Skip to contentSkip to conversation

clk: support hardware-specific debugfs entries

PATCH In review Alex Elder 2 participants 3 messages Dec 26, 2013 Common clk

lore
Share & tools
lore

Alex Elder

author
#0 lore raw

PATCH 1/1 clk: support hardware-specific debugfs entries

Add a new clk_ops->debug_init method to allow a clock hardware driver to populate the clock's debugfs directory with entries beyond those common for every clock.

---
 drivers/clk/clk.c            |    4 ++++
 include/linux/clk-provider.h |   10 ++++++++++
 2 files changed, 14 insertions(+)
drivers/clk/clk.c +0 -0
index 2cf2ea6..c82a1bc 100644
@@ -268,6 +268,10 @@ static int clk_debug_create_one(struct clk *clk,

struct dentry *pdentry)
if (!d)
goto err_out;

+ if (clk->ops->debug_init)
+ if (clk->ops->debug_init(clk->hw, clk->dentry))
+ goto err_out;
+
ret = 0;
goto out;

include/linux/clk-provider.h +0 -0
index 7e59253..4812512 100644
@@ -31,6 +31,7 @@
 #define CLK_SET_RATE_NO_REPARENT BIT(7) /* don't re-parent on rate

change */

struct clk_hw;
+struct dentry;

/**
* struct clk_ops - Callback operations for hardware clocks; these are to
@@ -108,6 +109,12 @@ struct clk_hw;
* which is likely helpful for most .set_rate implementation.
* Returns 0 on success, -EERROR otherwise.
*
+ * @debug_init: Set up type-specific debugfs entries for this clock. This
+ * is called once, after the debugfs directory entry for this
+ * clock has been created. The dentry pointer representing that
+ * directory is provided as an argument. Called with
+ * prepare_lock held. Returns 0 on success, -EERROR otherwise.
+ *
* The clk_enable/clk_disable and clk_prepare/clk_unprepare pairs allow
* implementations to split any work between atomic (enable) and sleepable
* (prepare) contexts. If enabling a clock requires code that might sleep,
@@ -140,6 +147,9 @@ struct clk_ops {
int (*set_rate)(struct clk_hw *hw, unsigned long,
unsigned long);
void (*init)(struct clk_hw *hw);
+#ifdef CONFIG_COMMON_CLK_DEBUG
+ int (*debug_init)(struct clk_hw *hw, struct dentry *dentry);
+#endif
};

/**

Signature · 1 line
1.7.9.5
#1 lore raw

Review struct clk_ops { · line 153

Quoted 9 lines · Alex Elder
(quoted diff) +3 -0
@@ -140,6 +147,9 @@ struct clk_ops {
 	int		(*set_rate)(struct clk_hw *hw, unsigned long,
 				    unsigned long);
 	void		(*init)(struct clk_hw *hw);
+#ifdef CONFIG_COMMON_CLK_DEBUG
+	int		(*debug_init)(struct clk_hw *hw, struct dentry *dentry);
+#endif
 };

Please note that this CONFIG_COMMON_CLK_DEBUG switch may go away (or get replaced) soon. Mike plans to effectively enable common clock debugging as soon as debugfs is in effect (which I welcome very much).

So you should monitor current development and adjust your patch, or be prepared for breakage and fixing afterwards. I'd suggest to just declare the .debug_init member unconditionally now. Since Mike's patch very likely gets accepted, and nothing will break when yours gets added later (regardless of the number of review iterations it will see).

virtually yours
Gerhard Sittig

Signature · 3 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80  Email: office at denx.de
author
#2 lore raw
Quoted 15 lines · Gerhard Sittig

Review struct clk_ops { · line 153

Quoted 9 lines · Alex Elder
(quoted diff) +3 -0
@@ -140,6 +147,9 @@ struct clk_ops {
 	int		(*set_rate)(struct clk_hw *hw, unsigned long,
 				    unsigned long);
 	void		(*init)(struct clk_hw *hw);
+#ifdef CONFIG_COMMON_CLK_DEBUG
+	int		(*debug_init)(struct clk_hw *hw, struct dentry *dentry);
+#endif
 };

Please note that this CONFIG_COMMON_CLK_DEBUG switch may go away (or get replaced) soon. Mike plans to effectively enable common clock debugging as soon as debugfs is in effect (which I welcome very much).

Fine with me. I'll wait for further feedback before sending out an updated patch. Thanks.

-Alex

Quoted 12 lines

So you should monitor current development and adjust your patch, or be prepared for breakage and fixing afterwards. I'd suggest to just declare the .debug_init member unconditionally now. Since Mike's patch very likely gets accepted, and nothing will break when yours gets added later (regardless of the number of review iterations it will see).

virtually yours
Gerhard Sittig