Use GENMASK along with FIELD_PREP and FIELD_GET to access bitfields in
registers to straighten register access and to drop a lot of defines.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-6-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
No need to have defines for the mask of single bits. Also shift is
unused. Drop all these unnecessary defines.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-5-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
All that the .set_sysclk hook in the micfil driver does is to pass
the sysclk frequency to fsl_micfil_set_mclk_rate(). This function
expects the sample rate as argument though, not any kind of sysclk
frequency. The resulting rate setting of the clock is overwritten
in hw_params anyway, so drop this altogether.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-4-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
In get_pdm_clk() REG_MICFIL_CTRL2 is read, but the result is never used.
Drop the unused code.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-3-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
in get_pdm_clk() REG_MICFIL_CTRL2 is read twice. Drop second read.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20220414162249.3934543-2-s.hauer@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>
Using pm_runtime_resume_and_get is more appropriate
for simplifing code
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Link: https://lore.kernel.org/r/20220412071030.2532230-1-chi.minghao@zte.com.cn
Signed-off-by: Mark Brown <broonie@kernel.org>
Using pm_runtime_resume_and_get is more appropriate
for simplifing code
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220418110259.2559144-1-chi.minghao@zte.com.cn
Signed-off-by: Mark Brown <broonie@kernel.org>
Each module on DSP side serves a processing purpose. Depending on its
purpose, it needs different information during its initialization. Add
functions responsible for creating instances of specific module types
given the information coming from the topology file.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-15-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
When binding modules to pins other than pin0, sometimes additional
preparations need to be made, depending on the module type.
Add function that prepares modules when necessary before binding them.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-14-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Creating the pipelines and instantiating the modules alone is
insufficient to have a fully operational stream. Before it can be run,
stream components need to be bound. Add arming functions to ensure all
necessary operations are completed before path is yielded back to the
avs_path_create() caller.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-13-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add functions to ease with state changing of all objects found in the
path. Each represents either a BIND/UNBIND or SET_PIPELINE_STATE IPC.
DSP pipelines follow simple state machine scheme:
CREATE -> RESET -> PAUSE -> RUNNING -> PAUSE -> RESET -> DELETE
There is no STOP, PAUSE serves that purpose instead.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-12-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
To implement ASoC PCM operations, DSP path handling is needed. With path
template concept present, information carried by topology file can be
converted into runtime path representation. Each may be composed of
several pipelines and each pipeline can contain a number of processing
modules inside. Number of templates and variants found within topology
may vastly outnumber the total amount of pipelines and modules supported
by AudioDSP firmware simultaneously (in runtime) so none of the IDs are
specified in the topology. These are assigned dynamically when needed
and account for limitations described by FIRMWARE_CONFIG and
HARDWARE_CONFIG basefw parameters.
Paths are created on ->hw_params() and are freed on ->hw_free() ALSA PCM
operations. This choice is based on firmware expectations - need for
complete set of information when attempting to instantiate pipelines and
modules on AudioDSP side. With DMA and audio format provided, search
mechanism tests all path variants available in given path template until
a matching variant is found. Once found, information already available
is combined with all avs_tplg_* pieces pointed by matching path variant.
This finally allows to begin a cascade of IPCs which goal is to reserve
resources and prepare DSP for upcoming audio streaming.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-11-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Declare representatives for all crucial elements which stream on ADSP
side is made of. That covers pipelines and modules subject which are
presented by struct avs_path_pipeline and avs_path_module respectively.
While struct avs_path_binding and struct avs_path do not represent any
object on firmware side directly, they are needed to help track the
interconnections and membership of every pipeline and module created.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-10-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Allow topology to specify formatted strings so machine board's
predefined ->link_mask can be used to specify SSP port number
automatically.
This is done to help reduce the amount of topology files as many I2S
configurations contain codec of the same type with little to no
differences in number of scenarios supported - the only difference being
the port number codec is connected to.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-9-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
AVS topology is split into two major parts: dictionaries - found within
ASoC topology manifest - and path templates.
Add custom handlers for a range of operations available in struct
snd_soc_tplg_ops to allow for actually loading the topology file.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-8-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Path template is similar to path description found in skylake-driver
and it describes how given path shall look like in runtime - number of
modules and pipelines that shape it and how they are laid out. It is
tied to DAPM widget's (representing either a FE or a BE) private data.
Depending on the number of audio formats supported, each path template
may carry one or more descriptions of given path. During runtime, when
audio format is known, description matching said format is selected and
used when instantiating path on ADSP firmware side through IPCs.
Add parsing helpers to support loading such information from the
topology file.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-7-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Shape of a path on DSP side, that is, the number and the layout of its
pipelines and modules is paramount for streaming to be efficient and low
power-consuming. Add parsing helpers to support loading such information
from the topology file.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-6-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Path in ADSP firmware is represented by one or more pipelines. Just like
modules, these are described by a config structure. Add parsing helpers
to support loading such information from the topology file.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-5-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Anything that goes beyond module base config is an extension config. It
covers all fields for all specific module types available in ADSP
firmware. Add parsing helpers to support loading such information from
the topology file.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-4-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
AVS topology is split into two major parts: dictionaries - found within
ASoC topology manifest - and path templates.
Dictionaries job is to reduce the total amount of memory
occupied by topology elements. Rather than having every pipeline and
module carry its own information, each refers to specific entry in
specific dictionary by provided (from topology file) indexes. In
consequence, most struct avs_tplg_xxx are made out of pointers.
To support the above, range of parsing helpers for all value-types known
to ALSA: uuid, bool, byte, short, word and string are added. Additional
handlers help translate pointer-types and more complex objects such as
audio formats and module base configs.
Path templates are similar to path descriptions found in skylake-driver
and they describe how given path shall look like in runtime - number of
modules and pipelines that shape it and how they are laid out.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-3-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Expose all vendor tokens that help shape AVS topology. Parsing helpers
introduced in follow up patches make use of these to know which block
they are currently dealing with and to verify their correctness.
Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220331135246.993089-2-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch adds an IP identifier for each Intel platform. The
identifier will be used to select different code branches or
constants.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-16-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Expose the code loader functions for re-use in new platforms
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-15-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add the rom_status_reg field to struct sof_intel_dsp_desc and define
it for HDA platforms. This will be used to check the ROM status during
FW boot.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-14-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Define the check_ipc_irq op for HDA platforms and use it
when checking if it is an IPC IRQ.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-13-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
BIT() macros should be used for all ADSPIC/IS registers.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-12-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Use the definition of the HDA_CL_STREAM_FORMAT macro from the header
file.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-11-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The dsp_ops are mostly common between platforms. Introduce a common
structure and an init function to set platform-specific values.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-10-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Now that we start having multiple platforms with minor variants, the
use of the const qualifier for 'dsp_ops' is starting to be
sub-optimal: the structures are copied across platforms, with only a
couple of members that differ.
This patch removes the const qualifier without any functionality
changes, and adds an optional initialization callback. In follow-up
patches, the dsp_ops will revisited for Intel HDaudio platforms, with
the differences added programmatically over a common baseline.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-9-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
If the topology filename is modified at a higher level, be it with a
DMI quirk or a kernel module parameter, we don't want to use the
default 'nocodec' topology name extracted from descriptors.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-8-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The existing 'tplg_path' module parameter can be used to load
alternate firmware files, be it for development or to handle
OEM-specific or board-specific releases. However the topology filename
is either hard-coded in machine descriptors or modified by specific
DMI-quirks.
For additional flexibility, this patch adds the 'tplg_filename' module
parameter to override topology names.
To avoid any confusion between DMI- and parameter-override, a variable
rename is added.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-7-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The existing 'fw_path' module parameter can be used to load alternate
firmware files, be it for development or to handle OEM-specific or
board-specific releases. The firmware name is however non-modifiable
and defined by platform-specific descriptors.
For additional flexibility during development or enable quirks, this
patch adds the 'fw_filename' module parameter to override default
firmware names.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-6-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
To avoid misleading file names, use different names for INTEL_IPC4
firmware files.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Chao Song <chao.song@intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch adds a default IPC type for each platform, along with file
paths to be used for each IPC type. To make reviews simpler, we only
modify platform descriptors in this table, the information will be
used in the next patch.
The Intel IPCv4 is only supported on Intel platforms after APL, and
not by default. In follow-up patches, support for SKL and KBL will be
added, and in those two cases the IPCv4 will be the default (and only
supported mode).
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
With the addition of the IPCv4, we need the ability to select
different paths for firmware and topologies.
First add an indirection. Follow-up patches will add mechanisms to
select a default IPC or override it.
No functionality change in this patch.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add enum type to allow for different types of IPCs. The IPCv4 is
intended for Intel only as a convergence path with firmware used in
Windows. Follow-up patches will introduce different abstractions with
.ops and different search paths for firmware and topology files.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220414184817.362215-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Tonghao Zhang says:
====================
net: sched: allow user to select txqueue
From: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Patch 1 allow user to select txqueue in clsact hook.
Patch 2 support skbhash to select txqueue.
====================
Link: https://lore.kernel.org/r/20220415164046.26636-1-xiangxia.m.yue@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
This patch fixes issue:
* If we install tc filters with act_skbedit in clsact hook.
It doesn't work, because netdev_core_pick_tx() overwrites
queue_mapping.
$ tc filter ... action skbedit queue_mapping 1
And this patch is useful:
* We can use FQ + EDT to implement efficient policies. Tx queues
are picked by xps, ndo_select_queue of netdev driver, or skb hash
in netdev_core_pick_tx(). In fact, the netdev driver, and skb
hash are _not_ under control. xps uses the CPUs map to select Tx
queues, but we can't figure out which task_struct of pod/containter
running on this cpu in most case. We can use clsact filters to classify
one pod/container traffic to one Tx queue. Why ?
In containter networking environment, there are two kinds of pod/
containter/net-namespace. One kind (e.g. P1, P2), the high throughput
is key in these applications. But avoid running out of network resource,
the outbound traffic of these pods is limited, using or sharing one
dedicated Tx queues assigned HTB/TBF/FQ Qdisc. Other kind of pods
(e.g. Pn), the low latency of data access is key. And the traffic is not
limited. Pods use or share other dedicated Tx queues assigned FIFO Qdisc.
This choice provides two benefits. First, contention on the HTB/FQ Qdisc
lock is significantly reduced since fewer CPUs contend for the same queue.
More importantly, Qdisc contention can be eliminated completely if each
CPU has its own FIFO Qdisc for the second kind of pods.
There must be a mechanism in place to support classifying traffic based on
pods/container to different Tx queues. Note that clsact is outside of Qdisc
while Qdisc can run a classifier to select a sub-queue under the lock.
In general recording the decision in the skb seems a little heavy handed.
This patch introduces a per-CPU variable, suggested by Eric.
The xmit.skip_txqueue flag is firstly cleared in __dev_queue_xmit().
- Tx Qdisc may install that skbedit actions, then xmit.skip_txqueue flag
is set in qdisc->enqueue() though tx queue has been selected in
netdev_tx_queue_mapping() or netdev_core_pick_tx(). That flag is cleared
firstly in __dev_queue_xmit(), is useful:
- Avoid picking Tx queue with netdev_tx_queue_mapping() in next netdev
in such case: eth0 macvlan - eth0.3 vlan - eth0 ixgbe-phy:
For example, eth0, macvlan in pod, which root Qdisc install skbedit
queue_mapping, send packets to eth0.3, vlan in host. In __dev_queue_xmit() of
eth0.3, clear the flag, does not select tx queue according to skb->queue_mapping
because there is no filters in clsact or tx Qdisc of this netdev.
Same action taked in eth0, ixgbe in Host.
- Avoid picking Tx queue for next packet. If we set xmit.skip_txqueue
in tx Qdisc (qdisc->enqueue()), the proper way to clear it is clearing it
in __dev_queue_xmit when processing next packets.
For performance reasons, use the static key. If user does not config the NET_EGRESS,
the patch will not be compiled.
+----+ +----+ +----+
| P1 | | P2 | | Pn |
+----+ +----+ +----+
| | |
+-----------+-----------+
|
| clsact/skbedit
| MQ
v
+-----------+-----------+
| q0 | q1 | qn
v v v
HTB/FQ HTB/FQ ... FIFO
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Jonathan Lemon <jonathan.lemon@gmail.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexander Lobakin <alobakin@pm.me>
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: Talal Ahmad <talalahmad@google.com>
Cc: Kevin Hao <haokexin@gmail.com>
Cc: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Cc: Antoine Tenart <atenart@kernel.org>
Cc: Wei Wang <weiwan@google.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Intel is subdividing the mobile segment with additional models
with the same codename. Using the Intel "N" and "P" suffices
for these will be less confusing than trying to map to some
different naming convention.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/YlS7n7Xtso9BXZA2@agluck-desk3.sc.intel.com
The dmc520 driver requires that at least one interrupt line, out of the
ten possible, is configured. The driver prints an error and returns
-EINVAL from its .probe function if there are no interrupt lines
configured.
Don't print a KERN_ERR level message for each interrupt line that's
unconfigured as that can confuse users into thinking that there is an
error condition.
Before this change, the following KERN_ERR level messages would be
reported if only dram_ecc_errc and dram_ecc_errd were configured in the
device tree:
dmc520 68000000.dmc: IRQ ram_ecc_errc not found
dmc520 68000000.dmc: IRQ ram_ecc_errd not found
dmc520 68000000.dmc: IRQ failed_access not found
dmc520 68000000.dmc: IRQ failed_prog not found
dmc520 68000000.dmc: IRQ link_err not
dmc520 68000000.dmc: IRQ temperature_event not found
dmc520 68000000.dmc: IRQ arch_fsm not found
dmc520 68000000.dmc: IRQ phy_request not found
Fixes: 1088750d78 ("EDAC: Add EDAC driver for DMC520")
Reported-by: Sinan Kaya <okaya@kernel.org>
Signed-off-by: Tyler Hicks <tyhicks@linux.microsoft.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20220111163800.22362-1-tyhicks@linux.microsoft.com
The memory size of ip_vs_conn_tab changed after we use hlist
instead of list.
Fixes: 731109e784 ("ipvs: use hlist instead of list")
Signed-off-by: Pengcheng Yang <yangpc@wangsu.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
When the user runs:
bridge link set dev $br_port mcast_flood on
this command should affect not only L2 multicast, but also IPv4 and IPv6
multicast.
In the Ocelot switch, unknown multicast gets flooded according to
different PGIDs according to its type, and PGID_MC only handles L2
multicast. Therefore, by leaving PGID_MCIPV4 and PGID_MCIPV6 at their
default value of 0, unknown IP multicast traffic is never flooded.
Fixes: 421741ea56 ("net: mscc: ocelot: offload bridge port flags to device")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/20220415151950.219660-1-vladimir.oltean@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
"make dtbs_check":
arch/arm/boot/dts/r9a06g032-rzn1d400-db.dt.yaml: timer: compatible: 'oneOf' conditional failed, one must be fixed:
['arm,cortex-a7-timer', 'arm,armv7-timer'] is too long
'arm,cortex-a7-timer' is not one of ['arm,cortex-a15-timer']
'arm,cortex-a7-timer' is not one of ['arm,armv7-timer']
'arm,cortex-a7-timer' is not one of ['arm,armv8-timer']
From schema: Documentation/devicetree/bindings/timer/arm,arch_timer.yaml
The Cortex-A7 timer should just declare compatibility with
"arm,armv7-timer".
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/a8e0cf00a983b4c539cdb1cfad5cc6b10b423c5b.1649680220.git.geert+renesas@glider.be
Document Renesas RZ/G2UL pinctrl bindings. RZ/G2UL GPIO block is
almost identical to RZ/G2L and has lesser pins compared to RZ/G2L.
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20220315152717.20045-1-biju.das.jz@bp.renesas.com
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
