Message ID | 87sgo2ilso.wl-kuninori.morimoto.gx@renesas.com (mailing list archive) |
---|---|
Headers | show |
Series | ASoC: soc-core cleanup - step 4 | expand |
On 10/8/19 11:29 PM, Kuninori Morimoto wrote: > > Hi Mark > Cc Pierre-Louis > > These are step4 of soc-core cleanup. > These related to soc-topology. > If my understanding and my code are correct, > there is no effect to topology side, but I can't test. > > I'm not sure who can test these, but I guess > Pierre-Louis knows ? > Pierre-Louis, can you help it ? we can use GitHub and create a pull request with these patches, that will be tested on a variety of platforms. We unfortunately have a random IPC error at the moment which sometimes makes it difficult to judge if errors reported are due to the patch being tested or just bad timing. I tried to apply the series but I get errors on the 2nd patch. I first need to rebase on Mark's tree. > > Kuninori Morimoto (21): > ASoC: soc-core: remove for_each_rtdcom_safe() > ASoC: soc-core: add for_each_rtd_components() and replace > ASoC: soc-core: move soc_init_dai_link() > ASoC: soc-core: rename soc_init_dai_link() to soc_dai_link_sanity_check() > ASoC: soc-core: remove duplicated soc_is_dai_link_bound() > ASoC: soc-core: call soc_bind_dai_link() under snd_soc_add_dai_link() > ASoC: soc-core: add soc_unbind_dai_link() > ASoC: soc-core: snd_soc_unbind_card() cleanup > ASoC: soc-core: remove unneeded snd_soc_tplg_component_remove() > ASoC: soc-core: move snd_soc_lookup_component() > ASoC: soc-core: add snd_soc_del_component() > ASoC: soc-core: use snd_soc_lookup_component() at snd_soc_unregister_component() > ASoC: soc-core: move snd_soc_register_dai() > ASoC: soc-core: have legacy_dai_naming at snd_soc_register_dai() > ASoC: soc-core: move snd_soc_unregister_dais() > ASoC: soc-core: add snd_soc_unregister_dai() > ASoC: soc-core: don't call snd_soc_dapm_new_dai_widgets() at snd_soc_register_dai() > ASoC: soc-core: use mutex_lock() at snd_soc_add_component() > ASoC: soc-core: call snd_soc_register_dai() from snd_soc_register_dais() > ASoC: soc-core: remove topology specific operation > ASoC: soc.h: dobj is used only when SND_SOC_TOPOLOGY > > include/sound/soc.h | 25 ++- > sound/soc/soc-component.c | 43 +--- > sound/soc/soc-compress.c | 52 ++--- > sound/soc/soc-core.c | 542 ++++++++++++++++++++-------------------------- > sound/soc/soc-pcm.c | 49 ++--- > sound/soc/soc-topology.c | 32 ++- > 6 files changed, 320 insertions(+), 423 deletions(-) >
On 10/8/19 11:29 PM, Kuninori Morimoto wrote: > > Hi Mark > Cc Pierre-Louis > > These are step4 of soc-core cleanup. > These related to soc-topology. > If my understanding and my code are correct, > there is no effect to topology side, but I can't test. > > I'm not sure who can test these, but I guess > Pierre-Louis knows ? > Pierre-Louis, can you help it ? This patchset was tested with the Intel SOF CI, and no regressions were seen. Full results here: https://github.com/thesofproject/linux/pull/1298 I added a couple of comments to help improve commit messages or code. Nothing major, more about fixing inconsistencies than about broken code. I could however not figure out if Patch 6 was correct or not, this is by far the most complicated in this series and additional folks more familiar with the code should look into it. > > Kuninori Morimoto (21): > ASoC: soc-core: remove for_each_rtdcom_safe() > ASoC: soc-core: add for_each_rtd_components() and replace > ASoC: soc-core: move soc_init_dai_link() > ASoC: soc-core: rename soc_init_dai_link() to soc_dai_link_sanity_check() > ASoC: soc-core: remove duplicated soc_is_dai_link_bound() > ASoC: soc-core: call soc_bind_dai_link() under snd_soc_add_dai_link() > ASoC: soc-core: add soc_unbind_dai_link() > ASoC: soc-core: snd_soc_unbind_card() cleanup > ASoC: soc-core: remove unneeded snd_soc_tplg_component_remove() > ASoC: soc-core: move snd_soc_lookup_component() > ASoC: soc-core: add snd_soc_del_component() > ASoC: soc-core: use snd_soc_lookup_component() at snd_soc_unregister_component() > ASoC: soc-core: move snd_soc_register_dai() > ASoC: soc-core: have legacy_dai_naming at snd_soc_register_dai() > ASoC: soc-core: move snd_soc_unregister_dais() > ASoC: soc-core: add snd_soc_unregister_dai() > ASoC: soc-core: don't call snd_soc_dapm_new_dai_widgets() at snd_soc_register_dai() > ASoC: soc-core: use mutex_lock() at snd_soc_add_component() > ASoC: soc-core: call snd_soc_register_dai() from snd_soc_register_dais() > ASoC: soc-core: remove topology specific operation > ASoC: soc.h: dobj is used only when SND_SOC_TOPOLOGY > > include/sound/soc.h | 25 ++- > sound/soc/soc-component.c | 43 +--- > sound/soc/soc-compress.c | 52 ++--- > sound/soc/soc-core.c | 542 ++++++++++++++++++++-------------------------- > sound/soc/soc-pcm.c | 49 ++--- > sound/soc/soc-topology.c | 32 ++- > 6 files changed, 320 insertions(+), 423 deletions(-) >
Hi Pierre-Louis Thank you for your review. > I added a couple of comments to help improve commit messages or > code. Nothing major, more about fixing inconsistencies than about > broken code. Thanks !! I will fixup these and post v2 > I could however not figure out if Patch 6 was correct or not, this is > by far the most complicated in this series and additional folks more > familiar with the code should look into it. It is the main patch of this patchset. If it was not correct, topology can't probe, I guess. But yes, more review is very welcome. Thank you for your help !! Best regards --- Kuninori Morimoto