Uh oh!
There was an error while loading. Please reload this page.
drivers: intel: remove remaining references to MN driver - #9671
Conversation
kv2019i
commented
Nov 21, 2024
Found in review of #9667 |
wszypelt
commented
Nov 25, 2024
Unfortunately we have a problem on TGL with IPC4, the test with SSP does not pass. The question is what do we do about it or should we throw away this test? I would like it to be agreed with the developers Issue: |
lgirdwood
commented
Nov 25, 2024
@wszypelt pls disable this test. I would not fully delete though, just disable. |
kv2019i
commented
Nov 26, 2024
@lgirdwood@wszypelt This is an expected result that some SSP tests will fail on cAVS2.5. This PR removes the SOF side CONFIG option, but we are still using old Zephyr here. so the SSP driver code for SSP2.5 clocks will be then partially disabled. I expect the tests to pass when Zephyr part is merged and can test together with this PR (and thus the DNM for this). |
cb77aa2 to
a172410Comparekv2019i
commented
Nov 26, 2024
New version pushed:
|
kv2019i
commented
Nov 27, 2024
DSP crashes on ace30, probably caused by latest Zephyr main: |
kv2019i
commented
Nov 28, 2024
Fix is identified for Zephyr upstream regression, but I'll mark this PR with DNM until the Zephyr part is merged and we can try again. |
kv2019i
commented
Dec 5, 2024
I'll rebase this once #9712 is merged. |
The Zephyr SSP driver has had a dependency to CONFIG_INTEL_NM defined in SOF, so this has not been removed before. This is now cleaned up on Zephyr side, so we can proceed to remove the remaining bits of the Intel MN driver on SOF side. Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
a172410 to
3b0271fComparekv2019i
commented
Dec 9, 2024
New version uploaded:
|
lgirdwood
commented
Dec 9, 2024
kv2019i
commented
Dec 9, 2024
@lgirdwood Ack, I think the LNL fails are unrelated. I think the BIOS update broke this DUT and it's loading the wrong config: |
wszypelt
commented
Dec 10, 2024
@kv2019i@lgirdwood Internal Intel CI System, all green |
The Zephyr SSP driver has had dependency to CONFIG_INTEL_NM defined in SOF, so this has not been removed before. This is now cleaned up on Zephyr side, so we can proceed to remove the remaining bits of the Intel MN driver on SOF side.