Skip to contentSkip to conversation

ASoC: WM8983: Initial driver

PATCH Mark Brown 2 participants 3 messages Jul 15, 2011 1 message not in the archive

lore
Share & tools
lore

Mark Brown

author
#1 lore raw
In reply to a message not in the archive
Quoted 4 lines · Dimitris Papastamos

The WM8983 is a low power, high quality stereo CODEC designed for portable multimedia applications. Highly flexible analogue mixing functions enable new application features, combining hi-fi quality audio with voice communication.

I've applied this but there's one issue you should fix up later:

Review

Quoted 4 lines
(quoted diff) +4 -0
+	/* mute all outputs and set PGAs to minimum gain */
+	for (i = WM8983_LOUT1_HP_VOLUME_CTRL;
+	     i <= WM8983_OUT4_MONO_MIX_CTRL; ++i)
+		snd_soc_update_bits(codec, i, 0x40, 0x40);

I know you're doing this to minimise artifacts on initial bias raise but we should really be doing this without affecting the default register settings.

#2 lore raw
Quoted 8 lines · Mark Brown
Quoted 4 lines · Dimitris Papastamos

The WM8983 is a low power, high quality stereo CODEC designed for portable multimedia applications. Highly flexible analogue mixing functions enable new application features, combining hi-fi quality audio with voice communication.

I've applied this but there's one issue you should fix up later:

Quoted 8 lines

Review

Quoted 4 lines
(quoted diff) +4 -0
+	/* mute all outputs and set PGAs to minimum gain */
+	for (i = WM8983_LOUT1_HP_VOLUME_CTRL;
+	     i <= WM8983_OUT4_MONO_MIX_CTRL; ++i)
+		snd_soc_update_bits(codec, i, 0x40, 0x40);

I know you're doing this to minimise artifacts on initial bias raise but we should really be doing this without affecting the default register settings.

Well, what about using the cache_bypass option to write directly to the HW?

Thanks,
Dimitris

author
#3 lore raw
Quoted 8 lines · Dimitris Papastamos
Quoted 3 lines · Mark Brown

I know you're doing this to minimise artifacts on initial bias raise but we should really be doing this without affecting the default register settings.

Well, what about using the cache_bypass option to write directly to the HW?

You still have the same issue at resume time at least.