WiFi modules often require 32kHz clock to function. Add support to
enable the clock to PCIe driver.

Co-developed-by: Ondrej Jirman <megi@xff.cz>
Signed-off-by: Ondrej Jirman <megi@xff.cz>
Signed-off-by: Jacobe Zang <jacobe.zang@wesion.com>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index e406e11481a62..4db188a41fd52 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -6,6 +6,7 @@
 #include <linux/of.h>
 #include <linux/of_irq.h>
 #include <linux/of_net.h>
+#include <linux/clk.h>
 
 #include <defs.h>
 #include "debug.h"
@@ -70,6 +71,7 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
 {
 	struct brcmfmac_sdio_pd *sdio = &settings->bus.sdio;
 	struct device_node *root, *np = dev->of_node;
+	struct clk *clk;
 	const char *prop;
 	int irq;
 	int err;
@@ -113,6 +115,13 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
 		of_node_put(root);
 	}
 
+	clk = devm_clk_get_optional_enabled(dev, "lpo");
+	if (IS_ERR(clk))
+	if (clk) {
+		brcmf_dbg(INFO, "enabling 32kHz clock\n");
+		clk_set_rate(clk, 32768);
+	}
+
 	if (!np || !of_device_is_compatible(np, "brcm,bcm4329-fmac"))
 		return;
 
-- 
2.34.1

Hello Jacobe,

Please see one comment below.

On 2024-07-29 09:01, Jacobe Zang wrote:
> WiFi modules often require 32kHz clock to function. Add support to
> enable the clock to PCIe driver.
> 
> Co-developed-by: Ondrej Jirman <megi@xff.cz>
> Signed-off-by: Ondrej Jirman <megi@xff.cz>
> Signed-off-by: Jacobe Zang <jacobe.zang@wesion.com>
> ---
>  drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> index e406e11481a62..4db188a41fd52 100644
> --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> @@ -6,6 +6,7 @@
>  #include <linux/of.h>
>  #include <linux/of_irq.h>
>  #include <linux/of_net.h>
> +#include <linux/clk.h>
> 
>  #include <defs.h>
>  #include "debug.h"
> @@ -70,6 +71,7 @@ void brcmf_of_probe(struct device *dev, enum
> brcmf_bus_type bus_type,
>  {
>  	struct brcmfmac_sdio_pd *sdio = &settings->bus.sdio;
>  	struct device_node *root, *np = dev->of_node;
> +	struct clk *clk;
>  	const char *prop;
>  	int irq;
>  	int err;
> @@ -113,6 +115,13 @@ void brcmf_of_probe(struct device *dev, enum
> brcmf_bus_type bus_type,
>  		of_node_put(root);
>  	}
> 
> +	clk = devm_clk_get_optional_enabled(dev, "lpo");
> +	if (IS_ERR(clk))
> +	if (clk) {

These two lines looks really confusing.  Shouldn't it be just a single
"if (!IS_ERR(clk)) {" line instead?

> +		brcmf_dbg(INFO, "enabling 32kHz clock\n");
> +		clk_set_rate(clk, 32768);
> +	}
> +
>  	if (!np || !of_device_is_compatible(np, "brcm,bcm4329-fmac"))
>  		return;

On Mon, Jul 29, 2024 at 09:12:20AM GMT, Dragan Simic wrote:
> Hello Jacobe,
> 
> [...]
>
> > 
> > +	clk = devm_clk_get_optional_enabled(dev, "lpo");
> > +	if (IS_ERR(clk))
> > +	if (clk) {
> 
> These two lines looks really confusing.  Shouldn't it be just a single
> "if (!IS_ERR(clk)) {" line instead?

It should be `!IS_ERR(clk) && clk` otherwise the debug message will be
incorrect.

Kind regards,
	o.

> > +		brcmf_dbg(INFO, "enabling 32kHz clock\n");
> > +		clk_set_rate(clk, 32768);
> > +	}
> > +
> >  	if (!np || !of_device_is_compatible(np, "brcm,bcm4329-fmac"))
> >  		return;

Hello Ondrej,

On 2024-07-29 10:44, Ondřej Jirman wrote:
> On Mon, Jul 29, 2024 at 09:12:20AM GMT, Dragan Simic wrote:
>> Hello Jacobe,
>> 
>> [...]
>> 
>> >
>> > +	clk = devm_clk_get_optional_enabled(dev, "lpo");
>> > +	if (IS_ERR(clk))
>> > +	if (clk) {
>> 
>> These two lines looks really confusing.  Shouldn't it be just a single
>> "if (!IS_ERR(clk)) {" line instead?
> 
> It should be `!IS_ERR(clk) && clk` otherwise the debug message will be
> incorrect.

Ah, I see now, thanks.  There's also IS_ERR_OR_NULL, so the condition
can actually be "!IS_ERR_OR_NULL(clk)".

>> > +		brcmf_dbg(INFO, "enabling 32kHz clock\n");
>> > +		clk_set_rate(clk, 32768);
>> > +	}
>> > +
>> >  	if (!np || !of_device_is_compatible(np, "brcm,bcm4329-fmac"))
>> >  		return;

On 7/29/2024 11:12 AM, Dragan Simic wrote:
> Hello Ondrej,
> 
> On 2024-07-29 10:44, Ondřej Jirman wrote:
>> On Mon, Jul 29, 2024 at 09:12:20AM GMT, Dragan Simic wrote:
>>> Hello Jacobe,
>>>
>>> [...]
>>>
>>> >
>>> > +    clk = devm_clk_get_optional_enabled(dev, "lpo");
>>> > +    if (IS_ERR(clk))
>>> > +    if (clk) {
>>>
>>> These two lines looks really confusing.  Shouldn't it be just a single
>>> "if (!IS_ERR(clk)) {" line instead?
>>
>> It should be `!IS_ERR(clk) && clk` otherwise the debug message will be
>> incorrect.
> 
> Ah, I see now, thanks.  There's also IS_ERR_OR_NULL, so the condition
> can actually be "!IS_ERR_OR_NULL(clk)".

++ best suggestion

>>> > +        brcmf_dbg(INFO, "enabling 32kHz clock\n");
>>> > +        clk_set_rate(clk, 32768);
>>> > +    }
>>> > +
>>> >      if (!np || !of_device_is_compatible(np, "brcm,bcm4329-fmac"))
>>> >          return;