[alsa-devel] [PATCH v2 09/25] ASoC: soc-core: tidyup for snd_soc_dapm_add_routes()

[Cezary Rojewski]

On 2019-08-20 16:05, Pierre-Louis Bossart wrote:
> 
> 
> On 8/20/19 8:38 AM, Cezary Rojewski wrote:
>> On 2019-08-20 14:36, Pierre-Louis Bossart wrote:
>>>
>>>
>>> On 8/20/19 6:18 AM, Cezary Rojewski wrote:
>>>> On 2019-08-07 03:31, Kuninori Morimoto wrote:
>>>>>
>>>>> From: Kuninori Morimoto <kuninori.morimoto.gx at renesas.com>
>>>>>
>>>>> snd_soc_dapm_add_routes() registers routes by using
>>>>> for(... i < num; ...). If routes was NULL, num should be zero.
>>>>> Thus, we don't need to check about route pointer.
>>>>> This patch also cares missing return value.
>>>>>
>>>>> Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx at renesas.com>
>>>>> ---
>>>>> v1 -> v2
>>>>>
>>>>>     - check return value
>>>>>     - change Subject
>>>>>
>>>>>   sound/soc/soc-core.c | 23 +++++++++++++----------
>>>>>   1 file changed, 13 insertions(+), 10 deletions(-)
>>>>>
>>>>> diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
>>>>> index 21cdd3c..ca1b04c 100644
>>>>> --- a/sound/soc/soc-core.c
>>>>> +++ b/sound/soc/soc-core.c
>>>>> @@ -1310,10 +1310,11 @@ static int soc_probe_component(struct 
>>>>> snd_soc_card *card,
>>>>>       if (ret < 0)
>>>>>           goto err_probe;
>>>>> -    if (component->driver->dapm_routes)
>>>>> -        snd_soc_dapm_add_routes(dapm,
>>>>> -                    component->driver->dapm_routes,
>>>>> -                    component->driver->num_dapm_routes);
>>>>> +    ret = snd_soc_dapm_add_routes(dapm,
>>>>> +                      component->driver->dapm_routes,
>>>>> +                      component->driver->num_dapm_routes);
>>>>> +    if (ret < 0)
>>>>> +        goto err_probe;
>>>>>       list_add(&dapm->list, &card->dapm_list);
>>>>>       /* see for_each_card_components */
>>>>> @@ -2060,13 +2061,15 @@ static int snd_soc_instantiate_card(struct 
>>>>> snd_soc_card *card)
>>>>>           snd_soc_add_card_controls(card, card->controls,
>>>>>                         card->num_controls);
>>>>> -    if (card->dapm_routes)
>>>>> -        snd_soc_dapm_add_routes(&card->dapm, card->dapm_routes,
>>>>> -                    card->num_dapm_routes);
>>>>> +    ret = snd_soc_dapm_add_routes(&card->dapm, card->dapm_routes,
>>>>> +                      card->num_dapm_routes);
>>>>> +    if (ret < 0)
>>>>> +        goto probe_end;
>>>>> -    if (card->of_dapm_routes)
>>>>> -        snd_soc_dapm_add_routes(&card->dapm, card->of_dapm_routes,
>>>>> -                    card->num_of_dapm_routes);
>>>>> +    ret = snd_soc_dapm_add_routes(&card->dapm, card->of_dapm_routes,
>>>>> +                      card->num_of_dapm_routes);
>>>>> +    if (ret < 0)
>>>>> +        goto probe_end;
>>>>>       /* try to set some sane longname if DMI is available */
>>>>>       snd_soc_set_dmi_name(card, NULL);
>>>>>
>>>>
>>>> Hello there,
>>>>
>>>> I've run a validation cycle on recent broonie/for-next and this 
>>>> commit caused regression. However, it may be simply an error on 
>>>> board side instead.
>>>>
>>>> Previously, ret from snd_soc_dapm_add_routes has been ignored thus 
>>>> it was permissive for addition of several routes to fail. As long as 
>>>> some routes succeeded, card was working just fine. Now it's no 
>>>> longer the case - behavior of the card initialization has changed: 
>>>> it is required for ALL routes to succeed before card can be fully 
>>>> instantiated.
>>>>
>>>> Must say collapsing snd_soc_instantiate_card is a wonderful way to 
>>>> test your card's removal flow (soc__cleanup_card_resources and 
>>>> friends)..
>>>>
>>>> Question is simple: are we staying with all-for-one/ one-for-all 
>>>> approach or we reverting to permissive behavior?
>>>
>>> Can you elaborate in which test case this patch creates a problem? 
>>> Just curious why the route addition fails in the first place.
>>
>> If snd_soc_instantiate_card fails so does any test, really. Red wall 
>> was easy to spot even for a hungry developer : )
>>
>> Our cnl_rt274 board declares several routes, yet our topology does not 
>> provide necessary info for all of them. And thus, addition of some 
>> routes fails. This was fine till now. That's also why I'd mentioned in 
>> the very first sentence: it might be simply a board issue. Maybe we 
>> should have never abused permissive behavior in the first place.
> 
> Yep, and that driver is not upstream as well so Intel can't complain 
> here...

??

It's not about complaining, rather starting a discussion. If I were 
using boards with topology not fully matching its board equivalent 
(because if has never been required of me) then there may be others who 
did the exact same trick. Your card won't be enumerated now == change of 
behavior.

Board bxt_rt298 is upstreamed and the exact same failures could be 
reproduced since topology has something to say here too..