Bluetooth: btmtksdio: Add the missed release_firmware() in mtk_setup_firmware()

[ Upstream commit b73b5781a85c03113476f62346c390f0277baa4b ]

mtk_setup_firmware() misses to call release_firmware() in an error
path. Jump to free_fw to fix it.

Fixes: 737cd06072a7 ("Bluetooth: btmtksdio: fix up firmware download sequence")
Signed-off-by: Jing Xiangfeng <jingxiangfeng@huawei.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

Bluetooth: btusb: Add the missed release_firmware() in btusb_mtk_setup_firmware()

[ Upstream commit d1e9d232e1e60fa63df1b836ec3ecba5abd3fa9d ]

btusb_mtk_setup_firmware() misses to call release_firmware() in an error
path. Jump to err_release_fw to fix it.

Fixes: f645125711c8 ("Bluetooth: btusb: fix up firmware download sequence")
Signed-off-by: Jing Xiangfeng <jingxiangfeng@huawei.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

spi: dw: Fix error return code in dw_spi_bt1_probe()

[ Upstream commit e748edd9841306908b4e02dddd0afd1aa1f8b973 ]

Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.

Fixes: abf00907538e ("spi: dw: Add Baikal-T1 SPI Controller glue driver")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhang Changzhong <zhangchangzhong@huawei.com>
Acked-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/1607071357-33378-1-git-send-email-zhangchangzhong@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

staging: greybus: audio: Fix possible leak free widgets in gbaudio_dapm_free_controls

[ Upstream commit e77b259f67ab99f1e22ce895b9b1c637fd5f2d4c ]

In gbaudio_dapm_free_controls(), if one of the widgets is not found, an error
will be returned directly, which will cause the rest to be unable to be freed,
resulting in leak.

This patch fixes the bug. If if one of them is not found, just skip and free the others.

Fixes: 510e340efe0c ("staging: greybus: audio: Add helper APIs for dynamic audio module")
Reported-by: Hulk Robot <hulkci@huawei.com>
Reviewed-by: Vaibhav Agarwal <vaibhav.sr@gmail.com>
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Link: https://lore.kernel.org/r/20201205103827.31244-1-wanghai38@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

staging: bcm2835: fix vchiq_mmal dependencies

[ Upstream commit 6c91799f59ed491a2b5bf5ef7c6d60306cda4e09 ]

When the MMAL code is built-in but the vchiq core config is
set to =m, the mmal code never gets built, which in turn can
lead to link errors:

ERROR: modpost: "vchiq_mmal_port_set_format" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_port_disable" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_port_parameter_set" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_component_finalise" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_port_connect_tunnel" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_component_enable" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_finalise" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_component_init" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_component_disable" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "mmal_vchi_buffer_init" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_port_enable" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_version" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_submit_buffer" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_init" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "mmal_vchi_buffer_cleanup" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!
ERROR: modpost: "vchiq_mmal_port_parameter_get" [drivers/staging/vc04_services/bcm2835-camera/bcm2835-v4l2.ko] undefined!

Change the Kconfig to depend on BCM2835_VCHIQ like the other drivers,
and remove the now redundant dependencies.

Fixes: b18ee53ad297 ("staging: bcm2835: Break MMAL support out from camera")
Acked-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20201203223836.1362313-1-arnd@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

macintosh/adb-iop: Send correct poll command

[ Upstream commit 10199e90ee20e68859f8128331ec8d85b036d349 ]

The behaviour of the IOP firmware is not well documented but we do know
that IOP message reply data can be used to issue new ADB commands.
Use the message reply to better control autopoll behaviour by sending
a Talk Register 0 command after every ADB response, not unlike the
algorithm in the via-macii driver. This poll command is addressed to
that device which last received a Talk command (explicit or otherwise).

Cc: Joshua Thompson <funaho@jurai.org>
Fixes: 32226e817043 ("macintosh/adb-iop: Implement idle -> sending state transition")
Tested-by: Stan Johnson <userm57@yahoo.com>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Link: https://lore.kernel.org/r/58bba4310da4c29b068345a4b36af8a531397ff7.1605847196.git.fthain@telegraphics.com.au
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

macintosh/adb-iop: Always wait for reply message from IOP

[ Upstream commit 2c9cfbadfa234b03473f1ef54e6f4772cc07a371 ]

A recent patch incorrectly altered the adb-iop state machine behaviour
and introduced a regression that can appear intermittently as a
malfunctioning ADB input device. This seems to be caused when reply
packets from different ADB commands become mixed up, especially during
the adb bus scan. Fix this by unconditionally entering the awaiting_reply
state after sending an explicit command, even when the ADB command won't
generate a reply from the ADB device.

It turns out that the IOP always generates reply messages, even when the
ADB command does not produce a reply packet (e.g. ADB Listen command).
So it's not really the ADB reply packets that are being mixed up, it's the
IOP messages that enclose them. The bug goes like this:

  1. CPU sends a message to the IOP, expecting no response because this
     message contains an ADB Listen command. The ADB command is now
     considered complete.

  2. CPU sends a second message to the IOP, this time expecting a
     response because this message contains an ADB Talk command. This
     ADB command needs a reply before it can be completed.

  3. adb-iop driver receives an IOP message and assumes that it relates
     to the Talk command. It's actually an empty one (with flags ==
     ADB_IOP_EXPLICIT|ADB_IOP_TIMEOUT) for the previous command. The
     Talk command is now considered complete but it gets the wrong reply
     data.

  4. adb-iop driver gets another IOP response message, which contains
     the actual reply data for the Talk command, but this is dropped
     (the driver is no longer in awaiting_reply state).

Cc: Joshua Thompson <funaho@jurai.org>
Fixes: e2954e5f727f ("macintosh/adb-iop: Implement sending -> idle state transition")
Tested-by: Stan Johnson <userm57@yahoo.com>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
Link: https://lore.kernel.org/r/0f0a25855391e7eaa53a50f651aea0124e8525dd.1605847196.git.fthain@telegraphics.com.au
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: imx: fix NVMEM_IMX_OCOTP dependency

[ Upstream commit fc928b901dc68481ba3e524860a641fe13e25dfe ]

A driver should not 'select' drivers from another subsystem.
If NVMEM is disabled, this one results in a warning:

WARNING: unmet direct dependencies detected for NVMEM_IMX_OCOTP
  Depends on [n]: NVMEM [=n] && (ARCH_MXC [=y] || COMPILE_TEST [=y]) && HAS_IOMEM [=y]
  Selected by [y]:
  - ARM_IMX6Q_CPUFREQ [=y] && CPU_FREQ [=y] && (ARM || ARM64 [=y]) && ARCH_MXC [=y] && REGULATOR_ANATOP [=y]

Change the 'select' to 'depends on' to prevent it from going wrong,
and allow compile-testing without that driver, since it is only
a runtime dependency.

Fixes: 2782ef34ed23 ("cpufreq: imx: Select NVMEM_IMX_OCOTP")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: vexpress-spc: Add missing MODULE_ALIAS

[ Upstream commit d15183991c2d53d7cecf27a1555c91b702cef1ea ]

This patch adds missing MODULE_ALIAS for automatic loading of this cpufreq
driver when it is compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 47ac9aa165540 ("cpufreq: arm_big_little: add vexpress SPC interface driver")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: scpi: Add missing MODULE_ALIAS

[ Upstream commit c0382d049d2def37b81e907a8b22661a4a4a6eb5 ]

This patch adds missing MODULE_ALIAS for automatic loading of this cpufreq
driver when it is compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 8def31034d033 ("cpufreq: arm_big_little: add SCPI interface driver")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: loongson1: Add missing MODULE_ALIAS

[ Upstream commit b9acab091842ca8b288882798bb809f7abf5408a ]

This patch adds missing MODULE_ALIAS for automatic loading of this cpufreq
driver when it is compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: a0a22cf14472f ("cpufreq: Loongson1: Add cpufreq driver for Loongson1B")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: sun50i: Add missing MODULE_DEVICE_TABLE

[ Upstream commit af2096f285077e3339eb835ad06c50bdd59f01b5 ]

This patch adds missing MODULE_DEVICE_TABLE definition which generates
correct modalias for automatic loading of this cpufreq driver when it is
compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: f328584f7bff8 ("cpufreq: Add sun50i nvmem based CPU scaling driver")
Reviewed-by: Yangtao Li <tiny.windzz@gmail.com>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: st: Add missing MODULE_DEVICE_TABLE

[ Upstream commit 183747ab52654eb406fc6b5bfb40806b75d31811 ]

This patch adds missing MODULE_DEVICE_TABLE definition which generates
correct modalias for automatic loading of this cpufreq driver when it is
compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: ab0ea257fc58d ("cpufreq: st: Provide runtime initialised driver for ST's platforms")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: qcom: Add missing MODULE_DEVICE_TABLE

[ Upstream commit a5a6031663bc1dd0a10babd49d1bcb3153a8327f ]

This patch adds missing MODULE_DEVICE_TABLE definition which generates
correct modalias for automatic loading of this cpufreq driver when it is
compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 46e2856b8e188 ("cpufreq: Add Kryo CPU scaling driver")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: mediatek: Add missing MODULE_DEVICE_TABLE

[ Upstream commit af6eca06501118af3e2ad46eee8edab20624b74e ]

This patch adds missing MODULE_DEVICE_TABLE definition which generates
correct modalias for automatic loading of this cpufreq driver when it is
compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 501c574f4e3a5 ("cpufreq: mediatek: Add support of cpufreq to MT2701/MT7623 SoC")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: highbank: Add missing MODULE_DEVICE_TABLE

[ Upstream commit 9433777a6e0aae27468d3434b75cd51bb88ff711 ]

This patch adds missing MODULE_DEVICE_TABLE definition which generates
correct modalias for automatic loading of this cpufreq driver when it is
compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 6754f556103be ("cpufreq / highbank: add support for highbank cpufreq")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

cpufreq: ap806: Add missing MODULE_DEVICE_TABLE

[ Upstream commit 925a5bcefe105f2790ecbdc252eb2315573f309d ]

This patch adds missing MODULE_DEVICE_TABLE definition which generates
correct modalias for automatic loading of this cpufreq driver when it is
compiled as an external module.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: f525a670533d9 ("cpufreq: ap806: add cpufreq driver for Armada 8K")
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

clocksource/drivers/arm_arch_timer: Correct fault programming of CNTKCTL_EL1.EVNTI

[ Upstream commit 8b7770b877d187bfdae1eaf587bd2b792479a31c ]

ARM virtual counter supports event stream, it can only trigger an event
when the trigger bit (the value of CNTKCTL_EL1.EVNTI) of CNTVCT_EL0 changes,
so the actual period of event stream is 2^(cntkctl_evnti + 1). For example,
when the trigger bit is 0, then virtual counter trigger an event for every
two cycles.

While we're at it, rework the way we compute the trigger bit position
by making it more obvious that when bits [n:n-1] are both set (with n
being the most significant bit), we pick bit (n + 1).

Fixes: 037f637767a8 ("drivers: clocksource: add support for ARM architected timer event stream")
Suggested-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com>
Acked-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: https://lore.kernel.org/r/20201204073126.6920-3-zhukeqian1@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clocksource/drivers/arm_arch_timer: Use stable count reader in erratum sne

[ Upstream commit d8cc3905b8073c7cfbff94af889fa8dc71f21dd5 ]

In commit 0ea415390cd3 ("clocksource/arm_arch_timer: Use arch_timer_read_counter
to access stable counters"), we separate stable and normal count reader to omit
unnecessary overhead on systems that have no timer erratum.

However, in erratum_set_next_event_tval_generic(), count reader becomes normal
reader. This converts it to stable reader.

Fixes: 0ea415390cd3 ("clocksource/arm_arch_timer: Use arch_timer_read_counter to access stable counters")
Acked-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: https://lore.kernel.org/r/20201204073126.6920-2-zhukeqian1@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/msm: add IOMMU_SUPPORT dependency

[ Upstream commit e319a1b956f785f618611857cd946dca2bb68542 ]

The iommu pgtable support is only available when IOMMU support
is built into the kernel:

WARNING: unmet direct dependencies detected for IOMMU_IO_PGTABLE
  Depends on [n]: IOMMU_SUPPORT [=n]
  Selected by [y]:
  - DRM_MSM [=y] && HAS_IOMEM [=y] && DRM [=y] && (ARCH_QCOM [=y] || SOC_IMX5 || ARM && COMPILE_TEST [=y]) && OF [=y] && COMMON_CLK [=y] && MMU [=y] && (QCOM_OCMEM [=y] || QCOM_OCMEM [=y]=n)

Fix the dependency accordingly. There is no need for depending on
CONFIG_MMU any more, as that is implied by the iommu support.

Fixes: b145c6e65eb0 ("drm/msm: Add support to create a local pagetable")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

drm/msm: a5xx: Make preemption reset case reentrant

[ Upstream commit 7cc29fcdfcc8784e97c5151c848e193800ec79ac ]

nr_rings is reset to 1, but when this function is called for a second
(and third!) time nr_rings > 1 is false, thus the else case is entered
to set up a buffer for the RPTR shadow and consequently written to
RB_RPTR_ADDR, hanging platforms without WHERE_AM_I firmware support.

Restructure the condition in such a way that shadow buffer setup only
ever happens when has_whereami is true; otherwise preemption is only
finalized when the number of ring buffers has not been reset to 1 yet.

Fixes: 8907afb476ac ("drm/msm: Allow a5xx to mark the RPTR shadow as privileged")
Signed-off-by: Marijn Suijten <marijn.suijten@somainline.org>
Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Reviewed-by: Jordan Crouse <jcrouse@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@chromium.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

memory: jz4780_nemc: Fix potential NULL dereference in jz4780_nemc_probe()

[ Upstream commit 4bfa07300b9334b487ed4f3d4901c35ebb31b7ca ]

platform_get_resource() may fail and return NULL, so we should
better check it's return value to avoid a NULL pointer dereference
a bit later in the code.

This is detected by Coccinelle semantic patch.

Fixes: 911a88829725 ("memory: jz4780-nemc: driver for the NEMC on JZ4780 SoCs")
Signed-off-by: Zhang Changzhong <zhangchangzhong@huawei.com>
Acked-by: Paul Cercueil <paul@crapouillou.net>
Link: https://lore.kernel.org/r/1607070717-32880-1-git-send-email-zhangchangzhong@huawei.com
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

memory: ti-emif-sram: only build for ARMv7

[ Upstream commit d77d22d701b0471584abe1871570bb43deb6e3c4 ]

The driver can be compile-tested on all ARM machines, but
causes a failure when built for ARMv7-M:

arm-linux-gnueabi-ld: error: drivers/memory/ti-emif-sram-pm.o: conflicting architecture profiles A/M

Limit the target machines to configurations that have ARMv7 enabled.

Fixes: ea0c0ad6b6eb ("memory: Enable compile testing for most of the drivers")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20201203230832.1481767-1-arnd@kernel.org
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

phy: renesas: rcar-gen3-usb2: disable runtime pm in case of failure

[ Upstream commit 51e339deab1e51443f6ac3b1bd5cd6cc8e8fe1d9 ]

pm_runtime_enable() will decrease power disable depth. Thus a pairing
increment is needed on the error handling path to keep it balanced.

Fixes: 5d8042e95fd4 ("phy: rcar-gen3-usb2: Add support for r8a77470")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wang Li <wangli74@huawei.com>
Link: https://lore.kernel.org/r/20201126024412.4046845-1-wangli74@huawei.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

phy: mediatek: allow compile-testing the hdmi phy

[ Upstream commit f5f6e01f9164040ba120f30522efdb4deceb529a ]

Compile-testing the DRM_MEDIATEK_HDMI driver shows two missing
dependencies, one results in a link failure:

arm-linux-gnueabi-ld: drivers/phy/mediatek/phy-mtk-hdmi.o: in function `mtk_hdmi_phy_probe':
phy-mtk-hdmi.c:(.text+0xd8): undefined reference to `__clk_get_name'
arm-linux-gnueabi-ld: phy-mtk-hdmi.c:(.text+0x12c): undefined reference to `devm_clk_register'
arm-linux-gnueabi-ld: phy-mtk-hdmi.c:(.text+0x250): undefined reference to `of_clk_add_provider'
arm-linux-gnueabi-ld: phy-mtk-hdmi.c:(.text+0x298): undefined reference to `of_clk_src_simple_get'

The other one is a harmless warning:

WARNING: unmet direct dependencies detected for PHY_MTK_HDMI
  Depends on [n]: ARCH_MEDIATEK [=n] && OF [=y]
  Selected by [y]:
  - DRM_MEDIATEK_HDMI [=y] && HAS_IOMEM [=y] && DRM_MEDIATEK [=y]

Fix these by adding dependencies on CONFIG_OF and CONFIG_COMMON_CLK.
With that done, there is also no reason against adding
CONFIG_COMPILE_TEST.

Fixes: b28be59a2e26 ("phy: mediatek: Move mtk_hdmi_phy driver into drivers/phy/mediatek folder")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20201204135650.2744481-1-arnd@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: qcom: fix QDSP6 dependencies, attempt #3

[ Upstream commit b1b8eb1283c90a953089d988930d7b6156418958 ]

The previous fix left another warning in randconfig builds:

WARNING: unmet direct dependencies detected for SND_SOC_QDSP6
  Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_QCOM [=y] && QCOM_APR [=y] && COMMON_CLK [=n]
  Selected by [y]:
  - SND_SOC_MSM8996 [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_SOC_QCOM [=y] && QCOM_APR [=y]

Add one more dependency for this one.

Fixes: 2bc8831b135c ("ASoC: qcom: fix SDM845 & QDSP6 dependencies more")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20201203231443.1483763-1-arnd@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: atmel: mchp-spdifrx needs COMMON_CLK

[ Upstream commit 29275309b0e32bb838d67158c6b6e687275f43e9 ]

Compile-testing this driver on an older platform without CONFIG_COMMON_CLK fails with

ERROR: modpost: "clk_set_min_rate" [sound/soc/atmel/snd-soc-mchp-spdifrx.ko] undefined!

Make this is a strict dependency.

Fixes: ef265c55c1ac ("ASoC: mchp-spdifrx: add driver for SPDIF RX")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20201203223815.1353451-1-arnd@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: cros_ec_codec: fix uninitialized memory read

[ Upstream commit 7061b8a52296e044eed47b605d136a48da1a7761 ]

gcc points out a memory area that is copied to a device
but not initialized:

sound/soc/codecs/cros_ec_codec.c: In function 'i2s_rx_event':
arch/x86/include/asm/string_32.h:83:20: error: '*((void *)&p+4)' may be used uninitialized in this function [-Werror=maybe-uninitialized]
   83 |   *((int *)to + 1) = *((int *)from + 1);

Initialize all the unused fields to zero.

Fixes: 727f1c71c780 ("ASoC: cros_ec_codec: refactor I2S RX")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Tzung-Bi Shih <tzungbi@google.com>
Link: https://lore.kernel.org/r/20201203225458.1477830-1-arnd@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

dm ioctl: fix error return code in target_message

[ Upstream commit 4d7659bfbe277a43399a4a2d90fca141e70f29e1 ]

Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.

Fixes: 2ca4c92f58f9 ("dm ioctl: prevent empty message")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Qinglang Miao <miaoqinglang@huawei.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: q6afe-clocks: Add missing parent clock rate

[ Upstream commit 7e20ae1208daaf6dad85c2dcb968fc590b6f3b99 ]

setting clock rate on child clocks without a parent clock rate will
result in zero clk rate for child. This also means that when audio
is started dsp will attempt to access registers without enabling
clock resulting in board boot up.

Fix this by adding the missing parent clock rate.

Fixes: 520a1c396d196 ("ASoC: q6afe-clocks: add q6afe clock controller")
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20201204164228.1826-1-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: jz4740-i2s: add missed checks for clk_get()

[ Upstream commit 1c1fb2653a0c2e3f310c07eacd8fc3a10e08c97a ]

jz4740_i2s_set_sysclk() does not check the return values of clk_get(),
while the file dereferences the pointers in clk_put().
Add the missed checks to fix it.

Fixes: 11bd3dd1b7c2 ("ASoC: Add JZ4740 ASoC support")
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Link: https://lore.kernel.org/r/20201203144227.418194-1-hslester96@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: fix tkip configuration for mt7615/7663 devices

[ Upstream commit 930e0eaddf810cfa90e114a0df02f48539e1346f ]

Fix Tx-Rx MIC overwrite during TKIP hw key configuration

Fixes: 01cfc1b45421 ("mt76: mt7615: add BIP_CMAC_128 cipher support")
Tested-by: YN Chen <yn.chen@mediatek.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: fix memory leak if device probing fails

[ Upstream commit bc348defcc6eceeb4f7784bf9a263ddb72fd3fb4 ]

Run mt76_free_device instead of ieee80211_free_hw if device probing
fails in order to remove the already allocated mt76 workqueue

Fixes: a86f1d01f5ce5 ("mt76: move mt76 workqueue in common code")
Fixes: f1d962369d568 ("mt76: mt7915: implement HE per-rate tx power support")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net/mlx5: Properly convey driver version to firmware

[ Upstream commit 907af0f0cab4ee5d5604f182ecec2c5b5119d294 ]

mlx5 firmware expects driver version in specific format X.X.X, so
make it always correct and based on real kernel version aligned with
the driver.

Fixes: 012e50e109fd ("net/mlx5: Set driver version into firmware")
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: dma: fix possible deadlock running mt76_dma_cleanup

[ Upstream commit 5f0ce5848733b86deaba1c6d102e20479230b20a ]

Fix the following possible deadlock reported by lockdep disabling BH
running mt76_free_pending_txwi()

================================
WARNING: inconsistent lock state
5.9.0-rc6 #14 Not tainted
--------------------------------
inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.
rmmod/1227 [HC0[0]:SC0[0]:HE1:SE1] takes:
ffff888156a83530 (&dev->lock#2){+.?.}-{2:2}, at: mt76_dma_cleanup+0x125/0x150 [mt76]
{IN-SOFTIRQ-W} state was registered at:
  __lock_acquire+0x20c/0x6b0
  lock_acquire+0x9d/0x220
  _raw_spin_lock+0x2c/0x70
  mt76_dma_tx_cleanup+0xc7/0x200 [mt76]
  mt76x02_poll_tx+0x31/0xb0 [mt76x02_lib]
  napi_poll+0x3a/0x100
  net_rx_action+0xa8/0x200
  __do_softirq+0xc4/0x430
  asm_call_on_stack+0xf/0x20
  do_softirq_own_stack+0x49/0x60
  irq_exit_rcu+0x9a/0xd0
  common_interrupt+0xa4/0x190
  asm_common_interrupt+0x1e/0x40
irq event stamp: 9915
hardirqs last  enabled at (9915): [<ffffffff8124e286>] __free_pages_ok+0x336/0x3b0
hardirqs last disabled at (9914): [<ffffffff8124e24e>] __free_pages_ok+0x2fe/0x3b0
softirqs last  enabled at (9912): [<ffffffffa03aa672>] mt76_dma_rx_cleanup+0xa2/0x120 [mt76]
softirqs last disabled at (9846): [<ffffffffa03aa5ea>] mt76_dma_rx_cleanup+0x1a/0x120 [mt76]

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(&dev->lock#2);
  <Interrupt>
    lock(&dev->lock#2);

 *** DEADLOCK ***

1 lock held by rmmod/1227:
 #0: ffff88815b5eb240 (&dev->mutex){....}-{3:3}, at: driver_detach+0xb5/0x110

stack backtrace:
CPU: 1 PID: 1227 Comm: rmmod Kdump: loaded Not tainted 5.9.0-rc6-wdn-src+ #14
Hardware name: Dell Inc. Studio XPS 1340/0K183D, BIOS A11 09/08/2009
Call Trace:
 dump_stack+0x77/0xa0
 mark_lock_irq.cold+0x15/0x39
 mark_lock+0x1fc/0x500
 mark_usage+0xc7/0x140
 __lock_acquire+0x20c/0x6b0
 ? find_held_lock+0x2b/0x80
 ? sched_clock_cpu+0xc/0xb0
 lock_acquire+0x9d/0x220
 ? mt76_dma_cleanup+0x125/0x150 [mt76]
 _raw_spin_lock+0x2c/0x70
 ? mt76_dma_cleanup+0x125/0x150 [mt76]
 mt76_dma_cleanup+0x125/0x150 [mt76]
 mt76x2_cleanup+0x5a/0x70 [mt76x2e]
 mt76x2e_remove+0x18/0x30 [mt76x2e]
 pci_device_remove+0x36/0xa0
 __device_release_driver+0x16c/0x220
 driver_detach+0xcf/0x110
 bus_remove_driver+0x56/0xca
 pci_unregister_driver+0x36/0x80
 __do_sys_delete_module.constprop.0+0x127/0x200
 ? syscall_enter_from_user_mode+0x1d/0x50
 ? trace_hardirqs_on+0x1c/0xe0
 do_syscall_64+0x33/0x80
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x7ff0da54e36b
Code: 73 01 c3 48 8b 0d 2d 0b 0c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa b8 b0 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d fd 0a 0c 00 f7 d8 64 89 01 48

Fixes: dd57a95cfddc ("mt76: move txwi handling code to dma.c, since it is mmio specific")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: set fops_tx_stats.owner to THIS_MODULE

[ Upstream commit f9df085ce1be5c599e4df590ff7ba853786c6d95 ]

If THIS_MODULE is not set, the module would be removed while debugfs is
being used.
It eventually makes kernel panic.

Fixes: e57b7901469f ("mt76: add mac80211 driver for MT7915 PCIe-based chipsets")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: mt7915: set fops_sta_stats.owner to THIS_MODULE

[ Upstream commit 5efbe3b1b8992d5f837388091920945c23212159 ]

If THIS_MODULE is not set, the module would be removed while debugfs is
being used.
It eventually makes kernel panic.

Fixes: ec9742a8f38e ("mt76: mt7915: add .sta_add_debugfs support")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mt76: mt7663s: fix a possible ple quota underflow

[ Upstream commit 1c79a190e94325e01811f653f770a34e816fdd8f ]

Properly account current consumed ple quota in mt7663s_tx_pick_quota
routine and avoid possible underflow.

Fixes: 6ef2d665f64d ("mt76: mt7663s: split mt7663s_tx_update_sched in mt7663s_tx_{pick,update}_quota")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Sasha Levin <sashal@kernel.org>

MIPS: Don't round up kernel sections size for memblock_add()

[ Upstream commit d121f125af22a16f0f679293756d28a9691fa46d ]

Linux doesn't own the memory immediately after the kernel image. On Octeon
bootloader places a shared structure right close after the kernel _end,
refer to "struct cvmx_bootinfo *octeon_bootinfo" in cavium-octeon/setup.c.

If check_kernel_sections_mem() rounds the PFNs up, first memblock_alloc()
inside early_init_dt_alloc_memory_arch() <= device_tree_init() returns
memory block overlapping with the above octeon_bootinfo structure, which
is being overwritten afterwards.

Fixes: a94e4f24ec83 ("MIPS: init: Drop boot_mem_map")
Signed-off-by: Alexander Sverdlin <alexander.sverdlin@nokia.com>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

memstick: r592: Fix error return in r592_probe()

[ Upstream commit db29d3d1c2451e673e29c7257471e3ce9d50383a ]

Fix to return a error code from the error handling case instead of 0.

Fixes: 926341250102 ("memstick: add driver for Ricoh R5C592 card reader")
Signed-off-by: Jing Xiangfeng <jingxiangfeng@huawei.com>
Link: https://lore.kernel.org/r/20201125014718.153563-1-jingxiangfeng@huawei.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

arm64: dts: rockchip: Fix UART pull-ups on rk3328

[ Upstream commit 94dad6bed3c86c00050bf7c2b2ad6b630facae31 ]

For UARTs, the local pull-ups should be on the RX pin, not the TX pin.
UARTs transmit active-low, so a disconnected RX pin should be pulled
high instead of left floating to prevent noise being interpreted as
transmissions.

This gets rid of bogus sysrq events when the UART console is not
connected.

Fixes: 52e02d377a72 ("arm64: dts: rockchip: add core dtsi file for RK3328 SoCs")
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Link: https://lore.kernel.org/r/20201204064805.6480-1-wens@kernel.org
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

soc: rockchip: io-domain: Fix error return code in rockchip_iodomain_probe()

[ Upstream commit c2867b2e710fc85bb39c6f6e5948450c48e8a33e ]

Fix to return a negative error code from the error handling
case instead of 0, as done elsewhere in this function.

Fixes: e943c43b32ce ("PM: AVS: rockchip-io: Move the driver to the rockchip specific drivers")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhang Changzhong <zhangchangzhong@huawei.com>
Link: https://lore.kernel.org/r/1607070805-33038-1-git-send-email-zhangchangzhong@huawei.com
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>

pinctrl: falcon: add missing put_device() call in pinctrl_falcon_probe()

[ Upstream commit 89cce2b3f247a434ee174ab6803698041df98014 ]

if of_find_device_by_node() succeed, pinctrl_falcon_probe() doesn't have
a corresponding put_device(). Thus add put_device() to fix the exception
handling for this function implementation.

Fixes: e316cb2b16bb ("OF: pinctrl: MIPS: lantiq: adds support for FALCON SoC")
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Link: https://lore.kernel.org/r/20201119011219.2248232-1-yukuai3@huawei.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests/bpf: Fix invalid use of strncat in test_sockmap

[ Upstream commit eceae70bdeaeb6b8ceb662983cf663ff352fbc96 ]

strncat()'s third argument is how many bytes will be added *in addition* to
already existing bytes in destination. Plus extra zero byte will be added
after that. So existing use in test_sockmap has many opportunities to overflow
the string and cause memory corruptions. And in this case, GCC complains for
a good reason.

Fixes: 16962b2404ac ("bpf: sockmap, add selftests")
Fixes: 73563aa3d977 ("selftests/bpf: test_sockmap, print additional test options")
Fixes: 1ade9abadfca ("bpf: test_sockmap, add options for msg_pop_data() helper")
Fixes: 463bac5f1ca7 ("bpf, selftests: Add test for ktls with skb bpf ingress policy")
Fixes: e9dd904708c4 ("bpf: add tls support for testing in test_sockmap")
Fixes: 753fb2ee0934 ("bpf: sockmap, add msg_peek tests to test_sockmap")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201203235440.2302137-2-andrii@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

bpf: Fix bpf_put_raw_tracepoint()'s use of __module_address()

[ Upstream commit 12cc126df82c96c89706aa207ad27c56f219047c ]

__module_address() needs to be called with preemption disabled or with
module_mutex taken. preempt_disable() is enough for read-only uses, which is
what this fix does. Also, module_put() does internal check for NULL, so drop
it as well.

Fixes: a38d1107f937 ("bpf: support raw tracepoints in modules")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20201203204634.1325171-2-andrii@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

scripts: kernel-doc: fix parsing function-like typedefs

[ Upstream commit 7d2c6b1edf790d96e9017a0b27be2425e1af1532 ]

Changeset 6b80975c6308 ("scripts: kernel-doc: fix typedef parsing")
added support for things like:

typedef unsigned long foo();

However, it caused a regression on this prototype:

typedef bool v4l2_check_dv_timings_fnc(const struct v4l2_dv_timings *t, void *handle);

This is only noticed after adding a patch that checks if the
kernel-doc identifier matches the typedef:

./scripts/kernel-doc -none $(git grep '^.. kernel-doc::' Documentation/ |cut -d ' ' -f 3|sort|uniq) 2>&1|grep expecting
include/media/v4l2-dv-timings.h:38: warning: expecting prototype for typedef v4l2_check_dv_timings_fnc. Prototype was for typedef nc instead

The problem is that, with the new parsing logic, it is not
checking for complete words at the type part.

Fix it by adding a \b at the end of each type word at the
regex.

fixes: 6b80975c6308 ("scripts: kernel-doc: fix typedef parsing")
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/218ff56dcb8e73755005d3fb64586eb1841a276b.1606896997.git.mchehab+huawei@kernel.org
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ARM: dts: at91: sama5d2: map securam as device

[ Upstream commit 9b5dcc8d427e2bcb84c49eb03ffefe11e7537a55 ]

Due to strobe signal not being propagated from CPU to securam
the securam needs to be mapped as device or strongly ordered memory
to work properly. Otherwise, updating to one offset may affect
the adjacent locations in securam.

Fixes: d4ce5f44d4409 ("ARM: dts: at91: sama5d2: Add securam node")
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/1606903025-14197-3-git-send-email-claudiu.beznea@microchip.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

ARM: dts: at91: sam9x60ek: remove bypass property

[ Upstream commit ab8a9bb41b2c330a0b280280bf37b6f3b1dd1e58 ]

atmel,osc-bypass property sets the bit 1 at main oscillator register.
On SAM9X60 this bit is not valid according to datasheet (chapter
28.16.9 PMC Clock Generator Main Oscillator Register).

Fixes: 1e5f532c2737 ("ARM: dts: at91: sam9x60: add device tree for soc and board")
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Cc: Marco Cardellini <marco.cardellini@microchip.com>
Link: https://lore.kernel.org/r/1606903025-14197-2-git-send-email-claudiu.beznea@microchip.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

libbpf: Sanitise map names before pinning

[ Upstream commit 9cf309c56f7910a81fbe053b6f11c3b1f0987b12 ]

When we added sanitising of map names before loading programs to libbpf, we
still allowed periods in the name. While the kernel will accept these for
the map names themselves, they are not allowed in file names when pinning
maps. This means that bpf_object__pin_maps() will fail if called on an
object that contains internal maps (such as sections .rodata).

Fix this by replacing periods with underscores when constructing map pin
paths. This only affects the paths generated by libbpf when
bpf_object__pin_maps() is called with a path argument. Any pin paths set
by bpf_map__set_pin_path() are unaffected, and it will still be up to the
caller to avoid invalid characters in those.

Fixes: 113e6b7e15e2 ("libbpf: Sanitise internal map names so they are not rejected by the kernel")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20201203093306.107676-1-toke@redhat.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

iio: hrtimer-trigger: Mark hrtimer to expire in hard interrupt context

[ Upstream commit 0178297c1e6898e2197fe169ef3be723e019b971 ]

On PREEMPT_RT enabled kernels unmarked hrtimers are moved into soft
interrupt expiry mode by default.

The IIO hrtimer-trigger needs to run in hard interrupt context since it
will end up calling generic_handle_irq() which has the requirement to run
in hard interrupt context.

Explicitly specify that the timer needs to run in hard interrupt context by
using the HRTIMER_MODE_REL_HARD flag.

Fixes: f5c2f0215e36 ("hrtimer: Move unmarked hrtimers to soft interrupt expiry on RT")
Reported-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/20201117103751.16131-1-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

arm64: mte: fix prctl(PR_GET_TAGGED_ADDR_CTRL) if TCF0=NONE

[ Upstream commit 929c1f3384d7e5cd319d03242cb925c3f91236f7 ]

Previously we were always returning a tag inclusion mask of zero via
PR_GET_TAGGED_ADDR_CTRL if TCF0 was set to NONE. Fix it by making
the code for the NONE case match the others.

Signed-off-by: Peter Collingbourne <pcc@google.com>
Link: https://linux-review.googlesource.com/id/Iefbea66cf7d2b4c80b82f9639b9ea7f33f7fac53
Fixes: af5ce95282dc ("arm64: mte: Allow user control of the generated random tags via prctl()")
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Link: https://lore.kernel.org/r/20201203075110.2781021-1-pcc@google.com
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

clocksource/drivers/riscv: Make RISCV_TIMER depends on RISCV_SBI

[ Upstream commit ab3105446f1ec4e98fadfc998ee24feec271c16c ]

The riscv timer is set via SBI timer call, let's make RISCV_TIMER
depends on RISCV_SBI, and it also fixes some build issue.

Fixes: d5be89a8d118 ("RISC-V: Resurrect the MMIO timer implementation for M-mode systems")
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Acked-by: Palmer Dabbelt <palmerdabbelt@google.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: https://lore.kernel.org/r/20201028131230.72907-1-wangkefeng.wang@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clocksource/drivers/ingenic: Fix section mismatch

[ Upstream commit 5bd7cb29eceb52e4b108917786fdbf2a2c2048ef ]

The function ingenic_tcu_get_clock() is annotated for the __init
section but it is actually called from the online cpu callback.

That will lead to a crash if a CPU is hotplugged after boot time.

Remove the __init annotation for the ingenic_tcu_get_clock()
function.

Fixes: f19d838d08fc (clocksource/drivers/ingenic: Add high resolution timer support for SMP/SMT)
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Reviewed-by: Paul Cercueil <paul@crapouillou.net>
Tested-by: 周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Link: https://lore.kernel.org/r/20201125102346.1816310-1-daniel.lezcano@linaro.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

clocksource/drivers/cadence_ttc: Fix memory leak in ttc_setup_clockevent()

[ Upstream commit eee422c46e6840a81c9db18a497b74387a557b29 ]

If clk_notifier_register() failed, ttc_setup_clockevent() will return
without freeing 'ttcce', which will leak memory.

Fixes: 70504f311d4b ("clocksource/drivers/cadence_ttc: Convert init function to return error")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: https://lore.kernel.org/r/20201116135123.2164033-1-yukuai3@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

clocksource/drivers/orion: Add missing clk_disable_unprepare() on error path

[ Upstream commit c1e6cad00aa2f17845e7270e38ff3cc82c7b022a ]

After calling clk_prepare_enable(), clk_disable_unprepare() need
be called on error path.

Fixes: fbe4b3566ddc ("clocksource/drivers/orion: Convert init function...")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Link: https://lore.kernel.org/r/20201111064706.3397156-1-yangyingliang@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/perf: Fix the PMU group constraints for threshold events in power10

[ Upstream commit 0263bbb377af0c2d38bc8b2ad2ff147e240094de ]

The PMU group constraints mask for threshold events covers
all thresholding bits which includes threshold control value
(start/stop), select value as well as thresh_cmp value (MMCRA[9:18].
In power9, thresh_cmp bits were part of the event code. But in case
of power10, thresh_cmp bits are not part of event code due to
inclusion of MMCR3 bits. Hence thresh_cmp is not valid for
group constraints for power10.

Fix the PMU group constraints checking for threshold events in
power10 by using constraint mask and value for only threshold control
and select bits.

Fixes: a64e697cef23 ("powerpc/perf: power10 Performance Monitoring support")
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606409684-1589-4-git-send-email-atrajeev@linux.vnet.ibm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/perf: Update the PMU group constraints for l2l3 events in power10

[ Upstream commit e924be7b0b0d1f37d0509c854a92c7a71e3cdfe7 ]

In Power9, L2/L3 bus events are always available as a
"bank" of 4 events. To obtain the counts for any of the
l2/l3 bus events in a given bank, the user will have to
program PMC4 with corresponding l2/l3 bus event for that
bank.

Commit 59029136d750 ("powerpc/perf: Add constraints for power9 l2/l3 bus events")
enforced this rule in Power9. But this is not valid for
Power10, since in Power10 Monitor Mode Control Register2
(MMCR2) has bits to configure l2/l3 event bits. Hence remove
this PMC4 constraint check from power10.

Since the l2/l3 bits in MMCR2 are not per-pmc, patch handles
group constrints checks for l2/l3 bits in MMCR2.

Fixes: a64e697cef23 ("powerpc/perf: power10 Performance Monitoring support")
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606409684-1589-3-git-send-email-atrajeev@linux.vnet.ibm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/perf: Fix to update radix_scope_qual in power10

[ Upstream commit d3afd28cd2f35b2a1046b76e0cf010b684da2e84 ]

power10 uses bit 9 of the raw event code as RADIX_SCOPE_QUAL.
This bit is used for enabling the radix process events.
Patch fixes the PMU counter support functions to program bit
18 of MMCR1 ( Monitor Mode Control Register1 ) with the
RADIX_SCOPE_QUAL bit value. Since this field is not per-pmc,
add this to PMU group constraints to make sure events in a
group will have same bit value for this field. Use bit 21 as
constraint bit field for radix_scope_qual. Patch also updates
the power10 raw event encoding layout information, format field
and constraints bit layout to include the radix_scope_qual bit.

Fixes: a64e697cef23 ("powerpc/perf: power10 Performance Monitoring support")
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606409684-1589-2-git-send-email-atrajeev@linux.vnet.ibm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/xmon: Fix build failure for 8xx

[ Upstream commit f3e90408019b353fd1fcd338091fb8d3c4a1c1a5 ]

With CONFIG_PPC_8xx and CONFIG_XMON set, kernel build fails with

  arch/powerpc/xmon/xmon.c:1379:12: error: 'find_free_data_bpt' defined
  but not used [-Werror=unused-function]

Fix it by enclosing find_free_data_bpt() inside #ifndef CONFIG_PPC_8xx.

Fixes: 30df74d67d48 ("powerpc/watchpoint/xmon: Support 2nd DAWR")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20201130034406.288047-1-ravi.bangoria@linux.ibm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/64: Fix an EMIT_BUG_ENTRY in head_64.S

[ Upstream commit fe18a35e685c9bdabc8b11b3e19deb85a068b75d ]

Commit 63ce271b5e37 ("powerpc/prom: convert PROM_BUG() to standard
trap") added an EMIT_BUG_ENTRY for the trap after the branch to
start_kernel(). The EMIT_BUG_ENTRY was for the address "0b", however the
trap was not labeled with "0". Hence the address used for bug is in
relative_toc() where the previous "0" label is. Label the trap as "0" so
the correct address is used.

Fixes: 63ce271b5e37 ("powerpc/prom: convert PROM_BUG() to standard trap")
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20201130004404.30953-1-jniethe5@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

powerpc/perf: Fix crash with is_sier_available when pmu is not set

[ Upstream commit f75e7d73bdf73f07b0701a6d21c111ef5d9021dd ]

On systems without any specific PMU driver support registered, running
'perf record' with —intr-regs  will crash ( perf record -I <workload> ).

The relevant portion from crash logs and Call Trace:

Unable to handle kernel paging request for data at address 0x00000068
Faulting instruction address: 0xc00000000013eb18
Oops: Kernel access of bad area, sig: 11 [#1]
CPU: 2 PID: 13435 Comm: kill Kdump: loaded Not tainted 4.18.0-193.el8.ppc64le #1
NIP:  c00000000013eb18 LR: c000000000139f2c CTR: c000000000393d80
REGS: c0000004a07ab4f0 TRAP: 0300   Not tainted  (4.18.0-193.el8.ppc64le)
NIP [c00000000013eb18] is_sier_available+0x18/0x30
LR [c000000000139f2c] perf_reg_value+0x6c/0xb0
Call Trace:
[c0000004a07ab770] [c0000004a07ab7c8] 0xc0000004a07ab7c8 (unreliable)
[c0000004a07ab7a0] [c0000000003aa77c] perf_output_sample+0x60c/0xac0
[c0000004a07ab840] [c0000000003ab3f0] perf_event_output_forward+0x70/0xb0
[c0000004a07ab8c0] [c00000000039e208] __perf_event_overflow+0x88/0x1a0
[c0000004a07ab910] [c00000000039e42c] perf_swevent_hrtimer+0x10c/0x1d0
[c0000004a07abc50] [c000000000228b9c] __hrtimer_run_queues+0x17c/0x480
[c0000004a07abcf0] [c00000000022aaf4] hrtimer_interrupt+0x144/0x520
[c0000004a07abdd0] [c00000000002a864] timer_interrupt+0x104/0x2f0
[c0000004a07abe30] [c0000000000091c4] decrementer_common+0x114/0x120

When perf record session is started with "-I" option, capturing registers
on each sample calls is_sier_available() to check for the
SIER (Sample Instruction Event Register) availability in the platform.
This function in core-book3s accesses 'ppmu->flags'. If a platform specific
PMU driver is not registered, ppmu is set to NULL and accessing its
members results in a crash. Fix the crash by returning false in
is_sier_available() if ppmu is not set.

Fixes: 333804dc3b7a ("powerpc/perf: Update perf_regs structure to include SIER")
Reported-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606185640-1720-1-git-send-email-atrajeev@linux.vnet.ibm.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: saa7146: fix array overflow in vidioc_s_audio()

[ Upstream commit 8e4d86e241cf035d6d3467cd346e7ce490681937 ]

The "a->index" value comes from the user via the ioctl.  The problem is
that the shift can wrap resulting in setting "mxb->cur_audinput" to an
invalid value, which later results in an array overflow.

Fixes: 6680427791c9 ("[media] mxb: fix audio handling")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: tvp5150: Fix wrong return value of tvp5150_parse_dt()

[ Upstream commit eb08c48132a1f594478ab9fa2b6ee646c3513a49 ]

If of_graph_get_endpoint_by_regs() return NULL, it will return 0 rather
than an errno, because we doesn't initialize the return value.

Fixes: 0556f1d580d4 ("media: tvp5150: add input source selection of_graph support")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Reviewed-by: Marco Felsch <m.felsch@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

f2fs: fix double free of unicode map

[ Upstream commit 89ff6005039a878afac87889fee748fa3f957c3a ]

In case of retrying fill_super with skip_recovery,
s_encoding for casefold would not be loaded again even though it's
already been freed because it's not NULL.
Set NULL after free to prevent double freeing when unmount.

Fixes: eca4873ee1b6 ("f2fs: Use generic casefolding support")
Signed-off-by: Hyeongseok Kim <hyeongseok@gmail.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

hwmon: (ina3221) Fix PM usage counter unbalance in ina3221_write_enable

[ Upstream commit bce776f10069c806290eaac712ba73432ae8ecd7 ]

pm_runtime_get_sync will increment pm usage counter
even it failed. Forgetting to putting operation will
result in reference leak here. We fix it by replacing
it with pm_runtime_resume_and_get to keep usage counter
balanced. It depends on the mainline commit[PM: runtime:
Add pm_runtime_resume_and_get to deal with usagecounter].

Fixes: 323aeb0eb5d9a ("hwmon: (ina3221) Add PM runtime support")
Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com>
Link: https://lore.kernel.org/r/20201202145320.1135614-1-zhangqilong3@huawei.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>

vfio-pci: Use io_remap_pfn_range() for PCI IO memory

[ Upstream commit 7b06a56d468b756ad6bb43ac21b11e474ebc54a0 ]

commit f8f6ae5d077a ("mm: always have io_remap_pfn_range() set
pgprot_decrypted()") allows drivers using mmap to put PCI memory mapped
BAR space into userspace to work correctly on AMD SME systems that default
to all memory encrypted.

Since vfio_pci_mmap_fault() is working with PCI memory mapped BAR space it
should be calling io_remap_pfn_range() otherwise it will not work on SME
systems.

Fixes: 11c4cd07ba11 ("vfio-pci: Fault mmaps to enable vma tracking")
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Acked-by: Peter Xu <peterx@redhat.com>
Tested-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

selftests/seccomp: Update kernel config

[ Upstream commit 2c07343abd8932200a45ff7b10950e71081e9e77 ]

seccomp_bpf.c uses unshare(CLONE_NEWPID), which requires CONFIG_PID_NS
to be set.

Cc: Kees Cook <keescook@chromium.org>
Cc: Shuah Khan <shuah@kernel.org>
Fixes: 6a21cc50f0c7 ("seccomp: add a return code to trap to userspace")
Signed-off-by: Mickaël Salaün <mic@linux.microsoft.com>
Acked-by: Tycho Andersen <tycho@tycho.pizza>
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20201202162643.249276-1-mic@digikod.net
Signed-off-by: Sasha Levin <sashal@kernel.org>

NFS: switch nfsiod to be an UNBOUND workqueue.

[ Upstream commit bf701b765eaa82dd164d65edc5747ec7288bb5c3 ]

nfsiod is currently a concurrency-managed workqueue (CMWQ).
This means that workitems scheduled to nfsiod on a given CPU are queued
behind all other work items queued on any CMWQ on the same CPU.  This
can introduce unexpected latency.

Occaionally nfsiod can even cause excessive latency.  If the work item
to complete a CLOSE request calls the final iput() on an inode, the
address_space of that inode will be dismantled.  This takes time
proportional to the number of in-memory pages, which on a large host
working on large files (e.g..  5TB), can be a large number of pages
resulting in a noticable number of seconds.

We can avoid these latency problems by switching nfsiod to WQ_UNBOUND.
This causes each concurrent work item to gets a dedicated thread which
can be scheduled to an idle CPU.

There is precedent for this as several other filesystems use WQ_UNBOUND
workqueue for handling various async events.

Signed-off-by: NeilBrown <neilb@suse.de>
Fixes: ada609ee2ac2 ("workqueue: use WQ_MEM_RECLAIM instead of WQ_RESCUER")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

lockd: don't use interval-based rebinding over TCP

[ Upstream commit 9b82d88d5976e5f2b8015d58913654856576ace5 ]

NLM uses an interval-based rebinding, i.e. it clears the transport's
binding under certain conditions if more than 60 seconds have elapsed
since the connection was last bound.

This rebinding is not necessary for an autobind RPC client over a
connection-oriented protocol like TCP.

It can also cause problems: it is possible for nlm_bind_host() to clear
XPRT_BOUND whilst a connection worker is in the middle of trying to
reconnect, after it had already been checked in xprt_connect().

When the connection worker notices that XPRT_BOUND has been cleared
under it, in xs_tcp_finish_connecting(), that results in:

xs_tcp_setup_socket: connect returned unhandled error -107

Worse, it's possible that the two can get into lockstep, resulting in
the same behaviour repeated indefinitely, with the above error every
300 seconds, without ever recovering, and the connection never being
established. This has been seen in practice, with a large number of NLM
client tasks, following a server restart.

The existing callers of nlm_bind_host & nlm_rebind_host should not need
to force the rebind, for TCP, so restrict the interval-based rebinding
to UDP only.

For TCP, we will still rebind when needed, e.g. on timeout, and connection
error (including closure), since connection-related errors on an existing
connection, ECONNREFUSED when trying to connect, and rpc_check_timeout(),
already unconditionally clear XPRT_BOUND.

To avoid having to add the fix, and explanation, to both nlm_bind_host()
and nlm_rebind_host(), remove the duplicate code from the former, and
have it call the latter.

Drop the dprintk, which adds no value over a trace.

Signed-off-by: Calum Mackay <calum.mackay@oracle.com>
Fixes: 35f5a422ce1a ("SUNRPC: new interface to force an RPC rebind")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

net: sunrpc: Fix 'snprintf' return value check in 'do_xprt_debugfs'

[ Upstream commit 35a6d396721e28ba161595b0fc9e8896c00399bb ]

'snprintf' returns the number of characters which would have been written
if enough space had been available, excluding the terminating null byte.
Thus, the return value of 'sizeof(buf)' means that the last character
has been dropped.

Signed-off-by: Fedor Tokarev <ftokarev@gmail.com>
Fixes: 2f34b8bfae19 ("SUNRPC: add links for all client xprts to debugfs")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

NFSv4: Fix the alignment of page data in the getdeviceinfo reply

[ Upstream commit 046e5ccb4198b990190e11fb52fd9cfd264402eb ]

We can fit the device_addr4 opaque data padding in the pages.

Fixes: cf500bac8fd4 ("SUNRPC: Introduce rpc_prepare_reply_pages()")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

SUNRPC: xprt_load_transport() needs to support the netid "rdma6"

[ Upstream commit d5aa6b22e2258f05317313ecc02efbb988ed6d38 ]

According to RFC5666, the correct netid for an IPv6 addressed RDMA
transport is "rdma6", which we've supported as a mount option since
Linux-4.7. The problem is when we try to load the module "xprtrdma6",
that will fail, since there is no modulealias of that name.

Fixes: 181342c5ebe8 ("xprtrdma: Add rdma6 option to support NFS/RDMA IPv6")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

NFSv4.2: condition READDIR's mask for security label based on LSM state

[ Upstream commit 05ad917561fca39a03338cb21fe9622f998b0f9c ]

Currently, the client will always ask for security_labels if the server
returns that it supports that feature regardless of any LSM modules
(such as Selinux) enforcing security policy. This adds performance
penalty to the READDIR operation.

Client adjusts superblock's support of the security_label based on
the server's support but also current client's configuration of the
LSM modules. Thus, prior to using the default bitmask in READDIR,
this patch checks the server's capabilities and then instructs
READDIR to remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.

v5: fixing silly mistakes of the rushed v4
v4: simplifying logic
v3: changing label's initialization per Ondrej's comment
v2: dropping selinux hook and using the sb cap.

Suggested-by: Ondrej Mosnacek <omosnace@redhat.com>
Suggested-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
Fixes: 2b0143b5c986 ("VFS: normal filesystems (and lustre): d_inode() annotations")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

SUNRPC: rpc_wake_up() should wake up tasks in the correct order

[ Upstream commit e4c72201b6ec3173dfe13fa2e2335a3ad78d4921 ]

Currently, we wake up the tasks by priority queue ordering, which means
that we ignore the batching that is supposed to help with QoS issues.

Fixes: c049f8ea9a0d ("SUNRPC: Remove the bh-safe lock requirement on the rpc_wait_queue->lock")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath10k: Release some resources in an error handling path

[ Upstream commit 6364e693f4a7a89a2fb3dd2cbd6cc06d5fd6e26d ]

Should an error occur after calling 'ath10k_usb_create()', it should be
undone by a corresponding 'ath10k_usb_destroy()' call

Fixes: 4db66499df91 ("ath10k: add initial USB support")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201122170358.1346065-1-christophe.jaillet@wanadoo.fr
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath10k: Fix an error handling path

[ Upstream commit ed3573bc3943c27d2d8e405a242f87ed14572ca1 ]

If 'ath10k_usb_create()' fails, we should release some resources and report
an error instead of silently continuing.

Fixes: 4db66499df91 ("ath10k: add initial USB support")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201122170342.1346011-1-christophe.jaillet@wanadoo.fr
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath10k: Fix the parsing error in service available event

[ Upstream commit c7cee9c0f499f27ec6de06bea664b61320534768 ]

The wmi service available event has been
extended to contain extra 128 bit for new services
to be indicated by firmware.

Currently the presence of any optional TLVs in
the wmi service available event leads to a parsing
error with the below error message:
ath10k_snoc 18800000.wifi: failed to parse svc_avail tlv: -71

The wmi service available event parsing should
not return error for the newly added optional TLV.
Fix this parsing for service available event message.

Tested-on: WCN3990 hw1.0 SNOC WLAN.HL.3.2.2-00720-QCAHLSWMTPL-1

Fixes: cea19a6ce8bf ("ath10k: add WMI_SERVICE_AVAILABLE_EVENT support")
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1605501291-23040-1-git-send-email-pillair@codeaurora.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath11k: Fix an error handling path

[ Upstream commit e7bcc145bcd035e56da7b97b033c463b32a5ff80 ]

If 'kzalloc' fails, we must return an error code.

While at it, remove a useless initialization of 'err' which could hide the
issue.

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201122173943.1366167-1-christophe.jaillet@wanadoo.fr
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath11k: Reset ath11k_skb_cb before setting new flags

[ Upstream commit 5da7acfec5ec55aa0b69b8760f1d2116b4e2ad26 ]

It was observed that the codepath for the ATH11K_SKB_HW_80211_ENCAP was
used even when the IEEE80211_TX_CTRL_HW_80211_ENCAP was not enabled for a
an skbuff. This became even more prominent when the QCAs wlan-open patchset
for ath11k [1] was applied and a sane looking fix just caused crashes when
injecting frames via a monitor interface (for example with ratechecker):

  [   86.963152] Unable to handle kernel NULL pointer dereference at virtual address 00000338
  [   86.963192] pgd = ffffffc0008f0000
  [   86.971034] [00000338] *pgd=0000000051706003, *pud=0000000051706003, *pmd=0000000051707003, *pte=00e800000b000707
  [   86.984292] Internal error: Oops: 96000006 [#1] PREEMPT SMP
  [...]
  [   87.713339] [<ffffffbffc802480>] ieee80211_tx_status_8023+0xf8/0x220 [mac80211]
  [   87.715654] [<ffffffbffc98bad4>] ath11k_dp_tx_completion_handler+0x42c/0xa10 [ath11k]
  [   87.722924] [<ffffffbffc989190>] ath11k_dp_service_srng+0x70/0x3c8 [ath11k]
  [   87.730831] [<ffffffbffca03460>] 0xffffffbffca03460
  [   87.737599] [<ffffffc00046ef58>] net_rx_action+0xf8/0x288
  [   87.742462] [<ffffffc000097554>] __do_softirq+0xfc/0x220
  [   87.748014] [<ffffffc000097900>] irq_exit+0x98/0xe8
  [   87.753396] [<ffffffc0000cf188>] __handle_domain_irq+0x90/0xb8
  [   87.757999] [<ffffffc000081ca4>] gic_handle_irq+0x6c/0xc8
  [   87.763899] Exception stack(0xffffffc00081bdc0 to 0xffffffc00081bef0)

Problem is that the state of ath11k_skb_cb->flags must be considered
unknown and could contain anything when it is not manually initialized. So
it could also contain ATH11K_SKB_HW_80211_ENCAP. And this can result in the
code to assume that the ath11k_skb_cb->vif is set - even when this is not
always the case for non ATH11K_SKB_HW_80211_ENCAP transmissions.

Tested-on: IPQ8074 hw2.0 WLAN.HK.2.4.0.1.r1-00026-QCAHKSWPL_SILICONZ-2

[1] https://source.codeaurora.org/quic/qsdk/oss/system/feeds/wlan-open/tree/mac80211/patches?h=NHSS.QSDK.11.4.r3
    (162 patches at the moment which are often not upstreamed but essential
     to get ath11k working)

Fixes: e7f33e0c52c0 ("ath11k: add tx hw 802.11 encapsulation offloading support")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201119154235.263250-2-sven@narfation.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

ath11k: Don't cast ath11k_skb_cb to ieee80211_tx_info.control

[ Upstream commit f4d291b43f809b74c66b21f5190cd578af43070b ]

The driver_data area of ieee80211_tx_info is used in ath11k for
ath11k_skb_cb. The first function in the TX patch which rewrites it to
ath11k_skb_cb is already ath11k_mac_op_tx. No one else in the code path
must use it for something else before it reinitializes it. Otherwise the
data has to be considered uninitialized or corrupt.

But the ieee80211_tx_info.control shares exactly the same area as
ieee80211_tx_info.driver_data and ath11k is still using it. This results in
best case in a

  ath11k c000000.wifi1: no vif found for mgmt frame, flags 0x0

or (slightly worse) in a kernel oops.

Instead, the interesting data must be moved first into the ath11k_skb_cb
and ieee80211_tx_info.control must then not be used anymore.

Tested-on: IPQ8074 hw2.0 WLAN.HK.2.4.0.1.r1-00026-QCAHKSWPL_SILICONZ-2

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201119154235.263250-1-sven@narfation.org
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: i2c: imx219: Selection compliance fixes

[ Upstream commit 1ed36ecd1459b653cced8929bfb37dba94b64c5d ]

To comply with the intended usage of the V4L2 selection target when
used to retrieve a sensor image properties, adjust the rectangles
returned by the imx219 driver.

The top/left crop coordinates of the TGT_CROP rectangle were set to
(0, 0) instead of (8, 8) which is the offset from the larger physical
pixel array rectangle. This was also a mismatch with the default values
crop rectangle value, so this is corrected. Found with v4l2-compliance.

While at it, add V4L2_SEL_TGT_CROP_BOUNDS support: CROP_DEFAULT and
CROP_BOUNDS have the same size as the non-active pixels are not readable
using the selection API. Found with v4l2-compliance.

[reword commit message, use macros for pixel offsets]

Fixes: e6d4ef7d58aa7 ("media: i2c: imx219: Implement get_selection")
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: rdacm20: Enable GPIO1 explicitly

[ Upstream commit 7fe1d4453fb6bf103d668a19d957a7b2fc21887c ]

The MAX9271 GPIO1 line that controls the sensor reset is by default
enabled after a serializer chip reset.

As rdacm20 does not go through an explicit serializer reset, make sure
GPIO1 is enabled to make the camera module driver more robust.

Fixes: 34009bffc1c6 ("media: i2c: Add RDACM20 driver")
Signed-off-by: Jacopo Mondi <jacopo+renesas@jmondi.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

media: max9271: Fix GPIO enable/disable

[ Upstream commit 909a0a189c677307edd461e21fd962784370d27f ]

Fix GPIO enable/disable operations which wrongly read the 0x0f register
to obtain the current mask of the enabled lines instead of using
the correct 0x0e register.

Also fix access to bit 0 of the register which is marked as reserved.

Fixes: 34009bffc1c6 ("media: i2c: Add RDACM20 driver")
Signed-off-by: Jacopo Mondi <jacopo+renesas@jmondi.org>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ASoC: Intel: Boards: tgl_max98373: update TDM slot_width

[ Upstream commit 0d7f2459ae926a964ab211aac413d72074131727 ]

Speaker amp's SSP bclk configuration was changed in the topology file to be
based on 12.288MHz and dai_ops->hw_params is based on s32le format.
But, the TDM slot size remained set to 24 bits.
This inconsistency created audible noises and needs to be corrected.
This patch updates TDM slot width to 32.

Fixes: bc7477fc2ab4 ("ASoC: Intel: Boards: tgl_max98373: Update TDM configuration in hw_params")

Signed-off-by: Sathyanarayana Nujella <sathyanarayana.nujella@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20201201211150.433472-1-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

platform/x86: dell-smbios-base: Fix error return code in dell_smbios_init

[ Upstream commit 2425ccd30fd78ce35237350fe8baac31dc18bd45 ]

Fix to return the error code -ENODEV when fails to init wmi and
smm.

Fixes: 41e36f2f85af ("platform/x86: dell-smbios: Link all dell-smbios-* modules together")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Qinglang Miao <miaoqinglang@huawei.com>
Reviewed-by: Mario Limonciello <mario.limonciello@dell.com>
Link: https://lore.kernel.org/r/20201125065032.154125-1-miaoqinglang@huawei.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

soundwire: master: use pm_runtime_set_active() on add

[ Upstream commit e04e60fce47e61743a8726d76b0149c1f4ad8957 ]

The 'master' device acts as a glue layer used during bus
initialization only, and it needs to be 'transparent' for pm_runtime
management. Its behavior should be that it becomes active when one of
its children becomes active, and suspends when all of its children are
suspended.

In our tests on Intel platforms, we routinely see these sort of
warnings on the initial boot:

[ 21.447345] rt715 sdw:3:25d:715:0: runtime PM trying to activate
child device sdw:3:25d:715:0 but parent (sdw-master-3) is not active

This is root-caused to a missing setup to make the device 'active' on
probe. Since we don't want the device to remain active forever after
the probe, the autosuspend configuration is also enabled at the end of
the probe - the device will actually autosuspend only in the case
where there are no devices physically attached. In practice, the
master device will suspend when all its children are no longer active.

Fixes: bd84256e86ecf ('soundwire: master: enable pm runtime')
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20201124130742.10986-1-yung-chuan.liao@linux.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

mailbox: arm_mhu_db: Fix mhu_db_shutdown by replacing kfree with devm_kfree

[ Upstream commit 9905f728b0bda737fe2c2afd7c24f3365a45cc7b ]

The mhu_db_channel info is allocated per channel using devm_kzalloc from
mhu_db_mbox_xlate which gets called from mbox_request_channel. However
we are releasing the allocated mhu_db_channel info using plain kfree from
mhu_db_shutdown which is called from mbox_free_channel.

This leads to random crashes when the channel is freed like below one:

  Unable to handle kernel paging request at virtual address 0080000400000008
  [0080000400000008] address between user and kernel address ranges
  Internal error: Oops: 96000044 [#1] PREEMPT SMP
  Modules linked in: scmi_module(-)
  CPU: 1 PID: 2212 Comm: rmmod Not tainted 5.10.0-rc5 #31
  Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno
   Development Platform, BIOS EDK II Nov 19 2020
  pstate: 20000085 (nzCv daIf -PAN -UAO -TCO BTYPE=--)
  pc : release_nodes+0x74/0x230
  lr : devres_release_all+0x40/0x68
  Call trace:
   release_nodes+0x74/0x230
   devres_release_all+0x40/0x68
   device_release_driver_internal+0x12c/0x1f8
   driver_detach+0x58/0xe8
   bus_remove_driver+0x64/0xe0
   driver_unregister+0x38/0x68
   platform_driver_unregister+0x1c/0x28
   scmi_driver_exit+0x38/0x44 [scmi_module]
   __arm64_sys_delete_module+0x188/0x260
   el0_svc_common.constprop.0+0x80/0x1a8
   do_el0_svc+0x2c/0x98
   el0_sync_handler+0x160/0x168
   el0_sync+0x174/0x180
  Code: 1400000d eb07009f 54000460 f9400486 (f90004a6)
  ---[ end trace c55ffd306c140233 ]---

Fix it by replacing kfree with devm_kfree as required.

Fixes: 7002ca237b21 ("mailbox: arm_mhu: Add ARM MHU doorbell driver")
Reported-by: Cristian Marussi <cristian.marussi@arm.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Tested-by: Cristian Marussi <cristian.marussi@arm.com>
Reviewed-by: Cristian Marussi <cristian.marussi@arm.com>
Signed-off-by: Jassi Brar <jaswinder.singh@linaro.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

RDMA/hns: Bugfix for calculation of extended sge

[ Upstream commit d34895c319faa1e0fc1a48c3b06bba6a8a39ba44 ]

Page alignment is required when setting the number of extended sge
according to the hardware's achivement. If the space of needed extended
sge is greater than one page, the roundup_pow_of_two() can ensure
that. But if the needed extended sge isn't 0 and can not be filled in a
whole page, the driver should align it specifically.

Fixes: 54d6638765b0 ("RDMA/hns: Optimize WQE buffer size calculating process")
Link: https://lore.kernel.org/r/1606558959-48510-3-git-send-email-liweihang@huawei.com
Signed-off-by: Yangyang Li <liyangyang20@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

RDMA/hns: Fix 0-length sge calculation error

[ Upstream commit 0fd0175e30e487f8d70ecb2cdd67fbb514fdf50f ]

One RC SQ WQE can store 2 sges but UD can't, so ignore 2 valid sges of
wr.sglist for RC which have been filled in WQE before setting extended
sge.  Either of RC and UD can not contain 0-length sges, so these 0-length
sges should be skipped.

Fixes: 54d6638765b0 ("RDMA/hns: Optimize WQE buffer size calculating process")
Link: https://lore.kernel.org/r/1606558959-48510-2-git-send-email-liweihang@huawei.com
Signed-off-by: Lang Cheng <chenglang@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

ARM: dts: at91: at91sam9rl: fix ADC triggers

[ Upstream commit 851a95da583c26e2ddeb7281e9b61f0d76ea5aba ]

The triggers for the ADC were taken from at91sam9260 dtsi but are not
correct.

Fixes: a4c1d6c75822 ("ARM: at91/dt: sam9rl: add lcd, adc, usb gadget and pwm support")
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20201128222818.1910764-10-alexandre.belloni@bootlin.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

spi: spi-fsl-dspi: Use max_native_cs instead of num_chipselect to set SPI_MCR

[ Upstream commit 2c2b3ad2c4c801bab1eec7264ea6991b1e4e8f2c ]

If cs-gpios property is used in devicetree then ctlr->num_chipselect value
may be changed by spi_get_gpio_descs().
So use ctlr->max_native_cs instead of ctlr->num_chipselect to set SPI_MCR

Fixes: 4fcc7c2292de (spi: spi-fsl-dspi: Don't access reserved fields in SPI_MCR)
Signed-off-by: Maxim Kochetkov <fido_max@inbox.ru>
Link: https://lore.kernel.org/r/20201201085916.63543-1-fido_max@inbox.ru
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: pm80xx: Do not sleep in atomic context

[ Upstream commit 4ba9e516573e60c471c01bb369144651f6f8d50b ]

hw_event_sas_phy_up() is used in hardirq/softirq context:

 pm8001_interrupt_handler_msix() || pm8001_interrupt_handler_intx() || pm8001_tasklet
   => PM8001_CHIP_DISP->isr() = pm80xx_chip_isr()
     => process_oq() [spin_lock_irqsave(&pm8001_ha->lock,)]
       => process_one_iomb()
         => mpi_hw_event()
           => hw_event_sas_phy_up()
             => msleep(200)

Revert the msleep() back to an mdelay() to avoid sleeping in atomic
context.

Link: https://lore.kernel.org/r/20201126132952.2287996-2-bigeasy@linutronix.de
Fixes: 4daf1ef3c681 ("scsi: pm80xx: Convert 'long' mdelay to msleep")
Cc: Vikram Auradkar <auradkar@google.com>
Cc: Jack Wang <jinpu.wang@cloud.ionos.com>
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

scsi: hisi_sas: Fix up probe error handling for v3 hw

[ Upstream commit 2ebde94f2ea4cffd812ece2f318c2f4922239b1d ]

Fix some rollbacks in function hisi_sas_v3_probe() and
interrupt_init_v3_hw().

Link: https://lore.kernel.org/r/1606207594-196362-3-git-send-email-john.garry@huawei.com
Fixes: 8d98416a55eb ("scsi: hisi_sas: Switch v3 hw to MQ")
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

soc: amlogic: canvas: add missing put_device() call in meson_canvas_get()

[ Upstream commit 28f851e6afa858f182802e23ac60c3ed7d1c04a1 ]

if of_find_device_by_node() succeed, meson_canvas_get() doesn't have
a corresponding put_device(). Thus add put_device() to fix the exception
handling for this function implementation.

Fixes: 382f8be04551 ("soc: amlogic: canvas: Fix meson_canvas_get when probe failed")
Signed-off-by: Yu Kuai <yukuai3@huawei.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>
Link: https://lore.kernel.org/r/20201117011322.522477-1-yukuai3@huawei.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

arm64: dts: meson-sm1: fix typo in opp table

[ Upstream commit b6a1c8a1eaa73b1e2ae251399308e9445d74cef7 ]

The freqency 1512000000 should be 1500000000.

Signed-off-by: Dongjin Kim <tobetter@gmail.com>
Fixes: 3d9e76483049 ("arm64: dts: meson-sm1-sei610: enable DVFS")
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>
Link: https://lore.kernel.org/r/20201130060320.GA30098@anyang-linuxfactory-or-kr
Signed-off-by: Sasha Levin <sashal@kernel.org>

arm64: dts: meson: fix spi-max-frequency on Khadas VIM2

[ Upstream commit b6c605e00ce8910d7ec3d9a54725d78b14db49b9 ]

The max frequency for the w25q32 (VIM v1.2) and w25q128 (VIM v1.4) spifc
chip should be 104Mhz not 30MHz.

Fixes: b8b74dda3908 ("ARM64: dts: meson-gxm: Add support for Khadas VIM2")
Signed-off-by: Artem Lapkin <art@khadas.com>
Reviewed-by: Neil Armstrong <narmstrong@baylibre.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>
Link: https://lore.kernel.org/r/20201125024001.19036-1-christianshewitt@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: iproc: Invalidate correct PAXB inbound windows

[ Upstream commit 89bbcaac3dff21f3567956b3416f5ec8b45f5555 ]

Second stage bootloaders prior to Linux boot may use all inbound windows
including IARR1/IMAP1. We need to ensure that all previous configuration
of inbound windows are invalidated during the initialization stage of
the Linux iProc PCIe driver so let's add a fix to define and invalidate
IARR1/IMAP1 because it is currently missing, fixing the issue.

Link: https://lore.kernel.org/r/20201001060054.6616-3-srinath.mannam@broadcom.com
Fixes: 9415743e4c8a ("PCI: iproc: Invalidate PAXB address mapping")
Signed-off-by: Roman Bacik <roman.bacik@broadcom.com>
Signed-off-by: Srinath Mannam <srinath.mannam@broadcom.com>
[lorenzo.pieralisi@arm.com: commit log]
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: iproc: Fix out-of-bound array accesses

[ Upstream commit a3ff529f5d368a17ff35ada8009e101162ebeaf9 ]

Declare the full size array for all revisions of PAX register sets
to avoid potentially out of bound access of the register array
when they are being initialized in iproc_pcie_rev_init().

Link: https://lore.kernel.org/r/20201001060054.6616-2-srinath.mannam@broadcom.com
Fixes: 06324ede76cdf ("PCI: iproc: Improve core register population")
Signed-off-by: Bharat Gooty <bharat.gooty@broadcom.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: Fix overflow in command-line resource alignment requests

[ Upstream commit cc73eb321d246776e5a9f7723d15708809aa3699 ]

The shift of 1 by align_order is evaluated using 32 bit arithmetic and the
result is assigned to a resource_size_t type variable that is a 64 bit
unsigned integer on 64 bit platforms. Fix an overflow before widening issue
by making the 1 a ULL.

Addresses-Coverity: ("Unintentional integer overflow")
Fixes: 32a9a682bef2 ("PCI: allow assignment of memory resources with a specified alignment")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

PCI: Bounds-check command-line resource alignment requests

[ Upstream commit 6534aac198b58309ff2337981d3f893e0be1d19d ]

32-bit BARs are limited to 2GB size (2^31).  By extension, I assume 64-bit
BARs are limited to 2^63 bytes.  Limit the alignment requested by the
"pci=resource_alignment=" command-line parameter to 2^63.

Link: https://lore.kernel.org/r/20201007123045.GS4282@kadam
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>