Oded Gabbay [Sat, 13 Nov 2021 15:58:43 +0000 (17:58 +0200)]
habanalabs: prevent false heartbeat message
If a device reset has started, there is a chance that the heartbeat
function will fail because the device is disabled at the beginning
of the reset function.
In that case, we don't want the error message to appear in the log.
Dani Liberman [Wed, 3 Nov 2021 08:09:59 +0000 (10:09 +0200)]
habanalabs: add support for fetching historic errors
A new uAPI is added for debug purposes of the user-space to retrieve
errors related data from previous session (before device reset was
performed).
Inforamtion is filled when a razwi or CS timeout happens and can
contain one of the following:
1. Retrieve timestamp of last time the device was opened and razwi or
CS timeout happened.
2. Retrieve information about last CS timeout.
3. Retrieve information about last razwi error.
This information doesn't contain user data, so no danger of data
leakage between users.
Ofir Bitton [Tue, 12 Oct 2021 17:52:46 +0000 (20:52 +0300)]
habanalabs: debugfs support for larger I2C transactions
I2C debugfs support is limited to 1 byte. We extend functionality
to more than 1 byte by using one of the pad fields as a length.
No backward compatibility issues as new F/W versions will treat 0
length as a 1 byte length transaction.
Oded Gabbay [Thu, 4 Nov 2021 07:48:22 +0000 (09:48 +0200)]
habanalabs: make hdev creation code more readable
Divide the code into 3 different parts:
- Copy kernel parameters
- Setting device behaivor per asic
- Fixup of various device parameters according to the device behaivor.
In addition, remove non-relevant code for upstream (simulator support).
farah kassabri [Wed, 3 Nov 2021 11:15:55 +0000 (13:15 +0200)]
habanalabs/gaudi: Fix collective wait bug
In Signaling-From-Graph case, the driver didn't set the hw_sob pointer
at the right place, which is needed for the cs completion
check prior to start sending all the master/slaves jobs to device.
Dani Liberman [Thu, 14 Oct 2021 19:38:41 +0000 (22:38 +0300)]
habanalabs: change wait for interrupt timeout to 64 bit
In order to increase maximum wait-for-interrupt timeout, change it
to 64 bit variable. This wait is used only by newer ASICs, so no
problem in changing this interface at this time.
Rajaravi Krishna Katta [Tue, 26 Oct 2021 11:11:06 +0000 (14:11 +0300)]
habanalabs: add dedicated message towards f/w to set power
CPUCP_PACKET_POWER_GET packet type was used for both
hl_get_power() and hl_set_power().
To align with other sensor functions hl_set_power()
should use CPUCP_PACKET_POWER_SET.
This packet will only be used with newer ASICs, so need to add
a compatibility flag to the asic properties to indicate whether to use
this packet or the GET packet.
habanalabs: handle abort scenario for user interrupt
In case of device reset, the driver does a force trigger on all waiting
users to release them from waiting. However, the driver does not handle
error scenario while waiting.
hl_interrupt_wait_ioctl() now exits the wait in case of an error with
abort status.
Ohad Sharabi [Tue, 26 Oct 2021 12:33:23 +0000 (15:33 +0300)]
habanalabs: use variable poll interval for fw loading
Using a variable poll interval for fw loading allows us to support
much slower environments (emulation) while changing only a single
line in the code, instead of choosing a different interval in each
function that polls.
Ohad Sharabi [Thu, 21 Oct 2021 08:24:41 +0000 (11:24 +0300)]
habanalabs: adding indication of boot fit loaded
Up until now the driver stored indication if Linux was loaded on the
device CPU. This was needed in order to coordinate some tasks that are
performed by the Linux.
In future ASICs, many of those tasks will be performed by the boot
fit, so now we need the same indication of boot fit load status.
Yuri Nudelman [Mon, 25 Oct 2021 08:37:25 +0000 (11:37 +0300)]
habanalabs: partly skip cache flush when in PMMU map flow
The PCI MMU cache is two layered. The upper layer, memcache, uses cache
lines, the bottom layer doesn't.
Hence, after PMMU map operation we have to invalidate memcache, to avoid
the situation where the new entry is already in the cache due to its
cache line being fully in the cache.
However, we do not have to invalidate the lower cache, and here we can
optimize, since cache invalidation is time consuming.
The enum vm_type was abused, used once as a value (indication
memory type for map) and once as a flag (for cache invalidation).
This makes it hard to add new and still keep it meaningful, hence it
is better to split into one enum for values and one for flags.
Yuri Nudelman [Thu, 14 Oct 2021 07:33:27 +0000 (10:33 +0300)]
habanalabs: wrong VA size calculation
VA blocks are currently stored in an inconsistent way. Sometimes block
end is inclusive, sometimes exclusive. This leads to wrong size
calculations in certain cases, plus could lead to a segmentation fault
in case mapping process fails in the middle and we try to roll it back.
Need to make this consistent - start inclusive till end inclusive.
For example, the regions table may now look like this:
0x0000 - 0x1fff : allocated
0x2000 - 0x2fff : free
0x3000 - 0x3fff : allocated
Ohad Sharabi [Sun, 17 Oct 2021 06:00:43 +0000 (09:00 +0300)]
habanalabs: revise and document use of boot status flags
The boot status flag "SRAM available" can be set by f/w Linux (in the
general case) or by f/w uboot (in some specific debug scenario) but
never by f/w preboot.
Hence, when polling the boot status flags in the preboot stage we do not
want to poll on "SRAM Avialable".
The special case in which uboot set this flag is when we are running
special debug scenario without Linux. In this case, at some point during
the boot, the uboot relocates its code to the DRAM and then set the
specified flag.
Oded Gabbay [Thu, 21 Oct 2021 11:02:40 +0000 (14:02 +0300)]
habanalabs/gaudi: recover from CPU WD event
There are rare cases where the device CPU's watchdog has expired and as
a result, the watchdog reset has happened and the CPU will now move to
running its preboot f/w.
When that happens, the driver will only know that a heartbeat failure
occurred. As a result, the driver will send a message to the CPU's main
f/w asking it to reset the device, but because the CPU is now running
preboot, it won't respond and the re-initialization process will later
fail when trying to load the f/w.
The solution is to send the request to the preboot as well, only if the
reset was caused because of HB failure.
Greg Kroah-Hartman [Wed, 22 Dec 2021 11:33:01 +0000 (12:33 +0100)]
Merge tag 'iio-for-5.17a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into char-misc-next
Jonathan writes:
1st set of IIO new device support, features and cleanup for 5.17
Includes some fixes that were either late breaking, low priority or
complex enough to not be good to rush in late in the cycle.
Tree rebased today to fix up some trivial issues + pull in a fix that
was previously on the fixes-togreg branch. Vast majority have been
in linux-next for some time now.
New device support
* adi,ad7293
- New driver and bindings for this Power Amplifier drain current
controller. A complex device with various related monitoring functions.
* adi,ad75513R
- New driver and bindings for this combined ADC and DAC device.
- A few follow up fixes.
* adi,admv8818
- New driver (and type) for this 2-18GHz filter device. Includes
bindings and ABI documentation to allow clk_notifier based auto
adjustment of the filters in appropriate applications.
* liteon,ltr501
- Support for the ltr303. ID and chip specific info table.
* xilinx,ams
- New generic firmware function fwnode_iomap() as used in this driver.
- New driver and bindings for this ADC and on-chip sensors as found
in various Xilinx devices.
Core
* Introduced IIO_VAL_INT_64 which uses val and val2 in IIO callbacks to
form a 64 bit integer when higher precision needed.
* Allow IIO_ENUM_AVAILABLE to be used with different shared values.
* Fix a long term issue with scheduling whilst atomic when iio_trig_poll()
is called but no trigger consumers are actually enabled and hence the
trigger may be reenabled from the interrupt handler. Seen in the wild
on the tsc2046.
* Mark iio_device_type const.
* buffer: Use a separate index variable to simplify code.
* buffer-dma: Clear out unused struct iio_buffer_block
* buffer-dmaengine: Switch to cheaper round_down() as power of 2 values.
Tests/tools
* format_value
- Check against NULL returns from allocations in tests.
- Add IIO_VAL_INT_64 test case.
* event_monitor
- Flush the output after event to given more consistent latency
when tool output piped to other programs.
Driver Features
* axp20x
- Add support for NTC thermistor channel and document TS pin binding.
* arm,scmi
- Add reading of raw channel values (using IIO_VAL_INT_64)
* liteon,ltr501
- Add proximity-near-level support and dt-binding.
Tree wide cleanup
* Remove no-op trigger ops from multiple drivers.
* Stop using dev_get_drvdata() on the iio_dev->dev in various drivers
and then stop assigning it to allow this to be used for other purposes.
We can always get to the indio_dev using dev_to_iio_dev() which is
a container_of() based approach. Also cleanup up some related unnecessary
convoluted cases.
- atmel,at91-sam5d2
- nxp,imx7d
- meas,ms5611
- st,st_sensors
* Where available (or easy to introduce) use the scan_type.* values
in place of a second copy for read_raw and similar paths.
- adi,ad7266
- bosch,bma220
- fsl,mac3110
- fsl,mma7455
- fsl,mpl3115
- kionix,kcjk-1013
- sensortek,stk8ba50
- sensortek,stk8312
- ti,adc12138
- ti,ads1015
- vti,sca3000
- xilinx,xadc-core
* Switch drives over to generic firmware properties including appropriate
header changes to avoid including of.h
- Various DACs had false CONFIG_OF dependencies.
- dpot-dac
- envelope-detector
- adi,ad5755
- adi,ad5758
- capella,cm3605
- maxim,max9611
- microchip,mcp41010
- microchip,mcp3911
- ti,adc12138
* Trivial clang warning fixes for W=1 warnings.
Driver specific cleanup and minor fixes
* adi,ad7606
- Comment fixes.
* ams,ad3935
- Drop pointless cast to the same type.
* atmel,at91-sama5d2
- Fix wrong cast of iio_dev->dev to platform_device that happened to
be harmless.
* fsl,mma7660
- Stop i2c remove() function returning an error code. Part of a rework
to eventually stop returning anything from these.
* fsl,mma8452
- Use correct type for local irqreturn_t.
* nxp,imx8mq
- Maintainer email address update.
* nxp,lpc18xx_adc
- Ensure clk_prepare_enable() called before clk_get_rate().
- Switch of.h for mod_devicetable.h to reflect no of specific functions,
just the id table.
* renesas,rzg2l
- Drop a dev_err() that just duplicates error printed in platform_get_irq()
* sgx,vz89x
- Drop pointless cast.
* st,lsm6dsx
- Make it possible to disable the sensorhub from DT to avoid a corner
case where the address of a slave device many be accidentally modified.
* st,stm32-adc
- Stop leaking an of_node in an error path.
* st,stmp2
- Avoid wrong sized type for bit field which could result in
over-reading (harmless). Precursor to enabling -Warray-bounds.
* ti,adc081c
- Put back some ACPI support for non standards compliant ADC081C
ID because it is known to be in the wild on some Aaeon boards.
* ti,ads8688
- Cleanup redundant local ret variable assignment.
* ti,ina2xx-adc
- Use helper macro kthread_run() to replace some boilerplate.
- Avoid double reference counting.
- Drop pointless cast.
* xilinx,xadc
- Make the IRQ optional as not always wired to the host system.
* tag 'iio-for-5.17a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio: (103 commits)
iio: adc: ti-adc081c: Partial revert of removal of ACPI IDs
iio:addac:ad74413r: Fix uninitialized ret in a path that won't be hit.
MAINTAINERS: Add maintainer for xilinx-ams
dt-bindings: iio: adc: Add Xilinx AMS binding documentation
iio: adc: Add Xilinx AMS driver
device property: Add fwnode_iomap()
iio:accel:kxcjk-1013: Mark struct __maybe_unused to avoid warning.
iio:accel:bmc150: Mark structure __maybe_unused as only needed with for pm ops.
iio:dummy: Drop set but unused variable len.
iio:magn:ak8975: Suppress clang W=1 warning about pointer to enum conversion.
iio:imu:inv_mpu6050: Suppress clang W=1 warning about pointer to enum conversion.
iio:imu:inv_icm42600: Suppress clang W=1 warning about pointer to enum conversion.
iio:dac:mcp4725: Suppress clang W=1 warning about pointer to enum conversion.
iio:amplifiers:hmc425a: Suppress clang W=1 warning about pointer to enum conversion.
iio:adc:ti-ads1015: Suppress clang W=1 warning about pointer to enum conversion.
iio:adc:rcar: Suppress clang W=1 warning about pointer to enum conversion.
iio:adc:ina2xx-adc: Suppress clang W=1 warning about pointer to enum conversion.
iio:accel:bma180: Suppress clang W=1 warning about pointer to enum conversion.
drivers:iio:dac: Add AD3552R driver support
dt-bindings: iio: dac: Add adi,ad3552r.yaml
...
Jonathan Cameron [Sun, 5 Dec 2021 17:27:28 +0000 (17:27 +0000)]
iio: adc: ti-adc081c: Partial revert of removal of ACPI IDs
Unfortuanately a non standards compliant ACPI ID is known to be
in the wild on some AAEON boards.
Partly revert the removal of these IDs so that ADC081C will again
work + add a comment to that affect for future reference.
Whilst here use generic firmware properties rather than the ACPI
specific handling previously found in this driver.
Reported-by: Kunyang Fan <Kunyang_Fan@aaeon.com.tw> Fixes: c458b7ca3fd0 ("iio:adc:ti-adc081c: Drop ACPI ids that seem very unlikely to be official.") Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Cc: Andy Shevchenko <andy.shevchenko@gmail.com> Tested-by: Kunyang Fan <Kunyang_Fan@aaeon.com.tw> #UP-extremei11 Link: https://lore.kernel.org/r/20211205172728.2826512-1-jic23@kernel.org Cc: <Stable@vger.kernel.org>
Jonathan Cameron [Mon, 20 Dec 2021 16:47:26 +0000 (16:47 +0000)]
iio:addac:ad74413r: Fix uninitialized ret in a path that won't be hit.
I don't believe it's possible to hit this, because we drop
out of __iio_update_buffers() earlier in the event of an empty
list. However, that is not visible to the compiler so lets
return an error if we do hit the loop with an empty bitmask.
Fixes: 5d97d9e9a703 ("iio: addac: ad74413r: fix off by one in ad74413r_parse_channel_config()") Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Cc: Cosmin Tanislav <cosmin.tanislav@analog.com> Link: https://lore.kernel.org/r/20211220164726.3136307-1-jic23@kernel.org
The AMS includes an ADC as well as on-chip sensors that can be used to
sample external voltages and monitor on-die operating conditions, such
as temperature and supply voltage levels. The AMS has two SYSMON blocks.
PL-SYSMON block is capable of monitoring off chip voltage and
temperature.
PL-SYSMON block has DRP, JTAG and I2C interface to enable monitoring
from an external master. Out of these interfaces currently only DRP is
supported. Other block PS-SYSMON is memory mapped to PS.
The AMS can use internal channels to monitor voltage and temperature as
well as one primary and up to 16 auxiliary channels for measuring
external voltages.
The voltage and temperature monitoring channels also have event capability
which allows to generate an interrupt when their value falls below or
raises above a set threshold.
This patch introduces a new helper routine - fwnode_iomap(), which
allows to map the memory mapped IO for a given device node.
This implementation does not cover the ACPI case and may be expanded
in the future. The main purpose here is to be able to develop resource
provider agnostic drivers.
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Anand Ashok Dumbre <anand.ashok.dumbre@xilinx.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Link: https://lore.kernel.org/r/20211203212358.31444-2-anand.ashok.dumbre@xilinx.com Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
A bug exists if the user executes a COUNTER_ADD_WATCH_IOCTL ioctl call,
and then executes a COUNTER_DISABLE_EVENTS_IOCTL ioctl call. Disabling
the events should disable the 104-QUAD-8 interrupts, but because of this
bug the interrupts are not disabling.
The reason this bug is occurring is because quad8_events_configure() is
called when COUNTER_DISABLE_EVENTS_IOCTL is handled, but the
next_irq_trigger[] array has not been cleared before it is checked in
the loop.
This patch fixes the bug by removing the next_irq_trigger array and
instead utilizing a different algorithm of walking the events_list list
for the current requested events. When a COUNTER_DISABLE_EVENTS_IOCTL is
handled, events_list will be empty and thus all device channels end up
with interrupts disabled.
Uwe Kleine-König [Tue, 21 Dec 2021 08:16:47 +0000 (17:16 +0900)]
counter: ti-eqep: Use container_of instead of struct counter_device::priv
Using counter->priv is a memory read and so more expensive than
container_of which is only an addition. (In this case even a noop
because the offset is 0.)
So container_of is expected to be a tad faster, it's type-safe, and
produces smaller code (ARCH=arm allmodconfig):
Rob Herring [Thu, 9 Dec 2021 17:42:35 +0000 (17:42 +0000)]
dt-bindings: nvmem: Add missing 'reg' property
With 'unevaluatedProperties' support implemented, the following warnings
are generated in the nvmem examples:
Documentation/devicetree/bindings/nvmem/st,stm32-romem.example.dt.yaml: efuse@1fff7800: Unevaluated properties are not allowed ('reg' was unexpected)
Documentation/devicetree/bindings/nvmem/rmem.example.dt.yaml: nvram@10000000: Unevaluated properties are not allowed ('reg' was unexpected)
Documentation/devicetree/bindings/nvmem/brcm,nvram.example.dt.yaml: nvram@1eff0000: Unevaluated properties are not allowed ('reg' was unexpected)
Instead of invoking a synchronize_rcu() to free a pointer
after a grace period we can directly make use of new API
that does the same but in more efficient way.
Tiezhu Yang [Thu, 16 Dec 2021 03:33:01 +0000 (11:33 +0800)]
rapidio: remove not used code about RIO_VID_TUNDRA
According to https://rapidio.org/vendor-id/, there is no 0x000d vendor id
in the complete and current list of VendorIDs, it means that the related
code is dead code now, so just remove them.
Tiezhu Yang [Thu, 16 Dec 2021 03:33:00 +0000 (11:33 +0800)]
rapidio: remove not used macro definition in rio_ids.h
The definition of RIO_VID_FREESCALE, RIO_DID_MPC8560, RIO_DID_TSI500,
RIO_DID_TSI576 and RIO_DID_TSI721 are not used for many years in the
current code, so just remove them.
Explicitly remove the file entries from sysfs before dropping the final
reference for symmetry reasons and for consistency with the rest of the
driver.
Johan Hovold [Wed, 1 Dec 2021 13:25:27 +0000 (14:25 +0100)]
firmware: qemu_fw_cfg: fix sysfs information leak
Make sure to always NUL-terminate file names retrieved from the firmware
to avoid accessing data beyond the entry slab buffer and exposing it
through sysfs in case the firmware data is corrupt.
Fixes: 75f3e8e47f38 ("firmware: introduce sysfs driver for QEMU's fw_cfg device") Cc: stable@vger.kernel.org # 4.6 Cc: Gabriel Somlo <somlo@cmu.edu> Signed-off-by: Johan Hovold <johan@kernel.org> Link: https://lore.kernel.org/r/20211201132528.30025-4-johan@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Johan Hovold [Wed, 1 Dec 2021 13:25:26 +0000 (14:25 +0100)]
firmware: qemu_fw_cfg: fix kobject leak in probe error path
An initialised kobject must be freed using kobject_put() to avoid
leaking associated resources (e.g. the object name).
Commit fe3c60684377 ("firmware: Fix a reference count leak.") "fixed"
the leak in the first error path of the file registration helper but
left the second one unchanged. This "fix" would however result in a NULL
pointer dereference due to the release function also removing the never
added entry from the fw_cfg_entry_cache list. This has now been
addressed.
Fix the remaining kobject leak by restoring the common error path and
adding the missing kobject_put().
Fixes: 75f3e8e47f38 ("firmware: introduce sysfs driver for QEMU's fw_cfg device") Cc: stable@vger.kernel.org # 4.6 Cc: Gabriel Somlo <somlo@cmu.edu> Signed-off-by: Johan Hovold <johan@kernel.org> Link: https://lore.kernel.org/r/20211201132528.30025-3-johan@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Johan Hovold [Wed, 1 Dec 2021 13:25:25 +0000 (14:25 +0100)]
firmware: qemu_fw_cfg: fix NULL-pointer deref on duplicate entries
Commit fe3c60684377 ("firmware: Fix a reference count leak.") "fixed"
a kobject leak in the file registration helper by properly calling
kobject_put() for the entry in case registration of the object fails
(e.g. due to a name collision).
This would however result in a NULL pointer dereference when the
release function tries to remove the never added entry from the
fw_cfg_entry_cache list.
Fix this by moving the list-removal out of the release function.
Note that the offending commit was one of the benign looking umn.edu
fixes which was reviewed but not reverted. [1][2]
Jason Wang [Sun, 12 Dec 2021 07:18:38 +0000 (15:18 +0800)]
applicom: unneed to initialise statics to 0
Static variables do not need to be initialised to 0, because compilers
will initialise all uninitialised statics to 0. Thus, remove the
unneeded initializations.
Kai Ye [Mon, 6 Dec 2021 10:47:24 +0000 (18:47 +0800)]
uacce: use sysfs_emit instead of sprintf
Use the sysfs_emit to replace sprintf. sprintf may cause
output defect in sysfs content, it is better to use new
added sysfs_emit function which knows the size of the
temporary buffer.
Ben Hutchings [Mon, 18 Jun 2018 22:55:40 +0000 (23:55 +0100)]
firmware: Update Kconfig help text for Google firmware
The help text for GOOGLE_FIRMWARE states that it should only be
enabled when building a kernel for Google's own servers. However,
many of the drivers dependent on it are also useful on Chromebooks or
on any platform using coreboot.
Update the help text to reflect this double duty.
Fixes: d384d6f43d1e ("firmware: google memconsole: Add coreboot support") Reviewed-by: Julius Werner <jwerner@chromium.org> Signed-off-by: Ben Hutchings <ben@decadent.org.uk> Link: https://lore.kernel.org/r/20180618225540.GD14131@decadent.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Greg Kroah-Hartman [Tue, 21 Dec 2021 09:05:01 +0000 (10:05 +0100)]
Merge tag 'fpga-for-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mdf/linux-fpga into char-misc-next
Moritz writes:
FPGA Manager changes for 5.17-rc1
Russ' patches rework the way we register FPGA managers, regions and
bridges by simplifying the functions into a single register call.
Nathan's patch addresses an unused variable warning that was introduced
by Russ' patches.
Yang's patch addresses a kernel doc warning.
All patches have been reviewed on the mailing list, and have been in the
last few linux-next releases (as part of our for-next branch) without issues.
Signed-off-by: Moritz Fischer <mdf@kernel.org>
* tag 'fpga-for-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mdf/linux-fpga:
fpga: region: fix kernel-doc
fpga: stratix10-soc: Do not use ret uninitialized in s10_probe()
fpga: region: Use standard dev_release for class driver
fpga: bridge: Use standard dev_release for class driver
fpga: mgr: Use standard dev_release for class driver
The qpnpint_irq_set_type() callback function configures the type
(edge vs level) and polarity (high, low, or both) of a particular
PMIC interrupt within a given peripheral. To do this, it reads
the three consecutive IRQ configuration registers, modifies the
specified IRQ bit within the register values, and finally writes
the three modified register values back to the PMIC. While a
spinlock is used to provide mutual exclusion on the SPMI bus
during the register read and write calls, there is no locking
around the overall read, modify, write sequence. This opens up
the possibility of a race condition if two tasks set the type of
a PMIC IRQ within the same peripheral simultaneously.
When the race condition is encountered, both tasks will read the
old value of the registers and IRQ bits set by one of the tasks
will be dropped upon the register write of the other task. This
then leads to PMIC IRQs being enabled with an incorrect type and
polarity configured. Such misconfiguration can lead to an IRQ
storm that overwhelms the system and causes it to crash.
This race condition and IRQ storm have been observed when using
a pair of pm8941-pwrkey devices to handle PMK8350 pwrkey and
resin interrupts. The independent devices probe asynchronously
in parallel and can simultaneously request and configure PMIC
IRQs in the same PMIC peripheral.
For a good case, the IRQ configuration calls end up serialized
due to timing deltas and the register read/write sequence looks
like this:
For a bad case, the IRQ configuration calls end up occurring
simultaneously and the race condition is encountered. The
register read/write sequence then looks like this:
This corresponds to the resin IRQ being configured for both
rising and falling edges, as expected. However, the pwrkey IRQ
is misconfigured as level type with both polarity high and low
set to disabled. The PMIC IRQ triggering hardware treats this
particular register configuration as if level low triggering is
enabled.
The raw pwrkey IRQ signal is low when the power key is not being
pressed. Thus, the pwrkey IRQ begins firing continuously in an
IRQ storm.
Fix the race condition by holding the spmi-pmic-arb spinlock for
the duration of the read, modify, write sequence performed in the
qpnpint_irq_set_type() function. Split the pmic_arb_read_cmd()
and pmic_arb_write_cmd() functions each into three parts so that
hardware register IO is decoupled from spinlock locking. This
allows a new function pmic_arb_masked_write() to be added which
locks the spinlock and then calls register IO functions to
perform SPMI read and write commands in a single atomic
operation.
Stephen Boyd [Thu, 16 Dec 2021 19:08:07 +0000 (11:08 -0800)]
spmi: pmic-arb: Add sid and address to error messages
It's useful to know what particular device/component is having trouble
accessing the bus. Add the sid and address to error messages here so
that debugging is a little simpler.
Kees Cook [Thu, 16 Dec 2021 08:12:26 +0000 (13:42 +0530)]
bus: mhi: core: Use correctly sized arguments for bit field
The find.h APIs are designed to be used only on unsigned long arguments.
This can technically result in a over-read, but it is harmless in this
case. Regardless, fix it to avoid the warning seen under -Warray-bounds,
which we'd like to enable globally:
In file included from ./include/linux/bitmap.h:9,
from ./include/linux/cpumask.h:12,
from ./arch/x86/include/asm/cpumask.h:5,
from ./arch/x86/include/asm/msr.h:11,
from ./arch/x86/include/asm/processor.h:22,
from ./arch/x86/include/asm/cpufeature.h:5,
from ./arch/x86/include/asm/thread_info.h:53,
from ./include/linux/thread_info.h:60,
from ./arch/x86/include/asm/preempt.h:7,
from ./include/linux/preempt.h:78,
from ./include/linux/spinlock.h:55,
from ./include/linux/wait.h:9,
from ./include/linux/wait_bit.h:8,
from ./include/linux/fs.h:6,
from ./include/linux/debugfs.h:15,
from drivers/bus/mhi/core/init.c:7:
drivers/bus/mhi/core/init.c: In function 'to_mhi_pm_state_str':
./include/linux/find.h:187:37: warning: array subscript 'long unsigned int[0]' is partly outside array bounds of 'enum mhi_pm_state[1]' [-Warray-bounds]
187 | unsigned long val = *addr & GENMASK(size - 1, 0);
| ^~~~~
drivers/bus/mhi/core/init.c:80:51: note: while referencing 'state'
80 | const char *to_mhi_pm_state_str(enum mhi_pm_state state)
| ~~~~~~~~~~~~~~~~~~^~~~~
Manivannan Sadhasivam [Thu, 16 Dec 2021 08:12:25 +0000 (13:42 +0530)]
bus: mhi: core: Add an API for auto queueing buffers for DL channel
Add a new API "mhi_prepare_for_transfer_autoqueue" for using with client
drivers like QRTR to request MHI core to autoqueue buffers for the DL
channel along with starting both UL and DL channels.
So far, the "auto_queue" flag specified by the controller drivers in
channel definition served this purpose but this will be removed at some
point in future.
Cc: netdev@vger.kernel.org Cc: Jakub Kicinski <kuba@kernel.org> Cc: David S. Miller <davem@davemloft.net> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Co-developed-by: Loic Poulain <loic.poulain@linaro.org> Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Loic Poulain <loic.poulain@linaro.org> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Link: https://lore.kernel.org/r/20211216081227.237749-9-manivannan.sadhasivam@linaro.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Manivannan Sadhasivam [Thu, 16 Dec 2021 08:12:24 +0000 (13:42 +0530)]
bus: mhi: core: Fix race while handling SYS_ERR at power up
During SYS_ERR condition, as a response to the MHI_RESET from host, some
devices tend to issue BHI interrupt without clearing the SYS_ERR state in
the device. This creates a race condition and causes a failure in booting
up the device.
The issue is seen on the Sierra Wireless EM9191 modem during SYS_ERR
handling in mhi_async_power_up(). Once the host detects that the device
is in SYS_ERR state, it issues MHI_RESET and waits for the device to
process the reset request. During this time, the device triggers the BHI
interrupt to the host without clearing SYS_ERR condition. So the host
starts handling the SYS_ERR condition again.
To fix this issue, let's register the IRQ handler only after handling the
SYS_ERR check to avoid getting spurious IRQs from the device.
Fixes: e18d4e9fa79b ("bus: mhi: core: Handle syserr during power_up") Cc: stable@vger.kernel.org Reported-by: Aleksander Morgado <aleksander@aleksander.es> Tested-by: Aleksander Morgado <aleksander@aleksander.es> Tested-by: Thomas Perrot <thomas.perrot@bootlin.com> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Link: https://lore.kernel.org/r/20211216081227.237749-8-manivannan.sadhasivam@linaro.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 'wake-capable' entry in channel configuration is not set when
parsing the configuration specified by the controller driver. Add
the missing entry to ensure channel is correctly specified as a
'wake-capable' channel.
Loic Poulain [Thu, 16 Dec 2021 08:12:19 +0000 (13:42 +0530)]
bus: mhi: pci_generic: Graceful shutdown on freeze
There is no reason for shutting down MHI ungracefully on freeze,
this causes the MHI host stack & device stack to not be aligned
anymore since the proper MHI reset sequence is not performed for
ungraceful shutdown.
Greg Kroah-Hartman [Fri, 17 Dec 2021 09:06:21 +0000 (10:06 +0100)]
Merge tag 'lkdtm-v5.17-rc1' of https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux into char-misc-next
Kees writes:
lkdtm updates for v5.17-rc1
- Fix printk() usage during recursion (Ard Biesheuvel)
- Fix rodata section to actually have contents (Christophe Leroy)
- Add notes about lkdtm_kernel_info usage (Kees Cook)
- Avoid stack-entropy selftest when LKDTM is disabled (Misono Tomohiro)
* tag 'lkdtm-v5.17-rc1' of https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
selftest/lkdtm: Skip stack-entropy test if lkdtm is not available
lkdtm: Fix content of section containing lkdtm_rodata_do_nothing()
lkdtm: avoid printk() in recursive_loop()
lkdtm: Note that lkdtm_kernel_info should be removed in the future
Ard Biesheuvel [Thu, 7 Oct 2021 08:12:35 +0000 (10:12 +0200)]
lkdtm: avoid printk() in recursive_loop()
The recursive_loop() function is intended as a diagnostic to ensure that
exhausting the stack is caught and mitigated. Currently, it uses
pr_info() to ensure that the function has side effects that the compiler
cannot simply optimize away, so that the stack footprint does not get
reduced inadvertently.
The typical mitigation for stack overflow is to kill the task, and this
overflow may occur inside the call to pr_info(), which means it could be
holding the console lock when this happens. This means that the console
lock is never going to be released again, preventing the diagnostic
prints related to the stack overflow handling from being visible on the
console.
So let's replace the call to pr_info() with a call to
memzero_explicit(), which is not a 'magic' function name like memset()
or memcpy(), which the compiler may replace with plain loads and stores.
To ensure that the stack frames are nested rather than tail-called, put
the call to memzero_explicit() after the recursive call.
lkdtm: Note that lkdtm_kernel_info should be removed in the future
As per Linus's request, remove lkdtm_kernel_info once sufficient
reporting exists in CI systems:
https://lore.kernel.org/lkml/CAHk-=wiFvfkoFixTapvvyPMN9pq5G-+Dys2eSyBa1vzDGAO5+A@mail.gmail.com
Mihail Chindris [Mon, 13 Dec 2021 11:08:25 +0000 (11:08 +0000)]
drivers:iio:dac: Add AD3552R driver support
The AD3552R-16 is a low drift ultrafast, 16-bit accuracy,
current output digital-to-analog converter (DAC) designed
to generate multiple output voltage span ranges.
The AD3552R-16 operates with a fixed 2.5V reference.
Mihail Chindris [Mon, 13 Dec 2021 11:08:24 +0000 (11:08 +0000)]
dt-bindings: iio: dac: Add adi,ad3552r.yaml
Add documentation for ad3552r and ad3542r
Signed-off-by: Mihail Chindris <mihail.chindris@analog.com> Reviewed-by: Rob Herring <robh@kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Antoniu Miclaus [Tue, 7 Dec 2021 15:54:43 +0000 (17:54 +0200)]
iio:filter:admv8818: add support for ADMV8818
The ADMV8818-EP is a fully monolithic microwave integrated
circuit (MMIC) that features a digitally selectable frequency of
operation. The device features four independently controlled high-
pass filters (HPFs) and four independently controlled low-pass
filters (LPFs) that span the 2 GHz to 18 GHz frequency range.
projects / linux.git / log