Remove cml_rt1011_rt5682 machine driver #4897
Conversation
brentlu
requested review from
lgirdwood,
plbossart,
ranj063,
kv2019i,
dbaluta,
bardliao and
ujfalusi
as code owners
| RT1011 + RT5682 I2S codec. | ||
| Say Y if you have such a device. | ||
| If unsure select "N". | ||
|
|
There was a problem hiding this comment.
if a .config file only selected this option, then with a make olddefconfig the sof_rt5682 machine would not be selected.
For true backwards compatibility I think we need to keep this Kconfig and use 'imply sof_rt5682' or something to keep the link.
|
SOFCI TEST |
|
The CI tests report a failure on Helios, that's concerning. |
|
SOFCI TEST |
|
CML FW failing to load again in https://sof-ci.01.org/linuxpr/PR4897/build2055/devicetest/index.html so this was not a fluke. |
|
We have a well-known bug where we don't get the kernel logs when the firmware does not load. It's very painful. Please upvote and comment to help prioritize it: I used ssh and I fetch some logs, see below. UTC time (as it always should be) |
|
@brentlu looks like we have a disconnect between machine driver and topology now, so this PR can't be merged until CI tests are ok. |
|
There is something wrong with the dmic dai link. May I know the configuration of the device? Is it running ChromeOS? What's the OS image version? |
|
I can reproduce this issue manually on jf-cml-hel-rt5682-05, this really isn't a CI problem. My guess is that we're now having a mismatch in topology, what's used by this PR is: |
|
Something wrong with dmic01 dai link which does not happen on my side. Does the CI run a clean build since there is change in Makefile? HW: helios |
|
@brentlu @plbossart Edit: is_legacy_cpu is true. That's why It let me wonder if CML_HEL_RT5682 really supports DMIC? |
|
Maybe we could test it after this patch got merged. |
Can you pick this patch on your branch and resubmit @brentlu ? |
I manually tested it on jf-cml-hel-rt5682-05 and it works. |
brentlu
force-pushed
the
helios-upgrade
branch
4 times, most recently
from
b332ca8
to
d82b703
Compare
|
RMMOD snd_soc_acpi_intel_match update the sof_remove script : |
Add support for boards with four ALC1011 amplifiers. Configuration is copied from cml_rt1011_rt5682 machine driver for backward compatibility with existing cml devices. Signed-off-by: Brent Lu <brent.lu@intel.com>
For cml boards, ALC1011 speaker amplifier is supported by machine driver cml_rt1011_rt5682. Use same driver name for backward compatibility with existing devices on market. Signed-off-by: Brent Lu <brent.lu@intel.com>
Delete this driver and use sof_rt5682 machine driver instead. Signed-off-by: Brent Lu <brent.lu@intel.com>
Initialize "is_legacy_cpu" to false to prevent an uninitialized variable bug. Fixes: 8efcd48 ("ASoC: Intel: sof_rt5682: use common module for sof_card_private initialization") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Link: https://msgid.link/r/6e16433f-1897-46a2-b3ee-e177e7d846b8@moroto.mountain Signed-off-by: Mark Brown <broonie@kernel.org>
|
SOFCI TEST |
|
tests look ok now, merging. |
plbossart
merged commit 46537ef
into
thesofproject:topic/sof-dev
This driver supports ALC5682 headphone codec with 2 or 4 ALC1011 speaker amplifier. Move the support to sof_rt5682 and sof_realtek_common module then delete the cml_rt1011_rt5682 machine driver. Tested on CML Chromebook Helios.