When set_config changes a line from input to output debounce is
implicitly disabled, as debounce makes no sense for outputs, but the
debounce period is not being cleared and is still reported in the
line info.
So clear the debounce period when the debouncer is stopped in
edge_detector_stop().
Fixes: 65cff70464 ("gpiolib: cdev: support setting debounce")
Cc: stable@vger.kernel.org
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
The kernel test robot reports the following warning in [1]:
drivers/gpio/gpiolib-cdev.c: In function 'gpio_ioctl':
>>drivers/gpio/gpiolib-cdev.c:1437:1: warning: the frame size of 1040 bytes is larger than 1024 bytes [-Wframe-larger-than=]
Refactor gpio_ioctl() to handle each ioctl in its own helper function
and so reduce the variables stored on the stack to those explicitly
required to service the ioctl at hand.
The lineinfo_get_v1() helper handles both the GPIO_GET_LINEINFO_IOCTL
and GPIO_GET_LINEINFO_WATCH_IOCTL, as per the corresponding v2
implementation - lineinfo_get().
[1] https://lore.kernel.org/lkml/202012270910.VW3qc1ER-lkp@intel.com/
Fixes: aad955842d ("gpiolib: cdev: support GPIO_V2_GET_LINEINFO_IOCTL and GPIO_V2_GET_LINEINFO_WATCH_IOCTL")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
- several refactoring patches of the core gpiolib code
- add support for NXP PCAL9554B/C to gpio-pca953x
- allow probing mockup devices from device tree
- refactoring and improvements to gpio-rcar
- improvements to locking in gpio-tegra
- code shrink in gpiolib devres
- get the irq offset from device tree in gpio-sifive
- major refactoring of gpio-exar
- convert gpio-mvebu pwm access to regmap
- create a new submenu for virtual GPIO drivers
- fix clang fall-through warnings treewide
- minor driver refactoring and tweaks sprinkled all over
-----BEGIN PGP SIGNATURE-----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=RmnM
-----END PGP SIGNATURE-----
Merge tag 'gpio-updates-for-v5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux into devel
gpio updates for v5.11-rc1
- several refactoring patches of the core gpiolib code
- add support for NXP PCAL9554B/C to gpio-pca953x
- allow probing mockup devices from device tree
- refactoring and improvements to gpio-rcar
- improvements to locking in gpio-tegra
- code shrink in gpiolib devres
- get the irq offset from device tree in gpio-sifive
- major refactoring of gpio-exar
- convert gpio-mvebu pwm access to regmap
- create a new submenu for virtual GPIO drivers
- fix clang fall-through warnings treewide
- minor driver refactoring and tweaks sprinkled all over
When reporting the state of a GPIO to userspace, we never check
for the actual validity of the line, meaning we report invalid
lines as being usable. A subsequent request will fail though,
which is an inconsistent behaviour from a userspace perspective.
Instead, let's check for the validity of the line and report it
as used if it is invalid. This allows a tool such as gpioinfo
to report something sensible:
gpiochip3 - 4 lines:
line 0: unnamed unused input active-high
line 1: unnamed kernel input active-high [used]
line 2: unnamed kernel input active-high [used]
line 3: unnamed unused input active-high
In this example, lines 1 and 2 are invalid, and cannot be used by
userspace.
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20201204164739.781812-2-maz@kernel.org
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Using CLOCK_REALTIME as the source for event timestamps is crucial for
some specific applications, particularly those requiring timetamps
relative to a PTP clock, so provide an option to switch the event
timestamp source from the default CLOCK_MONOTONIC to CLOCK_REALTIME.
Note that CLOCK_REALTIME was the default source clock for GPIO until
Linux 5.7 when it was changed to CLOCK_MONOTONIC due to issues with the
shifting of the realtime clock.
Providing this option maintains the CLOCK_MONOTONIC as the default,
while also providing a path forward for those dependent on the pre-5.7
behaviour.
Suggested-by: Jack Winch <sunt.un.morcov@gmail.com>
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Link: https://lore.kernel.org/r/20201014231158.34117-2-warthog618@gmail.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Add GPIO_V2_LINE_FLAG_EDGE_BOTH macro and use it in edge_irq_thread() to
improve readability of edge handling cases.
Suggested-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Use the new compat_need_64bit_alignment_fixup() helper to avoid
ugly ifdeffery in IOCTL compatibility code.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Kent Gibson <warthog618@gmail.com>
Depends-on: 527c412519 ("compat: add a compat_need_64bit_alignment_fixup() helper")
Link: https://lore.kernel.org/r/20201014103315.82662-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
The line.eflags field is shared so document this fact and highlight it
throughout using READ_ONCE() and WRITE_ONCE() accessors.
Also use a local copy of the eflags in edge_irq_thread() to ensure
consistent control flow even if eflags changes. This is only a defensive
measure as edge_irq_thread() is currently disabled when the eflags are
changed.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Core changes:
- The big core change is the updated (v2) userspace character
device API. This corrects badly designed 64-bit alignment around
the line events. We also add the debounce request feature.
This echoes the often quotes passage from Frederick Brooks
"The mythical man-month" to always throw one away, which we
have seen before in things such as V4L2. So we put in a new
one and deprecate and obsolete the old one.
- All example tools in tools/gpio/* are migrated to the new API
to set a good example. The libgpiod userspace library has been
augmented to use this new API pretty much from day 1.
- Some misc API hardening by using strn* function calls has been
added as well.
- Use the simpler IDA interface for GPIO chip instance enumeration.
- Add device core function for counting string arrays in
device properties.
- Provide a generic library function kfree_strarray() that can
be used throughout the kernel.
Driver enhancements:
- The DesignWare dwapb-gpio driver has been enhanced and now
uses the IRQ handling in the gpiolib core.
- The mockup and aggregator drivers have seen some substantial
code clean-up and now use more of the core kernel
inftrastructure.
- Misc cleanups using dev_err_probe().
- The MXC drivers (Freescale/NXP) can now be built modularized,
which makes modularized GKI Android kernels happy.
-----BEGIN PGP SIGNATURE-----
iQIzBAABCAAdFiEElDRnuGcz/wPCXQWMQRCzN7AZXXMFAl+FdjkACgkQQRCzN7AZ
XXMYgQ/+JgpHrp7yS1IkS1KiAxHdeIGnKzloTCQQo1JxYEymAnIeMwo/iWAk5wHu
NeJIEVxD0YzZwoI3BXbnO5Qy/62g1z7Ik8ToIa0TiFMwYxz5a7lqsiHwpBgHa50h
T2N8FRFdslVrhpUYBH4Q9wlfYxTki4FwdTD6aaoFFGcMwIVJXWyaYzE+o+qEUEne
VaPsGoNhRKTdKASP3c6+zbbPonzpZW7s/wvIBQAyBgPxEizlL97RzzX3bSSraoCX
i0NsDLHMe+9twqE064KN+CYu0Cy80etQSQsYcfnstVshMuY9+WC1YdyJqzYMciuQ
CYUIQBeskft86IBlsEU/fNCbV+FeAgrxRW6TJK7Hn+sUWZ5+UGdpJ03UE1hA3jjO
SniwG0vpqvZIkio49B6h51VdjNqVJn+AE8tN3hCzqpFknblXgJOVysD7RS7rNM6D
flV1bCsUYtC6jN43qsGFiRYLE9ml2iUxFFoBQUaAEh+pXgUzPTQqD7aSjyzmE3x2
uapKXgxN0dCNH+tFXij73Ro4bYf4ZTZhx3Z3XoEUNEyJpl8fE1bv1SZ2EykOmK8g
c78fAmT0vG3xYZvK10WZj4zuHV6GlPAYVm/MlhB7QHsrF3wa9vervOuqhEPmp2th
hTsVj/Zlz0SSDLncMQL64B7gbxOmzOYlVRxIkSrDEXUOFU7kiWE=
=8CE2
-----END PGP SIGNATURE-----
Merge tag 'gpio-v5.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO updates from Linus Walleij:
"This time very little driver changes but lots of core changes.
We have some interesting cooperative work for ARM and Intel alike,
making the GPIO subsystem more and more suitable for industrial
systems and the like, in addition to the in-kernel users.
We touch driver core (device properties) and lib/* by adding one
simple string array free function, these are authored by Andy
Shevchenko who is a well known and recognized core helpers maintainers
so this should be fine.
We also see some Android GKI-related modularization in the MXC
drivers.
Core changes:
- The big core change is the updated (v2) userspace character device
API.
This corrects badly designed 64-bit alignment around the line
events. We also add the debounce request feature. This echoes the
often quotes passage from Frederick Brooks "The mythical man-month"
to always throw one away, which we have seen before in things such
as V4L2. So we put in a new one and deprecate and obsolete the old
one.
- All example tools in tools/gpio/* are migrated to the new API to
set a good example. The libgpiod userspace library has been
augmented to use this new API pretty much from day 1.
- Some misc API hardening by using strn* function calls has been
added as well.
- Use the simpler IDA interface for GPIO chip instance enumeration.
- Add device core function for counting string arrays in device
properties.
- Provide a generic library function kfree_strarray() that can be
used throughout the kernel.
Driver enhancements:
- The DesignWare dwapb-gpio driver has been enhanced and now uses the
IRQ handling in the gpiolib core.
- The mockup and aggregator drivers have seen some substantial code
clean-up and now use more of the core kernel inftrastructure.
- Misc cleanups using dev_err_probe().
- The MXC drivers (Freescale/NXP) can now be built modularized, which
makes modularized GKI Android kernels happy"
* tag 'gpio-v5.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio: (73 commits)
gpiolib: Update header block in gpiolib-cdev.h
gpiolib: cdev: switch from kstrdup() to kstrndup()
docs: gpio: add a new document to its index.rst
gpio: pca953x: Add support for the NXP PCAL9554B/C
tools: gpio: add debounce support to gpio-event-mon
tools: gpio: add multi-line monitoring to gpio-event-mon
tools: gpio: port gpio-event-mon to v2 uAPI
tools: gpio: port gpio-hammer to v2 uAPI
tools: gpio: rename nlines to num_lines
tools: gpio: port gpio-watch to v2 uAPI
tools: gpio: port lsgpio to v2 uAPI
gpio: uapi: document uAPI v1 as deprecated
gpiolib: cdev: support setting debounce
gpiolib: cdev: support GPIO_V2_LINE_SET_VALUES_IOCTL
gpiolib: cdev: support GPIO_V2_LINE_SET_CONFIG_IOCTL
gpiolib: cdev: support edge detection for uAPI v2
gpiolib: cdev: support GPIO_V2_GET_LINEINFO_IOCTL and GPIO_V2_GET_LINEINFO_WATCH_IOCTL
gpiolib: cdev: support GPIO_V2_GET_LINE_IOCTL and GPIO_V2_LINE_GET_VALUES_IOCTL
gpiolib: add build option for CDEV v1 ABI
gpiolib: make cdev a build option
...
Use kstrndup() to copy line labels from the userspace provided char
array, rather than ensuring the char array contains a null terminator
and using kstrdup().
Note that the length provided to kstrndup() still assumes that the char
array does contain a null terminator, so the maximum string length is one
less than the array. This is consistent with the previous behaviour.
Suggested-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Link: https://lore.kernel.org/r/20201005070246.20927-1-warthog618@gmail.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
It appears that UML (arch/um) has no compat.h header defined and hence
can't compile a recently provided piece of code in GPIO library.
Disable compat ->read() code in UML case to avoid compilation errors.
While at it, use pattern which is already being used in the kernel elsewhere.
Fixes: 5ad284ab3a ("gpiolib: Fix line event handling in syscall compatible mode")
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20201005131044.87276-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Add support for setting debounce on a line via the GPIO uAPI.
Where debounce is not supported by hardware, a software debounce is
provided.
The implementation of the software debouncer waits for the line to be
stable for the debounce period before determining if a level change,
and a corresponding edge event, has occurred. This provides maximum
protection against glitches, but also introduces a debounce_period
latency to edge events.
The software debouncer is integrated with the edge detection as it
utilises the line interrupt, and integration is simpler than getting
the two to interwork. Where software debounce AND edge detection is
required, the debouncer provides both.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Add support for the GPIO_V2_LINE_SET_VALUES_IOCTL.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Add support for GPIO_V2_LINE_SET_CONFIG_IOCTL, the uAPI v2
line set config ioctl.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Add support for edge detection to lines requested using
GPIO_V2_GET_LINE_IOCTL.
The edge_detector implementation is based on the v1 lineevent
implementation. Unlike the v1 implementation, an overflow of the event
buffer results in discarding older events, rather than the most
recent, so the final event in a burst will correspond to the current
state of the line.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Add support for GPIO_V2_GET_LINEINFO_IOCTL and
GPIO_V2_GET_LINEINFO_WATCH_IOCTL.
The core of this change is the event kfifo switching to contain
struct gpioline_info_changed_v2, instead of v1 as v2 is richer.
The two uAPI versions are mostly independent - other than where they both
provide line info changes via reads on the chip fd. As the info change
structs differ between v1 and v2, the infowatch implementation tracks which
version of the infowatch ioctl, either GPIO_GET_LINEINFO_WATCH_IOCTL or
GPIO_V2_GET_LINEINFO_WATCH_IOCTL, initiates the initial watch and returns
the corresponding info change struct to the read. The version supported
on that fd locks to that version on the first watch request, so subsequent
watches from that process must use the same uAPI version.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Add support for requesting lines using the GPIO_V2_GET_LINE_IOCTL, and
returning their current values using GPIO_V2_LINE_GET_VALUES_IOCTL.
The struct linereq implementation is based on the v1 struct linehandle
implementation.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Replace usage of strncpy() with strscpy() to remove -Wstringop-truncation
warnings.
The structures being populated are zeroed, to prevent stack leakage as
they are returned to userspace, so strscpy() performs the equivalent
function without the warnings.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Set the value of the line info offset in gpio_desc_to_lineinfo(), rather
than relying on it being passed in the info. This makes the function
behave as you would expect from the name - it generates the line info
corresponding to a given GPIO desc.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
The introduced line event handling ABI in the commit
61f922db72 ("gpio: userspace ABI for reading GPIO line events")
missed the fact that 64-bit kernel may serve for 32-bit applications.
In such case the very first check in the lineevent_read() will fail
due to alignment differences.
To workaround this introduce lineevent_get_size() helper which returns actual
size of the structure in user space.
Fixes: 61f922db72 ("gpio: userspace ABI for reading GPIO line events")
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Tested-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Consolidate the cleanup of lineevents, currently duplicated in
lineevent_create and lineevent_release, into a helper function
lineevent_free.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Consolidate the cleanup of linehandles, currently duplicated in
linehandle_create and linehandle_release, into a helper function
linehandle_free.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Remove recalculation of offset from desc, where desc itself was calculated
from offset.
There is no benefit from the desc -> hwgpio conversion in this context.
The only implicit benefit of the offset -> desc -> hwgpio is
the range check in the offset -> desc, but where desc is required you
still get that, and where desc isn't required it is simpler to perform
the range check directly.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Merge separate usage of test_bit/set_bit into test_and_set_bit to remove
the possibility of a race between the test and set.
Similarly test_bit and clear_bit.
In the existing code it is possible for two threads to race past the
test_bit and then set or clear the watch bit, and neither return EBUSY.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Rename priv to cdev to improve readability.
The name "priv" indicates that the object is pointed to by
file->private_data, not what the object is actually is.
As it is always used to point to a struct gpio_chardev_data, renaming
it to cdev is more appropriate.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Replace usage of atomic_notifier_call_chain with
blocking_notifier_call_chain as the notifier function,
lineinfo_changed_notify, calls gpio_desc_to_lineinfo,
which calls pinctrl_gpio_can_use_line, which can sleep.
The chain isn't being called from an atomic context so the
the blocking notifier is a suitable substitute.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Remove pointless decrement of variable, and associated comment.
While i is used subsequently, it is re-initialized so this decrement
serves no purpose.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Rename numdescs to num_descs to be more consistent with the naming of
other counters and improve readability.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Rename 'filep' and 'filp' to 'file' to be consistent with other use
and improve readability.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Refactor the mapping from handle flags to desc flags into a helper
function.
The assign_bit is overkill where it is replacing the set_bit cases, as is
rechecking bits known to be clear in some circumstances, but the DRY
simplification more than makes up for any performance degradation,
especially as this is not a hot path.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Make indentation consistent with other use to improve readability.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Sort the includes of gpiolib-cdev.c to make it easier to identify if a
module is included and to avoid duplication.
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Split the cdev specific functionality out of gpiolib.c and into
gpiolib-cdev.c. This improves the readability and maintainability of both
the cdev and core gpiolib code.
Suggested-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Kent Gibson <warthog618@gmail.com>
Link: https://lore.kernel.org/r/20200616093615.5167-1-warthog618@gmail.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Kent Gibson