We should unlock before returning on this error path.

Fixes: 3a762dbd ('[media] Input: synaptics-rmi4 - add support for F54 diagnostics')
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Code obtained from https://raw.githubusercontent.com/mightybigcar/synaptics-rmi4/jf/drivers/input/rmi4/rmi_smbus.c
and updated to match upstream. And fixed to make it work.
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NAndrew Duggan <aduggan@synaptics.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

The rmi4 2D sensor functions report the tool type via
input_mt_report_slot_state(), but the abs parameter bit has not been
set so the tool type is not reported to userspace. This patch set
the ABS_MT_TOOL_TYPE bit.
Signed-off-by: NAndrew Duggan <aduggan@synaptics.com>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Implements reading and setting the dribble bit in F12's control registers.
Signed-off-by: NAndrew Duggan <aduggan@synaptics.com>
Acked-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

The rmi_f11 driver currently disables dribble packets and the palm detect
gesture for all devices. This patch creates a parameter in the 2d sensor
platform data for controlling this functionality on a per device basis.

For more information on dribble packets:
Commit 05ba999f ("HID: rmi: disable dribble packets on Synaptics
touchpads")

For more information on the palm detect gesture:
Commit f097deef ("HID: rmi: disable palm detect gesture when present")
Signed-off-by: NAndrew Duggan <aduggan@synaptics.com>
Reviewed-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Commit 5b65c2a0 ("HID: rmi: check sanity of the incoming report") added
support for handling incomplete HID reports do to the input data being
corrupted in transit. This patch reimplements this functionality in the
function drivers so they can handle getting less valid data then they
expect.
Signed-off-by: NAndrew Duggan <aduggan@synaptics.com>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

The attn IRQ is related to the chip, rather than the transport, so move
all handling of interrupts to the core driver. This also makes sure that
there are no races between interrupts and availability of the resources
used by the core driver.
Signed-off-by: NBjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Signed-off-by: NNick Dyer <nick@shmanahar.org>
Tested-by: NChris Healy <cphealy@gmail.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Signed-off-by: NNick Dyer <nick@shmanahar.org>
Signed-off-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Tested-by: NChris Healy <cphealy@gmail.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

We have encountered some RMI4 firmwares where there are blank pages in
between PDT pages which contain functions. This change makes them
correctly enumerate all functions on the device.

Tested on S7817 (has empty page 2).
Signed-off-by: NNick Dyer <nick@shmanahar.org>
[Tested successfully on S7817 and S7300 Synaptics touch controllers]
Tested-by: NChris Healy <cphealy@gmail.com>
Reviewed-by: NAndrew Duggan <aduggan@synaptics.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

Merge tag 'media/v4.9-3' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media into synaptics-rmi4

Sync up with media tree to bring up latest Synaptics and Atmel changes.

Remove this warning:

drivers/staging/media/bcm2048/radio-bcm2048.c: In function 'bcm2048_set_rds_no_lock':
drivers/staging/media/bcm2048/radio-bcm2048.c:467:6: warning: variable 'err' set but not used [-Wunused-but-set-variable]
  int err;
      ^~~

By returning the error code.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

drivers/media/usb/dvb-usb/pctv452e.c:115:2-3: Unneeded semicolon

 Remove unneeded semicolon.

Generated by: scripts/coccinelle/misc/semicolon.cocci
Signed-off-by: NFengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

While here, remove a dead function calling usb_control_msg().
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

There's no sense on decoding and generating a RC key code if
there was an error on the URB control message.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

Instead of silently ignoring the error, return it.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

There's no sense on decoding and generating a RC key code if
there was an error on the URB control message.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.

On this driver, most of the transfers are OK, but the I2C
one was using stack.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

There are some conditions on this driver that are tested with
BUG_ON() with are not serious enough to hang a machine.

So, just return an error if this happens.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

Add checks to avoid going out of the buffer.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

If something bad happens while an USB control message is
transfered, return an error code.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

If something goes wrong, return an error code, instead of
assuming that everything went fine.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

It is up to the frontend kthread to wait for lock.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

There's no sense on decoding and generating a RC key code if
there was an error on the URB control message.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

Be sure that I2C reads won't use stack by passing
a pointer to the state buffer, that we know it was
allocated via kmalloc, instead of relying on the buffer
allocated by an I2C client.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

dib0700_ctrl_rd() takes a RX and a TX pointer. Be sure that
both will point to a memory allocated via kmalloc().
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

The USB control messages require DMA to work. We cannot pass
a stack-allocated buffer, as it is not warranted that the
stack would be into a DMA enabled area.
Reviewed-by: NPatrick Boettcher <patrick.boettcher@posteo.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>

Instead of sending USB commands for every stats call, collect
them once, when status is updated. As the frontend kthread
will call it on every few seconds, the stats will still be
collected.

Besides reducing the amount of USB/I2C transfers, this also
warrants that all stats will be collected at the same time,
and makes easier to convert it to DVBv5 stats in the future.
Signed-off-by: NMauro Carvalho Chehab <mchehab@s-opensource.com>