On Wed, Jul 23, 2008 at 07:36:35PM -0400, Jon Smirl wrote:
Clean up ASOC functions to support 32b values instead of 16b
As it stands there's also a mix of unsigned int and u32 in here which you've already said you'll address.
Could you please provide a more detailed changelog when you resubmit this? Explaining the mechanics of the change you've made would make it a bit easier to review - obviously, much of it is replacing unsigned short in the register parameters but there's other changes in there like switching from signed to unsigned types. It may be that it could usefully be split into multiple patches, each doing one mechanical change but possibly not.
<petty>Also, it's spelt "ASoC".</petty>
@@ -221,7 +221,7 @@ static int dapm_update_bits(struct snd_soc_dapm_widget *widget)
power = widget->power; if (widget->invert)
power = (power ? 0:1);
power = (power ? 0 : 1);old = snd_soc_read(codec, widget->reg); new = (old & ~(0x1 << widget->shift)) | (power << widget->shift);
This change is reasonable enough but unrelated to the rest of the patch...
@@ -755,10 +755,10 @@ static int dapm_mixer_update_power(struct snd_soc_dapm_widget *widget, found = 1; if (val) /* new connection */
path->connect = invert ? 0:1;
path->connect = invert ? 0 : 1;
path->connect = invert ? 1:0;
path->connect = invert ? 1 : 0;
Ditto.