On Wed, Oct 20, 2010 at 1:01 PM, Mark Brown <broonie@xxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote: > On Wed, Oct 20, 2010 at 12:27:43PM +0900, Jassi Brar wrote: >> On Wed, Oct 20, 2010 at 12:13 PM, Mark Brown > >> > If we have something in the driver data struct specifying the masks to >> > use then set these at probe time rather than having the if statements - >> > probably the same mask can be used for playback & record if the >> > bitfields are lined up similarly. ÂThis would then be: > >> > Â Â Â Âcon |= data->active_mask; >> > Â Â Â Âcon &= ~data->pause_mask; > >> > or similar, possibly with some shifts. > >> Let me see if the space saved is worth the complication. > > I was thinking it should make something similar by making it clear that > we're just doing the same thing to a different location rather than lots > of conditional code to follow. In this case it doesn't fit very neatly. So I am dropping the suggestion, sorry. >> > How would the index be used with multi-component? > >> For example, please look at [Patch 23/25] > >> + Â Â Â Â Â Â Â /* Secondary is at offset MAX_I2S from Primary */ >> + Â Â Â Â Â Â Â str = (char *)smdk_dai[SEC_PLAYBACK].cpu_dai_name; >> + Â Â Â Â Â Â Â str[strlen(str) - 1] = '0' + MAX_I2S; > > Hrm, slightly fiddly. ÂPerhaps we want a function people can call which > will generate the secondary DAI name for you? ÂMight be more directly > what people want. We already have to decide on platform_device's id in probe while creating the secondary dai in CPU driver. So there must exist any such predefined offset in the common driver. So, I decided to reuse the symbol and save one addition to the API. _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
