| Message ID | 20231209205351.880797-12-cristian.ciocaltea@collabora.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | Improve SOF support for Steam Deck OLED | expand |
On 12/10/23 02:23, Cristian Ciocaltea wrote: > Commit efb931cdc4b9 ("ASoC: SOF: topology: Add support for AMD ACP > DAIs") registered "ACP" name for SOF_DAI_AMD_BT DAI type. However, some > boards, i.e. Steam Deck OLED, seem to require "ACPBT" for the same type: > > [ 467.489680] snd_sof_amd_vangogh 0000:04:00.5: ipc tx error for 0x30030000 (msg/reply size: 16/0): -22 > [ 467.492775] snd_sof_amd_vangogh 0000:04:00.5: sof_ipc3_route_setup: route ACPBT2.IN -> BUF5.0 failed > [ 467.495839] snd_sof_amd_vangogh 0000:04:00.5: sof_ipc3_set_up_all_pipelines: route set up failed > [ 467.499128] snd_sof_amd_vangogh 0000:04:00.5: error: tplg component load failed -22 > [ 467.502210] snd_sof_amd_vangogh 0000:04:00.5: error: failed to load DSP topology -22 > [ 467.505289] snd_sof_amd_vangogh 0000:04:00.5: ASoC: error at snd_soc_component_probe on 0000:04:00.5: -22 > [ 467.508430] sof_mach nau8821-max: ASoC: failed to instantiate card -22 > [ 467.511725] sof_mach nau8821-max: error -EINVAL: Failed to register card(sof-nau8821-max) > [ 467.514861] sof_mach: probe of nau8821-max failed with error -22 > > Add "ACPBT" alias for "ACP" SOF DAI type. > > Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com> > --- > sound/soc/sof/topology.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c > index e3e7fbe40fa6..73bf791e7520 100644 > --- a/sound/soc/sof/topology.c > +++ b/sound/soc/sof/topology.c > @@ -290,6 +290,7 @@ static const struct sof_dai_types sof_dais[] = { > {"SAI", SOF_DAI_IMX_SAI}, > {"ESAI", SOF_DAI_IMX_ESAI}, > {"ACP", SOF_DAI_AMD_BT}, > + {"ACPBT", SOF_DAI_AMD_BT}, No need to create the alias name, we can directly modify ACP to ACPBT as ACP is not using anywhere. > {"ACPSP", SOF_DAI_AMD_SP}, > {"ACPDMIC", SOF_DAI_AMD_DMIC}, > {"ACPHS", SOF_DAI_AMD_HS},
On 12/10/23 05:30, Venkata Prasad Potturu wrote: > > On 12/10/23 02:23, Cristian Ciocaltea wrote: >> Commit efb931cdc4b9 ("ASoC: SOF: topology: Add support for AMD ACP >> DAIs") registered "ACP" name for SOF_DAI_AMD_BT DAI type. However, some >> boards, i.e. Steam Deck OLED, seem to require "ACPBT" for the same type: >> >> [ 467.489680] snd_sof_amd_vangogh 0000:04:00.5: ipc tx error for >> 0x30030000 (msg/reply size: 16/0): -22 >> [ 467.492775] snd_sof_amd_vangogh 0000:04:00.5: sof_ipc3_route_setup: >> route ACPBT2.IN -> BUF5.0 failed >> [ 467.495839] snd_sof_amd_vangogh 0000:04:00.5: >> sof_ipc3_set_up_all_pipelines: route set up failed >> [ 467.499128] snd_sof_amd_vangogh 0000:04:00.5: error: tplg component >> load failed -22 >> [ 467.502210] snd_sof_amd_vangogh 0000:04:00.5: error: failed to load >> DSP topology -22 >> [ 467.505289] snd_sof_amd_vangogh 0000:04:00.5: ASoC: error at >> snd_soc_component_probe on 0000:04:00.5: -22 >> [ 467.508430] sof_mach nau8821-max: ASoC: failed to instantiate card -22 >> [ 467.511725] sof_mach nau8821-max: error -EINVAL: Failed to register >> card(sof-nau8821-max) >> [ 467.514861] sof_mach: probe of nau8821-max failed with error -22 >> >> Add "ACPBT" alias for "ACP" SOF DAI type. >> >> Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com> >> --- >> sound/soc/sof/topology.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c >> index e3e7fbe40fa6..73bf791e7520 100644 >> --- a/sound/soc/sof/topology.c >> +++ b/sound/soc/sof/topology.c >> @@ -290,6 +290,7 @@ static const struct sof_dai_types sof_dais[] = { >> {"SAI", SOF_DAI_IMX_SAI}, >> {"ESAI", SOF_DAI_IMX_ESAI}, >> {"ACP", SOF_DAI_AMD_BT}, >> + {"ACPBT", SOF_DAI_AMD_BT}, > No need to create the alias name, we can directly modify ACP to ACPBT as > ACP is not using anywhere. Great, thanks, will do in v2. >> {"ACPSP", SOF_DAI_AMD_SP}, >> {"ACPDMIC", SOF_DAI_AMD_DMIC}, >> {"ACPHS", SOF_DAI_AMD_HS},
On 12/10/23 14:38, Cristian Ciocaltea wrote: > On 12/10/23 05:30, Venkata Prasad Potturu wrote: >> On 12/10/23 02:23, Cristian Ciocaltea wrote: >>> Commit efb931cdc4b9 ("ASoC: SOF: topology: Add support for AMD ACP >>> DAIs") registered "ACP" name for SOF_DAI_AMD_BT DAI type. However, some >>> boards, i.e. Steam Deck OLED, seem to require "ACPBT" for the same type: >>> >>> [ 467.489680] snd_sof_amd_vangogh 0000:04:00.5: ipc tx error for >>> 0x30030000 (msg/reply size: 16/0): -22 >>> [ 467.492775] snd_sof_amd_vangogh 0000:04:00.5: sof_ipc3_route_setup: >>> route ACPBT2.IN -> BUF5.0 failed >>> [ 467.495839] snd_sof_amd_vangogh 0000:04:00.5: >>> sof_ipc3_set_up_all_pipelines: route set up failed >>> [ 467.499128] snd_sof_amd_vangogh 0000:04:00.5: error: tplg component >>> load failed -22 >>> [ 467.502210] snd_sof_amd_vangogh 0000:04:00.5: error: failed to load >>> DSP topology -22 >>> [ 467.505289] snd_sof_amd_vangogh 0000:04:00.5: ASoC: error at >>> snd_soc_component_probe on 0000:04:00.5: -22 >>> [ 467.508430] sof_mach nau8821-max: ASoC: failed to instantiate card -22 >>> [ 467.511725] sof_mach nau8821-max: error -EINVAL: Failed to register >>> card(sof-nau8821-max) >>> [ 467.514861] sof_mach: probe of nau8821-max failed with error -22 >>> >>> Add "ACPBT" alias for "ACP" SOF DAI type. >>> >>> Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com> >>> --- >>> sound/soc/sof/topology.c | 1 + >>> 1 file changed, 1 insertion(+) >>> >>> diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c >>> index e3e7fbe40fa6..73bf791e7520 100644 >>> --- a/sound/soc/sof/topology.c >>> +++ b/sound/soc/sof/topology.c >>> @@ -290,6 +290,7 @@ static const struct sof_dai_types sof_dais[] = { >>> {"SAI", SOF_DAI_IMX_SAI}, >>> {"ESAI", SOF_DAI_IMX_ESAI}, >>> {"ACP", SOF_DAI_AMD_BT}, >>> + {"ACPBT", SOF_DAI_AMD_BT}, >> No need to create the alias name, we can directly modify ACP to ACPBT as >> ACP is not using anywhere. > Great, thanks, will do in v2. This should send to SOF git repo for rewiew, once SOF reviewers approved this, again need to send to broonie git. All the changes in sound/soc/sof/ path should go to SOF git. > >>> {"ACPSP", SOF_DAI_AMD_SP}, >>> {"ACPDMIC", SOF_DAI_AMD_DMIC}, >>> {"ACPHS", SOF_DAI_AMD_HS},
On 12/10/23 11:51, Venkata Prasad Potturu wrote: > > On 12/10/23 14:38, Cristian Ciocaltea wrote: >> On 12/10/23 05:30, Venkata Prasad Potturu wrote: >>> On 12/10/23 02:23, Cristian Ciocaltea wrote: >>>> Commit efb931cdc4b9 ("ASoC: SOF: topology: Add support for AMD ACP >>>> DAIs") registered "ACP" name for SOF_DAI_AMD_BT DAI type. However, >>>> some >>>> boards, i.e. Steam Deck OLED, seem to require "ACPBT" for the same >>>> type: >>>> >>>> [ 467.489680] snd_sof_amd_vangogh 0000:04:00.5: ipc tx error for >>>> 0x30030000 (msg/reply size: 16/0): -22 >>>> [ 467.492775] snd_sof_amd_vangogh 0000:04:00.5: sof_ipc3_route_setup: >>>> route ACPBT2.IN -> BUF5.0 failed >>>> [ 467.495839] snd_sof_amd_vangogh 0000:04:00.5: >>>> sof_ipc3_set_up_all_pipelines: route set up failed >>>> [ 467.499128] snd_sof_amd_vangogh 0000:04:00.5: error: tplg component >>>> load failed -22 >>>> [ 467.502210] snd_sof_amd_vangogh 0000:04:00.5: error: failed to load >>>> DSP topology -22 >>>> [ 467.505289] snd_sof_amd_vangogh 0000:04:00.5: ASoC: error at >>>> snd_soc_component_probe on 0000:04:00.5: -22 >>>> [ 467.508430] sof_mach nau8821-max: ASoC: failed to instantiate >>>> card -22 >>>> [ 467.511725] sof_mach nau8821-max: error -EINVAL: Failed to register >>>> card(sof-nau8821-max) >>>> [ 467.514861] sof_mach: probe of nau8821-max failed with error -22 >>>> >>>> Add "ACPBT" alias for "ACP" SOF DAI type. >>>> >>>> Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com> >>>> --- >>>> sound/soc/sof/topology.c | 1 + >>>> 1 file changed, 1 insertion(+) >>>> >>>> diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c >>>> index e3e7fbe40fa6..73bf791e7520 100644 >>>> --- a/sound/soc/sof/topology.c >>>> +++ b/sound/soc/sof/topology.c >>>> @@ -290,6 +290,7 @@ static const struct sof_dai_types sof_dais[] = { >>>> {"SAI", SOF_DAI_IMX_SAI}, >>>> {"ESAI", SOF_DAI_IMX_ESAI}, >>>> {"ACP", SOF_DAI_AMD_BT}, >>>> + {"ACPBT", SOF_DAI_AMD_BT}, >>> No need to create the alias name, we can directly modify ACP to ACPBT as >>> ACP is not using anywhere. >> Great, thanks, will do in v2. > This should send to SOF git repo for rewiew, once SOF reviewers approved > this, again need to send to broonie git. > All the changes in sound/soc/sof/ path should go to SOF git. Unfortunately I'm not familiar with the SOF dev workflow. So it's not enough to have this patch cc-ed to sound-open-firmware@alsa-project.org? >>>> {"ACPSP", SOF_DAI_AMD_SP}, >>>> {"ACPDMIC", SOF_DAI_AMD_DMIC}, >>>> {"ACPHS", SOF_DAI_AMD_HS},
On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: > On 12/10/23 11:51, Venkata Prasad Potturu wrote: > > This should send to SOF git repo for rewiew, once SOF reviewers approved > > this, again need to send to broonie git. > > All the changes in sound/soc/sof/ path should go to SOF git. > Unfortunately I'm not familiar with the SOF dev workflow. So it's not > enough to have this patch cc-ed to sound-open-firmware@alsa-project.org? The SOF people basically do their own thing in github at https://github.com/thesofproject/linux with a github workflow and submit their patches upstream in batches a few times a release, however my understanding is that their workflow can cope with things going in directly upstream as well.
On 12/10/23 16:01, Mark Brown wrote: > On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >> On 12/10/23 11:51, Venkata Prasad Potturu wrote: > >>> This should send to SOF git repo for rewiew, once SOF reviewers approved >>> this, again need to send to broonie git. >>> All the changes in sound/soc/sof/ path should go to SOF git. > >> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >> enough to have this patch cc-ed to sound-open-firmware@alsa-project.org? > > The SOF people basically do their own thing in github at > > https://github.com/thesofproject/linux > > with a github workflow and submit their patches upstream in batches a > few times a release, however my understanding is that their workflow can > cope with things going in directly upstream as well. Thanks for clarifying, Mark! That would greatly simplify and speedup the whole process, at least for trivial patches like this one.
On 12/10/23 21:20, Cristian Ciocaltea wrote: > On 12/10/23 16:01, Mark Brown wrote: >> On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >>> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>>> This should send to SOF git repo for rewiew, once SOF reviewers approved >>>> this, again need to send to broonie git. >>>> All the changes in sound/soc/sof/ path should go to SOF git. >>> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >>> enough to have this patch cc-ed to sound-open-firmware@alsa-project.org? >> The SOF people basically do their own thing in github at >> >> https://github.com/thesofproject/linux >> >> with a github workflow and submit their patches upstream in batches a >> few times a release, however my understanding is that their workflow can >> cope with things going in directly upstream as well. > Thanks for clarifying, Mark! That would greatly simplify and speedup > the whole process, at least for trivial patches like this one. Hi Cristian, We have created a Pull request in SOF git hub for I2S BT support. please hold v2 version SOF patches till below PR get's merged. PR:- https://github.com/thesofproject/linux/pull/4742
On 10/12/23 19:31, Mark Brown wrote: > On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>> This should send to SOF git repo for rewiew, once SOF reviewers approved >>> this, again need to send to broonie git. >>> All the changes in sound/soc/sof/ path should go to SOF git. >> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >> enough to have this patch cc-ed to sound-open-firmware@alsa-project.org? > The SOF people basically do their own thing in github at > > https://github.com/thesofproject/linux > > with a github workflow and submit their patches upstream in batches a > few times a release, however my understanding is that their workflow can > cope with things going in directly upstream as well. If patches are directly pushed to alsa devel list instead of creating pull request for SOF patches It will break SOF github work flow. Validation across all the platforms is a potential challenge, and it will also create an overhead to pull the patches which got merged in to for-next branch, before the all the patches pulled in to SOF GitHub.
On 12/11/23 07:58, Venkata Prasad Potturu wrote: > > On 12/10/23 21:20, Cristian Ciocaltea wrote: >> On 12/10/23 16:01, Mark Brown wrote: >>> On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >>>> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>>>> This should send to SOF git repo for rewiew, once SOF reviewers >>>>> approved >>>>> this, again need to send to broonie git. >>>>> All the changes in sound/soc/sof/ path should go to SOF git. >>>> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >>>> enough to have this patch cc-ed to >>>> sound-open-firmware@alsa-project.org? >>> The SOF people basically do their own thing in github at >>> >>> https://github.com/thesofproject/linux >>> >>> with a github workflow and submit their patches upstream in batches a >>> few times a release, however my understanding is that their workflow can >>> cope with things going in directly upstream as well. >> Thanks for clarifying, Mark! That would greatly simplify and speedup >> the whole process, at least for trivial patches like this one. > > Hi Cristian, > > We have created a Pull request in SOF git hub for I2S BT support. > please hold v2 version SOF patches till below PR get's merged. > PR:- https://github.com/thesofproject/linux/pull/4742 Hi Venkata, If this is going to be handled via the github workflow, this patch should be removed from the series. Since there is no dependency on it, I cannot see a reason to put v2 on hold. Do I miss something? Thanks, Cristian
On 12/14/23 17:53, Cristian Ciocaltea wrote: > On 12/11/23 07:58, Venkata Prasad Potturu wrote: >> On 12/10/23 21:20, Cristian Ciocaltea wrote: >>> On 12/10/23 16:01, Mark Brown wrote: >>>> On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >>>>> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>>>>> This should send to SOF git repo for rewiew, once SOF reviewers >>>>>> approved >>>>>> this, again need to send to broonie git. >>>>>> All the changes in sound/soc/sof/ path should go to SOF git. >>>>> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >>>>> enough to have this patch cc-ed to >>>>> sound-open-firmware@alsa-project.org? >>>> The SOF people basically do their own thing in github at >>>> >>>> https://github.com/thesofproject/linux >>>> >>>> with a github workflow and submit their patches upstream in batches a >>>> few times a release, however my understanding is that their workflow can >>>> cope with things going in directly upstream as well. >>> Thanks for clarifying, Mark! That would greatly simplify and speedup >>> the whole process, at least for trivial patches like this one. >> Hi Cristian, >> >> We have created a Pull request in SOF git hub for I2S BT support. >> please hold v2 version SOF patches till below PR get's merged. >> PR:- https://github.com/thesofproject/linux/pull/4742 > Hi Venkata, > > If this is going to be handled via the github workflow, this patch > should be removed from the series. Since there is no dependency on it, > I cannot see a reason to put v2 on hold. > > Do I miss something? Yes, need to drop this patch. And it's better to follow the github workflow by creating a Pull request in SOF github for all sof driver related patches, rather than sending patches to broonie git directly. > > Thanks, > Cristian
On 12/14/23 17:53, Cristian Ciocaltea wrote: > On 12/11/23 07:58, Venkata Prasad Potturu wrote: >> On 12/10/23 21:20, Cristian Ciocaltea wrote: >>> On 12/10/23 16:01, Mark Brown wrote: >>>> On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >>>>> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>>>>> This should send to SOF git repo for rewiew, once SOF reviewers >>>>>> approved >>>>>> this, again need to send to broonie git. >>>>>> All the changes in sound/soc/sof/ path should go to SOF git. >>>>> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >>>>> enough to have this patch cc-ed to >>>>> sound-open-firmware@alsa-project.org? >>>> The SOF people basically do their own thing in github at >>>> >>>> https://github.com/thesofproject/linux >>>> >>>> with a github workflow and submit their patches upstream in batches a >>>> few times a release, however my understanding is that their workflow can >>>> cope with things going in directly upstream as well. >>> Thanks for clarifying, Mark! That would greatly simplify and speedup >>> the whole process, at least for trivial patches like this one. >> Hi Cristian, >> >> We have created a Pull request in SOF git hub for I2S BT support. >> please hold v2 version SOF patches till below PR get's merged. >> PR:- https://github.com/thesofproject/linux/pull/4742 > Hi Venkata, > > If this is going to be handled via the github workflow, this patch > should be removed from the series. Since there is no dependency on it, > I cannot see a reason to put v2 on hold. > > Do I miss something? Non-sof driver related patches can directly send to broonie git ad v2 series. SOF driver patches should send to SOF github to avoid merge conflicts as per guidelines of SOF community. > > Thanks, > Cristian
On 12/14/23 15:15, Venkata Prasad Potturu wrote: > > On 12/14/23 17:53, Cristian Ciocaltea wrote: >> On 12/11/23 07:58, Venkata Prasad Potturu wrote: >>> On 12/10/23 21:20, Cristian Ciocaltea wrote: >>>> On 12/10/23 16:01, Mark Brown wrote: >>>>> On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >>>>>> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>>>>>> This should send to SOF git repo for rewiew, once SOF reviewers >>>>>>> approved >>>>>>> this, again need to send to broonie git. >>>>>>> All the changes in sound/soc/sof/ path should go to SOF git. >>>>>> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >>>>>> enough to have this patch cc-ed to >>>>>> sound-open-firmware@alsa-project.org? >>>>> The SOF people basically do their own thing in github at >>>>> >>>>> https://github.com/thesofproject/linux >>>>> >>>>> with a github workflow and submit their patches upstream in batches a >>>>> few times a release, however my understanding is that their >>>>> workflow can >>>>> cope with things going in directly upstream as well. >>>> Thanks for clarifying, Mark! That would greatly simplify and speedup >>>> the whole process, at least for trivial patches like this one. >>> Hi Cristian, >>> >>> We have created a Pull request in SOF git hub for I2S BT support. >>> please hold v2 version SOF patches till below PR get's merged. >>> PR:- https://github.com/thesofproject/linux/pull/4742 >> Hi Venkata, >> >> If this is going to be handled via the github workflow, this patch >> should be removed from the series. Since there is no dependency on it, >> I cannot see a reason to put v2 on hold. >> >> Do I miss something? > Non-sof driver related patches can directly send to broonie git ad v2 > series. > SOF driver patches should send to SOF github to avoid merge conflicts > as per guidelines of SOF community. Honestly, I don't really see a high risk of conflicts, the patches are not that complex and can be simply cherry-picked when needed. Moreover, as we already had people reviewing this, splitting this up will only add confusion and unnecessary burden. Are there any specific changes you are concerned about and cannot be really handled here?
On 12/14/23 22:12, Cristian Ciocaltea wrote: > On 12/14/23 15:15, Venkata Prasad Potturu wrote: >> On 12/14/23 17:53, Cristian Ciocaltea wrote: >>> On 12/11/23 07:58, Venkata Prasad Potturu wrote: >>>> On 12/10/23 21:20, Cristian Ciocaltea wrote: >>>>> On 12/10/23 16:01, Mark Brown wrote: >>>>>> On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >>>>>>> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>>>>>>> This should send to SOF git repo for rewiew, once SOF reviewers >>>>>>>> approved >>>>>>>> this, again need to send to broonie git. >>>>>>>> All the changes in sound/soc/sof/ path should go to SOF git. >>>>>>> Unfortunately I'm not familiar with the SOF dev workflow. So it's not >>>>>>> enough to have this patch cc-ed to >>>>>>> sound-open-firmware@alsa-project.org? >>>>>> The SOF people basically do their own thing in github at >>>>>> >>>>>> https://github.com/thesofproject/linux >>>>>> >>>>>> with a github workflow and submit their patches upstream in batches a >>>>>> few times a release, however my understanding is that their >>>>>> workflow can >>>>>> cope with things going in directly upstream as well. >>>>> Thanks for clarifying, Mark! That would greatly simplify and speedup >>>>> the whole process, at least for trivial patches like this one. >>>> Hi Cristian, >>>> >>>> We have created a Pull request in SOF git hub for I2S BT support. >>>> please hold v2 version SOF patches till below PR get's merged. >>>> PR:- https://github.com/thesofproject/linux/pull/4742 >>> Hi Venkata, >>> >>> If this is going to be handled via the github workflow, this patch >>> should be removed from the series. Since there is no dependency on it, >>> I cannot see a reason to put v2 on hold. >>> >>> Do I miss something? >> Non-sof driver related patches can directly send to broonie git ad v2 >> series. >> SOF driver patches should send to SOF github to avoid merge conflicts >> as per guidelines of SOF community. > Honestly, I don't really see a high risk of conflicts, the patches are > not that complex and can be simply cherry-picked when needed. Moreover, > as we already had people reviewing this, splitting this up will only add > confusion and unnecessary burden. > > Are there any specific changes you are concerned about and cannot be > really handled here? This is not the concern about this patch series, Generally sof driver patches sends to SOF git hub as a PR, these are the guidelines by SOF maintainers. If you still want to send alsa devel list directly, please discuss with SOF maintainers.
On 12/15/23 11:58, Venkata Prasad Potturu wrote: > > On 12/14/23 22:12, Cristian Ciocaltea wrote: >> On 12/14/23 15:15, Venkata Prasad Potturu wrote: >>> On 12/14/23 17:53, Cristian Ciocaltea wrote: >>>> On 12/11/23 07:58, Venkata Prasad Potturu wrote: >>>>> On 12/10/23 21:20, Cristian Ciocaltea wrote: >>>>>> On 12/10/23 16:01, Mark Brown wrote: >>>>>>> On Sun, Dec 10, 2023 at 12:12:53PM +0200, Cristian Ciocaltea wrote: >>>>>>>> On 12/10/23 11:51, Venkata Prasad Potturu wrote: >>>>>>>>> This should send to SOF git repo for rewiew, once SOF reviewers >>>>>>>>> approved >>>>>>>>> this, again need to send to broonie git. >>>>>>>>> All the changes in sound/soc/sof/ path should go to SOF git. >>>>>>>> Unfortunately I'm not familiar with the SOF dev workflow. So >>>>>>>> it's not >>>>>>>> enough to have this patch cc-ed to >>>>>>>> sound-open-firmware@alsa-project.org? >>>>>>> The SOF people basically do their own thing in github at >>>>>>> >>>>>>> https://github.com/thesofproject/linux >>>>>>> >>>>>>> with a github workflow and submit their patches upstream in >>>>>>> batches a >>>>>>> few times a release, however my understanding is that their >>>>>>> workflow can >>>>>>> cope with things going in directly upstream as well. >>>>>> Thanks for clarifying, Mark! That would greatly simplify and speedup >>>>>> the whole process, at least for trivial patches like this one. >>>>> Hi Cristian, >>>>> >>>>> We have created a Pull request in SOF git hub for I2S BT support. >>>>> please hold v2 version SOF patches till below PR get's merged. >>>>> PR:- https://github.com/thesofproject/linux/pull/4742 >>>> Hi Venkata, >>>> >>>> If this is going to be handled via the github workflow, this patch >>>> should be removed from the series. Since there is no dependency on it, >>>> I cannot see a reason to put v2 on hold. >>>> >>>> Do I miss something? >>> Non-sof driver related patches can directly send to broonie git ad v2 >>> series. >>> SOF driver patches should send to SOF github to avoid merge conflicts >>> as per guidelines of SOF community. >> Honestly, I don't really see a high risk of conflicts, the patches are >> not that complex and can be simply cherry-picked when needed. Moreover, >> as we already had people reviewing this, splitting this up will only add >> confusion and unnecessary burden. >> >> Are there any specific changes you are concerned about and cannot be >> really handled here? > This is not the concern about this patch series, > Generally sof driver patches sends to SOF git hub as a PR, these are the > guidelines by SOF maintainers. > If you still want to send alsa devel list directly, please discuss with > SOF maintainers. I think this series makes sense as a whole and it's best to be handled here, as it only provides trivial fixes to issues found on mainline. If the SOF workflow is unable to integrate fixes submitted upstream, I would perceive that as a significant drawback of adhering to that process. It is hard to believe, though, that this is really the case. Hence, I kindly ask everyone here to shed some light and help move this forward. Thank you, Cristian
On Fri, Dec 15, 2023 at 12:57:34PM +0200, Cristian Ciocaltea wrote: > If the SOF workflow is unable to integrate fixes submitted upstream, I > would perceive that as a significant drawback of adhering to that > process. It is hard to believe, though, that this is really the case. As far as I'm aware they can cope fine with that, though it does help if people try to avoid needless collisions. It *does* cause trouble to use both github and upstream flows simultaneously and there's a preference for pushing anything substantial through github but picking one or the other works as far as I know.
diff --git a/sound/soc/sof/topology.c b/sound/soc/sof/topology.c index e3e7fbe40fa6..73bf791e7520 100644 --- a/sound/soc/sof/topology.c +++ b/sound/soc/sof/topology.c @@ -290,6 +290,7 @@ static const struct sof_dai_types sof_dais[] = { {"SAI", SOF_DAI_IMX_SAI}, {"ESAI", SOF_DAI_IMX_ESAI}, {"ACP", SOF_DAI_AMD_BT}, + {"ACPBT", SOF_DAI_AMD_BT}, {"ACPSP", SOF_DAI_AMD_SP}, {"ACPDMIC", SOF_DAI_AMD_DMIC}, {"ACPHS", SOF_DAI_AMD_HS},
Commit efb931cdc4b9 ("ASoC: SOF: topology: Add support for AMD ACP DAIs") registered "ACP" name for SOF_DAI_AMD_BT DAI type. However, some boards, i.e. Steam Deck OLED, seem to require "ACPBT" for the same type: [ 467.489680] snd_sof_amd_vangogh 0000:04:00.5: ipc tx error for 0x30030000 (msg/reply size: 16/0): -22 [ 467.492775] snd_sof_amd_vangogh 0000:04:00.5: sof_ipc3_route_setup: route ACPBT2.IN -> BUF5.0 failed [ 467.495839] snd_sof_amd_vangogh 0000:04:00.5: sof_ipc3_set_up_all_pipelines: route set up failed [ 467.499128] snd_sof_amd_vangogh 0000:04:00.5: error: tplg component load failed -22 [ 467.502210] snd_sof_amd_vangogh 0000:04:00.5: error: failed to load DSP topology -22 [ 467.505289] snd_sof_amd_vangogh 0000:04:00.5: ASoC: error at snd_soc_component_probe on 0000:04:00.5: -22 [ 467.508430] sof_mach nau8821-max: ASoC: failed to instantiate card -22 [ 467.511725] sof_mach nau8821-max: error -EINVAL: Failed to register card(sof-nau8821-max) [ 467.514861] sof_mach: probe of nau8821-max failed with error -22 Add "ACPBT" alias for "ACP" SOF DAI type. Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com> --- sound/soc/sof/topology.c | 1 + 1 file changed, 1 insertion(+)