| Message ID | 20181108083754.1746-1-peter.ujfalusi@ti.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | ASoC/ARM: Merge the davinci and omap audio directories | expand |
* Peter Ujfalusi <peter.ujfalusi@ti.com> [181108 08:36]: > I believe with v1 all of this can be merged via ASoC tree as the chances of > conflicts are low for the touched arch/arm files. But Sekhar and Tony can > correct me if I'm wrong. Good to see this :) Yes merging these changes via the ASoC tree makes sense to me. I will take a look and ack when you post the version without RFC. Regards, Tony
On 11/8/18 10:37 AM, Peter Ujfalusi wrote: > The series also includes patches to updated the arch/arm/ files (defconfigs, > source files using audio Kconfig symbols). > > I believe with v1 all of this can be merged via ASoC tree as the chances of > conflicts are low for the touched arch/arm files. But Sekhar and Tony can > correct me if I'm wrong. > > I think I have catched all (I hope) issues regarding to Kconfig thanks to the > various autobuild machines out there kindly testing my linux-next-wip branch. > > I would appreciate any testing comments before I send the v1! > I applied on top of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git commit abd97b7c53ac ("Merge branch 'asoc-4.21' into asoc-next"). Patches 2 and 6 didn't apply and 7-8 I skipped. For my functional testing none of those were required. For the sound/soc/{omap => ti} conversion you may add: Tested-by: Jarkko Nikula <jarkko.nikula@bitmer.com> Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
On 2018-11-11 18:37, Jarkko Nikula wrote: > On 11/8/18 10:37 AM, Peter Ujfalusi wrote: >> The series also includes patches to updated the arch/arm/ files (defconfigs, >> source files using audio Kconfig symbols). >> >> I believe with v1 all of this can be merged via ASoC tree as the chances of >> conflicts are low for the touched arch/arm files. But Sekhar and Tony can >> correct me if I'm wrong. >> >> I think I have catched all (I hope) issues regarding to Kconfig thanks to the >> various autobuild machines out there kindly testing my linux-next-wip branch. >> >> I would appreciate any testing comments before I send the v1! >> > I applied on top of > git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git commit > abd97b7c53ac ("Merge branch 'asoc-4.21' into asoc-next"). > > Patches 2 and 6 didn't apply and 7-8 I skipped. For my functional > testing none of those were required. Hrm, the series was based on next-20181019, but I will keep on eye on these! > For the sound/soc/{omap => ti} conversion you may add: > > Tested-by: Jarkko Nikula <jarkko.nikula@bitmer.com> > Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com> Thank you! - Péter Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki
On Thu, Nov 08, 2018 at 10:37:45AM +0200, Peter Ujfalusi wrote: > Hi, > > This RFC series is on top of the omap-mcbsp cleanup series: > http://mailman.alsa-project.org/pipermail/alsa-devel/2018-November/141604.html This is all fine by me.
On 08/11/18 8:46 PM, Tony Lindgren wrote: > * Peter Ujfalusi <peter.ujfalusi@ti.com> [181108 08:36]: >> I believe with v1 all of this can be merged via ASoC tree as the chances of >> conflicts are low for the touched arch/arm files. But Sekhar and Tony can >> correct me if I'm wrong. > > Good to see this :) Yes merging these changes via the ASoC tree makes > sense to me. I will take a look and ack when you post the version > without RFC. Looks good to me too. And fine with the series getting merged through ASoC. But will appreciate an immutable commit over v4.20-rc1 in case I do end up queuing something that clashes with changes here (especially worried about the defconfig changes - those are notorious for merge conflicts, and are pain to resolve). I will test this once the non-RFC version is posted. Thanks, Sekhar
Hi, This RFC series is on top of the omap-mcbsp cleanup series: http://mailman.alsa-project.org/pipermail/alsa-devel/2018-November/141604.html For Texas Instruments audio support we have two directories at the moment: davinci omap In reality TI have the following SoC lines supported by Linux: daVinci OMAP Sitara Keystone We already needed to have cross directory 'links' for DRA7x/AM5 since they use McASP (davinci directory) and sDMA (omap directory). Keystone II line of SoCs are using drivers from davinci directory (McASP and EDMA). Which is kind of fine as they are in the same place. With the new AM654 (http://www.ti.com/lit/pdf/spruid7) we will have completely new DMA and there is no place to put that as it does not belong under davinci or omap, but it makes no sense to create yet another TI related directory just for the small udma-pcm. Another issue with the current support is how the Kconfigs are written. The DAI drivers are depending on the ASoC platform drivers and it is a bit of a mess at the moment. For simple-sound-card one need to know what DMA is used and then select the DAI driver (if the Kconfig is right). I have also cleaned up the Kconfig prompts and comments a bit to be more precise. As can be seen in the diffstat I tried to do the merger with minimal code change, resisting the urge to rename the davinci-i2c driver to davinci-asp. With the merger I have created a new Kconfig structure: - Submenu for TI related settings for clarity - Only the DAI drivers are selectable and the platform drivers are selected based on the available (enabled) DMA engines - The remaining ASoC machine drivers are selecting the DAIs they need - The davinci-evm is now a single entry, combining all the various daVinci EVMS it supports and selects the needed DAIs based on the enabled boards The series also includes patches to updated the arch/arm/ files (defconfigs, source files using audio Kconfig symbols). I believe with v1 all of this can be merged via ASoC tree as the chances of conflicts are low for the touched arch/arm files. But Sekhar and Tony can correct me if I'm wrong. I think I have catched all (I hope) issues regarding to Kconfig thanks to the various autobuild machines out there kindly testing my linux-next-wip branch. I would appreciate any testing comments before I send the v1! Regards, Peter --- Peter Ujfalusi (9): ASoC: ti: Merge davinci and omap directories MAINTAINERS: Add entry for sound/soc/ti and update the OMAP audio support ARM: OMAP1: Makefile: Update for new MCBSP Kconfig option ARM: OMAP2: Update for new MCBSP Kconfig option ARM: davinci: dm365-evm: Update for the new ASoC Kcofnig options ARM: omap2plus_defconfig: Update the audio options ARM: omap1_defconfig: Do not select ASoC by default ARM: davinci_all_defconfig: Update the audio options ASoC: ti: Kconfig: Remove the deprecated options MAINTAINERS | 11 +- arch/arm/configs/davinci_all_defconfig | 5 +- arch/arm/configs/omap1_defconfig | 2 - arch/arm/configs/omap2plus_defconfig | 14 +- arch/arm/mach-davinci/board-dm365-evm.c | 4 +- arch/arm/mach-omap1/Makefile | 2 +- arch/arm/mach-omap2/Makefile | 2 +- arch/arm/mach-omap2/pdata-quirks.c | 4 +- sound/soc/Kconfig | 3 +- sound/soc/Makefile | 3 +- sound/soc/davinci/Kconfig | 106 --------- sound/soc/davinci/Makefile | 16 -- sound/soc/omap/Kconfig | 120 ---------- sound/soc/omap/Makefile | 30 --- sound/soc/ti/Kconfig | 209 ++++++++++++++++++ sound/soc/ti/Makefile | 44 ++++ sound/soc/{omap => ti}/ams-delta.c | 0 sound/soc/{davinci => ti}/davinci-evm.c | 4 +- sound/soc/{davinci => ti}/davinci-i2s.c | 0 sound/soc/{davinci => ti}/davinci-i2s.h | 0 sound/soc/{davinci => ti}/davinci-mcasp.c | 14 +- sound/soc/{davinci => ti}/davinci-mcasp.h | 0 sound/soc/{davinci => ti}/davinci-vcif.c | 0 sound/soc/{davinci => ti}/edma-pcm.c | 0 sound/soc/{davinci => ti}/edma-pcm.h | 4 +- sound/soc/{omap => ti}/n810.c | 0 sound/soc/{omap => ti}/omap-abe-twl6040.c | 0 sound/soc/{omap => ti}/omap-dmic.c | 0 sound/soc/{omap => ti}/omap-dmic.h | 0 .../omap-hdmi-audio.c => ti/omap-hdmi.c} | 0 sound/soc/{omap => ti}/omap-mcbsp-priv.h | 0 sound/soc/{omap => ti}/omap-mcbsp-st.c | 0 sound/soc/{omap => ti}/omap-mcbsp.c | 0 sound/soc/{omap => ti}/omap-mcbsp.h | 0 sound/soc/{omap => ti}/omap-mcpdm.c | 0 sound/soc/{omap => ti}/omap-mcpdm.h | 0 sound/soc/{omap => ti}/omap-twl4030.c | 0 sound/soc/{omap => ti}/omap3pandora.c | 0 sound/soc/{omap => ti}/osk5912.c | 0 sound/soc/{omap => ti}/rx51.c | 0 sound/soc/{omap => ti}/sdma-pcm.c | 0 sound/soc/{omap => ti}/sdma-pcm.h | 4 +- 42 files changed, 294 insertions(+), 307 deletions(-) delete mode 100644 sound/soc/davinci/Kconfig delete mode 100644 sound/soc/davinci/Makefile delete mode 100644 sound/soc/omap/Kconfig delete mode 100644 sound/soc/omap/Makefile create mode 100644 sound/soc/ti/Kconfig create mode 100644 sound/soc/ti/Makefile rename sound/soc/{omap => ti}/ams-delta.c (100%) rename sound/soc/{davinci => ti}/davinci-evm.c (99%) rename sound/soc/{davinci => ti}/davinci-i2s.c (100%) rename sound/soc/{davinci => ti}/davinci-i2s.h (100%) rename sound/soc/{davinci => ti}/davinci-mcasp.c (99%) rename sound/soc/{davinci => ti}/davinci-mcasp.h (100%) rename sound/soc/{davinci => ti}/davinci-vcif.c (100%) rename sound/soc/{davinci => ti}/edma-pcm.c (100%) rename sound/soc/{davinci => ti}/edma-pcm.h (91%) rename sound/soc/{omap => ti}/n810.c (100%) rename sound/soc/{omap => ti}/omap-abe-twl6040.c (100%) rename sound/soc/{omap => ti}/omap-dmic.c (100%) rename sound/soc/{omap => ti}/omap-dmic.h (100%) rename sound/soc/{omap/omap-hdmi-audio.c => ti/omap-hdmi.c} (100%) rename sound/soc/{omap => ti}/omap-mcbsp-priv.h (100%) rename sound/soc/{omap => ti}/omap-mcbsp-st.c (100%) rename sound/soc/{omap => ti}/omap-mcbsp.c (100%) rename sound/soc/{omap => ti}/omap-mcbsp.h (100%) rename sound/soc/{omap => ti}/omap-mcpdm.c (100%) rename sound/soc/{omap => ti}/omap-mcpdm.h (100%) rename sound/soc/{omap => ti}/omap-twl4030.c (100%) rename sound/soc/{omap => ti}/omap3pandora.c (100%) rename sound/soc/{omap => ti}/osk5912.c (100%) rename sound/soc/{omap => ti}/rx51.c (100%) rename sound/soc/{omap => ti}/sdma-pcm.c (100%) rename sound/soc/{omap => ti}/sdma-pcm.h (85%)