Add option to override how a core id, specified in SOF topology, is
mapped to host core. This allows to map a multithreaded pipeline
execution to specific range of host CPUs.
Implement the option via environment variable "SOF_HOST_CORE0", which is
understood by the CONFIG_LIBRARY implementation of ll-scheduler.
Document the usage to testbench help. mapping of DSP core ids specified
SOF_HOST_CORE
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
The testbench currently has a global structure that contains
per topology and per pipeline data. Refactor to move data to the
most appropriate structure for use.
This is not the final fix, further refactoring is needed.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
This patch changes the help print function. The print look and details
are improved. Executable behavior with option -h is fixed. Previously
the help text was printed twice.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
This is the first step in running a full DSP topology on multiple testbench
"virtual" cores. Testbench will use threads to virtualise a emulate a core
allowing topologies with more than one core to run simultaneously.
This patch makes the following changes.
1) Adds and passes a topology testbench context to all APIs instead of
relying on some globals.
2) Splits the testbench up into small functions that have a single purpose.
3) Creates a thread for each pipeline which in turn share a virtual core.
4) Adds the command line options to enable testing different cores and
pipelines.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Split pipeline functionality into smaller action to support more
complex use cases. Provide more useful logging.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Add LL scheduler emulation for testbench pipelines. Each core is
implemented as a thread.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
This patch adds to defaults of switch statements error prints
and error values returns to properly error non-supported
things.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
This patch limits file read from reading more than e.g. 48 frames
at time when scheduled every 1 ms for 48 kHz audio. It helps
the testbench to execute copy() operations similarly as in real
firmware and not process multiple periods of data in the same
pipeline copy. Some components contain internal limiting for data
processing amount but most do not.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
The file component new() sets function to s32. It's better to
initialize to default function that just errors if called.
The stream pointer is retrieved from buffer. There is no need to
apply twice list_first_item() function. The sample width is set
with get_sample_bytes(). The switch case for frame_fmt is changed
to use stream. It avoids an error to use 16 bit file write to
consume data from 32 bit buffer. This happens because ipc value is
from topology while stream format is from command line override. Only
the text file output becomes incorrect.
The comp_data struct contained unnecessary fields such as period_bytes,
frame_bytes, and frame_fmt.
The patch also contains some switch-case code cleanup for more compact
code and look.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
This patch replaces the sample by sample fread() and fwrite()
operations with max length block reads and writes without
circular wrap. Both binary and text format read and write
functions are replaced with more compact versions.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
This patch adds to command line switch -n for output channels count.
Existing -c is for in channels, new -n is for out channels count. Out
channels count is same as input if -n is not present. Switch -q is
added to quiet the trace output if it is not needed.
The Matlab language test scripts for components are updated to use
the -t config.sh interface of comp_run.sh. It allows more flexible
control of input and output streams.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
This commit:
- adds flags argument into scheduler_free function pointer in
scheduler_ops struct;
- adds SOF_SCHEDULER_FREE_IRQ_ONLY flag, which indicates to disable
only interrupts in scheduler_free() functions.
Signed-off-by: Bartosz Kokoszko <bartoszx.kokoszko@linux.intel.com>
Previously the file component stored component data to device
private data similarly as processing components. However the
pipeline treats file as DAI. Without this patch the missing driver
and DAI operation get_init_delay_ms() causes a segfault in pipeline
parameters walk. It is sufficient for the walk to pass to have just
rzalloc() cleared pointers in driver structure.
The file component data is now placed into DAI private data.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
Without CMP0079 we cannot conditionally include libraries against SOF in
sub directories without seriously restructuring the project. This is
because the old policy requires the link target must be created in the
same folder. This does not work well from a configuration standpoint for
3P audio libraries trying to keep their config in src/audio/*. Rather
than enable the policy, lets simply upgrade since 3.13 is widely
available.
With this upgrade we can also remove the two version dependent checks at
the top of our scripts.
Signed-off-by: Curtis Malainey <cujomalainey@chromium.org>
When the firmware receives a START or RELEASE IPC message, it
immediately triggers all involved components, which starts DMA.
Then it schedules the pipeline task, but since the scheduler can be
already running at that time, the task might be scheduled when DMA
data isn't available yet or has already overflowed. To fix this
change the control flow to also trigger all components from the task
during its first run. Actual data processing then begins with the
next period. Note, that this is currently only possible with
pipelines, using timer-based scheduling. Pipelines, using DMA
interrupts for scheduling are unaffected.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
We need to split the START trigger command into two commands because
some components have a long delay inside their START handling. This
patch introduces two new trigger commands: PRE_START and
PRE_RELEASE and a new state PRE_ACTIVE to prepare for that split.
For simmetry POST_STOP and POST_PAUSE are also added, however they
aren't used yet.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
This patch prevents topology parse failure when there are multiple
controls, e.g. enum controls.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
Buffer and params not initialised to zero like the other IPC structures
used by testbench. Fix this.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
This patch adds the initial IPC4 messaging structures as used by existing
host middle ware. These are mapped onto existing pipeline APIs to support
creation of certain pipeline elements using IPC4.
This patch only upstream the interfaces, the IPC handler is to follow.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Currently the DAI HW configuration is tightly bound to the IPC major
version and IPC structures. Provide a mechanism whereby different
IPC data structures can be passed for DAI configuration.
This change does the following changes.
1) Pass a common ipc_config_dai structure to all dai config call. This
allows retention of common logic that uses this common data.
2) Provide a IPC specific private data pointer to the dai config that
can be interpreted by the DAI as custom data. Today this is the
existing IPC, but it could also support an NHLT binary register blob in
the future.
3) Splits ipc specific code out of src/audio/dai.c and into
src/ipc/ipc3-dai.c.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Make sure component creation is not tightly coupled to a particular IPC
version. This is mostly a mechanical change of structures being passed
to the comp creation APIs away from IPC specific to general structures.
Highlevel changes
1) Pass a common component data object and a component
specific data object during create().
2) Mark the component IPC derived data as "ipc_config" within the
component device to help developers track the data source origin.
3) Pass component specific data during creation so that componets
can allocate and copy to thier private data.
4) Comp_dev no longer has component specific data appended to it.
Instead we can store all in the comp private data (and hence use the
compiler to access it rather than by developer access methods).
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Decouple the IPC ABI major version from pipeline creation. This moves
struct sof_ipc_pipe_new from the pipeline and replaces it with it's
members (saving 8 bytes as no header is needed).
Add new feature specific pipeline APIs to configure pipeline at creation.
Additionally align testbench and UTs to API change.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
IPC is aligned on at least 4 bytes, removes compiler warning about packed
data being aligned on 1 bytes.
In file included from /home/lrg/work/sof/sof/tools/testbench/file.c:23:
/home/lrg/work/sof/sof/tools/testbench/include/testbench/file.h:59:1: error: alignment 1 of ‘struct sof_ipc_comp_file’ is less than 4 [-Werror=packed-not-aligned]
59 | } __attribute__((packed));
| ^
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Give the compiler a chance to further optimise IPC data access since it's
all on a 4 byte alignment.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Split IPC APIs out by feature so that the IPC layer to help future
IPC infrastructure changes support more than one IPC ABI MAJOR version.
No code changes here, only code partitioning and Doxygen comments.
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Several files had multiple blank lines even before the removal of
platform_shared_commit(). Fix them with "cat -s"
Signed-off-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Add cmake -DINIT_CONFIG= option that can point at any initial file.
"make clean" does not delete .config any more.
Note reconfiguration does NOT causes recompilation because -imacros
hides the generated .h from CMake's dependency scan. This is not a
regression, that problems exists since -imacros was introduced. At least
it's now possible to "make clean" and rebuild without losing the .config
file.
Fix for #3617
Signed-off-by: Marc Herbert <marc.herbert@intel.com>
Kai Vehmanen