1 The Common Clk Framework
2 Mike Turquette <mturquette@ti.com>
4 This document endeavours to explain the common clk framework details,
5 and how to port a platform over to this framework. It is not yet a
6 detailed explanation of the clock api in include/linux/clk.h, but
7 perhaps someday it will include that information.
9 Part 1 - introduction and interface split
11 The common clk framework is an interface to control the clock nodes
12 available on various devices today. This may come in the form of clock
13 gating, rate adjustment, muxing or other operations. This framework is
14 enabled with the CONFIG_COMMON_CLK option.
16 The interface itself is divided into two halves, each shielded from the
17 details of its counterpart. First is the common definition of struct
18 clk which unifies the framework-level accounting and infrastructure that
19 has traditionally been duplicated across a variety of platforms. Second
20 is a common implementation of the clk.h api, defined in
21 drivers/clk/clk.c. Finally there is struct clk_ops, whose operations
22 are invoked by the clk api implementation.
24 The second half of the interface is comprised of the hardware-specific
25 callbacks registered with struct clk_ops and the corresponding
26 hardware-specific structures needed to model a particular clock. For
27 the remainder of this document any reference to a callback in struct
28 clk_ops, such as .enable or .set_rate, implies the hardware-specific
29 implementation of that code. Likewise, references to struct clk_foo
30 serve as a convenient shorthand for the implementation of the
31 hardware-specific bits for the hypothetical "foo" hardware.
33 Tying the two halves of this interface together is struct clk_hw, which
34 is defined in struct clk_foo and pointed to within struct clk. This
35 allows for easy navigation between the two discrete halves of the common
38 Part 2 - common data structures and api
40 Below is the common struct clk definition from
41 include/linux/clk-private.h, modified for brevity:
45 const struct clk_ops *ops;
50 struct hlist_head children;
51 struct hlist_node child_node;
55 The members above make up the core of the clk tree topology. The clk
56 api itself defines several driver-facing functions which operate on
57 struct clk. That api is documented in include/linux/clk.h.
59 Platforms and devices utilizing the common struct clk use the struct
60 clk_ops pointer in struct clk to perform the hardware-specific parts of
61 the operations defined in clk.h:
64 int (*prepare)(struct clk_hw *hw);
65 void (*unprepare)(struct clk_hw *hw);
66 int (*enable)(struct clk_hw *hw);
67 void (*disable)(struct clk_hw *hw);
68 int (*is_enabled)(struct clk_hw *hw);
69 unsigned long (*recalc_rate)(struct clk_hw *hw,
70 unsigned long parent_rate);
71 long (*round_rate)(struct clk_hw *hw,
73 unsigned long *parent_rate);
74 long (*determine_rate)(struct clk_hw *hw,
76 unsigned long min_rate,
77 unsigned long max_rate,
78 unsigned long *best_parent_rate,
79 struct clk_hw **best_parent_clk);
80 int (*set_parent)(struct clk_hw *hw, u8 index);
81 u8 (*get_parent)(struct clk_hw *hw);
82 int (*set_rate)(struct clk_hw *hw,
84 unsigned long parent_rate);
85 int (*set_rate_and_parent)(struct clk_hw *hw,
87 unsigned long parent_rate,
89 unsigned long (*recalc_accuracy)(struct clk_hw *hw,
90 unsigned long parent_accuracy);
91 void (*init)(struct clk_hw *hw);
92 int (*debug_init)(struct clk_hw *hw,
93 struct dentry *dentry);
96 Part 3 - hardware clk implementations
98 The strength of the common struct clk comes from its .ops and .hw pointers
99 which abstract the details of struct clk from the hardware-specific bits, and
100 vice versa. To illustrate consider the simple gateable clk implementation in
101 drivers/clk/clk-gate.c:
110 struct clk_gate contains struct clk_hw hw as well as hardware-specific
111 knowledge about which register and bit controls this clk's gating.
112 Nothing about clock topology or accounting, such as enable_count or
113 notifier_count, is needed here. That is all handled by the common
114 framework code and struct clk.
116 Let's walk through enabling this clk from driver code:
119 clk = clk_get(NULL, "my_gateable_clk");
124 The call graph for clk_enable is very simple:
127 clk->ops->enable(clk->hw);
130 [resolves struct clk gate with to_clk_gate(hw)]
131 clk_gate_set_bit(gate);
133 And the definition of clk_gate_set_bit:
135 static void clk_gate_set_bit(struct clk_gate *gate)
139 reg = __raw_readl(gate->reg);
140 reg |= BIT(gate->bit_idx);
141 writel(reg, gate->reg);
144 Note that to_clk_gate is defined as:
146 #define to_clk_gate(_hw) container_of(_hw, struct clk_gate, clk)
148 This pattern of abstraction is used for every clock hardware
151 Part 4 - supporting your own clk hardware
153 When implementing support for a new type of clock it only necessary to
154 include the following header:
156 #include <linux/clk-provider.h>
158 include/linux/clk.h is included within that header and clk-private.h
159 must never be included from the code which implements the operations for
160 a clock. More on that below in Part 5.
162 To construct a clk hardware structure for your platform you must define
167 ... hardware specific data goes here ...
170 To take advantage of your data you'll need to support valid operations
173 struct clk_ops clk_foo_ops {
174 .enable = &clk_foo_enable;
175 .disable = &clk_foo_disable;
178 Implement the above functions using container_of:
180 #define to_clk_foo(_hw) container_of(_hw, struct clk_foo, hw)
182 int clk_foo_enable(struct clk_hw *hw)
186 foo = to_clk_foo(hw);
188 ... perform magic on foo ...
193 Below is a matrix detailing which clk_ops are mandatory based upon the
194 hardware capabilities of that clock. A cell marked as "y" means
195 mandatory, a cell marked as "n" implies that either including that
196 callback is invalid or otherwise unnecessary. Empty cells are either
197 optional or must be evaluated on a case-by-case basis.
199 clock hardware characteristics
200 -----------------------------------------------------------
201 | gate | change rate | single parent | multiplexer | root |
202 |------|-------------|---------------|-------------|------|
204 .unprepare | | | | | |
206 .enable | y | | | | |
207 .disable | y | | | | |
208 .is_enabled | y | | | | |
210 .recalc_rate | | y | | | |
211 .round_rate | | y [1] | | | |
212 .determine_rate | | y [1] | | | |
213 .set_rate | | y | | | |
215 .set_parent | | | n | y | n |
216 .get_parent | | | n | y | n |
218 .recalc_accuracy| | | | | |
221 -----------------------------------------------------------
222 [1] either one of round_rate or determine_rate is required.
224 Finally, register your clock at run-time with a hardware-specific
225 registration function. This function simply populates struct clk_foo's
226 data and then passes the common struct clk parameters to the framework
231 See the basic clock types in drivers/clk/clk-*.c for examples.
233 Part 5 - static initialization of clock data
235 For platforms with many clocks (often numbering into the hundreds) it
236 may be desirable to statically initialize some clock data. This
237 presents a problem since the definition of struct clk should be hidden
238 from everyone except for the clock core in drivers/clk/clk.c.
240 To get around this problem struct clk's definition is exposed in
241 include/linux/clk-private.h along with some macros for more easily
242 initializing instances of the basic clock types. These clocks must
243 still be initialized with the common clock framework via a call to
246 clk-private.h must NEVER be included by code which implements struct
247 clk_ops callbacks, nor must it be included by any logic which pokes
248 around inside of struct clk at run-time. To do so is a layering
251 To better enforce this policy, always follow this simple rule: any
252 statically initialized clock data MUST be defined in a separate file
253 from the logic that implements its ops. Basically separate the logic
254 from the data and all is well.
256 Part 6 - Disabling clock gating of unused clocks
258 Sometimes during development it can be useful to be able to bypass the
259 default disabling of unused clocks. For example, if drivers aren't enabling
260 clocks properly but rely on them being on from the bootloader, bypassing
261 the disabling means that the driver will remain functional while the issues
264 To bypass this disabling, include "clk_ignore_unused" in the bootargs to the
269 The common clock framework uses two global locks, the prepare lock and the
272 The enable lock is a spinlock and is held across calls to the .enable,
273 .disable and .is_enabled operations. Those operations are thus not allowed to
274 sleep, and calls to the clk_enable(), clk_disable() and clk_is_enabled() API
275 functions are allowed in atomic context.
277 The prepare lock is a mutex and is held across calls to all other operations.
278 All those operations are allowed to sleep, and calls to the corresponding API
279 functions are not allowed in atomic context.
281 This effectively divides operations in two groups from a locking perspective.
283 Drivers don't need to manually protect resources shared between the operations
284 of one group, regardless of whether those resources are shared by multiple
285 clocks or not. However, access to resources that are shared between operations
286 of the two groups needs to be protected by the drivers. An example of such a
287 resource would be a register that controls both the clock rate and the clock
288 enable/disable state.
290 The clock framework is reentrant, in that a driver is allowed to call clock
291 framework functions from within its implementation of clock operations. This
292 can for instance cause a .set_rate operation of one clock being called from
293 within the .set_rate operation of another clock. This case must be considered
294 in the driver implementations, but the code flow is usually controlled by the
297 Note that locking must also be considered when code outside of the common
298 clock framework needs to access resources used by the clock operations. This
299 is considered out of scope of this document.