Re: [PATCH v2 10/12] ASoC: SOF: Intel: hda: make sure DAI widget is set up before IPC

[Péter Ujfalusi]

Hi Daniel,

On 24/09/2021 10:42, Daniel Baluta wrote:
> On Thu, Sep 23, 2021 at 4:04 PM Péter Ujfalusi
> <peter.ujfalusi@xxxxxxxxxxxxxxx> wrote:
>>
>>
>>
>> On 23/09/2021 15:58, Pierre-Louis Bossart wrote:
>>>
>>>>> +static struct sof_ipc_dai_config *hda_dai_update_config(struct snd_soc_dapm_widget *w,
>>>>> +                                                   int channel)
>>>>>  {
>>>>> +   struct snd_sof_widget *swidget = w->dobj.private;
>>>>>     struct sof_ipc_dai_config *config;
>>>>>     struct snd_sof_dai *sof_dai;
>>>>> -   struct sof_ipc_reply reply;
>>>>> -   int ret = 0;
>>>>>
>>>>> -   list_for_each_entry(sof_dai, &hda_stream->sdev->dai_list, list) {
>>>>> -           if (!sof_dai->cpu_dai_name)
>>>>> -                   continue;
>>>>> +   if (!swidget) {
>>>>> +           dev_err(swidget->scomp->dev, "error: No private data for widget %s\n", w->name);
>>>>
>>>> NULL pointer dereference, just return NULL without the print. The caller
>>>> is printing anyways.
>>>
>>> yes good catch, we need a v3 with the fixes suggested by Peter in
>>> https://github.com/thesofproject/linux/pull/3171/ applied.
>>
>> Only the second patch in the PR is applicable for upstream, but it
>> should be squashed in for v3.
> 
> Thanks Peter, will squash this in and send v3.

As we discussed, I'll send the v3 with the fix.

Thanks for sending the initial versions upstream!

-- 
Péter