1 The Common Clk Framework 2 Mike Turquette <mturquette@ti.com> 3 4This document endeavours to explain the common clk framework details, 5and how to port a platform over to this framework. It is not yet a 6detailed explanation of the clock api in include/linux/clk.h, but 7perhaps someday it will include that information. 8 9 Part 1 - introduction and interface split 10 11The common clk framework is an interface to control the clock nodes 12available on various devices today. This may come in the form of clock 13gating, rate adjustment, muxing or other operations. This framework is 14enabled with the CONFIG_COMMON_CLK option. 15 16The interface itself is divided into two halves, each shielded from the 17details of its counterpart. First is the common definition of struct 18clk which unifies the framework-level accounting and infrastructure that 19has traditionally been duplicated across a variety of platforms. Second 20is a common implementation of the clk.h api, defined in 21drivers/clk/clk.c. Finally there is struct clk_ops, whose operations 22are invoked by the clk api implementation. 23 24The second half of the interface is comprised of the hardware-specific 25callbacks registered with struct clk_ops and the corresponding 26hardware-specific structures needed to model a particular clock. For 27the remainder of this document any reference to a callback in struct 28clk_ops, such as .enable or .set_rate, implies the hardware-specific 29implementation of that code. Likewise, references to struct clk_foo 30serve as a convenient shorthand for the implementation of the 31hardware-specific bits for the hypothetical "foo" hardware. 32 33Tying the two halves of this interface together is struct clk_hw, which 34is defined in struct clk_foo and pointed to within struct clk. This 35allows for easy navigation between the two discrete halves of the common 36clock interface. 37 38 Part 2 - common data structures and api 39 40Below is the common struct clk definition from 41include/linux/clk-private.h, modified for brevity: 42 43 struct clk { 44 const char *name; 45 const struct clk_ops *ops; 46 struct clk_hw *hw; 47 char **parent_names; 48 struct clk **parents; 49 struct clk *parent; 50 struct hlist_head children; 51 struct hlist_node child_node; 52 ... 53 }; 54 55The members above make up the core of the clk tree topology. The clk 56api itself defines several driver-facing functions which operate on 57struct clk. That api is documented in include/linux/clk.h. 58 59Platforms and devices utilizing the common struct clk use the struct 60clk_ops pointer in struct clk to perform the hardware-specific parts of 61the operations defined in clk.h: 62 63 struct clk_ops { 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, 72 unsigned long rate, 73 unsigned long *parent_rate); 74 long (*determine_rate)(struct clk_hw *hw, 75 unsigned long rate, 76 unsigned long *best_parent_rate, 77 struct clk **best_parent_clk); 78 int (*set_parent)(struct clk_hw *hw, u8 index); 79 u8 (*get_parent)(struct clk_hw *hw); 80 int (*set_rate)(struct clk_hw *hw, 81 unsigned long rate, 82 unsigned long parent_rate); 83 int (*set_rate_and_parent)(struct clk_hw *hw, 84 unsigned long rate, 85 unsigned long parent_rate, 86 u8 index); 87 unsigned long (*recalc_accuracy)(struct clk_hw *hw, 88 unsigned long parent_accuracy); 89 void (*init)(struct clk_hw *hw); 90 int (*debug_init)(struct clk_hw *hw, 91 struct dentry *dentry); 92 }; 93 94 Part 3 - hardware clk implementations 95 96The strength of the common struct clk comes from its .ops and .hw pointers 97which abstract the details of struct clk from the hardware-specific bits, and 98vice versa. To illustrate consider the simple gateable clk implementation in 99drivers/clk/clk-gate.c: 100 101struct clk_gate { 102 struct clk_hw hw; 103 void __iomem *reg; 104 u8 bit_idx; 105 ... 106}; 107 108struct clk_gate contains struct clk_hw hw as well as hardware-specific 109knowledge about which register and bit controls this clk's gating. 110Nothing about clock topology or accounting, such as enable_count or 111notifier_count, is needed here. That is all handled by the common 112framework code and struct clk. 113 114Let's walk through enabling this clk from driver code: 115 116 struct clk *clk; 117 clk = clk_get(NULL, "my_gateable_clk"); 118 119 clk_prepare(clk); 120 clk_enable(clk); 121 122The call graph for clk_enable is very simple: 123 124clk_enable(clk); 125 clk->ops->enable(clk->hw); 126 [resolves to...] 127 clk_gate_enable(hw); 128 [resolves struct clk gate with to_clk_gate(hw)] 129 clk_gate_set_bit(gate); 130 131And the definition of clk_gate_set_bit: 132 133static void clk_gate_set_bit(struct clk_gate *gate) 134{ 135 u32 reg; 136 137 reg = __raw_readl(gate->reg); 138 reg |= BIT(gate->bit_idx); 139 writel(reg, gate->reg); 140} 141 142Note that to_clk_gate is defined as: 143 144#define to_clk_gate(_hw) container_of(_hw, struct clk_gate, clk) 145 146This pattern of abstraction is used for every clock hardware 147representation. 148 149 Part 4 - supporting your own clk hardware 150 151When implementing support for a new type of clock it only necessary to 152include the following header: 153 154#include <linux/clk-provider.h> 155 156include/linux/clk.h is included within that header and clk-private.h 157must never be included from the code which implements the operations for 158a clock. More on that below in Part 5. 159 160To construct a clk hardware structure for your platform you must define 161the following: 162 163struct clk_foo { 164 struct clk_hw hw; 165 ... hardware specific data goes here ... 166}; 167 168To take advantage of your data you'll need to support valid operations 169for your clk: 170 171struct clk_ops clk_foo_ops { 172 .enable = &clk_foo_enable; 173 .disable = &clk_foo_disable; 174}; 175 176Implement the above functions using container_of: 177 178#define to_clk_foo(_hw) container_of(_hw, struct clk_foo, hw) 179 180int clk_foo_enable(struct clk_hw *hw) 181{ 182 struct clk_foo *foo; 183 184 foo = to_clk_foo(hw); 185 186 ... perform magic on foo ... 187 188 return 0; 189}; 190 191Below is a matrix detailing which clk_ops are mandatory based upon the 192hardware capabilities of that clock. A cell marked as "y" means 193mandatory, a cell marked as "n" implies that either including that 194callback is invalid or otherwise unnecessary. Empty cells are either 195optional or must be evaluated on a case-by-case basis. 196 197 clock hardware characteristics 198 ----------------------------------------------------------- 199 | gate | change rate | single parent | multiplexer | root | 200 |------|-------------|---------------|-------------|------| 201.prepare | | | | | | 202.unprepare | | | | | | 203 | | | | | | 204.enable | y | | | | | 205.disable | y | | | | | 206.is_enabled | y | | | | | 207 | | | | | | 208.recalc_rate | | y | | | | 209.round_rate | | y [1] | | | | 210.determine_rate | | y [1] | | | | 211.set_rate | | y | | | | 212 | | | | | | 213.set_parent | | | n | y | n | 214.get_parent | | | n | y | n | 215 | | | | | | 216.recalc_accuracy| | | | | | 217 | | | | | | 218.init | | | | | | 219 ----------------------------------------------------------- 220[1] either one of round_rate or determine_rate is required. 221 222Finally, register your clock at run-time with a hardware-specific 223registration function. This function simply populates struct clk_foo's 224data and then passes the common struct clk parameters to the framework 225with a call to: 226 227clk_register(...) 228 229See the basic clock types in drivers/clk/clk-*.c for examples. 230 231 Part 5 - static initialization of clock data 232 233For platforms with many clocks (often numbering into the hundreds) it 234may be desirable to statically initialize some clock data. This 235presents a problem since the definition of struct clk should be hidden 236from everyone except for the clock core in drivers/clk/clk.c. 237 238To get around this problem struct clk's definition is exposed in 239include/linux/clk-private.h along with some macros for more easily 240initializing instances of the basic clock types. These clocks must 241still be initialized with the common clock framework via a call to 242__clk_init. 243 244clk-private.h must NEVER be included by code which implements struct 245clk_ops callbacks, nor must it be included by any logic which pokes 246around inside of struct clk at run-time. To do so is a layering 247violation. 248 249To better enforce this policy, always follow this simple rule: any 250statically initialized clock data MUST be defined in a separate file 251from the logic that implements its ops. Basically separate the logic 252from the data and all is well. 253 254 Part 6 - Disabling clock gating of unused clocks 255 256Sometimes during development it can be useful to be able to bypass the 257default disabling of unused clocks. For example, if drivers aren't enabling 258clocks properly but rely on them being on from the bootloader, bypassing 259the disabling means that the driver will remain functional while the issues 260are sorted out. 261 262To bypass this disabling, include "clk_ignore_unused" in the bootargs to the 263kernel. 264 265 Part 7 - Locking 266 267The common clock framework uses two global locks, the prepare lock and the 268enable lock. 269 270The enable lock is a spinlock and is held across calls to the .enable, 271.disable and .is_enabled operations. Those operations are thus not allowed to 272sleep, and calls to the clk_enable(), clk_disable() and clk_is_enabled() API 273functions are allowed in atomic context. 274 275The prepare lock is a mutex and is held across calls to all other operations. 276All those operations are allowed to sleep, and calls to the corresponding API 277functions are not allowed in atomic context. 278 279This effectively divides operations in two groups from a locking perspective. 280 281Drivers don't need to manually protect resources shared between the operations 282of one group, regardless of whether those resources are shared by multiple 283clocks or not. However, access to resources that are shared between operations 284of the two groups needs to be protected by the drivers. An example of such a 285resource would be a register that controls both the clock rate and the clock 286enable/disable state. 287 288The clock framework is reentrant, in that a driver is allowed to call clock 289framework functions from within its implementation of clock operations. This 290can for instance cause a .set_rate operation of one clock being called from 291within the .set_rate operation of another clock. This case must be considered 292in the driver implementations, but the code flow is usually controlled by the 293driver in that case. 294 295Note that locking must also be considered when code outside of the common 296clock framework needs to access resources used by the clock operations. This 297is considered out of scope of this document. 298