| Message ID | 20200218103824.26708-1-stephan@gerhold.net (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | a4877a6fb2bd2e356a5eaacd86d6b6d69ff84e69 |
| Headers | show |
| Series | ASoC: soc-pcm: fix regression in soc_new_pcm() | expand |
On Tue 18 Feb 2020 at 11:38, Stephan Gerhold <stephan@gerhold.net> wrote: > Commit af4bac11531f ("ASoC: soc-pcm: crash in snd_soc_dapm_new_dai") > swapped the SNDRV_PCM_STREAM_* parameter in the > snd_soc_dai_stream_valid(cpu_dai, ...) checks. But that works only > for codec2codec links. For normal links it breaks registration of > playback/capture-only PCM devices. > > E.g. on qcom/apq8016_sbc there is usually one playback-only and one > capture-only PCM device, but they disappeared after the commit. > > The codec2codec case was added in commit a342031cdd08 > ("ASoC: create pcm for codec2codec links as well") as an extra check > (e.g. `playback = playback && cpu_playback->channels_min`). That particular check was there when I worked on that change but I seems I messed up when I rebased on Kuninori's work regarding snd_soc_dai_stream_valid() which happened more or less at the same time. > > We should be able to simplify the code by checking directly for > the correct stream type in the loop. > This also fixes the regression because we check for PLAYBACK for > both codec and cpu dai again when codec2codec is not used. > > Cc: Sameer Pujar <spujar@nvidia.com> > Cc: Jerome Brunet <jbrunet@baylibre.com> > Fixes: af4bac11531f ("ASoC: soc-pcm: crash in snd_soc_dapm_new_dai") > Signed-off-by: Stephan Gerhold <stephan@gerhold.net> Looks good and works with the codec-to-codec links on Amlogic aiu. Thx ! Reviewed-by: Jerome Brunet <jbrunet@baylibre.com> Tested-by: Jerome Brunet <jbrunet@baylibre.com>
diff --git a/sound/soc/soc-pcm.c b/sound/soc/soc-pcm.c index ff1b7c7078e5..cfa24d214a9c 100644 --- a/sound/soc/soc-pcm.c +++ b/sound/soc/soc-pcm.c @@ -2888,22 +2888,19 @@ int soc_new_pcm(struct snd_soc_pcm_runtime *rtd, int num) capture = rtd->dai_link->dpcm_capture; } else { /* Adapt stream for codec2codec links */ - struct snd_soc_pcm_stream *cpu_capture = rtd->dai_link->params ? - &cpu_dai->driver->playback : &cpu_dai->driver->capture; - struct snd_soc_pcm_stream *cpu_playback = rtd->dai_link->params ? - &cpu_dai->driver->capture : &cpu_dai->driver->playback; + int cpu_capture = rtd->dai_link->params ? + SNDRV_PCM_STREAM_PLAYBACK : SNDRV_PCM_STREAM_CAPTURE; + int cpu_playback = rtd->dai_link->params ? + SNDRV_PCM_STREAM_CAPTURE : SNDRV_PCM_STREAM_PLAYBACK; for_each_rtd_codec_dai(rtd, i, codec_dai) { if (snd_soc_dai_stream_valid(codec_dai, SNDRV_PCM_STREAM_PLAYBACK) && - snd_soc_dai_stream_valid(cpu_dai, SNDRV_PCM_STREAM_CAPTURE)) + snd_soc_dai_stream_valid(cpu_dai, cpu_playback)) playback = 1; if (snd_soc_dai_stream_valid(codec_dai, SNDRV_PCM_STREAM_CAPTURE) && - snd_soc_dai_stream_valid(cpu_dai, SNDRV_PCM_STREAM_PLAYBACK)) + snd_soc_dai_stream_valid(cpu_dai, cpu_capture)) capture = 1; } - - capture = capture && cpu_capture->channels_min; - playback = playback && cpu_playback->channels_min; } if (rtd->dai_link->playback_only) {
Commit af4bac11531f ("ASoC: soc-pcm: crash in snd_soc_dapm_new_dai") swapped the SNDRV_PCM_STREAM_* parameter in the snd_soc_dai_stream_valid(cpu_dai, ...) checks. But that works only for codec2codec links. For normal links it breaks registration of playback/capture-only PCM devices. E.g. on qcom/apq8016_sbc there is usually one playback-only and one capture-only PCM device, but they disappeared after the commit. The codec2codec case was added in commit a342031cdd08 ("ASoC: create pcm for codec2codec links as well") as an extra check (e.g. `playback = playback && cpu_playback->channels_min`). We should be able to simplify the code by checking directly for the correct stream type in the loop. This also fixes the regression because we check for PLAYBACK for both codec and cpu dai again when codec2codec is not used. Cc: Sameer Pujar <spujar@nvidia.com> Cc: Jerome Brunet <jbrunet@baylibre.com> Fixes: af4bac11531f ("ASoC: soc-pcm: crash in snd_soc_dapm_new_dai") Signed-off-by: Stephan Gerhold <stephan@gerhold.net> --- Original report of the regression: https://lore.kernel.org/alsa-devel/20200217144120.GA243254@gerhold.net/ I'm still quite confused about the crash that was fixed by the commit that introduced the regression... If anything, the original code added in commit a342031cdd08 ("ASoC: create pcm for codec2codec links as well") should have been too "restrictive" since it checked for both PLAYBACK and CAPTURE for the cpu dai in the codec2codec case... Audio works fine gain on apq8016_sbc with this patch, but I don't have any device with codec2codec to test this. --- sound/soc/soc-pcm.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-)