The local 'current_bits' variable does not have to be initialized
because all cases in following switch() either return or initialize it.
Addresses-Coverity: Unused value
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Commit 60268b0e82 ("hwmon: (amd_energy) modify the visibility of
the counters") restricted visibility of AMD energy counters to work
around a side-channel attack using energy data to determine which
instructions are executed. The attack is described in 'PLATYPUS:
Software-based Power Side-Channel Attacks on x86'. It relies on quick
and accurate energy readings.
This change made the counters provided by the amd_energy driver
effectively unusable for non-provileged users. However, unprivileged
read access is the whole point of hardware monitoring attributes.
An attempt to remedy the situation by limiting and randomizing access
to chip registers was rejected by AMD. Since the driver is for all
practical purposes unusable, remove it.
Cc: Naveen Krishna Chatradhi <nchatrad@amd.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Hardware monitoring sysfs attributes are used and displayed by unrestricted
userspace applications. Standard attributes therefore have to be world
readable, since otherwise those userspace applications would either have
to run as super-user or display an error. None of those makes sense.
Clarify the expected scope of attribute access in the ABI document.
Cc: Naveen Krishna Chatradhi <nchatrad@amd.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Exported pmbus symbols are only supposed to be used from PMBus code.
Introduce PMBUS symbol namespace to prevent misuse from other code.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Add pmbus driver support for Maxim MAX15301 InTune Automatically
Compensated Digital PoL Controller with Driver and PMBus Telemetry
Even though the specification does not specifically mention it,
extensive empirical testing has revealed that auto-detection of
limit-registers will fail in a random fashion unless the delay
parameter is set to above about 80us. The default delay is set
to 100us to include some safety margin.
This patch is tested on a Flex BMR461 converter module.
Signed-off-by: Erik Rosen <erik.rosen@metormote.com>
Link: https://lore.kernel.org/r/20210419101251.24840-1-erik.rosen@metormote.com
[groeck: Added rationale for delay to driver header]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Calling remove() on error whould have only unregistered
the watchdog, and since a failure in registering him
is considered non-fatal and happens last, remove the
error path and return the error codes directly.
Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Link: https://lore.kernel.org/r/20210417210920.15496-3-W_Armin@gmx.de
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Use devm_hwmon_device_register_with_info() and remove hwmon_dev
from sch5627_data struct as it is not needed anymore.
Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Link: https://lore.kernel.org/r/20210417210920.15496-2-W_Armin@gmx.de
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Certain VRs might be configured to use only the first output channel and
so the mode for the second will be 0. Handle this gracefully.
Fixes: b9fa0a3acf ("hwmon: (pmbus/core) Add support for vid mode detection per page bases")
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Link: https://lore.kernel.org/r/20210416102926.13614-1-fercerpav@gmail.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
This patch adds support for these devices:
- YH-5151E - the PDU
- YM-2151E - the PSU
The device datasheet says that the devices support PMBus 1.2, but in my
testing, a lot of the commands aren't supported and if they are, they
sometimes behave strangely or inconsistently. For example, writes to the
PAGE command requires using PEC, otherwise the write won't work and the
page won't switch, even though, the standard says that PEC is optional.
On the other hand, writes to SMBALERT don't require PEC. Because of
this, the driver is mostly reverse engineered with the help of a tool
called pmbus_peek written by David Brownell (and later adopted by my
colleague Jan Kundrát).
The device also has some sort of a timing issue when switching pages,
which is explained further in the code.
Because of this, the driver support is limited. It exposes only the
values that have been tested to work correctly.
Signed-off-by: Václav Kubernát <kubernat@cesnet.cz>
Link: https://lore.kernel.org/r/20210414080019.3530794-1-kubernat@cesnet.cz
[groeck: Fixed up "missing braces around initializer" from 0-day]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Like the Intel N3000 card, the Intel D5005 has a MAX10 based
BMC. This commit adds support for the D5005 sensors that are
monitored by the MAX10 BMC.
Signed-off-by: Matthew Gerlach <matthew.gerlach@linux.intel.com>
Signed-off-by: Russ Weight <russell.h.weight@linux.intel.com>
Acked-by: Lee Jones <lee.jones@linaro.org>
Reviewed-by: Tom Rix <trix@redhat.com>
Link: https://lore.kernel.org/r/20210413225835.459662-3-matthew.gerlach@linux.intel.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
An error in sch5627_update_device() could cause sch5627_read()
to fail even if the error did not affect the target sensor type.
Split sch5627_update_device() to prevent that.
Tested on a Fujitsu Esprimo P720.
Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210411164225.11967-3-W_Armin@gmx.de
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
hwmon_device_register() is deprecated.
Convert driver to use hwmon_device_register_with_info() and
remove sysfs attributes which are now being handled by the
hwmon subsystem.
Channel handling was inspired by corsair-cpro.
Tested on a Fujitsu Esprimo P720.
Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210411164225.11967-2-W_Armin@gmx.de
[groeck: Replaced 0 with NULL]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
A pattern match for hardware monitoring registration functions ensures
that hardware monitoring maintainers are copied whenever hardware
monitoring drivers are added to the tree.
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Add support for fan drawer's capability and present registers in order
to set mapping between the fan drawers and tachometers. Some systems
are equipped with fan drawers with one tachometer inside. Others with
fan drawers with several tachometers inside. Using present register
along with tachometer-to-drawer mapping allows to skip reading missed
tachometers and expose input for them as zero, instead of exposing
fault code returned by hardware.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Link: https://lore.kernel.org/r/20210322172237.2213584-1-vadimp@nvidia.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Add support for TI TPS53676 controller to the tps53679 pmbus driver
The driver uses the USER_DATA_03 register to figure out how many phases
are enabled and to which channel they are assigned, and sets the number
of pages and phases accordingly.
Signed-off-by: Erik Rosen <erik.rosen@metormote.com>
Link: https://lore.kernel.org/r/20210322193734.75127-3-erik.rosen@metormote.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
The BPA-RS600 is a compact 600W AC to DC removable power supply module.
Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Link: https://lore.kernel.org/r/20210317040231.21490-3-chris.packham@alliedtelesis.co.nz
[groeck: Added bpa-rs600 to index.rst]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
coccicheck complains about the use of snprintf() in sysfs
show functions.
drivers/hwmon/ina3221.c:701:8-16: WARNING: use scnprintf or sprintf
This results in a large number of patch submissions. Fix it all in
one go using the following coccinelle rules. Use sysfs_emit instead
of scnprintf or sprintf since that makes more sense.
@depends on patch@
identifier show, dev, attr, buf;
@@
ssize_t show(struct device *dev, struct device_attribute *attr, char *buf)
{
<...
return
- snprintf(buf, \( PAGE_SIZE \| PAGE_SIZE - 1 \),
+ sysfs_emit(buf,
...);
...>
}
@depends on patch@
identifier show, dev, attr, buf, rc;
@@
ssize_t show(struct device *dev, struct device_attribute *attr, char *buf)
{
<...
rc =
- snprintf(buf, \( PAGE_SIZE \| PAGE_SIZE - 1 \),
+ sysfs_emit(buf,
...);
...>
}
While at it, remove unnecessary braces and as well as unnecessary
else after return statements to address checkpatch warnings in the
resulting patch.
Cc: Zihao Tang <tangzihao1@hisilicon.com>
Cc: Jay Fang <f.fangjian@huawei.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
The hwmon sysfs ABI requires that the `name` property doesn't include
any dashes. But when the pmbus code picks the name up from the device
tree it quite often does. Replace '-' with '_' before registering the
device.
Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Link: https://lore.kernel.org/r/20210317040231.21490-2-chris.packham@alliedtelesis.co.nz
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
These are "all-in-one" CPU liquid coolers that can be monitored and
controlled through a proprietary USB HID protocol.
While the models have differently sized radiators and come with varying
numbers of fans, they are all indistinguishable at the software level.
The driver exposes fan/pump speeds and coolant temperature through the
standard hwmon sysfs interface.
Fan and pump control, while supported by the devices, are not currently
exposed. The firmware accepts up to 61 trip points per channel
(fan/pump), but the same set of trip temperatures has to be maintained
for both; with pwmX_auto_point_Y_temp attributes, users would need to
maintain this invariant themselves.
Instead, fan and pump control, as well as LED control (which the device
also supports for 9 addressable RGB LEDs on the CPU water block) are
left for existing and already mature user-space tools, which can still
be used alongside the driver, thanks to hidraw. A link to one, which I
also maintain, is provided in the documentation.
The implementation is based on USB traffic analysis. It has been
runtime tested on x86_64, both as a built-in driver and as a module.
Signed-off-by: Jonas Malaco <jonas@protocubo.io>
Link: https://lore.kernel.org/r/20210319045544.416138-1-jonas@protocubo.io
[groeck: Removed unnecessary spinlock.h include]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Fix the following coccicheck warning:
drivers/hwmon/ina2xx.c:313:8-16: WARNING: use scnprintf or sprintf
drivers/hwmon/ina2xx.c:453:8-16: WARNING: use scnprintf or sprintf
drivers/hwmon/ina2xx.c:484:8-16: WARNING: use scnprintf or sprintf
drivers/hwmon/ina2xx.c:540:8-16: WARNING: use scnprintf or sprintf
Signed-off-by: Zihao Tang <tangzihao1@hisilicon.com>
Signed-off-by: Jay Fang <f.fangjian@huawei.com>
Link: https://lore.kernel.org/r/1615892457-35501-1-git-send-email-f.fangjian@huawei.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
coccinelle complains about
WARNING opportunity for kobj_to_dev()
in several files, resulting in one-by-one patch submissions.
Handle all remaining instances in one go.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Adds support for reading the critical values of the temperature sensors
and the rail sensors (voltage and current) once and caches them. Updates
the naming of the constants following a more clear scheme. Also updates
the documentation and fixes some typos. Updates is_visible and ops_read
functions to be more readable.
The new sensors output of a Corsair HX850i will look like this:
corsairpsu-hid-3-1
Adapter: HID adapter
v_in: 230.00 V
v_out +12v: 12.14 V (crit min = +8.41 V, crit max = +15.59 V)
v_out +5v: 5.03 V (crit min = +3.50 V, crit max = +6.50 V)
v_out +3.3v: 3.30 V (crit min = +2.31 V, crit max = +4.30 V)
psu fan: 0 RPM
vrm temp: +46.2°C (crit = +70.0°C)
case temp: +39.8°C (crit = +70.0°C)
power total: 152.00 W
power +12v: 108.00 W
power +5v: 41.00 W
power +3.3v: 5.00 W
curr +12v: 9.00 A (crit max = +85.00 A)
curr +5v: 8.31 A (crit max = +40.00 A)
curr +3.3v: 1.62 A (crit max = +40.00 A)
Signed-off-by: Wilken Gottwalt <wilken.gottwalt@posteo.net>
Link: https://lore.kernel.org/r/YFNg6vGk3sQmyqgB@monster.powergraphx.local
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Add hardware monitoring support for ST STPDDC60 Unversal Digital
Multicell Controller.
Signed-off-by: Erik Rosen <erik.rosen@metormote.com>
Link: https://lore.kernel.org/r/20210218115249.28513-3-erik.rosen@metormote.com
[groeck: Fixed whitespace error in Makefile]
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
For the STPDDC60 chip, the vout alarm-limits are represented as an offset
relative to the commanded output voltage. This means that the limits are
dynamic and must not be cached by the pmbus driver. This patch adds a
pmbus_set_sensor() function to pmbus_core to be able to set the update flag
on selected sensors after auto-detection of limit attributes.
Signed-off-by: Erik Rosen <erik.rosen@metormote.com>
Link: https://lore.kernel.org/r/20210218115249.28513-2-erik.rosen@metormote.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Add support for NCT6686D chip used in the Lenovo P620.
Signed-off-by: Jiqi Li <lijq9@lenovo.com>
Reviewed-by: Mark Pearson <markpearson@lenovo.com>
Link: https://lore.kernel.org/r/20210304104421.1912934-1-lijq9@lenovo.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Changes the way how LINEAR11 values are calculated. The new method
increases the precision of 2-3 digits.
old method:
corsairpsu-hid-3-1
Adapter: HID adapter
v_in: 230.00 V
v_out +12v: 12.00 V
v_out +5v: 5.00 V
v_out +3.3v: 3.00 V
psu fan: 0 RPM
vrm temp: +44.0°C
case temp: +37.0°C
power total: 152.00 W
power +12v: 112.00 W
power +5v: 38.00 W
power +3.3v: 5.00 W
curr in: N/A
curr +12v: 9.00 A
curr +5v: 7.00 A
curr +3.3v: 1000.00 mA
new method:
corsairpsu-hid-3-1
Adapter: HID adapter
v_in: 230.00 V
v_out +12v: 12.16 V
v_out +5v: 5.01 V
v_out +3.3v: 3.30 V
psu fan: 0 RPM
vrm temp: +44.5°C
case temp: +37.8°C
power total: 148.00 W
power +12v: 108.00 W
power +5v: 37.00 W
power +3.3v: 4.50 W
curr in: N/A
curr +12v: 9.25 A
curr +5v: 7.50 A
curr +3.3v: 1.50 A
Co-developed-by: Jack Doan <me@jackdoan.com>
Signed-off-by: Jack Doan <me@jackdoan.com>
Signed-off-by: Wilken Gottwalt <wilken.gottwalt@posteo.net>
Link: https://lore.kernel.org/r/YDoSMqFbgoTXyoru@monster.powergraphx.local
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
fixed the following coccicheck:
./drivers/hwmon/hwmon.c:82:60-61: WARNING opportunity for kobj_to_dev()
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Link: https://lore.kernel.org/r/1614071667-5665-1-git-send-email-yang.lee@linux.alibaba.com
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Also use regmap for register caching. This change reduces code and
data size by more than 40%.
While at it, fixed some warnings reported by checkpatch.
Cc: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
Reviewed-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Tested-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
We only use the pointer to i2c_client to access &client->dev.
Store the device pointer directly instead of retrieving it
from i2c_client.
Cc: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
Reviewed-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Tested-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Not detecting a chip in the detect function is normal and should not
generate any log messages, much less error messages.
Cc: Chris Packham <Chris.Packham@alliedtelesis.co.nz>
Reviewed-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Tested-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
The few quirks that deal with NO_MSI tend to be copy-paste heavy.
Refactor them so that the hierarchy of conditions is slightly
cleaner.
Link: https://lore.kernel.org/r/20210330151145.997953-15-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
We have now three ways of ending up with NO_MSI being set.
Document them.
Link: https://lore.kernel.org/r/20210330151145.997953-14-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Some Mediatek host bridges cannot handle MSIs, which is sad.
This also results in an ugly warning at device probe time,
as the core PCI code wasn't told that MSIs were not available.
Advertise this fact to the rest of the core PCI code by
using the 'msi_domain' attribute, which still opens the possibility
for another block to provide the MSI functionnality.
[maz: commit message, switched over to msi_domain attribute]
Link: https://lore.kernel.org/r/20210330151145.997953-13-maz@kernel.org
Reported-by: Frank Wunderlich <frank-w@public-files.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
The generic PCI host driver relies on MSI domains for MSIs to
be provided to its end-points. Make this dependency explicit.
This cures the warnings occuring on arm/arm64 VMs when booted
with PCI virtio devices and no MSI controller (no GICv3 ITS,
for example).
It is likely that other drivers will need to express the same
dependency.
Link: https://lore.kernel.org/r/20210330151145.997953-12-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
There is a whole class of host bridges that cannot know whether
MSIs will be provided or not, as they rely on other blocks
to provide the MSI functionnality, using MSI domains. This is
the case for example on systems that use the ARM GIC architecture.
Introduce a new attribute ('msi_domain') indicating that implicit
dependency, and use this property to set the NO_MSI flag when
no MSI domain is found at probe time.
Link: https://lore.kernel.org/r/20210330151145.997953-11-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
msi_controller had a good, long life as the abstraction for
a driver providing MSIs to PCI devices. But it has been replaced
in all drivers by the more expressive generic MSI framework.
Farewell, struct msi_controller.
Link: https://lore.kernel.org/r/20210330151145.997953-9-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
As there is no driver using msi_controller, we can now safely
remove its use from the PCI probe code.
Link: https://lore.kernel.org/r/20210330151145.997953-8-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
The Hyper-V PCI driver still makes use of a msi_controller structure,
but it looks more like a distant leftover than anything actually
useful, since it is initialised to 0 and never used for anything.
Just remove it.
Link: https://lore.kernel.org/r/20210330151145.997953-7-maz@kernel.org
Tested-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
In anticipation of the removal of the msi_controller structure, convert
the ancient xilinx host controller driver to MSI domains.
We end-up with the usual two domain structure, the top one being a
generic PCI/MSI domain, the bottom one being xilinx-specific and handling
the actual HW interrupt allocation.
This allows us to fix some of the most appaling MSI programming, where
the message programmed in the device is the virtual IRQ number instead
of the allocated vector number. The allocator is also made safe with
a mutex. This should allow support for MultiMSI, but I decided not to
even try, since I cannot test it.
Link: https://lore.kernel.org/r/20210330151145.997953-6-maz@kernel.org
Tested-by: Bharat Kumar Gogada <bharatku@xilinx.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
