HWE Next

~canonical-hwe-team/hwe-next/+git/sof:mtl-005-hotfix3

  1. Git
  2. lp:~canonical-hwe-team/hwe-next/+git/sof
  3. mtl-005-hotfix3
Last commit made on 2023-07-12
Get this branch:
git clone -b mtl-005-hotfix3 https://git.launchpad.net/~canonical-hwe-team/hwe-next/+git/sof

Branch merges

Branch information

Name:
mtl-005-hotfix3
Repository:
lp:~canonical-hwe-team/hwe-next/+git/sof

Recent commits

479c794... by "Dobrowolski, PawelX" <email address hidden>

iadk_modules refactor

Both native sof loadable modules and IADK are using same:
 - loading flow
 - registering
 - module API
That is why iadk prefix is removed from common parts.

Signed-off-by: Dobrowolski, PawelX <email address hidden>

c66ed3e... by "Dobrowolski, PawelX" <email address hidden>

iadk_modules: module_init: checking type of module

In manifest is information about version of API and
depending on it we load IADK or SOF module.
Flag is showing if module is loadable. According to this information
different api is being used in loading module.

Signed-off-by: Dobrowolski, PawelX <email address hidden>

2cdf38c... by "Dobrowolski, PawelX" <email address hidden>

module_adapter: native_system_agent: Introduce call and interface

This is part of fw which helps loaded module in communication with adsp
Also it provides basic operations for creation of component driver
Common part for native loadable modules and iadk is put together and
remade into native API. That way system services are using generic
methods defined in native system service.

Signed-off-by: Dobrowolski, PawelX <email address hidden>

e2ad42f... by "Dobrowolski, PawelX" <email address hidden>

loadable module: module api version structure

Building system for iadk modules and for native sof modules has built in
information about used api. That information we use to switch between
two different methods of registering modules.

Signed-off-by: Dobrowolski, PawelX <email address hidden>

70c257f... by Trevor Wu <email address hidden>

drivers: mtk: get memif format from config of DAI component

MTK memif(DMA) can support different format from BE dai, so we don't
need to rely on the transformation of SOF DAI. After the patch, the
format of memif follows DAI component, so the format can be kept the
same in the pipeline and do format transformation via AFE if
necessary.

Signed-off-by: Trevor Wu <email address hidden>

734dbd3... by Trevor Wu <email address hidden>

drivers: mtk: add missing argument

One argument is lost in the debug message. Add missing rptr in the
patch.

Signed-off-by: Trevor Wu <email address hidden>

6ddd655... by Rander Wang <email address hidden>

west.yml: upgrade Zephyr revision - Restore IDC interrupt on D3 exit

After exiting D3 state if IMR context save is enabled, IDC interrupt
must be re-enabled again for all cores. It fixes multicore CI test
issue.

431108d89e175: intel_adsp: ace: Restore IDC interrupt on D3 exit

Signed-off-by: Rander Wang <email address hidden>

e23857b... by Yong Zhi <email address hidden>

topology2: nocodec-multicore: test multi-stream capture

Enable more complex pipelines for multi-core use cases.

Signed-off-by: Yong Zhi <email address hidden>

192d1df... by Andy Ross <email address hidden>

ipc: Fix typo in error checking

Found a oss-fuzz error I thought I'd fixed being reported again.
Turns out the fix had a typo and only fixed half the problem space.
Need to check source and sink, not source and source!

Signed-off-by: Andy Ross <email address hidden>

eac223d... by Daniel Baluta <email address hidden>

audio: mixer: Set correct number of sources

After commit 867ada28f we only allocate 1 buffer for mixer even if there
could be 2 input sources.

So, set correct number of sources for mixer.

Fixes 867ada28f ("audio: module_adapter: Allocate max number of input/output buffers")
Signed-off-by: Daniel Baluta <email address hidden>