Jon Hunter [Fri, 17 Jun 2016 12:03:40 +0000 (13:03 +0100)]
pinctrl: pinconf: Add generic helper function for freeing mappings
The pinconf-generic.h file exposes functions for creating generic mappings
but it does not expose a function for freeing the mappings. Add a function
for freeing generic mappings.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Ben Dooks [Tue, 21 Jun 2016 17:38:21 +0000 (18:38 +0100)]
pinctrl: rockchip: make rockchip_irq_gc_mask_set_bit static
The rockchip_irq_gc_mask_set_bit() function is not exported our used
outside of ppinctrl-rockchip.c so fix the following sparse error by
making it static:
drivers/pinctrl/pinctrl-rockchip.c:2010:6: warning:
symbol 'rockchip_irq_gc_mask_set_bit' was not declared.
Should it be static?
Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Commit 546c6d79301 (pinctrl: digicolor: make it explicitly non-modular) removed
the platform_get_drvdata() call, so platform_set_drvdata() is no longer needed.
Paul Gortmaker [Mon, 13 Jun 2016 21:10:22 +0000 (17:10 -0400)]
pinctrl: as3722: convert PINCTRL_AS3722 from bool to tristate
The Kconfig currently controlling compilation of this code is:
config PINCTRL_AS3722
bool "Pinctrl and GPIO driver for ams AS3722 PMIC"
...meaning that it currently is not being built as a module by anyone.
During an audit for non-modular drivers using modular infrastructure
this driver showed up.
But rather than demodularize it, Laxman indicated that it would be
prefereable to instead convert the driver option to tristate.
This does that, and confirms that it will compile and modpost as
such. However, since I do not have the hardware to confirm that
no new runtime issues exist when modular, that remains untested.
Paul Gortmaker [Mon, 13 Jun 2016 21:10:21 +0000 (17:10 -0400)]
pinctrl: palmas: convert PINCTRL_PALMAS from bool to tristate
The Kconfig currently controlling compilation of this code is:
config PINCTRL_PALMAS
bool "Pinctrl driver for the PALMAS Series MFD devices"
...meaning that it currently is not being built as a module by anyone.
During an audit for non-modular drivers using modular infrastructure
this driver showed up.
But rather than demodularize it, Laxman indicated that it would be
prefereable to instead convert the driver option to tristate.
This does that, and confirms that it will compile and modpost as
such. However, since I do not have the hardware to confirm that
no new runtime issues exist when modular, that remains untested.
Jon Hunter [Thu, 16 Jun 2016 15:27:41 +0000 (16:27 +0100)]
pinctrl: OF: Don't create a pinctrl handle if no pinctrl entries exist
When pinctrl_get() is called for a device, it will return a valid handle
even if the device itself has no pinctrl state entries defined in
device-tree. This is caused by the function pinctrl_dt_to_map() which
will return success even if the first pinctrl state, 'pinctrl-0', is not
found in the device-tree node for a device.
According to the pinctrl device-tree binding documentation, pinctrl
states must be numbered starting from 0 and so 'pinctrl-0' should always
be present if a device uses pinctrl and therefore, if 'pinctrl-0' is not
present it seems valid that we should not return a valid pinctrl handle.
Fix this by returning an error code if the property 'pinctrl-0' is not
present for a device.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Mika Westerberg [Thu, 16 Jun 2016 08:25:37 +0000 (11:25 +0300)]
pinctrl: intel: Prevent force threading of the interrupt handler
The pinctrl-intel needs to use request_irq() instead of chained interrupt
handling because it shares the interrupt with multiple GPIO host
controllers found on Intel CPUs. In -rt all such interrupts are forced to
run in thread context which triggers following warning:
The handle_irq_event_* functions (and I suppose generic_handle_irq()) is
expected to be called with interrupts disabled and they rightfully complain
here because we run in thread context with interrupts enabled.
Fix this by adding IRQF_NO_THREAD flag when the master interrupt is
requested. This prevents forced threading of the interrupt used by the GPIO
host controllers.
Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
The reason why this happens is because intel_gpio_irq_ack() is called with
desc->lock raw_spinlock locked which cannot sleep but our normal spinlock
(which is converted to rtmutex in -rt) is allowed to sleep. This causes
might_sleep() to trigger.
Fix this by converting the normal spinlock to a raw_spinlock.
Reported-by: Kim Tatt Chuah <kim.tatt.chuah@intel.com> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Masahiro Yamada [Wed, 15 Jun 2016 08:18:05 +0000 (17:18 +0900)]
pinctrl: uniphier: prohibit drive control for pin 61-66 of PH1-LD11
According to the hardware document, setting the drive control is
prohibited for these pins (N-channel Open Drain pins). Set their
drive control attribute to "fixed".
Tan Jui Nee [Tue, 7 Jun 2016 06:55:51 +0000 (14:55 +0800)]
pinctrl/broxton: enable platform device in the absence of ACPI enumeration
This is to cater the need for non-ACPI system whereby
a platform device has to be created in order to bind
with the Apollo Lake Pinctrl GPIO platform driver.
Signed-off-by: Tan Jui Nee <jui.nee.tan@intel.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Arnd Bergmann [Mon, 13 Jun 2016 15:18:35 +0000 (17:18 +0200)]
pinctrl: max77620: select PINMUX
The recently added max77620 driver fails to build when CONFIG_PINMUX
is not set:
pinctrl/pinctrl-max77620.c:272:21: error: variable 'max77620_pinmux_ops' has initializer but incomplete type
static const struct pinmux_ops max77620_pinmux_ops = {
^~~~~~~~~~
pinctrl/pinctrl-max77620.c:273:2: error: unknown field 'get_functions_count' specified in initializer
This adds the Kconfig 'select' statement that was clearly meant
to be there and is used in all other pinmux drivers.
Dan O'Donovan [Fri, 10 Jun 2016 12:23:36 +0000 (13:23 +0100)]
pinctrl: cherryview: add handlers for pin_config_group_get/set
Pin config get/set handlers for pin groups were previously not
implemented by this driver. The pin_config_group_set is
particularly useful for applying a common config setting to all
pins in a specified group with a single call, without the caller
needing to reference each individual pin by name.
Signed-off-by: Dan O'Donovan <dan@emutex.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Dan O'Donovan [Fri, 10 Jun 2016 12:23:35 +0000 (13:23 +0100)]
pinctrl: cherryview: add option to set open-drain pin config
On some CHV platforms, we need an option to configure the
open-drain setting for these pins. This adds support for the
PIN_CONFIG_DRIVE_PUSH_PULL and PIN_CONFIG_DRIVE_OPEN_DRAIN to
disable/enable open-drain mode for a specific pin.
Signed-off-by: Dan O'Donovan <dan@emutex.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Dan O'Donovan [Fri, 10 Jun 2016 12:23:34 +0000 (13:23 +0100)]
pinctrl: cherryview: prevent concurrent access to GPIO controllers
Due to a silicon issue on the Atom X5-Z8000 "Cherry Trail" processor
series, a common lock must be used to prevent concurrent accesses
across the 4 GPIO controllers managed by this driver.
See Intel Atom Z8000 Processor Series Specification Update
(Rev. 005), errata #CHT34, for further information.
Cc: stable <stable@vger.kernel.org> Signed-off-by: Dan O'Donovan <dan@emutex.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Linus Walleij [Fri, 3 Jun 2016 11:55:00 +0000 (13:55 +0200)]
pinctrl: qcom-ssbi: support for PM8058
The PM8058 is found in connection to the APQ8060 on the APQ8060
Dragonboard. Works the same as all others, just add the compatible
string for this variant.
Cc: Bjorn Andersson <bjorn.andersson@linaro.org> Cc: Andy Gross <andy.gross@linaro.org> Reviewed-by: Stephen Boyd <sboyd@codeaurora.org> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
drivers/pinctrl/bcm/Makefile properly builds individual drivers based on
their respective Kconfig symbols. ARCH_BCM is currently a menuconfig
option from arch/arm/mach-bcm/Kconfig, which is fine, but prevents ARM64
platforms which do not have such menuconfig option from building their
pinctrl drivers, so let's get rid of that dependency.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Acked-by: Scott Branden <scott.branden@broadcom.com> Acked-by: Eric Anholt <eric@anholt.net> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Ben Dooks [Wed, 8 Jun 2016 16:07:42 +0000 (17:07 +0100)]
pinctrl: nomadik: fix warnings from unexported functions
There are five functions in the driver that are defined but
only used locally. Since these are not used in the current
kernel, delete them to avoid the following warnings:
drivers/pinctrl/nomadik/pinctrl-nomadik.c:1036:6: warning: symbol 'nmk_gpio_clocks_enable' was not declared. Should it be static?
drivers/pinctrl/nomadik/pinctrl-nomadik.c:1050:6: warning: symbol 'nmk_gpio_clocks_disable' was not declared. Should it be static?
drivers/pinctrl/nomadik/pinctrl-nomadik.c:1073:6: warning: symbol 'nmk_gpio_wakeups_suspend' was not declared. Should it be static?
drivers/pinctrl/nomadik/pinctrl-nomadik.c:1094:6: warning: symbol 'nmk_gpio_wakeups_resume' was not declared. Should it be static?
drivers/pinctrl/nomadik/pinctrl-nomadik.c:1120:6: warning: symbol 'nmk_gpio_read_pull' was not declared. Should it be static?
Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
...meaning that it currently is not being built as a module by anyone.
Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.
We explicitly disallow a driver unbind, since that doesn't have a
sensible use case anyway, and it allows us to drop the ".remove"
code for non-modular drivers.
Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.
Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code.
We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.
...meaning that it currently is not being built as a module by anyone.
Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.
We explicitly disallow a driver unbind, since that doesn't have a
sensible use case anyway, and it allows us to drop the ".remove"
code for non-modular drivers.
Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.
Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code.
Paul Gortmaker [Tue, 7 Jun 2016 02:43:04 +0000 (22:43 -0400)]
pinctrl: amd: make it explicitly non-modular
The Kconfig currently controlling compilation of this code is:
config PINCTRL_AMD
bool "AMD GPIO pin control"
...meaning that it currently is not being built as a module by anyone.
Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.
We explicitly disallow a driver unbind, since that doesn't have a
sensible use case anyway, and it allows us to drop the ".remove"
code for non-modular drivers.
Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.
Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code.
We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.
Cc: Linus Walleij <linus.walleij@linaro.org> Cc: linux-gpio@vger.kernel.org Cc: Ken Xue <Ken.Xue@amd.com> Cc: Jeff Wu <Jeff.Wu@amd.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
...meaning that it currently is not being built as a module by anyone.
Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.
We explicitly disallow a driver unbind, since that doesn't have a
sensible use case anyway, and it allows us to drop the ".remove"
code for non-modular drivers.
Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.
Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code.
We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.
Cc: Linus Walleij <linus.walleij@linaro.org> Cc: Joachim Eastwood <manabian@gmail.com> Cc: linux-gpio@vger.kernel.org Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
...meaning that it currently is not being built as a module by anyone.
Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.
We explicitly disallow a driver unbind, since that doesn't have a
sensible use case anyway, and it allows us to drop the ".remove"
code for non-modular drivers.
Since module_init() was already not in use in this driver, we don't
have any concerns with init ordering changes here.
Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code.
We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.
Cc: Heikki Krogerus <heikki.krogerus@linux.intel.com> Cc: linux-gpio@vger.kernel.org Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Ben Dooks [Tue, 7 Jun 2016 17:07:45 +0000 (18:07 +0100)]
pinctrl: at91-pio4: fix non-exported functions
The atmel_pctl_find_group_by_pin() and the atmel_pinctrl_remove()
functions are not exported, so fix the warnings about these
being exported without definitions by making them static.
drivers/pinctrl/pinctrl-at91-pio4.c:424:20: warning: symbol 'atmel_pctl_find_group_by_pin' was not declared. Should it be static?
drivers/pinctrl/pinctrl-at91-pio4.c:1077:5: warning: symbol 'atmel_pinctrl_remove' was not declared. Should it be static?
Ben Dooks [Tue, 7 Jun 2016 16:58:15 +0000 (17:58 +0100)]
pinctrl: nsp-gpio: fix non-static functions
Fixup warnings from functions that are not exported and
therefore should be marked static. Fixes:
drivers/pinctrl/bcm/pinctrl-nsp-gpio.c:461:5:
warning: symbol 'nsp_pin_config_group_get' was not declared.
Should it be static?
drivers/pinctrl/bcm/pinctrl-nsp-gpio.c:467:5:
warning: symbol 'nsp_pin_config_group_set' was not declared.
Should it be static?
Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk> Acked-by: Ray Jui <ray.jui@broadcom.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
An irq which is a wake up source maybe masked unexpectedly if the wake
up source irq was triggered after pinctrl irqchip suspend and before
suspend_device_irqs finished.
Use *_noirq callbacks to guarantee pinctrl irqchip suspend would be
called after suspend_devices_irqs.
Neil Armstrong [Tue, 31 May 2016 09:36:15 +0000 (11:36 +0200)]
pinctrl: oxnas: Rename pinctrl_utils_dt_free_map to pinctrl_utils_free_map
Rename pinctrl_utils_dt_free_map to pinctrl_utils_free_map, introduced in d32f7fd3bbc3 ("pinctrl: Rename pinctrl_utils_dt_free_map to pinctrl_utils_free_map")
but not reported into oxnas driver.
Fixes: 611dac1e48a4 ("pinctrl: Add Oxford Semiconductor OXNAS pinctrl and gpio driver") Signed-off-by: Neil Armstrong <narmstrong@baylibre.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Masahiro Yamada [Tue, 31 May 2016 08:05:21 +0000 (17:05 +0900)]
pinctrl: uniphier: allow to have pinctrl node under syscon node
Currently, the UniPhier pinctrl driver itself is a syscon, but it
turned out much more reasonable to make it a child node of a syscon
because our syscon node consists of a bunch of system configuration
registers, not only pinctrl, but also phy, and misc registers.
It is difficult to split the node.
To allow to migrate to the new DT structure, this commit adds new
compatible strings to not disturb the existing DT. After a while,
the old binding will be removed.
Masahiro Yamada [Tue, 31 May 2016 08:05:18 +0000 (17:05 +0900)]
pinctrl: uniphier: support pin configuration for dedicated pins
PH1-LD4 and PH1-sLD8 SoCs have pins that support pin configuration
(pin biasing, drive strength control), but not pin-muxing.
Allow to fill the mux value table with -1 for those pins; pins with
mux value -1 will be skipped in the pin-mux set function. The mux
value type should be changed from "unsigned" to "int" in order to
accommodate -1 as a special case.
Masahiro Yamada [Tue, 31 May 2016 08:05:17 +0000 (17:05 +0900)]
pinctrl: uniphier: support per-pin input enable for new SoCs
Upcoming new pinctrl drivers for PH1-LD11 and PH-LD20 support input
signal gating for each pin. (While, existing ones only support it
per pin-group.) This commit updates the core part for that.
Masahiro Yamada [Tue, 31 May 2016 08:05:16 +0000 (17:05 +0900)]
pinctrl: uniphier: introduce capability flag
The core part of the UniPhier pinctrl driver needs to support a new
capability for upcoming UniPhier ARMv8 SoCs. This sometimes happens
because pinctrl drivers include really SoC-specific stuff.
This commit intends to tidy up SoC-specific parameters of the existing
drivers before adding the new one. Having just one flag would be
better than adding a new struct member every time a new SoC-specific
capability comes up.
At this time, there is one flag, UNIPHIER_PINCTRL_CAPS_DBGMUX_SEPARATE.
This capability (I'd say rather quirk) was added for PH1-Pro4 and
PH1-Pro5 as requirement from a customer. For those SoCs, one pin-mux
setting is controlled by the combination of two separate registers; the
LSB bits at register offset (8 * N) and the MSB bits at (8 * N + 4).
Because it is impossible to update two separate registers atomically,
the LOAD_PINCTRL register should be set in order to make the pin-mux
settings really effective.
Masahiro Yamada [Tue, 31 May 2016 08:05:15 +0000 (17:05 +0900)]
pinctrl: uniphier: support pin configuration in sparse pin space
Unfortunately, the pin number of the new SoC, PH1-LD11, is not
contiguous. The base frame work must be adjusted to support the new
SoC pinctrl driver. The pin_desc_get() exploits radix-tree for pin
look-up, so it works more efficiently with sparse pin space.
Masahiro Yamada [Tue, 31 May 2016 08:05:14 +0000 (17:05 +0900)]
pinctrl: uniphier: support 3-bit drive strength control
The new ARMv8 SoC, PH1-LD20, supports more fine-grained drive
strength control. Drive strength of some pins are controlled by
3-bit width registers (8-level granularity).
Masahiro Yamada [Tue, 31 May 2016 08:05:12 +0000 (17:05 +0900)]
pinctrl: uniphier: allocate struct pinctrl_desc in probe function
Currently, every SoC driver defines struct pinctrl_desc statically,
i.e. it consumes memory footprint even if it is not probed.
In multi-platform, many pinctrl drivers are linked (generally as
built-in objects), although only one of them is actually used.
So, it is reasonable to allocate memory dynamically where possible.
Masahiro Yamada [Tue, 31 May 2016 08:05:11 +0000 (17:05 +0900)]
pinctrl: uniphier: set pinctrl_desc name in common probe function
Every SoC driver sets the same name for struct pinctrl_desc and
platform_driver. The common probe function can set desc->name
instead of duplicating strings in each SoC driver.
Peng Fan [Wed, 18 May 2016 09:31:59 +0000 (17:31 +0800)]
pinctrl: imx: fix initialization of imx_pinctrl_desc
To i.MX7D, there are two iomux controllers, iomuxc and iomuxc_lpsr.
They should not share one pin controller descriptor, otherwise
the value filled into imx_pinctrl_desc when probing the first
iomux controller will be overridden when probing the second one.
In this patch, discard the static allcoated imx_pinctrl_desc and
switch to dynamically allcate pin controller descriptor for each
iomux controller.
Signed-off-by: Peng Fan <van.freenix@gmail.com> Cc: Shawn Guo <shawnguo@kernel.org> Cc: Philipp Zabel <p.zabel@pengutronix.de> Cc: Stefan Agner <stefan@agner.ch> Cc: Vladimir Zapolskiy <vz@mleia.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Masahiro Yamada [Wed, 25 May 2016 05:09:31 +0000 (14:09 +0900)]
pinctrl: copy per-pin driver private data to struct pin_desc
Currently, struct pinctrl_pin_desc can have per-pin driver private
data, but it is not copied to struct pin_desc.
For a driver with sparse pin space, for-loop search like below would
be necessary in order to get the driver-specific data for a desired
pin number.
for (i = 0; i < pctldev->desc->npins; i++)
if (pin_number == pctldev->desc->pins[i].number)
return pctldev->desc->pins[i].drv_data;
This is not efficient for a driver with a large number of pins.
So, copy the data to struct pin_desc when each pin is registered
for the faster radix tree lookup.
Masahiro Yamada [Tue, 24 May 2016 05:26:26 +0000 (14:26 +0900)]
pinctrl: do not care about blank pin name
If a pin name is not specified in struct pinctrl_pin_desc,
pinctrl_register_one_pin() dynamically assigns its name.
So, desc->name is always a valid pointer here.
pinctrl: samsung: Suppress unbinding to prevent theoretical attacks
Although unbinding a pinctrl driver requires root privileges but it
still might be used theoretically in certain attacks (by triggering NULL
pointer exception or memory corruption).
Samsung pincontrol drivers are essential for system operation so their
removal is not expected. They do not implement remove() driver callback
and they are not buildable as modules.
Suppression of the unbinding will prevent triggering NULL pointer
exception like this (Odroid XU3):
Laxman Dewangan [Fri, 13 May 2016 05:19:15 +0000 (10:49 +0530)]
pinctrl: max77620: add pincontrol driver for MAX77620/MAX20024
MAXIM Semiconductor's PMIC, MAX77620/MAX20024 has 8 GPIO pins
which also act as the special function in alternate mode. Also
there is configuration like push-pull, open drain, FPS timing
etc for these pins.
Add pin control driver to configure these parameters through
pin control APIs.
George Spelvin [Sun, 29 May 2016 05:26:41 +0000 (01:26 -0400)]
Rename other copy of hash_string to hashlen_string
The original name was simply hash_string(), but that conflicted with a
function with that name in drivers/base/power/trace.c, and I decided
that calling it "hashlen_" was better anyway.
But you have to do it in two places.
[ This caused build errors for architectures that don't define
CONFIG_DCACHE_WORD_ACCESS - Linus ]
Mikulas Patocka [Tue, 24 May 2016 20:49:18 +0000 (22:49 +0200)]
hpfs: implement the show_options method
The HPFS filesystem used generic_show_options to produce string that is
displayed in /proc/mounts. However, there is a problem that the options
may disappear after remount. If we mount the filesystem with option1
and then remount it with option2, /proc/mounts should show both option1
and option2, however it only shows option2 because the whole option
string is replaced with replace_mount_options in hpfs_remount_fs.
To fix this bug, implement the hpfs_show_options function that prints
options that are currently selected.
Mikulas Patocka [Tue, 24 May 2016 20:48:33 +0000 (22:48 +0200)]
affs: fix remount failure when there are no options changed
Commit c8f33d0bec99 ("affs: kstrdup() memory handling") checks if the
kstrdup function returns NULL due to out-of-memory condition.
However, if we are remounting a filesystem with no change to
filesystem-specific options, the parameter data is NULL. In this case,
kstrdup returns NULL (because it was passed NULL parameter), although no
out of memory condition exists. The mount syscall then fails with
ENOMEM.
This patch fixes the bug. We fail with ENOMEM only if data is non-NULL.
The patch also changes the call to replace_mount_options - if we didn't
pass any filesystem-specific options, we don't call
replace_mount_options (thus we don't erase existing reported options).
Mikulas Patocka [Tue, 24 May 2016 20:47:00 +0000 (22:47 +0200)]
hpfs: fix remount failure when there are no options changed
Commit ce657611baf9 ("hpfs: kstrdup() out of memory handling") checks if
the kstrdup function returns NULL due to out-of-memory condition.
However, if we are remounting a filesystem with no change to
filesystem-specific options, the parameter data is NULL. In this case,
kstrdup returns NULL (because it was passed NULL parameter), although no
out of memory condition exists. The mount syscall then fails with
ENOMEM.
This patch fixes the bug. We fail with ENOMEM only if data is non-NULL.
The patch also changes the call to replace_mount_options - if we didn't
pass any filesystem-specific options, we don't call
replace_mount_options (thus we don't erase existing reported options).
Fixes: ce657611baf9 ("hpfs: kstrdup() out of memory handling") Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
VDSO:
- Build microMIPS VDSO for microMIPS kernels.
- Fix aliasing warning by building with `-fno-strict-aliasing' for
debugging but also tracing them might result in recursion.
Misc:
- Add missing FROZEN hotplug notifier transitions.
- Fix clk binding example for varioius PIC32 devices.
- Fix cpu interrupt controller node-names in the DT files.
- Fix XPA CPU feature separation.
- Fix write_gc0_* macros when writing zero.
- Add inline asm encoding helpers.
- Add missing VZ accessor microMIPS encodings.
- Fix little endian microMIPS MSA encodings.
- Add 64-bit HTW fields and fix its configuration.
- Fix sigreturn via VDSO on microMIPS kernel.
- Lots of typo fixes.
- Add definitions of SegCtl registers and use them"
Linus Torvalds [Sat, 28 May 2016 23:15:25 +0000 (16:15 -0700)]
Merge branch 'hash' of git://ftp.sciencehorizons.net/linux
Pull string hash improvements from George Spelvin:
"This series does several related things:
- Makes the dcache hash (fs/namei.c) useful for general kernel use.
(Thanks to Bruce for noticing the zero-length corner case)
- Converts the string hashes in <linux/sunrpc/svcauth.h> to use the
above.
- Avoids 64-bit multiplies in hash_64() on 32-bit platforms. Two
32-bit multiplies will do well enough.
- Rids the world of the bad hash multipliers in hash_32.
This finishes the job started in commit 689de1d6ca95 ("Minimal
fix-up of bad hashing behavior of hash_64()")
The vast majority of Linux architectures have hardware support for
32x32-bit multiply and so derive no benefit from "simplified"
multipliers.
The few processors that do not (68000, h8/300 and some models of
Microblaze) have arch-specific implementations added. Those
patches are last in the series.
- Overhauls the dcache hash mixing.
The patch in commit 0fed3ac866ea ("namei: Improve hash mixing if
CONFIG_DCACHE_WORD_ACCESS") was an off-the-cuff suggestion.
Replaced with a much more careful design that's simultaneously
faster and better. (My own invention, as there was noting suitable
in the literature I could find. Comments welcome!)
- Modify the hash_name() loop to skip the initial HASH_MIX(). This
would let us salt the hash if we ever wanted to.
- Sort out partial_name_hash().
The hash function is declared as using a long state, even though
it's truncated to 32 bits at the end and the extra internal state
contributes nothing to the result. And some callers do odd things:
- fs/hfs/string.c only allocates 32 bits of state
- fs/hfsplus/unicode.c uses it to hash 16-bit unicode symbols not bytes
- Modify bytemask_from_count to handle inputs of 1..sizeof(long)
rather than 0..sizeof(long)-1. This would simplify users other
than full_name_hash"
Special thanks to Bruce Fields for testing and finding bugs in v1. (I
learned some humbling lessons about "obviously correct" code.)
On the arch-specific front, the m68k assembly has been tested in a
standalone test harness, I've been in contact with the Microblaze
maintainers who mostly don't care, as the hardware multiplier is never
omitted in real-world applications, and I haven't heard anything from
the H8/300 world"
* 'hash' of git://ftp.sciencehorizons.net/linux:
h8300: Add <asm/hash.h>
microblaze: Add <asm/hash.h>
m68k: Add <asm/hash.h>
<linux/hash.h>: Add support for architecture-specific functions
fs/namei.c: Improve dcache hash function
Eliminate bad hash multipliers from hash_32() and hash_64()
Change hash_64() return value to 32 bits
<linux/sunrpc/svcauth.h>: Define hash_str() in terms of hashlen_string()
fs/namei.c: Add hashlen_string() function
Pull out string hash to <linux/stringhash.h>
George Spelvin [Wed, 25 May 2016 18:19:49 +0000 (14:19 -0400)]
h8300: Add <asm/hash.h>
This will improve the performance of hash_32() and hash_64(), but due
to complete lack of multi-bit shift instructions on H8, performance will
still be bad in surrounding code.
Designing H8-specific hash algorithms to work around that is a separate
project. (But if the maintainers would like to get in touch...)
Signed-off-by: George Spelvin <linux@sciencehorizons.net> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: uclinux-h8-devel@lists.sourceforge.jp
George Spelvin [Wed, 25 May 2016 15:06:09 +0000 (11:06 -0400)]
microblaze: Add <asm/hash.h>
Microblaze is an FPGA soft core that can be configured various ways.
If it is configured without a multiplier, the standard __hash_32()
will require a call to __mulsi3, which is a slow software loop.
Instead, use a shift-and-add sequence for the constant multiply.
GCC knows how to do this, but it's not as clever as some.
Signed-off-by: George Spelvin <linux@sciencehorizons.net> Cc: Alistair Francis <alistair.francis@xilinx.com> Cc: Michal Simek <michal.simek@xilinx.com>
George Spelvin [Fri, 27 May 2016 02:11:51 +0000 (22:11 -0400)]
<linux/hash.h>: Add support for architecture-specific functions
This is just the infrastructure; there are no users yet.
This is modelled on CONFIG_ARCH_RANDOM; a CONFIG_ symbol declares
the existence of <asm/hash.h>.
That file may define its own versions of various functions, and define
HAVE_* symbols (no CONFIG_ prefix!) to suppress the generic ones.
Included is a self-test (in lib/test_hash.c) that verifies the basics.
It is NOT in general required that the arch-specific functions compute
the same thing as the generic, but if a HAVE_* symbol is defined with
the value 1, then equality is tested.
Signed-off-by: George Spelvin <linux@sciencehorizons.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greg Ungerer <gerg@linux-m68k.org> Cc: Andreas Schwab <schwab@linux-m68k.org> Cc: Philippe De Muyter <phdm@macq.eu> Cc: linux-m68k@lists.linux-m68k.org Cc: Alistair Francis <alistai@xilinx.com> Cc: Michal Simek <michal.simek@xilinx.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: uclinux-h8-devel@lists.sourceforge.jp
George Spelvin [Mon, 23 May 2016 11:43:58 +0000 (07:43 -0400)]
fs/namei.c: Improve dcache hash function
Patch 0fed3ac866 improved the hash mixing, but the function is slower
than necessary; there's a 7-instruction dependency chain (10 on x86)
each loop iteration.
Word-at-a-time access is a very tight loop (which is good, because
link_path_walk() is one of the hottest code paths in the entire kernel),
and the hash mixing function must not have a longer latency to avoid
slowing it down.
There do not appear to be any published fast hash functions that:
1) Operate on the input a word at a time, and
2) Don't need to know the length of the input beforehand, and
3) Have a single iterated mixing function, not needing conditional
branches or unrolling to distinguish different loop iterations.
One of the algorithms which comes closest is Yann Collet's xxHash, but
that's two dependent multiplies per word, which is too much.
The key insights in this design are:
1) Barring expensive ops like multiplies, to diffuse one input bit
across 64 bits of hash state takes at least log2(64) = 6 sequentially
dependent instructions. That is more cycles than we'd like.
2) An operation like "hash ^= hash << 13" requires a second temporary
register anyway, and on a 2-operand machine like x86, it's three
instructions.
3) A better use of a second register is to hold a two-word hash state.
With careful design, no temporaries are needed at all, so it doesn't
increase register pressure. And this gets rid of register copying
on 2-operand machines, so the code is smaller and faster.
4) Using two words of state weakens the requirement for one-round mixing;
we now have two rounds of mixing before cancellation is possible.
5) A two-word hash state also allows operations on both halves to be
done in parallel, so on a superscalar processor we get more mixing
in fewer cycles.
I ended up using a mixing function inspired by the ChaCha and Speck
round functions. It is 6 simple instructions and 3 cycles per iteration
(assuming multiply by 9 can be done by an "lea" instruction):
x ^= *input++;
y ^= x; x = ROL(x, K1);
x += y; y = ROL(y, K2);
y *= 9;
Not only is this reversible, two consecutive rounds are reversible:
if you are given the initial and final states, but not the intermediate
state, it is possible to compute both input words. This means that at
least 3 words of input are required to create a collision.
(It also has the property, used by hash_name() to avoid a branch, that
it hashes all-zero to all-zero.)
The rotate constants K1 and K2 were found by experiment. The search took
a sample of random initial states (I used 1023) and considered the effect
of flipping each of the 64 input bits on each of the 128 output bits two
rounds later. Each of the 8192 pairs can be considered a biased coin, and
adding up the Shannon entropy of all of them produces a score.
The best-scoring shifts also did well in other tests (flipping bits in y,
trying 3 or 4 rounds of mixing, flipping all 64*63/2 pairs of input bits),
so the choice was made with the additional constraint that the sum of the
shifts is odd and not too close to the word size.
The final state is then folded into a 32-bit hash value by a less carefully
optimized multiply-based scheme. This also has to be fast, as pathname
components tend to be short (the most common case is one iteration!), but
there's some room for latency, as there is a fair bit of intervening logic
before the hash value is used for anything.
(Performance verified with "bonnie++ -s 0 -n 1536:-2" on tmpfs. I need
a better benchmark; the numbers seem to show a slight dip in performance
between 4.6.0 and this patch, but they're too noisy to quote.)
Special thanks to Bruce fields for diligent testing which uncovered a
nasty fencepost error in an earlier version of this patch.
[checkpatch.pl formatting complaints noted and respectfully disagreed with.]
Signed-off-by: George Spelvin <linux@sciencehorizons.net> Tested-by: J. Bruce Fields <bfields@redhat.com>
George Spelvin [Fri, 27 May 2016 03:00:23 +0000 (23:00 -0400)]
Eliminate bad hash multipliers from hash_32() and hash_64()
The "simplified" prime multipliers made very bad hash functions, so get rid
of them. This completes the work of 689de1d6ca.
To avoid the inefficiency which was the motivation for the "simplified"
multipliers, hash_64() on 32-bit systems is changed to use a different
algorithm. It makes two calls to hash_32() instead.
drivers/media/usb/dvb-usb-v2/af9015.c uses the old GOLDEN_RATIO_PRIME_32
for some horrible reason, so it inherits a copy of the old definition.
Signed-off-by: George Spelvin <linux@sciencehorizons.net> Cc: Antti Palosaari <crope@iki.fi> Cc: Mauro Carvalho Chehab <m.chehab@samsung.com>
George Spelvin [Fri, 27 May 2016 02:22:01 +0000 (22:22 -0400)]
Change hash_64() return value to 32 bits
That's all that's ever asked for, and it makes the return
type of hash_long() consistent.
It also allows (upcoming patch) an optimized implementation
of hash_64 on 32-bit machines.
I tried adding a BUILD_BUG_ON to ensure the number of bits requested
was never more than 32 (most callers use a compile-time constant), but
adding <linux/bug.h> to <linux/hash.h> breaks the tools/perf compiler
unless tools/perf/MANIFEST is updated, and understanding that code base
well enough to update it is too much trouble. I did the rest of an
allyesconfig build with such a check, and nothing tripped.
Signed-off-by: George Spelvin <linux@sciencehorizons.net>
George Spelvin [Fri, 20 May 2016 17:31:33 +0000 (13:31 -0400)]
<linux/sunrpc/svcauth.h>: Define hash_str() in terms of hashlen_string()
Finally, the first use of previous two patches: eliminate the
separate ad-hoc string hash functions in the sunrpc code.
Now hash_str() is a wrapper around hash_string(), and hash_mem() is
likewise a wrapper around full_name_hash().
Note that sunrpc code *does* call hash_mem() with a zero length, which
is why the previous patch needed to handle that in full_name_hash().
(Thanks, Bruce, for finding that!)
This also eliminates the only caller of hash_long which asks for
more than 32 bits of output.
The comment about the quality of hashlen_string() and full_name_hash()
is jumping the gun by a few patches; they aren't very impressive now,
but will be improved greatly later in the series.
Signed-off-by: George Spelvin <linux@sciencehorizons.net> Tested-by: J. Bruce Fields <bfields@redhat.com> Acked-by: J. Bruce Fields <bfields@redhat.com> Cc: Jeff Layton <jlayton@poochiereds.net> Cc: linux-nfs@vger.kernel.org
George Spelvin [Fri, 20 May 2016 12:41:37 +0000 (08:41 -0400)]
fs/namei.c: Add hashlen_string() function
We'd like to make more use of the highly-optimized dcache hash functions
throughout the kernel, rather than have every subsystem create its own,
and a function that hashes basic null-terminated strings is required
for that.
(The name is to emphasize that it returns both hash and length.)
It's actually useful in the dcache itself, specifically d_alloc_name().
Other uses in the next patch.
full_name_hash() is also tweaked to make it more generally useful:
1) Take a "char *" rather than "unsigned char *" argument, to
be consistent with hash_name().
2) Handle zero-length inputs. If we want more callers, we don't want
to make them worry about corner cases.
Signed-off-by: George Spelvin <linux@sciencehorizons.net>
Linus Torvalds [Sat, 28 May 2016 19:38:50 +0000 (12:38 -0700)]
Merge branch 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fix from Wolfram Sang:
"A fix for a regression introduced yesterday.
The regression didn't show up here locally because I did not have
PAGE_POISONING enabled. And buildbots discovered this only after it
hit your tree. Thanks to Dan for the quick response"
* 'i2c/for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: dev: use after free in detach
Linus Torvalds [Sat, 28 May 2016 19:32:01 +0000 (12:32 -0700)]
Merge tag 'chrome-platform' of git://git.kernel.org/pub/scm/linux/kernel/git/olof/chrome-platform
Pull chrome platform updates from Olof Johansson
"A handful of Chrome driver and binding changes this merge window:
- a few patches to fix probing and configuration of pstore
- a few patches adding Elan touchpad registration on a few devices
- EC changes: a security fix dealing with max message sizes and
addition of compat_ioctl support.
- keyboard backlight control support
There was also an accidential duplicate registration of trackpads on
'Leon', which was reverted just recently"
* tag 'chrome-platform' of git://git.kernel.org/pub/scm/linux/kernel/git/olof/chrome-platform:
Revert "platform/chrome: chromeos_laptop: Add Leon Touch"
platform/chrome: chromeos_laptop - Add Elan touchpad for Wolf
platform/chrome: chromeos_laptop - Add elan trackpad option for C720
platform/chrome: cros_ec_dev - Populate compat_ioctl
platform/chrome: cros_ec_lightbar - use name instead of ID to hide lightbar attributes
platform/chrome: cros_ec_dev - Fix security issue
platform/chrome: Add Chrome OS keyboard backlight LEDs support
platform/chrome: use to_platform_device()
platform/chrome: pstore: Move to larger record size.
platform/chrome: pstore: probe for ramoops buffer using acpi
platform/chrome: chromeos_laptop: Add Leon Touch
Linus Torvalds [Sat, 28 May 2016 19:23:12 +0000 (12:23 -0700)]
Merge tag 'sound-4.7-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull more sound updates from Takashi Iwai:
"This is the second update round for 4.7-rc1. Most of changes are
about the pending ASoC updates and fixes, including a few new drivers.
Below are some highlights:
ASoC:
- New drivers for MAX98371 and TAS5720
- SPI support for TLV320AIC32x4, along with the module split
- TDM support for STI Uniperf IPs
- Remaining topology API fixes / updates
HDA:
- A couple of Dell quirks and new Realtek codec support"
* tag 'sound-4.7-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (63 commits)
ALSA: hda - Fix headset mic detection problem for one Dell machine
spi: spi-ep93xx: Fix the PTR_ERR() argument
ALSA: hda/realtek - Add support for ALC295/ALC3254
ASoC: kirkwood: fix build failure
ALSA: hda - Fix headphone noise on Dell XPS 13 9360
ASoC: ak4642: Enable cache usage to fix crashes on resume
ASoC: twl6040: Disconnect AUX output pads on digital mute
ASoC: tlv320aic32x4: Properly implement the positive and negative pins into the mixers
rcar: src: skip disabled-SRC nodes
ASoC: max98371 Remove duplicate entry in max98371_reg
ASoC: twl6040: Select LPPLL during standby
ASoC: rsnd: don't use prohibited number to PDMACHCRn.SRS
ASoC: simple-card: Add pm callbacks to platform driver
ASoC: pxa: Fix module autoload for platform drivers
ASoC: topology: Fix memory leak in widget creation
ASoC: Add max98371 codec driver
ASoC: rsnd: count .probe/.remove for rsnd_mod_call()
ASoC: topology: Check size mismatch of ABI objects before parsing
ASoC: topology: Check failure to create a widget
ASoC: add support for TAS5720 digital amplifier
...
projects / linux-beck.git / log