Add two quirks to make it possible for usbhid module options to
override whether a device is ignored (HID_QUIRK_NO_IGNORE) and
whether to connect a hiddev device (HID_QUIRK_HIDDEV_FORCE).

Passing HID_QUIRK_NO_IGNORE for your device means that it will
not be ignored by the HID layer, even if present in a blacklist.

HID_QUIRK_HIDDEV_FORCE will force the creation of a hiddev for that
device, making it accessible from user-space.

Tested with an Apple IR Receiver, switching it from using appleir
to using lirc's macmini driver.
Signed-off-by: NBastien Nocera <hadess@hadess.net>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch (as1365) enables remote wakeup by default for USB keyboard
devices.  Keyboards in general are supposed to be wakeup devices, but
the correct place to enable it depends on the device's bus; no single
approach will work for all keyboard devices.  In particular, this
covers only USB keyboards (and then only those supporting the boot
protocol).
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Move the initialization of USB interface pointers from _start()
over to _probe() callback, which is where it belongs.

This fixes case where interface is NULL when parsing of report
descriptor fails.

LKML-Reference: <20100213135720.603e5f64@neptune.home>
Reported-by: NAlan Stern <stern@rowland.harvard.edu>
Tested-by: NBruno Prémont <bonbons@linux-vserver.org>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Some devices do not react to a control request (seen on APC UPS's) resulting in
a slow stream of messages, "generic-usb ... control queue full".  Therefore
request needs a timeout.

Cc: stable@kernel.org
Signed-off-by: NOliver Neukum <oliver@neukum.org>
Signed-off-by: NDavid Fries <david@fries.net>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

In commit 2da31939 ("Bluetooth: Implement raw output support for HIDP
layer"), support for Bluetooth hid_output_raw_report was added, but it
pushes the data to the intr socket instead of the ctrl one. This has been
fixed by 6bf8268f ("Bluetooth: Use the control channel for raw HID reports")

Still, it is necessary to distinguish whether the report in question should be
either FEATURE or OUTPUT. For this, we have to extend the generic HID API,
so that hid_output_raw_report() callback provides means to specify this
value so that it can be passed down to lower level hardware drivers (currently
Bluetooth and USB).

Based on original patch by Bastien Nocera <hadess@hadess.net>
Acked-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Signed-off-by: NJiri Kosina <jkosina@suse.cz>

The id_table field of the struct usb_device_id is constant in <linux/usb.h>
so it is worth to make the initialization data also constant.

The semantic match that finds this kind of pattern is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@r@
disable decl_init,const_decl_init;
identifier I1, I2, x;
@@
	struct I1 {
	  ...
	  const struct I2 *x;
	  ...
	};
@s@
identifier r.I1, y;
identifier r.x, E;
@@
	struct I1 y = {
	  .x = E,
	};
@c@
identifier r.I2;
identifier s.E;
@@
	const struct I2 E[] = ... ;
@depends on !c@
identifier r.I2;
identifier s.E;
@@
+	const
	struct I2 E[] = ...;
// </smpl>
Signed-off-by: NMárton Németh <nm127@freemail.hu>
Cc: Julia Lawall <julia@diku.dk>
Cc: cocci@diku.dk
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch (as1302) removes the auto_pm flag from struct usb_device.
The flag's only purpose was to distinguish between autosuspends and
external suspends, but that information is now available in the
pm_message_t argument passed to suspend methods.
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

NCR devices are terminally broken by design -- they claim themselves to contain
proper input applications in their HID report descriptor, but behave very badly
if treated in standard way.

According to NCR developers, the devices get confused when queried for reports
in a standard way, rendering them unusable.

NCR is shipping application called "RPSL" that can be used to drive these
devices through hiddev, under the assumption that in-kernel driver doesn't
perform initial report query.
If it does, neither in-kernel nor hiddev-based driver can operate with these
devices any more.

Introduce a quirk that skips the report query for all NCR devices. The previous
NOGET quirk was wrong and had been introduced because I misunderstood the nature
of brokenness of these devices.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

DRIVER_VERSION has no use whatosoever, it has been set to "2.6"
for ages. Remove it.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Remove unused (in usbhid module) DRIVER_AUTHOR macrco and properly
use multiple MODULE_AUTHOR() instances in both modules.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

HID core registers input, hidraw and hiddev devices, but leaves
unregistering it up to the individual driver, which is not really nice.
Let's move all the logic to the core.
Reported-by: NMarcel Holtmann <marcel@holtmann.org>
Reported-by: NBrian Rogers <brian@xyzw.org>
Acked-by: NMarcel Holtmann <marcel@holtmann.org>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

__usbhid_submit_report() is a local function wrapped by the exported
symbol usbhid_submit_report(). As such, it should be static.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc:  Jiri Kosina <jkosina@suse.cz>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Avoid double spin_lock_init on usbhid->lock.
Signed-off-by: NSergey Senozhatsky <sergey.senozhatsky@mail.by>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

* Remove smp_lock.h from files which don't need it (including some headers!)
* Add smp_lock.h to files which do need it
* Make smp_lock.h include conditional in hardirq.h
  It's needed only for one kernel_locked() usage which is under CONFIG_PREEMPT

  This will make hardirq.h inclusion cheaper for every PREEMPT=n config
  (which includes allmodconfig/allyesconfig, BTW)
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It is a little bit inconvenient for people who have some non-standard
HID hardware (usually violating the HID specification) to have to
recompile kernel with CONFIG_HID_DEBUG to be able to see kernel's perspective
of the HID report descriptor and observe the parsed events. Plus the messages
are then mixed up inconveniently with the rest of the dmesg stuff.

This patch implements /sys/kernel/debug/hid/<device>/rdesc file, which
represents the kernel's view of report descriptor (both the raw report
descriptor data and parsed contents).

With all the device-specific debug data being available through debugfs, there
is no need for keeping CONFIG_HID_DEBUG, as the 'debug' parameter to the
hid module will now only output only driver-specific debugging options, which has
absolutely minimal memory footprint, just a few error messages and one global
flag (hid_debug).

We use the current set of output formatting functions. The ones that need to be
used both for one-shot rdesc seq_file and also for continuous flow of data
(individual reports, as being sent by the device) distinguish according to the
passed seq_file parameter, and if it is NULL, it still output to kernel ringbuffer,
otherwise the corresponding seq_file is used for output.

The format of the output is preserved.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

No more reinitialization is needed in the post reset hook, remove
the FIXME comment.

While at it, clean up whitespaces in the immediate surrounding.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Device-specific quirks are set up correctly in their respective vendor-specific
driver, then get overwritten in usbhid_parse().

This is only issue for device-specific NOGET quirks being set by driver for a
few devices out there.
Signed-off-by: NZoltan Karcagi <zkr@freemail.hu>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch switches usbhid_close() from flush_scheduled_work() to canceling
the outstanding work. This fixes a possible deadlock due to work taking
the mutex usbhid_close() holds. Lockdep reported the problem.
Signed-off-by: NOliver Neukum <oliver@neukum.org>

--
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

When hid quirks were converted to specialized driver, the HID_QUIRK_IGNORE
has been moved completely, as the hid_ignore_list[] has been moved into the
generic code.

However userspace already got used to the possibility that modprobing
usbhid with

	'quirks=vid:pid:0x4'

makes the device ignored by usbhid driver. So keep this quirk flag in place
for backwards compatibility.
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This fixes a use of flush_scheduled_work() in USB HID's reset logic that can
deadlock.
Tested-by: NValdis Kletniks <Valdis.Kletnieks@vt.edu>
Signed-off-by: NOliver Neukum <oliver@neukum.name>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

There is no point of having hid_reset_resume() when CONFIG_PM is
not set, and even the corresponding .reset_resume pointer in hid_driver
struct is properly ifdefed.

Move the definition into the ifdef CONFIG_PM part of the source to avoid

drivers/hid/usbhid/hid-core.c:1337: warning: 'hid_reset_resume' defined but not used
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch fixes a bug caused by reset_resume not changing the internal
status flags for a device that is resumed via reset_resume. To do so the
reset handlers, which correctly assume that a device is awake, can no longer
do all the work of reset_resume handling.
Signed-off-by: NOliver Neukum <oneukum@suse.de>
Tested-by: NLaurent Riffard <laurent.riffard@free.fr>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Fixes this build breakage in case when CONFIG_PM is not set

drivers/hid/usbhid/hid-core.c: In function 'hid_suspend':
drivers/hid/usbhid/hid-core.c:1220: error: 'struct usb_device' has no member named 'auto_pm'
drivers/hid/usbhid/hid-core.c:1245: error: 'struct usb_device' has no member named 'auto_pm'
drivers/hid/usbhid/hid-core.c:1258: error: 'struct usb_device' has no member named 'auto_pm'

by throwing both the hid_suspend() and hid_resume() away completely
in such case, as they won't be used anyway.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This uses the USB busy mechanism for aggessive autosuspend of USB
HID devices. It autosuspends all opened devices supporting remote wakeup
after a timeout unless

- output is being done to the device
- a key is being held down (remote wakeup isn't triggered upon key release)
- LED(s) are lit
- hiddev is opened

As in the current driver closed devices will be autosuspended even if they
don't support remote wakeup.

The patch is quite large because output to devices is done in hard interrupt
context meaning a lot a queuing and locking had to be touched. The LED stuff
has been solved by means of a simple counter. Additions to the generic HID code
could be avoided. In addition it now covers hidraw. It contains an embryonic
version of an API to let the generic HID code tell the lower levels which
capabilities with respect to power management are needed.
Signed-off-by: NOliver Neukum <oneukum@suse.de>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch (as1161) changes the interface to
usb_lock_device_for_reset().  The existing interface is apparently not
very clear, judging from the fact that several of its callers don't
use it correctly.  The new interface always returns 0 for success and
it always requires the caller to unlock the device afterward.

The new routine will not return immediately if it is called while the
driver's probe method is running.  Instead it will wait until the
probe is over and the device has been unlocked.  This shouldn't cause
any problems; I don't know of any cases where drivers call
usb_lock_device_for_reset() during probe.
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Cc: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

We might sleep, so no problem to use GFP_KERNEL.

While at it bring the function to coding style.
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Use usb_endpoint_xfer_int() instead of direct use of constants.
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

This patch (as1146c) makes usbhid automatically call usbhid_set_leds()
for any device that supports the keyboard boot protocol.

In theory this should be perfectly safe.  BIOSes send the LED output
report as part of their normal device initialization, so any keyboard
device supporting the boot protocol has to be able to handle it.

As a side effect, the hid-dell and hid-bright drivers are no longer
needed, and the Logitech keyboard driver can be removed from hid-lg.

CC: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

When computing the maximal buffer size needed, we must take into
account that not only input reports can be numbered.

Pointed out in bugzilla #10467
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

It causes recursive locking warning and is unneeded after
introduction of STARTED flag.

* Resume vs. stop is effectively solved by DISCONNECT flag.
* No problem in suspend vs. start -- urb is submitted even after open
  which is possible after connect which is called after start.
* Resume vs. start solved by STARTED flag.
* Suspend vs. stop -- no problem in killing urb and timer twice.
Reported-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Some devices have no input interrupt endpoint. These won't be handled
by usbhid, but currently they are not refused and reside on hid bus.

Perform this checking earlier so that we refuse to control such
a device early enough (and not pass it to the hid bus at all).
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

`stop' left out usbhid->urb* pointers and so the next `start' thought
it needs to allocate nothing and used the memory pointers previously
pointed to. This led to memory corruption and device malfunction.

Also don't forget to clear disconnect flag on start which was left set
by the previous `stop'.

This fixes

	echo DEVICE > /sys/bus/hid/drivers/DRIVER/unbind
	echo DEVICE > /sys/bus/hid/drivers/DRIVER/bind

failures.
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

When suspending, make sure that the timer is not running
any more.
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Usbhid structure is allocated on start invoked only from probe
of some driver. When there is no driver, the structure is null
and causes null-dereference oopses.

Fix it by allocating the structure on probe and disconnect of
the device itself. Also make sure we won't race between start
and resume or stop and suspend respectively.

References: http://bugzilla.kernel.org/show_bug.cgi?id=11827Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Cc: Johannes Berg <johannes@sipsolutions.net>
Cc: Andreas Schwab <schwab@suse.de>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Add type to the hid structure to distinguish to which device type
(now only mouse) we are talking to. Needed for per device type ignore
list support.

Note: this patch leaves the type as unknown for bluetooth devices,
there is not support for this in the hidp code.
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

There were two stragglers that got missed in the last merge of the HID tree that forgot to change the warn() calls to dev_warn().  This patch fixes them up.
Acked-by: NJiri Kosina <jkosina@suse.cz>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

USB should not be having it's own printk macros, so remove warn() and
use the system-wide standard of dev_warn() wherever possible.  In the
few places that will not work out, use a basic printk().
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

USB should not be having it's own printk macros, so remove info() and
use the system-wide standard of dev_info() wherever possible.
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Commit 8006479c introduced a spinlock in
input_dev->event_lock, which is locked when handling input events.
However, the hid-pidff driver sleeps when handling events as it waits for
reports being sent to the device before changing the report contents
again.
This causes a system lockup when trying to use force feedback with a PID
device, a regression introduced in 2.6.24 and 2.6.23.15.

Fix it by extracting the raw report data from struct hid_report
immediately when hid_submit_report() is called, therefore allowing
drivers to change the contents of struct hid_report immediately without
affecting the already-queued transfer.

In hid-pidff, re-add the removed usbhid_wait_io() to
pidff_erase_effect() instead, to prevent a full report queue from causing
the submission to fail, thus not freeing up device memory.
pidff_erase_effect() is not called while dev->event_lock is held.
Signed-off-by: NAnssi Hannula <anssi.hannula@gmail.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>