Re: [PATCH v4 04/15] drm: bridge/dw_hdmi: add identification registers parse and record

[yakir <ykk@rock-chips.com>]

[-- Attachment #1.1: Type: text/plain, Size: 3400 bytes --]

Hi philipp,

On 2015年03月12日 18:29, Philipp Zabel wrote:
> Am Samstag, den 28.02.2015, 21:35 -0500 schrieb Yakir Yang:
>> By parsing the identification registers we can know what functions
>> are present on the hdmi ip.
>>
>> Signed-off-by: Yakir Yang <ykk@rock-chips.com>
>> ---
>> Changes in v4:
>> -Correct phy_type assignment bug
>>
>> Changes in v3:
>> - Add ID registers parse and record
>>
>> Changes in v2: None
>>
>>   drivers/gpu/drm/bridge/dw_hdmi.c | 59 ++++++++++++++++++++++++++++++++++++++++
>>   drivers/gpu/drm/bridge/dw_hdmi.h | 23 ++++++++++++++++
>>   2 files changed, 82 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
>> index 08f10da..937beed 100644
>> --- a/drivers/gpu/drm/bridge/dw_hdmi.c
>> +++ b/drivers/gpu/drm/bridge/dw_hdmi.c
>> @@ -79,6 +79,23 @@ static const u16 csc_coeff_rgb_in_eitu709[3][4] = {
>>   	{ 0x6756, 0x78ab, 0x2000, 0x0200 }
>>   };
>>   
>> +struct hdmi_id {
>> +	u8 design;
>> +	u8 revision;
>> +
>> +	bool prepen;
>> +	bool audspdif;
>> +	bool audi2s;
>> +	bool hdmi14;
>> +	bool csc;
>> +	bool hdcp;
>> +	bool hdmi20;
>> +	bool confapb;
>> +	bool ahbauddma;
>> +	bool gpaud;
>> +	u8 phy_type;
>> +};
>>
>>   struct hdmi_vmode {
>>   	bool mdvi;
>>   	bool mhsyncpolarity;
>> @@ -111,6 +128,8 @@ struct dw_hdmi {
>>   	struct clk *isfr_clk;
>>   	struct clk *iahb_clk;
>>   
>> +	struct hdmi_id id;
>> +
>>   	struct hdmi_data_info hdmi_data;
>>   	const struct dw_hdmi_plat_data *plat_data;
>>   
>> @@ -1259,6 +1278,36 @@ static int dw_hdmi_setup(struct dw_hdmi *hdmi, struct drm_display_mode *mode)
>>   	return 0;
>>   }
>>   
>> +static void hdmi_parse_id(struct dw_hdmi *hdmi)
>> +{
>> +	u8 config0_id, config1_id, config2_id, config3_id;
>> +
>> +	config0_id = hdmi_readb(hdmi, HDMI_CONFIG0_ID);
>> +	config1_id = hdmi_readb(hdmi, HDMI_CONFIG1_ID);
>> +	config2_id = hdmi_readb(hdmi, HDMI_CONFIG2_ID);
>> +	config3_id = hdmi_readb(hdmi, HDMI_CONFIG3_ID);
>> +
>> +	hdmi->id.prepen = config0_id & HDMI_CONFIG0_ID_PREPEN ? true : false;
>> +	hdmi->id.audi2s = config0_id & HDMI_CONFIG0_ID_AUDI2S ? true : false;
>> +	hdmi->id.hdmi14 = config0_id & HDMI_CONFIG0_ID_HDMI14 ? true : false;
>> +	hdmi->id.hdcp = config0_id & HDMI_CONFIG0_ID_HDCP ? true : false;
>> +	hdmi->id.csc = config0_id & HDMI_CONFIG0_ID_CSC ? true : false;
>> +	hdmi->id.audspdif = config0_id & HDMI_CONFIG0_ID_AUDSPDIF ?
>> +			    true : false;
>> +
>> +	hdmi->id.confapb = config1_id & HDMI_CONFIG1_ID_CONFAPB ? true : false;
>> +	hdmi->id.hdmi20 = config1_id & HDMI_CONFIG1_ID_HDMI20 ? true : false;
>> +
>> +	hdmi->id.phy_type = config2_id;
>> +
>> +	hdmi->id.gpaud = config3_id & HDMI_CONFIG3_ID_GPAUD ? true : false;
>> +	hdmi->id.ahbauddma = config3_id & HDMI_CONFIG3_ID_AHBAUDDMA ?
>> +			     true : false;
>> +
>> +	hdmi->id.design = hdmi_readb(hdmi, HDMI_DESIGN_ID);
>> +	hdmi->id.revision = hdmi_readb(hdmi, HDMI_REVISION_ID);
>> +}
> What is the purpose of creating a copy of all the feature bits? As far
> as I can tell you are only ever using hdmi_id.design afterwards.
Yeah, you are right. Seems hdmi audio have long long way to go.
As for now, we just need hdmi_id.design to separate registers.

Agree, I will simplify it in next verison.

>   
>
> regards
> Philipp
regards :)
Yakir
>
>
>


[-- Attachment #1.2: Type: text/html, Size: 4160 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel