[PATCH v2 0/2] clk: implement a new managed helper and add first user

[PATCH v2 0/2] clk: implement a new managed helper and add first user

[Bartosz Golaszewski]

I'm posting this as RFC to see if there's any interest. I noticed that
some drivers do: clk_get() -> clk_set_rate() -> clk_prepare_enable(). I
was wondering if it's worth factoring this out into dedicated helpers.

This series adds a new such helper for the "optional-enabled" use-case
and the first user. Let me know if this makes sense.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
---
Changes in v2:
- s/EXPORT_SYMBOL/EXPORT_SYMBOL_GPL/
- add a stub for !COMMON_CLK
- Link to v1: https://lore.kernel.org/r/20240801-clk-new-helper-v1-0-81e9338b7b17@linaro.org

---
Bartosz Golaszewski (2):
      clk: provide devm_clk_get_optional_enabled_with_rate()
      Bluetooth: hci_qca: use devm_clk_get_optional_enabled_with_rate()

 drivers/bluetooth/hci_qca.c | 24 ++----------------------
 drivers/clk/clk-devres.c    | 28 ++++++++++++++++++++++++++++
 include/linux/clk.h         | 33 +++++++++++++++++++++++++++++++++
 3 files changed, 63 insertions(+), 22 deletions(-)
---
base-commit: d6dbc9f56c3a70e915625b6f1887882c23dc5c91
change-id: 20240801-clk-new-helper-7853f662cda1

Best regards,
-- 
Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

[PATCH v2 1/2] clk: provide devm_clk_get_optional_enabled_with_rate()

[Bartosz Golaszewski]

From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

There are clock users in the kernel that can't use
devm_clk_get_optional_enabled() as they need to set rate after getting
the clock and before enabling it. Provide a managed helper that wraps
these operations in the correct order.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
---
 drivers/clk/clk-devres.c | 28 ++++++++++++++++++++++++++++
 include/linux/clk.h      | 33 +++++++++++++++++++++++++++++++++
 2 files changed, 61 insertions(+)

diff --git a/drivers/clk/clk-devres.c b/drivers/clk/clk-devres.c
index 90e6078fb6e1..82ae1f26e634 100644
--- a/drivers/clk/clk-devres.c
+++ b/drivers/clk/clk-devres.c
@@ -99,6 +99,34 @@ struct clk *devm_clk_get_optional_enabled(struct device *dev, const char *id)
 }
 EXPORT_SYMBOL_GPL(devm_clk_get_optional_enabled);
 
+struct clk *devm_clk_get_optional_enabled_with_rate(struct device *dev,
+						    const char *id,
+						    unsigned long rate)
+{
+	struct clk *clk;
+	int ret;
+
+	clk = __devm_clk_get(dev, id, clk_get_optional, NULL,
+			     clk_disable_unprepare);
+	if (IS_ERR(clk))
+		return ERR_CAST(clk);
+
+	ret = clk_set_rate(clk, rate);
+	if (ret)
+		goto out_put_clk;
+
+	ret = clk_prepare_enable(clk);
+	if (ret)
+		goto out_put_clk;
+
+	return clk;
+
+out_put_clk:
+	devm_clk_put(dev, clk);
+	return ERR_PTR(ret);
+}
+EXPORT_SYMBOL_GPL(devm_clk_get_optional_enabled_with_rate);
+
 struct clk_bulk_devres {
 	struct clk_bulk_data *clks;
 	int num_clks;
diff --git a/include/linux/clk.h b/include/linux/clk.h
index 0fa56d672532..851a0f2cf42c 100644
--- a/include/linux/clk.h
+++ b/include/linux/clk.h
@@ -640,6 +640,32 @@ struct clk *devm_clk_get_optional_prepared(struct device *dev, const char *id);
  */
 struct clk *devm_clk_get_optional_enabled(struct device *dev, const char *id);
 
+/**
+ * devm_clk_get_optional_enabled_with_rate - devm_clk_get_optional() +
+ *                                           clk_set_rate() +
+ *                                           clk_prepare_enable()
+ * @dev: device for clock "consumer"
+ * @id: clock consumer ID
+ * @rate: new clock rate
+ *
+ * Context: May sleep.
+ *
+ * Return: a struct clk corresponding to the clock producer, or
+ * valid IS_ERR() condition containing errno.  The implementation
+ * uses @dev and @id to determine the clock consumer, and thereby
+ * the clock producer.  If no such clk is found, it returns NULL
+ * which serves as a dummy clk.  That's the only difference compared
+ * to devm_clk_get_enabled().
+ *
+ * The returned clk (if valid) is prepared and enabled and rate was set.
+ *
+ * The clock will automatically be disabled, unprepared and freed
+ * when the device is unbound from the bus.
+ */
+struct clk *devm_clk_get_optional_enabled_with_rate(struct device *dev,
+						    const char *id,
+						    unsigned long rate);
+
 /**
  * devm_get_clk_from_child - lookup and obtain a managed reference to a
  *			     clock producer from child node.
@@ -982,6 +1008,13 @@ static inline struct clk *devm_clk_get_optional_enabled(struct device *dev,
 	return NULL;
 }
 
+static inline struct clk *
+devm_clk_get_optional_enabled_with_rate(struct device *dev, const char *id,
+					unsigned long rate)
+{
+	return NULL;
+}
+
 static inline int __must_check devm_clk_bulk_get(struct device *dev, int num_clks,
 						 struct clk_bulk_data *clks)
 {

-- 
2.43.0

Re: [PATCH v2 1/2] clk: provide devm_clk_get_optional_enabled_with_rate()

[Stephen Boyd]

Quoting Bartosz Golaszewski (2024-08-05 01:57:31)
> From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> 
> There are clock users in the kernel that can't use
> devm_clk_get_optional_enabled() as they need to set rate after getting
> the clock and before enabling it. Provide a managed helper that wraps
> these operations in the correct order.
> 
> Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> ---

Applied to clk-next

Re: [PATCH v2 1/2] clk: provide devm_clk_get_optional_enabled_with_rate()

[Bartosz Golaszewski]

On Thu, Sep 5, 2024 at 11:42 PM Stephen Boyd <sboyd@kernel.org> wrote:
>
> Quoting Bartosz Golaszewski (2024-08-05 01:57:31)
> > From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> >
> > There are clock users in the kernel that can't use
> > devm_clk_get_optional_enabled() as they need to set rate after getting
> > the clock and before enabling it. Provide a managed helper that wraps
> > these operations in the correct order.
> >
> > Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> > ---
>
> Applied to clk-next

Thanks, I will add some users in the next cycle. Unless Luiz really
wants to play with cross-tree merges again. :)

Bart

[PATCH v2 2/2] Bluetooth: hci_qca: use devm_clk_get_optional_enabled_with_rate()

[Bartosz Golaszewski]

From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

Use the new devm_clk_get_optional_enabled_with_rate() clock helper to
shrink the code a bit.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
---
 drivers/bluetooth/hci_qca.c | 24 ++----------------------
 1 file changed, 2 insertions(+), 22 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 45adc1560d94..7bee8102caf8 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -2293,13 +2293,6 @@ static int qca_init_regulators(struct qca_power *qca,
 	return 0;
 }
 
-static void qca_clk_disable_unprepare(void *data)
-{
-	struct clk *clk = data;
-
-	clk_disable_unprepare(clk);
-}
-
 static int qca_serdev_probe(struct serdev_device *serdev)
 {
 	struct qca_serdev *qcadev;
@@ -2432,25 +2425,12 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		if (!qcadev->bt_en)
 			power_ctrl_enabled = false;
 
-		qcadev->susclk = devm_clk_get_optional(&serdev->dev, NULL);
+		qcadev->susclk = devm_clk_get_optional_enabled_with_rate(
+					&serdev->dev, NULL, SUSCLK_RATE_32KHZ);
 		if (IS_ERR(qcadev->susclk)) {
 			dev_warn(&serdev->dev, "failed to acquire clk\n");
 			return PTR_ERR(qcadev->susclk);
 		}
-		err = clk_set_rate(qcadev->susclk, SUSCLK_RATE_32KHZ);
-		if (err)
-			return err;
-
-		err = clk_prepare_enable(qcadev->susclk);
-		if (err)
-			return err;
-
-		err = devm_add_action_or_reset(&serdev->dev,
-					       qca_clk_disable_unprepare,
-					       qcadev->susclk);
-		if (err)
-			return err;
-
 	}
 	
 	err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);

-- 
2.43.0

Re: [PATCH v2 0/2] clk: implement a new managed helper and add first user

[Bartosz Golaszewski]

On Mon, Aug 5, 2024 at 10:57 AM Bartosz Golaszewski <brgl@bgdev.pl> wrote:
>
> I'm posting this as RFC to see if there's any interest. I noticed that
> some drivers do: clk_get() -> clk_set_rate() -> clk_prepare_enable(). I
> was wondering if it's worth factoring this out into dedicated helpers.
>
> This series adds a new such helper for the "optional-enabled" use-case
> and the first user. Let me know if this makes sense.
>
> Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> ---
> Changes in v2:
> - s/EXPORT_SYMBOL/EXPORT_SYMBOL_GPL/
> - add a stub for !COMMON_CLK
> - Link to v1: https://lore.kernel.org/r/20240801-clk-new-helper-v1-0-81e9338b7b17@linaro.org
>
> ---
> Bartosz Golaszewski (2):
>       clk: provide devm_clk_get_optional_enabled_with_rate()
>       Bluetooth: hci_qca: use devm_clk_get_optional_enabled_with_rate()
>
>  drivers/bluetooth/hci_qca.c | 24 ++----------------------
>  drivers/clk/clk-devres.c    | 28 ++++++++++++++++++++++++++++
>  include/linux/clk.h         | 33 +++++++++++++++++++++++++++++++++
>  3 files changed, 63 insertions(+), 22 deletions(-)
> ---
> base-commit: d6dbc9f56c3a70e915625b6f1887882c23dc5c91
> change-id: 20240801-clk-new-helper-7853f662cda1
>
> Best regards,
> --
> Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
>

It's been a month. Any comments on this?

Bart