On 15/08/2025 17:56, Srinivas Kandagatla wrote:
> Thanks Krzysztof,
> On 8/15/25 12:39 PM, Krzysztof Kozlowski wrote:
>> If earlier opening of source graph fails (e.g. ADSP rejects due to
>
> I think you are referring to the err patch in prepare.
True I am working on feature relying on that other patch, but the code
here is not really relevant to that other patch, I think.
>
>> incorrect audioreach topology), the graph is closed and
>> "dai_data->graph[dai->id]" is assigned NULL. Preparing the DAI for sink
>> graph continues though and next call to q6apm_lpass_dai_prepare()
>> receives dai_data->graph[dai->id]=NULL leading to NULL pointer
>> exception:
>>
>> qcom-apm gprsvc:service:2:1: Error (1) Processing 0x01001002 cmd
>> qcom-apm gprsvc:service:2:1: DSP returned error[1001002] 1
>> q6apm-lpass-dais 30000000.remoteproc:glink-edge:gpr:service@1:bedais: fail to start APM port 78
>> q6apm-lpass-dais 30000000.remoteproc:glink-edge:gpr:service@1:bedais: ASoC: error at snd_soc_pcm_dai_prepare on TX_CODEC_DMA_TX_3: -22
>> Unable to handle kernel NULL pointer dereference at virtual address 00000000000000a8
>> ...
>> Call trace:
>> q6apm_graph_media_format_pcm+0x48/0x120 (P)
>> q6apm_lpass_dai_prepare+0x110/0x1b4
>> snd_soc_pcm_dai_prepare+0x74/0x108
>> __soc_pcm_prepare+0x44/0x160
>> dpcm_be_dai_prepare+0x124/0x1c0
>>
>> Fixes: 30ad723b93ad ("ASoC: qdsp6: audioreach: add q6apm lpass dai support")
>> Cc: <stable@xxxxxxxxxxxxxxx>
>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>
>> ---
>> sound/soc/qcom/qdsp6/q6apm-lpass-dais.c | 6 ++++++
>> 1 file changed, 6 insertions(+)
>>
>> diff --git a/sound/soc/qcom/qdsp6/q6apm-lpass-dais.c b/sound/soc/qcom/qdsp6/q6apm-lpass-dais.c
>> index f90628d9b90e..7520e6f024c3 100644
>> --- a/sound/soc/qcom/qdsp6/q6apm-lpass-dais.c
>> +++ b/sound/soc/qcom/qdsp6/q6apm-lpass-dais.c
>> @@ -191,6 +191,12 @@ static int q6apm_lpass_dai_prepare(struct snd_pcm_substream *substream, struct s
>> return rc;
>> }
>> dai_data->graph[graph_id] = graph;
>> + } else if (!dai_data->graph[dai->id]) {
>> + /*
>> + * Loading source graph failed before, so abort loading the sink
>> + * as well.
>> + */
>> + return -EINVAL;
>> }
> I guess this is the capture graph that is triggering the error, normally
> we do not open/close the capture graph in prepare, we do
> stop/prepare/start for capture graphs and handle open close in
> startup/shutdown.
>
> Can you try this change and see if it fixes the issue, as prepare could
> be called multiple times and your patch will not give chance for trying
> new parameters incase the failure was due to unsupported params.
Yes, this works.
Best regards,
Krzysztof
