linux-toradex.git/drivers/gpio/gpio-mpsse.c, branch v7.2-rc5

gpio: mpsse: drop redundant device reference

2026-03-09T09:15:00+00:00

Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.

Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.

Signed-off-by: Johan Hovold 
Reviewed-by: Linus Walleij 
Link: https://patch.msgid.link/20260305124945.10781-1-johan@kernel.org
Signed-off-by: Bartosz Golaszewski

gpio: mpsse: fix reference leak in gpio_mpsse_probe() error paths

2026-01-02T09:45:37+00:00

The reference obtained by calling usb_get_dev() is not released in the
gpio_mpsse_probe() error paths. Fix that by using device managed helper
functions. Also remove the usb_put_dev() call in the disconnect function
since now it will be released automatically.

Cc: stable@vger.kernel.org
Fixes: c46a74ff05c0 ("gpio: add support for FTDI's MPSSE as GPIO")
Signed-off-by: Abdun Nihaal 
Link: https://lore.kernel.org/r/20251226060414.20785-1-nihaal@cse.iitm.ac.in
Signed-off-by: Bartosz Golaszewski

gpio: mpsse: support bryx radio interface kit

2025-10-21T12:10:46+00:00

This device is powered by an FT232H, which is very similar to the
FT2232H this driver was written for. The key difference is it has only
one MPSSE instead of two. As a result, it presents only one USB
interface to the system, which conveniently "just works" out of the box
with this driver.

The brik exposes only two GPIO lines which are hardware limited to only
be useful in one direction. As a result, I've restricted things on the
driver side to refuse to configure any other lines.

This device, unlike the sealevel device I wrote this driver for
originally, is hotpluggable, which makes for all sorts of weird
edgecases. I've tried my best to stress-test the parts that could go
wrong, but given the new usecase, more heads taking a critical look at
the teardown and synchronization bits on the driver as a whole would be
appreciated.

Signed-off-by: Mary Strodl 
Reviewed-by: Linus Walleij 
Link: https://lore.kernel.org/r/20251014133530.3592716-5-mstrodl@csh.rit.edu
Signed-off-by: Bartosz Golaszewski

gpio: mpsse: add quirk support

2025-10-21T12:10:46+00:00

Builds out a facility for specifying compatible lines directions and
labels for MPSSE-based devices.

* dir_in/out are bitmask of lines that can go in/out. 1 means
  compatible, 0 means incompatible.
* names is an array of line names which will be exposed to userspace.

Also changes the chip label format to include some more useful
information about the device to help identify it from userspace.

Signed-off-by: Mary Strodl 
Reviewed-by: Dan Carpenter 
Reviewed-by: Linus Walleij 
Link: https://lore.kernel.org/r/20251014133530.3592716-4-mstrodl@csh.rit.edu
Signed-off-by: Bartosz Golaszewski

gpio: mpsse: ensure worker is torn down

2025-10-21T12:10:45+00:00

When an IRQ worker is running, unplugging the device would cause a
crash. The sealevel hardware this driver was written for was not
hotpluggable, so I never realized it.

This change uses a spinlock to protect a list of workers, which
it tears down on disconnect.

Signed-off-by: Mary Strodl 
Reviewed-by: Linus Walleij 
Link: https://lore.kernel.org/r/20251014133530.3592716-3-mstrodl@csh.rit.edu
Signed-off-by: Bartosz Golaszewski

gpio: mpsse: propagate error from direction_input

2025-10-21T12:10:45+00:00

Not sure how I missed this, but errors encountered when setting the
direction to input weren't being propagated to the caller.

Signed-off-by: Mary Strodl 
Reviewed-by: Linus Walleij 
Link: https://lore.kernel.org/r/20251014133530.3592716-2-mstrodl@csh.rit.edu
Signed-off-by: Bartosz Golaszewski

treewide: rename GPIO set callbacks back to their original names

2025-08-07T08:07:06+00:00

The conversion of all GPIO drivers to using the .set_rv() and
.set_multiple_rv() callbacks from struct gpio_chip (which - unlike their
predecessors - return an integer and allow the controller drivers to
indicate failures to users) is now complete and the legacy ones have
been removed. Rename the new callbacks back to their original names in
one sweeping change.

Signed-off-by: Bartosz Golaszewski

gpio: mpsse: use new GPIO line value setter callbacks

2025-06-17T09:09:30+00:00

struct gpio_chip now has callbacks for setting line values that return
an integer, allowing to indicate failures. Convert the driver to using
them.

Link: https://lore.kernel.org/r/20250610-gpiochip-set-rv-gpio-v1-6-3a9a3c1472ff@linaro.org
Signed-off-by: Bartosz Golaszewski

gpio: mpsse: Remove usage of the deprecated ida_simple_xx() API

2024-11-22T14:23:31+00:00

ida_alloc() and ida_free() should be preferred to the deprecated
ida_simple_get() and ida_simple_remove().

This is less verbose.

Signed-off-by: Christophe JAILLET 
Link: https://lore.kernel.org/r/2ce706d3242b9d3e4b9c20c0a7d9a8afcf8897ec.1729423829.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Bartosz Golaszewski

gpio: mpsse: Check for error code from devm_mutex_init() call

2024-10-31T12:47:36+00:00

Even if it's not critical, the avoidance of checking the error code
from devm_mutex_init() call today diminishes the point of using devm
variant of it. Tomorrow it may even leak something. Add the missed
check.

Fixes: c46a74ff05c0 ("gpio: add support for FTDI's MPSSE as GPIO")
Reviewed-by: Mary Strodl 
Signed-off-by: Andy Shevchenko 
Link: https://lore.kernel.org/r/20241030174132.2113286-2-andriy.shevchenko@linux.intel.com
Signed-off-by: Bartosz Golaszewski