Skip to contentSkip to conversation

clk: provide prepare/unprepare functions

PATCH New Russell King - ARM Linux 2 participants 3 messages Sep 22, 2011 – Sep 27, 2011 Clk API

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 clk: provide prepare/unprepare functions

As discussed previously, there's the need on some platforms to run some parts of clk_enable() in contexts which can schedule. The solution which was agreed upon was to provide clk_prepare() and clk_unprepare() to contain this parts, while clk_enable() and clk_disable() perform the atomic part.

This patch provides a common definition for clk_prepare() and clk_unprepare() in linux/clk.h, and provides an upgrade path for existing implementation and drivers: drivers can start using clk_prepare() and clk_unprepare() once this patch is merged without having to wait for platform support. Platforms can then start to provide these additional functions.

Eventually, HAVE_CLK_PREPARE will be removed from the kernel, and everyone will have to provide these new APIs.

---
 include/linux/clk.h |   38 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
include/linux/clk.h +37 -0
index 1d37f42..c7eb17d 100644
@@ -11,6 +11,8 @@
 #ifndef __LINUX_CLK_H
 #define __LINUX_CLK_H
 
+#include <linux/kernel.h>
+
 struct device;
 
 /*
@@ -41,11 +43,29 @@ struct clk;
 struct clk *clk_get(struct device *dev, const char *id);
 
 /**
+ * clk_prepare - prepare a clock source
+ * @clk: clock source
+ *
+ * This prepares the clock source for use.
+ */
+#ifdef CONFIG_HAVE_CLK_PREPARE
+int clk_prepare(struct clk *clk);
+#else
+static inline int clk_prepare(struct clk *clk)
+{
+	might_sleep();
+	return 0;
+}
+#endif
+
+/**
  * clk_enable - inform the system when the clock source should be running.
  * @clk: clock source
  *
  * If the clock can not be enabled/disabled, this should return success.
  *
+ * May be called from atomic contexts.
+ *
  * Returns success (0) or negative errno.
  */
 int clk_enable(struct clk *clk);
@@ -57,6 +77,8 @@ int clk_enable(struct clk *clk);
  * Inform the system that a clock source is no longer required by
  * a driver and may be shut down.
  *
+ * May be called from atomic contexts.
+ *
  * Implementation detail: if the clock source is shared between
  * multiple drivers, clk_enable() calls must be balanced by the
  * same number of clk_disable() calls for the clock source to be
@@ -64,6 +86,21 @@ int clk_enable(struct clk *clk);
  */
 void clk_disable(struct clk *clk);
 
+/**
+ * clk_unprepare - undo preparation of a clock source
+ * @clk: clock source
+ *
+ * This prepares the clock source for use.
+ */
+#ifdef CONFIG_HAVE_CLK_PREPARE
+void clk_unprepare(struct clk *clk);
+#else
+static inline void clk_unprepare(struct clk *clk)
+{
+	might_sleep();
+}
+#endif
+
 /**
  * clk_get_rate - obtain the current clock rate (in Hz) for a clock source.
  *		  This is only valid once the clock source has been enabled.
#1 lore raw

PATCH 1/1 clk: provide prepare/unprepare functions

Quoted 13 lines · Russell King - ARM Linux

As discussed previously, there's the need on some platforms to run some parts of clk_enable() in contexts which can schedule. The solution which was agreed upon was to provide clk_prepare() and clk_unprepare() to contain this parts, while clk_enable() and clk_disable() perform the atomic part.

This patch provides a common definition for clk_prepare() and clk_unprepare() in linux/clk.h, and provides an upgrade path for existing implementation and drivers: drivers can start using clk_prepare() and clk_unprepare() once this patch is merged without having to wait for platform support. Platforms can then start to provide these additional functions.

Quoted 2 lines

Eventually, HAVE_CLK_PREPARE will be removed from the kernel, and everyone will have to provide these new APIs.

Glad to see this. Now the drivers and individual arch's can start moving to these new APIs before we settle on the common infrastructure. Once the common infrastructure is in, it should hopefully be trivial to move each arch to it.

Quoted 19 lines
---
  include/linux/clk.h |   38 ++++++++++++++++++++++++++++++++++++++
  1 file changed, 38 insertions(+)
include/linux/clk.h +2 -0
index 1d37f42..c7eb17d 100644
@@ -11,6 +11,8 @@
  #ifndef __LINUX_CLK_H
  #define __LINUX_CLK_H
 
+#include<linux/kernel.h>
+
  struct device;
 
  /*

include/linux/clk.h struct clk; · line 49

Quoted 7 lines
include/linux/clk.h +4 -0
@@ -41,11 +43,29 @@ struct clk;
  struct clk *clk_get(struct device *dev, const char *id);
 
  /**
+ * clk_prepare - prepare a clock source
+ * @clk: clock source
+ *
+ * This prepares the clock source for use.

Do you want to add a "May NOT be called from atomic context"? Yeah, the stub makes it obvious, but I think it would be consistent to add the comment in. No strong opinion though.

Not sure if you care, but LGTM/Ack.

Thanks,
Saravana

Signature · 2 lines
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
#2 lore raw

PATCH 1/1 clk: provide prepare/unprepare functions

Quoted 9 lines · Saravana Kannan
Quoted 5 lines

/**
+ * clk_prepare - prepare a clock source
+ * @clk: clock source
+ *
+ * This prepares the clock source for use.

Do you want to add a "May NOT be called from atomic context"? Yeah, the stub makes it obvious, but I think it would be consistent to add the comment in. No strong opinion though.

The might_sleep() is sufficient to enforce that, but adding the comment for when the helper goes away would be a good idea.

Quoted 1 line

Not sure if you care, but LGTM/Ack.

Thanks.