Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Fix error output of boot instance creation failure

   Currently if a boot instance creation fails, instead of printing out
   the name of the instance that failed, it prints "(null)". That is
   because it prints "cur_str" that had already been processed by
   strsep(). Print the saved name instead.

   While at it, print the error code of the failure.

 - Fix use-after-free for same named historgrams

   Histograms can be named so that they can be used in multiple events.
   But if the named histogram has a variable attached, the second event
   that uses the named histogram which duplicates it and needs to free
   the original after duplication leaves the old variable in place and
   still visible. If another histogram uses than variable, it will use
   the stale one which will try to reference the freed duplicate
   histogram and crash the kernel.

   Free the duplicate variables along with the duplicated histogram
   data.

 - Check return value of kthread_run() in event self test

   The events self tests uses a kthread for testing but does not check
   if it succeeded in creating a kthread. If the kthread creation were
   to fail, the code will still try to call kthread_stop() on the error
   returned.

 - Fix race between reading trace_pipe and updating subbuffer size

   If a user is reading the trace_pipe file at the same time they update
   the ring buffer sub-buffer size, can cause the trace_pipe read to
   read stale data. Add trace_access_lock() around updating the ring
   buffer sub-buffer size.

 - Fix eventfs_inode on failure path in creation of the events directory

   In the creation of the "events" directory, if after allocating the
   eventfs_inode a failure is detected, it calls cleanup_ei() which
   calls free_ei(). The free_ei() will test if eventfs_inode being freed
   has no children. It is a bug if it does. But on the failure case of
   the creation of the "events" directory, the children lists have not
   yet been initialized and the free will trigger a warning because
   list_empty() on an uninitialized list returns false.

   Move the initialization into init_ei() where it makes more sense and
   makes sure that a created eventfs_inode has its lists initialized
   upon creation.

 - Check return value of kthread_run() in ftrace direct sample code

   The sample code that shows how to use the ftrace direct calls does
   not test the return of kthread_run() to see if it succeeds. Return a
   failure if the kthread_run() doesn't succeed.

 - Clear user events state on fork in case of alloc failure

   On fork, the child gets a pointer to the parent's user events state.
   It makes a copy of it then updates the child's pointer to it. But if
   the allocation fails, the duplication function leaves the child with
   a pointer to its parent's descriptor. When the child cleans up its
   data, it will free the parent's descriptor while the parent is still
   using it.

   In the duplication function, set the child's user_event_mm to NULL
   before testing if the allocation succeeded, and when it exits it will
   not free the parent's descriptor.

 - Fix retry exhaustion in simple ring buffer reader swap

   simple_ring_buffer_swap_reader_page() starts with retry set to 8 and
   post-decrements it only after a failed link replacement. On the final
   attempt, a successful replacement leaves retry at zero, while a
   failed replacement leaves it at -1.

   But the check for success expects the retry value to be non-zero and
   exits with an error on zero. This is the opposite result. Fix it.

 - Fail nicely when the remote swap_reader_page() returns an error

   Currently, if the swap_reader_page() of a remote buffer fails, it
   triggers a WARN_ON_ONCE() and continues normally. Instead, have it
   exit with an error and a pr_warn() print instead of a full WARNING.

* tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Stop remote reader update when page swap fails
  tracing: Fix retry exhaustion in simple ring buffer reader swap
  tracing/user_events: Clear copied tracing state before fork duplication
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify
  eventfs: Initialize ei->children and ei->list in init_ei()
  tracing: Fix use-after-free in trace_pipe read on sub-buffer order change
  tracing: Fix crash passing ERR_PTR to kthread_stop()
  tracing: Fix use-after-free with same-name named triggers
  tracing: Fix logged instance name on creation failure

1 files changed, 265 insertions, 0 deletions

diff --git a/Documentation/input/ff.rst b/Documentation/input/ff.rst
new file mode 100644
index 000000000..5a1da42c3
--- /dev/null
+++ b/Documentation/input/ff.rst
@@ -0,0 +1,265 @@
+========================
+Force feedback for Linux
+========================
+
+:Author: Johann Deneux <johann.deneux@gmail.com> on 2001/04/22.
+:Updated: Anssi Hannula <anssi.hannula@gmail.com> on 2006/04/09.
+
+You may redistribute this file. Please remember to include shape.svg and
+interactive.svg as well.
+
+Introduction
+~~~~~~~~~~~~
+
+This document describes how to use force feedback devices under Linux. The
+goal is not to support these devices as if they were simple input-only devices
+(as it is already the case), but to really enable the rendering of force
+effects.
+This document only describes the force feedback part of the Linux input
+interface. Please read joydev/joystick.rst and input.rst before reading further
+this document.
+
+Instructions to the user
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+To enable force feedback, you have to:
+
+1. have your kernel configured with evdev and a driver that supports your
+   device.
+2. make sure evdev module is loaded and /dev/input/event* device files are
+   created.
+
+Before you start, let me WARN you that some devices shake violently during the
+initialisation phase. This happens for example with my "AVB Top Shot Pegasus".
+To stop this annoying behaviour, move your joystick to its limits. Anyway, you
+should keep a hand on your device, in order to avoid it to break down if
+something goes wrong.
+
+If you have a serial iforce device, you need to start inputattach. See
+joydev/joystick.rst for details.
+
+Does it work ?
+--------------
+
+There is an utility called fftest that will allow you to test the driver::
+
+    % fftest /dev/input/eventXX
+
+Instructions to the developer
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+All interactions are done using the event API. That is, you can use ioctl()
+and write() on /dev/input/eventXX.
+This information is subject to change.
+
+Querying device capabilities
+----------------------------
+
+::
+
+    #include <linux/input.h>
+    #include <sys/ioctl.h>
+
+    #define BITS_TO_LONGS(x) \
+	    (((x) + 8 * sizeof (unsigned long) - 1) / (8 * sizeof (unsigned long)))
+    unsigned long features[BITS_TO_LONGS(FF_CNT)];
+    int ioctl(int file_descriptor, int request, unsigned long *features);
+
+"request" must be EVIOCGBIT(EV_FF, size of features array in bytes )
+
+Returns the features supported by the device. features is a bitfield with the
+following bits:
+
+- FF_CONSTANT	can render constant force effects
+- FF_PERIODIC	can render periodic effects with the following waveforms:
+
+  - FF_SQUARE	  square waveform
+  - FF_TRIANGLE	  triangle waveform
+  - FF_SINE	  sine waveform
+  - FF_SAW_UP	  sawtooth up waveform
+  - FF_SAW_DOWN	  sawtooth down waveform
+  - FF_CUSTOM	  custom waveform
+
+- FF_RAMP       can render ramp effects
+- FF_SPRING	can simulate the presence of a spring
+- FF_FRICTION	can simulate friction
+- FF_DAMPER	can simulate damper effects
+- FF_RUMBLE	rumble effects
+- FF_INERTIA    can simulate inertia
+- FF_GAIN	gain is adjustable
+- FF_AUTOCENTER	autocenter is adjustable
+
+.. note::
+
+    - In most cases you should use FF_PERIODIC instead of FF_RUMBLE. All
+      devices that support FF_RUMBLE support FF_PERIODIC (square, triangle,
+      sine) and the other way around.
+
+    - The exact syntax FF_CUSTOM is undefined for the time being as no driver
+      supports it yet.
+
+::
+
+    int ioctl(int fd, EVIOCGEFFECTS, int *n);
+
+Returns the number of effects the device can keep in its memory.
+
+Uploading effects to the device
+-------------------------------
+
+::
+
+    #include <linux/input.h>
+    #include <sys/ioctl.h>
+
+    int ioctl(int file_descriptor, int request, struct ff_effect *effect);
+
+"request" must be EVIOCSFF.
+
+"effect" points to a structure describing the effect to upload. The effect is
+uploaded, but not played.
+The content of effect may be modified. In particular, its field "id" is set
+to the unique id assigned by the driver. This data is required for performing
+some operations (removing an effect, controlling the playback).
+The "id" field must be set to -1 by the user in order to tell the driver to
+allocate a new effect.
+
+Effects are file descriptor specific.
+
+See <uapi/linux/input.h> for a description of the ff_effect struct.  You
+should also find help in a few sketches, contained in files shape.svg
+and interactive.svg:
+
+.. kernel-figure:: shape.svg
+
+    Shape
+
+.. kernel-figure:: interactive.svg
+
+    Interactive
+
+
+Removing an effect from the device
+----------------------------------
+
+::
+
+    int ioctl(int fd, EVIOCRMFF, effect.id);
+
+This makes room for new effects in the device's memory. Note that this also
+stops the effect if it was playing.
+
+Controlling the playback of effects
+-----------------------------------
+
+Control of playing is done with write(). Below is an example:
+
+::
+
+    #include <linux/input.h>
+    #include <unistd.h>
+
+	struct input_event play;
+	struct input_event stop;
+	struct ff_effect effect;
+	int fd;
+   ...
+	fd = open("/dev/input/eventXX", O_RDWR);
+   ...
+	/* Play three times */
+	play.type = EV_FF;
+	play.code = effect.id;
+	play.value = 3;
+
+	write(fd, (const void*) &play, sizeof(play));
+   ...
+	/* Stop an effect */
+	stop.type = EV_FF;
+	stop.code = effect.id;
+	stop.value = 0;
+
+	write(fd, (const void*) &stop, sizeof(stop));
+
+Setting the gain
+----------------
+
+Not all devices have the same strength. Therefore, users should set a gain
+factor depending on how strong they want effects to be. This setting is
+persistent across access to the driver.
+
+::
+
+    /* Set the gain of the device
+    int gain;		/* between 0 and 100 */
+    struct input_event ie;	/* structure used to communicate with the driver */
+
+    ie.type = EV_FF;
+    ie.code = FF_GAIN;
+    ie.value = 0xFFFFUL * gain / 100;
+
+    if (write(fd, &ie, sizeof(ie)) == -1)
+	perror("set gain");
+
+Enabling/Disabling autocenter
+-----------------------------
+
+The autocenter feature quite disturbs the rendering of effects in my opinion,
+and I think it should be an effect, which computation depends on the game
+type. But you can enable it if you want.
+
+::
+
+    int autocenter;		/* between 0 and 100 */
+    struct input_event ie;
+
+    ie.type = EV_FF;
+    ie.code = FF_AUTOCENTER;
+    ie.value = 0xFFFFUL * autocenter / 100;
+
+    if (write(fd, &ie, sizeof(ie)) == -1)
+	perror("set auto-center");
+
+A value of 0 means "no auto-center".
+
+Dynamic update of an effect
+---------------------------
+
+Proceed as if you wanted to upload a new effect, except that instead of
+setting the id field to -1, you set it to the wanted effect id.
+Normally, the effect is not stopped and restarted. However, depending on the
+type of device, not all parameters can be dynamically updated. For example,
+the direction of an effect cannot be updated with iforce devices. In this
+case, the driver stops the effect, up-load it, and restart it.
+
+Therefore it is recommended to dynamically change direction while the effect
+is playing only when it is ok to restart the effect with a replay count of 1.
+
+Information about the status of effects
+---------------------------------------
+
+Every time the status of an effect is changed, an event is sent. The values
+and meanings of the fields of the event are as follows::
+
+    struct input_event {
+    /* When the status of the effect changed */
+	    struct timeval time;
+
+    /* Set to EV_FF_STATUS */
+	    unsigned short type;
+
+    /* Contains the id of the effect */
+	    unsigned short code;
+
+    /* Indicates the status */
+	    unsigned int value;
+    };
+
+    FF_STATUS_STOPPED	The effect stopped playing
+    FF_STATUS_PLAYING	The effect started to play
+
+.. note::
+
+    - Status feedback is only supported by iforce driver. If you have
+      a really good reason to use this, please contact
+      linux-joystick@atrey.karlin.mff.cuni.cz or anssi.hannula@gmail.com
+      so that support for it can be added to the rest of the drivers.