[Intel-gfx] [RFC PATCH 5/5] ALSA: x86: hdmi: hack to enable DP audio on CHT

[Pierre-Louis Bossart]

Somehow the configuration register that the audio driver needs to use
depends on the port/pipe combination.
Adding the pipe information to the notify() call isn't helping, the
pipe value is -1 (illegal).

FIXME: does this require a change in the pdata or a handshake with i915?
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart at linux.intel.com>
---
 sound/x86/intel_hdmi_lpe_audio.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/sound/x86/intel_hdmi_lpe_audio.c b/sound/x86/intel_hdmi_lpe_audio.c
index 6ef0ff8..23e5b34 100644
--- a/sound/x86/intel_hdmi_lpe_audio.c
+++ b/sound/x86/intel_hdmi_lpe_audio.c
@@ -563,7 +563,13 @@ static int hdmi_lpe_audio_probe(struct platform_device *pdev)
 	if (pci_dev_present(cherryview_ids)) {
 		dev_dbg(&hlpe_pdev->dev, "%s: Cherrytrail LPE - Detected\n",
 				__func__);
-		ctx->had_config_offset = AUDIO_HDMI_CONFIG_C;
+		//ctx->had_config_offset = AUDIO_HDMI_CONFIG_C;
+		/* FIXME: hard-coding to CONFIG_A enables DP audio on CHT,
+		 *  how do I find out which config to use ?
+		 * the pipe is -1 (invalid) when the notify function is called,
+		 * so not sure how to go about this
+		 */
+		ctx->had_config_offset = AUDIO_HDMI_CONFIG_A;
 	} else {
 		dev_dbg(&hlpe_pdev->dev, "%s: Baytrail LPE - Assume\n",
 				__func__);
-- 
2.7.4