Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Don't destroy user event fields when removal fails

   User event fields are destroyed before the event is removed from
   visibility. But that can fail leaving the still visible event with no
   fields. Move the destroying of the fields to after the event is
   successfully removed from visibility.

 - Initialize function graph state is fork before calling
   copy_exec_state()

   For non-CLONE_VM forks, copy_exec_state() allocates a new
   task_exec_state. If that allocation fails, ftrace_graph_exit_task()
   will free the tasks ret_stack pointer. Since that pointer is still
   using the parent's ret_stack, it mistakenly frees the parent's
   pointer too.

   Call ftrace_graph_init() on the task first which will NULL out the
   new tasks's ret_stack and if the copy fails, it will not free
   anything.

 - Remove FGRAPH_MAX_INDEX

   The macro FGRAPH_MAX_INDEX was added but never used. Remove it.

 - Save ent_size in function graph printing of nested functions

   The function graph tracer needs to look at the next event to see if
   the next event is the return of the current function entry. If it is,
   it prints a single line:

	ktime_get();

   Otherwise it prints it like a nested function:

	tick_nohz_irq_exit() {
	    ktime_get();
	    kcpustat_irq_exit();
	}

   In order to look at the next event, it must save the current event so
   that it has the information to print from it. It saves the event in
   the iterator descriptor called "ent". What it doesn't save is the
   ent_size of the event which is now used to know if the function graph
   arguments are to be printed. The peek doesn't save the size so the
   size used happens to be that of the size of the last event that was
   seen.

   Save the entry event size in the iterator descriptor so that the
   correct size is used.

 - Fix several errors with freeing data in the histogram code

   The histogram code had a lot of leaked or or incorrect accounting
   when failures happen. Correct them.

 - Fix histogram regression of .percent and .graph modifiers

   Up until 6.3 histogram values could have "percent" or "graph"
   modifiers that changed how they were printed. But a change that added
   restricting histograms values from being strings, stack traces and
   other modifiers inadvertently prevented them from using the percent
   and graph modifiers, which were legal use cases for values.

   Put back the percent and graph modifiers.

 - Fix various typos in the comments

 - Set the trace_clock before initializing a histogram with clock
   argument

   The histogram API allows the user to specific which trace clock to
   use via a "clock=" string. The histogram is set up first before the
   clock is checked. If the passed in clock is not valid, it exits
   without fully fixing up the histogram leaving it on the list and a
   use-after-free can trigger.

   Update the clock argument first and if it fails then exit gracefully
   before the histogram trigger is placed on any lists.

 - Restore :mod: trailer after parsing in ftrace_set_clr_event

   The function ftrace_set_clr_event() modifies the parse string and
   needs to put it back to what was passed in. It searches for ":mod:"
   via a strsep() but fails to put back the first ':' in the string.

   Add back the ':' in the passed in string.

 - Take trace_array reference when opening a tracer options file

   The options files are dynamically created and some tracers add their
   own options. When a tracer adds their own list of options, the
   trace_array holding them has an array to hold the list of options for
   each tracer. This array increases in size via a krealloc(), and the
   new entry gets a newly allocated array to hold the options of the new
   tracer being added.

   The element in each entry of the tracer's option array holds a
   pointer back to the trace_array, a pointer to the tracer it is
   associated to, a pointer to the flags of the option.

   The issue is that these arrays are freed when the trace_array is
   freed when its instance it represents is removed from the instances
   directory. There's a race that an open of one of these options files
   can happen when the instance is being removed.

   Add a new helper function to be called by the open function of the
   options file to iterate all existing trace_arrays under a lock and
   find the one that has the given option element in one of it's tracer
   arrays. If found, then update the associated trace_array's reference
   counter to keep it from being freed. If not found, have the open call
   return -ENODEV.

 - Disable interrupts when acquiring the lock in rb_wake_up_waiters()

   The function rb_wake_up_waiters() assumes it will be called in
   interrupt context and does not disable irqs when taking
   cpu_buffer->reader_lock, which can be called in hard interrupt
   context. The issue is in PREEMPT_RT, this function is called in
   thread context leaving this lock open to a deadlock.

   Take the lock with interrupts disabled.

 - Use rcu_assign_pointer() for tmp_ops filter hash

   The tmp_ops used in update_ftrace_direct_mod() assigns its
   filter_hash field directly, but that field is annotated as __rcu and
   sparse complains. Assign it with rcu_assign_pointer()

 - Fix use-after-free in enable_trigger_private_data_free()

   The trace_event_call is accessed through the event_trigger_data's
   trace_event_file pointer to put the trace_event_call on freeing. The
   issue is that the trace_event_file data may have been freed already
   causing a use-after-free. Add a field to the event_trigger_data that
   points directly to the trace_event_call so that it can decrement its
   reference directly without needing to go through the
   trace_event_file.

 - Fix accounting of buffer data remote headers

   trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount
   the number of pages is needed for the asked for size as it doesn't
   take into account the meta data on each page. Add a helper function
   to do the calculation properly and use that in these functions.

 - Catch nr_page_va overflow in ring_buffer_desc sizing

   The number of pages per remote ring buffer is capped by
   ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to
   overflow that field would silently allocate a descriptor smaller than
   what was asked for.

 - Do not resize the subbuf order if any per_cpu buffer is disabled

   The mmapping of ring buffers disables resizing the subbuffers, but it
   is done per-cpu whereas the subbuf size change is done for all the
   per_cpu buffers under the buffer->mutex. It could change the size of
   some while the mapping is happening on others. Have the resize of the
   subbuf order check all the per_cpu buffers under the lock to see if
   any of them is disabled before starting and causing an inconsistency
   between buffers that are being mapped.

* tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits)
  ring-buffer: Check resize_disabled before publishing the new subbuf order
  tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing
  tracing/remotes: Account for ring buffer page header in size calculation
  tracing: Don't dereference trace_event_file in deferred trigger free
  ftrace: Use rcu_assign_pointer() for tmp_ops filter hash
  ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters()
  tracing: Take trace_array reference when opening a tracer options file
  tracing: Fix ring_buffer_read_page_size() kernel-doc
  tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event()
  tracing: Fix memory corruption from a "STACKTRACE" histogram key
  tracing: Fix memory corruption from the histogram stacktrace modifier
  tracing: Undo the registration when enabling the histogram trigger fails
  tracing: Take the reference before publishing the named histogram trigger
  tracing: Set the trace clock before registering the histogram trigger
  tracing: Fix typo "preceeded" in comment
  tracing: Fix typo "availabe" in comment
  tracing: Let histogram values keep the percent and graph modifiers
  tracing: Keep the entry count when the histogram stats allocation fails
  tracing: Free histogram the field rejected for a bad modifier
  tracing: Free histogram the var ref when its initialization fails
  ...

1 files changed, 367 insertions, 0 deletions

diff --git a/Documentation/input/input-programming.rst b/Documentation/input/input-programming.rst
new file mode 100644
index 000000000..2b3e6a34e
--- /dev/null
+++ b/Documentation/input/input-programming.rst
@@ -0,0 +1,367 @@
+===============================
+Creating an input device driver
+===============================
+
+The simplest example
+~~~~~~~~~~~~~~~~~~~~
+
+Here comes a very simple example of an input device driver. The device has
+just one button and the button is accessible at i/o port BUTTON_PORT. When
+pressed or released a BUTTON_IRQ happens. The driver could look like::
+
+    #include <linux/input.h>
+    #include <linux/module.h>
+    #include <linux/init.h>
+
+    #include <asm/irq.h>
+    #include <asm/io.h>
+
+    static struct input_dev *button_dev;
+
+    static irqreturn_t button_interrupt(int irq, void *dummy)
+    {
+	    input_report_key(button_dev, BTN_0, inb(BUTTON_PORT) & 1);
+	    input_sync(button_dev);
+	    return IRQ_HANDLED;
+    }
+
+    static int __init button_init(void)
+    {
+	    int error;
+
+	    if (request_irq(BUTTON_IRQ, button_interrupt, 0, "button", NULL)) {
+		    printk(KERN_ERR "button.c: Can't allocate irq %d\n", button_irq);
+		    return -EBUSY;
+	    }
+
+	    button_dev = input_allocate_device();
+	    if (!button_dev) {
+		    printk(KERN_ERR "button.c: Not enough memory\n");
+		    error = -ENOMEM;
+		    goto err_free_irq;
+	    }
+
+	    button_dev->evbit[0] = BIT_MASK(EV_KEY);
+	    button_dev->keybit[BIT_WORD(BTN_0)] = BIT_MASK(BTN_0);
+
+	    error = input_register_device(button_dev);
+	    if (error) {
+		    printk(KERN_ERR "button.c: Failed to register device\n");
+		    goto err_free_dev;
+	    }
+
+	    return 0;
+
+    err_free_dev:
+	    input_free_device(button_dev);
+    err_free_irq:
+	    free_irq(BUTTON_IRQ, button_interrupt);
+	    return error;
+    }
+
+    static void __exit button_exit(void)
+    {
+	    input_unregister_device(button_dev);
+	    free_irq(BUTTON_IRQ, button_interrupt);
+    }
+
+    module_init(button_init);
+    module_exit(button_exit);
+
+What the example does
+~~~~~~~~~~~~~~~~~~~~~
+
+First it has to include the <linux/input.h> file, which interfaces to the
+input subsystem. This provides all the definitions needed.
+
+In the _init function, which is called either upon module load or when
+booting the kernel, it grabs the required resources (it should also check
+for the presence of the device).
+
+Then it allocates a new input device structure with input_allocate_device()
+and sets up input bitfields. This way the device driver tells the other
+parts of the input systems what it is - what events can be generated or
+accepted by this input device. Our example device can only generate EV_KEY
+type events, and from those only BTN_0 event code. Thus we only set these
+two bits. We could have used::
+
+	set_bit(EV_KEY, button_dev->evbit);
+	set_bit(BTN_0, button_dev->keybit);
+
+as well, but with more than single bits the first approach tends to be
+shorter.
+
+Then the example driver registers the input device structure by calling::
+
+	input_register_device(button_dev);
+
+This adds the button_dev structure to linked lists of the input driver and
+calls device handler modules _connect functions to tell them a new input
+device has appeared. input_register_device() may sleep and therefore must
+not be called from an interrupt or with a spinlock held.
+
+While in use, the only used function of the driver is::
+
+	button_interrupt()
+
+which upon every interrupt from the button checks its state and reports it
+via the::
+
+	input_report_key()
+
+call to the input system. There is no need to check whether the interrupt
+routine isn't reporting two same value events (press, press for example) to
+the input system, because the input_report_* functions check that
+themselves.
+
+Then there is the::
+
+	input_sync()
+
+call to tell those who receive the events that we've sent a complete report.
+This doesn't seem important in the one button case, but is quite important
+for example for mouse movement, where you don't want the X and Y values
+to be interpreted separately, because that'd result in a different movement.
+
+dev->open() and dev->close()
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+In case the driver has to repeatedly poll the device, because it doesn't
+have an interrupt coming from it and the polling is too expensive to be done
+all the time, or if the device uses a valuable resource (e.g. interrupt), it
+can use the open and close callback to know when it can stop polling or
+release the interrupt and when it must resume polling or grab the interrupt
+again. To do that, we would add this to our example driver::
+
+    static int button_open(struct input_dev *dev)
+    {
+	    if (request_irq(BUTTON_IRQ, button_interrupt, 0, "button", NULL)) {
+		    printk(KERN_ERR "button.c: Can't allocate irq %d\n", button_irq);
+		    return -EBUSY;
+	    }
+
+	    return 0;
+    }
+
+    static void button_close(struct input_dev *dev)
+    {
+	    free_irq(IRQ_AMIGA_VERTB, button_interrupt);
+    }
+
+    static int __init button_init(void)
+    {
+	    ...
+	    button_dev->open = button_open;
+	    button_dev->close = button_close;
+	    ...
+    }
+
+Note that input core keeps track of number of users for the device and
+makes sure that dev->open() is called only when the first user connects
+to the device and that dev->close() is called when the very last user
+disconnects. Calls to both callbacks are serialized.
+
+The open() callback should return a 0 in case of success or any non-zero value
+in case of failure. The close() callback (which is void) must always succeed.
+
+Inhibiting input devices
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Inhibiting a device means ignoring input events from it. As such it is about
+maintaining relationships with input handlers - either already existing
+relationships, or relationships to be established while the device is in
+inhibited state.
+
+If a device is inhibited, no input handler will receive events from it.
+
+The fact that nobody wants events from the device is exploited further, by
+calling device's close() (if there are users) and open() (if there are users) on
+inhibit and uninhibit operations, respectively. Indeed, the meaning of close()
+is to stop providing events to the input core and that of open() is to start
+providing events to the input core.
+
+Calling the device's close() method on inhibit (if there are users) allows the
+driver to save power. Either by directly powering down the device or by
+releasing the runtime-PM reference it got in open() when the driver is using
+runtime-PM.
+
+Inhibiting and uninhibiting are orthogonal to opening and closing the device by
+input handlers. Userspace might want to inhibit a device in anticipation before
+any handler is positively matched against it.
+
+Inhibiting and uninhibiting are orthogonal to device's being a wakeup source,
+too. Being a wakeup source plays a role when the system is sleeping, not when
+the system is operating.  How drivers should program their interaction between
+inhibiting, sleeping and being a wakeup source is driver-specific.
+
+Taking the analogy with the network devices - bringing a network interface down
+doesn't mean that it should be impossible be wake the system up on LAN through
+this interface. So, there may be input drivers which should be considered wakeup
+sources even when inhibited. Actually, in many I2C input devices their interrupt
+is declared a wakeup interrupt and its handling happens in driver's core, which
+is not aware of input-specific inhibit (nor should it be).  Composite devices
+containing several interfaces can be inhibited on a per-interface basis and e.g.
+inhibiting one interface shouldn't affect the device's capability of being a
+wakeup source.
+
+If a device is to be considered a wakeup source while inhibited, special care
+must be taken when programming its suspend(), as it might need to call device's
+open(). Depending on what close() means for the device in question, not
+opening() it before going to sleep might make it impossible to provide any
+wakeup events. The device is going to sleep anyway.
+
+Basic event types
+~~~~~~~~~~~~~~~~~
+
+The most simple event type is EV_KEY, which is used for keys and buttons.
+It's reported to the input system via::
+
+	input_report_key(struct input_dev *dev, int code, int value)
+
+See uapi/linux/input-event-codes.h for the allowable values of code (from 0 to
+KEY_MAX). Value is interpreted as a truth value, i.e. any non-zero value means
+key pressed, zero value means key released. The input code generates events only
+in case the value is different from before.
+
+In addition to EV_KEY, there are two more basic event types: EV_REL and
+EV_ABS. They are used for relative and absolute values supplied by the
+device. A relative value may be for example a mouse movement in the X axis.
+The mouse reports it as a relative difference from the last position,
+because it doesn't have any absolute coordinate system to work in. Absolute
+events are namely for joysticks and digitizers - devices that do work in an
+absolute coordinate systems.
+
+Having the device report EV_REL buttons is as simple as with EV_KEY; simply
+set the corresponding bits and call the::
+
+	input_report_rel(struct input_dev *dev, int code, int value)
+
+function. Events are generated only for non-zero values.
+
+However EV_ABS requires a little special care. Before calling
+input_register_device, you have to fill additional fields in the input_dev
+struct for each absolute axis your device has. If our button device had also
+the ABS_X axis::
+
+	button_dev.absmin[ABS_X] = 0;
+	button_dev.absmax[ABS_X] = 255;
+	button_dev.absfuzz[ABS_X] = 4;
+	button_dev.absflat[ABS_X] = 8;
+
+Or, you can just say::
+
+	input_set_abs_params(button_dev, ABS_X, 0, 255, 4, 8);
+
+This setting would be appropriate for a joystick X axis, with the minimum of
+0, maximum of 255 (which the joystick *must* be able to reach, no problem if
+it sometimes reports more, but it must be able to always reach the min and
+max values), with noise in the data up to +- 4, and with a center flat
+position of size 8.
+
+If you don't need absfuzz and absflat, you can set them to zero, which mean
+that the thing is precise and always returns to exactly the center position
+(if it has any).
+
+BITS_TO_LONGS(), BIT_WORD(), BIT_MASK()
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+These three macros from bitops.h help some bitfield computations::
+
+	BITS_TO_LONGS(x) - returns the length of a bitfield array in longs for
+			   x bits
+	BIT_WORD(x)	 - returns the index in the array in longs for bit x
+	BIT_MASK(x)	 - returns the index in a long for bit x
+
+The id* and name fields
+~~~~~~~~~~~~~~~~~~~~~~~
+
+The dev->name should be set before registering the input device by the input
+device driver. It's a string like 'Generic button device' containing a
+user friendly name of the device.
+
+The id* fields contain the bus ID (PCI, USB, ...), vendor ID and device ID
+of the device. The bus IDs are defined in input.h. The vendor and device IDs
+are defined in pci_ids.h, usb_ids.h and similar include files. These fields
+should be set by the input device driver before registering it.
+
+The idtype field can be used for specific information for the input device
+driver.
+
+The id and name fields can be passed to userland via the evdev interface.
+
+The keycode, keycodemax, keycodesize fields
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+These three fields should be used by input devices that have dense keymaps.
+The keycode is an array used to map from scancodes to input system keycodes.
+The keycode max should contain the size of the array and keycodesize the
+size of each entry in it (in bytes).
+
+Userspace can query and alter current scancode to keycode mappings using
+EVIOCGKEYCODE and EVIOCSKEYCODE ioctls on corresponding evdev interface.
+When a device has all 3 aforementioned fields filled in, the driver may
+rely on kernel's default implementation of setting and querying keycode
+mappings.
+
+dev->getkeycode() and dev->setkeycode()
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+getkeycode() and setkeycode() callbacks allow drivers to override default
+keycode/keycodesize/keycodemax mapping mechanism provided by input core
+and implement sparse keycode maps.
+
+Key autorepeat
+~~~~~~~~~~~~~~
+
+... is simple. It is handled by the input.c module. Hardware autorepeat is
+not used, because it's not present in many devices and even where it is
+present, it is broken sometimes (at keyboards: Toshiba notebooks). To enable
+autorepeat for your device, just set EV_REP in dev->evbit. All will be
+handled by the input system.
+
+Other event types, handling output events
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The other event types up to now are:
+
+- EV_LED - used for the keyboard LEDs.
+- EV_SND - used for keyboard beeps.
+
+They are very similar to for example key events, but they go in the other
+direction - from the system to the input device driver. If your input device
+driver can handle these events, it has to set the respective bits in evbit,
+*and* also the callback routine::
+
+    button_dev->event = button_event;
+
+    int button_event(struct input_dev *dev, unsigned int type,
+		     unsigned int code, int value)
+    {
+	    if (type == EV_SND && code == SND_BELL) {
+		    outb(value, BUTTON_BELL);
+		    return 0;
+	    }
+	    return -1;
+    }
+
+This callback routine can be called from an interrupt or a BH (although that
+isn't a rule), and thus must not sleep, and must not take too long to finish.
+
+Polled input devices
+~~~~~~~~~~~~~~~~~~~~
+
+Input polling is set up by passing an input device struct and a callback to
+the function::
+
+    int input_setup_polling(struct input_dev *dev,
+        void (*poll_fn)(struct input_dev *dev))
+
+Within the callback, devices should use the regular input_report_* functions
+and input_sync as is used by other devices.
+
+There is also the function::
+
+    void input_set_poll_interval(struct input_dev *dev, unsigned int interval)
+
+which is used to configure the interval, in milliseconds, that the device will
+be polled at.