2018-09-25 10:08:48 +03:00
// SPDX-License-Identifier: GPL-2.0
2017-05-25 16:08:38 +03:00
# include <linux/bitmap.h>
2008-02-05 09:28:20 +03:00
# include <linux/kernel.h>
# include <linux/module.h>
2009-09-23 03:46:38 +04:00
# include <linux/interrupt.h>
2008-02-05 09:28:20 +03:00
# include <linux/irq.h>
# include <linux/spinlock.h>
2013-02-02 20:29:24 +04:00
# include <linux/list.h>
gpio: sysfs interface
This adds a simple sysfs interface for GPIOs.
/sys/class/gpio
/export ... asks the kernel to export a GPIO to userspace
/unexport ... to return a GPIO to the kernel
/gpioN ... for each exported GPIO #N
/value ... always readable, writes fail for input GPIOs
/direction ... r/w as: in, out (default low); write high, low
/gpiochipN ... for each gpiochip; #N is its first GPIO
/base ... (r/o) same as N
/label ... (r/o) descriptive, not necessarily unique
/ngpio ... (r/o) number of GPIOs; numbered N .. N+(ngpio - 1)
GPIOs claimed by kernel code may be exported by its owner using a new
gpio_export() call, which should be most useful for driver debugging.
Such exports may optionally be done without a "direction" attribute.
Userspace may ask to take over a GPIO by writing to a sysfs control file,
helping to cope with incomplete board support or other "one-off"
requirements that don't merit full kernel support:
echo 23 > /sys/class/gpio/export
... will gpio_request(23, "sysfs") and gpio_export(23);
use /sys/class/gpio/gpio-23/direction to (re)configure it,
when that GPIO can be used as both input and output.
echo 23 > /sys/class/gpio/unexport
... will gpio_free(23), when it was exported as above
The extra D-space footprint is a few hundred bytes, except for the sysfs
resources associated with each exported GPIO. The additional I-space
footprint is about two thirds of the current size of gpiolib (!). Since
no /dev node creation is involved, no "udev" support is needed.
Related changes:
* This adds a device pointer to "struct gpio_chip". When GPIO
providers initialize that, sysfs gpio class devices become children of
that device instead of being "virtual" devices.
* The (few) gpio_chip providers which have such a device node have
been updated.
* Some gpio_chip drivers also needed to update their module "owner"
field ... for which missing kerneldoc was added.
* Some gpio_chips don't support input GPIOs. Those GPIOs are now
flagged appropriately when the chip is registered.
Based on previous patches, and discussion both on and off LKML.
A Documentation/ABI/testing/sysfs-gpio update is ready to submit once this
merges to mainline.
[akpm@linux-foundation.org: a few maintenance build fixes]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Guennadi Liakhovetski <g.liakhovetski@pengutronix.de>
Cc: Greg KH <greg@kroah.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 12:46:07 +04:00
# include <linux/device.h>
# include <linux/err.h>
# include <linux/debugfs.h>
# include <linux/seq_file.h>
# include <linux/gpio.h>
2009-09-23 03:46:38 +04:00
# include <linux/idr.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 11:04:11 +03:00
# include <linux/slab.h>
ACPI / driver core: Store an ACPI device pointer in struct acpi_dev_node
Modify struct acpi_dev_node to contain a pointer to struct acpi_device
associated with the given device object (that is, its ACPI companion
device) instead of an ACPI handle corresponding to it. Introduce two
new macros for manipulating that pointer in a CONFIG_ACPI-safe way,
ACPI_COMPANION() and ACPI_COMPANION_SET(), and rework the
ACPI_HANDLE() macro to take the above changes into account.
Drop the ACPI_HANDLE_SET() macro entirely and rework its users to
use ACPI_COMPANION_SET() instead. For some of them who used to
pass the result of acpi_get_child() directly to ACPI_HANDLE_SET()
introduce a helper routine acpi_preset_companion() doing an
equivalent thing.
The main motivation for doing this is that there are things
represented by struct acpi_device objects that don't have valid
ACPI handles (so called fixed ACPI hardware features, such as
power and sleep buttons) and we would like to create platform
device objects for them and "glue" them to their ACPI companions
in the usual way (which currently is impossible due to the
lack of valid ACPI handles). However, there are more reasons
why it may be useful.
First, struct acpi_device pointers allow of much better type checking
than void pointers which are ACPI handles, so it should be more
difficult to write buggy code using modified struct acpi_dev_node
and the new macros. Second, the change should help to reduce (over
time) the number of places in which the result of ACPI_HANDLE() is
passed to acpi_bus_get_device() in order to obtain a pointer to the
struct acpi_device associated with the given "physical" device,
because now that pointer is returned by ACPI_COMPANION() directly.
Finally, the change should make it easier to write generic code that
will build both for CONFIG_ACPI set and unset without adding explicit
compiler directives to it.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> # on Haswell
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Aaron Lu <aaron.lu@intel.com> # for ATA and SDIO part
2013-11-12 01:41:56 +04:00
# include <linux/acpi.h>
2013-11-16 16:44:52 +04:00
# include <linux/gpio/driver.h>
2014-07-24 22:08:55 +04:00
# include <linux/gpio/machine.h>
2015-10-11 18:34:15 +03:00
# include <linux/pinctrl/consumer.h>
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
# include <linux/cdev.h>
# include <linux/fs.h>
# include <linux/uaccess.h>
2016-05-27 15:24:04 +03:00
# include <linux/compat.h>
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
# include <linux/anon_inodes.h>
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
# include <linux/file.h>
2016-06-02 12:30:15 +03:00
# include <linux/kfifo.h>
# include <linux/poll.h>
# include <linux/timekeeping.h>
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
# include <uapi/linux/gpio.h>
2008-02-05 09:28:20 +03:00
2014-01-08 14:40:54 +04:00
# include "gpiolib.h"
2019-07-17 10:10:01 +03:00
# include "gpiolib-of.h"
2019-07-30 13:43:36 +03:00
# include "gpiolib-acpi.h"
2014-01-08 14:40:54 +04:00
2011-05-20 10:40:19 +04:00
# define CREATE_TRACE_POINTS
# include <trace/events/gpio.h>
2008-02-05 09:28:20 +03:00
2013-10-17 21:21:36 +04:00
/* Implementation infrastructure for GPIO interfaces.
2008-02-05 09:28:20 +03:00
*
2013-10-17 21:21:36 +04:00
* The GPIO programming interface allows for inlining speed - critical
* get / set operations for common cases , so that access to SOC - integrated
* GPIOs can sometimes cost only an instruction or two per bit .
2008-02-05 09:28:20 +03:00
*/
/* When debugging, extend minimal trust to callers and platform code.
* Also emit diagnostic messages that may help initial bringup , when
* board setup or driver bugs are most common .
*
* Otherwise , minimize overhead in what may be bitbanging codepaths .
*/
# ifdef DEBUG
# define extra_checks 1
# else
# define extra_checks 0
# endif
2015-10-20 12:10:38 +03:00
/* Device and char device-related information */
static DEFINE_IDA ( gpio_ida ) ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
static dev_t gpio_devt ;
# define GPIO_DEV_MAX 256 /* 256 GPIO chip devices supported */
static struct bus_type gpio_bus_type = {
. name = " gpio " ,
} ;
2015-10-20 12:10:38 +03:00
2018-05-21 20:57:07 +03:00
/*
* Number of GPIOs to use for the fast path in set array
*/
# define FASTPATH_NGPIO CONFIG_GPIOLIB_FASTPATH_LIMIT
2008-02-05 09:28:20 +03:00
/* gpio_lock prevents conflicts during gpio_desc[] table updates.
* While any GPIO is requested , its gpio_chip is not removable ;
* each GPIO ' s " requested " flag serves as a lock and refcount .
*/
2014-07-01 09:45:15 +04:00
DEFINE_SPINLOCK ( gpio_lock ) ;
2008-02-05 09:28:20 +03:00
2013-10-17 21:21:38 +04:00
static DEFINE_MUTEX ( gpio_lookup_lock ) ;
static LIST_HEAD ( gpio_lookup_list ) ;
2015-10-20 12:10:38 +03:00
LIST_HEAD ( gpio_devices ) ;
2015-05-04 18:23:25 +03:00
2018-04-10 23:30:28 +03:00
static DEFINE_MUTEX ( gpio_machine_hogs_mutex ) ;
static LIST_HEAD ( gpio_machine_hogs ) ;
2015-05-04 18:23:25 +03:00
static void gpiochip_free_hogs ( struct gpio_chip * chip ) ;
2017-11-07 21:15:59 +03:00
static int gpiochip_add_irqchip ( struct gpio_chip * gpiochip ,
2017-12-02 20:11:04 +03:00
struct lock_class_key * lock_key ,
struct lock_class_key * request_key ) ;
2015-05-04 18:23:25 +03:00
static void gpiochip_irqchip_remove ( struct gpio_chip * gpiochip ) ;
2019-10-09 17:34:44 +03:00
static int gpiochip_irqchip_init_hw ( struct gpio_chip * gpiochip ) ;
2016-09-20 15:15:21 +03:00
static int gpiochip_irqchip_init_valid_mask ( struct gpio_chip * gpiochip ) ;
static void gpiochip_irqchip_free_valid_mask ( struct gpio_chip * gpiochip ) ;
2015-05-04 18:23:25 +03:00
2016-03-31 18:11:30 +03:00
static bool gpiolib_initialized ;
2015-05-04 18:23:25 +03:00
2008-02-05 09:28:20 +03:00
static inline void desc_set_label ( struct gpio_desc * d , const char * label )
{
d - > label = label ;
}
2013-02-02 20:29:29 +04:00
/**
2017-07-24 17:57:22 +03:00
* gpio_to_desc - Convert a GPIO number to its descriptor
* @ gpio : global GPIO number
*
* Returns :
* The GPIO descriptor associated with the given GPIO , or % NULL if no GPIO
* with the given number exists in the system .
2013-02-02 20:29:29 +04:00
*/
2013-10-17 21:21:36 +04:00
struct gpio_desc * gpio_to_desc ( unsigned gpio )
2013-02-02 20:29:29 +04:00
{
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev ;
2014-11-19 10:51:27 +03:00
unsigned long flags ;
spin_lock_irqsave ( & gpio_lock , flags ) ;
2015-10-20 12:10:38 +03:00
list_for_each_entry ( gdev , & gpio_devices , list ) {
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
if ( gdev - > base < = gpio & &
gdev - > base + gdev - > ngpio > gpio ) {
2014-11-19 10:51:27 +03:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
return & gdev - > descs [ gpio - gdev - > base ] ;
2014-11-19 10:51:27 +03:00
}
}
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2014-12-02 17:15:05 +03:00
if ( ! gpio_is_valid ( gpio ) )
WARN ( 1 , " invalid GPIO %d \n " , gpio ) ;
2014-11-19 10:51:27 +03:00
return NULL ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpio_to_desc ) ;
2013-02-02 20:29:29 +04:00
2013-09-24 13:54:38 +04:00
/**
2017-07-24 17:57:22 +03:00
* gpiochip_get_desc - get the GPIO descriptor corresponding to the given
* hardware number for this chip
* @ chip : GPIO chip
* @ hwnum : hardware number of the GPIO for this chip
*
* Returns :
* A pointer to the GPIO descriptor or % ERR_PTR ( - EINVAL ) if no GPIO exists
* in the given chip for the specified hardware number .
2013-09-24 13:54:38 +04:00
*/
2014-02-09 12:43:54 +04:00
struct gpio_desc * gpiochip_get_desc ( struct gpio_chip * chip ,
u16 hwnum )
2013-09-24 13:54:38 +04:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_device * gdev = chip - > gpiodev ;
if ( hwnum > = gdev - > ngpio )
2013-12-03 07:31:11 +04:00
return ERR_PTR ( - EINVAL ) ;
2013-09-24 13:54:38 +04:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
return & gdev - > descs [ hwnum ] ;
2013-09-24 13:54:38 +04:00
}
2013-02-02 20:29:29 +04:00
/**
2017-07-24 17:57:22 +03:00
* desc_to_gpio - convert a GPIO descriptor to the integer namespace
* @ desc : GPIO descriptor
*
2013-02-02 20:29:29 +04:00
* This should disappear in the future but is needed since we still
2017-07-24 17:57:22 +03:00
* use GPIO numbers for error messages and sysfs nodes .
*
* Returns :
* The global GPIO number for the GPIO specified by its descriptor .
2013-02-02 20:29:29 +04:00
*/
2013-10-17 21:21:36 +04:00
int desc_to_gpio ( const struct gpio_desc * desc )
2013-02-02 20:29:29 +04:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
return desc - > gdev - > base + ( desc - & desc - > gdev - > descs [ 0 ] ) ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( desc_to_gpio ) ;
2013-02-02 20:29:29 +04:00
2013-10-17 21:21:36 +04:00
/**
* gpiod_to_chip - Return the GPIO chip to which a GPIO descriptor belongs
* @ desc : descriptor to return the chip of
*/
struct gpio_chip * gpiod_to_chip ( const struct gpio_desc * desc )
2013-02-02 20:29:29 +04:00
{
2017-12-21 19:37:30 +03:00
if ( ! desc | | ! desc - > gdev )
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
return NULL ;
return desc - > gdev - > chip ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_to_chip ) ;
2008-02-05 09:28:20 +03:00
2008-04-28 13:14:46 +04:00
/* dynamic allocation of GPIOs, e.g. on a hotplugged device */
static int gpiochip_find_base ( int ngpio )
{
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev ;
2013-02-02 20:29:28 +04:00
int base = ARCH_NR_GPIOS - ngpio ;
2008-04-28 13:14:46 +04:00
2015-10-20 12:10:38 +03:00
list_for_each_entry_reverse ( gdev , & gpio_devices , list ) {
2013-02-02 20:29:28 +04:00
/* found a free space? */
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
if ( gdev - > base + gdev - > ngpio < = base )
2013-02-02 20:29:28 +04:00
break ;
else
/* nope, check the space right before the chip */
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
base = gdev - > base - ngpio ;
2008-04-28 13:14:46 +04:00
}
2013-02-02 20:29:28 +04:00
if ( gpio_is_valid ( base ) ) {
2008-04-28 13:14:46 +04:00
pr_debug ( " %s: found new base at %d \n " , __func__ , base ) ;
2013-02-02 20:29:28 +04:00
return base ;
} else {
pr_err ( " %s: cannot find free range \n " , __func__ ) ;
return - ENOSPC ;
2008-04-28 13:14:47 +04:00
}
}
2013-10-17 21:21:36 +04:00
/**
* gpiod_get_direction - return the current direction of a GPIO
* @ desc : GPIO to get the direction of
*
2018-01-09 14:35:53 +03:00
* Returns 0 for output , 1 for input , or an error code in case of error .
2013-10-17 21:21:36 +04:00
*
* This function may sleep if gpiod_cansleep ( ) is true .
*/
2014-11-25 11:16:31 +03:00
int gpiod_get_direction ( struct gpio_desc * desc )
2012-10-24 18:25:27 +04:00
{
2018-07-11 19:33:19 +03:00
struct gpio_chip * chip ;
unsigned offset ;
2019-07-16 12:11:45 +03:00
int ret ;
2012-10-24 18:25:27 +04:00
2013-02-02 20:29:29 +04:00
chip = gpiod_to_chip ( desc ) ;
offset = gpio_chip_hwgpio ( desc ) ;
2012-10-24 18:25:27 +04:00
if ( ! chip - > get_direction )
2018-07-11 19:33:19 +03:00
return - ENOTSUPP ;
2012-10-24 18:25:27 +04:00
2019-07-16 12:11:45 +03:00
ret = chip - > get_direction ( chip , offset ) ;
if ( ret > 0 ) {
2012-10-24 18:25:27 +04:00
/* GPIOF_DIR_IN, or other positive */
2019-07-16 12:11:45 +03:00
ret = 1 ;
2014-11-25 11:16:31 +03:00
clear_bit ( FLAG_IS_OUT , & desc - > flags ) ;
2012-10-24 18:25:27 +04:00
}
2019-07-16 12:11:45 +03:00
if ( ret = = 0 ) {
2012-10-24 18:25:27 +04:00
/* GPIOF_DIR_OUT */
2014-11-25 11:16:31 +03:00
set_bit ( FLAG_IS_OUT , & desc - > flags ) ;
2012-10-24 18:25:27 +04:00
}
2019-07-16 12:11:45 +03:00
return ret ;
2012-10-24 18:25:27 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_get_direction ) ;
2012-10-24 18:25:27 +04:00
2013-02-02 20:29:24 +04:00
/*
* Add a new chip to the global chips list , keeping the list of chips sorted
2015-11-16 08:02:46 +03:00
* by range ( means [ base , base + ngpio - 1 ] ) order .
2013-02-02 20:29:24 +04:00
*
* Return - EBUSY if the new chip overlaps with some other chip ' s integer
* space .
*/
2015-10-20 12:10:38 +03:00
static int gpiodev_add_to_list ( struct gpio_device * gdev )
2013-02-02 20:29:24 +04:00
{
2016-02-26 17:37:14 +03:00
struct gpio_device * prev , * next ;
2013-02-02 20:29:24 +04:00
2015-10-20 12:10:38 +03:00
if ( list_empty ( & gpio_devices ) ) {
2016-02-26 17:37:14 +03:00
/* initial entry in list */
2015-10-20 12:10:38 +03:00
list_add_tail ( & gdev - > list , & gpio_devices ) ;
2015-12-27 16:36:50 +03:00
return 0 ;
2013-02-02 20:29:24 +04:00
}
2016-02-26 17:37:14 +03:00
next = list_entry ( gpio_devices . next , struct gpio_device , list ) ;
if ( gdev - > base + gdev - > ngpio < = next - > base ) {
/* add before first entry */
list_add ( & gdev - > list , & gpio_devices ) ;
return 0 ;
2013-02-02 20:29:24 +04:00
}
2016-02-26 17:37:14 +03:00
prev = list_entry ( gpio_devices . prev , struct gpio_device , list ) ;
if ( prev - > base + prev - > ngpio < = gdev - > base ) {
/* add behind last entry */
list_add_tail ( & gdev - > list , & gpio_devices ) ;
2016-01-08 00:46:45 +03:00
return 0 ;
2013-02-02 20:29:24 +04:00
}
2016-02-26 17:37:14 +03:00
list_for_each_entry_safe ( prev , next , & gpio_devices , list ) {
/* at the end of the list */
if ( & next - > list = = & gpio_devices )
break ;
2013-02-02 20:29:24 +04:00
2016-02-26 17:37:14 +03:00
/* add between prev and next */
if ( prev - > base + prev - > ngpio < = gdev - > base
& & gdev - > base + gdev - > ngpio < = next - > base ) {
list_add ( & gdev - > list , & prev - > list ) ;
return 0 ;
}
}
dev_err ( & gdev - > dev , " GPIO integer space overlap, cannot add chip \n " ) ;
return - EBUSY ;
2013-02-02 20:29:24 +04:00
}
2017-07-24 17:57:22 +03:00
/*
2015-09-24 02:20:43 +03:00
* Convert a GPIO name to its descriptor
*/
static struct gpio_desc * gpio_name_to_desc ( const char * const name )
{
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev ;
2015-09-24 02:20:43 +03:00
unsigned long flags ;
spin_lock_irqsave ( & gpio_lock , flags ) ;
2015-10-20 12:10:38 +03:00
list_for_each_entry ( gdev , & gpio_devices , list ) {
2015-09-24 02:20:43 +03:00
int i ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
for ( i = 0 ; i ! = gdev - > ngpio ; + + i ) {
struct gpio_desc * desc = & gdev - > descs [ i ] ;
2015-09-24 02:20:43 +03:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
if ( ! desc - > name | | ! name )
2015-09-24 02:20:43 +03:00
continue ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
if ( ! strcmp ( desc - > name , name ) ) {
2015-09-24 02:20:43 +03:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
return desc ;
2015-09-24 02:20:43 +03:00
}
}
}
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
return NULL ;
}
2015-08-14 17:11:00 +03:00
/*
* Takes the names from gc - > names and checks if they are all unique . If they
* are , they are assigned to their gpio descriptors .
*
2015-11-14 11:43:20 +03:00
* Warning if one of the names is already used for a different GPIO .
2015-08-14 17:11:00 +03:00
*/
static int gpiochip_set_desc_names ( struct gpio_chip * gc )
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_device * gdev = gc - > gpiodev ;
2015-08-14 17:11:00 +03:00
int i ;
if ( ! gc - > names )
return 0 ;
/* First check all names if they are unique */
for ( i = 0 ; i ! = gc - > ngpio ; + + i ) {
struct gpio_desc * gpio ;
gpio = gpio_name_to_desc ( gc - > names [ i ] ) ;
2015-09-24 02:20:43 +03:00
if ( gpio )
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
dev_warn ( & gdev - > dev ,
2015-10-20 12:31:54 +03:00
" Detected name collision for GPIO name '%s' \n " ,
2015-09-24 02:20:43 +03:00
gc - > names [ i ] ) ;
2015-08-14 17:11:00 +03:00
}
/* Then add all names to the GPIO descriptors */
for ( i = 0 ; i ! = gc - > ngpio ; + + i )
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
gdev - > descs [ i ] . name = gc - > names [ i ] ;
2015-08-14 17:11:00 +03:00
return 0 ;
}
2018-03-23 19:34:50 +03:00
static unsigned long * gpiochip_allocate_mask ( struct gpio_chip * chip )
{
unsigned long * p ;
2019-07-18 09:51:01 +03:00
p = bitmap_alloc ( chip - > ngpio , GFP_KERNEL ) ;
2018-03-23 19:34:50 +03:00
if ( ! p )
return NULL ;
/* Assume by default all GPIOs are valid */
bitmap_fill ( p , chip - > ngpio ) ;
return p ;
}
2019-07-17 10:10:01 +03:00
static int gpiochip_alloc_valid_mask ( struct gpio_chip * gc )
2018-03-23 19:34:52 +03:00
{
2019-08-19 12:30:58 +03:00
if ( ! ( of_gpio_need_valid_mask ( gc ) | | gc - > init_valid_mask ) )
2018-03-23 19:34:52 +03:00
return 0 ;
2019-07-17 10:10:01 +03:00
gc - > valid_mask = gpiochip_allocate_mask ( gc ) ;
if ( ! gc - > valid_mask )
2018-03-23 19:34:52 +03:00
return - ENOMEM ;
return 0 ;
}
2019-08-19 11:49:04 +03:00
static int gpiochip_init_valid_mask ( struct gpio_chip * gc )
2018-10-05 09:52:58 +03:00
{
2019-08-19 11:49:04 +03:00
if ( gc - > init_valid_mask )
return gc - > init_valid_mask ( gc ,
gc - > valid_mask ,
gc - > ngpio ) ;
2018-10-05 09:52:58 +03:00
return 0 ;
}
2018-03-23 19:34:52 +03:00
static void gpiochip_free_valid_mask ( struct gpio_chip * gpiochip )
{
2019-07-18 09:51:01 +03:00
bitmap_free ( gpiochip - > valid_mask ) ;
2018-03-23 19:34:52 +03:00
gpiochip - > valid_mask = NULL ;
}
bool gpiochip_line_is_valid ( const struct gpio_chip * gpiochip ,
unsigned int offset )
{
/* No mask means all valid */
if ( likely ( ! gpiochip - > valid_mask ) )
return true ;
return test_bit ( offset , gpiochip - > valid_mask ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_line_is_valid ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
/*
* GPIO line handle management
*/
/**
* struct linehandle_state - contains the state of a userspace handle
* @ gdev : the GPIO device the handle pertains to
* @ label : consumer label used to tag descriptors
* @ descs : the GPIO descriptors held by this handle
* @ numdescs : the number of descriptors held in the descs array
*/
struct linehandle_state {
struct gpio_device * gdev ;
const char * label ;
struct gpio_desc * descs [ GPIOHANDLES_MAX ] ;
u32 numdescs ;
} ;
2016-10-18 17:54:05 +03:00
# define GPIOHANDLE_REQUEST_VALID_FLAGS \
( GPIOHANDLE_REQUEST_INPUT | \
GPIOHANDLE_REQUEST_OUTPUT | \
GPIOHANDLE_REQUEST_ACTIVE_LOW | \
2019-09-21 13:25:23 +03:00
GPIOHANDLE_REQUEST_PULL_UP | \
GPIOHANDLE_REQUEST_PULL_DOWN | \
2016-10-18 17:54:05 +03:00
GPIOHANDLE_REQUEST_OPEN_DRAIN | \
GPIOHANDLE_REQUEST_OPEN_SOURCE )
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
static long linehandle_ioctl ( struct file * filep , unsigned int cmd ,
unsigned long arg )
{
struct linehandle_state * lh = filep - > private_data ;
void __user * ip = ( void __user * ) arg ;
struct gpiohandle_data ghd ;
2018-09-06 00:50:05 +03:00
DECLARE_BITMAP ( vals , GPIOHANDLES_MAX ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
int i ;
if ( cmd = = GPIOHANDLE_GET_LINE_VALUES_IOCTL ) {
2018-07-16 11:34:24 +03:00
/* NOTE: It's ok to read values of output lines. */
2017-10-12 13:40:10 +03:00
int ret = gpiod_get_array_value_complex ( false ,
true ,
lh - > numdescs ,
lh - > descs ,
2018-09-06 00:50:07 +03:00
NULL ,
2017-10-12 13:40:10 +03:00
vals ) ;
if ( ret )
return ret ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
2016-10-18 17:54:02 +03:00
memset ( & ghd , 0 , sizeof ( ghd ) ) ;
2017-10-12 13:40:10 +03:00
for ( i = 0 ; i < lh - > numdescs ; i + + )
2018-09-06 00:50:05 +03:00
ghd . values [ i ] = test_bit ( i , vals ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
if ( copy_to_user ( ip , & ghd , sizeof ( ghd ) ) )
return - EFAULT ;
return 0 ;
} else if ( cmd = = GPIOHANDLE_SET_LINE_VALUES_IOCTL ) {
2018-07-16 11:34:23 +03:00
/*
* All line descriptors were created at once with the same
* flags so just check if the first one is really output .
*/
if ( ! test_bit ( FLAG_IS_OUT , & lh - > descs [ 0 ] - > flags ) )
return - EPERM ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
if ( copy_from_user ( & ghd , ip , sizeof ( ghd ) ) )
return - EFAULT ;
/* Clamp all values to [0,1] */
for ( i = 0 ; i < lh - > numdescs ; i + + )
2018-09-06 00:50:05 +03:00
__assign_bit ( i , vals , ghd . values [ i ] ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
/* Reuse the array setting function */
2018-05-21 20:57:07 +03:00
return gpiod_set_array_value_complex ( false ,
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
true ,
lh - > numdescs ,
lh - > descs ,
2018-09-06 00:50:07 +03:00
NULL ,
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
vals ) ;
}
return - EINVAL ;
}
# ifdef CONFIG_COMPAT
static long linehandle_ioctl_compat ( struct file * filep , unsigned int cmd ,
unsigned long arg )
{
return linehandle_ioctl ( filep , cmd , ( unsigned long ) compat_ptr ( arg ) ) ;
}
# endif
static int linehandle_release ( struct inode * inode , struct file * filep )
{
struct linehandle_state * lh = filep - > private_data ;
struct gpio_device * gdev = lh - > gdev ;
int i ;
for ( i = 0 ; i < lh - > numdescs ; i + + )
gpiod_free ( lh - > descs [ i ] ) ;
kfree ( lh - > label ) ;
kfree ( lh ) ;
put_device ( & gdev - > dev ) ;
return 0 ;
}
static const struct file_operations linehandle_fileops = {
. release = linehandle_release ,
. owner = THIS_MODULE ,
. llseek = noop_llseek ,
. unlocked_ioctl = linehandle_ioctl ,
# ifdef CONFIG_COMPAT
. compat_ioctl = linehandle_ioctl_compat ,
# endif
} ;
static int linehandle_create ( struct gpio_device * gdev , void __user * ip )
{
struct gpiohandle_request handlereq ;
struct linehandle_state * lh ;
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
struct file * file ;
2018-03-29 21:29:12 +03:00
int fd , i , count = 0 , ret ;
2017-10-16 12:32:29 +03:00
u32 lflags ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
if ( copy_from_user ( & handlereq , ip , sizeof ( handlereq ) ) )
return - EFAULT ;
if ( ( handlereq . lines = = 0 ) | | ( handlereq . lines > GPIOHANDLES_MAX ) )
return - EINVAL ;
2017-10-16 12:32:29 +03:00
lflags = handlereq . flags ;
/* Return an error if an unknown flag is set */
if ( lflags & ~ GPIOHANDLE_REQUEST_VALID_FLAGS )
return - EINVAL ;
2019-09-09 06:22:18 +03:00
/*
* Do not allow both INPUT & OUTPUT flags to be set as they are
* contradictory .
*/
if ( ( lflags & GPIOHANDLE_REQUEST_INPUT ) & &
( lflags & GPIOHANDLE_REQUEST_OUTPUT ) )
return - EINVAL ;
2017-11-15 18:47:43 +03:00
/*
* Do not allow OPEN_SOURCE & OPEN_DRAIN flags in a single request . If
* the hardware actually supports enabling both at the same time the
* electrical result would be disastrous .
*/
if ( ( lflags & GPIOHANDLE_REQUEST_OPEN_DRAIN ) & &
( lflags & GPIOHANDLE_REQUEST_OPEN_SOURCE ) )
return - EINVAL ;
2017-10-16 12:32:30 +03:00
/* OPEN_DRAIN and OPEN_SOURCE flags only make sense for output mode. */
if ( ! ( lflags & GPIOHANDLE_REQUEST_OUTPUT ) & &
( ( lflags & GPIOHANDLE_REQUEST_OPEN_DRAIN ) | |
( lflags & GPIOHANDLE_REQUEST_OPEN_SOURCE ) ) )
return - EINVAL ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
lh = kzalloc ( sizeof ( * lh ) , GFP_KERNEL ) ;
if ( ! lh )
return - ENOMEM ;
lh - > gdev = gdev ;
get_device ( & gdev - > dev ) ;
/* Make sure this is terminated */
handlereq . consumer_label [ sizeof ( handlereq . consumer_label ) - 1 ] = ' \0 ' ;
if ( strlen ( handlereq . consumer_label ) ) {
lh - > label = kstrdup ( handlereq . consumer_label ,
GFP_KERNEL ) ;
if ( ! lh - > label ) {
ret = - ENOMEM ;
goto out_free_lh ;
}
}
/* Request each GPIO */
for ( i = 0 ; i < handlereq . lines ; i + + ) {
u32 offset = handlereq . lineoffsets [ i ] ;
struct gpio_desc * desc ;
2016-10-18 17:54:01 +03:00
if ( offset > = gdev - > ngpio ) {
ret = - EINVAL ;
goto out_free_descs ;
}
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
desc = & gdev - > descs [ offset ] ;
ret = gpiod_request ( desc , lh - > label ) ;
if ( ret )
goto out_free_descs ;
lh - > descs [ i ] = desc ;
2018-09-13 16:37:04 +03:00
count = i + 1 ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
if ( lflags & GPIOHANDLE_REQUEST_ACTIVE_LOW )
set_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) ;
if ( lflags & GPIOHANDLE_REQUEST_OPEN_DRAIN )
set_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) ;
if ( lflags & GPIOHANDLE_REQUEST_OPEN_SOURCE )
set_bit ( FLAG_OPEN_SOURCE , & desc - > flags ) ;
2019-09-21 13:25:23 +03:00
if ( lflags & GPIOHANDLE_REQUEST_PULL_DOWN )
set_bit ( FLAG_PULL_DOWN , & desc - > flags ) ;
if ( lflags & GPIOHANDLE_REQUEST_PULL_UP )
set_bit ( FLAG_PULL_UP , & desc - > flags ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
2017-11-30 06:55:24 +03:00
ret = gpiod_set_transitory ( desc , false ) ;
if ( ret < 0 )
goto out_free_descs ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
/*
* Lines have to be requested explicitly for input
* or output , else the line will be treated " as is " .
*/
if ( lflags & GPIOHANDLE_REQUEST_OUTPUT ) {
int val = ! ! handlereq . default_values [ i ] ;
ret = gpiod_direction_output ( desc , val ) ;
if ( ret )
goto out_free_descs ;
} else if ( lflags & GPIOHANDLE_REQUEST_INPUT ) {
ret = gpiod_direction_input ( desc ) ;
if ( ret )
goto out_free_descs ;
}
dev_dbg ( & gdev - > dev , " registered chardev handle for line %d \n " ,
offset ) ;
}
2016-06-18 11:56:43 +03:00
/* Let i point at the last handle */
i - - ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
lh - > numdescs = handlereq . lines ;
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
fd = get_unused_fd_flags ( O_RDONLY | O_CLOEXEC ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
if ( fd < 0 ) {
ret = fd ;
goto out_free_descs ;
}
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
file = anon_inode_getfile ( " gpio-linehandle " ,
& linehandle_fileops ,
lh ,
O_RDONLY | O_CLOEXEC ) ;
if ( IS_ERR ( file ) ) {
ret = PTR_ERR ( file ) ;
goto out_put_unused_fd ;
}
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
handlereq . fd = fd ;
2016-07-04 14:13:04 +03:00
if ( copy_to_user ( ip , & handlereq , sizeof ( handlereq ) ) ) {
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
/*
* fput ( ) will trigger the release ( ) callback , so do not go onto
* the regular error cleanup path here .
*/
fput ( file ) ;
put_unused_fd ( fd ) ;
return - EFAULT ;
2016-07-04 14:13:04 +03:00
}
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
fd_install ( fd , file ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
dev_dbg ( & gdev - > dev , " registered chardev handle for %d lines \n " ,
lh - > numdescs ) ;
return 0 ;
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
out_put_unused_fd :
put_unused_fd ( fd ) ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
out_free_descs :
2018-03-29 21:29:12 +03:00
for ( i = 0 ; i < count ; i + + )
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
gpiod_free ( lh - > descs [ i ] ) ;
kfree ( lh - > label ) ;
out_free_lh :
kfree ( lh ) ;
put_device ( & gdev - > dev ) ;
return ret ;
}
2016-06-02 12:30:15 +03:00
/*
* GPIO line event management
*/
/**
* struct lineevent_state - contains the state of a userspace event
* @ gdev : the GPIO device the event pertains to
* @ label : consumer label used to tag descriptors
* @ desc : the GPIO descriptor held by this event
* @ eflags : the event flags this line was requested with
* @ irq : the interrupt that trigger in response to events on this GPIO
* @ wait : wait queue that handles blocking reads of events
* @ events : KFIFO for the GPIO events
* @ read_lock : mutex lock to protect reads from colliding with adding
* new events to the FIFO
2017-11-30 12:23:27 +03:00
* @ timestamp : cache for the timestamp storing it between hardirq
* and IRQ thread , used to bring the timestamp close to the actual
* event
2016-06-02 12:30:15 +03:00
*/
struct lineevent_state {
struct gpio_device * gdev ;
const char * label ;
struct gpio_desc * desc ;
u32 eflags ;
int irq ;
wait_queue_head_t wait ;
DECLARE_KFIFO ( events , struct gpioevent_data , 16 ) ;
struct mutex read_lock ;
2017-11-30 12:23:27 +03:00
u64 timestamp ;
2016-06-02 12:30:15 +03:00
} ;
2016-10-18 17:54:06 +03:00
# define GPIOEVENT_REQUEST_VALID_FLAGS \
( GPIOEVENT_REQUEST_RISING_EDGE | \
GPIOEVENT_REQUEST_FALLING_EDGE )
2017-07-03 13:39:46 +03:00
static __poll_t lineevent_poll ( struct file * filep ,
2016-06-02 12:30:15 +03:00
struct poll_table_struct * wait )
{
struct lineevent_state * le = filep - > private_data ;
2017-07-03 13:39:46 +03:00
__poll_t events = 0 ;
2016-06-02 12:30:15 +03:00
poll_wait ( filep , & le - > wait , wait ) ;
if ( ! kfifo_is_empty ( & le - > events ) )
2018-02-12 01:34:03 +03:00
events = EPOLLIN | EPOLLRDNORM ;
2016-06-02 12:30:15 +03:00
return events ;
}
static ssize_t lineevent_read ( struct file * filep ,
char __user * buf ,
size_t count ,
loff_t * f_ps )
{
struct lineevent_state * le = filep - > private_data ;
unsigned int copied ;
int ret ;
if ( count < sizeof ( struct gpioevent_data ) )
return - EINVAL ;
do {
if ( kfifo_is_empty ( & le - > events ) ) {
if ( filep - > f_flags & O_NONBLOCK )
return - EAGAIN ;
ret = wait_event_interruptible ( le - > wait ,
! kfifo_is_empty ( & le - > events ) ) ;
if ( ret )
return ret ;
}
if ( mutex_lock_interruptible ( & le - > read_lock ) )
return - ERESTARTSYS ;
ret = kfifo_to_user ( & le - > events , buf , count , & copied ) ;
mutex_unlock ( & le - > read_lock ) ;
if ( ret )
return ret ;
/*
* If we couldn ' t read anything from the fifo ( a different
* thread might have been faster ) we either return - EAGAIN if
* the file descriptor is non - blocking , otherwise we go back to
* sleep and wait for more data to arrive .
*/
if ( copied = = 0 & & ( filep - > f_flags & O_NONBLOCK ) )
return - EAGAIN ;
} while ( copied = = 0 ) ;
return copied ;
}
static int lineevent_release ( struct inode * inode , struct file * filep )
{
struct lineevent_state * le = filep - > private_data ;
struct gpio_device * gdev = le - > gdev ;
free_irq ( le - > irq , le ) ;
gpiod_free ( le - > desc ) ;
kfree ( le - > label ) ;
kfree ( le ) ;
put_device ( & gdev - > dev ) ;
return 0 ;
}
static long lineevent_ioctl ( struct file * filep , unsigned int cmd ,
unsigned long arg )
{
struct lineevent_state * le = filep - > private_data ;
void __user * ip = ( void __user * ) arg ;
struct gpiohandle_data ghd ;
/*
* We can get the value for an event line but not set it ,
* because it is input by definition .
*/
if ( cmd = = GPIOHANDLE_GET_LINE_VALUES_IOCTL ) {
int val ;
2016-10-18 17:54:04 +03:00
memset ( & ghd , 0 , sizeof ( ghd ) ) ;
2016-06-02 12:30:15 +03:00
val = gpiod_get_value_cansleep ( le - > desc ) ;
if ( val < 0 )
return val ;
ghd . values [ 0 ] = val ;
if ( copy_to_user ( ip , & ghd , sizeof ( ghd ) ) )
return - EFAULT ;
return 0 ;
}
return - EINVAL ;
}
# ifdef CONFIG_COMPAT
static long lineevent_ioctl_compat ( struct file * filep , unsigned int cmd ,
unsigned long arg )
{
return lineevent_ioctl ( filep , cmd , ( unsigned long ) compat_ptr ( arg ) ) ;
}
# endif
static const struct file_operations lineevent_fileops = {
. release = lineevent_release ,
. read = lineevent_read ,
. poll = lineevent_poll ,
. owner = THIS_MODULE ,
. llseek = noop_llseek ,
. unlocked_ioctl = lineevent_ioctl ,
# ifdef CONFIG_COMPAT
. compat_ioctl = lineevent_ioctl_compat ,
# endif
} ;
2016-06-17 18:03:13 +03:00
static irqreturn_t lineevent_irq_thread ( int irq , void * p )
2016-06-02 12:30:15 +03:00
{
struct lineevent_state * le = p ;
struct gpioevent_data ge ;
2018-08-20 09:32:53 +03:00
int ret ;
2016-06-02 12:30:15 +03:00
2018-01-22 15:19:28 +03:00
/* Do not leak kernel stack to userspace */
memset ( & ge , 0 , sizeof ( ge ) ) ;
2019-01-04 13:24:20 +03:00
/*
* We may be running from a nested threaded interrupt in which case
* we didn ' t get the timestamp from lineevent_irq_handler ( ) .
*/
if ( ! le - > timestamp )
ge . timestamp = ktime_get_real_ns ( ) ;
else
ge . timestamp = le - > timestamp ;
2016-06-02 12:30:15 +03:00
2017-06-23 14:45:16 +03:00
if ( le - > eflags & GPIOEVENT_REQUEST_RISING_EDGE
& & le - > eflags & GPIOEVENT_REQUEST_FALLING_EDGE ) {
2018-08-20 09:32:53 +03:00
int level = gpiod_get_value_cansleep ( le - > desc ) ;
2016-06-02 12:30:15 +03:00
if ( level )
/* Emit low-to-high event */
ge . id = GPIOEVENT_EVENT_RISING_EDGE ;
else
/* Emit high-to-low event */
ge . id = GPIOEVENT_EVENT_FALLING_EDGE ;
2018-08-20 09:32:53 +03:00
} else if ( le - > eflags & GPIOEVENT_REQUEST_RISING_EDGE ) {
2016-06-02 12:30:15 +03:00
/* Emit low-to-high event */
ge . id = GPIOEVENT_EVENT_RISING_EDGE ;
2018-08-20 09:32:53 +03:00
} else if ( le - > eflags & GPIOEVENT_REQUEST_FALLING_EDGE ) {
2016-06-02 12:30:15 +03:00
/* Emit high-to-low event */
ge . id = GPIOEVENT_EVENT_FALLING_EDGE ;
2016-06-16 12:02:41 +03:00
} else {
return IRQ_NONE ;
2016-06-02 12:30:15 +03:00
}
ret = kfifo_put ( & le - > events , ge ) ;
2019-09-07 20:39:10 +03:00
if ( ret )
2018-02-12 01:34:03 +03:00
wake_up_poll ( & le - > wait , EPOLLIN ) ;
2016-06-02 12:30:15 +03:00
return IRQ_HANDLED ;
}
2017-11-30 12:23:27 +03:00
static irqreturn_t lineevent_irq_handler ( int irq , void * p )
{
struct lineevent_state * le = p ;
/*
* Just store the timestamp in hardirq context so we get it as
* close in time as possible to the actual event .
*/
le - > timestamp = ktime_get_real_ns ( ) ;
return IRQ_WAKE_THREAD ;
}
2016-06-02 12:30:15 +03:00
static int lineevent_create ( struct gpio_device * gdev , void __user * ip )
{
struct gpioevent_request eventreq ;
struct lineevent_state * le ;
struct gpio_desc * desc ;
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
struct file * file ;
2016-06-02 12:30:15 +03:00
u32 offset ;
u32 lflags ;
u32 eflags ;
int fd ;
int ret ;
int irqflags = 0 ;
if ( copy_from_user ( & eventreq , ip , sizeof ( eventreq ) ) )
return - EFAULT ;
2019-09-16 12:46:23 +03:00
offset = eventreq . lineoffset ;
lflags = eventreq . handleflags ;
eflags = eventreq . eventflags ;
if ( offset > = gdev - > ngpio )
return - EINVAL ;
/* Return an error if a unknown flag is set */
if ( ( lflags & ~ GPIOHANDLE_REQUEST_VALID_FLAGS ) | |
( eflags & ~ GPIOEVENT_REQUEST_VALID_FLAGS ) )
return - EINVAL ;
/* This is just wrong: we don't look for events on output lines */
if ( ( lflags & GPIOHANDLE_REQUEST_OUTPUT ) | |
( lflags & GPIOHANDLE_REQUEST_OPEN_DRAIN ) | |
( lflags & GPIOHANDLE_REQUEST_OPEN_SOURCE ) )
return - EINVAL ;
2016-06-02 12:30:15 +03:00
le = kzalloc ( sizeof ( * le ) , GFP_KERNEL ) ;
if ( ! le )
return - ENOMEM ;
le - > gdev = gdev ;
get_device ( & gdev - > dev ) ;
/* Make sure this is terminated */
eventreq . consumer_label [ sizeof ( eventreq . consumer_label ) - 1 ] = ' \0 ' ;
if ( strlen ( eventreq . consumer_label ) ) {
le - > label = kstrdup ( eventreq . consumer_label ,
GFP_KERNEL ) ;
if ( ! le - > label ) {
ret = - ENOMEM ;
goto out_free_le ;
}
}
desc = & gdev - > descs [ offset ] ;
ret = gpiod_request ( desc , le - > label ) ;
if ( ret )
2018-04-16 14:17:53 +03:00
goto out_free_label ;
2016-06-02 12:30:15 +03:00
le - > desc = desc ;
le - > eflags = eflags ;
if ( lflags & GPIOHANDLE_REQUEST_ACTIVE_LOW )
set_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) ;
ret = gpiod_direction_input ( desc ) ;
if ( ret )
goto out_free_desc ;
le - > irq = gpiod_to_irq ( desc ) ;
if ( le - > irq < = 0 ) {
ret = - ENODEV ;
goto out_free_desc ;
}
if ( eflags & GPIOEVENT_REQUEST_RISING_EDGE )
gpiolib: fix incorrect IRQ requesting of an active-low lineevent
When a pin is active-low, logical trigger edge should be inverted to match
the same interrupt opportunity.
For example, a button pushed triggers falling edge in ACTIVE_HIGH case; in
ACTIVE_LOW case, the button pushed triggers rising edge. For user space the
IRQ requesting doesn't need to do any modification except to configuring
GPIOHANDLE_REQUEST_ACTIVE_LOW.
For example, we want to catch the event when the button is pushed. The
button on the original board drives level to be low when it is pushed, and
drives level to be high when it is released.
In user space we can do:
req.handleflags = GPIOHANDLE_REQUEST_INPUT;
req.eventflags = GPIOEVENT_REQUEST_FALLING_EDGE;
while (1) {
read(fd, &dat, sizeof(dat));
if (dat.id == GPIOEVENT_EVENT_FALLING_EDGE)
printf("button pushed\n");
}
Run the same logic on another board which the polarity of the button is
inverted; it drives level to be high when pushed, and level to be low when
released. For this inversion we add flag GPIOHANDLE_REQUEST_ACTIVE_LOW:
req.handleflags = GPIOHANDLE_REQUEST_INPUT |
GPIOHANDLE_REQUEST_ACTIVE_LOW;
req.eventflags = GPIOEVENT_REQUEST_FALLING_EDGE;
At the result, there are no any events caught when the button is pushed.
By the way, button releasing will emit a "falling" event. The timing of
"falling" catching is not expected.
Cc: stable@vger.kernel.org
Signed-off-by: Michael Wu <michael.wu@vatics.com>
Tested-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2019-07-08 08:23:08 +03:00
irqflags | = test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) ?
IRQF_TRIGGER_FALLING : IRQF_TRIGGER_RISING ;
2016-06-02 12:30:15 +03:00
if ( eflags & GPIOEVENT_REQUEST_FALLING_EDGE )
gpiolib: fix incorrect IRQ requesting of an active-low lineevent
When a pin is active-low, logical trigger edge should be inverted to match
the same interrupt opportunity.
For example, a button pushed triggers falling edge in ACTIVE_HIGH case; in
ACTIVE_LOW case, the button pushed triggers rising edge. For user space the
IRQ requesting doesn't need to do any modification except to configuring
GPIOHANDLE_REQUEST_ACTIVE_LOW.
For example, we want to catch the event when the button is pushed. The
button on the original board drives level to be low when it is pushed, and
drives level to be high when it is released.
In user space we can do:
req.handleflags = GPIOHANDLE_REQUEST_INPUT;
req.eventflags = GPIOEVENT_REQUEST_FALLING_EDGE;
while (1) {
read(fd, &dat, sizeof(dat));
if (dat.id == GPIOEVENT_EVENT_FALLING_EDGE)
printf("button pushed\n");
}
Run the same logic on another board which the polarity of the button is
inverted; it drives level to be high when pushed, and level to be low when
released. For this inversion we add flag GPIOHANDLE_REQUEST_ACTIVE_LOW:
req.handleflags = GPIOHANDLE_REQUEST_INPUT |
GPIOHANDLE_REQUEST_ACTIVE_LOW;
req.eventflags = GPIOEVENT_REQUEST_FALLING_EDGE;
At the result, there are no any events caught when the button is pushed.
By the way, button releasing will emit a "falling" event. The timing of
"falling" catching is not expected.
Cc: stable@vger.kernel.org
Signed-off-by: Michael Wu <michael.wu@vatics.com>
Tested-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2019-07-08 08:23:08 +03:00
irqflags | = test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) ?
IRQF_TRIGGER_RISING : IRQF_TRIGGER_FALLING ;
2016-06-02 12:30:15 +03:00
irqflags | = IRQF_ONESHOT ;
INIT_KFIFO ( le - > events ) ;
init_waitqueue_head ( & le - > wait ) ;
mutex_init ( & le - > read_lock ) ;
/* Request a thread to read the events */
ret = request_threaded_irq ( le - > irq ,
2017-11-30 12:23:27 +03:00
lineevent_irq_handler ,
2016-06-02 12:30:15 +03:00
lineevent_irq_thread ,
irqflags ,
le - > label ,
le ) ;
if ( ret )
goto out_free_desc ;
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
fd = get_unused_fd_flags ( O_RDONLY | O_CLOEXEC ) ;
2016-06-02 12:30:15 +03:00
if ( fd < 0 ) {
ret = fd ;
goto out_free_irq ;
}
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
file = anon_inode_getfile ( " gpio-event " ,
& lineevent_fileops ,
le ,
O_RDONLY | O_CLOEXEC ) ;
if ( IS_ERR ( file ) ) {
ret = PTR_ERR ( file ) ;
goto out_put_unused_fd ;
}
2016-06-02 12:30:15 +03:00
eventreq . fd = fd ;
2016-07-04 14:13:04 +03:00
if ( copy_to_user ( ip , & eventreq , sizeof ( eventreq ) ) ) {
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
/*
* fput ( ) will trigger the release ( ) callback , so do not go onto
* the regular error cleanup path here .
*/
fput ( file ) ;
put_unused_fd ( fd ) ;
return - EFAULT ;
2016-07-04 14:13:04 +03:00
}
2016-06-02 12:30:15 +03:00
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
fd_install ( fd , file ) ;
2016-06-02 12:30:15 +03:00
return 0 ;
gpio: GPIO_GET_LINE{HANDLE,EVENT}_IOCTL: Fix file descriptor leak
When allocating a new line handle or event a file is allocated that it is
associated to. The file is attached to a file descriptor of the current
process and the file descriptor is returned to userspace using
copy_to_user(). If this copy operation fails the line handle or event
allocation is aborted, all acquired resources are freed and an error is
returned.
But the file struct is not freed and left attached to the userspace
application and even though the file descriptor number was not copied it is
trivial to guess. If a userspace application performs a IOCTL on such a
left over file descriptor it will trigger a use-after-free and if the file
descriptor is closed (latest when the application exits) a double-free is
triggered.
anon_inode_getfd() performs 3 tasks, allocate a file struct, allocate a
file descriptor for the current process and install the file struct in the
file descriptor. As soon as the file struct is installed in the file
descriptor it is accessible by userspace (even if the IOCTL itself hasn't
completed yet), this means uninstalling the fd on the error path is not an
option, since userspace might already got a reference to the file.
Instead anon_inode_getfd() needs to be broken into its individual steps.
The allocation of the file struct and file descriptor is done first, then
the copy_to_user() is executed and only if it succeeds the file is
installed.
Since the file struct is reference counted it can not be just freed, but
its reference needs to be dropped, which will also call the release()
callback, which will free the state attached to the file. So in this case
the normal error cleanup path should not be taken.
Cc: stable@vger.kernel.org
Fixes: d932cd49182f ("gpio: free handles in fringe cases")
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-10-24 14:59:15 +03:00
out_put_unused_fd :
put_unused_fd ( fd ) ;
2016-06-02 12:30:15 +03:00
out_free_irq :
free_irq ( le - > irq , le ) ;
out_free_desc :
gpiod_free ( le - > desc ) ;
out_free_label :
kfree ( le - > label ) ;
out_free_le :
kfree ( le ) ;
put_device ( & gdev - > dev ) ;
return ret ;
}
2017-07-24 17:57:22 +03:00
/*
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
* gpio_ioctl ( ) - ioctl handler for the GPIO chardev
*/
static long gpio_ioctl ( struct file * filp , unsigned int cmd , unsigned long arg )
{
struct gpio_device * gdev = filp - > private_data ;
struct gpio_chip * chip = gdev - > chip ;
2016-05-27 15:24:04 +03:00
void __user * ip = ( void __user * ) arg ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
/* We fail any subsequent ioctl():s when the chip is gone */
if ( ! chip )
return - ENODEV ;
2016-02-13 00:25:22 +03:00
/* Fill in the struct and pass to userspace */
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
if ( cmd = = GPIO_GET_CHIPINFO_IOCTL ) {
2016-02-13 00:25:22 +03:00
struct gpiochip_info chipinfo ;
2016-10-18 17:54:00 +03:00
memset ( & chipinfo , 0 , sizeof ( chipinfo ) ) ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
strncpy ( chipinfo . name , dev_name ( & gdev - > dev ) ,
sizeof ( chipinfo . name ) ) ;
chipinfo . name [ sizeof ( chipinfo . name ) - 1 ] = ' \0 ' ;
2016-02-12 16:48:23 +03:00
strncpy ( chipinfo . label , gdev - > label ,
sizeof ( chipinfo . label ) ) ;
chipinfo . label [ sizeof ( chipinfo . label ) - 1 ] = ' \0 ' ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
chipinfo . lines = gdev - > ngpio ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
if ( copy_to_user ( ip , & chipinfo , sizeof ( chipinfo ) ) )
return - EFAULT ;
return 0 ;
2016-02-13 00:25:22 +03:00
} else if ( cmd = = GPIO_GET_LINEINFO_IOCTL ) {
struct gpioline_info lineinfo ;
struct gpio_desc * desc ;
if ( copy_from_user ( & lineinfo , ip , sizeof ( lineinfo ) ) )
return - EFAULT ;
2016-10-18 17:53:59 +03:00
if ( lineinfo . line_offset > = gdev - > ngpio )
2016-02-13 00:25:22 +03:00
return - EINVAL ;
desc = & gdev - > descs [ lineinfo . line_offset ] ;
if ( desc - > name ) {
strncpy ( lineinfo . name , desc - > name ,
sizeof ( lineinfo . name ) ) ;
lineinfo . name [ sizeof ( lineinfo . name ) - 1 ] = ' \0 ' ;
} else {
lineinfo . name [ 0 ] = ' \0 ' ;
}
if ( desc - > label ) {
2016-02-25 23:01:48 +03:00
strncpy ( lineinfo . consumer , desc - > label ,
sizeof ( lineinfo . consumer ) ) ;
lineinfo . consumer [ sizeof ( lineinfo . consumer ) - 1 ] = ' \0 ' ;
2016-02-13 00:25:22 +03:00
} else {
2016-02-25 23:01:48 +03:00
lineinfo . consumer [ 0 ] = ' \0 ' ;
2016-02-13 00:25:22 +03:00
}
/*
* Userspace only need to know that the kernel is using
* this GPIO so it can ' t use it .
*/
lineinfo . flags = 0 ;
2016-02-22 15:44:53 +03:00
if ( test_bit ( FLAG_REQUESTED , & desc - > flags ) | |
test_bit ( FLAG_IS_HOGGED , & desc - > flags ) | |
test_bit ( FLAG_USED_AS_IRQ , & desc - > flags ) | |
test_bit ( FLAG_EXPORT , & desc - > flags ) | |
2019-08-14 14:00:35 +03:00
test_bit ( FLAG_SYSFS , & desc - > flags ) | |
! pinctrl_gpio_can_use_line ( chip - > base + lineinfo . line_offset ) )
2016-02-13 00:25:22 +03:00
lineinfo . flags | = GPIOLINE_FLAG_KERNEL ;
2016-02-22 15:44:53 +03:00
if ( test_bit ( FLAG_IS_OUT , & desc - > flags ) )
2016-02-13 00:25:22 +03:00
lineinfo . flags | = GPIOLINE_FLAG_IS_OUT ;
2016-02-22 15:44:53 +03:00
if ( test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) )
2016-02-13 00:25:22 +03:00
lineinfo . flags | = GPIOLINE_FLAG_ACTIVE_LOW ;
2016-02-22 15:44:53 +03:00
if ( test_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) )
2019-08-06 14:41:51 +03:00
lineinfo . flags | = ( GPIOLINE_FLAG_OPEN_DRAIN |
GPIOLINE_FLAG_IS_OUT ) ;
2016-02-22 15:44:53 +03:00
if ( test_bit ( FLAG_OPEN_SOURCE , & desc - > flags ) )
2019-08-06 14:41:51 +03:00
lineinfo . flags | = ( GPIOLINE_FLAG_OPEN_SOURCE |
GPIOLINE_FLAG_IS_OUT ) ;
2019-09-21 13:25:23 +03:00
if ( test_bit ( FLAG_PULL_DOWN , & desc - > flags ) )
lineinfo . flags | = GPIOLINE_FLAG_PULL_DOWN ;
if ( test_bit ( FLAG_PULL_UP , & desc - > flags ) )
lineinfo . flags | = GPIOLINE_FLAG_PULL_UP ;
2016-02-13 00:25:22 +03:00
if ( copy_to_user ( ip , & lineinfo , sizeof ( lineinfo ) ) )
return - EFAULT ;
return 0 ;
gpio: userspace ABI for reading/writing GPIO lines
This adds a userspace ABI for reading and writing GPIO lines.
The mechanism returns an anonymous file handle to a request
to read/write n offsets from a gpiochip. This file handle
in turn accepts two ioctl()s: one that reads and one that
writes values to the selected lines.
- Handles can be requested as input/output, active low,
open drain, open source, however when you issue a request
for n lines with GPIO_GET_LINEHANDLE_IOCTL, they must all
have the same flags, i.e. all inputs or all outputs, all
open drain etc. If a granular control of the flags for
each line is desired, they need to be requested
individually, not in a batch.
- The GPIOHANDLE_GET_LINE_VALUES_IOCTL read ioctl() can be
issued also to output lines to verify that the hardware
is in the expected state.
- It reads and writes up to GPIOHANDLES_MAX lines at once,
utilizing the .set_multiple() call in the driver if
possible, making the call efficient if several lines
can be written with a single register update.
The limitation of GPIOHANDLES_MAX to 64 lines is done under
the assumption that we may expect hardware that can issue a
transaction updating 64 bits at an instant but unlikely
anything larger than that.
ChangeLog v2->v3:
- Use gpiod_get_value_cansleep() so we support also slowpath
GPIO drivers.
- Fix up the UAPI docs kerneldoc.
- Allocate the anonymous fd last, so that the release
function don't get called until that point of something
fails. After this point, skip the errorpath.
ChangeLog v1->v2:
- Handle ioctl_compat() properly based on a similar patch
to the other ioctl() handling code.
- Use _IOWR() as we pass pointers both in and out of the
ioctl()
- Use kmalloc() and kfree() for the linehandled, do not
try to be fancy with devm_* it doesn't work the way I
thought.
- Fix const-correctness on the linehandle name field.
Acked-by: Michael Welling <mwelling@ieee.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-04-26 11:35:29 +03:00
} else if ( cmd = = GPIO_GET_LINEHANDLE_IOCTL ) {
return linehandle_create ( gdev , ip ) ;
2016-06-02 12:30:15 +03:00
} else if ( cmd = = GPIO_GET_LINEEVENT_IOCTL ) {
return lineevent_create ( gdev , ip ) ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
}
return - EINVAL ;
}
2016-05-27 15:24:04 +03:00
# ifdef CONFIG_COMPAT
static long gpio_ioctl_compat ( struct file * filp , unsigned int cmd ,
unsigned long arg )
{
return gpio_ioctl ( filp , cmd , ( unsigned long ) compat_ptr ( arg ) ) ;
}
# endif
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
/**
* gpio_chrdev_open ( ) - open the chardev for ioctl operations
* @ inode : inode for this chardev
* @ filp : file struct for storing private data
* Returns 0 on success
*/
static int gpio_chrdev_open ( struct inode * inode , struct file * filp )
{
struct gpio_device * gdev = container_of ( inode - > i_cdev ,
struct gpio_device , chrdev ) ;
/* Fail on open if the backing gpiochip is gone */
2017-01-09 22:47:44 +03:00
if ( ! gdev - > chip )
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
return - ENODEV ;
get_device ( & gdev - > dev ) ;
filp - > private_data = gdev ;
2016-11-30 15:05:21 +03:00
return nonseekable_open ( inode , filp ) ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
}
/**
* gpio_chrdev_release ( ) - close chardev after ioctl operations
* @ inode : inode for this chardev
* @ filp : file struct for storing private data
* Returns 0 on success
*/
static int gpio_chrdev_release ( struct inode * inode , struct file * filp )
{
struct gpio_device * gdev = container_of ( inode - > i_cdev ,
struct gpio_device , chrdev ) ;
put_device ( & gdev - > dev ) ;
return 0 ;
}
static const struct file_operations gpio_fileops = {
. release = gpio_chrdev_release ,
. open = gpio_chrdev_open ,
. owner = THIS_MODULE ,
2016-11-30 15:05:21 +03:00
. llseek = no_llseek ,
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
. unlocked_ioctl = gpio_ioctl ,
2016-05-27 15:24:04 +03:00
# ifdef CONFIG_COMPAT
. compat_ioctl = gpio_ioctl_compat ,
# endif
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
} ;
2015-10-20 12:10:38 +03:00
static void gpiodevice_release ( struct device * dev )
{
struct gpio_device * gdev = dev_get_drvdata ( dev ) ;
list_del ( & gdev - > list ) ;
ida_simple_remove ( & gpio_ida , gdev - > id ) ;
2017-12-14 17:29:20 +03:00
kfree_const ( gdev - > label ) ;
2016-03-31 18:11:29 +03:00
kfree ( gdev - > descs ) ;
2016-02-09 16:27:42 +03:00
kfree ( gdev ) ;
2015-10-20 12:10:38 +03:00
}
2016-03-31 18:11:30 +03:00
static int gpiochip_setup_dev ( struct gpio_device * gdev )
{
2019-07-16 12:11:45 +03:00
int ret ;
2016-03-31 18:11:30 +03:00
cdev_init ( & gdev - > chrdev , & gpio_fileops ) ;
gdev - > chrdev . owner = THIS_MODULE ;
gdev - > dev . devt = MKDEV ( MAJOR ( gpio_devt ) , gdev - > id ) ;
2017-03-17 21:48:12 +03:00
2019-07-16 12:11:45 +03:00
ret = cdev_device_add ( & gdev - > chrdev , & gdev - > dev ) ;
if ( ret )
return ret ;
2017-03-17 21:48:12 +03:00
chip_dbg ( gdev - > chip , " added GPIO chardev (%d:%d) \n " ,
MAJOR ( gpio_devt ) , gdev - > id ) ;
2016-03-31 18:11:30 +03:00
2019-07-16 12:11:45 +03:00
ret = gpiochip_sysfs_register ( gdev ) ;
if ( ret )
2016-03-31 18:11:30 +03:00
goto err_remove_device ;
/* From this point, the .release() function cleans up gpio_device */
gdev - > dev . release = gpiodevice_release ;
pr_debug ( " %s: registered GPIOs %d to %d on device: %s (%s) \n " ,
__func__ , gdev - > base , gdev - > base + gdev - > ngpio - 1 ,
dev_name ( & gdev - > dev ) , gdev - > chip - > label ? : " generic " ) ;
return 0 ;
err_remove_device :
2017-03-17 21:48:12 +03:00
cdev_device_del ( & gdev - > chrdev , & gdev - > dev ) ;
2019-07-16 12:11:45 +03:00
return ret ;
2016-03-31 18:11:30 +03:00
}
2018-04-10 23:30:28 +03:00
static void gpiochip_machine_hog ( struct gpio_chip * chip , struct gpiod_hog * hog )
{
struct gpio_desc * desc ;
int rv ;
desc = gpiochip_get_desc ( chip , hog - > chip_hwnum ) ;
if ( IS_ERR ( desc ) ) {
pr_err ( " %s: unable to get GPIO desc: %ld \n " ,
__func__ , PTR_ERR ( desc ) ) ;
return ;
}
2018-05-01 10:36:39 +03:00
if ( test_bit ( FLAG_IS_HOGGED , & desc - > flags ) )
2018-04-10 23:30:28 +03:00
return ;
rv = gpiod_hog ( desc , hog - > line_name , hog - > lflags , hog - > dflags ) ;
if ( rv )
pr_err ( " %s: unable to hog GPIO line (%s:%u): %d \n " ,
__func__ , chip - > label , hog - > chip_hwnum , rv ) ;
}
static void machine_gpiochip_add ( struct gpio_chip * chip )
{
struct gpiod_hog * hog ;
mutex_lock ( & gpio_machine_hogs_mutex ) ;
list_for_each_entry ( hog , & gpio_machine_hogs , list ) {
if ( ! strcmp ( chip - > label , hog - > chip_label ) )
gpiochip_machine_hog ( chip , hog ) ;
}
mutex_unlock ( & gpio_machine_hogs_mutex ) ;
}
2016-03-31 18:11:30 +03:00
static void gpiochip_setup_devs ( void )
{
struct gpio_device * gdev ;
2019-07-16 12:11:45 +03:00
int ret ;
2016-03-31 18:11:30 +03:00
list_for_each_entry ( gdev , & gpio_devices , list ) {
2019-07-16 12:11:45 +03:00
ret = gpiochip_setup_dev ( gdev ) ;
if ( ret )
2016-03-31 18:11:30 +03:00
pr_err ( " %s: Failed to initialize gpio device (%d) \n " ,
2019-07-16 12:11:45 +03:00
dev_name ( & gdev - > dev ) , ret ) ;
2016-03-31 18:11:30 +03:00
}
}
2017-11-07 21:15:59 +03:00
int gpiochip_add_data_with_key ( struct gpio_chip * chip , void * data ,
2017-12-02 20:11:04 +03:00
struct lock_class_key * lock_key ,
struct lock_class_key * request_key )
2008-02-05 09:28:20 +03:00
{
unsigned long flags ;
2019-07-16 12:11:45 +03:00
int ret = 0 ;
2015-10-20 12:10:38 +03:00
unsigned i ;
2008-04-28 13:14:46 +04:00
int base = chip - > base ;
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev ;
2008-02-05 09:28:20 +03:00
2015-10-20 12:10:38 +03:00
/*
* First : allocate and populate the internal stat container , and
* set up the struct device .
*/
2016-02-18 01:44:15 +03:00
gdev = kzalloc ( sizeof ( * gdev ) , GFP_KERNEL ) ;
2015-10-20 12:10:38 +03:00
if ( ! gdev )
2014-11-19 10:51:27 +03:00
return - ENOMEM ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
gdev - > dev . bus = & gpio_bus_type ;
2015-10-20 12:10:38 +03:00
gdev - > chip = chip ;
chip - > gpiodev = gdev ;
if ( chip - > parent ) {
gdev - > dev . parent = chip - > parent ;
gdev - > dev . of_node = chip - > parent - > of_node ;
2016-07-05 15:11:14 +03:00
}
2015-10-20 12:10:38 +03:00
# ifdef CONFIG_OF_GPIO
/* If the gpiochip has an assigned OF node this takes precedence */
2016-07-05 15:11:14 +03:00
if ( chip - > of_node )
gdev - > dev . of_node = chip - > of_node ;
2018-08-06 12:48:01 +03:00
else
chip - > of_node = gdev - > dev . of_node ;
2015-10-20 12:10:38 +03:00
# endif
2016-07-05 15:11:14 +03:00
2015-10-20 12:10:38 +03:00
gdev - > id = ida_simple_get ( & gpio_ida , 0 , 0 , GFP_KERNEL ) ;
if ( gdev - > id < 0 ) {
2019-07-16 12:11:45 +03:00
ret = gdev - > id ;
2015-10-20 12:10:38 +03:00
goto err_free_gdev ;
}
dev_set_name ( & gdev - > dev , " gpiochip%d " , gdev - > id ) ;
device_initialize ( & gdev - > dev ) ;
dev_set_drvdata ( & gdev - > dev , gdev ) ;
if ( chip - > parent & & chip - > parent - > driver )
gdev - > owner = chip - > parent - > driver - > owner ;
else if ( chip - > owner )
/* TODO: remove chip->owner */
gdev - > owner = chip - > owner ;
else
gdev - > owner = THIS_MODULE ;
2008-02-05 09:28:20 +03:00
2016-03-31 18:11:29 +03:00
gdev - > descs = kcalloc ( chip - > ngpio , sizeof ( gdev - > descs [ 0 ] ) , GFP_KERNEL ) ;
2016-02-09 15:51:59 +03:00
if ( ! gdev - > descs ) {
2019-07-16 12:11:45 +03:00
ret = - ENOMEM ;
2018-11-02 16:39:43 +03:00
goto err_free_ida ;
2015-10-20 12:10:38 +03:00
}
2015-11-16 08:02:47 +03:00
if ( chip - > ngpio = = 0 ) {
chip_err ( chip , " tried to insert a GPIO chip with zero lines \n " ) ;
2019-07-16 12:11:45 +03:00
ret = - EINVAL ;
2016-03-31 18:11:30 +03:00
goto err_free_descs ;
2015-11-16 08:02:47 +03:00
}
2016-02-12 16:48:23 +03:00
2018-05-21 20:57:07 +03:00
if ( chip - > ngpio > FASTPATH_NGPIO )
chip_warn ( chip , " line cnt %u is greater than fast path cnt %u \n " ,
chip - > ngpio , FASTPATH_NGPIO ) ;
2017-12-14 17:29:20 +03:00
gdev - > label = kstrdup_const ( chip - > label ? : " unknown " , GFP_KERNEL ) ;
2016-02-12 16:48:23 +03:00
if ( ! gdev - > label ) {
2019-07-16 12:11:45 +03:00
ret = - ENOMEM ;
2016-03-31 18:11:29 +03:00
goto err_free_descs ;
2016-02-12 16:48:23 +03:00
}
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
gdev - > ngpio = chip - > ngpio ;
2016-02-11 13:37:48 +03:00
gdev - > data = data ;
2015-11-16 08:02:47 +03:00
2008-02-05 09:28:20 +03:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
/*
* TODO : this allocates a Linux GPIO number base in the global
* GPIO numberspace for this chip . In the long run we want to
* get * rid * of this numberspace and use only descriptors , but
* it may be a pipe dream . It will not happen before we get rid
* of the sysfs interface anyways .
*/
2008-04-28 13:14:46 +04:00
if ( base < 0 ) {
base = gpiochip_find_base ( chip - > ngpio ) ;
if ( base < 0 ) {
2019-07-16 12:11:45 +03:00
ret = base ;
2015-01-12 19:12:25 +03:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2016-03-31 18:11:29 +03:00
goto err_free_label ;
2008-04-28 13:14:46 +04:00
}
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
/*
* TODO : it should not be necessary to reflect the assigned
* base outside of the GPIO subsystem . Go over drivers and
* see if anyone makes use of this , else drop this and assign
* a poison instead .
*/
2008-04-28 13:14:46 +04:00
chip - > base = base ;
}
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
gdev - > base = base ;
2008-04-28 13:14:46 +04:00
2019-07-16 12:11:45 +03:00
ret = gpiodev_add_to_list ( gdev ) ;
if ( ret ) {
2015-01-12 19:12:26 +03:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2016-03-31 18:11:29 +03:00
goto err_free_label ;
2015-01-12 19:12:26 +03:00
}
2013-02-02 20:29:24 +04:00
2016-05-30 18:11:59 +03:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2018-10-12 09:11:36 +03:00
for ( i = 0 ; i < chip - > ngpio ; i + + )
gdev - > descs [ i ] . gdev = gdev ;
2014-11-19 10:51:27 +03:00
2012-10-27 13:51:36 +04:00
# ifdef CONFIG_PINCTRL
2016-02-11 13:03:06 +03:00
INIT_LIST_HEAD ( & gdev - > pin_ranges ) ;
2012-10-27 13:51:36 +04:00
# endif
2019-07-16 12:11:45 +03:00
ret = gpiochip_set_desc_names ( chip ) ;
if ( ret )
2015-08-14 17:11:00 +03:00
goto err_remove_from_list ;
2019-07-16 12:11:45 +03:00
ret = gpiochip_alloc_valid_mask ( chip ) ;
if ( ret )
2019-08-20 11:05:27 +03:00
goto err_remove_from_list ;
2017-11-07 21:15:54 +03:00
2019-07-16 12:11:45 +03:00
ret = of_gpiochip_add ( chip ) ;
if ( ret )
2019-08-20 11:05:27 +03:00
goto err_free_gpiochip_mask ;
2015-07-14 11:29:54 +03:00
2019-07-16 12:11:45 +03:00
ret = gpiochip_init_valid_mask ( chip ) ;
if ( ret )
2019-04-24 16:59:33 +03:00
goto err_remove_of_chip ;
2018-10-05 09:52:58 +03:00
2018-10-05 09:53:00 +03:00
for ( i = 0 ; i < chip - > ngpio ; i + + ) {
struct gpio_desc * desc = & gdev - > descs [ i ] ;
2019-07-07 23:35:58 +03:00
if ( chip - > get_direction & & gpiochip_line_is_valid ( chip , i ) ) {
if ( ! chip - > get_direction ( chip , i ) )
set_bit ( FLAG_IS_OUT , & desc - > flags ) ;
else
clear_bit ( FLAG_IS_OUT , & desc - > flags ) ;
} else {
if ( ! chip - > direction_input )
set_bit ( FLAG_IS_OUT , & desc - > flags ) ;
else
clear_bit ( FLAG_IS_OUT , & desc - > flags ) ;
}
2018-10-05 09:53:00 +03:00
}
2014-01-08 14:40:54 +04:00
acpi_gpiochip_add ( chip ) ;
2010-06-08 17:48:17 +04:00
2018-04-10 23:30:28 +03:00
machine_gpiochip_add ( chip ) ;
2019-10-30 15:29:14 +03:00
ret = gpiochip_irqchip_init_valid_mask ( chip ) ;
2019-10-09 17:34:44 +03:00
if ( ret )
goto err_remove_acpi_chip ;
2019-10-30 15:29:14 +03:00
ret = gpiochip_irqchip_init_hw ( chip ) ;
2019-09-06 13:05:35 +03:00
if ( ret )
2019-08-20 11:05:27 +03:00
goto err_remove_acpi_chip ;
2019-09-06 13:05:35 +03:00
ret = gpiochip_add_irqchip ( chip , lock_key , request_key ) ;
if ( ret )
2019-08-20 11:05:27 +03:00
goto err_remove_irqchip_mask ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
/*
* By first adding the chardev , and then adding the device ,
* we get a device node entry in sysfs under
* / sys / bus / gpio / devices / gpiochipN / dev that can be used for
* coldplug of device nodes and other udev business .
2016-03-31 18:11:30 +03:00
* We can do this only if gpiolib has been initialized .
* Otherwise , defer until later .
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
*/
2016-03-31 18:11:30 +03:00
if ( gpiolib_initialized ) {
2019-07-16 12:11:45 +03:00
ret = gpiochip_setup_dev ( gdev ) ;
if ( ret )
2019-08-20 11:05:27 +03:00
goto err_remove_irqchip ;
2016-03-31 18:11:30 +03:00
}
2010-06-08 17:48:15 +04:00
return 0 ;
2013-06-09 07:08:32 +04:00
2019-08-20 11:05:27 +03:00
err_remove_irqchip :
gpiochip_irqchip_remove ( chip ) ;
err_remove_irqchip_mask :
gpiochip_irqchip_free_valid_mask ( chip ) ;
2019-04-24 16:59:33 +03:00
err_remove_acpi_chip :
2015-01-12 19:12:25 +03:00
acpi_gpiochip_remove ( chip ) ;
2019-04-24 16:59:33 +03:00
err_remove_of_chip :
2015-05-04 18:23:25 +03:00
gpiochip_free_hogs ( chip ) ;
2015-01-12 19:12:25 +03:00
of_gpiochip_remove ( chip ) ;
2019-04-24 16:59:33 +03:00
err_free_gpiochip_mask :
2018-03-23 19:34:52 +03:00
gpiochip_free_valid_mask ( chip ) ;
2015-08-14 17:11:00 +03:00
err_remove_from_list :
2015-01-12 19:12:25 +03:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
2015-10-20 12:10:38 +03:00
list_del ( & gdev - > list ) ;
2013-06-09 07:08:32 +04:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2016-03-31 18:11:29 +03:00
err_free_label :
2017-12-14 17:29:20 +03:00
kfree_const ( gdev - > label ) ;
2016-03-31 18:11:29 +03:00
err_free_descs :
kfree ( gdev - > descs ) ;
2018-11-02 16:39:43 +03:00
err_free_ida :
2015-10-20 12:10:38 +03:00
ida_simple_remove ( & gpio_ida , gdev - > id ) ;
2018-11-02 16:39:43 +03:00
err_free_gdev :
2008-02-05 09:28:20 +03:00
/* failures here can mean systems won't boot... */
2018-07-20 10:54:49 +03:00
pr_err ( " %s: GPIOs %d..%d (%s) failed to register, %d \n " , __func__ ,
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
gdev - > base , gdev - > base + gdev - > ngpio - 1 ,
2019-07-16 12:11:45 +03:00
chip - > label ? : " generic " , ret ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
kfree ( gdev ) ;
2019-07-16 12:11:45 +03:00
return ret ;
2008-02-05 09:28:20 +03:00
}
2017-11-07 21:15:59 +03:00
EXPORT_SYMBOL_GPL ( gpiochip_add_data_with_key ) ;
2008-02-05 09:28:20 +03:00
2016-02-11 13:37:48 +03:00
/**
* gpiochip_get_data ( ) - get per - subdriver data for the chip
2017-07-24 17:57:22 +03:00
* @ chip : GPIO chip
*
* Returns :
* The per - subdriver data for the chip .
2016-02-11 13:37:48 +03:00
*/
void * gpiochip_get_data ( struct gpio_chip * chip )
{
return chip - > gpiodev - > data ;
}
EXPORT_SYMBOL_GPL ( gpiochip_get_data ) ;
2008-02-05 09:28:20 +03:00
/**
* gpiochip_remove ( ) - unregister a gpio_chip
* @ chip : the chip to unregister
*
* A gpio_chip with any GPIOs still requested may not be removed .
*/
2014-07-05 20:28:50 +04:00
void gpiochip_remove ( struct gpio_chip * chip )
2008-02-05 09:28:20 +03:00
{
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev = chip - > gpiodev ;
2015-05-04 18:10:27 +03:00
struct gpio_desc * desc ;
2008-02-05 09:28:20 +03:00
unsigned long flags ;
2016-02-09 15:51:59 +03:00
unsigned i ;
2015-05-04 18:10:27 +03:00
bool requested = false ;
2008-02-05 09:28:20 +03:00
2015-10-20 12:10:38 +03:00
/* FIXME: should the legacy sysfs handling be moved to gpio_device? */
2016-02-09 15:21:06 +03:00
gpiochip_sysfs_unregister ( gdev ) ;
2016-12-19 20:29:23 +03:00
gpiochip_free_hogs ( chip ) ;
2016-02-20 08:13:19 +03:00
/* Numb the device, cancelling all outstanding operations */
gdev - > chip = NULL ;
2015-01-12 19:12:27 +03:00
gpiochip_irqchip_remove ( chip ) ;
2014-03-10 16:54:53 +04:00
acpi_gpiochip_remove ( chip ) ;
2012-11-06 18:15:44 +04:00
gpiochip_remove_pin_ranges ( chip ) ;
2010-06-08 17:48:17 +04:00
of_gpiochip_remove ( chip ) ;
2018-03-23 19:34:52 +03:00
gpiochip_free_valid_mask ( chip ) ;
2016-02-11 13:37:48 +03:00
/*
* We accept no more calls into the driver from this point , so
* NULL the driver data pointer
*/
gdev - > data = NULL ;
2010-06-08 17:48:17 +04:00
2015-01-12 19:12:28 +03:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
for ( i = 0 ; i < gdev - > ngpio ; i + + ) {
2016-02-09 15:51:59 +03:00
desc = & gdev - > descs [ i ] ;
2015-05-04 18:10:27 +03:00
if ( test_bit ( FLAG_REQUESTED , & desc - > flags ) )
requested = true ;
2008-02-05 09:28:20 +03:00
}
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2014-11-19 10:51:27 +03:00
2015-05-04 18:10:27 +03:00
if ( requested )
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
dev_crit ( & gdev - > dev ,
2015-11-04 11:56:26 +03:00
" REMOVING GPIOCHIP WITH GPIOS STILL REQUESTED \n " ) ;
2015-05-04 18:10:27 +03:00
2015-10-20 12:10:38 +03:00
/*
* The gpiochip side puts its use of the device to rest here :
* if there are no userspace clients , the chardev and device will
* be removed , else it will be dangling until the last user is
* gone .
*/
2017-03-17 21:48:12 +03:00
cdev_device_del ( & gdev - > chrdev , & gdev - > dev ) ;
2015-10-20 12:10:38 +03:00
put_device ( & gdev - > dev ) ;
2008-02-05 09:28:20 +03:00
}
EXPORT_SYMBOL_GPL ( gpiochip_remove ) ;
2016-02-15 14:02:09 +03:00
static void devm_gpio_chip_release ( struct device * dev , void * res )
{
struct gpio_chip * chip = * ( struct gpio_chip * * ) res ;
gpiochip_remove ( chip ) ;
}
/**
2017-12-21 19:56:34 +03:00
* devm_gpiochip_add_data ( ) - Resource manager gpiochip_add_data ( )
2018-10-08 12:34:03 +03:00
* @ dev : pointer to the device that gpio_chip belongs to .
2016-02-15 14:02:09 +03:00
* @ chip : the chip to register , with chip - > base initialized
2017-07-24 17:57:22 +03:00
* @ data : driver - private data associated with this chip
2016-02-15 14:02:09 +03:00
*
2017-07-24 17:57:22 +03:00
* Context : potentially before irqs will work
2016-02-15 14:02:09 +03:00
*
* The gpio chip automatically be released when the device is unbound .
2017-07-24 17:57:22 +03:00
*
* Returns :
* A negative errno if the chip can ' t be registered , such as because the
* chip - > base is invalid or already associated with a different chip .
* Otherwise it returns zero as a success code .
2016-02-15 14:02:09 +03:00
*/
int devm_gpiochip_add_data ( struct device * dev , struct gpio_chip * chip ,
void * data )
{
struct gpio_chip * * ptr ;
int ret ;
ptr = devres_alloc ( devm_gpio_chip_release , sizeof ( * ptr ) ,
GFP_KERNEL ) ;
if ( ! ptr )
return - ENOMEM ;
ret = gpiochip_add_data ( chip , data ) ;
if ( ret < 0 ) {
devres_free ( ptr ) ;
return ret ;
}
* ptr = chip ;
devres_add ( dev , ptr ) ;
return 0 ;
}
EXPORT_SYMBOL_GPL ( devm_gpiochip_add_data ) ;
2010-06-08 17:48:16 +04:00
/**
* gpiochip_find ( ) - iterator for locating a specific gpio_chip
* @ data : data to pass to match function
2017-07-24 17:57:22 +03:00
* @ match : Callback function to check gpio_chip
2010-06-08 17:48:16 +04:00
*
* Similar to bus_find_device . It returns a reference to a gpio_chip as
* determined by a user supplied @ match callback . The callback should return
* 0 if the device doesn ' t match and non - zero if it does . If the callback is
* non - zero , this function will return to the caller and not iterate over any
* more gpio_chips .
*/
2012-05-19 09:01:05 +04:00
struct gpio_chip * gpiochip_find ( void * data ,
2012-03-03 02:56:03 +04:00
int ( * match ) ( struct gpio_chip * chip ,
2012-05-17 23:54:40 +04:00
void * data ) )
2010-06-08 17:48:16 +04:00
{
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev ;
2016-08-11 19:21:58 +03:00
struct gpio_chip * chip = NULL ;
2010-06-08 17:48:16 +04:00
unsigned long flags ;
spin_lock_irqsave ( & gpio_lock , flags ) ;
2015-10-20 12:10:38 +03:00
list_for_each_entry ( gdev , & gpio_devices , list )
2016-08-11 19:21:58 +03:00
if ( gdev - > chip & & match ( gdev - > chip , data ) ) {
chip = gdev - > chip ;
2010-06-08 17:48:16 +04:00
break ;
2016-08-11 19:21:58 +03:00
}
2015-10-20 12:10:38 +03:00
2010-06-08 17:48:16 +04:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
return chip ;
}
2011-05-20 10:40:18 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_find ) ;
2008-02-05 09:28:20 +03:00
2013-11-16 16:39:32 +04:00
static int gpiochip_match_name ( struct gpio_chip * chip , void * data )
{
const char * name = data ;
return ! strcmp ( chip - > label , name ) ;
}
static struct gpio_chip * find_chip_by_name ( const char * name )
{
return gpiochip_find ( ( void * ) name , gpiochip_match_name ) ;
}
2014-03-25 13:40:18 +04:00
# ifdef CONFIG_GPIOLIB_IRQCHIP
/*
* The following is irqchip helper code for gpiochips .
*/
2019-10-09 17:34:44 +03:00
static int gpiochip_irqchip_init_hw ( struct gpio_chip * gc )
{
struct gpio_irq_chip * girq = & gc - > irq ;
if ( ! girq - > init_hw )
return 0 ;
return girq - > init_hw ( gc ) ;
}
2019-09-04 17:01:04 +03:00
static int gpiochip_irqchip_init_valid_mask ( struct gpio_chip * gc )
2016-09-20 15:15:21 +03:00
{
2019-09-04 17:01:04 +03:00
struct gpio_irq_chip * girq = & gc - > irq ;
if ( ! girq - > init_valid_mask )
2016-09-20 15:15:21 +03:00
return 0 ;
2019-09-04 17:01:04 +03:00
girq - > valid_mask = gpiochip_allocate_mask ( gc ) ;
if ( ! girq - > valid_mask )
2016-09-20 15:15:21 +03:00
return - ENOMEM ;
2019-09-04 17:01:04 +03:00
girq - > init_valid_mask ( gc , girq - > valid_mask , gc - > ngpio ) ;
2016-09-20 15:15:21 +03:00
return 0 ;
}
static void gpiochip_irqchip_free_valid_mask ( struct gpio_chip * gpiochip )
{
2019-07-18 09:51:01 +03:00
bitmap_free ( gpiochip - > irq . valid_mask ) ;
2017-11-07 21:15:52 +03:00
gpiochip - > irq . valid_mask = NULL ;
2016-09-20 15:15:21 +03:00
}
2018-01-10 04:58:46 +03:00
bool gpiochip_irqchip_irq_valid ( const struct gpio_chip * gpiochip ,
unsigned int offset )
2016-09-20 15:15:21 +03:00
{
2018-03-23 19:34:52 +03:00
if ( ! gpiochip_line_is_valid ( gpiochip , offset ) )
return false ;
2016-09-20 15:15:21 +03:00
/* No mask means all valid */
2017-11-07 21:15:52 +03:00
if ( likely ( ! gpiochip - > irq . valid_mask ) )
2016-09-20 15:15:21 +03:00
return true ;
2017-11-07 21:15:52 +03:00
return test_bit ( offset , gpiochip - > irq . valid_mask ) ;
2016-09-20 15:15:21 +03:00
}
2018-01-10 04:58:46 +03:00
EXPORT_SYMBOL_GPL ( gpiochip_irqchip_irq_valid ) ;
2016-09-20 15:15:21 +03:00
2014-03-25 13:40:18 +04:00
/**
2016-11-24 12:57:25 +03:00
* gpiochip_set_cascaded_irqchip ( ) - connects a cascaded irqchip to a gpiochip
2019-06-14 11:12:26 +03:00
* @ gc : the gpiochip to set the irqchip chain to
2014-03-25 13:40:18 +04:00
* @ parent_irq : the irq number corresponding to the parent IRQ for this
* chained irqchip
* @ parent_handler : the parent interrupt handler for the accumulated IRQ
2014-09-26 16:19:52 +04:00
* coming out of the gpiochip . If the interrupt is nested rather than
* cascaded , pass NULL in this handler argument
2014-03-25 13:40:18 +04:00
*/
2019-06-14 11:12:26 +03:00
static void gpiochip_set_cascaded_irqchip ( struct gpio_chip * gc ,
2017-04-03 19:05:21 +03:00
unsigned int parent_irq ,
2016-11-24 12:57:25 +03:00
irq_flow_handler_t parent_handler )
2014-03-25 13:40:18 +04:00
{
2019-06-14 11:12:26 +03:00
struct gpio_irq_chip * girq = & gc - > irq ;
struct device * dev = & gc - > gpiodev - > dev ;
if ( ! girq - > domain ) {
chip_err ( gc , " called %s before setting up irqchip \n " ,
2014-09-26 15:50:12 +04:00
__func__ ) ;
2014-04-09 15:34:39 +04:00
return ;
}
2014-09-26 16:19:52 +04:00
if ( parent_handler ) {
2019-06-14 11:12:26 +03:00
if ( gc - > can_sleep ) {
chip_err ( gc ,
2018-07-03 03:39:03 +03:00
" you cannot have chained interrupts on a chip that may sleep \n " ) ;
2014-09-26 16:19:52 +04:00
return ;
}
2019-06-14 11:12:26 +03:00
girq - > parents = devm_kcalloc ( dev , 1 ,
sizeof ( * girq - > parents ) ,
GFP_KERNEL ) ;
if ( ! girq - > parents ) {
chip_err ( gc , " out of memory allocating parent IRQ \n " ) ;
return ;
}
girq - > parents [ 0 ] = parent_irq ;
girq - > num_parents = 1 ;
2014-09-26 16:19:52 +04:00
/*
* The parent irqchip is already using the chip_data for this
* irqchip , so our callbacks simply use the handler_data .
*/
2015-06-21 22:10:48 +03:00
irq_set_chained_handler_and_data ( parent_irq , parent_handler ,
2019-06-14 11:12:26 +03:00
gc ) ;
2014-09-26 16:19:52 +04:00
}
2014-03-25 13:40:18 +04:00
}
2016-11-24 12:57:25 +03:00
/**
* gpiochip_set_chained_irqchip ( ) - connects a chained irqchip to a gpiochip
* @ gpiochip : the gpiochip to set the irqchip chain to
* @ irqchip : the irqchip to chain to the gpiochip
* @ parent_irq : the irq number corresponding to the parent IRQ for this
* chained irqchip
* @ parent_handler : the parent interrupt handler for the accumulated IRQ
2018-10-08 19:32:16 +03:00
* coming out of the gpiochip .
2016-11-24 12:57:25 +03:00
*/
void gpiochip_set_chained_irqchip ( struct gpio_chip * gpiochip ,
struct irq_chip * irqchip ,
2017-04-03 19:05:21 +03:00
unsigned int parent_irq ,
2016-11-24 12:57:25 +03:00
irq_flow_handler_t parent_handler )
{
2017-11-07 21:15:57 +03:00
if ( gpiochip - > irq . threaded ) {
chip_err ( gpiochip , " tried to chain a threaded gpiochip \n " ) ;
return ;
}
2018-10-08 19:32:15 +03:00
gpiochip_set_cascaded_irqchip ( gpiochip , parent_irq , parent_handler ) ;
2016-11-24 12:57:25 +03:00
}
2014-03-25 13:40:18 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_set_chained_irqchip ) ;
2016-11-24 12:57:25 +03:00
/**
* gpiochip_set_nested_irqchip ( ) - connects a nested irqchip to a gpiochip
* @ gpiochip : the gpiochip to set the irqchip nested handler to
* @ irqchip : the irqchip to nest to the gpiochip
* @ parent_irq : the irq number corresponding to the parent IRQ for this
* nested irqchip
*/
void gpiochip_set_nested_irqchip ( struct gpio_chip * gpiochip ,
struct irq_chip * irqchip ,
2017-04-03 19:05:21 +03:00
unsigned int parent_irq )
2016-11-24 12:57:25 +03:00
{
2018-10-08 19:32:15 +03:00
gpiochip_set_cascaded_irqchip ( gpiochip , parent_irq , NULL ) ;
2016-11-24 12:57:25 +03:00
}
EXPORT_SYMBOL_GPL ( gpiochip_set_nested_irqchip ) ;
2019-08-08 15:32:37 +03:00
# ifdef CONFIG_IRQ_DOMAIN_HIERARCHY
/**
* gpiochip_set_hierarchical_irqchip ( ) - connects a hierarchical irqchip
* to a gpiochip
* @ gc : the gpiochip to set the irqchip hierarchical handler to
* @ irqchip : the irqchip to handle this level of the hierarchy , the interrupt
* will then percolate up to the parent
*/
static void gpiochip_set_hierarchical_irqchip ( struct gpio_chip * gc ,
struct irq_chip * irqchip )
{
/* DT will deal with mapping each IRQ as we go along */
if ( is_of_node ( gc - > irq . fwnode ) )
return ;
/*
* This is for legacy and boardfile " irqchip " fwnodes : allocate
* irqs upfront instead of dynamically since we don ' t have the
* dynamic type of allocation that hardware description languages
* provide . Once all GPIO drivers using board files are gone from
* the kernel we can delete this code , but for a transitional period
* it is necessary to keep this around .
*/
if ( is_fwnode_irqchip ( gc - > irq . fwnode ) ) {
int i ;
int ret ;
for ( i = 0 ; i < gc - > ngpio ; i + + ) {
struct irq_fwspec fwspec ;
unsigned int parent_hwirq ;
unsigned int parent_type ;
struct gpio_irq_chip * girq = & gc - > irq ;
/*
* We call the child to parent translation function
* only to check if the child IRQ is valid or not .
* Just pick the rising edge type here as that is what
* we likely need to support .
*/
ret = girq - > child_to_parent_hwirq ( gc , i ,
IRQ_TYPE_EDGE_RISING ,
& parent_hwirq ,
& parent_type ) ;
if ( ret ) {
chip_err ( gc , " skip set-up on hwirq %d \n " ,
i ) ;
continue ;
}
fwspec . fwnode = gc - > irq . fwnode ;
/* This is the hwirq for the GPIO line side of things */
fwspec . param [ 0 ] = girq - > child_offset_to_irq ( gc , i ) ;
/* Just pick something */
fwspec . param [ 1 ] = IRQ_TYPE_EDGE_RISING ;
fwspec . param_count = 2 ;
ret = __irq_domain_alloc_irqs ( gc - > irq . domain ,
/* just pick something */
- 1 ,
1 ,
NUMA_NO_NODE ,
& fwspec ,
false ,
NULL ) ;
if ( ret < 0 ) {
chip_err ( gc ,
" can not allocate irq for GPIO line %d parent hwirq %d in hierarchy domain: %d \n " ,
i , parent_hwirq ,
ret ) ;
}
}
}
chip_err ( gc , " %s unknown fwnode type proceed anyway \n " , __func__ ) ;
return ;
}
static int gpiochip_hierarchy_irq_domain_translate ( struct irq_domain * d ,
struct irq_fwspec * fwspec ,
unsigned long * hwirq ,
unsigned int * type )
{
/* We support standard DT translation */
if ( is_of_node ( fwspec - > fwnode ) & & fwspec - > param_count = = 2 ) {
return irq_domain_translate_twocell ( d , fwspec , hwirq , type ) ;
}
/* This is for board files and others not using DT */
if ( is_fwnode_irqchip ( fwspec - > fwnode ) ) {
int ret ;
ret = irq_domain_translate_twocell ( d , fwspec , hwirq , type ) ;
if ( ret )
return ret ;
WARN_ON ( * type = = IRQ_TYPE_NONE ) ;
return 0 ;
}
return - EINVAL ;
}
static int gpiochip_hierarchy_irq_domain_alloc ( struct irq_domain * d ,
unsigned int irq ,
unsigned int nr_irqs ,
void * data )
{
struct gpio_chip * gc = d - > host_data ;
irq_hw_number_t hwirq ;
unsigned int type = IRQ_TYPE_NONE ;
struct irq_fwspec * fwspec = data ;
struct irq_fwspec parent_fwspec ;
unsigned int parent_hwirq ;
unsigned int parent_type ;
struct gpio_irq_chip * girq = & gc - > irq ;
int ret ;
/*
* The nr_irqs parameter is always one except for PCI multi - MSI
* so this should not happen .
*/
WARN_ON ( nr_irqs ! = 1 ) ;
ret = gc - > irq . child_irq_domain_ops . translate ( d , fwspec , & hwirq , & type ) ;
if ( ret )
return ret ;
chip_info ( gc , " allocate IRQ %d, hwirq %lu \n " , irq , hwirq ) ;
ret = girq - > child_to_parent_hwirq ( gc , hwirq , type ,
& parent_hwirq , & parent_type ) ;
if ( ret ) {
chip_err ( gc , " can't look up hwirq %lu \n " , hwirq ) ;
return ret ;
}
chip_info ( gc , " found parent hwirq %u \n " , parent_hwirq ) ;
/*
* We set handle_bad_irq because the . set_type ( ) should
* always be invoked and set the right type of handler .
*/
irq_domain_set_info ( d ,
irq ,
hwirq ,
gc - > irq . chip ,
gc ,
girq - > handler ,
NULL , NULL ) ;
irq_set_probe ( irq ) ;
/*
* Create a IRQ fwspec to send up to the parent irqdomain :
* specify the hwirq we address on the parent and tie it
* all together up the chain .
*/
parent_fwspec . fwnode = d - > parent - > fwnode ;
/* This parent only handles asserted level IRQs */
girq - > populate_parent_fwspec ( gc , & parent_fwspec , parent_hwirq ,
parent_type ) ;
chip_info ( gc , " alloc_irqs_parent for %d parent hwirq %d \n " ,
irq , parent_hwirq ) ;
ret = irq_domain_alloc_irqs_parent ( d , irq , 1 , & parent_fwspec ) ;
if ( ret )
chip_err ( gc ,
" failed to allocate parent hwirq %d for hwirq %lu \n " ,
parent_hwirq , hwirq ) ;
return ret ;
}
static unsigned int gpiochip_child_offset_to_irq_noop ( struct gpio_chip * chip ,
unsigned int offset )
{
return offset ;
}
static void gpiochip_hierarchy_setup_domain_ops ( struct irq_domain_ops * ops )
{
ops - > activate = gpiochip_irq_domain_activate ;
ops - > deactivate = gpiochip_irq_domain_deactivate ;
ops - > alloc = gpiochip_hierarchy_irq_domain_alloc ;
ops - > free = irq_domain_free_irqs_common ;
/*
* We only allow overriding the translate ( ) function for
* hierarchical chips , and this should only be done if the user
* really need something other than 1 : 1 translation .
*/
if ( ! ops - > translate )
ops - > translate = gpiochip_hierarchy_irq_domain_translate ;
}
static int gpiochip_hierarchy_add_domain ( struct gpio_chip * gc )
{
if ( ! gc - > irq . child_to_parent_hwirq | |
! gc - > irq . fwnode ) {
chip_err ( gc , " missing irqdomain vital data \n " ) ;
return - EINVAL ;
}
if ( ! gc - > irq . child_offset_to_irq )
gc - > irq . child_offset_to_irq = gpiochip_child_offset_to_irq_noop ;
if ( ! gc - > irq . populate_parent_fwspec )
gc - > irq . populate_parent_fwspec =
gpiochip_populate_parent_fwspec_twocell ;
gpiochip_hierarchy_setup_domain_ops ( & gc - > irq . child_irq_domain_ops ) ;
gc - > irq . domain = irq_domain_create_hierarchy (
gc - > irq . parent_domain ,
0 ,
gc - > ngpio ,
gc - > irq . fwnode ,
& gc - > irq . child_irq_domain_ops ,
gc ) ;
if ( ! gc - > irq . domain )
return - ENOMEM ;
gpiochip_set_hierarchical_irqchip ( gc , gc - > irq . chip ) ;
return 0 ;
}
static bool gpiochip_hierarchy_is_hierarchical ( struct gpio_chip * gc )
{
return ! ! gc - > irq . parent_domain ;
}
void gpiochip_populate_parent_fwspec_twocell ( struct gpio_chip * chip ,
struct irq_fwspec * fwspec ,
unsigned int parent_hwirq ,
unsigned int parent_type )
{
fwspec - > param_count = 2 ;
fwspec - > param [ 0 ] = parent_hwirq ;
fwspec - > param [ 1 ] = parent_type ;
}
EXPORT_SYMBOL_GPL ( gpiochip_populate_parent_fwspec_twocell ) ;
void gpiochip_populate_parent_fwspec_fourcell ( struct gpio_chip * chip ,
struct irq_fwspec * fwspec ,
unsigned int parent_hwirq ,
unsigned int parent_type )
{
fwspec - > param_count = 4 ;
fwspec - > param [ 0 ] = 0 ;
fwspec - > param [ 1 ] = parent_hwirq ;
fwspec - > param [ 2 ] = 0 ;
fwspec - > param [ 3 ] = parent_type ;
}
EXPORT_SYMBOL_GPL ( gpiochip_populate_parent_fwspec_fourcell ) ;
# else
static int gpiochip_hierarchy_add_domain ( struct gpio_chip * gc )
{
return - EINVAL ;
}
static bool gpiochip_hierarchy_is_hierarchical ( struct gpio_chip * gc )
{
return false ;
}
# endif /* CONFIG_IRQ_DOMAIN_HIERARCHY */
2014-03-25 13:40:18 +04:00
/**
* gpiochip_irq_map ( ) - maps an IRQ into a GPIO irqchip
* @ d : the irqdomain used by this irqchip
* @ irq : the global irq number used by this GPIO irqchip irq
* @ hwirq : the local IRQ / GPIO line offset on this gpiochip
*
* This function will set up the mapping for a certain IRQ line on a
* gpiochip by assigning the gpiochip as chip data , and using the irqchip
* stored inside the gpiochip .
*/
2017-11-07 21:15:55 +03:00
int gpiochip_irq_map ( struct irq_domain * d , unsigned int irq ,
irq_hw_number_t hwirq )
2014-03-25 13:40:18 +04:00
{
struct gpio_chip * chip = d - > host_data ;
2019-07-16 12:11:45 +03:00
int ret = 0 ;
2014-03-25 13:40:18 +04:00
2017-07-21 19:49:00 +03:00
if ( ! gpiochip_irqchip_irq_valid ( chip , hwirq ) )
return - ENXIO ;
2014-03-25 13:40:18 +04:00
irq_set_chip_data ( irq , chip ) ;
gpiolib: irqchip: use different lockdep class for each gpio irqchip
Since IRQ chip helpers were introduced drivers lose ability to
register separate lockdep classes for each registered GPIO IRQ
chip and the gpiolib now is using shared lockdep class for
all GPIO IRQ chips (gpiochip_irq_lock_class).
As result, lockdep will produce warning when there are min two
stacked GPIO chips and all of them are interrupt controllers.
HW configuration which generates lockdep warning (TI dra7-evm):
[SOC GPIO bankA.gpioX]
<- irq - [pcf875x.gpioY]
<- irq - DevZ.enable_irq_wake(pcf_gpioY_irq);
The issue was reported in [1] and discussed [2].
=============================================
[ INFO: possible recursive locking detected ]
4.2.0-rc6-00013-g5d050ed-dirty #55 Not tainted
---------------------------------------------
sh/63 is trying to acquire lock:
(class){......}, at: [<c009b91c>] __irq_get_desc_lock+0x50/0x94
but task is already holding lock:
(class){......}, at: [<c009b91c>] __irq_get_desc_lock+0x50/0x94
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(class);
lock(class);
*** DEADLOCK ***
May be due to missing lock nesting notation
7 locks held by sh/63:
#0: (sb_writers#4){.+.+.+}, at: [<c016bbb8>] vfs_write+0x13c/0x164
#1: (&of->mutex){+.+.+.}, at: [<c01debf4>] kernfs_fop_write+0x4c/0x1a0
#2: (s_active#36){.+.+.+}, at: [<c01debfc>] kernfs_fop_write+0x54/0x1a0
#3: (pm_mutex){+.+.+.}, at: [<c009758c>] pm_suspend+0xec/0x4c4
#4: (&dev->mutex){......}, at: [<c03f77f8>] __device_suspend+0xd4/0x398
#5: (&gpio->lock){+.+.+.}, at: [<c009b940>] __irq_get_desc_lock+0x74/0x94
#6: (class){......}, at: [<c009b91c>] __irq_get_desc_lock+0x50/0x94
stack backtrace:
CPU: 0 PID: 63 Comm: sh Not tainted 4.2.0-rc6-00013-g5d050ed-dirty #55
Hardware name: Generic DRA74X (Flattened Device Tree)
[<c0016e24>] (unwind_backtrace) from [<c0013338>] (show_stack+0x10/0x14)
[<c0013338>] (show_stack) from [<c05f6b24>] (dump_stack+0x84/0x9c)
[<c05f6b24>] (dump_stack) from [<c00903f4>] (__lock_acquire+0x19c0/0x1e20)
[<c00903f4>] (__lock_acquire) from [<c0091098>] (lock_acquire+0xa8/0x128)
[<c0091098>] (lock_acquire) from [<c05fd61c>] (_raw_spin_lock_irqsave+0x38/0x4c)
[<c05fd61c>] (_raw_spin_lock_irqsave) from [<c009b91c>] (__irq_get_desc_lock+0x50/0x94)
[<c009b91c>] (__irq_get_desc_lock) from [<c009c4f4>] (irq_set_irq_wake+0x20/0xfc)
[<c009c4f4>] (irq_set_irq_wake) from [<c0393ac4>] (pcf857x_irq_set_wake+0x24/0x54)
[<c0393ac4>] (pcf857x_irq_set_wake) from [<c009c560>] (irq_set_irq_wake+0x8c/0xfc)
[<c009c560>] (irq_set_irq_wake) from [<c04a02ac>] (gpio_keys_suspend+0x70/0xd4)
[<c04a02ac>] (gpio_keys_suspend) from [<c03f6a00>] (dpm_run_callback+0x50/0x124)
[<c03f6a00>] (dpm_run_callback) from [<c03f7830>] (__device_suspend+0x10c/0x398)
[<c03f7830>] (__device_suspend) from [<c03f90f0>] (dpm_suspend+0x134/0x2f4)
[<c03f90f0>] (dpm_suspend) from [<c0096e20>] (suspend_devices_and_enter+0xa8/0x728)
[<c0096e20>] (suspend_devices_and_enter) from [<c00977cc>] (pm_suspend+0x32c/0x4c4)
[<c00977cc>] (pm_suspend) from [<c0096060>] (state_store+0x64/0xb8)
[<c0096060>] (state_store) from [<c01dec64>] (kernfs_fop_write+0xbc/0x1a0)
[<c01dec64>] (kernfs_fop_write) from [<c016b280>] (__vfs_write+0x20/0xd8)
[<c016b280>] (__vfs_write) from [<c016bb0c>] (vfs_write+0x90/0x164)
[<c016bb0c>] (vfs_write) from [<c016c330>] (SyS_write+0x44/0x9c)
[<c016c330>] (SyS_write) from [<c000f500>] (ret_fast_syscall+0x0/0x54)
Lets fix it by using separate lockdep class for each registered GPIO
IRQ Chip. This is done by wrapping gpiochip_irqchip_add call into macros.
The implementation of this patch inspired by solution done by Nicolas
Boichat for regmap [3]
[1] http://www.spinics.net/lists/linux-gpio/msg05844.html
[2] http://www.spinics.net/lists/linux-gpio/msg06021.html
[3] http://www.spinics.net/lists/arm-kernel/msg429834.html
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roger Quadros <rogerq@ti.com>
Reported-by: Roger Quadros <rogerq@ti.com>
Tested-by: Roger Quadros <rogerq@ti.com>
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-08-17 15:35:23 +03:00
/*
* This lock class tells lockdep that GPIO irqs are in a different
* category than their parents , so it won ' t report false recursion .
*/
2017-12-02 20:11:04 +03:00
irq_set_lockdep_class ( irq , chip - > irq . lock_key , chip - > irq . request_key ) ;
2017-11-07 21:15:48 +03:00
irq_set_chip_and_handler ( irq , chip - > irq . chip , chip - > irq . handler ) ;
2016-11-24 12:57:25 +03:00
/* Chips that use nested thread handlers have them marked */
2017-11-07 21:15:57 +03:00
if ( chip - > irq . threaded )
2014-04-09 15:34:39 +04:00
irq_set_nested_thread ( irq , 1 ) ;
2014-03-25 13:40:18 +04:00
irq_set_noprobe ( irq ) ;
2015-07-27 23:55:16 +03:00
2017-11-07 21:15:54 +03:00
if ( chip - > irq . num_parents = = 1 )
2019-07-16 12:11:45 +03:00
ret = irq_set_parent ( irq , chip - > irq . parents [ 0 ] ) ;
2017-11-07 21:15:54 +03:00
else if ( chip - > irq . map )
2019-07-16 12:11:45 +03:00
ret = irq_set_parent ( irq , chip - > irq . map [ hwirq ] ) ;
2017-11-07 21:15:54 +03:00
2019-07-16 12:11:45 +03:00
if ( ret < 0 )
return ret ;
2017-11-07 21:15:54 +03:00
2014-04-23 18:45:12 +04:00
/*
* No set - up of the hardware will happen if IRQ_TYPE_NONE
* is passed as default type .
*/
2017-11-07 21:15:49 +03:00
if ( chip - > irq . default_type ! = IRQ_TYPE_NONE )
irq_set_irq_type ( irq , chip - > irq . default_type ) ;
2014-03-25 13:40:18 +04:00
return 0 ;
}
2017-11-07 21:15:55 +03:00
EXPORT_SYMBOL_GPL ( gpiochip_irq_map ) ;
2014-03-25 13:40:18 +04:00
2017-11-07 21:15:55 +03:00
void gpiochip_irq_unmap ( struct irq_domain * d , unsigned int irq )
2014-03-28 23:42:01 +04:00
{
2014-04-09 15:34:39 +04:00
struct gpio_chip * chip = d - > host_data ;
2017-11-07 21:15:57 +03:00
if ( chip - > irq . threaded )
2014-04-09 15:34:39 +04:00
irq_set_nested_thread ( irq , 0 ) ;
2014-03-28 23:42:01 +04:00
irq_set_chip_and_handler ( irq , NULL , NULL ) ;
irq_set_chip_data ( irq , NULL ) ;
}
2017-11-07 21:15:55 +03:00
EXPORT_SYMBOL_GPL ( gpiochip_irq_unmap ) ;
2014-03-28 23:42:01 +04:00
2014-03-25 13:40:18 +04:00
static const struct irq_domain_ops gpiochip_domain_ops = {
. map = gpiochip_irq_map ,
2014-03-28 23:42:01 +04:00
. unmap = gpiochip_irq_unmap ,
2014-03-25 13:40:18 +04:00
/* Virtually all GPIO irqchips are twocell:ed */
. xlate = irq_domain_xlate_twocell ,
} ;
2019-08-08 15:32:37 +03:00
/*
* TODO : move these activate / deactivate in under the hierarchicial
* irqchip implementation as static once SPMI and SSBI ( all external
* users ) are phased over .
*/
2019-01-19 23:42:42 +03:00
/**
* gpiochip_irq_domain_activate ( ) - Lock a GPIO to be used as an IRQ
* @ domain : The IRQ domain used by this IRQ chip
* @ data : Outermost irq_data associated with the IRQ
* @ reserve : If set , only reserve an interrupt vector instead of assigning one
*
* This function is a wrapper that calls gpiochip_lock_as_irq ( ) and is to be
* used as the activate function for the & struct irq_domain_ops . The host_data
* for the IRQ domain must be the & struct gpio_chip .
*/
int gpiochip_irq_domain_activate ( struct irq_domain * domain ,
struct irq_data * data , bool reserve )
{
struct gpio_chip * chip = domain - > host_data ;
return gpiochip_lock_as_irq ( chip , data - > hwirq ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_irq_domain_activate ) ;
/**
* gpiochip_irq_domain_deactivate ( ) - Unlock a GPIO used as an IRQ
* @ domain : The IRQ domain used by this IRQ chip
* @ data : Outermost irq_data associated with the IRQ
*
* This function is a wrapper that will call gpiochip_unlock_as_irq ( ) and is to
* be used as the deactivate function for the & struct irq_domain_ops . The
* host_data for the IRQ domain must be the & struct gpio_chip .
*/
void gpiochip_irq_domain_deactivate ( struct irq_domain * domain ,
struct irq_data * data )
{
struct gpio_chip * chip = domain - > host_data ;
return gpiochip_unlock_as_irq ( chip , data - > hwirq ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_irq_domain_deactivate ) ;
2018-09-08 12:23:14 +03:00
static int gpiochip_to_irq ( struct gpio_chip * chip , unsigned offset )
2014-03-25 13:40:18 +04:00
{
2019-08-08 15:32:37 +03:00
struct irq_domain * domain = chip - > irq . domain ;
2018-09-08 12:23:14 +03:00
if ( ! gpiochip_irqchip_irq_valid ( chip , offset ) )
return - ENXIO ;
2015-06-25 20:30:50 +03:00
2019-08-08 15:32:37 +03:00
# ifdef CONFIG_IRQ_DOMAIN_HIERARCHY
if ( irq_domain_is_hierarchy ( domain ) ) {
struct irq_fwspec spec ;
spec . fwnode = domain - > fwnode ;
spec . param_count = 2 ;
spec . param [ 0 ] = chip - > irq . child_offset_to_irq ( chip , offset ) ;
spec . param [ 1 ] = IRQ_TYPE_NONE ;
return irq_create_fwspec_mapping ( & spec ) ;
}
# endif
return irq_create_mapping ( domain , offset ) ;
2014-03-25 13:40:18 +04:00
}
static int gpiochip_irq_reqres ( struct irq_data * d )
{
struct gpio_chip * chip = irq_data_get_irq_chip_data ( d ) ;
2015-06-25 20:30:50 +03:00
2018-09-08 12:23:14 +03:00
return gpiochip_reqres_irq ( chip , d - > hwirq ) ;
2014-03-25 13:40:18 +04:00
}
static void gpiochip_irq_relres ( struct irq_data * d )
{
struct gpio_chip * chip = irq_data_get_irq_chip_data ( d ) ;
2018-09-08 12:23:14 +03:00
gpiochip_relres_irq ( chip , d - > hwirq ) ;
2014-03-25 13:40:18 +04:00
}
2018-09-08 12:23:17 +03:00
static void gpiochip_irq_enable ( struct irq_data * d )
2014-03-25 13:40:18 +04:00
{
2018-09-08 12:23:17 +03:00
struct gpio_chip * chip = irq_data_get_irq_chip_data ( d ) ;
2017-11-07 21:15:54 +03:00
2018-09-08 12:23:17 +03:00
gpiochip_enable_irq ( chip , d - > hwirq ) ;
if ( chip - > irq . irq_enable )
chip - > irq . irq_enable ( d ) ;
else
chip - > irq . chip - > irq_unmask ( d ) ;
}
static void gpiochip_irq_disable ( struct irq_data * d )
{
struct gpio_chip * chip = irq_data_get_irq_chip_data ( d ) ;
if ( chip - > irq . irq_disable )
chip - > irq . irq_disable ( d ) ;
else
chip - > irq . chip - > irq_mask ( d ) ;
gpiochip_disable_irq ( chip , d - > hwirq ) ;
}
2018-09-08 12:23:15 +03:00
static void gpiochip_set_irq_hooks ( struct gpio_chip * gpiochip )
{
struct irq_chip * irqchip = gpiochip - > irq . chip ;
if ( ! irqchip - > irq_request_resources & &
! irqchip - > irq_release_resources ) {
irqchip - > irq_request_resources = gpiochip_irq_reqres ;
irqchip - > irq_release_resources = gpiochip_irq_relres ;
}
2018-09-08 12:23:17 +03:00
if ( WARN_ON ( gpiochip - > irq . irq_enable ) )
return ;
2018-09-14 11:36:39 +03:00
/* Check if the irqchip already has this hook... */
if ( irqchip - > irq_enable = = gpiochip_irq_enable ) {
/*
* . . . and if so , give a gentle warning that this is bad
* practice .
*/
chip_info ( gpiochip ,
" detected irqchip that is shared with multiple gpiochips: please fix the driver. \n " ) ;
return ;
}
2018-09-08 12:23:17 +03:00
gpiochip - > irq . irq_enable = irqchip - > irq_enable ;
gpiochip - > irq . irq_disable = irqchip - > irq_disable ;
irqchip - > irq_enable = gpiochip_irq_enable ;
irqchip - > irq_disable = gpiochip_irq_disable ;
2014-03-25 13:40:18 +04:00
}
2017-11-07 21:15:54 +03:00
/**
* gpiochip_add_irqchip ( ) - adds an IRQ chip to a GPIO chip
* @ gpiochip : the GPIO chip to add the IRQ chip to
2017-12-02 20:11:04 +03:00
* @ lock_key : lockdep class for IRQ lock
* @ request_key : lockdep class for IRQ request
2017-11-07 21:15:54 +03:00
*/
2017-11-07 21:15:59 +03:00
static int gpiochip_add_irqchip ( struct gpio_chip * gpiochip ,
2017-12-02 20:11:04 +03:00
struct lock_class_key * lock_key ,
struct lock_class_key * request_key )
2017-11-07 21:15:54 +03:00
{
struct irq_chip * irqchip = gpiochip - > irq . chip ;
2019-08-08 15:32:37 +03:00
const struct irq_domain_ops * ops = NULL ;
2017-11-07 21:15:54 +03:00
struct device_node * np ;
unsigned int type ;
unsigned int i ;
if ( ! irqchip )
return 0 ;
if ( gpiochip - > irq . parent_handler & & gpiochip - > can_sleep ) {
2018-07-03 03:39:03 +03:00
chip_err ( gpiochip , " you cannot have chained interrupts on a chip that may sleep \n " ) ;
2017-11-07 21:15:54 +03:00
return - EINVAL ;
}
np = gpiochip - > gpiodev - > dev . of_node ;
type = gpiochip - > irq . default_type ;
/*
* Specifying a default trigger is a terrible idea if DT or ACPI is
* used to configure the interrupts , as you may end up with
* conflicting triggers . Tell the user , and reset to NONE .
*/
if ( WARN ( np & & type ! = IRQ_TYPE_NONE ,
" %s: Ignoring %u default trigger \n " , np - > full_name , type ) )
type = IRQ_TYPE_NONE ;
if ( has_acpi_companion ( gpiochip - > parent ) & & type ! = IRQ_TYPE_NONE ) {
acpi_handle_warn ( ACPI_HANDLE ( gpiochip - > parent ) ,
" Ignoring %u default trigger \n " , type ) ;
type = IRQ_TYPE_NONE ;
}
gpiochip - > to_irq = gpiochip_to_irq ;
gpiochip - > irq . default_type = type ;
2017-11-07 21:15:59 +03:00
gpiochip - > irq . lock_key = lock_key ;
2017-12-02 20:11:04 +03:00
gpiochip - > irq . request_key = request_key ;
2017-11-07 21:15:54 +03:00
2019-08-08 15:32:37 +03:00
/* If a parent irqdomain is provided, let's build a hierarchy */
if ( gpiochip_hierarchy_is_hierarchical ( gpiochip ) ) {
int ret = gpiochip_hierarchy_add_domain ( gpiochip ) ;
if ( ret )
return ret ;
} else {
/* Some drivers provide custom irqdomain ops */
if ( gpiochip - > irq . domain_ops )
ops = gpiochip - > irq . domain_ops ;
if ( ! ops )
ops = & gpiochip_domain_ops ;
gpiochip - > irq . domain = irq_domain_add_simple ( np ,
gpiochip - > ngpio ,
gpiochip - > irq . first ,
ops , gpiochip ) ;
if ( ! gpiochip - > irq . domain )
return - EINVAL ;
}
2017-11-07 21:15:54 +03:00
if ( gpiochip - > irq . parent_handler ) {
void * data = gpiochip - > irq . parent_handler_data ? : gpiochip ;
for ( i = 0 ; i < gpiochip - > irq . num_parents ; i + + ) {
/*
* The parent IRQ chip is already using the chip_data
* for this IRQ chip , so our callbacks simply use the
* handler_data .
*/
irq_set_chained_handler_and_data ( gpiochip - > irq . parents [ i ] ,
gpiochip - > irq . parent_handler ,
data ) ;
}
}
2018-09-08 12:23:15 +03:00
gpiochip_set_irq_hooks ( gpiochip ) ;
2017-11-07 21:15:54 +03:00
acpi_gpiochip_request_interrupts ( gpiochip ) ;
return 0 ;
}
2014-03-25 13:40:18 +04:00
/**
* gpiochip_irqchip_remove ( ) - removes an irqchip added to a gpiochip
* @ gpiochip : the gpiochip to remove the irqchip from
*
* This is called only from gpiochip_remove ( )
*/
static void gpiochip_irqchip_remove ( struct gpio_chip * gpiochip )
{
2018-09-08 12:23:15 +03:00
struct irq_chip * irqchip = gpiochip - > irq . chip ;
2017-11-07 21:15:50 +03:00
unsigned int offset ;
2014-03-28 23:42:01 +04:00
2014-07-25 10:54:48 +04:00
acpi_gpiochip_free_interrupts ( gpiochip ) ;
2018-09-08 12:23:15 +03:00
if ( irqchip & & gpiochip - > irq . parent_handler ) {
2017-11-07 21:15:50 +03:00
struct gpio_irq_chip * irq = & gpiochip - > irq ;
unsigned int i ;
for ( i = 0 ; i < irq - > num_parents ; i + + )
irq_set_chained_handler_and_data ( irq - > parents [ i ] ,
NULL , NULL ) ;
2015-05-12 20:12:23 +03:00
}
2014-03-28 23:42:01 +04:00
/* Remove all IRQ mappings and delete the domain */
2017-11-07 21:15:47 +03:00
if ( gpiochip - > irq . domain ) {
2017-11-07 21:15:50 +03:00
unsigned int irq ;
2016-09-20 15:15:21 +03:00
for ( offset = 0 ; offset < gpiochip - > ngpio ; offset + + ) {
if ( ! gpiochip_irqchip_irq_valid ( gpiochip , offset ) )
continue ;
2017-11-07 21:15:47 +03:00
irq = irq_find_mapping ( gpiochip - > irq . domain , offset ) ;
irq_dispose_mapping ( irq ) ;
2016-09-20 15:15:21 +03:00
}
2017-11-07 21:15:47 +03:00
irq_domain_remove ( gpiochip - > irq . domain ) ;
2014-03-28 23:42:01 +04:00
}
2014-03-25 13:40:18 +04:00
2018-09-08 12:23:17 +03:00
if ( irqchip ) {
if ( irqchip - > irq_request_resources = = gpiochip_irq_reqres ) {
irqchip - > irq_request_resources = NULL ;
irqchip - > irq_release_resources = NULL ;
}
if ( irqchip - > irq_enable = = gpiochip_irq_enable ) {
irqchip - > irq_enable = gpiochip - > irq . irq_enable ;
irqchip - > irq_disable = gpiochip - > irq . irq_disable ;
}
2014-03-25 13:40:18 +04:00
}
2018-09-08 12:23:17 +03:00
gpiochip - > irq . irq_enable = NULL ;
gpiochip - > irq . irq_disable = NULL ;
2018-09-08 12:23:15 +03:00
gpiochip - > irq . chip = NULL ;
2016-09-20 15:15:21 +03:00
gpiochip_irqchip_free_valid_mask ( gpiochip ) ;
2014-03-25 13:40:18 +04:00
}
/**
2017-01-11 15:37:07 +03:00
* gpiochip_irqchip_add_key ( ) - adds an irqchip to a gpiochip
2014-03-25 13:40:18 +04:00
* @ gpiochip : the gpiochip to add the irqchip to
* @ irqchip : the irqchip to add to the gpiochip
* @ first_irq : if not dynamically assigned , the base ( first ) IRQ to
* allocate gpiochip irqs from
* @ handler : the irq handler to use ( often a predefined irq core function )
2014-04-23 18:45:12 +04:00
* @ type : the default type for IRQs on this irqchip , pass IRQ_TYPE_NONE
* to have the core avoid setting up any default type in the hardware .
2017-11-07 21:15:57 +03:00
* @ threaded : whether this irqchip uses a nested thread handler
2017-12-02 20:11:04 +03:00
* @ lock_key : lockdep class for IRQ lock
* @ request_key : lockdep class for IRQ request
2014-03-25 13:40:18 +04:00
*
* This function closely associates a certain irqchip with a certain
* gpiochip , providing an irq domain to translate the local IRQs to
* global irqs in the gpiolib core , and making sure that the gpiochip
* is passed as chip data to all related functions . Driver callbacks
2015-12-07 17:31:58 +03:00
* need to use gpiochip_get_data ( ) to get their local state containers back
2014-03-25 13:40:18 +04:00
* from the gpiochip passed as chip data . An irqdomain will be stored
* in the gpiochip that shall be used by the driver to handle IRQ number
* translation . The gpiochip will need to be initialized and registered
* before calling this function .
*
2014-03-28 23:42:01 +04:00
* This function will handle two cell : ed simple IRQs and assumes all
* the pins on the gpiochip can generate a unique IRQ . Everything else
2014-03-25 13:40:18 +04:00
* need to be open coded .
*/
2017-01-11 15:37:07 +03:00
int gpiochip_irqchip_add_key ( struct gpio_chip * gpiochip ,
struct irq_chip * irqchip ,
unsigned int first_irq ,
irq_flow_handler_t handler ,
unsigned int type ,
2017-11-07 21:15:57 +03:00
bool threaded ,
2017-12-02 20:11:04 +03:00
struct lock_class_key * lock_key ,
struct lock_class_key * request_key )
2014-03-25 13:40:18 +04:00
{
struct device_node * of_node ;
if ( ! gpiochip | | ! irqchip )
return - EINVAL ;
2015-11-04 11:56:26 +03:00
if ( ! gpiochip - > parent ) {
2014-03-25 13:40:18 +04:00
pr_err ( " missing gpiochip .dev parent pointer \n " ) ;
return - EINVAL ;
}
2017-11-07 21:15:57 +03:00
gpiochip - > irq . threaded = threaded ;
2015-11-04 11:56:26 +03:00
of_node = gpiochip - > parent - > of_node ;
2014-03-25 13:40:18 +04:00
# ifdef CONFIG_OF_GPIO
/*
2015-05-18 21:41:43 +03:00
* If the gpiochip has an assigned OF node this takes precedence
2015-11-18 12:07:07 +03:00
* FIXME : get rid of this and use gpiochip - > parent - > of_node
* everywhere
2014-03-25 13:40:18 +04:00
*/
if ( gpiochip - > of_node )
of_node = gpiochip - > of_node ;
# endif
2016-09-07 11:12:11 +03:00
/*
2016-09-12 14:29:51 +03:00
* Specifying a default trigger is a terrible idea if DT or ACPI is
2016-09-07 11:12:11 +03:00
* used to configure the interrupts , as you may end - up with
* conflicting triggers . Tell the user , and reset to NONE .
*/
if ( WARN ( of_node & & type ! = IRQ_TYPE_NONE ,
2017-07-19 00:43:03 +03:00
" %pOF: Ignoring %d default trigger \n " , of_node , type ) )
2016-09-07 11:12:11 +03:00
type = IRQ_TYPE_NONE ;
2016-09-12 14:29:51 +03:00
if ( has_acpi_companion ( gpiochip - > parent ) & & type ! = IRQ_TYPE_NONE ) {
acpi_handle_warn ( ACPI_HANDLE ( gpiochip - > parent ) ,
" Ignoring %d default trigger \n " , type ) ;
type = IRQ_TYPE_NONE ;
}
2016-09-07 11:12:11 +03:00
2017-11-07 21:15:46 +03:00
gpiochip - > irq . chip = irqchip ;
2017-11-07 21:15:48 +03:00
gpiochip - > irq . handler = handler ;
2017-11-07 21:15:49 +03:00
gpiochip - > irq . default_type = type ;
2014-03-25 13:40:18 +04:00
gpiochip - > to_irq = gpiochip_to_irq ;
2017-11-07 21:15:53 +03:00
gpiochip - > irq . lock_key = lock_key ;
2017-12-02 20:11:04 +03:00
gpiochip - > irq . request_key = request_key ;
2017-11-07 21:15:47 +03:00
gpiochip - > irq . domain = irq_domain_add_simple ( of_node ,
2014-03-25 13:40:18 +04:00
gpiochip - > ngpio , first_irq ,
& gpiochip_domain_ops , gpiochip ) ;
2017-11-07 21:15:47 +03:00
if ( ! gpiochip - > irq . domain ) {
2017-11-07 21:15:46 +03:00
gpiochip - > irq . chip = NULL ;
2014-03-25 13:40:18 +04:00
return - EINVAL ;
}
2015-07-31 15:48:56 +03:00
2018-09-08 12:23:15 +03:00
gpiochip_set_irq_hooks ( gpiochip ) ;
2014-03-25 13:40:18 +04:00
2014-07-25 10:54:48 +04:00
acpi_gpiochip_request_interrupts ( gpiochip ) ;
2014-03-25 13:40:18 +04:00
return 0 ;
}
2017-01-11 15:37:07 +03:00
EXPORT_SYMBOL_GPL ( gpiochip_irqchip_add_key ) ;
2014-03-25 13:40:18 +04:00
# else /* CONFIG_GPIOLIB_IRQCHIP */
2017-11-07 21:15:59 +03:00
static inline int gpiochip_add_irqchip ( struct gpio_chip * gpiochip ,
2017-12-02 20:11:04 +03:00
struct lock_class_key * lock_key ,
struct lock_class_key * request_key )
2017-11-07 21:15:54 +03:00
{
return 0 ;
}
2014-03-25 13:40:18 +04:00
static void gpiochip_irqchip_remove ( struct gpio_chip * gpiochip ) { }
2019-10-09 17:34:44 +03:00
static inline int gpiochip_irqchip_init_hw ( struct gpio_chip * gpiochip )
{
return 0 ;
}
2016-09-20 15:15:21 +03:00
static inline int gpiochip_irqchip_init_valid_mask ( struct gpio_chip * gpiochip )
{
return 0 ;
}
static inline void gpiochip_irqchip_free_valid_mask ( struct gpio_chip * gpiochip )
{ }
2014-03-25 13:40:18 +04:00
# endif /* CONFIG_GPIOLIB_IRQCHIP */
2015-10-11 18:34:15 +03:00
/**
* gpiochip_generic_request ( ) - request the gpio function for a pin
* @ chip : the gpiochip owning the GPIO
* @ offset : the offset of the GPIO to request for GPIO function
*/
int gpiochip_generic_request ( struct gpio_chip * chip , unsigned offset )
{
2017-09-22 12:02:10 +03:00
return pinctrl_gpio_request ( chip - > gpiodev - > base + offset ) ;
2015-10-11 18:34:15 +03:00
}
EXPORT_SYMBOL_GPL ( gpiochip_generic_request ) ;
/**
* gpiochip_generic_free ( ) - free the gpio function from a pin
* @ chip : the gpiochip to request the gpio function for
* @ offset : the offset of the GPIO to free from GPIO function
*/
void gpiochip_generic_free ( struct gpio_chip * chip , unsigned offset )
{
2017-09-22 12:02:10 +03:00
pinctrl_gpio_free ( chip - > gpiodev - > base + offset ) ;
2015-10-11 18:34:15 +03:00
}
EXPORT_SYMBOL_GPL ( gpiochip_generic_free ) ;
2017-01-23 15:34:34 +03:00
/**
* gpiochip_generic_config ( ) - apply configuration for a pin
* @ chip : the gpiochip owning the GPIO
* @ offset : the offset of the GPIO to apply the configuration
* @ config : the configuration to be applied
*/
int gpiochip_generic_config ( struct gpio_chip * chip , unsigned offset ,
unsigned long config )
{
return pinctrl_gpio_set_config ( chip - > gpiodev - > base + offset , config ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_generic_config ) ;
2012-10-27 13:51:36 +04:00
# ifdef CONFIG_PINCTRL
2012-11-06 17:49:39 +04:00
2013-10-15 17:37:54 +04:00
/**
* gpiochip_add_pingroup_range ( ) - add a range for GPIO < - > pin mapping
* @ chip : the gpiochip to add the range for
2015-06-17 16:42:11 +03:00
* @ pctldev : the pin controller to map to
2013-10-15 17:37:54 +04:00
* @ gpio_offset : the start offset in the current gpio_chip number space
* @ pin_group : name of the pin group inside the pin controller
2018-05-21 23:57:39 +03:00
*
* Calling this function directly from a DeviceTree - supported
* pinctrl driver is DEPRECATED . Please see Section 2.1 of
* Documentation / devicetree / bindings / gpio / gpio . txt on how to
* bind pinctrl and gpio drivers via the " gpio-ranges " property .
2013-10-15 17:37:54 +04:00
*/
int gpiochip_add_pingroup_range ( struct gpio_chip * chip ,
struct pinctrl_dev * pctldev ,
unsigned int gpio_offset , const char * pin_group )
{
struct gpio_pin_range * pin_range ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_device * gdev = chip - > gpiodev ;
2013-10-15 17:37:54 +04:00
int ret ;
pin_range = kzalloc ( sizeof ( * pin_range ) , GFP_KERNEL ) ;
if ( ! pin_range ) {
2013-12-05 13:26:24 +04:00
chip_err ( chip , " failed to allocate pin ranges \n " ) ;
2013-10-15 17:37:54 +04:00
return - ENOMEM ;
}
/* Use local offset as range ID */
pin_range - > range . id = gpio_offset ;
pin_range - > range . gc = chip ;
pin_range - > range . name = chip - > label ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
pin_range - > range . base = gdev - > base + gpio_offset ;
2013-10-15 17:37:54 +04:00
pin_range - > pctldev = pctldev ;
ret = pinctrl_get_group_pins ( pctldev , pin_group ,
& pin_range - > range . pins ,
& pin_range - > range . npins ) ;
2013-11-14 00:20:39 +04:00
if ( ret < 0 ) {
kfree ( pin_range ) ;
2013-10-15 17:37:54 +04:00
return ret ;
2013-11-14 00:20:39 +04:00
}
2013-10-15 17:37:54 +04:00
pinctrl_add_gpio_range ( pctldev , & pin_range - > range ) ;
2013-12-05 13:26:24 +04:00
chip_dbg ( chip , " created GPIO range %d->%d ==> %s PINGRP %s \n " ,
gpio_offset , gpio_offset + pin_range - > range . npins - 1 ,
2013-10-15 17:37:54 +04:00
pinctrl_dev_get_devname ( pctldev ) , pin_group ) ;
2016-02-11 13:03:06 +03:00
list_add_tail ( & pin_range - > node , & gdev - > pin_ranges ) ;
2013-10-15 17:37:54 +04:00
return 0 ;
}
EXPORT_SYMBOL_GPL ( gpiochip_add_pingroup_range ) ;
2012-11-20 15:40:15 +04:00
/**
* gpiochip_add_pin_range ( ) - add a range for GPIO < - > pin mapping
* @ chip : the gpiochip to add the range for
2017-07-24 17:57:22 +03:00
* @ pinctl_name : the dev_name ( ) of the pin controller to map to
2012-11-21 11:48:09 +04:00
* @ gpio_offset : the start offset in the current gpio_chip number space
* @ pin_offset : the start offset in the pin controller number space
2012-11-20 15:40:15 +04:00
* @ npins : the number of pins from the offset of each pin space ( GPIO and
* pin controller ) to accumulate in this range
2017-07-24 17:57:22 +03:00
*
* Returns :
* 0 on success , or a negative error - code on failure .
2018-05-21 23:57:39 +03:00
*
* Calling this function directly from a DeviceTree - supported
* pinctrl driver is DEPRECATED . Please see Section 2.1 of
* Documentation / devicetree / bindings / gpio / gpio . txt on how to
* bind pinctrl and gpio drivers via the " gpio-ranges " property .
2012-11-20 15:40:15 +04:00
*/
2012-11-06 19:03:35 +04:00
int gpiochip_add_pin_range ( struct gpio_chip * chip , const char * pinctl_name ,
2012-11-21 11:48:09 +04:00
unsigned int gpio_offset , unsigned int pin_offset ,
2012-11-20 15:40:15 +04:00
unsigned int npins )
2012-10-27 13:51:36 +04:00
{
struct gpio_pin_range * pin_range ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_device * gdev = chip - > gpiodev ;
2012-11-21 10:33:56 +04:00
int ret ;
2012-10-27 13:51:36 +04:00
2012-11-20 15:40:15 +04:00
pin_range = kzalloc ( sizeof ( * pin_range ) , GFP_KERNEL ) ;
2012-10-27 13:51:36 +04:00
if ( ! pin_range ) {
2013-12-05 13:26:24 +04:00
chip_err ( chip , " failed to allocate pin ranges \n " ) ;
2012-11-06 19:03:35 +04:00
return - ENOMEM ;
2012-10-27 13:51:36 +04:00
}
2012-11-20 15:40:15 +04:00
/* Use local offset as range ID */
2012-11-21 11:48:09 +04:00
pin_range - > range . id = gpio_offset ;
2012-11-20 15:40:15 +04:00
pin_range - > range . gc = chip ;
2012-10-27 13:51:36 +04:00
pin_range - > range . name = chip - > label ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
pin_range - > range . base = gdev - > base + gpio_offset ;
2012-11-21 11:48:09 +04:00
pin_range - > range . pin_base = pin_offset ;
2012-10-27 13:51:36 +04:00
pin_range - > range . npins = npins ;
2012-11-20 17:03:37 +04:00
pin_range - > pctldev = pinctrl_find_and_add_gpio_range ( pinctl_name ,
2012-10-27 13:51:36 +04:00
& pin_range - > range ) ;
2012-11-20 17:56:25 +04:00
if ( IS_ERR ( pin_range - > pctldev ) ) {
2012-11-21 10:33:56 +04:00
ret = PTR_ERR ( pin_range - > pctldev ) ;
2013-12-05 13:26:24 +04:00
chip_err ( chip , " could not create pin range \n " ) ;
2012-11-20 15:40:15 +04:00
kfree ( pin_range ) ;
2012-11-21 10:33:56 +04:00
return ret ;
2012-11-20 15:40:15 +04:00
}
2013-12-05 13:26:24 +04:00
chip_dbg ( chip , " created GPIO range %d->%d ==> %s PIN %d->%d \n " ,
gpio_offset , gpio_offset + npins - 1 ,
2012-11-21 11:48:09 +04:00
pinctl_name ,
pin_offset , pin_offset + npins - 1 ) ;
2012-10-27 13:51:36 +04:00
2016-02-11 13:03:06 +03:00
list_add_tail ( & pin_range - > node , & gdev - > pin_ranges ) ;
2012-11-06 19:03:35 +04:00
return 0 ;
2012-10-27 13:51:36 +04:00
}
2012-11-06 17:49:39 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_add_pin_range ) ;
2012-10-27 13:51:36 +04:00
2012-11-20 15:40:15 +04:00
/**
* gpiochip_remove_pin_ranges ( ) - remove all the GPIO < - > pin mappings
* @ chip : the chip to remove all the mappings for
*/
2012-10-27 13:51:36 +04:00
void gpiochip_remove_pin_ranges ( struct gpio_chip * chip )
{
struct gpio_pin_range * pin_range , * tmp ;
2016-02-11 13:03:06 +03:00
struct gpio_device * gdev = chip - > gpiodev ;
2012-10-27 13:51:36 +04:00
2016-02-11 13:03:06 +03:00
list_for_each_entry_safe ( pin_range , tmp , & gdev - > pin_ranges , node ) {
2012-10-27 13:51:36 +04:00
list_del ( & pin_range - > node ) ;
pinctrl_remove_gpio_range ( pin_range - > pctldev ,
& pin_range - > range ) ;
2012-11-20 15:40:15 +04:00
kfree ( pin_range ) ;
2012-10-27 13:51:36 +04:00
}
}
2012-11-06 17:49:39 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_remove_pin_ranges ) ;
# endif /* CONFIG_PINCTRL */
2012-10-27 13:51:36 +04:00
2008-02-05 09:28:20 +03:00
/* These "optional" allocation calls help prevent drivers from stomping
* on each other , and help provide better diagnostics in debugfs .
* They ' re called even less than the " set direction " calls .
*/
2017-09-26 21:58:28 +03:00
static int gpiod_request_commit ( struct gpio_desc * desc , const char * label )
2008-02-05 09:28:20 +03:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_chip * chip = desc - > gdev - > chip ;
2019-07-16 12:11:45 +03:00
int ret ;
2008-02-05 09:28:20 +03:00
unsigned long flags ;
2018-08-07 10:15:18 +03:00
unsigned offset ;
2008-02-05 09:28:20 +03:00
2018-11-01 16:12:50 +03:00
if ( label ) {
label = kstrdup_const ( label , GFP_KERNEL ) ;
if ( ! label )
return - ENOMEM ;
}
2013-02-15 09:46:14 +04:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
2008-02-05 09:28:20 +03:00
/* NOTE: gpio_request() can be called in early boot,
2008-10-16 09:03:16 +04:00
* before IRQs are enabled , for non - sleeping ( SOC ) GPIOs .
2008-02-05 09:28:20 +03:00
*/
if ( test_and_set_bit ( FLAG_REQUESTED , & desc - > flags ) = = 0 ) {
desc_set_label ( desc , label ? : " ? " ) ;
2019-07-16 12:11:45 +03:00
ret = 0 ;
2008-04-28 13:14:44 +04:00
} else {
2018-11-01 16:12:50 +03:00
kfree_const ( label ) ;
2019-07-16 12:11:45 +03:00
ret = - EBUSY ;
2009-01-30 01:25:12 +03:00
goto done ;
2008-10-16 09:03:16 +04:00
}
if ( chip - > request ) {
/* chip->request may sleep */
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2018-08-07 10:15:18 +03:00
offset = gpio_chip_hwgpio ( desc ) ;
if ( gpiochip_line_is_valid ( chip , offset ) )
2019-07-16 12:11:45 +03:00
ret = chip - > request ( chip , offset ) ;
2018-08-07 10:15:18 +03:00
else
2019-07-16 12:11:45 +03:00
ret = - EINVAL ;
2008-10-16 09:03:16 +04:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
2019-07-16 12:11:45 +03:00
if ( ret < 0 ) {
2008-10-16 09:03:16 +04:00
desc_set_label ( desc , NULL ) ;
2018-11-01 16:12:50 +03:00
kfree_const ( label ) ;
2008-10-16 09:03:16 +04:00
clear_bit ( FLAG_REQUESTED , & desc - > flags ) ;
2012-10-24 18:25:27 +04:00
goto done ;
2008-10-16 09:03:16 +04:00
}
2008-04-28 13:14:44 +04:00
}
2012-10-24 18:25:27 +04:00
if ( chip - > get_direction ) {
/* chip->get_direction may sleep */
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2013-02-02 20:29:29 +04:00
gpiod_get_direction ( desc ) ;
2012-10-24 18:25:27 +04:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
}
2014-03-10 16:54:50 +04:00
done :
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2019-07-16 12:11:45 +03:00
return ret ;
2014-03-10 16:54:50 +04:00
}
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
/*
* This descriptor validation needs to be inserted verbatim into each
* function taking a descriptor , so we need to use a preprocessor
2016-05-30 17:48:39 +03:00
* macro to avoid endless duplication . If the desc is NULL it is an
* optional GPIO and calls should just bail out .
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
*/
2017-12-21 03:27:04 +03:00
static int validate_desc ( const struct gpio_desc * desc , const char * func )
{
if ( ! desc )
return 0 ;
if ( IS_ERR ( desc ) ) {
pr_warn ( " %s: invalid GPIO (errorpointer) \n " , func ) ;
return PTR_ERR ( desc ) ;
}
if ( ! desc - > gdev ) {
pr_warn ( " %s: invalid GPIO (no device) \n " , func ) ;
return - EINVAL ;
}
if ( ! desc - > gdev - > chip ) {
dev_warn ( & desc - > gdev - > dev ,
" %s: backing chip is gone \n " , func ) ;
return 0 ;
}
return 1 ;
}
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
# define VALIDATE_DESC(desc) do { \
2017-12-21 03:27:04 +03:00
int __valid = validate_desc ( desc , __func__ ) ; \
if ( __valid < = 0 ) \
return __valid ; \
} while ( 0 )
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
# define VALIDATE_DESC_VOID(desc) do { \
2017-12-21 03:27:04 +03:00
int __valid = validate_desc ( desc , __func__ ) ; \
if ( __valid < = 0 ) \
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
return ; \
2017-12-21 03:27:04 +03:00
} while ( 0 )
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
2014-07-01 09:45:15 +04:00
int gpiod_request ( struct gpio_desc * desc , const char * label )
2014-03-10 16:54:50 +04:00
{
2019-07-16 12:11:45 +03:00
int ret = - EPROBE_DEFER ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_device * gdev ;
2014-03-10 16:54:50 +04:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
gdev = desc - > gdev ;
2014-03-10 16:54:50 +04:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
if ( try_module_get ( gdev - > owner ) ) {
2019-07-16 12:11:45 +03:00
ret = gpiod_request_commit ( desc , label ) ;
if ( ret < 0 )
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
module_put ( gdev - > owner ) ;
2016-02-11 12:28:44 +03:00
else
get_device ( & gdev - > dev ) ;
2014-03-10 16:54:50 +04:00
}
2019-07-16 12:11:45 +03:00
if ( ret )
gpiod_dbg ( desc , " %s: status %d \n " , __func__ , ret ) ;
2014-03-10 16:54:50 +04:00
2019-07-16 12:11:45 +03:00
return ret ;
2008-02-05 09:28:20 +03:00
}
2013-02-02 20:29:29 +04:00
2017-09-26 21:58:28 +03:00
static bool gpiod_free_commit ( struct gpio_desc * desc )
2008-02-05 09:28:20 +03:00
{
2014-03-10 16:54:50 +04:00
bool ret = false ;
2008-02-05 09:28:20 +03:00
unsigned long flags ;
2008-10-16 09:03:16 +04:00
struct gpio_chip * chip ;
2008-02-05 09:28:20 +03:00
2008-10-16 09:03:12 +04:00
might_sleep ( ) ;
2013-02-02 20:29:29 +04:00
gpiod_unexport ( desc ) ;
gpio: sysfs interface
This adds a simple sysfs interface for GPIOs.
/sys/class/gpio
/export ... asks the kernel to export a GPIO to userspace
/unexport ... to return a GPIO to the kernel
/gpioN ... for each exported GPIO #N
/value ... always readable, writes fail for input GPIOs
/direction ... r/w as: in, out (default low); write high, low
/gpiochipN ... for each gpiochip; #N is its first GPIO
/base ... (r/o) same as N
/label ... (r/o) descriptive, not necessarily unique
/ngpio ... (r/o) number of GPIOs; numbered N .. N+(ngpio - 1)
GPIOs claimed by kernel code may be exported by its owner using a new
gpio_export() call, which should be most useful for driver debugging.
Such exports may optionally be done without a "direction" attribute.
Userspace may ask to take over a GPIO by writing to a sysfs control file,
helping to cope with incomplete board support or other "one-off"
requirements that don't merit full kernel support:
echo 23 > /sys/class/gpio/export
... will gpio_request(23, "sysfs") and gpio_export(23);
use /sys/class/gpio/gpio-23/direction to (re)configure it,
when that GPIO can be used as both input and output.
echo 23 > /sys/class/gpio/unexport
... will gpio_free(23), when it was exported as above
The extra D-space footprint is a few hundred bytes, except for the sysfs
resources associated with each exported GPIO. The additional I-space
footprint is about two thirds of the current size of gpiolib (!). Since
no /dev node creation is involved, no "udev" support is needed.
Related changes:
* This adds a device pointer to "struct gpio_chip". When GPIO
providers initialize that, sysfs gpio class devices become children of
that device instead of being "virtual" devices.
* The (few) gpio_chip providers which have such a device node have
been updated.
* Some gpio_chip drivers also needed to update their module "owner"
field ... for which missing kerneldoc was added.
* Some gpio_chips don't support input GPIOs. Those GPIOs are now
flagged appropriately when the chip is registered.
Based on previous patches, and discussion both on and off LKML.
A Documentation/ABI/testing/sysfs-gpio update is ready to submit once this
merges to mainline.
[akpm@linux-foundation.org: a few maintenance build fixes]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Guennadi Liakhovetski <g.liakhovetski@pengutronix.de>
Cc: Greg KH <greg@kroah.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-25 12:46:07 +04:00
2008-02-05 09:28:20 +03:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
chip = desc - > gdev - > chip ;
2008-10-16 09:03:16 +04:00
if ( chip & & test_bit ( FLAG_REQUESTED , & desc - > flags ) ) {
if ( chip - > free ) {
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2010-08-11 05:02:24 +04:00
might_sleep_if ( chip - > can_sleep ) ;
2013-02-02 20:29:29 +04:00
chip - > free ( chip , gpio_chip_hwgpio ( desc ) ) ;
2008-10-16 09:03:16 +04:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
}
2018-11-01 16:12:50 +03:00
kfree_const ( desc - > label ) ;
2008-02-05 09:28:20 +03:00
desc_set_label ( desc , NULL ) ;
2009-12-16 03:46:20 +03:00
clear_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) ;
2008-10-16 09:03:16 +04:00
clear_bit ( FLAG_REQUESTED , & desc - > flags ) ;
2012-02-17 18:56:21 +04:00
clear_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) ;
2012-02-17 18:56:22 +04:00
clear_bit ( FLAG_OPEN_SOURCE , & desc - > flags ) ;
2019-09-21 13:25:23 +03:00
clear_bit ( FLAG_PULL_UP , & desc - > flags ) ;
clear_bit ( FLAG_PULL_DOWN , & desc - > flags ) ;
2015-02-02 20:44:44 +03:00
clear_bit ( FLAG_IS_HOGGED , & desc - > flags ) ;
2014-03-10 16:54:50 +04:00
ret = true ;
}
2008-02-05 09:28:20 +03:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2014-03-10 16:54:50 +04:00
return ret ;
}
2014-07-01 09:45:15 +04:00
void gpiod_free ( struct gpio_desc * desc )
2014-03-10 16:54:50 +04:00
{
2017-09-26 21:58:28 +03:00
if ( desc & & desc - > gdev & & gpiod_free_commit ( desc ) ) {
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
module_put ( desc - > gdev - > owner ) ;
2016-02-11 12:28:44 +03:00
put_device ( & desc - > gdev - > dev ) ;
} else {
2014-03-10 16:54:50 +04:00
WARN_ON ( extra_checks ) ;
2016-02-11 12:28:44 +03:00
}
2008-02-05 09:28:20 +03:00
}
2013-02-02 20:29:29 +04:00
2008-02-05 09:28:20 +03:00
/**
* gpiochip_is_requested - return string iff signal was requested
* @ chip : controller managing the signal
* @ offset : of signal within controller ' s 0. . ( ngpio - 1 ) range
*
* Returns NULL if the GPIO is not currently requested , else a string .
2014-07-01 09:45:14 +04:00
* The string returned is the label passed to gpio_request ( ) ; if none has been
* passed it is a meaningless , non - NULL constant .
2008-02-05 09:28:20 +03:00
*
* This function is for use by GPIO controller drivers . The label can
* help with diagnostics , and knowing that the signal is used as a GPIO
* can help avoid accidentally multiplexing it to another controller .
*/
const char * gpiochip_is_requested ( struct gpio_chip * chip , unsigned offset )
{
2013-02-02 20:29:30 +04:00
struct gpio_desc * desc ;
2008-02-05 09:28:20 +03:00
2015-08-18 19:02:32 +03:00
if ( offset > = chip - > ngpio )
2008-02-05 09:28:20 +03:00
return NULL ;
2013-02-02 20:29:30 +04:00
2016-02-09 15:51:59 +03:00
desc = & chip - > gpiodev - > descs [ offset ] ;
2013-02-02 20:29:30 +04:00
2013-02-02 20:29:29 +04:00
if ( test_bit ( FLAG_REQUESTED , & desc - > flags ) = = 0 )
2008-02-05 09:28:20 +03:00
return NULL ;
2013-02-02 20:29:29 +04:00
return desc - > label ;
2008-02-05 09:28:20 +03:00
}
EXPORT_SYMBOL_GPL ( gpiochip_is_requested ) ;
2014-03-10 16:54:50 +04:00
/**
* gpiochip_request_own_desc - Allow GPIO chip to request its own descriptor
2017-07-24 17:57:22 +03:00
* @ chip : GPIO chip
* @ hwnum : hardware number of the GPIO for which to request the descriptor
2014-03-10 16:54:50 +04:00
* @ label : label for the GPIO
2019-04-26 15:40:18 +03:00
* @ lflags : lookup flags for this GPIO or 0 if default , this can be used to
* specify things like line inversion semantics with the machine flags
* such as GPIO_OUT_LOW
* @ dflags : descriptor request flags for this GPIO or 0 if default , this
* can be used to specify consumer semantics such as open drain
2014-03-10 16:54:50 +04:00
*
* Function allows GPIO chip drivers to request and use their own GPIO
* descriptors via gpiolib API . Difference to gpiod_request ( ) is that this
* function will not increase reference count of the GPIO chip module . This
* allows the GPIO chip module to be unloaded as needed ( we assume that the
* GPIO chip driver handles freeing the GPIOs it has requested ) .
2017-07-24 17:57:22 +03:00
*
* Returns :
* A pointer to the GPIO descriptor , or an ERR_PTR ( ) - encoded negative error
* code on failure .
2014-03-10 16:54:50 +04:00
*/
2014-08-19 21:06:09 +04:00
struct gpio_desc * gpiochip_request_own_desc ( struct gpio_chip * chip , u16 hwnum ,
2018-09-04 14:31:45 +03:00
const char * label ,
2019-04-26 15:40:18 +03:00
enum gpio_lookup_flags lflags ,
enum gpiod_flags dflags )
2014-03-10 16:54:50 +04:00
{
2014-08-19 21:06:09 +04:00
struct gpio_desc * desc = gpiochip_get_desc ( chip , hwnum ) ;
2019-07-16 12:11:45 +03:00
int ret ;
2014-03-10 16:54:50 +04:00
2014-08-19 21:06:09 +04:00
if ( IS_ERR ( desc ) ) {
chip_err ( chip , " failed to get GPIO descriptor \n " ) ;
return desc ;
}
2019-07-16 12:11:45 +03:00
ret = gpiod_request_commit ( desc , label ) ;
if ( ret < 0 )
return ERR_PTR ( ret ) ;
2014-03-10 16:54:50 +04:00
2019-07-16 12:11:45 +03:00
ret = gpiod_configure_flags ( desc , label , lflags , dflags ) ;
if ( ret ) {
2018-09-04 14:31:45 +03:00
chip_err ( chip , " setup of own GPIO %s failed \n " , label ) ;
gpiod_free_commit ( desc ) ;
2019-07-16 12:11:45 +03:00
return ERR_PTR ( ret ) ;
2018-09-04 14:31:45 +03:00
}
2014-08-19 21:06:09 +04:00
return desc ;
2014-03-10 16:54:50 +04:00
}
2014-07-22 19:01:01 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_request_own_desc ) ;
2014-03-10 16:54:50 +04:00
/**
* gpiochip_free_own_desc - Free GPIO requested by the chip driver
* @ desc : GPIO descriptor to free
*
* Function frees the given GPIO requested previously with
* gpiochip_request_own_desc ( ) .
*/
void gpiochip_free_own_desc ( struct gpio_desc * desc )
{
if ( desc )
2017-09-26 21:58:28 +03:00
gpiod_free_commit ( desc ) ;
2014-03-10 16:54:50 +04:00
}
2014-07-22 19:01:01 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_free_own_desc ) ;
2008-02-05 09:28:20 +03:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
/*
* Drivers MUST set GPIO direction before making get / set calls . In
2008-02-05 09:28:20 +03:00
* some cases this is done in early boot , before IRQs are enabled .
*
* As a rule these aren ' t called more than once ( except for drivers
* using the open - drain emulation idiom ) so these are natural places
* to accumulate extra debugging checks . Note that we can ' t ( yet )
* rely on gpio_request ( ) having been called beforehand .
*/
2019-02-07 19:28:56 +03:00
static int gpio_set_config ( struct gpio_chip * gc , unsigned offset ,
enum pin_config_param mode )
{
2019-03-14 22:32:50 +03:00
unsigned long config ;
unsigned arg ;
2019-02-07 19:28:56 +03:00
2019-03-14 22:32:50 +03:00
switch ( mode ) {
case PIN_CONFIG_BIAS_PULL_DOWN :
case PIN_CONFIG_BIAS_PULL_UP :
arg = 1 ;
break ;
default :
arg = 0 ;
}
config = PIN_CONF_PACKED ( mode , arg ) ;
2019-02-07 19:28:56 +03:00
return gc - > set_config ? gc - > set_config ( gc , offset , config ) : - ENOTSUPP ;
}
2013-10-17 21:21:36 +04:00
/**
* gpiod_direction_input - set the GPIO direction to input
* @ desc : GPIO to set to input
*
* Set the direction of the passed GPIO to input , such as gpiod_get_value ( ) can
* be called safely on it .
*
* Return 0 in case of success , else an error code .
*/
int gpiod_direction_input ( struct gpio_desc * desc )
2008-02-05 09:28:20 +03:00
{
struct gpio_chip * chip ;
2019-07-16 12:11:45 +03:00
int ret = 0 ;
2008-02-05 09:28:20 +03:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
chip = desc - > gdev - > chip ;
2013-02-15 09:46:14 +04:00
2018-09-25 10:54:14 +03:00
/*
* It is legal to have no . get ( ) and . direction_input ( ) specified if
* the chip is output - only , but you can ' t specify . direction_input ( )
* and not support the . get ( ) operation , that doesn ' t make sense .
*/
2018-09-21 13:36:03 +03:00
if ( ! chip - > get & & chip - > direction_input ) {
2013-09-09 13:33:49 +04:00
gpiod_warn ( desc ,
2018-09-25 10:54:14 +03:00
" %s: missing get() but have direction_input() \n " ,
__func__ ) ;
2013-08-30 11:41:45 +04:00
return - EIO ;
}
2018-09-25 10:54:14 +03:00
/*
* If we have a . direction_input ( ) callback , things are simple ,
* just call it . Else we are some input - only chip so try to check the
* direction ( if . get_direction ( ) is supported ) else we silently
* assume we are in input mode after this .
*/
2018-09-21 13:36:03 +03:00
if ( chip - > direction_input ) {
2019-07-16 12:11:45 +03:00
ret = chip - > direction_input ( chip , gpio_chip_hwgpio ( desc ) ) ;
2018-09-21 13:36:03 +03:00
} else if ( chip - > get_direction & &
( chip - > get_direction ( chip , gpio_chip_hwgpio ( desc ) ) ! = 1 ) ) {
gpiod_warn ( desc ,
2018-09-25 10:54:14 +03:00
" %s: missing direction_input() operation and line is output \n " ,
__func__ ) ;
2018-09-21 13:36:03 +03:00
return - EIO ;
}
2019-07-16 12:11:45 +03:00
if ( ret = = 0 )
2008-02-05 09:28:20 +03:00
clear_bit ( FLAG_IS_OUT , & desc - > flags ) ;
2011-05-20 10:40:19 +04:00
2019-02-07 19:28:58 +03:00
if ( test_bit ( FLAG_PULL_UP , & desc - > flags ) )
gpio_set_config ( chip , gpio_chip_hwgpio ( desc ) ,
PIN_CONFIG_BIAS_PULL_UP ) ;
else if ( test_bit ( FLAG_PULL_DOWN , & desc - > flags ) )
gpio_set_config ( chip , gpio_chip_hwgpio ( desc ) ,
PIN_CONFIG_BIAS_PULL_DOWN ) ;
2019-07-16 12:11:45 +03:00
trace_gpio_direction ( desc_to_gpio ( desc ) , 1 , ret ) ;
2014-07-22 11:17:43 +04:00
2019-07-16 12:11:45 +03:00
return ret ;
2008-02-05 09:28:20 +03:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_direction_input ) ;
2013-02-02 20:29:29 +04:00
2017-09-26 21:58:28 +03:00
static int gpiod_direction_output_raw_commit ( struct gpio_desc * desc , int value )
2008-02-05 09:28:20 +03:00
{
2016-03-22 12:51:16 +03:00
struct gpio_chip * gc = desc - > gdev - > chip ;
2016-11-14 01:02:44 +03:00
int val = ! ! value ;
2018-09-21 13:36:03 +03:00
int ret = 0 ;
2008-02-05 09:28:20 +03:00
2018-09-25 10:54:14 +03:00
/*
* It ' s OK not to specify . direction_output ( ) if the gpiochip is
* output - only , but if there is then not even a . set ( ) operation it
* is pretty tricky to drive the output line .
*/
2018-09-21 13:36:03 +03:00
if ( ! gc - > set & & ! gc - > direction_output ) {
2013-09-09 13:33:49 +04:00
gpiod_warn ( desc ,
2018-09-25 10:54:14 +03:00
" %s: missing set() and direction_output() operations \n " ,
__func__ ) ;
2013-08-30 11:41:45 +04:00
return - EIO ;
}
2018-09-21 13:36:03 +03:00
if ( gc - > direction_output ) {
ret = gc - > direction_output ( gc , gpio_chip_hwgpio ( desc ) , val ) ;
} else {
2018-09-25 10:54:14 +03:00
/* Check that we are in output mode if we can */
2018-09-21 13:36:03 +03:00
if ( gc - > get_direction & &
gc - > get_direction ( gc , gpio_chip_hwgpio ( desc ) ) ) {
gpiod_warn ( desc ,
" %s: missing direction_output() operation \n " ,
__func__ ) ;
return - EIO ;
}
2018-09-25 10:54:14 +03:00
/*
* If we can ' t actively set the direction , we are some
* output - only chip , so just drive the output as desired .
*/
2018-09-21 13:36:03 +03:00
gc - > set ( gc , gpio_chip_hwgpio ( desc ) , val ) ;
}
2016-03-22 12:51:16 +03:00
if ( ! ret )
2008-02-05 09:28:20 +03:00
set_bit ( FLAG_IS_OUT , & desc - > flags ) ;
2016-11-14 01:02:44 +03:00
trace_gpio_value ( desc_to_gpio ( desc ) , 0 , val ) ;
2016-03-22 12:51:16 +03:00
trace_gpio_direction ( desc_to_gpio ( desc ) , 0 , ret ) ;
return ret ;
2008-02-05 09:28:20 +03:00
}
2014-01-07 15:34:11 +04:00
/**
* gpiod_direction_output_raw - set the GPIO direction to output
* @ desc : GPIO to set to output
* @ value : initial output value of the GPIO
*
* Set the direction of the passed GPIO to output , such as gpiod_set_value ( ) can
* be called safely on it . The initial value of the output must be specified
* as raw value on the physical line without regard for the ACTIVE_LOW status .
*
* Return 0 in case of success , else an error code .
*/
int gpiod_direction_output_raw ( struct gpio_desc * desc , int value )
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
2017-09-26 21:58:28 +03:00
return gpiod_direction_output_raw_commit ( desc , value ) ;
2014-01-07 15:34:11 +04:00
}
EXPORT_SYMBOL_GPL ( gpiod_direction_output_raw ) ;
/**
2014-02-08 10:25:25 +04:00
* gpiod_direction_output - set the GPIO direction to output
2014-01-07 15:34:11 +04:00
* @ desc : GPIO to set to output
* @ value : initial output value of the GPIO
*
* Set the direction of the passed GPIO to output , such as gpiod_set_value ( ) can
* be called safely on it . The initial value of the output must be specified
* as the logical value of the GPIO , i . e . taking its ACTIVE_LOW status into
* account .
*
* Return 0 in case of success , else an error code .
*/
int gpiod_direction_output ( struct gpio_desc * desc , int value )
{
2017-12-21 19:37:24 +03:00
struct gpio_chip * gc ;
2017-09-26 22:20:23 +03:00
int ret ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
2014-01-07 15:34:11 +04:00
if ( test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) )
value = ! value ;
2016-11-14 01:02:44 +03:00
else
value = ! ! value ;
2017-09-26 22:20:23 +03:00
2018-09-08 12:23:16 +03:00
/* GPIOs used for enabled IRQs shall not be set as output */
if ( test_bit ( FLAG_USED_AS_IRQ , & desc - > flags ) & &
test_bit ( FLAG_IRQ_IS_ENABLED , & desc - > flags ) ) {
2017-09-26 22:20:23 +03:00
gpiod_err ( desc ,
" %s: tried to set a GPIO tied to an IRQ as output \n " ,
__func__ ) ;
return - EIO ;
}
2017-12-21 19:37:24 +03:00
gc = desc - > gdev - > chip ;
2017-09-26 22:20:23 +03:00
if ( test_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) ) {
/* First see if we can enable open drain in hardware */
2019-02-07 19:28:56 +03:00
ret = gpio_set_config ( gc , gpio_chip_hwgpio ( desc ) ,
PIN_CONFIG_DRIVE_OPEN_DRAIN ) ;
2017-09-26 22:20:23 +03:00
if ( ! ret )
goto set_output_value ;
/* Emulate open drain by not actively driving the line high */
2019-10-01 12:44:53 +03:00
if ( value ) {
ret = gpiod_direction_input ( desc ) ;
goto set_output_flag ;
}
2017-09-26 22:20:23 +03:00
}
else if ( test_bit ( FLAG_OPEN_SOURCE , & desc - > flags ) ) {
2019-02-07 19:28:56 +03:00
ret = gpio_set_config ( gc , gpio_chip_hwgpio ( desc ) ,
PIN_CONFIG_DRIVE_OPEN_SOURCE ) ;
2017-09-26 22:20:23 +03:00
if ( ! ret )
goto set_output_value ;
/* Emulate open source by not actively driving the line low */
2019-10-01 12:44:53 +03:00
if ( ! value ) {
ret = gpiod_direction_input ( desc ) ;
goto set_output_flag ;
}
2017-09-26 22:20:23 +03:00
} else {
2019-02-07 19:28:56 +03:00
gpio_set_config ( gc , gpio_chip_hwgpio ( desc ) ,
PIN_CONFIG_DRIVE_PUSH_PULL ) ;
2017-09-26 22:20:23 +03:00
}
set_output_value :
2017-09-26 21:58:28 +03:00
return gpiod_direction_output_raw_commit ( desc , value ) ;
2019-10-01 12:44:53 +03:00
set_output_flag :
/*
* When emulating open - source or open - drain functionalities by not
* actively driving the line ( setting mode to input ) we still need to
* set the IS_OUT flag or otherwise we won ' t be able to set the line
* value anymore .
*/
if ( ret = = 0 )
set_bit ( FLAG_IS_OUT , & desc - > flags ) ;
return ret ;
2014-01-07 15:34:11 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_direction_output ) ;
2008-02-05 09:28:20 +03:00
2010-05-27 01:42:23 +04:00
/**
2017-07-24 17:57:22 +03:00
* gpiod_set_debounce - sets @ debounce time for a GPIO
* @ desc : descriptor of the GPIO for which to set debounce time
* @ debounce : debounce time in microseconds
2013-09-04 16:17:08 +04:00
*
2017-07-24 17:57:22 +03:00
* Returns :
* 0 on success , % - ENOTSUPP if the controller doesn ' t support setting the
* debounce time .
2010-05-27 01:42:23 +04:00
*/
2013-10-17 21:21:36 +04:00
int gpiod_set_debounce ( struct gpio_desc * desc , unsigned debounce )
2010-05-27 01:42:23 +04:00
{
struct gpio_chip * chip ;
2017-01-23 15:34:34 +03:00
unsigned long config ;
2010-05-27 01:42:23 +04:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
chip = desc - > gdev - > chip ;
2017-01-23 15:34:34 +03:00
if ( ! chip - > set | | ! chip - > set_config ) {
2013-09-09 13:33:49 +04:00
gpiod_dbg ( desc ,
2017-01-23 15:34:34 +03:00
" %s: missing set() or set_config() operations \n " ,
2013-09-09 13:33:49 +04:00
__func__ ) ;
2013-09-04 16:17:08 +04:00
return - ENOTSUPP ;
2013-08-30 11:41:45 +04:00
}
2017-01-23 15:34:34 +03:00
config = pinconf_to_config_packed ( PIN_CONFIG_INPUT_DEBOUNCE , debounce ) ;
Revert "gpio: use new gpio_set_config() helper in more places"
gpio-aspeed implements support for PIN_CONFIG_INPUT_DEBOUNCE. As of
v5.1-rc1 we're seeing the following when booting a Romulus BMC kernel:
> [ 21.373137] ------------[ cut here ]------------
> [ 21.374545] WARNING: CPU: 0 PID: 1 at drivers/gpio/gpio-aspeed.c:834 unregister_allocated_timer+0x38/0x94
> [ 21.376181] No timer allocated to offset 74
> [ 21.377672] CPU: 0 PID: 1 Comm: swapper Not tainted 5.1.0-rc1-dirty #6
> [ 21.378800] Hardware name: Generic DT based system
> [ 21.379965] Backtrace:
> [ 21.381024] [<80107d44>] (dump_backtrace) from [<80107f78>] (show_stack+0x20/0x24)
> [ 21.382713] r7:8038b720 r6:00000009 r5:00000000 r4:87897c64
> [ 21.383815] [<80107f58>] (show_stack) from [<80656398>] (dump_stack+0x20/0x28)
> [ 21.385042] [<80656378>] (dump_stack) from [<80115f1c>] (__warn.part.3+0xb4/0xdc)
> [ 21.386253] [<80115e68>] (__warn.part.3) from [<80115fb0>] (warn_slowpath_fmt+0x6c/0x90)
> [ 21.387471] r6:00000342 r5:807f8758 r4:80a07008
> [ 21.388278] [<80115f48>] (warn_slowpath_fmt) from [<8038b720>] (unregister_allocated_timer+0x38/0x94)
> [ 21.389809] r3:0000004a r2:807f8774
> [ 21.390526] r7:00000000 r6:0000000a r5:60000153 r4:0000004a
> [ 21.391601] [<8038b6e8>] (unregister_allocated_timer) from [<8038baac>] (aspeed_gpio_set_config+0x330/0x48c)
> [ 21.393248] [<8038b77c>] (aspeed_gpio_set_config) from [<803840b0>] (gpiod_set_debounce+0xe8/0x114)
> [ 21.394745] r10:82ee2248 r9:00000000 r8:87b63a00 r7:00001388 r6:87947320 r5:80729310
> [ 21.396030] r4:879f64a0
> [ 21.396499] [<80383fc8>] (gpiod_set_debounce) from [<804b4350>] (gpio_keys_probe+0x69c/0x8e0)
> [ 21.397715] r7:845d94b8 r6:00000001 r5:00000000 r4:87b63a1c
> [ 21.398618] [<804b3cb4>] (gpio_keys_probe) from [<8040eeec>] (platform_dev_probe+0x44/0x80)
> [ 21.399834] r10:00000003 r9:80a3a8b0 r8:00000000 r7:00000000 r6:80a7f9dc r5:80a3a8b0
> [ 21.401163] r4:8796bc10
> [ 21.401634] [<8040eea8>] (platform_drv_probe) from [<8040d0d4>] (really_probe+0x208/0x3dc)
> [ 21.402786] r5:80a7f8d0 r4:8796bc10
> [ 21.403547] [<8040cecc>] (really_probe) from [<8040d7a4>] (driver_probe_device+0x130/0x170)
> [ 21.404744] r10:0000007b r9:8093683c r8:00000000 r7:80a07008 r6:80a3a8b0 r5:8796bc10
> [ 21.405854] r4:80a3a8b0
> [ 21.406324] [<8040d674>] (driver_probe_device) from [<8040da8c>] (device_driver_attach+0x68/0x70)
> [ 21.407568] r9:8093683c r8:00000000 r7:80a07008 r6:80a3a8b0 r5:00000000 r4:8796bc10
> [ 21.408877] [<8040da24>] (device_driver_attach) from [<8040db14>] (__driver_attach+0x80/0x150)
> [ 21.410327] r7:80a07008 r6:8796bc10 r5:00000001 r4:80a3a8b0
> [ 21.411294] [<8040da94>] (__driver_attach) from [<8040b20c>] (bus_for_each_dev+0x80/0xc4)
> [ 21.412641] r7:80a07008 r6:8040da94 r5:80a3a8b0 r4:87966f30
> [ 21.413580] [<8040b18c>] (bus_for_each_dev) from [<8040dc0c>] (driver_attach+0x28/0x30)
> [ 21.414943] r7:00000000 r6:87b411e0 r5:80a33fc8 r4:80a3a8b0
> [ 21.415927] [<8040dbe4>] (driver_attach) from [<8040bbf0>] (bus_add_driver+0x14c/0x200)
> [ 21.417289] [<8040baa4>] (bus_add_driver) from [<8040e2b4>] (driver_register+0x84/0x118)
> [ 21.418652] r7:80a60ae0 r6:809226b8 r5:80a07008 r4:80a3a8b0
> [ 21.419652] [<8040e230>] (driver_register) from [<8040fc28>] (__platform_driver_register+0x3c/0x50)
> [ 21.421193] r5:80a07008 r4:809525f8
> [ 21.421990] [<8040fbec>] (__platform_driver_register) from [<809226d8>] (gpio_keys_init+0x20/0x28)
> [ 21.423447] [<809226b8>] (gpio_keys_init) from [<8090128c>] (do_one_initcall+0x80/0x180)
> [ 21.424886] [<8090120c>] (do_one_initcall) from [<80901538>] (kernel_init_freeable+0x1ac/0x26c)
> [ 21.426354] r8:80a60ae0 r7:80a60ae0 r6:8093685c r5:00000008 r4:809525f8
> [ 21.427579] [<8090138c>] (kernel_init_freeable) from [<8066d9a0>] (kernel_init+0x18/0x11c)
> [ 21.428819] r10:00000000 r9:00000000 r8:00000000 r7:00000000 r6:00000000 r5:8066d988
> [ 21.429947] r4:00000000
> [ 21.430415] [<8066d988>] (kernel_init) from [<801010e8>] (ret_from_fork+0x14/0x2c)
> [ 21.431666] Exception stack(0x87897fb0 to 0x87897ff8)
> [ 21.432877] 7fa0: 00000000 00000000 00000000 00000000
> [ 21.434446] 7fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> [ 21.436052] 7fe0: 00000000 00000000 00000000 00000000 00000013 00000000
> [ 21.437308] r5:8066d988 r4:00000000
> [ 21.438102] ---[ end trace d7d7ac3a80567d0e ]---
We only hit unregister_allocated_timer() if the argument to
aspeed_gpio_set_config() is 0, but we can't be calling through
gpiod_set_debounce() from gpio_keys_probe() unless the gpio-keys button has a
non-zero debounce interval.
Commit 6581eaf0e890 ("gpio: use new gpio_set_config() helper in more places")
spreads the use of gpio_set_config() to the debounce and transitory
state configuration paths. The implementation of gpio_set_config() is:
> static int gpio_set_config(struct gpio_chip *gc, unsigned offset,
> enum pin_config_param mode)
> {
> unsigned long config = { PIN_CONF_PACKED(mode, 0) };
>
> return gc->set_config ? gc->set_config(gc, offset, config) : -ENOTSUPP;
> }
Here it packs its own config value with a fixed argument of 0; this is
incorrect behaviour for implementing the debounce and transitory functions, and
the debounce and transitory gpio_set_config() call-sites now have an undetected
type mismatch as they both already pack their own config parameter (i.e. what
gets passed is not an `enum pin_config_param`). Indeed this can be seen in the
small diff for 6581eaf0e890:
> diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
> index de595fa31a1a..1f239aac43df 100644
> --- a/drivers/gpio/gpiolib.c
> +++ b/drivers/gpio/gpiolib.c
> @@ -2725,7 +2725,7 @@ int gpiod_set_debounce(struct gpio_desc *desc, unsigned debounce)
> }
>
> config = pinconf_to_config_packed(PIN_CONFIG_INPUT_DEBOUNCE, debounce);
> - return chip->set_config(chip, gpio_chip_hwgpio(desc), config);
> + return gpio_set_config(chip, gpio_chip_hwgpio(desc), config);
> }
> EXPORT_SYMBOL_GPL(gpiod_set_debounce);
>
> @@ -2762,7 +2762,7 @@ int gpiod_set_transitory(struct gpio_desc *desc, bool transitory)
> packed = pinconf_to_config_packed(PIN_CONFIG_PERSIST_STATE,
> !transitory);
> gpio = gpio_chip_hwgpio(desc);
> - rc = chip->set_config(chip, gpio, packed);
> + rc = gpio_set_config(chip, gpio, packed);
> if (rc == -ENOTSUPP) {
> dev_dbg(&desc->gdev->dev, "Persistence not supported for GPIO %d\n",
> gpio);
Revert commit 6581eaf0e890 ("gpio: use new gpio_set_config() helper in
more places") to restore correct behaviour for gpiod_set_debounce() and
gpiod_set_transitory().
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2019-03-26 07:49:54 +03:00
return chip - > set_config ( chip , gpio_chip_hwgpio ( desc ) , config ) ;
2010-05-27 01:42:23 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_set_debounce ) ;
2013-02-02 20:29:29 +04:00
2017-11-30 06:55:24 +03:00
/**
* gpiod_set_transitory - Lose or retain GPIO state on suspend or reset
* @ desc : descriptor of the GPIO for which to configure persistence
* @ transitory : True to lose state on suspend or reset , false for persistence
*
* Returns :
* 0 on success , otherwise a negative error code .
*/
int gpiod_set_transitory ( struct gpio_desc * desc , bool transitory )
{
struct gpio_chip * chip ;
unsigned long packed ;
int gpio ;
int rc ;
2017-12-21 19:37:35 +03:00
VALIDATE_DESC ( desc ) ;
2017-11-30 06:55:24 +03:00
/*
* Handle FLAG_TRANSITORY first , enabling queries to gpiolib for
* persistence state .
*/
if ( transitory )
set_bit ( FLAG_TRANSITORY , & desc - > flags ) ;
else
clear_bit ( FLAG_TRANSITORY , & desc - > flags ) ;
/* If the driver supports it, set the persistence state now */
chip = desc - > gdev - > chip ;
if ( ! chip - > set_config )
return 0 ;
packed = pinconf_to_config_packed ( PIN_CONFIG_PERSIST_STATE ,
! transitory ) ;
gpio = gpio_chip_hwgpio ( desc ) ;
Revert "gpio: use new gpio_set_config() helper in more places"
gpio-aspeed implements support for PIN_CONFIG_INPUT_DEBOUNCE. As of
v5.1-rc1 we're seeing the following when booting a Romulus BMC kernel:
> [ 21.373137] ------------[ cut here ]------------
> [ 21.374545] WARNING: CPU: 0 PID: 1 at drivers/gpio/gpio-aspeed.c:834 unregister_allocated_timer+0x38/0x94
> [ 21.376181] No timer allocated to offset 74
> [ 21.377672] CPU: 0 PID: 1 Comm: swapper Not tainted 5.1.0-rc1-dirty #6
> [ 21.378800] Hardware name: Generic DT based system
> [ 21.379965] Backtrace:
> [ 21.381024] [<80107d44>] (dump_backtrace) from [<80107f78>] (show_stack+0x20/0x24)
> [ 21.382713] r7:8038b720 r6:00000009 r5:00000000 r4:87897c64
> [ 21.383815] [<80107f58>] (show_stack) from [<80656398>] (dump_stack+0x20/0x28)
> [ 21.385042] [<80656378>] (dump_stack) from [<80115f1c>] (__warn.part.3+0xb4/0xdc)
> [ 21.386253] [<80115e68>] (__warn.part.3) from [<80115fb0>] (warn_slowpath_fmt+0x6c/0x90)
> [ 21.387471] r6:00000342 r5:807f8758 r4:80a07008
> [ 21.388278] [<80115f48>] (warn_slowpath_fmt) from [<8038b720>] (unregister_allocated_timer+0x38/0x94)
> [ 21.389809] r3:0000004a r2:807f8774
> [ 21.390526] r7:00000000 r6:0000000a r5:60000153 r4:0000004a
> [ 21.391601] [<8038b6e8>] (unregister_allocated_timer) from [<8038baac>] (aspeed_gpio_set_config+0x330/0x48c)
> [ 21.393248] [<8038b77c>] (aspeed_gpio_set_config) from [<803840b0>] (gpiod_set_debounce+0xe8/0x114)
> [ 21.394745] r10:82ee2248 r9:00000000 r8:87b63a00 r7:00001388 r6:87947320 r5:80729310
> [ 21.396030] r4:879f64a0
> [ 21.396499] [<80383fc8>] (gpiod_set_debounce) from [<804b4350>] (gpio_keys_probe+0x69c/0x8e0)
> [ 21.397715] r7:845d94b8 r6:00000001 r5:00000000 r4:87b63a1c
> [ 21.398618] [<804b3cb4>] (gpio_keys_probe) from [<8040eeec>] (platform_dev_probe+0x44/0x80)
> [ 21.399834] r10:00000003 r9:80a3a8b0 r8:00000000 r7:00000000 r6:80a7f9dc r5:80a3a8b0
> [ 21.401163] r4:8796bc10
> [ 21.401634] [<8040eea8>] (platform_drv_probe) from [<8040d0d4>] (really_probe+0x208/0x3dc)
> [ 21.402786] r5:80a7f8d0 r4:8796bc10
> [ 21.403547] [<8040cecc>] (really_probe) from [<8040d7a4>] (driver_probe_device+0x130/0x170)
> [ 21.404744] r10:0000007b r9:8093683c r8:00000000 r7:80a07008 r6:80a3a8b0 r5:8796bc10
> [ 21.405854] r4:80a3a8b0
> [ 21.406324] [<8040d674>] (driver_probe_device) from [<8040da8c>] (device_driver_attach+0x68/0x70)
> [ 21.407568] r9:8093683c r8:00000000 r7:80a07008 r6:80a3a8b0 r5:00000000 r4:8796bc10
> [ 21.408877] [<8040da24>] (device_driver_attach) from [<8040db14>] (__driver_attach+0x80/0x150)
> [ 21.410327] r7:80a07008 r6:8796bc10 r5:00000001 r4:80a3a8b0
> [ 21.411294] [<8040da94>] (__driver_attach) from [<8040b20c>] (bus_for_each_dev+0x80/0xc4)
> [ 21.412641] r7:80a07008 r6:8040da94 r5:80a3a8b0 r4:87966f30
> [ 21.413580] [<8040b18c>] (bus_for_each_dev) from [<8040dc0c>] (driver_attach+0x28/0x30)
> [ 21.414943] r7:00000000 r6:87b411e0 r5:80a33fc8 r4:80a3a8b0
> [ 21.415927] [<8040dbe4>] (driver_attach) from [<8040bbf0>] (bus_add_driver+0x14c/0x200)
> [ 21.417289] [<8040baa4>] (bus_add_driver) from [<8040e2b4>] (driver_register+0x84/0x118)
> [ 21.418652] r7:80a60ae0 r6:809226b8 r5:80a07008 r4:80a3a8b0
> [ 21.419652] [<8040e230>] (driver_register) from [<8040fc28>] (__platform_driver_register+0x3c/0x50)
> [ 21.421193] r5:80a07008 r4:809525f8
> [ 21.421990] [<8040fbec>] (__platform_driver_register) from [<809226d8>] (gpio_keys_init+0x20/0x28)
> [ 21.423447] [<809226b8>] (gpio_keys_init) from [<8090128c>] (do_one_initcall+0x80/0x180)
> [ 21.424886] [<8090120c>] (do_one_initcall) from [<80901538>] (kernel_init_freeable+0x1ac/0x26c)
> [ 21.426354] r8:80a60ae0 r7:80a60ae0 r6:8093685c r5:00000008 r4:809525f8
> [ 21.427579] [<8090138c>] (kernel_init_freeable) from [<8066d9a0>] (kernel_init+0x18/0x11c)
> [ 21.428819] r10:00000000 r9:00000000 r8:00000000 r7:00000000 r6:00000000 r5:8066d988
> [ 21.429947] r4:00000000
> [ 21.430415] [<8066d988>] (kernel_init) from [<801010e8>] (ret_from_fork+0x14/0x2c)
> [ 21.431666] Exception stack(0x87897fb0 to 0x87897ff8)
> [ 21.432877] 7fa0: 00000000 00000000 00000000 00000000
> [ 21.434446] 7fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
> [ 21.436052] 7fe0: 00000000 00000000 00000000 00000000 00000013 00000000
> [ 21.437308] r5:8066d988 r4:00000000
> [ 21.438102] ---[ end trace d7d7ac3a80567d0e ]---
We only hit unregister_allocated_timer() if the argument to
aspeed_gpio_set_config() is 0, but we can't be calling through
gpiod_set_debounce() from gpio_keys_probe() unless the gpio-keys button has a
non-zero debounce interval.
Commit 6581eaf0e890 ("gpio: use new gpio_set_config() helper in more places")
spreads the use of gpio_set_config() to the debounce and transitory
state configuration paths. The implementation of gpio_set_config() is:
> static int gpio_set_config(struct gpio_chip *gc, unsigned offset,
> enum pin_config_param mode)
> {
> unsigned long config = { PIN_CONF_PACKED(mode, 0) };
>
> return gc->set_config ? gc->set_config(gc, offset, config) : -ENOTSUPP;
> }
Here it packs its own config value with a fixed argument of 0; this is
incorrect behaviour for implementing the debounce and transitory functions, and
the debounce and transitory gpio_set_config() call-sites now have an undetected
type mismatch as they both already pack their own config parameter (i.e. what
gets passed is not an `enum pin_config_param`). Indeed this can be seen in the
small diff for 6581eaf0e890:
> diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
> index de595fa31a1a..1f239aac43df 100644
> --- a/drivers/gpio/gpiolib.c
> +++ b/drivers/gpio/gpiolib.c
> @@ -2725,7 +2725,7 @@ int gpiod_set_debounce(struct gpio_desc *desc, unsigned debounce)
> }
>
> config = pinconf_to_config_packed(PIN_CONFIG_INPUT_DEBOUNCE, debounce);
> - return chip->set_config(chip, gpio_chip_hwgpio(desc), config);
> + return gpio_set_config(chip, gpio_chip_hwgpio(desc), config);
> }
> EXPORT_SYMBOL_GPL(gpiod_set_debounce);
>
> @@ -2762,7 +2762,7 @@ int gpiod_set_transitory(struct gpio_desc *desc, bool transitory)
> packed = pinconf_to_config_packed(PIN_CONFIG_PERSIST_STATE,
> !transitory);
> gpio = gpio_chip_hwgpio(desc);
> - rc = chip->set_config(chip, gpio, packed);
> + rc = gpio_set_config(chip, gpio, packed);
> if (rc == -ENOTSUPP) {
> dev_dbg(&desc->gdev->dev, "Persistence not supported for GPIO %d\n",
> gpio);
Revert commit 6581eaf0e890 ("gpio: use new gpio_set_config() helper in
more places") to restore correct behaviour for gpiod_set_debounce() and
gpiod_set_transitory().
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2019-03-26 07:49:54 +03:00
rc = chip - > set_config ( chip , gpio , packed ) ;
2017-11-30 06:55:24 +03:00
if ( rc = = - ENOTSUPP ) {
dev_dbg ( & desc - > gdev - > dev , " Persistence not supported for GPIO %d \n " ,
gpio ) ;
return 0 ;
}
return rc ;
}
EXPORT_SYMBOL_GPL ( gpiod_set_transitory ) ;
2013-10-17 21:21:36 +04:00
/**
* gpiod_is_active_low - test whether a GPIO is active - low or not
* @ desc : the gpio descriptor to test
*
* Returns 1 if the GPIO is active - low , 0 otherwise .
*/
int gpiod_is_active_low ( const struct gpio_desc * desc )
2013-02-02 20:29:29 +04:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
2013-10-17 21:21:36 +04:00
return test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_is_active_low ) ;
2008-02-05 09:28:20 +03:00
/* I/O calls are only valid after configuration completed; the relevant
* " is this a valid GPIO " error checks should already have been done .
*
* " Get " operations are often inlinable as reading a pin value register ,
* and masking the relevant bit in that register .
*
* When " set " operations are inlinable , they involve writing that mask to
* one register to set a low value , or a different register to set it high .
* Otherwise locking is needed , so there may be little value to inlining .
*
* - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
*
* IMPORTANT ! ! ! The hot paths - - get / set value - - assume that callers
* have requested the GPIO . That can include implicit requesting by
* a direction setting call . Marking a gpio as requested locks its chip
* in memory , guaranteeing that these table lookups need no more locking
* and that gpiochip_remove ( ) will fail .
*
* REVISIT when debugging , consider adding some instrumentation to ensure
* that the GPIO was actually requested .
*/
2017-09-26 21:58:28 +03:00
static int gpiod_get_raw_value_commit ( const struct gpio_desc * desc )
2008-02-05 09:28:20 +03:00
{
struct gpio_chip * chip ;
2013-02-02 20:29:29 +04:00
int offset ;
2015-08-28 19:44:18 +03:00
int value ;
2008-02-05 09:28:20 +03:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
chip = desc - > gdev - > chip ;
2013-02-02 20:29:29 +04:00
offset = gpio_chip_hwgpio ( desc ) ;
2015-08-28 19:44:18 +03:00
value = chip - > get ? chip - > get ( chip , offset ) : - EIO ;
2015-12-22 01:10:12 +03:00
value = value < 0 ? value : ! ! value ;
2013-02-02 20:29:29 +04:00
trace_gpio_value ( desc_to_gpio ( desc ) , 1 , value ) ;
2011-05-20 10:40:19 +04:00
return value ;
2008-02-05 09:28:20 +03:00
}
2013-02-02 20:29:29 +04:00
2017-10-12 13:40:10 +03:00
static int gpio_chip_get_multiple ( struct gpio_chip * chip ,
unsigned long * mask , unsigned long * bits )
{
if ( chip - > get_multiple ) {
return chip - > get_multiple ( chip , mask , bits ) ;
} else if ( chip - > get ) {
int i , value ;
for_each_set_bit ( i , mask , chip - > ngpio ) {
value = chip - > get ( chip , i ) ;
if ( value < 0 )
return value ;
__assign_bit ( i , bits , value ) ;
}
return 0 ;
}
return - EIO ;
}
int gpiod_get_array_value_complex ( bool raw , bool can_sleep ,
unsigned int array_size ,
struct gpio_desc * * desc_array ,
2018-09-06 00:50:07 +03:00
struct gpio_array * array_info ,
2018-09-06 00:50:05 +03:00
unsigned long * value_bitmap )
2017-10-12 13:40:10 +03:00
{
2019-07-16 12:11:45 +03:00
int ret , i = 0 ;
2018-09-06 00:50:08 +03:00
/*
* Validate array_info against desc_array and its size .
* It should immediately follow desc_array if both
* have been obtained from the same gpiod_get_array ( ) call .
*/
if ( array_info & & array_info - > desc = = desc_array & &
array_size < = array_info - > size & &
( void * ) array_info = = desc_array + array_info - > size ) {
if ( ! can_sleep )
WARN_ON ( array_info - > chip - > can_sleep ) ;
2019-07-16 12:11:45 +03:00
ret = gpio_chip_get_multiple ( array_info - > chip ,
2018-09-06 00:50:08 +03:00
array_info - > get_mask ,
value_bitmap ) ;
2019-07-16 12:11:45 +03:00
if ( ret )
return ret ;
2018-09-06 00:50:08 +03:00
if ( ! raw & & ! bitmap_empty ( array_info - > invert_mask , array_size ) )
bitmap_xor ( value_bitmap , value_bitmap ,
array_info - > invert_mask , array_size ) ;
if ( bitmap_full ( array_info - > get_mask , array_size ) )
return 0 ;
i = find_first_zero_bit ( array_info - > get_mask , array_size ) ;
} else {
array_info = NULL ;
}
2017-10-12 13:40:10 +03:00
while ( i < array_size ) {
struct gpio_chip * chip = desc_array [ i ] - > gdev - > chip ;
2018-05-21 20:57:07 +03:00
unsigned long fastpath [ 2 * BITS_TO_LONGS ( FASTPATH_NGPIO ) ] ;
unsigned long * mask , * bits ;
2017-10-12 13:40:10 +03:00
int first , j , ret ;
2018-05-21 20:57:07 +03:00
if ( likely ( chip - > ngpio < = FASTPATH_NGPIO ) ) {
mask = fastpath ;
} else {
mask = kmalloc_array ( 2 * BITS_TO_LONGS ( chip - > ngpio ) ,
sizeof ( * mask ) ,
can_sleep ? GFP_KERNEL : GFP_ATOMIC ) ;
if ( ! mask )
return - ENOMEM ;
}
bits = mask + BITS_TO_LONGS ( chip - > ngpio ) ;
bitmap_zero ( mask , chip - > ngpio ) ;
2017-10-12 13:40:10 +03:00
if ( ! can_sleep )
WARN_ON ( chip - > can_sleep ) ;
/* collect all inputs belonging to the same chip */
first = i ;
do {
const struct gpio_desc * desc = desc_array [ i ] ;
int hwgpio = gpio_chip_hwgpio ( desc ) ;
__set_bit ( hwgpio , mask ) ;
i + + ;
2018-09-06 00:50:08 +03:00
if ( array_info )
2018-09-24 02:53:35 +03:00
i = find_next_zero_bit ( array_info - > get_mask ,
array_size , i ) ;
2017-10-12 13:40:10 +03:00
} while ( ( i < array_size ) & &
( desc_array [ i ] - > gdev - > chip = = chip ) ) ;
ret = gpio_chip_get_multiple ( chip , mask , bits ) ;
2018-05-21 20:57:07 +03:00
if ( ret ) {
if ( mask ! = fastpath )
kfree ( mask ) ;
2017-10-12 13:40:10 +03:00
return ret ;
2018-05-21 20:57:07 +03:00
}
2017-10-12 13:40:10 +03:00
2018-09-06 00:50:08 +03:00
for ( j = first ; j < i ; ) {
2017-10-12 13:40:10 +03:00
const struct gpio_desc * desc = desc_array [ j ] ;
int hwgpio = gpio_chip_hwgpio ( desc ) ;
int value = test_bit ( hwgpio , bits ) ;
if ( ! raw & & test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) )
value = ! value ;
2018-09-06 00:50:05 +03:00
__assign_bit ( j , value_bitmap , value ) ;
2017-10-12 13:40:10 +03:00
trace_gpio_value ( desc_to_gpio ( desc ) , 1 , value ) ;
2018-09-29 15:20:22 +03:00
j + + ;
2018-09-06 00:50:08 +03:00
if ( array_info )
2018-09-24 02:53:35 +03:00
j = find_next_zero_bit ( array_info - > get_mask , i ,
j ) ;
2017-10-12 13:40:10 +03:00
}
2018-05-21 20:57:07 +03:00
if ( mask ! = fastpath )
kfree ( mask ) ;
2017-10-12 13:40:10 +03:00
}
return 0 ;
}
2008-02-05 09:28:20 +03:00
/**
2013-10-17 21:21:36 +04:00
* gpiod_get_raw_value ( ) - return a gpio ' s raw value
* @ desc : gpio whose value will be returned
2008-02-05 09:28:20 +03:00
*
2013-10-17 21:21:36 +04:00
* Return the GPIO ' s raw value , i . e . the value of the physical line disregarding
2015-08-28 19:44:18 +03:00
* its ACTIVE_LOW status , or negative errno on failure .
2013-10-17 21:21:36 +04:00
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep , and will
2013-10-17 21:21:36 +04:00
* complain if the GPIO chip functions potentially sleep .
2008-02-05 09:28:20 +03:00
*/
2013-10-17 21:21:36 +04:00
int gpiod_get_raw_value ( const struct gpio_desc * desc )
2008-02-05 09:28:20 +03:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
2019-07-01 17:27:38 +03:00
/* Should be using gpiod_get_raw_value_cansleep() */
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
WARN_ON ( desc - > gdev - > chip - > can_sleep ) ;
2017-09-26 21:58:28 +03:00
return gpiod_get_raw_value_commit ( desc ) ;
2008-02-05 09:28:20 +03:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_get_raw_value ) ;
2013-02-02 20:29:29 +04:00
2013-10-17 21:21:36 +04:00
/**
* gpiod_get_value ( ) - return a gpio ' s value
* @ desc : gpio whose value will be returned
*
* Return the GPIO ' s logical value , i . e . taking the ACTIVE_LOW status into
2015-08-28 19:44:18 +03:00
* account , or negative errno on failure .
2013-10-17 21:21:36 +04:00
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep , and will
2013-10-17 21:21:36 +04:00
* complain if the GPIO chip functions potentially sleep .
*/
int gpiod_get_value ( const struct gpio_desc * desc )
2013-02-02 20:29:29 +04:00
{
2013-10-17 21:21:36 +04:00
int value ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
2019-07-01 17:27:38 +03:00
/* Should be using gpiod_get_value_cansleep() */
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
WARN_ON ( desc - > gdev - > chip - > can_sleep ) ;
2013-10-17 21:21:36 +04:00
2017-09-26 21:58:28 +03:00
value = gpiod_get_raw_value_commit ( desc ) ;
2015-08-28 19:44:18 +03:00
if ( value < 0 )
return value ;
2013-10-17 21:21:36 +04:00
if ( test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) )
value = ! value ;
return value ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_get_value ) ;
2008-02-05 09:28:20 +03:00
2017-10-12 13:40:10 +03:00
/**
* gpiod_get_raw_array_value ( ) - read raw values from an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2017-10-12 13:40:10 +03:00
* @ desc_array : array of GPIO descriptors whose values will be read
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap to store the read values
2017-10-12 13:40:10 +03:00
*
* Read the raw values of the GPIOs , i . e . the values of the physical lines
* without regard for their ACTIVE_LOW status . Return 0 in case of success ,
* else an error code .
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep ,
2017-10-12 13:40:10 +03:00
* and it will complain if the GPIO chip functions potentially sleep .
*/
int gpiod_get_raw_array_value ( unsigned int array_size ,
2018-09-06 00:50:05 +03:00
struct gpio_desc * * desc_array ,
2018-09-06 00:50:07 +03:00
struct gpio_array * array_info ,
2018-09-06 00:50:05 +03:00
unsigned long * value_bitmap )
2017-10-12 13:40:10 +03:00
{
if ( ! desc_array )
return - EINVAL ;
return gpiod_get_array_value_complex ( true , false , array_size ,
2018-09-06 00:50:07 +03:00
desc_array , array_info ,
value_bitmap ) ;
2017-10-12 13:40:10 +03:00
}
EXPORT_SYMBOL_GPL ( gpiod_get_raw_array_value ) ;
/**
* gpiod_get_array_value ( ) - read values from an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2017-10-12 13:40:10 +03:00
* @ desc_array : array of GPIO descriptors whose values will be read
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap to store the read values
2017-10-12 13:40:10 +03:00
*
* Read the logical values of the GPIOs , i . e . taking their ACTIVE_LOW status
* into account . Return 0 in case of success , else an error code .
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep ,
2017-10-12 13:40:10 +03:00
* and it will complain if the GPIO chip functions potentially sleep .
*/
int gpiod_get_array_value ( unsigned int array_size ,
2018-09-06 00:50:05 +03:00
struct gpio_desc * * desc_array ,
2018-09-06 00:50:07 +03:00
struct gpio_array * array_info ,
2018-09-06 00:50:05 +03:00
unsigned long * value_bitmap )
2017-10-12 13:40:10 +03:00
{
if ( ! desc_array )
return - EINVAL ;
return gpiod_get_array_value_complex ( false , false , array_size ,
2018-09-06 00:50:07 +03:00
desc_array , array_info ,
value_bitmap ) ;
2017-10-12 13:40:10 +03:00
}
EXPORT_SYMBOL_GPL ( gpiod_get_array_value ) ;
2012-02-17 18:56:21 +04:00
/*
2017-09-26 21:58:28 +03:00
* gpio_set_open_drain_value_commit ( ) - Set the open drain gpio ' s value .
2013-10-17 21:21:36 +04:00
* @ desc : gpio descriptor whose state need to be set .
2015-05-18 21:41:43 +03:00
* @ value : Non - zero for setting it HIGH otherwise it will set to LOW .
2012-02-17 18:56:21 +04:00
*/
2017-09-26 21:58:28 +03:00
static void gpio_set_open_drain_value_commit ( struct gpio_desc * desc , bool value )
2012-02-17 18:56:21 +04:00
{
2019-07-16 12:11:45 +03:00
int ret = 0 ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_chip * chip = desc - > gdev - > chip ;
2013-02-02 20:29:29 +04:00
int offset = gpio_chip_hwgpio ( desc ) ;
2012-02-17 18:56:21 +04:00
if ( value ) {
2019-07-16 12:11:45 +03:00
ret = chip - > direction_input ( chip , offset ) ;
2012-02-17 18:56:21 +04:00
} else {
2019-07-16 12:11:45 +03:00
ret = chip - > direction_output ( chip , offset , 0 ) ;
if ( ! ret )
2013-02-02 20:29:29 +04:00
set_bit ( FLAG_IS_OUT , & desc - > flags ) ;
2012-02-17 18:56:21 +04:00
}
2019-07-16 12:11:45 +03:00
trace_gpio_direction ( desc_to_gpio ( desc ) , value , ret ) ;
if ( ret < 0 )
2013-09-09 13:33:49 +04:00
gpiod_err ( desc ,
" %s: Error in set_value for open drain err %d \n " ,
2019-07-16 12:11:45 +03:00
__func__ , ret ) ;
2012-02-17 18:56:21 +04:00
}
2012-02-17 18:56:22 +04:00
/*
2013-10-17 21:21:36 +04:00
* _gpio_set_open_source_value ( ) - Set the open source gpio ' s value .
* @ desc : gpio descriptor whose state need to be set .
2015-05-18 21:41:43 +03:00
* @ value : Non - zero for setting it HIGH otherwise it will set to LOW .
2012-02-17 18:56:22 +04:00
*/
2017-09-26 21:58:28 +03:00
static void gpio_set_open_source_value_commit ( struct gpio_desc * desc , bool value )
2012-02-17 18:56:22 +04:00
{
2019-07-16 12:11:45 +03:00
int ret = 0 ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_chip * chip = desc - > gdev - > chip ;
2013-02-02 20:29:29 +04:00
int offset = gpio_chip_hwgpio ( desc ) ;
2012-02-17 18:56:22 +04:00
if ( value ) {
2019-07-16 12:11:45 +03:00
ret = chip - > direction_output ( chip , offset , 1 ) ;
if ( ! ret )
2013-02-02 20:29:29 +04:00
set_bit ( FLAG_IS_OUT , & desc - > flags ) ;
2012-02-17 18:56:22 +04:00
} else {
2019-07-16 12:11:45 +03:00
ret = chip - > direction_input ( chip , offset ) ;
2012-02-17 18:56:22 +04:00
}
2019-07-16 12:11:45 +03:00
trace_gpio_direction ( desc_to_gpio ( desc ) , ! value , ret ) ;
if ( ret < 0 )
2013-09-09 13:33:49 +04:00
gpiod_err ( desc ,
" %s: Error in set_value for open source err %d \n " ,
2019-07-16 12:11:45 +03:00
__func__ , ret ) ;
2012-02-17 18:56:22 +04:00
}
2017-09-26 21:58:28 +03:00
static void gpiod_set_raw_value_commit ( struct gpio_desc * desc , bool value )
2008-02-05 09:28:20 +03:00
{
struct gpio_chip * chip ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
chip = desc - > gdev - > chip ;
2013-02-02 20:29:29 +04:00
trace_gpio_value ( desc_to_gpio ( desc ) , 0 , value ) ;
2017-09-26 22:20:23 +03:00
chip - > set ( chip , gpio_chip_hwgpio ( desc ) , value ) ;
2013-02-02 20:29:29 +04:00
}
2014-11-04 19:12:06 +03:00
/*
* set multiple outputs on the same chip ;
* use the chip ' s set_multiple function if available ;
* otherwise set the outputs sequentially ;
* @ mask : bit mask array ; one bit per output ; BITS_PER_LONG bits per word
* defines which outputs are to be changed
* @ bits : bit value array ; one bit per output ; BITS_PER_LONG bits per word
* defines the values the outputs specified by mask are to be set to
*/
static void gpio_chip_set_multiple ( struct gpio_chip * chip ,
unsigned long * mask , unsigned long * bits )
{
if ( chip - > set_multiple ) {
chip - > set_multiple ( chip , mask , bits ) ;
} else {
2017-01-03 20:01:17 +03:00
unsigned int i ;
/* set outputs if the corresponding mask bit is set */
for_each_set_bit ( i , mask , chip - > ngpio )
chip - > set ( chip , i , test_bit ( i , bits ) ) ;
2014-11-04 19:12:06 +03:00
}
}
2018-05-21 20:57:07 +03:00
int gpiod_set_array_value_complex ( bool raw , bool can_sleep ,
2018-09-27 14:38:10 +03:00
unsigned int array_size ,
struct gpio_desc * * desc_array ,
struct gpio_array * array_info ,
unsigned long * value_bitmap )
2014-11-04 19:12:06 +03:00
{
int i = 0 ;
2018-09-06 00:50:08 +03:00
/*
* Validate array_info against desc_array and its size .
* It should immediately follow desc_array if both
* have been obtained from the same gpiod_get_array ( ) call .
*/
if ( array_info & & array_info - > desc = = desc_array & &
array_size < = array_info - > size & &
( void * ) array_info = = desc_array + array_info - > size ) {
if ( ! can_sleep )
WARN_ON ( array_info - > chip - > can_sleep ) ;
if ( ! raw & & ! bitmap_empty ( array_info - > invert_mask , array_size ) )
bitmap_xor ( value_bitmap , value_bitmap ,
array_info - > invert_mask , array_size ) ;
gpio_chip_set_multiple ( array_info - > chip , array_info - > set_mask ,
value_bitmap ) ;
if ( bitmap_full ( array_info - > set_mask , array_size ) )
return 0 ;
i = find_first_zero_bit ( array_info - > set_mask , array_size ) ;
} else {
array_info = NULL ;
}
2014-11-04 19:12:06 +03:00
while ( i < array_size ) {
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_chip * chip = desc_array [ i ] - > gdev - > chip ;
2018-05-21 20:57:07 +03:00
unsigned long fastpath [ 2 * BITS_TO_LONGS ( FASTPATH_NGPIO ) ] ;
unsigned long * mask , * bits ;
2014-11-04 19:12:06 +03:00
int count = 0 ;
2018-05-21 20:57:07 +03:00
if ( likely ( chip - > ngpio < = FASTPATH_NGPIO ) ) {
mask = fastpath ;
} else {
mask = kmalloc_array ( 2 * BITS_TO_LONGS ( chip - > ngpio ) ,
sizeof ( * mask ) ,
can_sleep ? GFP_KERNEL : GFP_ATOMIC ) ;
if ( ! mask )
return - ENOMEM ;
}
bits = mask + BITS_TO_LONGS ( chip - > ngpio ) ;
bitmap_zero ( mask , chip - > ngpio ) ;
2015-06-10 16:26:27 +03:00
if ( ! can_sleep )
2014-11-04 19:12:06 +03:00
WARN_ON ( chip - > can_sleep ) ;
2015-06-10 16:26:27 +03:00
2014-11-04 19:12:06 +03:00
do {
struct gpio_desc * desc = desc_array [ i ] ;
int hwgpio = gpio_chip_hwgpio ( desc ) ;
2018-09-06 00:50:05 +03:00
int value = test_bit ( i , value_bitmap ) ;
2014-11-04 19:12:06 +03:00
2018-09-06 00:50:08 +03:00
/*
* Pins applicable for fast input but not for
* fast output processing may have been already
* inverted inside the fast path , skip them .
*/
if ( ! raw & & ! ( array_info & &
test_bit ( i , array_info - > invert_mask ) ) & &
test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) )
2014-11-04 19:12:06 +03:00
value = ! value ;
trace_gpio_value ( desc_to_gpio ( desc ) , 0 , value ) ;
/*
* collect all normal outputs belonging to the same chip
* open drain and open source outputs are set individually
*/
2017-09-26 22:20:23 +03:00
if ( test_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) & & ! raw ) {
2017-09-26 21:58:28 +03:00
gpio_set_open_drain_value_commit ( desc , value ) ;
2017-09-26 22:20:23 +03:00
} else if ( test_bit ( FLAG_OPEN_SOURCE , & desc - > flags ) & & ! raw ) {
2017-09-26 21:58:28 +03:00
gpio_set_open_source_value_commit ( desc , value ) ;
2014-11-04 19:12:06 +03:00
} else {
__set_bit ( hwgpio , mask ) ;
2015-06-10 16:26:27 +03:00
if ( value )
2014-11-04 19:12:06 +03:00
__set_bit ( hwgpio , bits ) ;
2015-06-10 16:26:27 +03:00
else
2014-11-04 19:12:06 +03:00
__clear_bit ( hwgpio , bits ) ;
count + + ;
}
i + + ;
2018-09-06 00:50:08 +03:00
if ( array_info )
2018-09-24 02:53:35 +03:00
i = find_next_zero_bit ( array_info - > set_mask ,
array_size , i ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
} while ( ( i < array_size ) & &
( desc_array [ i ] - > gdev - > chip = = chip ) ) ;
2014-11-04 19:12:06 +03:00
/* push collected bits to outputs */
2015-06-10 16:26:27 +03:00
if ( count ! = 0 )
2014-11-04 19:12:06 +03:00
gpio_chip_set_multiple ( chip , mask , bits ) ;
2018-05-21 20:57:07 +03:00
if ( mask ! = fastpath )
kfree ( mask ) ;
2014-11-04 19:12:06 +03:00
}
2018-05-21 20:57:07 +03:00
return 0 ;
2014-11-04 19:12:06 +03:00
}
2008-02-05 09:28:20 +03:00
/**
2013-10-17 21:21:36 +04:00
* gpiod_set_raw_value ( ) - assign a gpio ' s raw value
* @ desc : gpio whose value will be assigned
2008-02-05 09:28:20 +03:00
* @ value : value to assign
*
2013-10-17 21:21:36 +04:00
* Set the raw value of the GPIO , i . e . the value of its physical line without
* regard for its ACTIVE_LOW status .
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep , and will
2013-10-17 21:21:36 +04:00
* complain if the GPIO chip functions potentially sleep .
2008-02-05 09:28:20 +03:00
*/
2013-10-17 21:21:36 +04:00
void gpiod_set_raw_value ( struct gpio_desc * desc , int value )
2013-02-02 20:29:29 +04:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC_VOID ( desc ) ;
2019-07-01 17:27:38 +03:00
/* Should be using gpiod_set_raw_value_cansleep() */
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
WARN_ON ( desc - > gdev - > chip - > can_sleep ) ;
2017-09-26 21:58:28 +03:00
gpiod_set_raw_value_commit ( desc , value ) ;
2008-02-05 09:28:20 +03:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_set_raw_value ) ;
2008-02-05 09:28:20 +03:00
2018-01-09 21:08:21 +03:00
/**
* gpiod_set_value_nocheck ( ) - set a GPIO line value without checking
* @ desc : the descriptor to set the value on
* @ value : value to set
*
* This sets the value of a GPIO line backing a descriptor , applying
* different semantic quirks like active low and open drain / source
* handling .
*/
static void gpiod_set_value_nocheck ( struct gpio_desc * desc , int value )
{
if ( test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) )
value = ! value ;
if ( test_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) )
gpio_set_open_drain_value_commit ( desc , value ) ;
else if ( test_bit ( FLAG_OPEN_SOURCE , & desc - > flags ) )
gpio_set_open_source_value_commit ( desc , value ) ;
else
gpiod_set_raw_value_commit ( desc , value ) ;
}
2008-02-05 09:28:20 +03:00
/**
2013-10-17 21:21:36 +04:00
* gpiod_set_value ( ) - assign a gpio ' s value
* @ desc : gpio whose value will be assigned
* @ value : value to assign
*
2017-09-26 22:20:23 +03:00
* Set the logical value of the GPIO , i . e . taking its ACTIVE_LOW ,
* OPEN_DRAIN and OPEN_SOURCE flags into account .
2008-02-05 09:28:20 +03:00
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep , and will
2013-10-17 21:21:36 +04:00
* complain if the GPIO chip functions potentially sleep .
2008-02-05 09:28:20 +03:00
*/
2013-10-17 21:21:36 +04:00
void gpiod_set_value ( struct gpio_desc * desc , int value )
2008-02-05 09:28:20 +03:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC_VOID ( desc ) ;
2019-07-01 17:27:38 +03:00
/* Should be using gpiod_set_value_cansleep() */
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
WARN_ON ( desc - > gdev - > chip - > can_sleep ) ;
2018-01-09 21:08:21 +03:00
gpiod_set_value_nocheck ( desc , value ) ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_set_value ) ;
2008-02-05 09:28:20 +03:00
2014-11-04 19:12:06 +03:00
/**
2015-05-13 12:04:56 +03:00
* gpiod_set_raw_array_value ( ) - assign values to an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2014-11-04 19:12:06 +03:00
* @ desc_array : array of GPIO descriptors whose values will be assigned
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap of values to assign
2014-11-04 19:12:06 +03:00
*
* Set the raw values of the GPIOs , i . e . the values of the physical lines
* without regard for their ACTIVE_LOW status .
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep , and will
2014-11-04 19:12:06 +03:00
* complain if the GPIO chip functions potentially sleep .
*/
2018-05-21 20:57:07 +03:00
int gpiod_set_raw_array_value ( unsigned int array_size ,
2018-09-27 14:38:10 +03:00
struct gpio_desc * * desc_array ,
struct gpio_array * array_info ,
unsigned long * value_bitmap )
2014-11-04 19:12:06 +03:00
{
if ( ! desc_array )
2018-05-21 20:57:07 +03:00
return - EINVAL ;
return gpiod_set_array_value_complex ( true , false , array_size ,
2018-09-06 00:50:07 +03:00
desc_array , array_info , value_bitmap ) ;
2014-11-04 19:12:06 +03:00
}
2015-05-13 12:04:56 +03:00
EXPORT_SYMBOL_GPL ( gpiod_set_raw_array_value ) ;
2014-11-04 19:12:06 +03:00
/**
2015-05-13 12:04:56 +03:00
* gpiod_set_array_value ( ) - assign values to an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2014-11-04 19:12:06 +03:00
* @ desc_array : array of GPIO descriptors whose values will be assigned
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap of values to assign
2014-11-04 19:12:06 +03:00
*
* Set the logical values of the GPIOs , i . e . taking their ACTIVE_LOW status
* into account .
*
2019-07-01 17:28:09 +03:00
* This function can be called from contexts where we cannot sleep , and will
2014-11-04 19:12:06 +03:00
* complain if the GPIO chip functions potentially sleep .
*/
2018-09-27 14:38:09 +03:00
int gpiod_set_array_value ( unsigned int array_size ,
struct gpio_desc * * desc_array ,
struct gpio_array * array_info ,
unsigned long * value_bitmap )
2014-11-04 19:12:06 +03:00
{
if ( ! desc_array )
2018-09-27 14:38:09 +03:00
return - EINVAL ;
return gpiod_set_array_value_complex ( false , false , array_size ,
desc_array , array_info ,
value_bitmap ) ;
2014-11-04 19:12:06 +03:00
}
2015-05-13 12:04:56 +03:00
EXPORT_SYMBOL_GPL ( gpiod_set_array_value ) ;
2014-11-04 19:12:06 +03:00
2008-02-05 09:28:20 +03:00
/**
2013-10-17 21:21:36 +04:00
* gpiod_cansleep ( ) - report whether gpio value access may sleep
* @ desc : gpio to check
2008-02-05 09:28:20 +03:00
*
*/
2013-10-17 21:21:36 +04:00
int gpiod_cansleep ( const struct gpio_desc * desc )
2013-02-02 20:29:29 +04:00
{
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
return desc - > gdev - > chip - > can_sleep ;
2008-02-05 09:28:20 +03:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_cansleep ) ;
2008-02-05 09:28:20 +03:00
2018-06-01 14:21:27 +03:00
/**
* gpiod_set_consumer_name ( ) - set the consumer name for the descriptor
* @ desc : gpio to set the consumer name on
* @ name : the new consumer name
*/
2018-11-01 16:12:50 +03:00
int gpiod_set_consumer_name ( struct gpio_desc * desc , const char * name )
2018-06-01 14:21:27 +03:00
{
2018-11-01 16:12:50 +03:00
VALIDATE_DESC ( desc ) ;
if ( name ) {
name = kstrdup_const ( name , GFP_KERNEL ) ;
if ( ! name )
return - ENOMEM ;
}
kfree_const ( desc - > label ) ;
desc_set_label ( desc , name ) ;
return 0 ;
2018-06-01 14:21:27 +03:00
}
EXPORT_SYMBOL_GPL ( gpiod_set_consumer_name ) ;
2008-10-16 09:03:14 +04:00
/**
2013-10-17 21:21:36 +04:00
* gpiod_to_irq ( ) - return the IRQ corresponding to a GPIO
* @ desc : gpio whose IRQ will be returned ( already requested )
2008-10-16 09:03:14 +04:00
*
2013-10-17 21:21:36 +04:00
* Return the IRQ corresponding to the passed GPIO , or an error code in case of
* error .
2008-10-16 09:03:14 +04:00
*/
2013-10-17 21:21:36 +04:00
int gpiod_to_irq ( const struct gpio_desc * desc )
2008-10-16 09:03:14 +04:00
{
2016-05-02 14:13:10 +03:00
struct gpio_chip * chip ;
int offset ;
2008-10-16 09:03:14 +04:00
2016-06-15 23:57:38 +03:00
/*
* Cannot VALIDATE_DESC ( ) here as gpiod_to_irq ( ) consumer semantics
* requires this function to not return zero on an invalid descriptor
* but rather a negative error number .
*/
2016-06-16 12:55:55 +03:00
if ( ! desc | | IS_ERR ( desc ) | | ! desc - > gdev | | ! desc - > gdev - > chip )
2016-06-15 23:57:38 +03:00
return - EINVAL ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
chip = desc - > gdev - > chip ;
2013-02-02 20:29:29 +04:00
offset = gpio_chip_hwgpio ( desc ) ;
2016-05-02 14:13:10 +03:00
if ( chip - > to_irq ) {
int retirq = chip - > to_irq ( chip , offset ) ;
/* Zero means NO_IRQ */
if ( ! retirq )
return - ENXIO ;
return retirq ;
}
return - ENXIO ;
2008-10-16 09:03:14 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_to_irq ) ;
2008-10-16 09:03:14 +04:00
2013-09-24 13:54:38 +04:00
/**
2014-10-23 12:27:07 +04:00
* gpiochip_lock_as_irq ( ) - lock a GPIO to be used as IRQ
2014-07-22 11:17:42 +04:00
* @ chip : the chip the GPIO to lock belongs to
* @ offset : the offset of the GPIO to lock as IRQ
2013-09-24 13:54:38 +04:00
*
* This is used directly by GPIO drivers that want to lock down
2014-03-07 06:12:49 +04:00
* a certain GPIO line to be used for IRQs .
2013-09-24 13:54:38 +04:00
*/
2014-10-23 12:27:07 +04:00
int gpiochip_lock_as_irq ( struct gpio_chip * chip , unsigned int offset )
2013-02-02 20:29:29 +04:00
{
2016-05-25 11:56:03 +03:00
struct gpio_desc * desc ;
desc = gpiochip_get_desc ( chip , offset ) ;
if ( IS_ERR ( desc ) )
return PTR_ERR ( desc ) ;
2016-11-12 17:01:09 +03:00
/*
* If it ' s fast : flush the direction setting if something changed
* behind our back
*/
if ( ! chip - > can_sleep & & chip - > get_direction ) {
2018-07-09 21:47:21 +03:00
int dir = gpiod_get_direction ( desc ) ;
2016-05-25 11:56:03 +03:00
2018-07-03 03:38:31 +03:00
if ( dir < 0 ) {
chip_err ( chip , " %s: cannot get GPIO direction \n " ,
__func__ ) ;
return dir ;
}
2016-05-25 11:56:03 +03:00
}
2013-09-24 13:54:38 +04:00
2016-05-25 11:56:03 +03:00
if ( test_bit ( FLAG_IS_OUT , & desc - > flags ) ) {
2014-07-22 11:17:42 +04:00
chip_err ( chip ,
2018-07-03 03:39:03 +03:00
" %s: tried to flag a GPIO set as output for IRQ \n " ,
__func__ ) ;
2013-09-24 13:54:38 +04:00
return - EIO ;
}
2016-05-25 11:56:03 +03:00
set_bit ( FLAG_USED_AS_IRQ , & desc - > flags ) ;
2018-09-08 12:23:16 +03:00
set_bit ( FLAG_IRQ_IS_ENABLED , & desc - > flags ) ;
2016-11-14 02:09:07 +03:00
/*
* If the consumer has not set up a label ( such as when the
* IRQ is referenced from . to_irq ( ) ) we set up a label here
* so it is clear this is used as an interrupt .
*/
if ( ! desc - > label )
desc_set_label ( desc , " interrupt " ) ;
2013-09-24 13:54:38 +04:00
return 0 ;
2013-02-02 20:29:29 +04:00
}
2014-10-23 12:27:07 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_lock_as_irq ) ;
2008-02-05 09:28:20 +03:00
2013-09-24 13:54:38 +04:00
/**
2014-10-23 12:27:07 +04:00
* gpiochip_unlock_as_irq ( ) - unlock a GPIO used as IRQ
2014-07-22 11:17:42 +04:00
* @ chip : the chip the GPIO to lock belongs to
* @ offset : the offset of the GPIO to lock as IRQ
2013-09-24 13:54:38 +04:00
*
* This is used directly by GPIO drivers that want to indicate
* that a certain GPIO is no longer used exclusively for IRQ .
2008-02-05 09:28:20 +03:00
*/
2014-10-23 12:27:07 +04:00
void gpiochip_unlock_as_irq ( struct gpio_chip * chip , unsigned int offset )
2013-09-24 13:54:38 +04:00
{
2016-11-14 02:09:07 +03:00
struct gpio_desc * desc ;
desc = gpiochip_get_desc ( chip , offset ) ;
if ( IS_ERR ( desc ) )
2013-09-24 13:54:38 +04:00
return ;
2008-02-05 09:28:20 +03:00
2016-11-14 02:09:07 +03:00
clear_bit ( FLAG_USED_AS_IRQ , & desc - > flags ) ;
2018-09-08 12:23:16 +03:00
clear_bit ( FLAG_IRQ_IS_ENABLED , & desc - > flags ) ;
2016-11-14 02:09:07 +03:00
/* If we only had this marking, erase it */
if ( desc - > label & & ! strcmp ( desc - > label , " interrupt " ) )
desc_set_label ( desc , NULL ) ;
2013-09-24 13:54:38 +04:00
}
2014-10-23 12:27:07 +04:00
EXPORT_SYMBOL_GPL ( gpiochip_unlock_as_irq ) ;
2013-09-24 13:54:38 +04:00
2018-09-08 12:23:16 +03:00
void gpiochip_disable_irq ( struct gpio_chip * chip , unsigned int offset )
{
struct gpio_desc * desc = gpiochip_get_desc ( chip , offset ) ;
if ( ! IS_ERR ( desc ) & &
! WARN_ON ( ! test_bit ( FLAG_USED_AS_IRQ , & desc - > flags ) ) )
clear_bit ( FLAG_IRQ_IS_ENABLED , & desc - > flags ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_disable_irq ) ;
void gpiochip_enable_irq ( struct gpio_chip * chip , unsigned int offset )
{
struct gpio_desc * desc = gpiochip_get_desc ( chip , offset ) ;
if ( ! IS_ERR ( desc ) & &
! WARN_ON ( ! test_bit ( FLAG_USED_AS_IRQ , & desc - > flags ) ) ) {
WARN_ON ( test_bit ( FLAG_IS_OUT , & desc - > flags ) ) ;
set_bit ( FLAG_IRQ_IS_ENABLED , & desc - > flags ) ;
}
}
EXPORT_SYMBOL_GPL ( gpiochip_enable_irq ) ;
2016-02-11 22:16:45 +03:00
bool gpiochip_line_is_irq ( struct gpio_chip * chip , unsigned int offset )
{
if ( offset > = chip - > ngpio )
return false ;
return test_bit ( FLAG_USED_AS_IRQ , & chip - > gpiodev - > descs [ offset ] . flags ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_line_is_irq ) ;
2018-09-08 12:23:14 +03:00
int gpiochip_reqres_irq ( struct gpio_chip * chip , unsigned int offset )
{
int ret ;
if ( ! try_module_get ( chip - > gpiodev - > owner ) )
return - ENODEV ;
ret = gpiochip_lock_as_irq ( chip , offset ) ;
if ( ret ) {
chip_err ( chip , " unable to lock HW IRQ %u for IRQ \n " , offset ) ;
module_put ( chip - > gpiodev - > owner ) ;
return ret ;
}
return 0 ;
}
EXPORT_SYMBOL_GPL ( gpiochip_reqres_irq ) ;
void gpiochip_relres_irq ( struct gpio_chip * chip , unsigned int offset )
{
gpiochip_unlock_as_irq ( chip , offset ) ;
module_put ( chip - > gpiodev - > owner ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_relres_irq ) ;
2016-02-16 17:41:42 +03:00
bool gpiochip_line_is_open_drain ( struct gpio_chip * chip , unsigned int offset )
{
if ( offset > = chip - > ngpio )
return false ;
return test_bit ( FLAG_OPEN_DRAIN , & chip - > gpiodev - > descs [ offset ] . flags ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_line_is_open_drain ) ;
bool gpiochip_line_is_open_source ( struct gpio_chip * chip , unsigned int offset )
{
if ( offset > = chip - > ngpio )
return false ;
return test_bit ( FLAG_OPEN_SOURCE , & chip - > gpiodev - > descs [ offset ] . flags ) ;
}
EXPORT_SYMBOL_GPL ( gpiochip_line_is_open_source ) ;
2017-05-23 17:47:29 +03:00
bool gpiochip_line_is_persistent ( struct gpio_chip * chip , unsigned int offset )
{
if ( offset > = chip - > ngpio )
return false ;
2017-11-30 06:55:24 +03:00
return ! test_bit ( FLAG_TRANSITORY , & chip - > gpiodev - > descs [ offset ] . flags ) ;
2017-05-23 17:47:29 +03:00
}
EXPORT_SYMBOL_GPL ( gpiochip_line_is_persistent ) ;
2013-10-17 21:21:36 +04:00
/**
* gpiod_get_raw_value_cansleep ( ) - return a gpio ' s raw value
* @ desc : gpio whose value will be returned
*
* Return the GPIO ' s raw value , i . e . the value of the physical line disregarding
2015-08-28 19:44:18 +03:00
* its ACTIVE_LOW status , or negative errno on failure .
2013-10-17 21:21:36 +04:00
*
* This function is to be called from contexts that can sleep .
2008-02-05 09:28:20 +03:00
*/
2013-10-17 21:21:36 +04:00
int gpiod_get_raw_value_cansleep ( const struct gpio_desc * desc )
2008-02-05 09:28:20 +03:00
{
might_sleep_if ( extra_checks ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
2017-09-26 21:58:28 +03:00
return gpiod_get_raw_value_commit ( desc ) ;
2008-02-05 09:28:20 +03:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_get_raw_value_cansleep ) ;
2013-02-02 20:29:29 +04:00
2013-10-17 21:21:36 +04:00
/**
* gpiod_get_value_cansleep ( ) - return a gpio ' s value
* @ desc : gpio whose value will be returned
*
* Return the GPIO ' s logical value , i . e . taking the ACTIVE_LOW status into
2015-08-28 19:44:18 +03:00
* account , or negative errno on failure .
2013-10-17 21:21:36 +04:00
*
* This function is to be called from contexts that can sleep .
*/
int gpiod_get_value_cansleep ( const struct gpio_desc * desc )
2008-02-05 09:28:20 +03:00
{
2011-05-20 10:40:19 +04:00
int value ;
2008-02-05 09:28:20 +03:00
might_sleep_if ( extra_checks ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC ( desc ) ;
2017-09-26 21:58:28 +03:00
value = gpiod_get_raw_value_commit ( desc ) ;
2015-08-28 19:44:18 +03:00
if ( value < 0 )
return value ;
2013-10-17 21:21:36 +04:00
if ( test_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) )
value = ! value ;
2011-05-20 10:40:19 +04:00
return value ;
2008-02-05 09:28:20 +03:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_get_value_cansleep ) ;
2013-02-02 20:29:29 +04:00
2017-10-12 13:40:10 +03:00
/**
* gpiod_get_raw_array_value_cansleep ( ) - read raw values from an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2017-10-12 13:40:10 +03:00
* @ desc_array : array of GPIO descriptors whose values will be read
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap to store the read values
2017-10-12 13:40:10 +03:00
*
* Read the raw values of the GPIOs , i . e . the values of the physical lines
* without regard for their ACTIVE_LOW status . Return 0 in case of success ,
* else an error code .
*
* This function is to be called from contexts that can sleep .
*/
int gpiod_get_raw_array_value_cansleep ( unsigned int array_size ,
struct gpio_desc * * desc_array ,
2018-09-06 00:50:07 +03:00
struct gpio_array * array_info ,
2018-09-06 00:50:05 +03:00
unsigned long * value_bitmap )
2017-10-12 13:40:10 +03:00
{
might_sleep_if ( extra_checks ) ;
if ( ! desc_array )
return - EINVAL ;
return gpiod_get_array_value_complex ( true , true , array_size ,
2018-09-06 00:50:07 +03:00
desc_array , array_info ,
value_bitmap ) ;
2017-10-12 13:40:10 +03:00
}
EXPORT_SYMBOL_GPL ( gpiod_get_raw_array_value_cansleep ) ;
/**
* gpiod_get_array_value_cansleep ( ) - read values from an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2017-10-12 13:40:10 +03:00
* @ desc_array : array of GPIO descriptors whose values will be read
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap to store the read values
2017-10-12 13:40:10 +03:00
*
* Read the logical values of the GPIOs , i . e . taking their ACTIVE_LOW status
* into account . Return 0 in case of success , else an error code .
*
* This function is to be called from contexts that can sleep .
*/
int gpiod_get_array_value_cansleep ( unsigned int array_size ,
struct gpio_desc * * desc_array ,
2018-09-06 00:50:07 +03:00
struct gpio_array * array_info ,
2018-09-06 00:50:05 +03:00
unsigned long * value_bitmap )
2017-10-12 13:40:10 +03:00
{
might_sleep_if ( extra_checks ) ;
if ( ! desc_array )
return - EINVAL ;
return gpiod_get_array_value_complex ( false , true , array_size ,
2018-09-06 00:50:07 +03:00
desc_array , array_info ,
value_bitmap ) ;
2017-10-12 13:40:10 +03:00
}
EXPORT_SYMBOL_GPL ( gpiod_get_array_value_cansleep ) ;
2013-10-17 21:21:36 +04:00
/**
* gpiod_set_raw_value_cansleep ( ) - assign a gpio ' s raw value
* @ desc : gpio whose value will be assigned
* @ value : value to assign
*
* Set the raw value of the GPIO , i . e . the value of its physical line without
* regard for its ACTIVE_LOW status .
*
* This function is to be called from contexts that can sleep .
*/
void gpiod_set_raw_value_cansleep ( struct gpio_desc * desc , int value )
2013-02-02 20:29:29 +04:00
{
2008-02-05 09:28:20 +03:00
might_sleep_if ( extra_checks ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC_VOID ( desc ) ;
2017-09-26 21:58:28 +03:00
gpiod_set_raw_value_commit ( desc , value ) ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_set_raw_value_cansleep ) ;
2008-02-05 09:28:20 +03:00
2013-10-17 21:21:36 +04:00
/**
* gpiod_set_value_cansleep ( ) - assign a gpio ' s value
* @ desc : gpio whose value will be assigned
* @ value : value to assign
*
* Set the logical value of the GPIO , i . e . taking its ACTIVE_LOW status into
* account
*
* This function is to be called from contexts that can sleep .
*/
void gpiod_set_value_cansleep ( struct gpio_desc * desc , int value )
2008-02-05 09:28:20 +03:00
{
might_sleep_if ( extra_checks ) ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
VALIDATE_DESC_VOID ( desc ) ;
2018-01-09 21:08:21 +03:00
gpiod_set_value_nocheck ( desc , value ) ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:36 +04:00
EXPORT_SYMBOL_GPL ( gpiod_set_value_cansleep ) ;
2008-02-05 09:28:20 +03:00
2014-11-04 19:12:06 +03:00
/**
2015-05-13 12:04:56 +03:00
* gpiod_set_raw_array_value_cansleep ( ) - assign values to an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2014-11-04 19:12:06 +03:00
* @ desc_array : array of GPIO descriptors whose values will be assigned
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap of values to assign
2014-11-04 19:12:06 +03:00
*
* Set the raw values of the GPIOs , i . e . the values of the physical lines
* without regard for their ACTIVE_LOW status .
*
* This function is to be called from contexts that can sleep .
*/
2018-05-21 20:57:07 +03:00
int gpiod_set_raw_array_value_cansleep ( unsigned int array_size ,
2018-09-27 14:38:10 +03:00
struct gpio_desc * * desc_array ,
struct gpio_array * array_info ,
unsigned long * value_bitmap )
2014-11-04 19:12:06 +03:00
{
might_sleep_if ( extra_checks ) ;
if ( ! desc_array )
2018-05-21 20:57:07 +03:00
return - EINVAL ;
return gpiod_set_array_value_complex ( true , true , array_size , desc_array ,
2018-09-06 00:50:07 +03:00
array_info , value_bitmap ) ;
2014-11-04 19:12:06 +03:00
}
2015-05-13 12:04:56 +03:00
EXPORT_SYMBOL_GPL ( gpiod_set_raw_array_value_cansleep ) ;
2014-11-04 19:12:06 +03:00
2017-08-15 07:59:55 +03:00
/**
* gpiod_add_lookup_tables ( ) - register GPIO device consumers
* @ tables : list of tables of consumers to register
* @ n : number of tables in the list
*/
void gpiod_add_lookup_tables ( struct gpiod_lookup_table * * tables , size_t n )
{
unsigned int i ;
mutex_lock ( & gpio_lookup_lock ) ;
for ( i = 0 ; i < n ; i + + )
list_add_tail ( & tables [ i ] - > list , & gpio_lookup_list ) ;
mutex_unlock ( & gpio_lookup_lock ) ;
}
2014-11-04 19:12:06 +03:00
/**
2015-05-13 12:04:56 +03:00
* gpiod_set_array_value_cansleep ( ) - assign values to an array of GPIOs
2018-09-06 00:50:05 +03:00
* @ array_size : number of elements in the descriptor array / value bitmap
2014-11-04 19:12:06 +03:00
* @ desc_array : array of GPIO descriptors whose values will be assigned
2018-09-06 00:50:07 +03:00
* @ array_info : information on applicability of fast bitmap processing path
2018-09-06 00:50:05 +03:00
* @ value_bitmap : bitmap of values to assign
2014-11-04 19:12:06 +03:00
*
* Set the logical values of the GPIOs , i . e . taking their ACTIVE_LOW status
* into account .
*
* This function is to be called from contexts that can sleep .
*/
2018-09-27 14:38:09 +03:00
int gpiod_set_array_value_cansleep ( unsigned int array_size ,
struct gpio_desc * * desc_array ,
struct gpio_array * array_info ,
unsigned long * value_bitmap )
2014-11-04 19:12:06 +03:00
{
might_sleep_if ( extra_checks ) ;
if ( ! desc_array )
2018-09-27 14:38:09 +03:00
return - EINVAL ;
return gpiod_set_array_value_complex ( false , true , array_size ,
desc_array , array_info ,
value_bitmap ) ;
2014-11-04 19:12:06 +03:00
}
2015-05-13 12:04:56 +03:00
EXPORT_SYMBOL_GPL ( gpiod_set_array_value_cansleep ) ;
2014-11-04 19:12:06 +03:00
2013-10-17 21:21:38 +04:00
/**
2013-12-03 07:20:11 +04:00
* gpiod_add_lookup_table ( ) - register GPIO device consumers
* @ table : table of consumers to register
2013-10-17 21:21:38 +04:00
*/
2013-12-03 07:20:11 +04:00
void gpiod_add_lookup_table ( struct gpiod_lookup_table * table )
2013-10-17 21:21:38 +04:00
{
mutex_lock ( & gpio_lookup_lock ) ;
2013-12-03 07:20:11 +04:00
list_add_tail ( & table - > list , & gpio_lookup_list ) ;
2013-10-17 21:21:38 +04:00
mutex_unlock ( & gpio_lookup_lock ) ;
}
2017-04-21 00:23:20 +03:00
EXPORT_SYMBOL_GPL ( gpiod_add_lookup_table ) ;
2013-10-17 21:21:38 +04:00
2015-06-26 12:02:04 +03:00
/**
* gpiod_remove_lookup_table ( ) - unregister GPIO device consumers
* @ table : table of consumers to unregister
*/
void gpiod_remove_lookup_table ( struct gpiod_lookup_table * table )
{
mutex_lock ( & gpio_lookup_lock ) ;
list_del ( & table - > list ) ;
mutex_unlock ( & gpio_lookup_lock ) ;
}
2017-04-21 00:23:20 +03:00
EXPORT_SYMBOL_GPL ( gpiod_remove_lookup_table ) ;
2015-06-26 12:02:04 +03:00
2018-04-10 23:30:28 +03:00
/**
* gpiod_add_hogs ( ) - register a set of GPIO hogs from machine code
* @ hogs : table of gpio hog entries with a zeroed sentinel at the end
*/
void gpiod_add_hogs ( struct gpiod_hog * hogs )
{
struct gpio_chip * chip ;
struct gpiod_hog * hog ;
mutex_lock ( & gpio_machine_hogs_mutex ) ;
for ( hog = & hogs [ 0 ] ; hog - > chip_label ; hog + + ) {
list_add_tail ( & hog - > list , & gpio_machine_hogs ) ;
/*
* The chip may have been registered earlier , so check if it
* exists and , if so , try to hog the line now .
*/
chip = find_chip_by_name ( hog - > chip_label ) ;
if ( chip )
gpiochip_machine_hog ( chip , hog ) ;
}
mutex_unlock ( & gpio_machine_hogs_mutex ) ;
}
EXPORT_SYMBOL_GPL ( gpiod_add_hogs ) ;
2013-12-03 07:20:11 +04:00
static struct gpiod_lookup_table * gpiod_find_lookup_table ( struct device * dev )
2013-10-17 21:21:38 +04:00
{
const char * dev_id = dev ? dev_name ( dev ) : NULL ;
2013-12-03 07:20:11 +04:00
struct gpiod_lookup_table * table ;
2013-10-17 21:21:38 +04:00
mutex_lock ( & gpio_lookup_lock ) ;
2013-12-03 07:20:11 +04:00
list_for_each_entry ( table , & gpio_lookup_list , list ) {
if ( table - > dev_id & & dev_id ) {
/*
* Valid strings on both ends , must be identical to have
* a match
*/
if ( ! strcmp ( table - > dev_id , dev_id ) )
goto found ;
} else {
/*
* One of the pointers is NULL , so both must be to have
* a match
*/
if ( dev_id = = table - > dev_id )
goto found ;
}
}
table = NULL ;
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
found :
mutex_unlock ( & gpio_lookup_lock ) ;
return table ;
}
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
static struct gpio_desc * gpiod_find ( struct device * dev , const char * con_id ,
2019-04-10 18:39:16 +03:00
unsigned int idx , unsigned long * flags )
2013-12-03 07:20:11 +04:00
{
2013-12-11 06:32:28 +04:00
struct gpio_desc * desc = ERR_PTR ( - ENOENT ) ;
2013-12-03 07:20:11 +04:00
struct gpiod_lookup_table * table ;
struct gpiod_lookup * p ;
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
table = gpiod_find_lookup_table ( dev ) ;
if ( ! table )
return desc ;
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
for ( p = & table - > table [ 0 ] ; p - > chip_label ; p + + ) {
struct gpio_chip * chip ;
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
/* idx must always match exactly */
2013-10-17 21:21:38 +04:00
if ( p - > idx ! = idx )
continue ;
2013-12-03 07:20:11 +04:00
/* If the lookup entry has a con_id, require exact match */
if ( p - > con_id & & ( ! con_id | | strcmp ( p - > con_id , con_id ) ) )
continue ;
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
chip = find_chip_by_name ( p - > chip_label ) ;
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
if ( ! chip ) {
2018-07-04 01:18:19 +03:00
/*
* As the lookup table indicates a chip with
* p - > chip_label should exist , assume it may
* still appear later and let the interested
* consumer be probed again or let the Deferred
* Probe infrastructure handle the error .
*/
dev_warn ( dev , " cannot find GPIO chip %s, deferring \n " ,
p - > chip_label ) ;
return ERR_PTR ( - EPROBE_DEFER ) ;
2013-12-03 07:20:11 +04:00
}
2013-10-17 21:21:38 +04:00
2013-12-03 07:20:11 +04:00
if ( chip - > ngpio < = p - > chip_hwnum ) {
2013-12-11 06:32:28 +04:00
dev_err ( dev ,
" requested GPIO %d is out of range [0..%d] for chip %s \n " ,
idx , chip - > ngpio , chip - > label ) ;
return ERR_PTR ( - EINVAL ) ;
2013-10-17 21:21:38 +04:00
}
2014-02-09 12:43:54 +04:00
desc = gpiochip_get_desc ( chip , p - > chip_hwnum ) ;
2013-12-03 07:20:11 +04:00
* flags = p - > flags ;
2013-10-17 21:21:38 +04:00
2013-12-11 06:32:28 +04:00
return desc ;
2013-10-17 21:21:38 +04:00
}
return desc ;
}
2015-02-11 19:27:58 +03:00
static int platform_gpio_count ( struct device * dev , const char * con_id )
{
struct gpiod_lookup_table * table ;
struct gpiod_lookup * p ;
unsigned int count = 0 ;
table = gpiod_find_lookup_table ( dev ) ;
if ( ! table )
return - ENOENT ;
for ( p = & table - > table [ 0 ] ; p - > chip_label ; p + + ) {
if ( ( con_id & & p - > con_id & & ! strcmp ( con_id , p - > con_id ) ) | |
( ! con_id & & ! p - > con_id ) )
count + + ;
}
if ( ! count )
return - ENOENT ;
return count ;
}
2019-09-13 06:22:39 +03:00
/**
* fwnode_gpiod_get_index - obtain a GPIO from firmware node
* @ fwnode : handle of the firmware node
* @ con_id : function within the GPIO consumer
* @ index : index of the GPIO to obtain for the consumer
* @ flags : GPIO initialization flags
* @ label : label to attach to the requested GPIO
*
* This function can be used for drivers that get their configuration
* from opaque firmware .
*
* The function properly finds the corresponding GPIO using whatever is the
* underlying firmware interface and then makes sure that the GPIO
* descriptor is requested before it is returned to the caller .
*
* Returns :
* On successful request the GPIO pin is configured in accordance with
* provided @ flags .
*
* In case of error an ERR_PTR ( ) is returned .
*/
struct gpio_desc * fwnode_gpiod_get_index ( struct fwnode_handle * fwnode ,
const char * con_id , int index ,
enum gpiod_flags flags ,
const char * label )
{
struct gpio_desc * desc ;
char prop_name [ 32 ] ; /* 32 is max size of property name */
unsigned int i ;
for ( i = 0 ; i < ARRAY_SIZE ( gpio_suffixes ) ; i + + ) {
if ( con_id )
snprintf ( prop_name , sizeof ( prop_name ) , " %s-%s " ,
con_id , gpio_suffixes [ i ] ) ;
else
snprintf ( prop_name , sizeof ( prop_name ) , " %s " ,
gpio_suffixes [ i ] ) ;
desc = fwnode_get_named_gpiod ( fwnode , prop_name , index , flags ,
label ) ;
if ( ! IS_ERR ( desc ) | | ( PTR_ERR ( desc ) ! = - ENOENT ) )
break ;
}
return desc ;
}
EXPORT_SYMBOL_GPL ( fwnode_gpiod_get_index ) ;
2015-02-11 19:27:58 +03:00
/**
* gpiod_count - return the number of GPIOs associated with a device / function
* or - ENOENT if no GPIO has been assigned to the requested function
* @ dev : GPIO consumer , can be NULL for system - global GPIOs
* @ con_id : function within the GPIO consumer
*/
int gpiod_count ( struct device * dev , const char * con_id )
{
int count = - ENOENT ;
if ( IS_ENABLED ( CONFIG_OF ) & & dev & & dev - > of_node )
2019-07-17 10:10:01 +03:00
count = of_gpio_get_count ( dev , con_id ) ;
2015-02-11 19:27:58 +03:00
else if ( IS_ENABLED ( CONFIG_ACPI ) & & dev & & ACPI_HANDLE ( dev ) )
count = acpi_gpio_count ( dev , con_id ) ;
if ( count < 0 )
count = platform_gpio_count ( dev , con_id ) ;
return count ;
}
EXPORT_SYMBOL_GPL ( gpiod_count ) ;
2013-10-17 21:21:38 +04:00
/**
2014-04-25 18:54:22 +04:00
* gpiod_get - obtain a GPIO for a given GPIO function
2013-12-03 07:20:11 +04:00
* @ dev : GPIO consumer , can be NULL for system - global GPIOs
2013-10-17 21:21:38 +04:00
* @ con_id : function within the GPIO consumer
2014-07-25 18:38:36 +04:00
* @ flags : optional GPIO initialization flags
2013-10-17 21:21:38 +04:00
*
* Return the GPIO descriptor corresponding to the function con_id of device
2013-12-11 06:32:28 +04:00
* dev , - ENOENT if no GPIO has been assigned to the requested function , or
2015-05-18 21:41:43 +03:00
* another IS_ERR ( ) code if an error occurred while trying to acquire the GPIO .
2013-10-17 21:21:38 +04:00
*/
2015-02-11 13:52:37 +03:00
struct gpio_desc * __must_check gpiod_get ( struct device * dev , const char * con_id ,
2014-07-25 18:38:36 +04:00
enum gpiod_flags flags )
2013-10-17 21:21:38 +04:00
{
2014-07-25 18:38:36 +04:00
return gpiod_get_index ( dev , con_id , 0 , flags ) ;
2013-10-17 21:21:38 +04:00
}
2015-02-11 13:52:37 +03:00
EXPORT_SYMBOL_GPL ( gpiod_get ) ;
2013-10-17 21:21:38 +04:00
2014-04-25 19:10:06 +04:00
/**
* gpiod_get_optional - obtain an optional GPIO for a given GPIO function
* @ dev : GPIO consumer , can be NULL for system - global GPIOs
* @ con_id : function within the GPIO consumer
2014-07-25 18:38:36 +04:00
* @ flags : optional GPIO initialization flags
2014-04-25 19:10:06 +04:00
*
* This is equivalent to gpiod_get ( ) , except that when no GPIO was assigned to
* the requested function it will return NULL . This is convenient for drivers
* that need to handle optional GPIOs .
*/
2015-02-11 13:52:37 +03:00
struct gpio_desc * __must_check gpiod_get_optional ( struct device * dev ,
2014-07-25 18:38:36 +04:00
const char * con_id ,
enum gpiod_flags flags )
2014-04-25 19:10:06 +04:00
{
2014-07-25 18:38:36 +04:00
return gpiod_get_index_optional ( dev , con_id , 0 , flags ) ;
2014-04-25 19:10:06 +04:00
}
2015-02-11 13:52:37 +03:00
EXPORT_SYMBOL_GPL ( gpiod_get_optional ) ;
2014-04-25 19:10:06 +04:00
2015-02-02 20:44:44 +03:00
/**
* gpiod_configure_flags - helper function to configure a given GPIO
* @ desc : gpio whose value will be assigned
* @ con_id : function within the GPIO consumer
2019-04-10 18:39:16 +03:00
* @ lflags : bitmask of gpio_lookup_flags GPIO_ * values - returned from
* of_find_gpio ( ) or of_get_gpio_hog ( )
2015-02-02 20:44:44 +03:00
* @ dflags : gpiod_flags - optional GPIO initialization flags
*
* Return 0 on success , - ENOENT if no GPIO has been assigned to the
* requested function and / or index , or another IS_ERR ( ) code if an error
* occurred while trying to acquire the GPIO .
*/
2017-05-23 20:03:16 +03:00
int gpiod_configure_flags ( struct gpio_desc * desc , const char * con_id ,
2016-07-03 19:32:05 +03:00
unsigned long lflags , enum gpiod_flags dflags )
2015-02-02 20:44:44 +03:00
{
2019-07-16 12:11:45 +03:00
int ret ;
2015-02-02 20:44:44 +03:00
2016-07-03 19:32:05 +03:00
if ( lflags & GPIO_ACTIVE_LOW )
set_bit ( FLAG_ACTIVE_LOW , & desc - > flags ) ;
2017-09-10 20:26:22 +03:00
2016-07-03 19:32:05 +03:00
if ( lflags & GPIO_OPEN_DRAIN )
set_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) ;
2017-09-10 20:26:22 +03:00
else if ( dflags & GPIOD_FLAGS_BIT_OPEN_DRAIN ) {
/*
* This enforces open drain mode from the consumer side .
* This is necessary for some busses like I2C , but the lookup
* should * REALLY * have specified them as open drain in the
* first place , so print a little warning here .
*/
set_bit ( FLAG_OPEN_DRAIN , & desc - > flags ) ;
gpiod_warn ( desc ,
" enforced open drain please flag it properly in DT/ACPI DSDT/board file \n " ) ;
}
2016-07-03 19:32:05 +03:00
if ( lflags & GPIO_OPEN_SOURCE )
set_bit ( FLAG_OPEN_SOURCE , & desc - > flags ) ;
2017-11-30 06:55:24 +03:00
2019-02-07 19:28:58 +03:00
if ( ( lflags & GPIO_PULL_UP ) & & ( lflags & GPIO_PULL_DOWN ) ) {
gpiod_err ( desc ,
" both pull-up and pull-down enabled, invalid configuration \n " ) ;
return - EINVAL ;
}
if ( lflags & GPIO_PULL_UP )
set_bit ( FLAG_PULL_UP , & desc - > flags ) ;
else if ( lflags & GPIO_PULL_DOWN )
set_bit ( FLAG_PULL_DOWN , & desc - > flags ) ;
2019-07-16 12:11:45 +03:00
ret = gpiod_set_transitory ( desc , ( lflags & GPIO_TRANSITORY ) ) ;
if ( ret < 0 )
return ret ;
2016-07-03 19:32:05 +03:00
2015-02-02 20:44:44 +03:00
/* No particular flag request, return here... */
if ( ! ( dflags & GPIOD_FLAGS_BIT_DIR_SET ) ) {
pr_debug ( " no flags found for %s \n " , con_id ) ;
return 0 ;
}
/* Process flags */
if ( dflags & GPIOD_FLAGS_BIT_DIR_OUT )
2019-07-16 12:11:45 +03:00
ret = gpiod_direction_output ( desc ,
2016-11-14 01:02:44 +03:00
! ! ( dflags & GPIOD_FLAGS_BIT_DIR_VAL ) ) ;
2015-02-02 20:44:44 +03:00
else
2019-07-16 12:11:45 +03:00
ret = gpiod_direction_input ( desc ) ;
2015-02-02 20:44:44 +03:00
2019-07-16 12:11:45 +03:00
return ret ;
2015-02-02 20:44:44 +03:00
}
2013-10-17 21:21:38 +04:00
/**
* gpiod_get_index - obtain a GPIO from a multi - index GPIO function
2013-12-05 13:26:26 +04:00
* @ dev : GPIO consumer , can be NULL for system - global GPIOs
2013-10-17 21:21:38 +04:00
* @ con_id : function within the GPIO consumer
* @ idx : index of the GPIO to obtain in the consumer
2014-07-25 18:38:36 +04:00
* @ flags : optional GPIO initialization flags
2013-10-17 21:21:38 +04:00
*
* This variant of gpiod_get ( ) allows to access GPIOs other than the first
* defined one for functions that define several GPIOs .
*
2013-12-11 06:32:28 +04:00
* Return a valid GPIO descriptor , - ENOENT if no GPIO has been assigned to the
* requested function and / or index , or another IS_ERR ( ) code if an error
2015-05-18 21:41:43 +03:00
* occurred while trying to acquire the GPIO .
2013-10-17 21:21:38 +04:00
*/
2015-02-11 13:52:37 +03:00
struct gpio_desc * __must_check gpiod_get_index ( struct device * dev ,
2013-10-17 21:21:38 +04:00
const char * con_id ,
2014-07-25 18:38:36 +04:00
unsigned int idx ,
enum gpiod_flags flags )
2013-10-17 21:21:38 +04:00
{
2019-04-10 18:39:17 +03:00
unsigned long lookupflags = GPIO_LOOKUP_FLAGS_DEFAULT ;
2013-11-23 14:34:50 +04:00
struct gpio_desc * desc = NULL ;
2019-07-16 12:11:45 +03:00
int ret ;
2018-01-16 10:29:50 +03:00
/* Maybe we have a device name, maybe not */
const char * devname = dev ? dev_name ( dev ) : " ? " ;
2013-10-17 21:21:38 +04:00
dev_dbg ( dev , " GPIO lookup for consumer %s \n " , con_id ) ;
2015-03-11 01:08:57 +03:00
if ( dev ) {
/* Using device tree? */
if ( IS_ENABLED ( CONFIG_OF ) & & dev - > of_node ) {
dev_dbg ( dev , " using device tree for GPIO lookup \n " ) ;
desc = of_find_gpio ( dev , con_id , idx , & lookupflags ) ;
} else if ( ACPI_COMPANION ( dev ) ) {
dev_dbg ( dev , " using ACPI for GPIO lookup \n " ) ;
2017-05-23 20:03:23 +03:00
desc = acpi_find_gpio ( dev , con_id , idx , & flags , & lookupflags ) ;
2015-03-11 01:08:57 +03:00
}
2013-11-23 14:34:50 +04:00
}
/*
* Either we are not using DT or ACPI , or their lookup did not return
* a result . In that case , use platform lookup as a fallback .
*/
2013-12-11 06:32:28 +04:00
if ( ! desc | | desc = = ERR_PTR ( - ENOENT ) ) {
2014-09-19 11:39:25 +04:00
dev_dbg ( dev , " using lookup tables for GPIO lookup \n " ) ;
2014-07-25 18:38:36 +04:00
desc = gpiod_find ( dev , con_id , idx , & lookupflags ) ;
2013-10-17 21:21:38 +04:00
}
if ( IS_ERR ( desc ) ) {
2018-02-27 11:12:13 +03:00
dev_dbg ( dev , " No GPIO consumer %s found \n " , con_id ) ;
2013-10-17 21:21:38 +04:00
return desc ;
}
2018-01-16 10:29:50 +03:00
/*
* If a connection label was passed use that , else attempt to use
* the device name as label
*/
2019-07-16 12:11:45 +03:00
ret = gpiod_request ( desc , con_id ? con_id : devname ) ;
if ( ret < 0 ) {
if ( ret = = - EBUSY & & flags & GPIOD_FLAGS_BIT_NONEXCLUSIVE ) {
2018-10-12 15:54:12 +03:00
/*
* This happens when there are several consumers for
* the same GPIO line : we just return here without
* further initialization . It is a bit if a hack .
* This is necessary to support fixed regulators .
*
* FIXME : Make this more sane and safe .
*/
dev_info ( dev , " nonexclusive access to GPIO for %s \n " ,
con_id ? con_id : devname ) ;
return desc ;
} else {
2019-07-16 12:11:45 +03:00
return ERR_PTR ( ret ) ;
2018-10-12 15:54:12 +03:00
}
}
2013-10-17 21:21:38 +04:00
2019-07-16 12:11:45 +03:00
ret = gpiod_configure_flags ( desc , con_id , lookupflags , flags ) ;
2017-12-29 04:07:54 +03:00
if ( ret < 0 ) {
2014-07-25 18:38:36 +04:00
dev_dbg ( dev , " setup of GPIO %s failed \n " , con_id ) ;
2017-12-29 04:07:54 +03:00
gpiod_put ( desc ) ;
return ERR_PTR ( ret ) ;
}
return desc ;
}
2015-02-11 13:52:37 +03:00
EXPORT_SYMBOL_GPL ( gpiod_get_index ) ;
2017-12-29 04:07:54 +03:00
2014-10-21 15:33:59 +04:00
/**
* fwnode_get_named_gpiod - obtain a GPIO from firmware node
* @ fwnode : handle of the firmware node
* @ propname : name of the firmware property representing the GPIO
2017-12-29 04:07:54 +03:00
* @ index : index of the GPIO to obtain for the consumer
2017-01-09 17:02:28 +03:00
* @ dflags : GPIO initialization flags
2017-07-24 17:57:22 +03:00
* @ label : label to attach to the requested GPIO
2014-10-21 15:33:59 +04:00
*
* This function can be used for drivers that get their configuration
2017-12-29 04:07:54 +03:00
* from opaque firmware .
2014-10-21 15:33:59 +04:00
*
2017-12-29 04:07:54 +03:00
* The function properly finds the corresponding GPIO using whatever is the
2014-10-21 15:33:59 +04:00
* underlying firmware interface and then makes sure that the GPIO
* descriptor is requested before it is returned to the caller .
*
2017-07-24 17:57:22 +03:00
* Returns :
2017-02-28 18:03:12 +03:00
* On successful request the GPIO pin is configured in accordance with
2017-01-09 17:02:28 +03:00
* provided @ dflags .
*
2014-10-21 15:33:59 +04:00
* In case of error an ERR_PTR ( ) is returned .
*/
struct gpio_desc * fwnode_get_named_gpiod ( struct fwnode_handle * fwnode ,
2017-02-02 16:53:11 +03:00
const char * propname , int index ,
2017-01-12 19:39:24 +03:00
enum gpiod_flags dflags ,
const char * label )
2014-10-21 15:33:59 +04:00
{
2019-04-10 18:39:17 +03:00
unsigned long lflags = GPIO_LOOKUP_FLAGS_DEFAULT ;
2014-10-21 15:33:59 +04:00
struct gpio_desc * desc = ERR_PTR ( - ENODEV ) ;
int ret ;
if ( ! fwnode )
return ERR_PTR ( - EINVAL ) ;
if ( is_of_node ( fwnode ) ) {
2017-12-29 04:07:54 +03:00
desc = gpiod_get_from_of_node ( to_of_node ( fwnode ) ,
propname , index ,
dflags ,
label ) ;
return desc ;
2014-10-21 15:33:59 +04:00
} else if ( is_acpi_node ( fwnode ) ) {
struct acpi_gpio_info info ;
2017-02-02 16:53:11 +03:00
desc = acpi_node_get_gpiod ( fwnode , propname , index , & info ) ;
2017-12-29 04:07:54 +03:00
if ( IS_ERR ( desc ) )
return desc ;
2014-10-21 15:33:59 +04:00
2017-12-29 04:07:54 +03:00
acpi_gpio_update_gpiod_flags ( & dflags , & info ) ;
2019-04-10 18:39:20 +03:00
acpi_gpio_update_gpiod_lookup_flags ( & lflags , & info ) ;
2017-12-29 04:07:54 +03:00
}
2014-10-21 15:33:59 +04:00
2017-12-29 04:07:54 +03:00
/* Currently only ACPI takes this path */
2017-01-12 19:39:24 +03:00
ret = gpiod_request ( desc , label ) ;
2016-07-03 19:32:05 +03:00
if ( ret )
return ERR_PTR ( ret ) ;
2017-01-09 17:02:28 +03:00
ret = gpiod_configure_flags ( desc , propname , lflags , dflags ) ;
if ( ret < 0 ) {
gpiod_put ( desc ) ;
return ERR_PTR ( ret ) ;
2015-10-13 00:20:21 +03:00
}
2014-10-21 15:33:59 +04:00
return desc ;
}
EXPORT_SYMBOL_GPL ( fwnode_get_named_gpiod ) ;
2014-04-25 19:10:06 +04:00
/**
* gpiod_get_index_optional - obtain an optional GPIO from a multi - index GPIO
* function
* @ dev : GPIO consumer , can be NULL for system - global GPIOs
* @ con_id : function within the GPIO consumer
* @ index : index of the GPIO to obtain in the consumer
2014-07-25 18:38:36 +04:00
* @ flags : optional GPIO initialization flags
2014-04-25 19:10:06 +04:00
*
* This is equivalent to gpiod_get_index ( ) , except that when no GPIO with the
* specified index was assigned to the requested function it will return NULL .
* This is convenient for drivers that need to handle optional GPIOs .
*/
2015-02-11 13:52:37 +03:00
struct gpio_desc * __must_check gpiod_get_index_optional ( struct device * dev ,
2014-04-25 19:10:06 +04:00
const char * con_id ,
2014-07-25 18:38:36 +04:00
unsigned int index ,
enum gpiod_flags flags )
2014-04-25 19:10:06 +04:00
{
struct gpio_desc * desc ;
2014-07-25 18:38:36 +04:00
desc = gpiod_get_index ( dev , con_id , index , flags ) ;
2014-04-25 19:10:06 +04:00
if ( IS_ERR ( desc ) ) {
if ( PTR_ERR ( desc ) = = - ENOENT )
return NULL ;
}
return desc ;
}
2015-02-11 13:52:37 +03:00
EXPORT_SYMBOL_GPL ( gpiod_get_index_optional ) ;
2014-04-25 19:10:06 +04:00
2015-02-02 20:44:44 +03:00
/**
* gpiod_hog - Hog the specified GPIO desc given the provided flags
* @ desc : gpio whose value will be assigned
* @ name : gpio line name
2019-04-10 18:39:16 +03:00
* @ lflags : bitmask of gpio_lookup_flags GPIO_ * values - returned from
* of_find_gpio ( ) or of_get_gpio_hog ( )
2015-02-02 20:44:44 +03:00
* @ dflags : gpiod_flags - optional GPIO initialization flags
*/
int gpiod_hog ( struct gpio_desc * desc , const char * name ,
unsigned long lflags , enum gpiod_flags dflags )
{
struct gpio_chip * chip ;
struct gpio_desc * local_desc ;
int hwnum ;
2019-07-16 12:11:45 +03:00
int ret ;
2015-02-02 20:44:44 +03:00
chip = gpiod_to_chip ( desc ) ;
hwnum = gpio_chip_hwgpio ( desc ) ;
2019-04-26 15:40:18 +03:00
local_desc = gpiochip_request_own_desc ( chip , hwnum , name ,
lflags , dflags ) ;
2015-02-02 20:44:44 +03:00
if ( IS_ERR ( local_desc ) ) {
2019-07-16 12:11:45 +03:00
ret = PTR_ERR ( local_desc ) ;
2016-03-11 16:43:21 +03:00
pr_err ( " requesting hog GPIO %s (chip %s, offset %d) failed, %d \n " ,
2019-07-16 12:11:45 +03:00
name , chip - > label , hwnum , ret ) ;
return ret ;
2015-02-02 20:44:44 +03:00
}
/* Mark GPIO as hogged so it can be identified and removed later */
set_bit ( FLAG_IS_HOGGED , & desc - > flags ) ;
pr_info ( " GPIO line %d (%s) hogged as %s%s \n " ,
desc_to_gpio ( desc ) , name ,
( dflags & GPIOD_FLAGS_BIT_DIR_OUT ) ? " output " : " input " ,
( dflags & GPIOD_FLAGS_BIT_DIR_OUT ) ?
( dflags & GPIOD_FLAGS_BIT_DIR_VAL ) ? " /high " : " /low " : " " ) ;
return 0 ;
}
/**
* gpiochip_free_hogs - Scan gpio - controller chip and release GPIO hog
* @ chip : gpio chip to act on
*/
static void gpiochip_free_hogs ( struct gpio_chip * chip )
{
int id ;
for ( id = 0 ; id < chip - > ngpio ; id + + ) {
2016-02-09 15:51:59 +03:00
if ( test_bit ( FLAG_IS_HOGGED , & chip - > gpiodev - > descs [ id ] . flags ) )
gpiochip_free_own_desc ( & chip - > gpiodev - > descs [ id ] ) ;
2015-02-02 20:44:44 +03:00
}
}
2015-02-11 19:27:58 +03:00
/**
* gpiod_get_array - obtain multiple GPIOs from a multi - index GPIO function
* @ dev : GPIO consumer , can be NULL for system - global GPIOs
* @ con_id : function within the GPIO consumer
* @ flags : optional GPIO initialization flags
*
* This function acquires all the GPIOs defined under a given function .
*
* Return a struct gpio_descs containing an array of descriptors , - ENOENT if
* no GPIO has been assigned to the requested function , or another IS_ERR ( )
* code if an error occurred while trying to acquire the GPIOs .
*/
struct gpio_descs * __must_check gpiod_get_array ( struct device * dev ,
const char * con_id ,
enum gpiod_flags flags )
{
struct gpio_desc * desc ;
struct gpio_descs * descs ;
2018-09-06 00:50:06 +03:00
struct gpio_array * array_info = NULL ;
struct gpio_chip * chip ;
int count , bitmap_size ;
2015-02-11 19:27:58 +03:00
count = gpiod_count ( dev , con_id ) ;
if ( count < 0 )
return ERR_PTR ( count ) ;
treewide: Use struct_size() for kmalloc()-family
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
struct foo {
int stuff;
void *entry[];
};
instance = kmalloc(sizeof(struct foo) + sizeof(void *) * count, GFP_KERNEL);
Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:
instance = kmalloc(struct_size(instance, entry, count), GFP_KERNEL);
This patch makes the changes for kmalloc()-family (and kvmalloc()-family)
uses. It was done via automatic conversion with manual review for the
"CHECKME" non-standard cases noted below, using the following Coccinelle
script:
// pkey_cache = kmalloc(sizeof *pkey_cache + tprops->pkey_tbl_len *
// sizeof *pkey_cache->table, GFP_KERNEL);
@@
identifier alloc =~ "kmalloc|kzalloc|kvmalloc|kvzalloc";
expression GFP;
identifier VAR, ELEMENT;
expression COUNT;
@@
- alloc(sizeof(*VAR) + COUNT * sizeof(*VAR->ELEMENT), GFP)
+ alloc(struct_size(VAR, ELEMENT, COUNT), GFP)
// mr = kzalloc(sizeof(*mr) + m * sizeof(mr->map[0]), GFP_KERNEL);
@@
identifier alloc =~ "kmalloc|kzalloc|kvmalloc|kvzalloc";
expression GFP;
identifier VAR, ELEMENT;
expression COUNT;
@@
- alloc(sizeof(*VAR) + COUNT * sizeof(VAR->ELEMENT[0]), GFP)
+ alloc(struct_size(VAR, ELEMENT, COUNT), GFP)
// Same pattern, but can't trivially locate the trailing element name,
// or variable name.
@@
identifier alloc =~ "kmalloc|kzalloc|kvmalloc|kvzalloc";
expression GFP;
expression SOMETHING, COUNT, ELEMENT;
@@
- alloc(sizeof(SOMETHING) + COUNT * sizeof(ELEMENT), GFP)
+ alloc(CHECKME_struct_size(&SOMETHING, ELEMENT, COUNT), GFP)
Signed-off-by: Kees Cook <keescook@chromium.org>
2018-05-08 23:45:50 +03:00
descs = kzalloc ( struct_size ( descs , desc , count ) , GFP_KERNEL ) ;
2015-02-11 19:27:58 +03:00
if ( ! descs )
return ERR_PTR ( - ENOMEM ) ;
for ( descs - > ndescs = 0 ; descs - > ndescs < count ; ) {
desc = gpiod_get_index ( dev , con_id , descs - > ndescs , flags ) ;
if ( IS_ERR ( desc ) ) {
gpiod_put_array ( descs ) ;
return ERR_CAST ( desc ) ;
}
2018-09-06 00:50:06 +03:00
2015-02-11 19:27:58 +03:00
descs - > desc [ descs - > ndescs ] = desc ;
2018-09-06 00:50:06 +03:00
chip = gpiod_to_chip ( desc ) ;
/*
2018-09-24 02:53:36 +03:00
* If pin hardware number of array member 0 is also 0 , select
* its chip as a candidate for fast bitmap processing path .
2018-09-06 00:50:06 +03:00
*/
2018-09-24 02:53:36 +03:00
if ( descs - > ndescs = = 0 & & gpio_chip_hwgpio ( desc ) = = 0 ) {
2018-09-06 00:50:06 +03:00
struct gpio_descs * array ;
bitmap_size = BITS_TO_LONGS ( chip - > ngpio > count ?
chip - > ngpio : count ) ;
array = kzalloc ( struct_size ( descs , desc , count ) +
struct_size ( array_info , invert_mask ,
3 * bitmap_size ) , GFP_KERNEL ) ;
if ( ! array ) {
gpiod_put_array ( descs ) ;
return ERR_PTR ( - ENOMEM ) ;
}
memcpy ( array , descs ,
struct_size ( descs , desc , descs - > ndescs + 1 ) ) ;
kfree ( descs ) ;
descs = array ;
array_info = ( void * ) ( descs - > desc + count ) ;
array_info - > get_mask = array_info - > invert_mask +
bitmap_size ;
array_info - > set_mask = array_info - > get_mask +
bitmap_size ;
array_info - > desc = descs - > desc ;
array_info - > size = count ;
array_info - > chip = chip ;
bitmap_set ( array_info - > get_mask , descs - > ndescs ,
count - descs - > ndescs ) ;
bitmap_set ( array_info - > set_mask , descs - > ndescs ,
count - descs - > ndescs ) ;
descs - > info = array_info ;
}
2018-09-24 02:53:36 +03:00
/* Unmark array members which don't belong to the 'fast' chip */
if ( array_info & & array_info - > chip ! = chip ) {
2018-09-06 00:50:06 +03:00
__clear_bit ( descs - > ndescs , array_info - > get_mask ) ;
__clear_bit ( descs - > ndescs , array_info - > set_mask ) ;
2018-09-24 02:53:36 +03:00
}
/*
* Detect array members which belong to the ' fast ' chip
* but their pins are not in hardware order .
*/
else if ( array_info & &
gpio_chip_hwgpio ( desc ) ! = descs - > ndescs ) {
/*
* Don ' t use fast path if all array members processed so
* far belong to the same chip as this one but its pin
* hardware number is different from its array index .
*/
if ( bitmap_full ( array_info - > get_mask , descs - > ndescs ) ) {
array_info = NULL ;
} else {
__clear_bit ( descs - > ndescs ,
array_info - > get_mask ) ;
__clear_bit ( descs - > ndescs ,
array_info - > set_mask ) ;
}
2018-09-06 00:50:06 +03:00
} else if ( array_info ) {
/* Exclude open drain or open source from fast output */
if ( gpiochip_line_is_open_drain ( chip , descs - > ndescs ) | |
gpiochip_line_is_open_source ( chip , descs - > ndescs ) )
__clear_bit ( descs - > ndescs ,
array_info - > set_mask ) ;
/* Identify 'fast' pins which require invertion */
if ( gpiod_is_active_low ( desc ) )
__set_bit ( descs - > ndescs ,
array_info - > invert_mask ) ;
}
2015-02-11 19:27:58 +03:00
descs - > ndescs + + ;
}
2018-09-06 00:50:06 +03:00
if ( array_info )
dev_dbg ( dev ,
" GPIO array info: chip=%s, size=%d, get_mask=%lx, set_mask=%lx, invert_mask=%lx \n " ,
array_info - > chip - > label , array_info - > size ,
* array_info - > get_mask , * array_info - > set_mask ,
* array_info - > invert_mask ) ;
2015-02-11 19:27:58 +03:00
return descs ;
}
EXPORT_SYMBOL_GPL ( gpiod_get_array ) ;
/**
* gpiod_get_array_optional - obtain multiple GPIOs from a multi - index GPIO
* function
* @ dev : GPIO consumer , can be NULL for system - global GPIOs
* @ con_id : function within the GPIO consumer
* @ flags : optional GPIO initialization flags
*
* This is equivalent to gpiod_get_array ( ) , except that when no GPIO was
* assigned to the requested function it will return NULL .
*/
struct gpio_descs * __must_check gpiod_get_array_optional ( struct device * dev ,
const char * con_id ,
enum gpiod_flags flags )
{
struct gpio_descs * descs ;
descs = gpiod_get_array ( dev , con_id , flags ) ;
if ( IS_ERR ( descs ) & & ( PTR_ERR ( descs ) = = - ENOENT ) )
return NULL ;
return descs ;
}
EXPORT_SYMBOL_GPL ( gpiod_get_array_optional ) ;
2013-10-17 21:21:38 +04:00
/**
* gpiod_put - dispose of a GPIO descriptor
* @ desc : GPIO descriptor to dispose of
*
* No descriptor can be used after gpiod_put ( ) has been called on it .
*/
void gpiod_put ( struct gpio_desc * desc )
{
2019-03-26 18:21:14 +03:00
if ( desc )
gpiod_free ( desc ) ;
2013-02-02 20:29:29 +04:00
}
2013-10-17 21:21:38 +04:00
EXPORT_SYMBOL_GPL ( gpiod_put ) ;
2008-02-05 09:28:20 +03:00
2015-02-11 19:27:58 +03:00
/**
* gpiod_put_array - dispose of multiple GPIO descriptors
* @ descs : struct gpio_descs containing an array of descriptors
*/
void gpiod_put_array ( struct gpio_descs * descs )
{
unsigned int i ;
for ( i = 0 ; i < descs - > ndescs ; i + + )
gpiod_put ( descs - > desc [ i ] ) ;
kfree ( descs ) ;
}
EXPORT_SYMBOL_GPL ( gpiod_put_array ) ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
static int __init gpiolib_dev_init ( void )
{
int ret ;
/* Register GPIO sysfs bus */
2018-07-03 03:39:03 +03:00
ret = bus_register ( & gpio_bus_type ) ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
if ( ret < 0 ) {
pr_err ( " gpiolib: could not register GPIO bus type \n " ) ;
return ret ;
}
ret = alloc_chrdev_region ( & gpio_devt , 0 , GPIO_DEV_MAX , " gpiochip " ) ;
if ( ret < 0 ) {
pr_err ( " gpiolib: failed to allocate char dev region \n " ) ;
bus_unregister ( & gpio_bus_type ) ;
2016-03-31 18:11:30 +03:00
} else {
gpiolib_initialized = true ;
gpiochip_setup_devs ( ) ;
gpio: add a userspace chardev ABI for GPIOs
A new chardev that is to be used for userspace GPIO access is
added in this patch. It is intended to gradually replace the
horribly broken sysfs ABI.
Using a chardev has many upsides:
- All operations are per-gpiochip, which is the actual
device underlying the GPIOs, making us tie in to the
kernel device model properly.
- Hotpluggable GPIO controllers can come and go, as this
kind of problem has been know to userspace for character
devices since ages, and if a gpiochip handle is held in
userspace we know we will break something, whereas the
sysfs is stateless.
- The one-value-per-file rule of sysfs is really hard to
maintain when you want to twist more than one knob at a time,
for example have in-kernel APIs to switch several GPIO
lines at the same time, and this will be possible to do
with a single ioctl() from userspace, saving a lot of
context switching.
We also need to add a new bus type for GPIO. This is
necessary for example for userspace coldplug, where sysfs is
traversed to find the boot-time device nodes and create the
character devices in /dev.
This new chardev ABI is *non* *optional* and can be counted
on to be present in the future, emphasizing the preference
of this ABI.
The ABI only implements one single ioctl() to get the name
and number of GPIO lines of a chip. Even this is debatable:
see it as a minimal example for review. This ABI shall be
ruthlessly reviewed and etched in stone.
The old /sys/class/gpio is still optional to compile in,
but will be deprecated.
Unique device IDs are created using IDR, which is overkill
and insanely scalable, but also well tested.
Cc: Johan Hovold <johan@kernel.org>
Cc: Michael Welling <mwelling@ieee.org>
Cc: Markus Pargmann <mpa@pengutronix.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2015-10-21 16:29:53 +03:00
}
return ret ;
}
core_initcall ( gpiolib_dev_init ) ;
2008-02-05 09:28:20 +03:00
# ifdef CONFIG_DEBUG_FS
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
static void gpiolib_dbg_show ( struct seq_file * s , struct gpio_device * gdev )
2008-02-05 09:28:20 +03:00
{
unsigned i ;
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
struct gpio_chip * chip = gdev - > chip ;
unsigned gpio = gdev - > base ;
struct gpio_desc * gdesc = & gdev - > descs [ 0 ] ;
2018-10-02 00:06:17 +03:00
bool is_out ;
bool is_irq ;
bool active_low ;
2008-02-05 09:28:20 +03:00
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
for ( i = 0 ; i < gdev - > ngpio ; i + + , gpio + + , gdesc + + ) {
2015-08-14 17:11:02 +03:00
if ( ! test_bit ( FLAG_REQUESTED , & gdesc - > flags ) ) {
if ( gdesc - > name ) {
seq_printf ( s , " gpio-%-3d (%-20.20s) \n " ,
gpio , gdesc - > name ) ;
}
2008-02-05 09:28:20 +03:00
continue ;
2015-08-14 17:11:02 +03:00
}
2008-02-05 09:28:20 +03:00
2013-02-02 20:29:29 +04:00
gpiod_get_direction ( gdesc ) ;
2008-02-05 09:28:20 +03:00
is_out = test_bit ( FLAG_IS_OUT , & gdesc - > flags ) ;
2013-09-24 13:54:38 +04:00
is_irq = test_bit ( FLAG_USED_AS_IRQ , & gdesc - > flags ) ;
2018-10-02 00:06:17 +03:00
active_low = test_bit ( FLAG_ACTIVE_LOW , & gdesc - > flags ) ;
seq_printf ( s , " gpio-%-3d (%-20.20s|%-20.20s) %s %s %s%s " ,
2015-08-14 17:11:02 +03:00
gpio , gdesc - > name ? gdesc - > name : " " , gdesc - > label ,
2008-02-05 09:28:20 +03:00
is_out ? " out " : " in " ,
2018-07-12 20:36:42 +03:00
chip - > get ? ( chip - > get ( chip , i ) ? " hi " : " lo " ) : " ? " ,
2018-10-02 00:06:17 +03:00
is_irq ? " IRQ " : " " ,
active_low ? " ACTIVE LOW " : " " ) ;
2008-02-05 09:28:20 +03:00
seq_printf ( s , " \n " ) ;
}
}
2012-04-12 15:26:01 +04:00
static void * gpiolib_seq_start ( struct seq_file * s , loff_t * pos )
2008-02-05 09:28:20 +03:00
{
2013-02-09 13:41:49 +04:00
unsigned long flags ;
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev = NULL ;
2013-02-02 20:29:27 +04:00
loff_t index = * pos ;
2008-02-05 09:28:20 +03:00
2012-04-12 15:26:01 +04:00
s - > private = " " ;
2008-02-05 09:28:20 +03:00
2013-02-09 13:41:49 +04:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
2015-10-20 12:10:38 +03:00
list_for_each_entry ( gdev , & gpio_devices , list )
2013-02-09 13:41:49 +04:00
if ( index - - = = 0 ) {
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2015-10-20 12:10:38 +03:00
return gdev ;
2012-04-12 15:26:01 +04:00
}
2013-02-09 13:41:49 +04:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2012-04-12 15:26:01 +04:00
2013-02-02 20:29:27 +04:00
return NULL ;
2012-04-12 15:26:01 +04:00
}
static void * gpiolib_seq_next ( struct seq_file * s , void * v , loff_t * pos )
{
2013-02-09 13:41:49 +04:00
unsigned long flags ;
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev = v ;
2012-04-12 15:26:01 +04:00
void * ret = NULL ;
2013-02-09 13:41:49 +04:00
spin_lock_irqsave ( & gpio_lock , flags ) ;
2015-10-20 12:10:38 +03:00
if ( list_is_last ( & gdev - > list , & gpio_devices ) )
2013-02-02 20:29:27 +04:00
ret = NULL ;
else
2015-10-20 12:10:38 +03:00
ret = list_entry ( gdev - > list . next , struct gpio_device , list ) ;
2013-02-09 13:41:49 +04:00
spin_unlock_irqrestore ( & gpio_lock , flags ) ;
2012-04-12 15:26:01 +04:00
s - > private = " \n " ;
+ + * pos ;
return ret ;
}
static void gpiolib_seq_stop ( struct seq_file * s , void * v )
{
}
static int gpiolib_seq_show ( struct seq_file * s , void * v )
{
2015-10-20 12:10:38 +03:00
struct gpio_device * gdev = v ;
struct gpio_chip * chip = gdev - > chip ;
struct device * parent ;
if ( ! chip ) {
seq_printf ( s , " %s%s: (dangling chip) " , ( char * ) s - > private ,
dev_name ( & gdev - > dev ) ) ;
return 0 ;
}
2012-04-12 15:26:01 +04:00
2015-10-20 12:10:38 +03:00
seq_printf ( s , " %s%s: GPIOs %d-%d " , ( char * ) s - > private ,
dev_name ( & gdev - > dev ) ,
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
gdev - > base , gdev - > base + gdev - > ngpio - 1 ) ;
2015-10-20 12:10:38 +03:00
parent = chip - > parent ;
if ( parent )
seq_printf ( s , " , parent: %s/%s " ,
parent - > bus ? parent - > bus - > name : " no-bus " ,
dev_name ( parent ) ) ;
2012-04-12 15:26:01 +04:00
if ( chip - > label )
seq_printf ( s , " , %s " , chip - > label ) ;
if ( chip - > can_sleep )
seq_printf ( s , " , can sleep " ) ;
seq_printf ( s , " : \n " ) ;
if ( chip - > dbg_show )
chip - > dbg_show ( s , chip ) ;
else
gpio: reflect base and ngpio into gpio_device
Some information about the GPIO chip need to stay around also
after the gpio_chip has been removed and only the gpio_device
persist. The base and ngpio are such things, for example we
don't want a new chip arriving to overlap the number space
of a dangling gpio_device, and the chardev may still query
the device for the number of lines etc.
Note that the code that assigns base and insert gpio_device
into the global list no longer check for a missing gpio_chip:
we respect the number space allocated by any other gpio_device.
As a consequence of the gdev being referenced directly from
the gpio_desc, we need to verify it differently from all
in-kernel API calls that fall through to direct queries to
the gpio_chip vtable: we first check that desc is !NULL, then
that desc->gdev is !NULL, then, if desc->gdev->chip is NULL,
we *BAIL OUT* without any error, so as to manage the case
where operations are requested on a device that is gone.
These checks were non-uniform and partly missing in the past:
so to simplify: create the macros VALIDATE_DESC() that will
return -EINVAL if the desc or desc->gdev is missing and just
0 if the chip is gone, and conversely VALIDATE_DESC_VOID()
for the case where the function does not return an error.
By using these macros, we get warning messages about missing
gdev with reference to the right function in the kernel log.
Despite the macro business this simplifies the code and make
it more readable than if we copy/paste the same descriptor
checking code into all code ABI call sites (IMHO).
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2016-02-10 12:57:36 +03:00
gpiolib_dbg_show ( s , gdev ) ;
2012-04-12 15:26:01 +04:00
2008-02-05 09:28:20 +03:00
return 0 ;
}
2012-04-12 15:26:01 +04:00
static const struct seq_operations gpiolib_seq_ops = {
. start = gpiolib_seq_start ,
. next = gpiolib_seq_next ,
. stop = gpiolib_seq_stop ,
. show = gpiolib_seq_show ,
} ;
2008-02-05 09:28:20 +03:00
static int gpiolib_open ( struct inode * inode , struct file * file )
{
2012-04-12 15:26:01 +04:00
return seq_open ( file , & gpiolib_seq_ops ) ;
2008-02-05 09:28:20 +03:00
}
2009-10-02 02:43:56 +04:00
static const struct file_operations gpiolib_operations = {
2012-04-12 15:26:01 +04:00
. owner = THIS_MODULE ,
2008-02-05 09:28:20 +03:00
. open = gpiolib_open ,
. read = seq_read ,
. llseek = seq_lseek ,
2012-04-12 15:26:01 +04:00
. release = seq_release ,
2008-02-05 09:28:20 +03:00
} ;
static int __init gpiolib_debugfs_init ( void )
{
/* /sys/kernel/debug/gpio */
2019-06-18 18:50:45 +03:00
debugfs_create_file ( " gpio " , S_IFREG | S_IRUGO , NULL , NULL ,
& gpiolib_operations ) ;
2008-02-05 09:28:20 +03:00
return 0 ;
}
subsys_initcall ( gpiolib_debugfs_init ) ;
# endif /* DEBUG_FS */