On Fri, Mar 19, 2010 at 08:08:53AM +0100, Daniel Mack wrote:
On Fri, Mar 19, 2010 at 08:56:40AM +0200, Peter Ujfalusi wrote:
b/sound/soc/davinci/davinci-i2s.c index 6362ca0..4aad7ec 100644 --- a/sound/soc/davinci/davinci-i2s.c +++ b/sound/soc/davinci/davinci-i2s.c @@ -585,7 +585,8 @@ static int davinci_i2s_probe(struct platform_device *pdev) dev->dma_params[SNDRV_PCM_STREAM_CAPTURE].channel = res->start;
davinci_i2s_dai.private_data = dev;
davinci_i2s_dai.dma_data = dev->dma_params;
davinci_i2s_dai.capture.dma_data = dev->dma_params;davinci_i2s_dai.playback.dma_data = dev->dma_params;
Would not be more consistent around the places to actually use the snd_soc_dai_set_dma_data(...); function instead of direct assignment here, and there randomly?
Sure. Now that this is possible, the davinvi code should be cleaned up. But as I say - I couldn't even compile-test these changes, so I didn't want to break the code flow logic as well. Davinci is special though, other drivers should be more consistent.
I agree with both of you, we can always tidy the code up later but cross platform changes need some sync. I've applied the patch, thanks once again for analysing this and to everyone else for their review.