It seems that since inception, this driver never called mutex_destroy in
_remove. Use devm to handle this automatically.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://patch.msgid.link/20241030203727.6039-3-rosenp@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Enqueue packets in dql after dma engine starts causes race condition.
Tx transfer starts once dma engine is started and may execute dql dequeue
in completion before it gets queued. It results in following kernel crash
while running iperf stress test:
kernel BUG at lib/dynamic_queue_limits.c:99!
<snip>
Internal error: Oops - BUG: 00000000f2000800 [#1] SMP
pc : dql_completed+0x238/0x248
lr : dql_completed+0x3c/0x248
Call trace:
dql_completed+0x238/0x248
axienet_dma_tx_cb+0xa0/0x170
xilinx_dma_do_tasklet+0xdc/0x290
tasklet_action_common+0xf8/0x11c
tasklet_action+0x30/0x3c
handle_softirqs+0xf8/0x230
<snip>
Start dmaengine after enqueue in dql fixes the crash.
Fixes: 6a91b846af ("net: axienet: Introduce dmaengine support")
Signed-off-by: Suraj Gupta <suraj.gupta2@amd.com>
Link: https://patch.msgid.link/20241030062533.2527042-2-suraj.gupta2@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use ip4h_dscp() to get the DSCP from the IPv4 header, then convert the
dscp_t value to __u8 with inet_dscp_to_dsfield().
Then, when we'll convert .flowi4_tos to dscp_t, we'll just have to drop
the inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/6be084229008dcfa7a4e2758befccfd2217a331e.1730294788.git.gnault@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use ip4h_dscp() to get the DSCP from the IPv4 header, then convert the
dscp_t value to __u8 with inet_dscp_to_dsfield().
Then, when we'll convert .flowi4_tos to dscp_t, we'll just have to drop
the inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/f48335504a05b3587e0081a9b4511e0761571ca5.1730292157.git.gnault@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pass the "mdio" child node directly to `macb_mdiobus_register` to avoid
performing the node lookup twice.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20241030085224.2632426-1-o.rempel@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The driver and description indicate "snps,kbbe" is a boolean, not an
uint32.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://patch.msgid.link/20241101211331.24605-2-robh@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This commit pushes the grace-period-end checks further down into
rcu_dump_cpu_stacks(), and also uses lockless checks coupled with
finer-grained locking.
The result is that the current leaf rcu_node structure's ->lock is
acquired only if a stack backtrace might be needed from the current CPU,
and is held across only that CPU's backtrace. As a result, if there are
no stalled CPUs associated with a given rcu_node structure, then its
->lock will not be acquired at all. On large systems, it is usually
(though not always) the case that a small number of CPUs are stalling
the current grace period, which means that the ->lock need be acquired
only for a small fraction of the rcu_node structures.
[ paulmck: Apply Dan Carpenter feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
The check on ret has already been performed a few statements earlier
and ret has not been re-assigned and so the re-checking is redundant.
Clean up the code by removing the redundant check.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://patch.msgid.link/20241031135042.3250614-1-colin.i.king@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Introduce port_setup_tc callback in mt7530 dsa driver in order to enable
dsa ports rate shaping via hw Token Bucket Filter (TBF) for hw switched
traffic.
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20241031-mt7530-tc-offload-v2-1-cb242ad954a0@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In the previous implementation, vf_state is allocated memory only when VF
is enabled. However, net_device_ops::ndo_set_vf_mac() may be called before
VF is enabled to configure the MAC address of VF. If this is the case,
enetc_pf_set_vf_mac() will access vf_state, resulting in access to a null
pointer. The simplified error log is as follows.
root@ls1028ardb:~# ip link set eno0 vf 1 mac 00:0c:e7:66:77:89
[ 173.543315] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000004
[ 173.637254] pc : enetc_pf_set_vf_mac+0x3c/0x80 Message from sy
[ 173.641973] lr : do_setlink+0x4a8/0xec8
[ 173.732292] Call trace:
[ 173.734740] enetc_pf_set_vf_mac+0x3c/0x80
[ 173.738847] __rtnl_newlink+0x530/0x89c
[ 173.742692] rtnl_newlink+0x50/0x7c
[ 173.746189] rtnetlink_rcv_msg+0x128/0x390
[ 173.750298] netlink_rcv_skb+0x60/0x130
[ 173.754145] rtnetlink_rcv+0x18/0x24
[ 173.757731] netlink_unicast+0x318/0x380
[ 173.761665] netlink_sendmsg+0x17c/0x3c8
Fixes: d4fd0404c1 ("enetc: Introduce basic PF and VF ENETC ethernet drivers")
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://patch.msgid.link/20241031060247.1290941-2-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net_dim() is currently passed a struct dim_sample argument by value.
struct dim_sample is 24 bytes. Since this is greater 16 bytes, x86-64
passes it on the stack. All callers have already initialized dim_sample
on the stack, so passing it by value requires pushing a duplicated copy
to the stack. Either witing to the stack and immediately reading it, or
perhaps dereferencing addresses relative to the stack pointer in a chain
of push instructions, seems to perform quite poorly.
In a heavy TCP workload, mlx5e_handle_rx_dim() consumes 3% of CPU time,
94% of which is attributed to the first push instruction to copy
dim_sample on the stack for the call to net_dim():
// Call ktime_get()
0.26 |4ead2: call 4ead7 <mlx5e_handle_rx_dim+0x47>
// Pass the address of struct dim in %rdi
|4ead7: lea 0x3d0(%rbx),%rdi
// Set dim_sample.pkt_ctr
|4eade: mov %r13d,0x8(%rsp)
// Set dim_sample.byte_ctr
|4eae3: mov %r12d,0xc(%rsp)
// Set dim_sample.event_ctr
0.15 |4eae8: mov %bp,0x10(%rsp)
// Duplicate dim_sample on the stack
94.16 |4eaed: push 0x10(%rsp)
2.79 |4eaf1: push 0x10(%rsp)
0.07 |4eaf5: push %rax
// Call net_dim()
0.21 |4eaf6: call 4eafb <mlx5e_handle_rx_dim+0x6b>
To allow the caller to reuse the struct dim_sample already on the stack,
pass the struct dim_sample by reference to net_dim().
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Arthur Kiyanovski <akiyano@amazon.com>
Reviewed-by: Louis Peens <louis.peens@corigine.com>
Link: https://patch.msgid.link/20241031002326.3426181-2-csander@purestorage.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Make the start and end arguments to dim_calc_stats() const pointers
to clarify that the function does not modify their values.
Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Arthur Kiyanovski <akiyano@amazon.com>
Link: https://patch.msgid.link/20241031002326.3426181-1-csander@purestorage.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add functionality to inject a specified amount of current to the heating
plate before the start of the gas measurement to allow the sensor to reach
faster to the requested temperature.
Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com>
Link: https://patch.msgid.link/20241102131311.36210-5-vassilisamir@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Add triggered buffer and soft timestamp support. The available scan mask
enables all the channels of the sensor in order to follow the operation of
the sensor. The sensor basically starts to capture from all channels
as long as it enters into FORCED mode.
The bulk read, reads a total of 15 registers from the sensor, 0x1D..0x2B.
Even though some of those registers are not reported in the register map
of the device, this is how the BME680 Sensor API [1] proposes to do it.
This allows to have one bulk read instead of multiple ones.
Link: https://github.com/boschsensortec/BME68x_SensorAPI/blob/v4.4.8/bme68x.c#L1200
Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com>
Link: https://patch.msgid.link/20241102131311.36210-4-vassilisamir@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Add SCALE,RAW channels to the device. Even though PROCESSED should be
kept for backwards compatibility add comment to avoid using it if the
value is not actually reported in IIO values.
Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com>
Link: https://patch.msgid.link/20241102131311.36210-3-vassilisamir@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Refactorize the set_mode() function to use an external enum that
describes the possible modes of the BME680 device instead of using
true/false variables for selecting SLEEPING/FORCED mode.
Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com>
Link: https://patch.msgid.link/20241102131311.36210-2-vassilisamir@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Make IIO_EVENT_CODE "private" by adding a leading underscore.
There are no more users of this macro in the kernel so we can make it
"private" and encourage developers to use the specialized versions of
the macro instead.
Signed-off-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20241101-iio-fix-event-macro-use-v1-3-0000c5d09f6d@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The IIO_DIFF_EVENT_CODE macro helper was introduced to provide a more
specific alternative to the generic IIO_EVENT_CODE macro for handling
differential channels. This commit updates the code to use
IIO_DIFF_EVENT_CODE for better clarity and maintainability.
However, the current implementation incorrectly sets both chan1 and
chan2 to 0. To maintain compatibility and avoid breaking existing
user space applications, this behavior is preserved for now.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Reviewed-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20241028-iio-add-macro-for-even-identifier-for-differential-channels-v1-2-b452c90f7ea6@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Currently, there are 3 helper macros in iio/events.h to create event
identifiers:
- IIO_EVENT_CODE : create generic event identifier for differential and non
differential channels
- IIO_MOD_EVENT_CODE : create event identifier for modified (non
differential) channels
- IIO_UNMOD_EVENT_CODE : create event identifier for unmodified (non
differential) channels
For differential channels, drivers are expected to use IIO_EVENT_CODE.
However, only one driver in drivers/iio currently uses it correctly,
leading to inconsistent event identifiers for differential channels that
don’t match the intended attributes (such as max1363.c that supports
differential channels, but only uses IIO_UNMOD_EVENT_CODE).
To prevent such issues in future drivers, a new helper macro,
IIO_DIFF_EVENT_CODE, is introduced to specifically create event identifiers
for differential channels. Only one helper is needed for differential
channels since they cannot have modifiers.
Additionally, the descriptions for IIO_MOD_EVENT_CODE and
IIO_UNMOD_EVENT_CODE have been updated to clarify that they are intended
for non-differential channels,
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Reviewed-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20241028-iio-add-macro-for-even-identifier-for-differential-channels-v1-1-b452c90f7ea6@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The Allegro MicroSystems ALS31300 is a 3-D Linear Hall Effect Sensor
mainly used for 3D head-on motion sensing applications.
The device is configured over I2C, and as part of the Sensor data the
temperature core is also provided.
While the device provides an IRQ gpio, it depends on a configuration
programmed into the internal EEPROM, thus only the default mode is
supported and buffered input via trigger is also supported to allow
streaming values with the same sensing timestamp.
The device can be configured with different sensitivities in factory,
but the sensitivity value used to calculate value into the Gauss
unit is not available from registers, thus the sensitivity is provided
by the compatible/device-id string which is based on the part number
as described in the datasheet page 2.
Reviewed-by: Andy Shevchenko <andy@kernel.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patch.msgid.link/20241030-topic-input-upstream-als31300-v4-3-494297c9e50a@linaro.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Document the bindings for the Allegro MicroSystems ALS31300 3-D Linear
Hall Effect Sensor controller by an I2C interface, mainly used in 3D
head-on motion sensing applications.
The device can be configured with different sensitivities in factory,
but the sensitivity value used to calculate value into the Gauss unit
is not available from registers, thus the sensitivity is provided by
the compatible/device-id string which is based on the part number as
described in the datasheet page 2.
Datasheet: https://www.allegromicro.com/-/media/files/datasheets/als31300-datasheet.pdf
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://patch.msgid.link/20241030-topic-input-upstream-als31300-v4-2-494297c9e50a@linaro.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Since the write_event_config callback now uses a bool for the state
parameter, update apds9300_set_intr_state accordingly and change intr_en
to bool.
Also update apds9300_set_power_state and power_state for consistency.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-12-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
write_event_config callback use an int for state, but it is actually a
boolean. iio_ev_state_store is actually using kstrtobool to check user
input, then gives the converted boolean value to write_event_config.
Fix signature and update all iio drivers to use the new signature.
This patch has been partially written using coccinelle with the
following script:
$ cat iio-bool.cocci
// Options: --all-includes
virtual patch
@c1@
identifier iioinfo;
identifier wecfunc;
@@
static const struct iio_info iioinfo = {
...,
.write_event_config =
(
wecfunc
|
&wecfunc
),
...,
};
@@
identifier c1.wecfunc;
identifier indio_dev, chan, type, dir, state;
@@
int wecfunc(struct iio_dev *indio_dev, const struct iio_chan_spec *chan, enum iio_event_type type, enum iio_event_direction dir,
-int
+bool
state) {
...
}
make coccicheck MODE=patch COCCI=iio-bool.cocci M=drivers/iio
Unfortunately, this script didn't match all files:
* all write_event_config callbacks using iio_device_claim_direct_scoped
were not detected and not patched.
* all files that do not assign and declare the write_event_config
callback in the same file.
iio.h was also manually updated.
The patch was build tested using allmodconfig config.
cc: Julia Lawall <julia.lawall@inria.fr>
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-7-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
state parameter is currently an int, but it is actually a boolean.
iio_ev_state_store is actually using kstrtobool to check user input,
then gives the converted boolean value to write_event_config. The code
in adux1020_write_event_config re-uses state parameter to store an
integer value. To prepare for updating the write_event_config signature
to use a boolean for state, introduce a new local int variable.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-6-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
iio_ev_state_store is actually using kstrtobool to check user
input, then gives the converted boolean value to the write_event_config
callback.
Remove useless code in write_event_config callback.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-5-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
iio_ev_state_store is actually using kstrtobool to check user
input, then gives the converted boolean value to the write_event_config
callback.
Remove useless code in write_event_config callback.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-4-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
iio_ev_state_store is actually using kstrtobool to check user
input, then gives the converted boolean value to the write_event_config
callback.
Remove useless code in write_event_config callback.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-3-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
iio_ev_state_store is actually using kstrtobool to check user
input, then gives the converted boolean value to the write_event_config
callback.
Remove useless code in write_event_config callback.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-2-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
iio_ev_state_store is actually using kstrtobool to check user
input, then gives the converted boolean value to the write_event_config
callback.
Remove useless code in write_event_config callback.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20241031-iio-fix-write-event-config-signature-v2-1-2bcacbb517a2@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Egis620 supports 3 channels: presense, proximity and attention.
Modify the driver so it can read those channels as well.
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://patch.msgid.link/20241101-hpd-v3-5-e9c80b7c7164@chromium.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
We are going to introduce more proximity channels. Make proximity a
indexed channel now, in a simple patch, so the change can be easily
bisected if there are any issues.
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://patch.msgid.link/20241101-hpd-v3-4-e9c80b7c7164@chromium.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Add a new channel type representing if the user's attention state to the
the system. This usually means if the user is looking at the screen or
not.
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://patch.msgid.link/20241101-hpd-v3-3-e9c80b7c7164@chromium.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The HID Usage Table at https://usb.org/sites/default/files/hut1_5.pdf
reserves:
- 0x4b2 for Human Proximity Range
Distance between a human and the computer. Default unit of
measure is meters;
https://www.usb.org/sites/default/files/hutrr39b_0.pdf
- 0x4bd for Human Attention Detected
Human-Presence sensors detect the presence of humans in the sensor’s
field-of-view using diverse and evolving technologies. Some presence
sensors are implemented with low resolution video cameras, which can
additionally track a subject’s attention (i.e. if the
user is ‘looking’ at the system with the integrated sensor).
A Human-Presence sensor, providing a Host with the user’s attention
state, allows the Host to optimize its behavior. For example, to
brighten/dim the system display, based on the user’s attention to the
system (potentially prolonging battery life). Default unit is
true/false;
https://www.usb.org/sites/default/files/hutrr107-humanpresenceattention_1.pdf
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://patch.msgid.link/20241101-hpd-v3-1-e9c80b7c7164@chromium.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The integration time of the veml6070 depends on an external resistor
(called Rset in the datasheet) and the value configured in the IT
field of the command register, whose supported values are 1/2x, 1x,
2x and 4x.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Link: https://patch.msgid.link/20241031-veml6070-integration-time-v4-2-c66da6788256@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The veml6070 provides a configurable integration time by means of an
external resistor (Rset in the datasheet) with values between 75 and
1200 kohms.
Document vishay,rset-ohms to select the integration time.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/20241031-veml6070-integration-time-v4-1-c66da6788256@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
