4 The V4L2 control API seems simple enough, but quickly becomes very hard to
5 implement correctly in drivers. But much of the code needed to handle controls
6 is actually not driver specific and can be moved to the V4L core framework.
8 After all, the only part that a driver developer is interested in is:
10 1) How do I add a control?
11 2) How do I set the control's value? (i.e. s_ctrl)
15 3) How do I get the control's value? (i.e. g_volatile_ctrl)
16 4) How do I validate the user's proposed control value? (i.e. try_ctrl)
18 All the rest is something that can be done centrally.
20 The control framework was created in order to implement all the rules of the
21 V4L2 specification with respect to controls in a central place. And to make
22 life as easy as possible for the driver developer.
24 Note that the control framework relies on the presence of a struct v4l2_device
25 for V4L2 drivers and struct v4l2_subdev for sub-device drivers.
28 Objects in the framework
29 ========================
31 There are two main objects:
33 The v4l2_ctrl object describes the control properties and keeps track of the
34 control's value (both the current value and the proposed new value).
36 v4l2_ctrl_handler is the object that keeps track of controls. It maintains a
37 list of v4l2_ctrl objects that it owns and another list of references to
38 controls, possibly to controls owned by other handlers.
41 Basic usage for V4L2 and sub-device drivers
42 ===========================================
44 1) Prepare the driver:
46 1.1) Add the handler to your driver's top-level struct:
50 struct v4l2_ctrl_handler ctrl_handler;
56 1.2) Initialize the handler:
58 v4l2_ctrl_handler_init(&foo->ctrl_handler, nr_of_controls);
60 The second argument is a hint telling the function how many controls this
61 handler is expected to handle. It will allocate a hashtable based on this
62 information. It is a hint only.
64 1.3) Hook the control handler into the driver:
66 1.3.1) For V4L2 drivers do this:
70 struct v4l2_device v4l2_dev;
72 struct v4l2_ctrl_handler ctrl_handler;
76 foo->v4l2_dev.ctrl_handler = &foo->ctrl_handler;
78 Where foo->v4l2_dev is of type struct v4l2_device.
80 Finally, remove all control functions from your v4l2_ioctl_ops:
81 vidioc_queryctrl, vidioc_querymenu, vidioc_g_ctrl, vidioc_s_ctrl,
82 vidioc_g_ext_ctrls, vidioc_try_ext_ctrls and vidioc_s_ext_ctrls.
83 Those are now no longer needed.
85 1.3.2) For sub-device drivers do this:
89 struct v4l2_subdev sd;
91 struct v4l2_ctrl_handler ctrl_handler;
95 foo->sd.ctrl_handler = &foo->ctrl_handler;
97 Where foo->sd is of type struct v4l2_subdev.
99 And set all core control ops in your struct v4l2_subdev_core_ops to these
102 .queryctrl = v4l2_subdev_queryctrl,
103 .querymenu = v4l2_subdev_querymenu,
104 .g_ctrl = v4l2_subdev_g_ctrl,
105 .s_ctrl = v4l2_subdev_s_ctrl,
106 .g_ext_ctrls = v4l2_subdev_g_ext_ctrls,
107 .try_ext_ctrls = v4l2_subdev_try_ext_ctrls,
108 .s_ext_ctrls = v4l2_subdev_s_ext_ctrls,
110 Note: this is a temporary solution only. Once all V4L2 drivers that depend
111 on subdev drivers are converted to the control framework these helpers will
114 1.4) Clean up the handler at the end:
116 v4l2_ctrl_handler_free(&foo->ctrl_handler);
121 You add non-menu controls by calling v4l2_ctrl_new_std:
123 struct v4l2_ctrl *v4l2_ctrl_new_std(struct v4l2_ctrl_handler *hdl,
124 const struct v4l2_ctrl_ops *ops,
125 u32 id, s32 min, s32 max, u32 step, s32 def);
127 Menu controls are added by calling v4l2_ctrl_new_std_menu:
129 struct v4l2_ctrl *v4l2_ctrl_new_std_menu(struct v4l2_ctrl_handler *hdl,
130 const struct v4l2_ctrl_ops *ops,
131 u32 id, s32 max, s32 skip_mask, s32 def);
133 These functions are typically called right after the v4l2_ctrl_handler_init:
135 v4l2_ctrl_handler_init(&foo->ctrl_handler, nr_of_controls);
136 v4l2_ctrl_new_std(&foo->ctrl_handler, &foo_ctrl_ops,
137 V4L2_CID_BRIGHTNESS, 0, 255, 1, 128);
138 v4l2_ctrl_new_std(&foo->ctrl_handler, &foo_ctrl_ops,
139 V4L2_CID_CONTRAST, 0, 255, 1, 128);
140 v4l2_ctrl_new_std_menu(&foo->ctrl_handler, &foo_ctrl_ops,
141 V4L2_CID_POWER_LINE_FREQUENCY,
142 V4L2_CID_POWER_LINE_FREQUENCY_60HZ, 0,
143 V4L2_CID_POWER_LINE_FREQUENCY_DISABLED);
145 if (foo->ctrl_handler.error) {
146 int err = foo->ctrl_handler.error;
148 v4l2_ctrl_handler_free(&foo->ctrl_handler);
152 The v4l2_ctrl_new_std function returns the v4l2_ctrl pointer to the new
153 control, but if you do not need to access the pointer outside the control ops,
154 then there is no need to store it.
156 The v4l2_ctrl_new_std function will fill in most fields based on the control
157 ID except for the min, max, step and default values. These are passed in the
158 last four arguments. These values are driver specific while control attributes
159 like type, name, flags are all global. The control's current value will be set
160 to the default value.
162 The v4l2_ctrl_new_std_menu function is very similar but it is used for menu
163 controls. There is no min argument since that is always 0 for menu controls,
164 and instead of a step there is a skip_mask argument: if bit X is 1, then menu
167 Note that if something fails, the function will return NULL or an error and
168 set ctrl_handler->error to the error code. If ctrl_handler->error was already
169 set, then it will just return and do nothing. This is also true for
170 v4l2_ctrl_handler_init if it cannot allocate the internal data structure.
172 This makes it easy to init the handler and just add all controls and only check
173 the error code at the end. Saves a lot of repetitive error checking.
175 It is recommended to add controls in ascending control ID order: it will be
176 a bit faster that way.
178 3) Optionally force initial control setup:
180 v4l2_ctrl_handler_setup(&foo->ctrl_handler);
182 This will call s_ctrl for all controls unconditionally. Effectively this
183 initializes the hardware to the default control values. It is recommended
184 that you do this as this ensures that both the internal data structures and
185 the hardware are in sync.
187 4) Finally: implement the v4l2_ctrl_ops
189 static const struct v4l2_ctrl_ops foo_ctrl_ops = {
190 .s_ctrl = foo_s_ctrl,
193 Usually all you need is s_ctrl:
195 static int foo_s_ctrl(struct v4l2_ctrl *ctrl)
197 struct foo *state = container_of(ctrl->handler, struct foo, ctrl_handler);
200 case V4L2_CID_BRIGHTNESS:
201 write_reg(0x123, ctrl->val);
203 case V4L2_CID_CONTRAST:
204 write_reg(0x456, ctrl->val);
210 The control ops are called with the v4l2_ctrl pointer as argument.
211 The new control value has already been validated, so all you need to do is
212 to actually update the hardware registers.
214 You're done! And this is sufficient for most of the drivers we have. No need
215 to do any validation of control values, or implement QUERYCTRL/QUERYMENU. And
216 G/S_CTRL as well as G/TRY/S_EXT_CTRLS are automatically supported.
219 ==============================================================================
221 The remainder of this document deals with more advanced topics and scenarios.
222 In practice the basic usage as described above is sufficient for most drivers.
224 ===============================================================================
230 When a sub-device is registered with a V4L2 driver by calling
231 v4l2_device_register_subdev() and the ctrl_handler fields of both v4l2_subdev
232 and v4l2_device are set, then the controls of the subdev will become
233 automatically available in the V4L2 driver as well. If the subdev driver
234 contains controls that already exist in the V4L2 driver, then those will be
235 skipped (so a V4L2 driver can always override a subdev control).
237 What happens here is that v4l2_device_register_subdev() calls
238 v4l2_ctrl_add_handler() adding the controls of the subdev to the controls
242 Accessing Control Values
243 ========================
245 The v4l2_ctrl struct contains these two unions:
247 /* The current control value. */
254 /* The new control value. */
261 Within the control ops you can freely use these. The val and val64 speak for
262 themselves. The string pointers point to character buffers of length
263 ctrl->maximum + 1, and are always 0-terminated.
265 In most cases 'cur' contains the current cached control value. When you create
266 a new control this value is made identical to the default value. After calling
267 v4l2_ctrl_handler_setup() this value is passed to the hardware. It is generally
268 a good idea to call this function.
270 Whenever a new value is set that new value is automatically cached. This means
271 that most drivers do not need to implement the g_volatile_ctrl() op. The
272 exception is for controls that return a volatile register such as a signal
273 strength read-out that changes continuously. In that case you will need to
274 implement g_volatile_ctrl like this:
276 static int foo_g_volatile_ctrl(struct v4l2_ctrl *ctrl)
279 case V4L2_CID_BRIGHTNESS:
280 ctrl->val = read_reg(0x123);
285 Note that you use the 'new value' union as well in g_volatile_ctrl. In general
286 controls that need to implement g_volatile_ctrl are read-only controls.
288 To mark a control as volatile you have to set the is_volatile flag:
290 ctrl = v4l2_ctrl_new_std(&sd->ctrl_handler, ...);
292 ctrl->is_volatile = 1;
294 For try/s_ctrl the new values (i.e. as passed by the user) are filled in and
295 you can modify them in try_ctrl or set them in s_ctrl. The 'cur' union
296 contains the current value, which you can use (but not change!) as well.
298 If s_ctrl returns 0 (OK), then the control framework will copy the new final
299 values to the 'cur' union.
301 While in g_volatile/s/try_ctrl you can access the value of all controls owned
302 by the same handler since the handler's lock is held. If you need to access
303 the value of controls owned by other handlers, then you have to be very careful
304 not to introduce deadlocks.
306 Outside of the control ops you have to go through to helper functions to get
307 or set a single control value safely in your driver:
309 s32 v4l2_ctrl_g_ctrl(struct v4l2_ctrl *ctrl);
310 int v4l2_ctrl_s_ctrl(struct v4l2_ctrl *ctrl, s32 val);
312 These functions go through the control framework just as VIDIOC_G/S_CTRL ioctls
313 do. Don't use these inside the control ops g_volatile/s/try_ctrl, though, that
314 will result in a deadlock since these helpers lock the handler as well.
316 You can also take the handler lock yourself:
318 mutex_lock(&state->ctrl_handler.lock);
319 printk(KERN_INFO "String value is '%s'\n", ctrl1->cur.string);
320 printk(KERN_INFO "Integer value is '%s'\n", ctrl2->cur.val);
321 mutex_unlock(&state->ctrl_handler.lock);
327 The v4l2_ctrl struct contains this union:
334 For menu controls menu_skip_mask is used. What it does is that it allows you
335 to easily exclude certain menu items. This is used in the VIDIOC_QUERYMENU
336 implementation where you can return -EINVAL if a certain menu item is not
337 present. Note that VIDIOC_QUERYCTRL always returns a step value of 1 for
340 A good example is the MPEG Audio Layer II Bitrate menu control where the
341 menu is a list of standardized possible bitrates. But in practice hardware
342 implementations will only support a subset of those. By setting the skip
343 mask you can tell the framework which menu items should be skipped. Setting
344 it to 0 means that all menu items are supported.
346 You set this mask either through the v4l2_ctrl_config struct for a custom
347 control, or by calling v4l2_ctrl_new_std_menu().
353 Driver specific controls can be created using v4l2_ctrl_new_custom():
355 static const struct v4l2_ctrl_config ctrl_filter = {
356 .ops = &ctrl_custom_ops,
357 .id = V4L2_CID_MPEG_CX2341X_VIDEO_SPATIAL_FILTER,
358 .name = "Spatial Filter",
359 .type = V4L2_CTRL_TYPE_INTEGER,
360 .flags = V4L2_CTRL_FLAG_SLIDER,
365 ctrl = v4l2_ctrl_new_custom(&foo->ctrl_handler, &ctrl_filter, NULL);
367 The last argument is the priv pointer which can be set to driver-specific
370 The v4l2_ctrl_config struct also has fields to set the is_private and is_volatile
373 If the name field is not set, then the framework will assume this is a standard
374 control and will fill in the name, type and flags fields accordingly.
377 Active and Grabbed Controls
378 ===========================
380 If you get more complex relationships between controls, then you may have to
381 activate and deactivate controls. For example, if the Chroma AGC control is
382 on, then the Chroma Gain control is inactive. That is, you may set it, but
383 the value will not be used by the hardware as long as the automatic gain
384 control is on. Typically user interfaces can disable such input fields.
386 You can set the 'active' status using v4l2_ctrl_activate(). By default all
387 controls are active. Note that the framework does not check for this flag.
388 It is meant purely for GUIs. The function is typically called from within
391 The other flag is the 'grabbed' flag. A grabbed control means that you cannot
392 change it because it is in use by some resource. Typical examples are MPEG
393 bitrate controls that cannot be changed while capturing is in progress.
395 If a control is set to 'grabbed' using v4l2_ctrl_grab(), then the framework
396 will return -EBUSY if an attempt is made to set this control. The
397 v4l2_ctrl_grab() function is typically called from the driver when it
398 starts or stops streaming.
404 By default all controls are independent from the others. But in more
405 complex scenarios you can get dependencies from one control to another.
406 In that case you need to 'cluster' them:
409 struct v4l2_ctrl_handler ctrl_handler;
410 #define AUDIO_CL_VOLUME (0)
411 #define AUDIO_CL_MUTE (1)
412 struct v4l2_ctrl *audio_cluster[2];
416 state->audio_cluster[AUDIO_CL_VOLUME] =
417 v4l2_ctrl_new_std(&state->ctrl_handler, ...);
418 state->audio_cluster[AUDIO_CL_MUTE] =
419 v4l2_ctrl_new_std(&state->ctrl_handler, ...);
420 v4l2_ctrl_cluster(ARRAY_SIZE(state->audio_cluster), state->audio_cluster);
422 From now on whenever one or more of the controls belonging to the same
423 cluster is set (or 'gotten', or 'tried'), only the control ops of the first
424 control ('volume' in this example) is called. You effectively create a new
425 composite control. Similar to how a 'struct' works in C.
427 So when s_ctrl is called with V4L2_CID_AUDIO_VOLUME as argument, you should set
428 all two controls belonging to the audio_cluster:
430 static int foo_s_ctrl(struct v4l2_ctrl *ctrl)
432 struct foo *state = container_of(ctrl->handler, struct foo, ctrl_handler);
435 case V4L2_CID_AUDIO_VOLUME: {
436 struct v4l2_ctrl *mute = ctrl->cluster[AUDIO_CL_MUTE];
438 write_reg(0x123, mute->val ? 0 : ctrl->val);
441 case V4L2_CID_CONTRAST:
442 write_reg(0x456, ctrl->val);
448 In the example above the following are equivalent for the VOLUME case:
450 ctrl == ctrl->cluster[AUDIO_CL_VOLUME] == state->audio_cluster[AUDIO_CL_VOLUME]
451 ctrl->cluster[AUDIO_CL_MUTE] == state->audio_cluster[AUDIO_CL_MUTE]
453 In practice using cluster arrays like this becomes very tiresome. So instead
454 the following equivalent method is used:
458 struct v4l2_ctrl *volume;
459 struct v4l2_ctrl *mute;
462 The anonymous struct is used to clearly 'cluster' these two control pointers,
463 but it serves no other purpose. The effect is the same as creating an
464 array with two control pointers. So you can just do:
466 state->volume = v4l2_ctrl_new_std(&state->ctrl_handler, ...);
467 state->mute = v4l2_ctrl_new_std(&state->ctrl_handler, ...);
468 v4l2_ctrl_cluster(2, &state->volume);
470 And in foo_s_ctrl you can use these pointers directly: state->mute->val.
472 Note that controls in a cluster may be NULL. For example, if for some
473 reason mute was never added (because the hardware doesn't support that
474 particular feature), then mute will be NULL. So in that case we have a
475 cluster of 2 controls, of which only 1 is actually instantiated. The
476 only restriction is that the first control of the cluster must always be
477 present, since that is the 'master' control of the cluster. The master
478 control is the one that identifies the cluster and that provides the
479 pointer to the v4l2_ctrl_ops struct that is used for that cluster.
481 Obviously, all controls in the cluster array must be initialized to either
482 a valid control or to NULL.
484 In rare cases you might want to know which controls of a cluster actually
485 were set explicitly by the user. For this you can check the 'is_new' flag of
486 each control. For example, in the case of a volume/mute cluster the 'is_new'
487 flag of the mute control would be set if the user called VIDIOC_S_CTRL for
488 mute only. If the user would call VIDIOC_S_EXT_CTRLS for both mute and volume
489 controls, then the 'is_new' flag would be 1 for both controls.
491 The 'is_new' flag is always 1 when called from v4l2_ctrl_handler_setup().
494 Handling autogain/gain-type Controls with Auto Clusters
495 =======================================================
497 A common type of control cluster is one that handles 'auto-foo/foo'-type
498 controls. Typical examples are autogain/gain, autoexposure/exposure,
499 autowhitebalance/red balance/blue balance. In all cases you have one controls
500 that determines whether another control is handled automatically by the hardware,
501 or whether it is under manual control from the user.
503 If the cluster is in automatic mode, then the manual controls should be
504 marked inactive. When the volatile controls are read the g_volatile_ctrl
505 operation should return the value that the hardware's automatic mode set up
508 If the cluster is put in manual mode, then the manual controls should become
509 active again and the is_volatile flag should be ignored (so g_volatile_ctrl is
510 no longer called while in manual mode).
512 Finally the V4L2_CTRL_FLAG_UPDATE should be set for the auto control since
513 changing that control affects the control flags of the manual controls.
515 In order to simplify this a special variation of v4l2_ctrl_cluster was
518 void v4l2_ctrl_auto_cluster(unsigned ncontrols, struct v4l2_ctrl **controls,
519 u8 manual_val, bool set_volatile);
521 The first two arguments are identical to v4l2_ctrl_cluster. The third argument
522 tells the framework which value switches the cluster into manual mode. The
523 last argument will optionally set the is_volatile flag for the non-auto controls.
525 The first control of the cluster is assumed to be the 'auto' control.
527 Using this function will ensure that you don't need to handle all the complex
528 flag and volatile handling.
531 VIDIOC_LOG_STATUS Support
532 =========================
534 This ioctl allow you to dump the current status of a driver to the kernel log.
535 The v4l2_ctrl_handler_log_status(ctrl_handler, prefix) can be used to dump the
536 value of the controls owned by the given handler to the log. You can supply a
537 prefix as well. If the prefix didn't end with a space, then ': ' will be added
541 Different Handlers for Different Video Nodes
542 ============================================
544 Usually the V4L2 driver has just one control handler that is global for
545 all video nodes. But you can also specify different control handlers for
546 different video nodes. You can do that by manually setting the ctrl_handler
547 field of struct video_device.
549 That is no problem if there are no subdevs involved but if there are, then
550 you need to block the automatic merging of subdev controls to the global
551 control handler. You do that by simply setting the ctrl_handler field in
552 struct v4l2_device to NULL. Now v4l2_device_register_subdev() will no longer
553 merge subdev controls.
555 After each subdev was added, you will then have to call v4l2_ctrl_add_handler
556 manually to add the subdev's control handler (sd->ctrl_handler) to the desired
557 control handler. This control handler may be specific to the video_device or
558 for a subset of video_device's. For example: the radio device nodes only have
559 audio controls, while the video and vbi device nodes share the same control
560 handler for the audio and video controls.
562 If you want to have one handler (e.g. for a radio device node) have a subset
563 of another handler (e.g. for a video device node), then you should first add
564 the controls to the first handler, add the other controls to the second
565 handler and finally add the first handler to the second. For example:
567 v4l2_ctrl_new_std(&radio_ctrl_handler, &radio_ops, V4L2_CID_AUDIO_VOLUME, ...);
568 v4l2_ctrl_new_std(&radio_ctrl_handler, &radio_ops, V4L2_CID_AUDIO_MUTE, ...);
569 v4l2_ctrl_new_std(&video_ctrl_handler, &video_ops, V4L2_CID_BRIGHTNESS, ...);
570 v4l2_ctrl_new_std(&video_ctrl_handler, &video_ops, V4L2_CID_CONTRAST, ...);
571 v4l2_ctrl_add_handler(&video_ctrl_handler, &radio_ctrl_handler);
573 Or you can add specific controls to a handler:
575 volume = v4l2_ctrl_new_std(&video_ctrl_handler, &ops, V4L2_CID_AUDIO_VOLUME, ...);
576 v4l2_ctrl_new_std(&video_ctrl_handler, &ops, V4L2_CID_BRIGHTNESS, ...);
577 v4l2_ctrl_new_std(&video_ctrl_handler, &ops, V4L2_CID_CONTRAST, ...);
578 v4l2_ctrl_add_ctrl(&radio_ctrl_handler, volume);
580 What you should not do is make two identical controls for two handlers.
583 v4l2_ctrl_new_std(&radio_ctrl_handler, &radio_ops, V4L2_CID_AUDIO_MUTE, ...);
584 v4l2_ctrl_new_std(&video_ctrl_handler, &video_ops, V4L2_CID_AUDIO_MUTE, ...);
586 This would be bad since muting the radio would not change the video mute
587 control. The rule is to have one control for each hardware 'knob' that you
594 Normally you have created the controls yourself and you can store the struct
595 v4l2_ctrl pointer into your own struct.
597 But sometimes you need to find a control from another handler that you do
598 not own. For example, if you have to find a volume control from a subdev.
600 You can do that by calling v4l2_ctrl_find:
602 struct v4l2_ctrl *volume;
604 volume = v4l2_ctrl_find(sd->ctrl_handler, V4L2_CID_AUDIO_VOLUME);
606 Since v4l2_ctrl_find will lock the handler you have to be careful where you
607 use it. For example, this is not a good idea:
609 struct v4l2_ctrl_handler ctrl_handler;
611 v4l2_ctrl_new_std(&ctrl_handler, &video_ops, V4L2_CID_BRIGHTNESS, ...);
612 v4l2_ctrl_new_std(&ctrl_handler, &video_ops, V4L2_CID_CONTRAST, ...);
614 ...and in video_ops.s_ctrl:
616 case V4L2_CID_BRIGHTNESS:
617 contrast = v4l2_find_ctrl(&ctrl_handler, V4L2_CID_CONTRAST);
620 When s_ctrl is called by the framework the ctrl_handler.lock is already taken, so
621 attempting to find another control from the same handler will deadlock.
623 It is recommended not to use this function from inside the control ops.
629 When one control handler is added to another using v4l2_ctrl_add_handler, then
630 by default all controls from one are merged to the other. But a subdev might
631 have low-level controls that make sense for some advanced embedded system, but
632 not when it is used in consumer-level hardware. In that case you want to keep
633 those low-level controls local to the subdev. You can do this by simply
634 setting the 'is_private' flag of the control to 1:
636 static const struct v4l2_ctrl_config ctrl_private = {
637 .ops = &ctrl_custom_ops,
639 .name = "Some Private Control",
640 .type = V4L2_CTRL_TYPE_INTEGER,
646 ctrl = v4l2_ctrl_new_custom(&foo->ctrl_handler, &ctrl_private, NULL);
648 These controls will now be skipped when v4l2_ctrl_add_handler is called.
651 V4L2_CTRL_TYPE_CTRL_CLASS Controls
652 ==================================
654 Controls of this type can be used by GUIs to get the name of the control class.
655 A fully featured GUI can make a dialog with multiple tabs with each tab
656 containing the controls belonging to a particular control class. The name of
657 each tab can be found by querying a special control with ID <control class | 1>.
659 Drivers do not have to care about this. The framework will automatically add
660 a control of this type whenever the first control belonging to a new control
664 Differences from the Spec
665 =========================
667 There are a few places where the framework acts slightly differently from the
668 V4L2 Specification. Those differences are described in this section. We will
669 have to see whether we need to adjust the spec or not.
671 1) It is no longer required to have all controls contained in a
672 v4l2_ext_control array be from the same control class. The framework will be
673 able to handle any type of control in the array. You need to set ctrl_class
674 to 0 in order to enable this. If ctrl_class is non-zero, then it will still
675 check that all controls belong to that control class.
677 If you set ctrl_class to 0 and count to 0, then it will only return an error
678 if there are no controls at all.
680 2) Clarified the way error_idx works. For get and set it will be equal to
681 count if nothing was done yet. If it is less than count then only the controls
682 up to error_idx-1 were successfully applied.
684 3) When attempting to read a button control the framework will return -EACCES
685 instead of -EINVAL as stated in the spec. It seems to make more sense since
686 button controls are write-only controls.
688 4) Attempting to write to a read-only control will return -EACCES instead of
689 -EINVAL as the spec says.
691 5) The spec does not mention what should happen when you try to set/get a
692 control class controls. The framework will return -EACCES.
695 Proposals for Extensions
696 ========================
698 Some ideas for future extensions to the spec:
700 1) Add a V4L2_CTRL_FLAG_HEX to have values shown as hexadecimal instead of
701 decimal. Useful for e.g. video_mute_yuv.
703 2) It is possible to mark in the controls array which controls have been
704 successfully written and which failed by for example adding a bit to the
705 control ID. Not sure if it is worth the effort, though.
707 3) Trying to set volatile inactive controls should result in -EACCESS.
709 4) Add a new flag to mark volatile controls. Any application that wants
710 to store the state of the controls can then skip volatile inactive controls.
711 Currently it is not possible to detect such controls.