char
> > *cur) {
> > const void *curv = cur;
> > diff --git a/include/linux/of.h b/include/linux/of.h index
> > 13cf7a43b473..464eca6a4636 100644
> > --- a/include/linux/of.h
> > +++ b/include/linux/of.h
> > @@ -439,6 +439,18 @@ extern int of_detach_node(struct
> device_node *);
> > */
> > const __be32 *of_prop_next_u32(struct property *prop, const
> __be32 *cur,
> > u32 *pu);
> > +
> > +/*
> > + * struct property *prop;
> > + * const __be32 *p;
> > + * u64 u;
> > + *
> > + * of_property_for_each_u64(np, "propname", prop, p, u)
> > + * printk("U64 value: %llx\n", u);
> > + */
> > +const __be32 *of_prop_next_u64(struct property *prop, const
> __be32 *cur,
> > + u64 *pu);
> > +
> > /*
> > * struct property *prop;
> > * const char *s;
> > @@ -834,6 +846,12 @@ static inline const __be32
> *of_prop_next_u32(struct property *prop,
> > return NULL;
> > }
> >
> > +static inline const __be32 *of_prop_next_u64(struct property *prop,
> > + const __be32 *cur, u64 *pu)
> > +{
> > + return NULL;
> > +}
> > +
> > static inline const char *of_prop_next_string(struct property *prop,
> > const char *cur)
> > {
> > @@ -1437,6 +1455,12 @@ static inline int
> of_property_read_s32(const struct device_node *np,
> > p; \
> > p = of_prop_next_u32(prop, p, &u))
> >
> > +#define of_property_for_each_u64(np, propname, prop, p, u) \
> > + for (prop = of_find_property(np, propname, NULL), \
> > + p = of_prop_next_u64(prop, NULL, &u); \
> > + p; \
> > + p = of_prop_next_u64(prop, p, &u))
>
> I think we want to define this differently to avoid exposing struct
> property and the property data directly. Like this:
>
> #define of_property_for_each_u64(np, propname, u) \
> for (struct property *_prop = of_find_property(np, propname, NULL),
> const __be32 *_p = of_prop_next_u64(_prop, NULL, &u);
> _p;
> _p = of_prop_next_u64(_prop, _p, &u))
>
Sure, I will fix in v2.
Thanks,
Peng.
> See this discussion for context[1].
>
> Rob
>
> [1] https://lore.kernel.org/all/20240624232122.3cfe03f8@booty/
^ permalink raw reply [flat|nested] 8+ messages in thread
* RE: [PATCH 1/2] of: property: add of_property_for_each_u64
2024-06-28 12:33 ` Peng Fan
@ 2024-06-28 13:10 ` Peng Fan
2024-06-28 14:16 ` Luca Ceresoli
0 siblings, 1 reply; 8+ messages in thread
From: Peng Fan @ 2024-06-28 13:10 UTC (permalink / raw)
To: Rob Herring, Peng Fan (OSS), Luca Ceresoli
Cc: Saravana Kannan, Michael Turquette, Stephen Boyd,
devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-clk@vger.kernel.org
> Subject: RE: [PATCH 1/2] of: property: add of_property_for_each_u64
>
> > Subject: Re: [PATCH 1/2] of: property: add of_property_for_each_u64
> >
> > +Luca
> >
> > On Fri, Jun 21, 2024 at 08:36:39PM +0800, Peng Fan (OSS) wrote:
> > > From: Peng Fan <peng.fan@nxp.com>
> > >
> > > Preparing for assigned-clock-rates-u64 support, add function
> > > of_property_for_each_u64 to iterate each u64 value
> > >
> > > Signed-off-by: Peng Fan <peng.fan@nxp.com>
> > > ---
> > > drivers/of/property.c | 23 +++++++++++++++++++++++
> > > include/linux/of.h | 24 ++++++++++++++++++++++++
> > > 2 files changed, 47 insertions(+)
> > >
> > > diff --git a/drivers/of/property.c b/drivers/of/property.c index
> > > 164d77cb9445..b89c3ab01d44 100644
> > > --- a/drivers/of/property.c
> > > +++ b/drivers/of/property.c
> > > @@ -548,6 +548,29 @@ const __be32 *of_prop_next_u32(struct
> > property
> > > *prop, const __be32 *cur, }
> > EXPORT_SYMBOL_GPL(of_prop_next_u32);
> > >
> > > +const __be32 *of_prop_next_u64(struct property *prop, const
> > __be32 *cur,
> > > + u64 *pu)
> >
> > struct property can be const
>
> Fix in v2. BTW, I am thinking something as below:
>
> const __be64 *of_prop_next_u64(const struct property *prop, const
> __be64 *cur,
> u64 *pu)
> {
> const void *curv = cur;
>
> if (!prop)
> return NULL;
>
> if (!cur) {
> curv = prop->value;
> goto out_val;
> }
>
> curv += sizeof(*cur);
> if (curv >= prop->value + prop->length)
> return NULL;
>
> out_val:
> *pu = be64_to_cpup(curv);
> return curv;
> }
> EXPORT_SYMBOL_GPL(of_prop_next_u64);
>
> >
> > > +{
> > > + const void *curv = cur;
> > > +
> > > + if (!prop)
> > > + return NULL;
> > > +
> > > + if (!cur) {
> > > + curv = prop->value;
> > > + goto out_val;
> > > + }
> > > +
> > > + curv += sizeof(*cur) * 2;
> > > + if (curv >= prop->value + prop->length)
> > > + return NULL;
> > > +
> > > +out_val:
> > > + *pu = of_read_number(curv, 2);
> > > + return curv;
> > > +}
> > > +EXPORT_SYMBOL_GPL(of_prop_next_u64);
> > > +
> > > const char *of_prop_next_string(struct property *prop, const char
> > > *cur) {
> > > const void *curv = cur;
> > > diff --git a/include/linux/of.h b/include/linux/of.h index
> > > 13cf7a43b473..464eca6a4636 100644
> > > --- a/include/linux/of.h
> > > +++ b/include/linux/of.h
> > > @@ -439,6 +439,18 @@ extern int of_detach_node(struct
> > device_node *);
> > > */
> > > const __be32 *of_prop_next_u32(struct property *prop, const
> > __be32 *cur,
> > > u32 *pu);
> > > +
> > > +/*
> > > + * struct property *prop;
> > > + * const __be32 *p;
> > > + * u64 u;
> > > + *
> > > + * of_property_for_each_u64(np, "propname", prop, p, u)
> > > + * printk("U64 value: %llx\n", u);
> > > + */
> > > +const __be32 *of_prop_next_u64(struct property *prop, const
> > __be32 *cur,
> > > + u64 *pu);
> > > +
> > > /*
> > > * struct property *prop;
> > > * const char *s;
> > > @@ -834,6 +846,12 @@ static inline const __be32
> > *of_prop_next_u32(struct property *prop,
> > > return NULL;
> > > }
> > >
> > > +static inline const __be32 *of_prop_next_u64(struct property
> *prop,
> > > + const __be32 *cur, u64 *pu)
> > > +{
> > > + return NULL;
> > > +}
> > > +
> > > static inline const char *of_prop_next_string(struct property *prop,
> > > const char *cur)
> > > {
> > > @@ -1437,6 +1455,12 @@ static inline int
> > of_property_read_s32(const struct device_node *np,
> > > p; \
> > > p = of_prop_next_u32(prop, p, &u))
> > >
> > > +#define of_property_for_each_u64(np, propname, prop, p, u) \
> > > + for (prop = of_find_property(np, propname, NULL), \
> > > + p = of_prop_next_u64(prop, NULL, &u); \
> > > + p; \
> > > + p = of_prop_next_u64(prop, p, &u))
> >
> > I think we want to define this differently to avoid exposing struct
> > property and the property data directly. Like this:
> >
> > #define of_property_for_each_u64(np, propname, u) \
> > for (struct property *_prop = of_find_property(np, propname, NULL),
> > const __be32 *_p = of_prop_next_u64(_prop, NULL, &u);
> > _p;
> > _p = of_prop_next_u64(_prop, _p, &u))
This will trigger a compilation error, because C not allow
declare two variables with different types as for loop expression 1.
Need to think about other methods.
Thanks,
Peng.
> >
>
> Sure, I will fix in v2.
>
> Thanks,
> Peng.
>
> > See this discussion for context[1].
> >
> > Rob
> >
> > [1] https://lore.kernel.org/all/20240624232122.3cfe03f8@booty/
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/2] of: property: add of_property_for_each_u64
2024-06-28 13:10 ` Peng Fan
@ 2024-06-28 14:16 ` Luca Ceresoli
2024-07-03 10:42 ` Luca Ceresoli
0 siblings, 1 reply; 8+ messages in thread
From: Luca Ceresoli @ 2024-06-28 14:16 UTC (permalink / raw)
To: Peng Fan
Cc: Rob Herring, Peng Fan (OSS), Saravana Kannan, Michael Turquette,
Stephen Boyd, devicetree@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-clk@vger.kernel.org
Hello Peng,
On Fri, 28 Jun 2024 13:10:34 +0000
Peng Fan <peng.fan@nxp.com> wrote:
> > Subject: RE: [PATCH 1/2] of: property: add of_property_for_each_u64
> >
> > > Subject: Re: [PATCH 1/2] of: property: add of_property_for_each_u64
> > >
> > > +Luca
> > >
> > > On Fri, Jun 21, 2024 at 08:36:39PM +0800, Peng Fan (OSS) wrote:
> > > > From: Peng Fan <peng.fan@nxp.com>
> > > >
> > > > Preparing for assigned-clock-rates-u64 support, add function
> > > > of_property_for_each_u64 to iterate each u64 value
> > > >
> > > > Signed-off-by: Peng Fan <peng.fan@nxp.com>
> > > > ---
> > > > drivers/of/property.c | 23 +++++++++++++++++++++++
> > > > include/linux/of.h | 24 ++++++++++++++++++++++++
> > > > 2 files changed, 47 insertions(+)
> > > >
> > > > diff --git a/drivers/of/property.c b/drivers/of/property.c index
> > > > 164d77cb9445..b89c3ab01d44 100644
> > > > --- a/drivers/of/property.c
> > > > +++ b/drivers/of/property.c
> > > > @@ -548,6 +548,29 @@ const __be32 *of_prop_next_u32(struct
> > > property
> > > > *prop, const __be32 *cur, }
> > > EXPORT_SYMBOL_GPL(of_prop_next_u32);
> > > >
> > > > +const __be32 *of_prop_next_u64(struct property *prop, const
> > > __be32 *cur,
> > > > + u64 *pu)
> > >
> > > struct property can be const
> >
> > Fix in v2. BTW, I am thinking something as below:
> >
> > const __be64 *of_prop_next_u64(const struct property *prop, const
> > __be64 *cur,
> > u64 *pu)
> > {
> > const void *curv = cur;
> >
> > if (!prop)
> > return NULL;
> >
> > if (!cur) {
> > curv = prop->value;
> > goto out_val;
> > }
> >
> > curv += sizeof(*cur);
> > if (curv >= prop->value + prop->length)
> > return NULL;
> >
> > out_val:
> > *pu = be64_to_cpup(curv);
> > return curv;
> > }
> > EXPORT_SYMBOL_GPL(of_prop_next_u64);
> >
> > >
> > > > +{
> > > > + const void *curv = cur;
> > > > +
> > > > + if (!prop)
> > > > + return NULL;
> > > > +
> > > > + if (!cur) {
> > > > + curv = prop->value;
> > > > + goto out_val;
> > > > + }
> > > > +
> > > > + curv += sizeof(*cur) * 2;
> > > > + if (curv >= prop->value + prop->length)
> > > > + return NULL;
> > > > +
> > > > +out_val:
> > > > + *pu = of_read_number(curv, 2);
> > > > + return curv;
> > > > +}
> > > > +EXPORT_SYMBOL_GPL(of_prop_next_u64);
> > > > +
> > > > const char *of_prop_next_string(struct property *prop, const char
> > > > *cur) {
> > > > const void *curv = cur;
> > > > diff --git a/include/linux/of.h b/include/linux/of.h index
> > > > 13cf7a43b473..464eca6a4636 100644
> > > > --- a/include/linux/of.h
> > > > +++ b/include/linux/of.h
> > > > @@ -439,6 +439,18 @@ extern int of_detach_node(struct
> > > device_node *);
> > > > */
> > > > const __be32 *of_prop_next_u32(struct property *prop, const
> > > __be32 *cur,
> > > > u32 *pu);
> > > > +
> > > > +/*
> > > > + * struct property *prop;
> > > > + * const __be32 *p;
> > > > + * u64 u;
> > > > + *
> > > > + * of_property_for_each_u64(np, "propname", prop, p, u)
> > > > + * printk("U64 value: %llx\n", u);
> > > > + */
> > > > +const __be32 *of_prop_next_u64(struct property *prop, const
> > > __be32 *cur,
> > > > + u64 *pu);
> > > > +
> > > > /*
> > > > * struct property *prop;
> > > > * const char *s;
> > > > @@ -834,6 +846,12 @@ static inline const __be32
> > > *of_prop_next_u32(struct property *prop,
> > > > return NULL;
> > > > }
> > > >
> > > > +static inline const __be32 *of_prop_next_u64(struct property
> > *prop,
> > > > + const __be32 *cur, u64 *pu)
> > > > +{
> > > > + return NULL;
> > > > +}
> > > > +
> > > > static inline const char *of_prop_next_string(struct property *prop,
> > > > const char *cur)
> > > > {
> > > > @@ -1437,6 +1455,12 @@ static inline int
> > > of_property_read_s32(const struct device_node *np,
> > > > p; \
> > > > p = of_prop_next_u32(prop, p, &u))
> > > >
> > > > +#define of_property_for_each_u64(np, propname, prop, p, u) \
> > > > + for (prop = of_find_property(np, propname, NULL), \
> > > > + p = of_prop_next_u64(prop, NULL, &u); \
> > > > + p; \
> > > > + p = of_prop_next_u64(prop, p, &u))
> > >
> > > I think we want to define this differently to avoid exposing struct
> > > property and the property data directly. Like this:
> > >
> > > #define of_property_for_each_u64(np, propname, u) \
> > > for (struct property *_prop = of_find_property(np, propname, NULL),
> > > const __be32 *_p = of_prop_next_u64(_prop, NULL, &u);
> > > _p;
> > > _p = of_prop_next_u64(_prop, _p, &u))
>
> This will trigger a compilation error, because C not allow
> declare two variables with different types as for loop expression 1.
> Need to think about other methods.
I have a working draft here where I solved it somehow, let me just find
the proper branch and send it. Perhaps next week, but I'm striving to do
that by Mon-Tue.
Luca
--
Luca Ceresoli, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [PATCH 1/2] of: property: add of_property_for_each_u64
2024-06-28 14:16 ` Luca Ceresoli
@ 2024-07-03 10:42 ` Luca Ceresoli
0 siblings, 0 replies; 8+ messages in thread
From: Luca Ceresoli @ 2024-07-03 10:42 UTC (permalink / raw)
To: Peng Fan
Cc: Rob Herring, Peng Fan (OSS), Saravana Kannan, Michael Turquette,
Stephen Boyd, devicetree@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-clk@vger.kernel.org
Hello Peng,
On Fri, 28 Jun 2024 16:16:17 +0200
Luca Ceresoli <luca.ceresoli@bootlin.com> wrote:
[...]
> > > > > +#define of_property_for_each_u64(np, propname, prop, p, u) \
> > > > > + for (prop = of_find_property(np, propname, NULL), \
> > > > > + p = of_prop_next_u64(prop, NULL, &u); \
> > > > > + p; \
> > > > > + p = of_prop_next_u64(prop, p, &u))
> > > >
> > > > I think we want to define this differently to avoid exposing struct
> > > > property and the property data directly. Like this:
> > > >
> > > > #define of_property_for_each_u64(np, propname, u) \
> > > > for (struct property *_prop = of_find_property(np, propname, NULL),
> > > > const __be32 *_p = of_prop_next_u64(_prop, NULL, &u);
> > > > _p;
> > > > _p = of_prop_next_u64(_prop, _p, &u))
> >
> > This will trigger a compilation error, because C not allow
> > declare two variables with different types as for loop expression 1.
> > Need to think about other methods.
>
> I have a working draft here where I solved it somehow, let me just find
> the proper branch and send it. Perhaps next week, but I'm striving to do
> that by Mon-Tue.
Ok, that slipped to Wednesday, but here it is:
https://lore.kernel.org/all/20240703-of_property_for_each_u32-v1-1-42c1fc0b82aa@bootlin.com/
I think you can reuse the technique I used in that patch to write
of_property_for_each_u64(np, propname, u), taking only 3 parameters.
Luca
--
Luca Ceresoli, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2024-07-03 10:42 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-06-21 12:36 [PATCH 0/2] clk: add assigned-clock-rates-u64 Peng Fan (OSS)
2024-06-21 12:36 ` [PATCH 1/2] of: property: add of_property_for_each_u64 Peng Fan (OSS)
2024-06-27 21:43 ` Rob Herring
2024-06-28 12:33 ` Peng Fan
2024-06-28 13:10 ` Peng Fan
2024-06-28 14:16 ` Luca Ceresoli
2024-07-03 10:42 ` Luca Ceresoli
2024-06-21 12:36 ` [PATCH 2/2] clk: clk-conf: support assigned-clock-rates-u64 Peng Fan (OSS)
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).