Thread (4 messages) flat view 4 messages, 2 authors, 2014-01-11

[PATCH v2 12/28] drm/i2c: tda998x: add DT support

From: Russell King - ARM Linux <hidden>
Date: 2014-01-11 18:03:21
Also in: dri-devel, linux-devicetree, lkml

On Thu, Jan 09, 2014 at 12:03:24PM +0100, Jean-Francois Moine wrote:
This patch adds DT support to the tda998x.

As a side effect, now, the audio sample rate is always 48kHz and the
audio clock is always set.

Signed-off-by: Jean-Francois Moine <redacted>
This patch breaks audio through the renaming of variable names.  When you
do such changes, *never* change all three together.  Do them one at a time,
replacing all instances of one variable before moving on to the next one.
This avoids getting the names muddled up.
-	struct tda998x_encoder_params params;
+
+	u8 audio_type;		/* audio type */
+	u8 audio_frame[6];
+	u32 audio_port;
"audio type" is a pointless comment for a variable called "audio_type".
Explain what it is.  It's the input format, which may be SPDIF or I2S.
 	/* Set audio input source */
-	switch (p->audio_format) {
+	switch (priv->audio_type) {
So, "audio_format" has become "audio_type" here.
quoted hunk ↗ jump to hunk
@@ -698,7 +703,13 @@ tda998x_encoder_set_config(struct drm_encoder *encoder, void *params)
 			    VIP_CNTRL_2_SWAP_F(p->swap_f) |
 			    (p->mirr_f ? VIP_CNTRL_2_MIRR_F : 0);
 
-	priv->params = *p;
+	memcpy(priv->audio_frame, p->audio_frame,
+			sizeof priv->audio_frame);
+
+	if (p->audio_cfg) {
+		priv->audio_port = p->audio_cfg;
+		priv->audio_type = p->audio_format;
+	}
Here, audio_port and audio_type are only written to if audio_cfg is
non-zero, which is not quite what is intended here.  If you want DT
to override this, then make that explicit.

Moreover, we can see that this confirms that "audio_format" becomes
"audio_type", and "audio_cfg" becomes "audio_port".
quoted hunk ↗ jump to hunk
@@ -947,8 +958,8 @@ tda998x_encoder_mode_set(struct drm_encoder *encoder,
 
 		tda998x_write_avi(priv, mode);
 
-		if (priv->params.audio_cfg)
-			tda998x_configure_audio(priv, mode, &priv->params);
+		if (priv->audio_type)
+			tda998x_configure_audio(priv, mode);
And here we have the real bug.  "audio_cfg" has become "audio_type",
which in the case of SPDIF, is zero.  Hence, with a SPDIF source,
tda998x_configure_audio() is no longer called.

-- 
FTTC broadband for 0.8mile line: 5.8Mbps down 500kbps up.  Estimation
in database were 13.1 to 19Mbit for a good line, about 7.5+ for a bad.
Estimate before purchase was "up to 13.2Mbit".
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help