platform/kernel/linux-rpi.git
7 years agoMerge branch 'mcp23s08' into devel
Linus Walleij [Tue, 23 May 2017 07:52:08 +0000 (09:52 +0200)]
Merge branch 'mcp23s08' into devel

7 years agopinctrl: mcp23s08: fix comment for mcp23s08_platform_data.base
Sebastian Reichel [Mon, 15 May 2017 09:24:38 +0000 (11:24 +0200)]
pinctrl: mcp23s08: fix comment for mcp23s08_platform_data.base

The comment does not match the driver, which actually supports
automatic assignment. Fix this by updating the comment.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: drop comment about missing irq support
Sebastian Reichel [Mon, 15 May 2017 09:24:37 +0000 (11:24 +0200)]
pinctrl: mcp23s08: drop comment about missing irq support

The driver supports using mcp23xxx as interrupt controller, so
let's drop all comments stating otherwise.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: simplify spi_present_mask handling
Sebastian Reichel [Mon, 15 May 2017 09:24:36 +0000 (11:24 +0200)]
pinctrl: mcp23s08: simplify spi_present_mask handling

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: generalize irq property handling
Sebastian Reichel [Mon, 15 May 2017 09:24:35 +0000 (11:24 +0200)]
pinctrl: mcp23s08: generalize irq property handling

This moves irq property handling from spi/i2c specific code into
the generic mcp23s08_probe_one. This is possible because the
device properties are named equally.

As a side-effect this drops support for setting the properties via
pdata, which has no mainline users. If boardcode wants to enable
the chip as interrupt controller it can attach the device properties
instead.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: simplify spi pdata handling
Sebastian Reichel [Mon, 15 May 2017 09:24:34 +0000 (11:24 +0200)]
pinctrl: mcp23s08: simplify spi pdata handling

Simplify spi pdata handling, so that it uses pdata when available
and falls back to reading device properties otherwise.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: simplify i2c pdata handling
Sebastian Reichel [Mon, 15 May 2017 09:24:33 +0000 (11:24 +0200)]
pinctrl: mcp23s08: simplify i2c pdata handling

Simplify i2c pdata handling, so that it uses pdata when available
and falls back to reading device properties otherwise.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: switch to devm_gpiochip_add_data
Sebastian Reichel [Mon, 15 May 2017 09:24:32 +0000 (11:24 +0200)]
pinctrl: mcp23s08: switch to devm_gpiochip_add_data

Switching to devm_gpiochip_add_data simplifies the driver's
cleanup routine and safes a few loc.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: use managed kzalloc for mcp
Sebastian Reichel [Mon, 15 May 2017 09:24:31 +0000 (11:24 +0200)]
pinctrl: mcp23s08: use managed kzalloc for mcp

Let's remove a few lines of code by using managed memory for mcp
variable.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: irq mapping is already done
Sebastian Reichel [Mon, 15 May 2017 09:24:30 +0000 (11:24 +0200)]
pinctrl: mcp23s08: irq mapping is already done

i2c-core and spi-core already assign the irq, so we
can drop the additional call from the mcp driver.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: drop OF_GPIO dependency
Sebastian Reichel [Mon, 15 May 2017 09:24:29 +0000 (11:24 +0200)]
pinctrl: mcp23s08: drop OF_GPIO dependency

The driver compiles & works perfectly fine without OF_GPIO on x86,
so lets drop the dependency.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: switch to regmap caching
Sebastian Reichel [Mon, 15 May 2017 09:24:28 +0000 (11:24 +0200)]
pinctrl: mcp23s08: switch to regmap caching

Instead of using custom caching, this switches to regmap based
caching. Before the conversion the debugfs file used uncached
values, so that it was easily possible to see power-loss related
problems. The new code will check and recover at this place.

The patch will also ensure, that irqs are not cleared by checking
register status in debugfs.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: drop pullup config from pdata
Sebastian Reichel [Mon, 15 May 2017 09:24:27 +0000 (11:24 +0200)]
pinctrl: mcp23s08: drop pullup config from pdata

mcp23s08 support configuration of the pullups using the
pinconf framework. This removes the custom pullup configuration
from platform data, which has no upstream users.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mcp23s08: add pinconf support
Sebastian Reichel [Mon, 15 May 2017 09:24:26 +0000 (11:24 +0200)]
pinctrl: mcp23s08: add pinconf support

mcp23xxx device have configurable 100k pullup resistors. This adds
support for enabling them using pinctrl's pinconf interface.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Tested-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agogpio: mcp23s08: move to pinctrl
Sebastian Reichel [Mon, 15 May 2017 09:24:25 +0000 (11:24 +0200)]
gpio: mcp23s08: move to pinctrl

This moves the mcp23s08 driver from gpio to pinctrl. Actual
pinctrl support for configuration of the pull-up resistors
follows in its own patch.

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Acked-by: Sylvain Lemieux <slemieux.tyco@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agoMerge branch 'ingenic' into devel
Linus Walleij [Mon, 22 May 2017 15:27:18 +0000 (17:27 +0200)]
Merge branch 'ingenic' into devel

7 years agoMIPS: jz4740: Remove custom GPIO code
Paul Cercueil [Fri, 12 May 2017 16:53:07 +0000 (18:53 +0200)]
MIPS: jz4740: Remove custom GPIO code

All the drivers for the various hardware elements of the jz4740 SoC have
been modified to use the pinctrl framework for their pin configuration
needs.
As such, this platform code is now unused and can be deleted.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agoMIPS: JZ4780: CI20: Add pinctrl configuration for several drivers
Paul Cercueil [Fri, 12 May 2017 16:53:02 +0000 (18:53 +0200)]
MIPS: JZ4780: CI20: Add pinctrl configuration for several drivers

We set the pin configuration for the jz4780-nand and jz4780-uart
drivers.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agoMIPS: JZ4740: Qi LB60: Add pinctrl configuration for several drivers
Paul Cercueil [Fri, 12 May 2017 16:53:01 +0000 (18:53 +0200)]
MIPS: JZ4740: Qi LB60: Add pinctrl configuration for several drivers

We set the pin configuration for the jz4740-nand, jz4740-mmc,
jz4740-fb, jz4740-pwm and jz4740-uart drivers.

This will permit those drivers to be cleaned out of the custom GPIO code
that they currently use.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agoMIPS: jz4780: DTS: Add nodes for ingenic pinctrl and gpio drivers
Paul Cercueil [Fri, 12 May 2017 16:53:00 +0000 (18:53 +0200)]
MIPS: jz4780: DTS: Add nodes for ingenic pinctrl and gpio drivers

For a description of the devicetree node, please read
Documentation/devicetree/bindings/pinctrl/ingenic,pinctrl.txt

For a description of the gpio devicetree nodes, please read
Documentation/devicetree/bindings/gpio/ingenic,gpio.txt

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agoMIPS: jz4740: DTS: Add nodes for ingenic pinctrl and gpio drivers
Paul Cercueil [Fri, 12 May 2017 16:52:59 +0000 (18:52 +0200)]
MIPS: jz4740: DTS: Add nodes for ingenic pinctrl and gpio drivers

For a description of the pinctrl devicetree node, please read
Documentation/devicetree/bindings/pinctrl/ingenic,pinctrl.txt

For a description of the gpio devicetree nodes, please read
Documentation/devicetree/bindings/gpio/ingenic,gpio.txt

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agoMIPS: ingenic: Enable pinctrl for all ingenic SoCs
Paul Cercueil [Fri, 12 May 2017 16:52:58 +0000 (18:52 +0200)]
MIPS: ingenic: Enable pinctrl for all ingenic SoCs

There is a pinctrl driver for each of the Ingenic SoCs supported by the
upstream Linux kernel. In order to switch away from the old GPIO
platform code, we now enable the pinctrl drivers by default for the
Ingenic SoCs.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agomtd: nand: jz4740: Let the pinctrl driver configure the pins
Paul Cercueil [Fri, 12 May 2017 16:53:04 +0000 (18:53 +0200)]
mtd: nand: jz4740: Let the pinctrl driver configure the pins

Before, this NAND driver would set itself the configuration of the
chip-select pins for the various NAND banks.

Now that the JZ4740 and similar SoCs have a pinctrl driver, we rely on
the pins being properly configured before the driver probes.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agofbdev: jz4740-fb: Let the pinctrl driver configure the pins
Paul Cercueil [Fri, 12 May 2017 16:53:05 +0000 (18:53 +0200)]
fbdev: jz4740-fb: Let the pinctrl driver configure the pins

Now that the JZ4740 and similar SoCs have a pinctrl driver, we rely on
the pins being properly configured before the driver probes.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Acked-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopwm: jz4740: Let the pinctrl driver configure the pins
Paul Cercueil [Fri, 12 May 2017 16:53:06 +0000 (18:53 +0200)]
pwm: jz4740: Let the pinctrl driver configure the pins

Now that the JZ4740 and similar SoCs have a pinctrl driver, we rely on
the pins being properly configured before the driver probes.

One inherent problem of this new approach is that the pinctrl framework
does not allow us to configure each pin on demand, when the various PWM
channels are requested or released. For instance, the PWM channels can
be configured from sysfs, which would require all PWM pins to be configured
properly beforehand for the PWM function, eventually causing conflicts
with other platform or board drivers.

The proper solution here would be to modify the pwm-jz4740 driver to
handle only one PWM channel, and create an instance of this driver
for each one of the 8 PWM channels. Then, it could use the pinctrl
framework to dynamically configure the PWM pin it controls.

Until this can be done, the only jz4740 board supported upstream
(Qi lb60) can configure all of its connected PWM pins in PWM function
mode, since those are not used by other drivers nor by GPIOs on the
board.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Acked-by: Thierry Reding <thierry.reding@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agommc: jz4740: Let the pinctrl driver configure the pins
Paul Cercueil [Fri, 12 May 2017 16:53:03 +0000 (18:53 +0200)]
mmc: jz4740: Let the pinctrl driver configure the pins

Now that the JZ4740 and similar SoCs have a pinctrl driver, we rely on
the pins being properly configured before the driver probes.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agogpio: Add gpio-ingenic driver
Paul Cercueil [Fri, 12 May 2017 16:52:57 +0000 (18:52 +0200)]
gpio: Add gpio-ingenic driver

This driver handles the GPIOs of all the Ingenic JZ47xx SoCs
currently supported by the upsteam Linux kernel.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: add a pinctrl driver for the Ingenic jz47xx SoCs
Paul Cercueil [Fri, 12 May 2017 16:52:56 +0000 (18:52 +0200)]
pinctrl: add a pinctrl driver for the Ingenic jz47xx SoCs

This driver handles pin configuration and pin muxing for the
JZ4740 and JZ4780 SoCs from Ingenic.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agodt/bindings: Document gpio-ingenic
Paul Cercueil [Fri, 12 May 2017 16:52:55 +0000 (18:52 +0200)]
dt/bindings: Document gpio-ingenic

This commit adds documentation for the devicetree bindings of the
gpio-ingenic driver, which handles GPIOs of the Ingenic SoCs
currently supported by the Linux kernel.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agodt/bindings: Document pinctrl-ingenic
Paul Cercueil [Fri, 12 May 2017 16:52:54 +0000 (18:52 +0200)]
dt/bindings: Document pinctrl-ingenic

This commit adds documentation for the devicetree bindings of the
pinctrl-ingenic driver, which handles pin configuration and pin
muxing of the Ingenic SoCs currently supported by the Linux kernel.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: When claiming hog, skip maps not served by same device
Nikita Yushchenko [Thu, 11 May 2017 20:02:11 +0000 (23:02 +0300)]
pinctrl: When claiming hog, skip maps not served by same device

When pinctrl device registers, it automatically claims hogs, that is,
maps that pinctrl device serves for itself.

It is possible that in addition to SoC's pinctrl device, other pinctrl
devices get registered. E.g. some gpio expander devies are registered
as pinctrl devices. For such devices, pinctrl maps could be defined
that set up SoC's pins (e.g. interrupt pin for gpio expander). Such
a map will have target device set to gpio expander.

Here is device tree snippet that causes this scenario:

&i2c0 {
sx1503@20 {
compatible = "semtech,sx1503q";
pinctrl-names = "default";
pinctrl-0 = <&pinctrl_sx1503_20>;
...
};
};
...
&iomuxc {
pinctrl_sx1503_20: pinctrl-sx1503-20 {
fsl,pins = <
VF610_PAD_PTB1__GPIO_23         0x219d
>;
};
};

Such a map will have target device set to gpio expander. However is not
a hog, it is a regular map that is claimed by core before gpio expander
device is probed.

Thus when looking for hogs, it is not enough to check that map's target
device is set to pinctrl device being registered. Need also check that
map's control device is also set to the same.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: imx: add soc specific mux_mode mask and shift property
Dong Aisheng [Fri, 19 May 2017 07:05:43 +0000 (15:05 +0800)]
pinctrl: imx: add soc specific mux_mode mask and shift property

MX7ULP MUX mode mask and shift bit is different from VF610.
Let's make it a platform specific property for the later easy of
adding MX7ULP support.

One trick in exist code that Vybrid hardcoded the config part
as 0xffff because its mux_config register BIT[15-0] are all configs
part. But it's not true in ULP, so use mux_mask instead to address
the difference.

Cc: Stefan Agner <stefan@agner.ch>
Cc: Bai Ping <ping.bai@nxp.com>
Signed-off-by: Fugang Duan <fugang.duan@nxp.com>
Signed-off-by: Dong Aisheng <aisheng.dong@nxp.com>
Acked-by: Shawn Guo <shawnguo@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: imx: add generic pin config core support
Dong Aisheng [Fri, 19 May 2017 07:05:42 +0000 (15:05 +0800)]
pinctrl: imx: add generic pin config core support

The design is based on the exist architecture that the core will
provide a uniformed way to decode the generic pin config into platform
config register raw data according to the imx_cfg_params_decode maps
registered by platform.

Two useful macros, IMX_CFG_PARAMS_DECODE and IMX_CFG_PARAMS_DECODE_INVERT,
are created for platform to register decode map conveniently.

In order to cope with some special case, a platform specific fixup()
function is also available to use.

Note that rather than fully utilizing the generic pinconf support
provided by pinctrl core, IMX only adopts the device tree bindings
of generic pinconf. The config used in .pin_config_get[set] are raw
register data instead of generic one which makes us align the exist
using. And that's also why we cannot set pinconf_ops.is_generic.

Cc: Bai Ping <ping.bai@nxp.com>
Signed-off-by: Dong Aisheng <aisheng.dong@nxp.com>
Acked-by: Shawn Guo <shawnguo@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: imx: fix debug message for SHARE_MUX_CONF_REG case
Dong Aisheng [Fri, 19 May 2017 07:05:41 +0000 (15:05 +0800)]
pinctrl: imx: fix debug message for SHARE_MUX_CONF_REG case

The original implemented debug message does not work for
SHARE_MUX_CONF_REG case. This patch fixes it.

Fixes: bf5a530971af ("pinctrl: imx: add VF610 support to imx pinctrl framework")
Signed-off-by: Dong Aisheng <aisheng.dong@nxp.com>
Acked-by: Shawn Guo <shawnguo@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: imx: Check for memory allocation failure
Christophe JAILLET [Sat, 6 May 2017 08:23:59 +0000 (10:23 +0200)]
pinctrl: imx: Check for memory allocation failure

If 'devm_kzalloc' fails, a NULL pointer will be dereferenced.
Return -ENOMEM instead, as done for the other memory allocation just a
few lines below.
BTW, change the 'devm_kzalloc' into a 'devm_kcalloc'.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: mediatek: reuse pinctrl driver for mt7623
Sean Wang [Mon, 1 May 2017 07:54:34 +0000 (15:54 +0800)]
pinctrl: mediatek: reuse pinctrl driver for mt7623

mt7623 pinctrl hardware can be compatible with mt2701 driver,
so the patch lets the pinctrl on mt7623 SoC reuse the driver
and deletes those redundant ones.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Acked-by: John Crispin <john@phrozen.org>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: meson: meson8: add the AO HDMI CEC pin
Martin Blumenstingl [Sat, 6 May 2017 16:57:51 +0000 (18:57 +0200)]
pinctrl: meson: meson8: add the AO HDMI CEC pin

This adds another missing pin found in the Meson8 SoCs. Currently
there's no driver which would use this pin yet.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: meson: meson8: add support for the AO remote output pin
Martin Blumenstingl [Sat, 6 May 2017 16:57:50 +0000 (18:57 +0200)]
pinctrl: meson: meson8: add support for the AO remote output pin

This adds another missing pin found in the Meson8 SoCs. Currently
there's no driver which would use this pin yet.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: meson: meson8: add support for the I2S and SPDIF pins
Martin Blumenstingl [Sat, 6 May 2017 16:57:49 +0000 (18:57 +0200)]
pinctrl: meson: meson8: add support for the I2S and SPDIF pins

This adds support for the I2S and SPDIF input and output pins, similar
to what we have on GXBB and GXL.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: meson: meson8: add the PWM pins
Martin Blumenstingl [Sat, 6 May 2017 16:57:48 +0000 (18:57 +0200)]
pinctrl: meson: meson8: add the PWM pins

This adds the missing pins for the PWM controllers found in Amlogic
Meson8 SoCs. This includes the pins for PWM_A, PWM_B, PWM_C, PWM_D,
PWM_E and PWM_F controllers.
There is an additional PWM function with the name PWM_VS in the vendor
kernel sources which seems to be used for external video input. Thus
it's not part of this change as the IP block behind the pwm-meson driver
is not responsible for these pins.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: Adjust nine checks for null pointers
Markus Elfring [Tue, 2 May 2017 09:04:55 +0000 (11:04 +0200)]
pinctrl: Adjust nine checks for null pointers

MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The script “checkpatch.pl” pointed information out like the following.

Comparison to NULL could be written …

Thus fix the affected source code places.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: Use seq_putc() in pinctrl_maps_show()
Markus Elfring [Tue, 2 May 2017 08:47:35 +0000 (10:47 +0200)]
pinctrl: Use seq_putc() in pinctrl_maps_show()

A single character (line break) should be put into a sequence.
Thus use the corresponding function "seq_putc".

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: Add spaces for better code readability
Markus Elfring [Tue, 2 May 2017 08:32:19 +0000 (10:32 +0200)]
pinctrl: Add spaces for better code readability

The script "checkpatch.pl" pointed information out like the following.

CHECK: spaces preferred around that '+' (ctx:VxV)

Thus fix the affected source code places.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: Combine substrings for a message in pin_config_group_get()
Markus Elfring [Tue, 2 May 2017 08:22:47 +0000 (10:22 +0200)]
pinctrl: Combine substrings for a message in pin_config_group_get()

The script "checkpatch.pl" pointed information out like the following.

WARNING: quoted string split across lines

Thus fix the affected source code place.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: Adjust five checks for null pointers
Markus Elfring [Tue, 2 May 2017 08:01:57 +0000 (10:01 +0200)]
pinctrl: Adjust five checks for null pointers

MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The script “checkpatch.pl” pointed information out like the following.

Comparison to NULL could be written !…

Thus fix the affected source code places.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: Replace two seq_printf() calls by seq_puts() in pinconf_show_map()
Markus Elfring [Tue, 2 May 2017 07:52:50 +0000 (09:52 +0200)]
pinctrl: Replace two seq_printf() calls by seq_puts() in pinconf_show_map()

Strings which did not contain data format specifications should be put
into a sequence. Thus use the corresponding function "seq_puts".

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl/amd: Update contact information for AMD pinctrl/amd
Shyam Sundar S K [Wed, 3 May 2017 06:29:11 +0000 (11:59 +0530)]
pinctrl/amd: Update contact information for AMD pinctrl/amd

Updating the point of contact for AMD GPIO driver.

Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
Signed-off-by: Nehal Shah <Nehal-bakulchandra.Shah@amd.com>
Cc: Ken Xue <Ken.Xue@amd.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: Use seq_putc() in three functions
Markus Elfring [Mon, 1 May 2017 20:24:29 +0000 (22:24 +0200)]
pinctrl: Use seq_putc() in three functions

A single character (line break) should be put into a sequence.
Thus use the corresponding function "seq_putc".

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: armada-37xx: Add irqchip support
Gregory CLEMENT [Fri, 28 Apr 2017 14:01:33 +0000 (16:01 +0200)]
pinctrl: armada-37xx: Add irqchip support

The Armada 37xx SoCs can handle interrupt through GPIO. However it can
only manage the edge ones.

The way the interrupt are managed is classical so we can use the generic
interrupt chip model.

The only unusual "feature" is that many interrupts are connected to the
parent interrupt controller. But we do not take advantage of this and use
the chained irq with all of them.

Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: rockchip: remove unneeded (void *) casts in of_match_table
Masahiro Yamada [Fri, 28 Apr 2017 12:50:35 +0000 (21:50 +0900)]
pinctrl: rockchip: remove unneeded (void *) casts in of_match_table

of_device_id::data is an opaque pointer.  No explicit cast is needed.

Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agopinctrl: add ZTE ZX pinctrl driver support
Shawn Guo [Mon, 24 Apr 2017 13:01:13 +0000 (21:01 +0800)]
pinctrl: add ZTE ZX pinctrl driver support

The pin controller on ZTE ZX platforms is kinda of hybrid.  It consists
of a main controller and an auxiliary one.  For example, on ZX296718 SoC,
the main controller is TOP_PMM and the auxiliary one is AON_IOCFG.  Both
controllers work together to control pin multiplexing and configuration.

For most of pins, the pinmux function is controlled by main controller
only, and this type of pins are meant by term 'TOP pins'.  For other
pins, the pinmux is controlled by both main and auxiliary controllers,
as the available multiplexing functions for the pin spread in both
controllers.  This type of pins are called 'AON pins'.  Though pinmux
implementation is quite different, pinconf is same for both types of
pins.  Both are controlled by auxiliary controller, i.e. AON_IOCFG on
ZX296718.

The patch adds the ZTE ZX core pinctrl driver to support this hybrid
pin controller as well as ZX296718 SoC specific pin data.

Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agodt-bindings: add bindings doc for ZTE pinctrl
Shawn Guo [Mon, 24 Apr 2017 13:01:12 +0000 (21:01 +0800)]
dt-bindings: add bindings doc for ZTE pinctrl

It adds device tree bindings for ZTE pin controller found on ZX2967xx
family SoCs.

Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
7 years agoLinux 4.12-rc2
Linus Torvalds [Mon, 22 May 2017 02:30:23 +0000 (19:30 -0700)]
Linux 4.12-rc2

7 years agox86: fix 32-bit case of __get_user_asm_u64()
Linus Torvalds [Mon, 22 May 2017 01:26:54 +0000 (18:26 -0700)]
x86: fix 32-bit case of __get_user_asm_u64()

The code to fetch a 64-bit value from user space was entirely buggered,
and has been since the code was merged in early 2016 in commit
b2f680380ddf ("x86/mm/32: Add support for 64-bit __get_user() on 32-bit
kernels").

Happily the buggered routine is almost certainly entirely unused, since
the normal way to access user space memory is just with the non-inlined
"get_user()", and the inlined version didn't even historically exist.

The normal "get_user()" case is handled by external hand-written asm in
arch/x86/lib/getuser.S that doesn't have either of these issues.

There were two independent bugs in __get_user_asm_u64():

 - it still did the STAC/CLAC user space access marking, even though
   that is now done by the wrapper macros, see commit 11f1a4b9755f
   ("x86: reorganize SMAP handling in user space accesses").

   This didn't result in a semantic error, it just means that the
   inlined optimized version was hugely less efficient than the
   allegedly slower standard version, since the CLAC/STAC overhead is
   quite high on modern Intel CPU's.

 - the double register %eax/%edx was marked as an output, but the %eax
   part of it was touched early in the asm, and could thus clobber other
   inputs to the asm that gcc didn't expect it to touch.

   In particular, that meant that the generated code could look like
   this:

        mov    (%eax),%eax
        mov    0x4(%eax),%edx

   where the load of %edx obviously was _supposed_ to be from the 32-bit
   word that followed the source of %eax, but because %eax was
   overwritten by the first instruction, the source of %edx was
   basically random garbage.

The fixes are trivial: remove the extraneous STAC/CLAC entries, and mark
the 64-bit output as early-clobber to let gcc know that no inputs should
alias with the output register.

Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: stable@kernel.org # v4.8+
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agoClean up x86 unsafe_get/put_user() type handling
Linus Torvalds [Sun, 21 May 2017 22:25:46 +0000 (15:25 -0700)]
Clean up x86 unsafe_get/put_user() type handling

Al noticed that unsafe_put_user() had type problems, and fixed them in
commit a7cc722fff0b ("fix unsafe_put_user()"), which made me look more
at those functions.

It turns out that unsafe_get_user() had a type issue too: it limited the
largest size of the type it could handle to "unsigned long".  Which is
fine with the current users, but doesn't match our existing normal
get_user() semantics, which can also handle "u64" even when that does
not fit in a long.

While at it, also clean up the type cast in unsafe_put_user().  We
actually want to just make it an assignment to the expected type of the
pointer, because we actually do want warnings from types that don't
convert silently.  And it makes the code more readable by not having
that one very long and complex line.

[ This patch might become stable material if we ever end up back-porting
  any new users of the unsafe uaccess code, but as things stand now this
  doesn't matter for any current existing uses. ]

Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Sun, 21 May 2017 19:06:44 +0000 (12:06 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/viro/vfs

Pull misc uaccess fixes from Al Viro:
 "Fix for unsafe_put_user() (no callers currently in mainline, but
  anyone starting to use it will step into that) + alpha osf_wait4()
  infoleak fix"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  osf_wait4(): fix infoleak
  fix unsafe_put_user()

7 years agoMerge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 21 May 2017 18:52:00 +0000 (11:52 -0700)]
Merge branch 'sched-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull scheduler fix from Thomas Gleixner:
 "A single scheduler fix:

  Prevent idle task from ever being preempted. That makes sure that
  synchronize_rcu_tasks() which is ignoring idle task does not pretend
  that no task is stuck in preempted state. If that happens and idle was
  preempted on a ftrace trampoline the machine crashes due to
  inconsistent state"

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/core: Call __schedule() from do_idle() without enabling preemption

7 years agoMerge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 21 May 2017 18:45:26 +0000 (11:45 -0700)]
Merge branch 'irq-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull irq fixes from Thomas Gleixner:
 "A set of small fixes for the irq subsystem:

   - Cure a data ordering problem with chained interrupts

   - Three small fixlets for the mbigen irq chip"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  genirq: Fix chained interrupt data ordering
  irqchip/mbigen: Fix the clear register offset calculation
  irqchip/mbigen: Fix potential NULL dereferencing
  irqchip/mbigen: Fix memory mapping code

7 years agoosf_wait4(): fix infoleak
Al Viro [Mon, 15 May 2017 01:47:25 +0000 (21:47 -0400)]
osf_wait4(): fix infoleak

failing sys_wait4() won't fill struct rusage...

Cc: stable@vger.kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
7 years agofix unsafe_put_user()
Al Viro [Sun, 21 May 2017 17:08:42 +0000 (13:08 -0400)]
fix unsafe_put_user()

__put_user_size() relies upon its first argument having the same type as what
the second one points to; the only other user makes sure of that and
unsafe_put_user() should do the same.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
7 years agoMerge tag 'trace-v4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt...
Linus Torvalds [Sun, 21 May 2017 06:39:03 +0000 (23:39 -0700)]
Merge tag 'trace-v4.12-rc1' of git://git./linux/kernel/git/rostedt/linux-trace

Pull tracing fixes from Steven Rostedt:

 - Fix a bug caused by not cleaning up the new instance unique triggers
   when deleting an instance. It also creates a selftest that triggers
   that bug.

 - Fix the delayed optimization happening after kprobes boot up self
   tests being removed by freeing of init memory.

 - Comment kprobes on why the delay optimization is not a problem for
   removal of modules, to keep other developers from searching that
   riddle.

 - Fix another case of rcu not watching in stack trace tracing.

* tag 'trace-v4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Make sure RCU is watching before calling a stack trace
  kprobes: Document how optimized kprobes are removed from module unload
  selftests/ftrace: Add test to remove instance with active event triggers
  selftests/ftrace: Fix bashisms
  ftrace: Remove #ifdef from code and add clear_ftrace_function_probes() stub
  ftrace/instances: Clear function triggers when removing instances
  ftrace: Simplify glob handling in unregister_ftrace_function_probe_func()
  tracing/kprobes: Enforce kprobes teardown after testing
  tracing: Move postpone selftests to core from early_initcall

7 years agoMerge branch 'for-linus' of git://git.kernel.dk/linux-block
Linus Torvalds [Sat, 20 May 2017 23:12:30 +0000 (16:12 -0700)]
Merge branch 'for-linus' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "A small collection of fixes that should go into this cycle.

   - a pull request from Christoph for NVMe, which ended up being
     manually applied to avoid pulling in newer bits in master. Mostly
     fibre channel fixes from James, but also a few fixes from Jon and
     Vijay

   - a pull request from Konrad, with just a single fix for xen-blkback
     from Gustavo.

   - a fuseblk bdi fix from Jan, fixing a regression in this series with
     the dynamic backing devices.

   - a blktrace fix from Shaohua, replacing sscanf() with kstrtoull().

   - a request leak fix for drbd from Lars, fixing a regression in the
     last series with the kref changes. This will go to stable as well"

* 'for-linus' of git://git.kernel.dk/linux-block:
  nvmet: release the sq ref on rdma read errors
  nvmet-fc: remove target cpu scheduling flag
  nvme-fc: stop queues on error detection
  nvme-fc: require target or discovery role for fc-nvme targets
  nvme-fc: correct port role bits
  nvme: unmap CMB and remove sysfs file in reset path
  blktrace: fix integer parse
  fuseblk: Fix warning in super_setup_bdi_name()
  block: xen-blkback: add null check to avoid null pointer dereference
  drbd: fix request leak introduced by locking/atomic, kref: Kill kref_sub()

7 years agonvmet: release the sq ref on rdma read errors
Vijay Immanuel [Mon, 8 May 2017 23:38:35 +0000 (16:38 -0700)]
nvmet: release the sq ref on rdma read errors

On rdma read errors, release the sq ref that was taken
when the req was initialized. This avoids a hang in
nvmet_sq_destroy() when the queue is being freed.

Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
7 years agonvmet-fc: remove target cpu scheduling flag
James Smart [Tue, 25 Apr 2017 23:23:09 +0000 (16:23 -0700)]
nvmet-fc: remove target cpu scheduling flag

Remove NVMET_FCTGTFEAT_NEEDS_CMD_CPUSCHED. It's unnecessary.

Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
7 years agonvme-fc: stop queues on error detection
James Smart [Tue, 25 Apr 2017 22:32:01 +0000 (15:32 -0700)]
nvme-fc: stop queues on error detection

Per the recommendation by Sagi on:
http://lists.infradead.org/pipermail/linux-nvme/2017-April/009261.html

Rather than waiting for reset work thread to stop queues and abort the ios,
immediately stop the queues on error detection. Reset thread will restop
the queues (as it's called on other paths), but it does not appear to have
a side effect.

Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
7 years agonvme-fc: require target or discovery role for fc-nvme targets
James Smart [Fri, 5 May 2017 23:13:15 +0000 (16:13 -0700)]
nvme-fc: require target or discovery role for fc-nvme targets

In order to create an association, the remoteport must be
serving either a target role or a discovery role.

Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
7 years agonvme-fc: correct port role bits
James Smart [Fri, 5 May 2017 23:13:02 +0000 (16:13 -0700)]
nvme-fc: correct port role bits

FC Port roles is a bit mask, not individual values.
Correct nvme definitions to unique bits.

Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
7 years agonvme: unmap CMB and remove sysfs file in reset path
Jon Derrick [Fri, 5 May 2017 20:52:06 +0000 (14:52 -0600)]
nvme: unmap CMB and remove sysfs file in reset path

CMB doesn't get unmapped until removal while getting remapped on every
reset. Add the unmapping and sysfs file removal to the reset path in
nvme_pci_disable to match the mapping path in nvme_pci_enable.

Fixes: 202021c1a ("nvme : Add sysfs entry for NVMe CMBs when appropriate")

Signed-off-by: Jon Derrick <jonathan.derrick@intel.com>
Acked-by: Keith Busch <keith.busch@intel.com>
Reviewed-By: Stephen Bates <sbates@raithlin.com>
Cc: <stable@vger.kernel.org> # 4.9+
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
7 years agoMerge tag 'staging-4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Sat, 20 May 2017 16:02:27 +0000 (09:02 -0700)]
Merge tag 'staging-4.12-rc2' of git://git./linux/kernel/git/gregkh/staging

Pull staging driver fixes from Greg KH:
 "Here are a number of staging driver fixes for 4.12-rc2

  Most of them are typec driver fixes found by reviewers and users of
  the code. There are also some removals of files no longer needed in
  the tree due to the ion driver rewrite in 4.12-rc1, as well as some
  wifi driver fixes. And to round it out, a MAINTAINERS file update.

  All have been in linux-next with no reported issues"

* tag 'staging-4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging: (22 commits)
  MAINTAINERS: greybus-dev list is members-only
  staging: fsl-dpaa2/eth: add ETHERNET dependency
  staging: typec: fusb302: refactor resume retry mechanism
  staging: typec: fusb302: reset i2c_busy state in error
  staging: rtl8723bs: remove re-positioned call to kfree in os_dep/ioctl_cfg80211.c
  staging: rtl8192e: GetTs Fix invalid TID 7 warning.
  staging: rtl8192e: rtl92e_get_eeprom_size Fix read size of EPROM_CMD.
  staging: rtl8192e: fix 2 byte alignment of register BSSIDR.
  staging: rtl8192e: rtl92e_fill_tx_desc fix write to mapped out memory.
  staging: vc04_services: Fix bulk cache maintenance
  staging: ccree: remove extraneous spin_unlock_bh() in error handler
  staging: typec: Fix sparse warnings about incorrect types
  staging: typec: fusb302: do not free gpio from managed resource
  staging: typec: tcpm: Fix Port Power Role field in PS_RDY messages
  staging: typec: tcpm: Respond to Discover Identity commands
  staging: typec: tcpm: Set correct flags in PD request messages
  staging: typec: tcpm: Drop duplicate PD messages
  staging: typec: fusb302: Fix chip->vbus_present init value
  staging: typec: fusb302: Fix module autoload
  staging: typec: tcpci: declare private structure as static
  ...

7 years agoMerge tag 'usb-4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Sat, 20 May 2017 15:52:34 +0000 (08:52 -0700)]
Merge tag 'usb-4.12-rc2' of git://git./linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are a number of small USB fixes for 4.12-rc2

  Most of them come from Johan, in his valiant quest to fix up all
  drivers that could be affected by "malicious" USB devices. There's
  also some fixes for more "obscure" drivers to handle some of the
  vmalloc stack fallout (which for USB drivers, was always the case, but
  very few people actually ran those systems...)

  Other than that, the normal set of xhci and gadget and musb driver
  fixes as well.

  All have been in linux-next with no reported issues"

* tag 'usb-4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (42 commits)
  usb: musb: tusb6010_omap: Do not reset the other direction's packet size
  usb: musb: Fix trying to suspend while active for OTG configurations
  usb: host: xhci-plat: propagate return value of platform_get_irq()
  xhci: Fix command ring stop regression in 4.11
  xhci: remove GFP_DMA flag from allocation
  USB: xhci: fix lock-inversion problem
  usb: host: xhci-ring: don't need to clear interrupt pending for MSI enabled hcd
  usb: host: xhci-mem: allocate zeroed Scratchpad Buffer
  xhci: apply PME_STUCK_QUIRK and MISSING_CAS quirk for Denverton
  usb: xhci: trace URB before giving it back instead of after
  USB: serial: qcserial: add more Lenovo EM74xx device IDs
  USB: host: xhci: use max-port define
  USB: hub: fix SS max number of ports
  USB: hub: fix non-SS hub-descriptor handling
  USB: hub: fix SS hub-descriptor handling
  USB: usbip: fix nonconforming hub descriptor
  USB: gadget: dummy_hcd: fix hub-descriptor removable fields
  doc-rst: fixed kernel-doc directives in usb/typec.rst
  USB: core: of: document reference taken by companion helper
  USB: ehci-platform: fix companion-device leak
  ...

7 years agoMerge tag 'char-misc-4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregk...
Linus Torvalds [Sat, 20 May 2017 15:44:22 +0000 (08:44 -0700)]
Merge tag 'char-misc-4.12-rc2' of git://git./linux/kernel/git/gregkh/char-misc

Pull char/misc driver fixes from Greg KH:
 "Here are five small bugfixes for reported issues with 4.12-rc1 and
  earlier kernels. Nothing huge here, just a lp, mem, vpd, and uio
  driver fix, along with a Kconfig fixup for one of the misc drivers.

  All of these have been in linux-next with no reported issues"

* tag 'char-misc-4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  firmware: Google VPD: Fix memory allocation error handling
  drivers: char: mem: Check for address space wraparound with mmap()
  uio: fix incorrect memory leak cleanup
  misc: pci_endpoint_test: select CRC32
  char: lp: fix possible integer overflow in lp_setup()

7 years agoMerge git://www.linux-watchdog.org/linux-watchdog
Linus Torvalds [Sat, 20 May 2017 15:35:27 +0000 (08:35 -0700)]
Merge git://www.linux-watchdog.org/linux-watchdog

Pull watchdog fixes from Wim Van Sebroeck:
 - orion_wdt compile-test dependencies
 - sama5d4_wdt: WDDIS handling and a race confition
 - pcwd_usb: fix NULL-deref at probe
 - cadence_wdt: fix timeout setting
 - wdt_pci: fix build error if SOFTWARE_REBOOT is defined
 - iTCO_wdt: all versions count down twice
 - zx2967: remove redundant dev_err call in zx2967_wdt_probe()
 - bcm281xx: Fix use of uninitialized spinlock

* git://www.linux-watchdog.org/linux-watchdog:
  watchdog: bcm281xx: Fix use of uninitialized spinlock.
  watchdog: zx2967: remove redundant dev_err call in zx2967_wdt_probe()
  iTCO_wdt: all versions count down twice
  watchdog: wdt_pci: fix build error if define SOFTWARE_REBOOT
  watchdog: cadence_wdt: fix timeout setting
  watchdog: pcwd_usb: fix NULL-deref at probe
  watchdog: sama5d4: fix race condition
  watchdog: sama5d4: fix WDDIS handling
  watchdog: orion: fix compile-test dependencies

7 years agoMerge tag 'drm-fixes-for-v4.12-rc2' of git://people.freedesktop.org/~airlied/linux
Linus Torvalds [Sat, 20 May 2017 15:29:30 +0000 (08:29 -0700)]
Merge tag 'drm-fixes-for-v4.12-rc2' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "Mostly nouveau and i915, fairly quiet as usual for rc2"

* tag 'drm-fixes-for-v4.12-rc2' of git://people.freedesktop.org/~airlied/linux:
  drm/atmel-hlcdc: Fix output initialization
  gpu: host1x: select IOMMU_IOVA
  drm/nouveau/fifo/gk104-: Silence a locking warning
  drm/nouveau/secboot: plug memory leak in ls_ucode_img_load_gr() error path
  drm/nouveau: Fix drm poll_helper handling
  drm/i915: don't do allocate_va_range again on PIN_UPDATE
  drm/i915: Fix rawclk readout for g4x
  drm/i915: Fix runtime PM for LPE audio
  drm/i915/glk: Fix DSI "*ERROR* ULPS is still active" messages
  drm/i915/gvt: avoid unnecessary vgpu switch
  drm/i915/gvt: not to restore in-context mmio
  drm/etnaviv: don't put fence in case of submit failure
  drm/i915/gvt: fix typo: "supporte" -> "support"
  drm: hdlcd: Fix the calculation of the scanout start address

7 years agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Sat, 20 May 2017 00:46:51 +0000 (17:46 -0700)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "This is the first sweep of mostly minor fixes. There's one security
  one: the read past the end of a buffer in qedf, and a panic fix for
  lpfc SLI-3 adapters, but the rest are a set of include and build
  dependency tidy ups and assorted other small fixes and updates"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: pmcraid: remove redundant check to see if request_size is less than zero
  scsi: lpfc: ensure els_wq is being checked before destroying it
  scsi: cxlflash: Select IRQ_POLL
  scsi: qedf: Avoid reading past end of buffer
  scsi: qedf: Cleanup the type of io_log->op
  scsi: lpfc: double lock typo in lpfc_ns_rsp()
  scsi: qedf: properly update arguments position in function call
  scsi: scsi_lib: Add #include <scsi/scsi_transport.h>
  scsi: MAINTAINERS: update OSD entries
  scsi: Skip deleted devices in __scsi_device_lookup
  scsi: lpfc: Fix panic on BFS configuration
  scsi: libfc: do not flood console with messages 'libfc: queue full ...'

7 years agoMerge branch 'libnvdimm-for-next' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 20 May 2017 00:35:34 +0000 (17:35 -0700)]
Merge branch 'libnvdimm-for-next' of git://git./linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm fixes from Dan Williams:
 "A couple of compile fixes.

  With the removal of the ->direct_access() method from
  block_device_operations in favor of a new dax_device + dax_operations
  we broke two configurations.

  The CONFIG_BLOCK=n case is fixed by compiling out the block+dax
  helpers in the dax core. Configurations with FS_DAX=n EXT4=y / XFS=y
  and DAX=m fail due to the helpers the builtin filesystem needs being
  in a module, so we stub out the helpers in the FS_DAX=n case."

* 'libnvdimm-for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  dax, xfs, ext4: compile out iomap-dax paths in the FS_DAX=n case
  dax: fix false CONFIG_BLOCK dependency

7 years agoMerge branch 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa...
Linus Torvalds [Sat, 20 May 2017 00:33:08 +0000 (17:33 -0700)]
Merge branch 'i2c/for-current' of git://git./linux/kernel/git/wsa/linux

Pull i2c fix from Wolfram Sang:
 "A regression fix for I2C that would be great to have in rc2"

* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: designware: don't infer timings described by ACPI from clock rate

7 years agoMerge tag 'iommu-fixes-v4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 20 May 2017 00:27:28 +0000 (17:27 -0700)]
Merge tag 'iommu-fixes-v4.12-rc1' of git://git./linux/kernel/git/joro/iommu

Pull IOMMU fixes from Joerg Roedel:

 - another compile-fix as a fallout of the recent header-file cleanup

 - add a missing IO/TLB flush to the Intel VT-d kdump code path

 - a fix for ARM64 dma code to only access initialized iova_domain
   members

* tag 'iommu-fixes-v4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/mediatek: Include linux/dma-mapping.h
  iommu/vt-d: Flush the IOTLB to get rid of the initial kdump mappings
  iommu/dma: Don't touch invalid iova_domain members

7 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Fri, 19 May 2017 22:13:13 +0000 (15:13 -0700)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull KVM fixes from Radim Krčmář:
 "ARM:
   - a fix for a build failure introduced in -rc1 when tracepoints are
     enabled on 32-bit ARM.

   - disable use of stack pointer protection in the hyp code which can
     cause panics.

   - a handful of VGIC fixes.

   - a fix to the init of the redistributors on GICv3 systems that
     prevented boot with kvmtool on GICv3 systems introduced in -rc1.

   - a number of race conditions fixed in our MMU handling code.

   - a fix for the guest being able to program the debug extensions for
     the host on the 32-bit side.

  PPC:
   - fixes for build failures with PR KVM configurations.

   - a fix for a host crash that can occur on POWER9 with radix guests.

  x86:
   - fixes for nested PML and nested EPT.

   - a fix for crashes caused by reserved bits in SSE MXCSR that could
     have been set by userspace.

   - an optimization of halt polling that fixes high CPU overhead.

   - fixes for four reports from Dan Carpenter's static checker.

   - a protection around code that shouldn't have been preemptible.

   - a fix for port IO emulation"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (27 commits)
  KVM: x86: prevent uninitialized variable warning in check_svme()
  KVM: x86/vPMU: fix undefined shift in intel_pmu_refresh()
  KVM: x86: zero base3 of unusable segments
  KVM: X86: Fix read out-of-bounds vulnerability in kvm pio emulation
  KVM: x86: Fix potential preemption when get the current kvmclock timestamp
  KVM: Silence underflow warning in avic_get_physical_id_entry()
  KVM: arm/arm64: Hold slots_lock when unregistering kvm io bus devices
  KVM: arm/arm64: Fix bug when registering redist iodevs
  KVM: x86: lower default for halt_poll_ns
  kvm: arm/arm64: Fix use after free of stage2 page table
  kvm: arm/arm64: Force reading uncached stage2 PGD
  KVM: nVMX: fix EPT permissions as reported in exit qualification
  KVM: VMX: Don't enable EPT A/D feature if EPT feature is disabled
  KVM: x86: Fix load damaged SSEx MXCSR register
  kvm: nVMX: off by one in vmx_write_pml_buffer()
  KVM: arm: rename pm_fake handler to trap_raz_wi
  KVM: arm: plug potential guest hardware debug leakage
  kvm: arm/arm64: Fix race in resetting stage2 PGD
  KVM: arm/arm64: vgic-v3: Use PREbits to infer the number of ICH_APxRn_EL2 registers
  KVM: arm/arm64: vgic-v3: Do not use Active+Pending state for a HW interrupt
  ...

7 years agoMerge tag 'for-linus-4.12b-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 19 May 2017 22:06:48 +0000 (15:06 -0700)]
Merge tag 'for-linus-4.12b-rc2-tag' of git://git./linux/kernel/git/xen/tip

Pull xen fixes from Juergen Gross:
 "Some fixes for the new Xen 9pfs frontend and some minor cleanups"

* tag 'for-linus-4.12b-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen: make xen_flush_tlb_all() static
  xen: cleanup pvh leftovers from pv-only sources
  xen/9pfs: p9_trans_xen_init and p9_trans_xen_exit can be static
  xen/9pfs: fix return value check in xen_9pfs_front_probe()

7 years agoMerge tag 'devicetree-fixes-for-4.12' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 19 May 2017 22:03:24 +0000 (15:03 -0700)]
Merge tag 'devicetree-fixes-for-4.12' of git://git./linux/kernel/git/robh/linux

Pull DeviceTree fixes from Rob Herring:

 - fix missing allocation failure handling in fdt code

 - fix dtc compile error on 32-bit hosts

 - revert bad sparse changes causing GCC7 warnings

* tag 'devicetree-fixes-for-4.12' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
  of: fdt: add missing allocation-failure check
  dtc: check.c fix compile error
  Partially Revert "of: fix sparse warnings in fdt, irq, reserved mem, and resolver code"

7 years agoMerge tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Linus Torvalds [Fri, 19 May 2017 20:36:56 +0000 (13:36 -0700)]
Merge tag 'armsoc-fixes' of git://git./linux/kernel/git/arm/arm-soc

Pull ARM SoC fixes from Olof Johansson:
 "We had a small batch of fixes before -rc1, but here is a larger one.
  It contains a backmerge of 4.12-rc1 since some of the downstream
  branches we merge had that as base; at the same time we already had
  merged contents before -rc1 and rebase wasn't the right solution.

  A mix of random smaller fixes and a few things worth pointing out:

   - We've started telling people to avoid cross-tree shared branches if
     all they're doing is picking up one or two DT-used constants from a
     shared include file, and instead to use the numeric values on first
     submission. Follow-up moving over to symbolic names are sent in
     right after -rc1, i.e. here. It's only a few minor patches of this
     type.

   - Linus Walleij and others are resurrecting the 'Gemini' platform,
     and wanted a cut-down platform-specific defconfig for it. So I
     picked that up for them.

   - Rob Herring ran 'savedefconfig' on arm64, it's a bit churny but it
     helps people to prepare patches since it's a pain when defconfig
     and current savedefconfig contents differs too much.

   - Devicetree additions for some pinctrl drivers for Armada that were
     merged this window. I'd have preferred to see those earlier but
     it's not a huge deail.

  The biggest change worth pointing out though since it's touching other
  parts of the tree: We added prefixes to be used when cross-including
  DT contents between arm64 and arm, allowing someone to #include
  <arm/foo.dtsi> from arm64, and likewise. As part of that, we needed
  arm/foo.dtsi to work on arm as well. The way I suggested this to Heiko
  resulted in a recursive symlink.

  Instead, I've now moved it out of arch/*/boot/dts/include, into a
  shared location under scripts/dtc. While I was at it, I consolidated
  so all architectures now behave the same way in this manner.

  Rob Herring (DT maintainer) has acked it. I cc:d most other arch
  maintainers but nobody seems to care much; it doesn't really affect
  them since functionality is unchanged for them by default"

* tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (29 commits)
  arm64: dts: rockchip: fix include reference
  firmware: ti_sci: fix strncat length check
  ARM: remove duplicate 'const' annotations'
  arm64: defconfig: enable options needed for QCom DB410c board
  arm64: defconfig: sync with savedefconfig
  ARM: configs: add a gemini defconfig
  devicetree: Move include prefixes from arch to separate directory
  ARM: dts: dra7: Reduce cpu thermal shutdown temperature
  memory: omap-gpmc: Fix debug output for access width
  ARM: dts: LogicPD Torpedo: Fix camera pin mux
  ARM: dts: omap4: enable CEC pin for Pandaboard A4 and ES
  ARM: dts: gta04: fix polarity of clocks for mcbsp4
  ARM: dts: dra7: Add power hold and power controller properties to palmas
  soc: imx: add PM dependency for IMX7_PM_DOMAINS
  ARM: dts: imx6sx-sdb: Remove OPP override
  ARM: dts: imx53-qsrb: Pulldown PMIC IRQ pin
  soc: bcm: brcmstb: Correctly match 7435 SoC
  tee: add ARM_SMCCC dependency
  ARM: omap2+: make omap4_get_cpu1_ns_pa_addr declaration usable
  ARM64: dts: mediatek: configure some fixed mmc parameters
  ...

7 years agoMerge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Linus Torvalds [Fri, 19 May 2017 20:34:34 +0000 (13:34 -0700)]
Merge tag 'arm64-fixes' of git://git./linux/kernel/git/arm64/linux

Pull arm64 fixes/cleanups from Catalin Marinas:

 - Avoid taking a mutex in the secondary CPU bring-up path when
   interrupts are disabled

 - Ignore perf exclude_hv when the kernel is running in Hyp mode

 - Remove redundant instruction in cmpxchg

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64/cpufeature: don't use mutex in bringup path
  arm64: perf: Ignore exclude_hv when kernel is running in HYP
  arm64: Remove redundant mov from LL/SC cmpxchg

7 years agoMerge branch 'for-upstream/hdlcd' of git://linux-arm.org/linux-ld into drm-fixes
Dave Airlie [Fri, 19 May 2017 20:00:49 +0000 (06:00 +1000)]
Merge branch 'for-upstream/hdlcd' of git://linux-arm.org/linux-ld into drm-fixes

single hdlcd fix
* 'for-upstream/hdlcd' of git://linux-arm.org/linux-ld:
  drm: hdlcd: Fix the calculation of the scanout start address

7 years agoMerge tag 'powerpc-4.12-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc...
Linus Torvalds [Fri, 19 May 2017 18:31:38 +0000 (11:31 -0700)]
Merge tag 'powerpc-4.12-3' of git://git./linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:
 "The headliner is a fix for FP/VMX register corruption when using
  transactional memory, and a new selftest to go with it.

  Then there's the virt_addr_valid() fix, currently HARDENDED_USERCOPY
  is tripping on that causing some machines to crash.

  A few other fairly minor fixes for long tail things, and a couple of
  fixes for code we just merged.

  Thanks to: Breno Leitao, Gautham Shenoy, Michael Neuling, Naveen Rao.
  Nicholas Piggin, Paul Mackerras"

* tag 'powerpc-4.12-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/mm: Fix virt_addr_valid() etc. on 64-bit hash
  powerpc/mm: Fix crash in page table dump with huge pages
  powerpc/kprobes: Fix handling of instruction emulation on probe re-entry
  powerpc/powernv: Set NAPSTATELOST after recovering paca on P9 DD1
  selftests/powerpc: Test TM and VMX register state
  powerpc/tm: Fix FP and VMX register corruption
  powerpc/modules: If mprofile-kernel is enabled add it to vermagic

7 years agoKVM: x86: prevent uninitialized variable warning in check_svme()
Radim Krčmář [Thu, 18 May 2017 17:37:32 +0000 (19:37 +0200)]
KVM: x86: prevent uninitialized variable warning in check_svme()

get_msr() of MSR_EFER is currently always going to succeed, but static
checker doesn't see that far.

Don't complicate stuff and just use 0 for the fallback -- it means that
the feature is not present.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
7 years agoKVM: x86/vPMU: fix undefined shift in intel_pmu_refresh()
Radim Krčmář [Thu, 18 May 2017 17:37:31 +0000 (19:37 +0200)]
KVM: x86/vPMU: fix undefined shift in intel_pmu_refresh()

Static analysis noticed that pmu->nr_arch_gp_counters can be 32
(INTEL_PMC_MAX_GENERIC) and therefore cannot be used to shift 'int'.

I didn't add BUILD_BUG_ON for it as we have a better checker.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: 25462f7f5295 ("KVM: x86/vPMU: Define kvm_pmu_ops to support vPMU function dispatch")
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
7 years agoKVM: x86: zero base3 of unusable segments
Radim Krčmář [Thu, 18 May 2017 17:37:30 +0000 (19:37 +0200)]
KVM: x86: zero base3 of unusable segments

Static checker noticed that base3 could be used uninitialized if the
segment was not present (useable).  Random stack values probably would
not pass VMCS entry checks.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: 1aa366163b8b ("KVM: x86 emulator: consolidate segment accessors")
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
7 years agoKVM: X86: Fix read out-of-bounds vulnerability in kvm pio emulation
Wanpeng Li [Fri, 19 May 2017 09:46:56 +0000 (02:46 -0700)]
KVM: X86: Fix read out-of-bounds vulnerability in kvm pio emulation

Huawei folks reported a read out-of-bounds vulnerability in kvm pio emulation.

- "inb" instruction to access PIT Mod/Command register (ioport 0x43, write only,
  a read should be ignored) in guest can get a random number.
- "rep insb" instruction to access PIT register port 0x43 can control memcpy()
  in emulator_pio_in_emulated() to copy max 0x400 bytes but only read 1 bytes,
  which will disclose the unimportant kernel memory in host but no crash.

The similar test program below can reproduce the read out-of-bounds vulnerability:

void hexdump(void *mem, unsigned int len)
{
        unsigned int i, j;

        for(i = 0; i < len + ((len % HEXDUMP_COLS) ? (HEXDUMP_COLS - len % HEXDUMP_COLS) : 0); i++)
        {
                /* print offset */
                if(i % HEXDUMP_COLS == 0)
                {
                        printf("0x%06x: ", i);
                }

                /* print hex data */
                if(i < len)
                {
                        printf("%02x ", 0xFF & ((char*)mem)[i]);
                }
                else /* end of block, just aligning for ASCII dump */
                {
                        printf("   ");
                }

                /* print ASCII dump */
                if(i % HEXDUMP_COLS == (HEXDUMP_COLS - 1))
                {
                        for(j = i - (HEXDUMP_COLS - 1); j <= i; j++)
                        {
                                if(j >= len) /* end of block, not really printing */
                                {
                                        putchar(' ');
                                }
                                else if(isprint(((char*)mem)[j])) /* printable char */
                                {
                                        putchar(0xFF & ((char*)mem)[j]);
                                }
                                else /* other char */
                                {
                                        putchar('.');
                                }
                        }
                        putchar('\n');
                }
        }
}

int main(void)
{
int i;
if (iopl(3))
{
err(1, "set iopl unsuccessfully\n");
return -1;
}
static char buf[0x40];

/* test ioport 0x40,0x41,0x42,0x43,0x44,0x45 */

memset(buf, 0xab, sizeof(buf));

asm volatile("push %rdi;");
asm volatile("mov %0, %%rdi;"::"q"(buf));

asm volatile ("mov $0x40, %rdx;");
asm volatile ("in %dx,%al;");
asm volatile ("stosb;");

asm volatile ("mov $0x41, %rdx;");
asm volatile ("in %dx,%al;");
asm volatile ("stosb;");

asm volatile ("mov $0x42, %rdx;");
asm volatile ("in %dx,%al;");
asm volatile ("stosb;");

asm volatile ("mov $0x43, %rdx;");
asm volatile ("in %dx,%al;");
asm volatile ("stosb;");

asm volatile ("mov $0x44, %rdx;");
asm volatile ("in %dx,%al;");
asm volatile ("stosb;");

asm volatile ("mov $0x45, %rdx;");
asm volatile ("in %dx,%al;");
asm volatile ("stosb;");

asm volatile ("pop %rdi;");
hexdump(buf, 0x40);

printf("\n");

/* ins port 0x40 */

memset(buf, 0xab, sizeof(buf));

asm volatile("push %rdi;");
asm volatile("mov %0, %%rdi;"::"q"(buf));

asm volatile ("mov $0x20, %rcx;");
asm volatile ("mov $0x40, %rdx;");
asm volatile ("rep insb;");

asm volatile ("pop %rdi;");
hexdump(buf, 0x40);

printf("\n");

/* ins port 0x43 */

memset(buf, 0xab, sizeof(buf));

asm volatile("push %rdi;");
asm volatile("mov %0, %%rdi;"::"q"(buf));

asm volatile ("mov $0x20, %rcx;");
asm volatile ("mov $0x43, %rdx;");
asm volatile ("rep insb;");

asm volatile ("pop %rdi;");
hexdump(buf, 0x40);

printf("\n");
return 0;
}

The vcpu->arch.pio_data buffer is used by both in/out instrutions emulation
w/o clear after using which results in some random datas are left over in
the buffer. Guest reads port 0x43 will be ignored since it is write only,
however, the function kernel_pio() can't distigush this ignore from successfully
reads data from device's ioport. There is no new data fill the buffer from
port 0x43, however, emulator_pio_in_emulated() will copy the stale data in
the buffer to the guest unconditionally. This patch fixes it by clearing the
buffer before in instruction emulation to avoid to grant guest the stale data
in the buffer.

In addition, string I/O is not supported for in kernel device. So there is no
iteration to read ioport %RCX times for string I/O. The function kernel_pio()
just reads one round, and then copy the io size * %RCX to the guest unconditionally,
actually it copies the one round ioport data w/ other random datas which are left
over in the vcpu->arch.pio_data buffer to the guest. This patch fixes it by
introducing the string I/O support for in kernel device in order to grant the right
ioport datas to the guest.

Before the patch:

0x000000: fe 38 93 93 ff ff ab ab .8......
0x000008: ab ab ab ab ab ab ab ab ........
0x000010: ab ab ab ab ab ab ab ab ........
0x000018: ab ab ab ab ab ab ab ab ........
0x000020: ab ab ab ab ab ab ab ab ........
0x000028: ab ab ab ab ab ab ab ab ........
0x000030: ab ab ab ab ab ab ab ab ........
0x000038: ab ab ab ab ab ab ab ab ........

0x000000: f6 00 00 00 00 00 00 00 ........
0x000008: 00 00 00 00 00 00 00 00 ........
0x000010: 00 00 00 00 4d 51 30 30 ....MQ00
0x000018: 30 30 20 33 20 20 20 20 00 3
0x000020: ab ab ab ab ab ab ab ab ........
0x000028: ab ab ab ab ab ab ab ab ........
0x000030: ab ab ab ab ab ab ab ab ........
0x000038: ab ab ab ab ab ab ab ab ........

0x000000: f6 00 00 00 00 00 00 00 ........
0x000008: 00 00 00 00 00 00 00 00 ........
0x000010: 00 00 00 00 4d 51 30 30 ....MQ00
0x000018: 30 30 20 33 20 20 20 20 00 3
0x000020: ab ab ab ab ab ab ab ab ........
0x000028: ab ab ab ab ab ab ab ab ........
0x000030: ab ab ab ab ab ab ab ab ........
0x000038: ab ab ab ab ab ab ab ab ........

After the patch:

0x000000: 1e 02 f8 00 ff ff ab ab ........
0x000008: ab ab ab ab ab ab ab ab ........
0x000010: ab ab ab ab ab ab ab ab ........
0x000018: ab ab ab ab ab ab ab ab ........
0x000020: ab ab ab ab ab ab ab ab ........
0x000028: ab ab ab ab ab ab ab ab ........
0x000030: ab ab ab ab ab ab ab ab ........
0x000038: ab ab ab ab ab ab ab ab ........

0x000000: d2 e2 d2 df d2 db d2 d7 ........
0x000008: d2 d3 d2 cf d2 cb d2 c7 ........
0x000010: d2 c4 d2 c0 d2 bc d2 b8 ........
0x000018: d2 b4 d2 b0 d2 ac d2 a8 ........
0x000020: ab ab ab ab ab ab ab ab ........
0x000028: ab ab ab ab ab ab ab ab ........
0x000030: ab ab ab ab ab ab ab ab ........
0x000038: ab ab ab ab ab ab ab ab ........

0x000000: 00 00 00 00 00 00 00 00 ........
0x000008: 00 00 00 00 00 00 00 00 ........
0x000010: 00 00 00 00 00 00 00 00 ........
0x000018: 00 00 00 00 00 00 00 00 ........
0x000020: ab ab ab ab ab ab ab ab ........
0x000028: ab ab ab ab ab ab ab ab ........
0x000030: ab ab ab ab ab ab ab ab ........
0x000038: ab ab ab ab ab ab ab ab ........

Reported-by: Moguofang <moguofang@huawei.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Moguofang <moguofang@huawei.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
7 years agoKVM: x86: Fix potential preemption when get the current kvmclock timestamp
Wanpeng Li [Fri, 12 May 2017 01:12:05 +0000 (18:12 -0700)]
KVM: x86: Fix potential preemption when get the current kvmclock timestamp

 BUG: using __this_cpu_read() in preemptible [00000000] code: qemu-system-x86/2809
 caller is __this_cpu_preempt_check+0x13/0x20
 CPU: 2 PID: 2809 Comm: qemu-system-x86 Not tainted 4.11.0+ #13
 Call Trace:
  dump_stack+0x99/0xce
  check_preemption_disabled+0xf5/0x100
  __this_cpu_preempt_check+0x13/0x20
  get_kvmclock_ns+0x6f/0x110 [kvm]
  get_time_ref_counter+0x5d/0x80 [kvm]
  kvm_hv_process_stimers+0x2a1/0x8a0 [kvm]
  ? kvm_hv_process_stimers+0x2a1/0x8a0 [kvm]
  ? kvm_arch_vcpu_ioctl_run+0xac9/0x1ce0 [kvm]
  kvm_arch_vcpu_ioctl_run+0x5bf/0x1ce0 [kvm]
  kvm_vcpu_ioctl+0x384/0x7b0 [kvm]
  ? kvm_vcpu_ioctl+0x384/0x7b0 [kvm]
  ? __fget+0xf3/0x210
  do_vfs_ioctl+0xa4/0x700
  ? __fget+0x114/0x210
  SyS_ioctl+0x79/0x90
  entry_SYSCALL_64_fastpath+0x23/0xc2
 RIP: 0033:0x7f9d164ed357
  ? __this_cpu_preempt_check+0x13/0x20

This can be reproduced by run kvm-unit-tests/hyperv_stimer.flat w/
CONFIG_PREEMPT and CONFIG_DEBUG_PREEMPT enabled.

Safe access to per-CPU data requires a couple of constraints, though: the
thread working with the data cannot be preempted and it cannot be migrated
while it manipulates per-CPU variables. If the thread is preempted, the
thread that replaces it could try to work with the same variables; migration
to another CPU could also cause confusion. However there is no preemption
disable when reads host per-CPU tsc rate to calculate the current kvmclock
timestamp.

This patch fixes it by utilizing get_cpu/put_cpu pair to guarantee both
__this_cpu_read() and rdtsc() are not preempted.

Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
7 years agoblktrace: fix integer parse
Shaohua Li [Fri, 19 May 2017 15:04:59 +0000 (08:04 -0700)]
blktrace: fix integer parse

sscanf is a very poor way to parse integer. For example, I input
"discard" for act_mask, it gets 0xd and completely messes up. Using
correct API to do integer parse.

This patch also makes attributes accept any base of integer.

Signed-off-by: Shaohua Li <shli@fb.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
7 years agoi2c: designware: don't infer timings described by ACPI from clock rate
Ard Biesheuvel [Fri, 19 May 2017 08:56:40 +0000 (09:56 +0100)]
i2c: designware: don't infer timings described by ACPI from clock rate

Commit bd698d24b1b57 ("i2c: designware: Get selected speed mode
sda-hold-time via ACPI") updated the logic that reads the timing
parameters for various I2C bus rates from the DSDT, to only read
the timing parameters for the currently selected mode.

This causes a WARN_ON() splat on platforms that legally omit the clock
frequency from the ACPI description, because in the new situation, the
core I2C designware driver still accesses the fields in the driver
struct that we no longer populate, and proceeds to calculate them from
the clock frequency. Since the clock frequency is unspecified, the
driver complains loudly using a WARN_ON().

So revert back to the old situation, where the struct fields for all
timings are populated, but retain the new logic which chooses the SDA
hold time from the timing mode that is currently in use.

Fixes: bd698d24b1b57 ("i2c: designware: Get selected speed mode ...")
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reported-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
7 years agoarm64: dts: rockchip: fix include reference
Arnd Bergmann [Fri, 19 May 2017 12:12:00 +0000 (14:12 +0200)]
arm64: dts: rockchip: fix include reference

The way we handle include paths for DT has changed a bit, which
broke a file that had an unconventional way to reference a common
header file:

arch/arm64/boot/dts/rockchip/rk3399-gru-kevin.dts:47:10: fatal error: include/dt-bindings/input/linux-event-codes.h: No such file or directory

This removes the leading "include/" from the path name, which fixes it.

Fixes: d5d332d3f7e8 ("devicetree: Move include prefixes from arch to separate directory")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
7 years agowatchdog: bcm281xx: Fix use of uninitialized spinlock.
Eric Anholt [Fri, 28 Apr 2017 01:02:32 +0000 (18:02 -0700)]
watchdog: bcm281xx: Fix use of uninitialized spinlock.

The bcm_kona_wdt_set_resolution_reg() call takes the spinlock, so
initialize it earlier.  Fixes a warning at boot with lock debugging
enabled.

Fixes: 6adb730dc208 ("watchdog: bcm281xx: Watchdog Driver")
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
7 years agowatchdog: zx2967: remove redundant dev_err call in zx2967_wdt_probe()
Wei Yongjun [Tue, 25 Apr 2017 16:17:33 +0000 (16:17 +0000)]
watchdog: zx2967: remove redundant dev_err call in zx2967_wdt_probe()

There is a error message within devm_ioremap_resource
already, so remove the dev_err call to avoid redundant
error message.

Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
7 years agoiTCO_wdt: all versions count down twice
Paolo Bonzini [Wed, 5 Apr 2017 11:41:15 +0000 (13:41 +0200)]
iTCO_wdt: all versions count down twice

The ICH9 is listed as having TCO v2, and indeed the behavior in the
datasheet corresponds to v2 (for example the NO_REBOOT flag is
accessible via the 16KiB-aligned Root Complex Base Address).

However, the TCO counts twice just like in v1; the documentation
of the SECOND_TO_STS bit says: "ICH9 sets this bit to 1 to indicate
that the TIMEOUT bit had been (or is currently) set and a second
timeout occurred before the TCO_RLD register was written. If this
bit is set and the NO_REBOOT config bit is 0, then the ICH9 will
reboot the system after the second timeout.  The same can be found
in the BayTrail (Atom E3800) datasheet, and even HOWTOs around
the Internet say that it will reboot after _twice_ the specified
heartbeat.

I did not find the Apollo Lake datasheet, but because v4/v5 has
a SECOND_TO_STS bit just like the previous version I'm enabling
this for Apollo Lake as well.

Cc: linux-watchdog@vger.kernel.org
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
7 years agofirmware: ti_sci: fix strncat length check
Arnd Bergmann [Wed, 11 Jan 2017 11:53:05 +0000 (12:53 +0100)]
firmware: ti_sci: fix strncat length check

gcc-7 notices that the length we pass to strncat is wrong:

drivers/firmware/ti_sci.c: In function 'ti_sci_probe':
drivers/firmware/ti_sci.c:204:32: error: specified bound 50 equals the size of the destination [-Werror=stringop-overflow=]

Instead of the total length, we must pass the length of the
remaining space here.

Fixes: aa276781a64a ("firmware: Add basic support for TI System Control Interface (TI-SCI) protocol")
Cc: stable@vger.kernel.org
Acked-by: Nishanth Menon <nm@ti.com>
Acked-by: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
7 years agoARM: remove duplicate 'const' annotations'
Arnd Bergmann [Thu, 11 May 2017 11:50:16 +0000 (13:50 +0200)]
ARM: remove duplicate 'const' annotations'

gcc-7 warns about some declarations that are more 'const' than necessary:

arch/arm/mach-at91/pm.c:338:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier]
 static const struct of_device_id const ramc_ids[] __initconst = {
arch/arm/mach-bcm/bcm_kona_smc.c:36:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier]
 static const struct of_device_id const bcm_kona_smc_ids[] __initconst = {
arch/arm/mach-spear/time.c:207:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier]
 static const struct of_device_id const timer_of_match[] __initconst = {
arch/arm/mach-omap2/prm_common.c:714:34: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier]
 static const struct of_device_id const omap_prcm_dt_match_table[] __initconst = {
arch/arm/mach-omap2/vc.c:562:35: error: duplicate 'const' declaration specifier [-Werror=duplicate-decl-specifier]
 static const struct i2c_init_data const omap4_i2c_timing_data[] __initconst = {

The ones in arch/arm were apparently all introduced accidentally by one
commit that correctly marked a lot of variables as __initconst.

Fixes: 19c233b79d1a ("ARM: appropriate __init annotation for const data")
Acked-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Acked-by: Nicolas Pitre <nico@linaro.org>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Krzysztof Hałasa <khalasa@piap.pl>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
7 years agoMerge tag 'usb-serial-4.12-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Greg Kroah-Hartman [Fri, 19 May 2017 08:10:07 +0000 (10:10 +0200)]
Merge tag 'usb-serial-4.12-rc2' of git://git./linux/kernel/git/johan/usb-serial into usb-linus

Johan writes:

USB-serial fixes for v4.12-rc2

Here's a fix for a long-standing issue in the ftdi_sio driver that
prevented unprivileged users from updating the low-latency flag,
something which became apparent after a recent change that restored the
older setting of not using low-latency mode by default.

A run of sparse revealed a couple of endianness issues that are now
fixed, and addressed is also a user-triggerable division-by-zero in
io_ti when debugging is enabled.

Finally there are some new device ids, including a simplification of how
we deal with a couple of older Olimex JTAG adapters.

All have been in linux-next with no reported issues.

Signed-off-by: Johan Hovold <johan@kernel.org>
7 years agoMerge tag 'omap-for-v4.12/fixes-v2-signed' of git://git.kernel.org/pub/scm/linux...
Olof Johansson [Fri, 19 May 2017 07:02:16 +0000 (00:02 -0700)]
Merge tag 'omap-for-v4.12/fixes-v2-signed' of git://git./linux/kernel/git/tmlind/linux-omap into fixes

Fixes for omaps for v4.12-rc cycle most consisting of few minor dts fixes
for various devices. Also included is a memory controller (GPMC) debug output
fix as without that the shown bootloader configured GPMC bus width will
be wrong and won't work for kernel timings:

- Add dra7 powerhold configuration to be able to shut down pmic correctly
- Fix polarity for gta04 mcbsp4 clocks for modem
- Fix Pandaboard CEC pin pull making it usable
- Fix LogicPD Torpedo camera pin mux
- Fix GPMC debug bus width
- Reduce cpu thermal shutdown temperature

* tag 'omap-for-v4.12/fixes-v2-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
  ARM: dts: dra7: Reduce cpu thermal shutdown temperature
  memory: omap-gpmc: Fix debug output for access width
  ARM: dts: LogicPD Torpedo: Fix camera pin mux
  ARM: dts: omap4: enable CEC pin for Pandaboard A4 and ES
  ARM: dts: gta04: fix polarity of clocks for mcbsp4
  ARM: dts: dra7: Add power hold and power controller properties to palmas

Signed-off-by: Olof Johansson <olof@lixom.net>
7 years agoMerge tag 'imx-fixes-4.12' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo...
Olof Johansson [Fri, 19 May 2017 07:02:04 +0000 (00:02 -0700)]
Merge tag 'imx-fixes-4.12' of git://git./linux/kernel/git/shawnguo/linux into fixes

i.MX fixes for 4.12:
 - A fix on GPCv2 power domain driver Kconfig which causes a build
   failure when CONFIG_PM is not set.
 - Pull down PMIC IRQ pin for imx53-qsrb board to prevent spurious
   PMIC interrupts from happening.
 - Remove board level OPP override for imx6sx-sdb to fix a boot crash
   seen on Rev.C boards.

* tag 'imx-fixes-4.12' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:
  soc: imx: add PM dependency for IMX7_PM_DOMAINS
  ARM: dts: imx6sx-sdb: Remove OPP override
  ARM: dts: imx53-qsrb: Pulldown PMIC IRQ pin

Signed-off-by: Olof Johansson <olof@lixom.net>