The i.MX pinctrl driver uses 5 different arrays for storing the
informations for pins. This requires five allocations. Instead,
use a struct type which is more cache friendly, readable and
requires less allocations. One array of integers is still needed
since the pinctrl framework forces us to maintain it.
This also adds checks whether the allocations are succesful which
were missing.
Sascha Hauer [Fri, 9 Aug 2013 12:20:52 +0000 (14:20 +0200)]
pinctrl: imx: do not fail when parsing a group fails
The i.MX pinctrl driver completely bails out when it detects an error
in the pinctrl nodes. This usually means that whatever error a
devicetree has the user is left blind because even the console cannot
be initialized without working pinmux.
Instead of bailing out completely, just continue probing. This makes
the pinctrl driver work, only the erroneous groups will fail later
during pin request time.
- res = platform_get_resource(pdev, IORESOURCE_MEM, n);
... when != res
- if (res == NULL) { ... \(goto l;\|return ret;\) }
... when != res
+ res = platform_get_resource(pdev, IORESOURCE_MEM, n);
e = devm_ioremap_resource(e1, res);
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Linus Walleij [Thu, 15 Aug 2013 19:38:49 +0000 (21:38 +0200)]
pinctrl: add includes and ifdefs for non-DT builds
Commit e81c8f18afc4fdd6e34d8c83814b8b5134dbb30f
"pinctrl: pinconf-generic: add generic APIs for mapping pinctrl node"
Added function prototypes with implicit dependencies
on other header files causing build warnings like this:
In file included from
arch/arm/mach-ux500/board-mop500-pins.c:12:0:
include/linux/pinctrl/pinconf-generic.h:142:3:
warning: 'struct device_node' declared inside parameter list [enabled
by default]
unsigned *reserved_maps, unsigned *num_maps);
^
include/linux/pinctrl/pinconf-generic.h:142:3:
warning: its scope is only this definition or declaration, which is
probably not what you want [enabled by default]
include/linux/pinctrl/pinconf-generic.h:142:3:
warning: 'struct pinctrl_dev' declared inside parameter list [enabled
by default]
include/linux/pinctrl/pinconf-generic.h:145:3:
warning: 'struct device_node' declared inside parameter list [enabled
by default]
unsigned *num_maps);
^
Let's just add ifdefs for non-DT systems (the actual code is
already ifdefed) and #include <linux/device.h> to get the
most important structs and forward-declare the pinctrl
core structs.
In file included from include/linux/seqlock.h:29:0,
from include/linux/time.h:5,
from include/linux/stat.h:18,
from include/linux/module.h:10,
from
drivers/pinctrl/pinctrl-nomadik.c:14:
drivers/pinctrl/pinctrl-nomadik.c: In function 'nmk_pmx_enable':
include/linux/spinlock.h:348:122: warning:
'flags' may be used uninitialized in this function
[-Wmaybe-uninitialized]
raw_spin_unlock_irqrestore(&lock->rlock, flags);
^
drivers/pinctrl/pinctrl-nomadik.c:1515:16: note:
'flags' was declared here
unsigned long flags;
The function is question was never changed but it appears
the semantic checker could previously determine that the code
path that would use the flags was going to either use it or
not, but now it can't for some reason. Just fix it up.
Reported-by: Olof Johansson <olof@lixom.net> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Stephen Warren [Fri, 9 Aug 2013 16:57:46 +0000 (10:57 -0600)]
gpio: clean up gpio-ranges documentation
This change makes documentation of the the gpio-ranges property shorter
and more succinct, more consistent with the style of the rest of the
document, and not mention Linux-specifics such as the API
pinctrl_request_gpio(); DT binding documents should be OS independant
where at all possible. As part of this, the gpio-ranges property's format
is described in BNF form, in order to match the rest of the document.
This change also deprecates the #gpio-range-cells property. Such
properties are useful when one node references a second node, and that
second node dictates the format of the reference. However, that is not
the case here; the definition of gpio-ranges itself always dictates its
format entirely, and hence the value #gpio-range-cells must always be 3,
and hence there is no point requiring any referenced node to include
this property. The only remaining need for this property is to ensure
compatibility of DTs with older SW that was written to support the
previous version of the binding.
v4:
* Mention #gpio-range-cells as being deprecated, rather than removing all
documentation of that property. This allows DTs to be written in a
backwards-compatible way if desired, and also allows older DTs to be
interpreted fully using the latest documentation.
v3:
* Mention BNF in commit description.
* Fixed typo.
* Dropped patch that removed the deprecated property from *.dts, since
it's required to boot older kernels.
Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Mark Rutland <mark.rutland@arm.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Linus Walleij [Wed, 14 Aug 2013 16:23:33 +0000 (18:23 +0200)]
pinctrl: improve warning messages
Print out the affected group name on activation of pin mux
settings, and warn if you cannot free a pin that should have
been part of a certain setting.
ChangeLog v1->v2:
- Also print the pin name in the error messages.
Cc: Sonic Zhang <sonic.zhang@analog.com> Reviewed-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Laxman Dewangan [Tue, 6 Aug 2013 13:12:35 +0000 (18:42 +0530)]
pinctrl: palmas: add pincontrol driver
TI Palmas series Power Management IC have multiple pins which can be
configured for different functionality. This pins can be configured
for different function. Also their properties like pull up/down,
open drain enable/disable are configurable.
Add support for pincontrol driver Palmas series device like TPS65913,
TPS80036. The driver supports to be register from DT only.
Changes from V1:
- Add generic property for pins and functions in pinconf-generic.
- Add APIs to map the DT and subnode.
- Move common utils APIs to the pinctrl-utils from this file.
- Update the binding document accordingly.
Changes from V2:
- Add ack by Lee.
- Correct the binding docs.
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com> Acked-by: Lee Jones <lee.jones@linaro.org> Reviewed-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Laxman Dewangan [Tue, 6 Aug 2013 13:12:34 +0000 (18:42 +0530)]
pinctrl: pinconf-generic: add generic APIs for mapping pinctrl node
Add generic APIs to map the DT node and its sub node in pinconf generic
driver. These APIs can be used from driver to parse the DT node who
uses the pinconf generic APIs for defining their nodes.
Changes from V1:
- Add generic property for pins and functions in pinconf-generic.
- Add APIs to map the DT and subnode.
- Move common utils APIs to the pinctrl-utils from this file.
- Update the binding document accordingly.
Changes from V2:
- Rebased the pinctrl binding doc on top of Stephen's cleanup.
- Rename properties "pinctrl-pins" and "pinctrl-function" to
"pins" and "function".
Laxman Dewangan [Tue, 6 Aug 2013 13:12:33 +0000 (18:42 +0530)]
pinctrl: add utility functions for add map/configs
Some of pincontrol driver needs the utility function to create map
list. The utility function needed for adding mux, configs etc.
In place of duplicating this in each driver, add the common utility
function in common file and use from device specific driver. This will
reduce the duplicating of code across drivers.
Changes from V1:
- Add this files in this patch and add common utility APIs to here.
Changes from V2:
- Nothing in code.
- Added Reviewed by Stephen.
Sonic Zhang [Wed, 14 Aug 2013 05:26:43 +0000 (13:26 +0800)]
pinctrl: pinmux: Don't free pins requested by other devices in pinmux_disable_setting.
One peripheral may share part of its pins with the 2nd
peripheral and the other pins with the 3rd. If it requests all pins
when part of them has already be requested and owned by the 2nd
peripheral, this request fails and pinmux_disable_setting() is called.
The pinmux_disable_setting() frees all pins of the first peripheral
without checking if the pin is owned by itself or the 2nd, which
results in the malfunction of the 2nd peripheral driver.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Linus Walleij [Wed, 7 Aug 2013 20:21:57 +0000 (22:21 +0200)]
pinctrl: nomadik: delete ancient pin control API
The pin control subsystem was created to do away with custom pin
control APIs such as this one. It was kept for backward-compatibility
but is completely unused in the current kernel, so let's delete
it.
Stephen Warren [Tue, 6 Aug 2013 17:10:32 +0000 (11:10 -0600)]
pinctrl: add generic pins and functions properties
pinctrl bindings can benefit from generic property names that define
which pins a "pin configuration node" affects, and which mux function
to select onto those pins. Document new properties for this purpose so
that other bindings may refer to them.
Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Stephen Warren [Mon, 5 Aug 2013 21:55:59 +0000 (15:55 -0600)]
pinctrl: clean up pinconfig-generic documentation
Reword the section of pinctrl-bindings.txt that describes generic
properties that pinctrl bindings may use. The aim is to make the text
clearer, and more explicitly call out the responsibility of individual
bindings that use the generic properties to define which of the
properties are used, and how.
Signed-off-by: Stephen Warren <swarren@nvidia.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Boris BREZILLON [Sat, 20 Jul 2013 14:51:33 +0000 (16:51 +0200)]
pinctrl: at91: choose appropriate handler for level interrupts
The current implementation handle both edge and level interrupts with the
'handle_simple_irq' handler.
Level interrupts are active as long as the pin stays at the configured
level (low or high). In this case we have to use 'handle_level_irq' which
mask the interrupt until the handle has treated it.
Shawn Guo [Sun, 4 Aug 2013 13:39:23 +0000 (21:39 +0800)]
pinctrl: imx: work around select input quirk
The select input for some pin may not be implemented using the regular
select input register but the general purpose register. A real example
is that imx6q designers found the select input for USB OTG ID pin is
missing at the very late stage, and can not add a new select input
register but have to use a general purpose register bit to implement it.
The patch adds a workaround for such select input quirk by interpreting
the input_val cell of pin function ID in a different way, so that all
the info that needed for setting up select input bits in general purpose
register could be decoded from there.
*map should be tested for NULL instead of map as kmalloc pointer
is assigned to it. This also fixes a potential null pointer dereference
bug later in the code.
So, I notice that we get a couple of warnings from the pinctrl code:
drivers/pinctrl/pinconf.c: In function 'pinconf_dbg_config_print':
drivers/pinctrl/pinconf.c:433:36: warning: 'configs' may be used uninitialized in this function
drivers/pinctrl/pinconf.c: In function 'pinconf_dbg_config_write':
drivers/pinctrl/pinconf.c:511:36: warning: 'configs' may be used uninitialized in this function
While the compiler might not be able to work out that "configs" is
safe, the code doesn't lend itself very well to identifying that
fact when reading it either. This can be trivially solved by a slight
restructuring of the code - which also reduces the LOC.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
sh-pfc: sh7372: Replace <mach/irqs.h> with <linux/sh_intc.h>
The mach/irqs.h header is included only to get the evt2irq macro
definition. The macro is defined in linux/sh_intc.h, include it directly
instead of the mach-specific header.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Acked-by: Magnus Damm <damm@opensource.se>
sh-pfc: Support pins not associated with a GPIO port
Pins with selectable functions but without a GPIO port can't be named
PORT_# or GP_#_#. Add a SH_PFC_PIN_NAMED macro to declare such pins in
the pinmux pins array, naming them with the PIN_ prefix followed by the
pin physical position.
In order to make sure not to register those pins as GPIOs, add a
SH_PFC_PIN_CFG_NO_GPIO pin flag to denote pins without a GPIO port.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
Remove the manually specified ranges from PFC SoC data and compute the
ranges automatically. This prevents ranges from being out-of-sync with
pins definitions.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
sh-pfc: Rename struct sh_pfc nr_pins field to nr_gpio_pins
The field contains the number of pins with an associated GPIO port. This
is currently equal to the total number of pins but will be modified when
adding support for pins without a GPIO port. Rename the field
accordingly.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
Laurent Pinchart [Wed, 13 Feb 2013 23:41:57 +0000 (00:41 +0100)]
sh-pfc: Pass the pin number down to the port function macro
The PORT_1 macro invokes a macro passed as a parameter. Pass the pin
number down to that macro at the bottom of the call stack. This will be
used to compute the pin ranges automatically.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
sh-pfc: Don't duplicate argument to PINMUX_GPIO macro
The PINMUX_GPIO macro takes a port name and a data mark, respectively of
the form GPIO_name and name_DATA. Modify the macro to take the name as a
single argument and derive the port name and data mark from it.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
sh-pfc: Rename gpio arguments to be consistent with the rest of the code
The gpio_get_data_reg() and gpio_setup_data_reg() functions both take an
argument named gpio. The argument contains a GPIO offset for the first
function and a pin index for the second one. Rename them to offset and
idx respectively to match the rest of the code.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
The typedef only conceals the real variable type without bringing any
additional value (see Documentation/CodingStyle, section 5.b). Moreover,
it polutes the pinmux namespace. Replace it with the integer type it
used to hide.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
sh-pfc: Don't overallocate memory for the GPIO chip pins array
The GPIO driver uses an array of sh_pfc_gpio_pin structures to store
per-GPIO pin data. The array size is miscomputed at allocation time by
using the number of the last pin instead of the number of pins. When the
pin space contains holes this leads to memory overallocation. Fix it.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Yusuke Goda <yusuke.goda.sx@renesas.com>
The PFC SHX3 SoC data contains a input_pu range used to configure
pull-up resistors using the legacy non-pinconf API. That API has been
removed from the driver, the range is thus not used anymore. Remove it.
If required, configuring pull-up resistors for the SHX3 can be
implemented using the pinconf API, as done for the SH-Mobile, R-Mobile
and R-Car platforms.
The PFC SH7786 SoC data contains a input_pu range used to configure
pull-up resistors using the legacy non-pinconf API. That API has been
removed from the driver, the range is thus not used anymore. Remove it.
If required, configuring pull-up resistors for the SH7786 can be
implemented using the pinconf API, as done for the SH-Mobile, R-Mobile
and R-Car platforms.
The PFC SH7785 SoC data contains a input_pu range used to configure
pull-up resistors using the legacy non-pinconf API. That API has been
removed from the driver, the range is thus not used anymore. Remove it.
If required, configuring pull-up resistors for the SH7785 can be
implemented using the pinconf API, as done for the SH-Mobile, R-Mobile
and R-Car platforms.
The PFC SH7757 SoC data contains a input_pu range used to configure
pull-up resistors using the legacy non-pinconf API. That API has been
removed from the driver, the range is thus not used anymore. Remove it.
If required, configuring pull-up resistors for the SH7757 can be
implemented using the pinconf API, as done for the SH-Mobile, R-Mobile
and R-Car platforms.
The PFC SH7724 SoC data contains a input_pu range used to configure
pull-up resistors using the legacy non-pinconf API. That API has been
removed from the driver, the range is thus not used anymore. Remove it.
If required, configuring pull-up resistors for the SH7724 can be
implemented using the pinconf API, as done for the SH-Mobile, R-Mobile
and R-Car platforms.
The PFC SH7723 SoC data contains a input_pu range used to configure
pull-up resistors using the legacy non-pinconf API. That API has been
removed from the driver, the range is thus not used anymore. Remove it.
If required, configuring pull-up resistors for the SH7723 can be
implemented using the pinconf API, as done for the SH-Mobile, R-Mobile
and R-Car platforms.
sh-pfc: sh7722: Remove unused input_pd and input_pu ranges
The PFC SH7722 SoC data contains input_pd and input_pu ranges used to
configure pull-down and pull-up resistors using the legacy non-pinconf
API. That API has been removed from the driver, the ranges are thus not
used anymore. Remove them.
If required, configuring pull-down and pull-up resistors for the SH7722
can be implemented using the pinconf API, as done for the SH-Mobile,
R-Mobile and R-Car platforms.
The PFC SH7720 SoC data contains a input_pu range used to configure
pull-up resistors using the legacy non-pinconf API. That API has been
removed from the driver, the range is thus not used anymore. Remove it.
If required, configuring pull-up resistors for the SH7720 can be
implemented using the pinconf API, as done for the SH-Mobile, R-Mobile
and R-Car platforms.
pinctrl: pinctrl-imx: Remove unneeded check for platform_get_resource()
As devm_ioremap_resource() is used on probe, there is no need to explicitly
check the return value from platform_get_resource(), as this is something that
devm_ioremap_resource() takes care by itself.
Tony Lindgren [Thu, 18 Jul 2013 15:15:04 +0000 (08:15 -0700)]
pinctrl: Remove duplicate code in pinctrl_pm_select_state functions
There's no need to duplicate essentially the same functions. Let's
introduce static int pinctrl_pm_select_state() and make the other
related functions call that.
This allows us to add support later on for multiple active states,
and more optimized dynamic remuxing.
Note that we still need to export the various pinctrl_pm_select
functions as we want to keep struct pinctrl_state private to the
pinctrl code, and cannot replace those with inline functions.
Cc: Felipe Balbi <balbi@ti.com> Cc: Stephen Warren <swarren@wwwdotorg.org> Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Merge branch 'slab/for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/linux
Pull slab update from Pekka Enberg:
"Highlights:
- Fix for boot-time problems on some architectures due to
init_lock_keys() not respecting kmalloc_caches boundaries
(Christoph Lameter)
- CONFIG_SLUB_CPU_PARTIAL requested by RT folks (Joonsoo Kim)
- Fix for excessive slab freelist draining (Wanpeng Li)
- SLUB and SLOB cleanups and fixes (various people)"
I ended up editing the branch, and this avoids two commits at the end
that were immediately reverted, and I instead just applied the oneliner
fix in between myself.
* 'slab/for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/linux
slub: Check for page NULL before doing the node_match check
mm/slab: Give s_next and s_stop slab-specific names
slob: Check for NULL pointer before calling ctor()
slub: Make cpu partial slab support configurable
slab: add kmalloc() to kernel API documentation
slab: fix init_lock_keys
slob: use DIV_ROUND_UP where possible
slub: do not put a slab to cpu partial list when cpu_partial is 0
mm/slub: Use node_nr_slabs and node_nr_objs in get_slabinfo
mm/slub: Drop unnecessary nr_partials
mm/slab: Fix /proc/slabinfo unwriteable for slab
mm/slab: Sharing s_next and s_stop between slab and slub
mm/slab: Fix drain freelist excessively
slob: Rework #ifdeffery in slab.h
mm, slab: moved kmem_cache_alloc_node comment to correct place
Now, this uses SLUB pretty much unmodified, but as it is the -rt kernel
with CONFIG_PREEMPT_RT set, spinlocks are mutexes, although they do
disable migration. But the SLUB code is relatively lockless, and the
spin_locks there are raw_spin_locks (not converted to mutexes), thus I
believe this bug can happen in mainline without -rt features. The -rt
patch is just good at triggering mainline bugs ;-)
Anyway, looking at where this crashed, it seems that the page variable
can be NULL when passed to the node_match() function (which does not
check if it is NULL). When this happens we get the above panic.
As page is only used in slab_alloc() to check if the node matches, if
it's NULL I'm assuming that we can say it doesn't and call the
__slab_alloc() code. Is this a correct assumption?
Acked-by: Christoph Lameter <cl@linux.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Pekka Enberg <penberg@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull more vfs stuff from Al Viro:
"O_TMPFILE ABI changes, Oleg's fput() series, misc cleanups, including
making simple_lookup() usable for filesystems with non-NULL s_d_op,
which allows us to get rid of quite a bit of ugliness"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
sunrpc: now we can just set ->s_d_op
cgroup: we can use simple_lookup() now
efivarfs: we can use simple_lookup() now
make simple_lookup() usable for filesystems that set ->s_d_op
configfs: don't open-code d_alloc_name()
__rpc_lookup_create_exclusive: pass string instead of qstr
rpc_create_*_dir: don't bother with qstr
llist: llist_add() can use llist_add_batch()
llist: fix/simplify llist_add() and llist_add_batch()
fput: turn "list_head delayed_fput_list" into llist_head
fs/file_table.c:fput(): add comment
Safer ABI for O_TMPFILE
projects / linux-beck.git / log