Re: [PATCH 2/4] [media] platform: Add Synopsys Designware HDMI RX Controller Driver

[Jose Abreu]

On 13-06-2017 11:54, Hans Verkuil wrote:
> On 06/13/17 12:31, Hans Verkuil wrote:
>> On 06/13/17 12:06, Jose Abreu wrote:
>>> Hi Hans,
>>>
>>>
>>> On 13-06-2017 11:01, Jose Abreu wrote:
>>>
>>> [snip]
>>>> Changes from RFC:
>>>> 	- Added support for HDCP 1.4
>>> [snip]
>>>> +
>>>> +/* HDCP 1.4 */
>>>> +#define DW_HDMI_HDCP14_BKSV_SIZE	2
>>>> +#define DW_HDMI_HDCP14_KEYS_SIZE	(2 * 40)
>>>> +
>>>> +struct dw_hdmi_hdcp14_key {
>>>> +	u32 seed;
>>>> +	u32 bksv[DW_HDMI_HDCP14_BKSV_SIZE];
>>>> +	u32 keys[DW_HDMI_HDCP14_KEYS_SIZE];
>>>> +	bool keys_valid;
>>>> +};
>>>> +
>>>> +struct dw_hdmi_rx_pdata {
>>>> +	/* Controller configuration */
>>>> +	unsigned int iref_clk; /* MHz */
>>>> +	struct dw_hdmi_hdcp14_key hdcp14_keys;
>>>> +	/* 5V sense interface */
>>>> +	bool (*dw_5v_status)(void __iomem *regs, int input);
>>>> +	void (*dw_5v_clear)(void __iomem *regs);
>>>> +	void __iomem *dw_5v_arg;
>>>> +	/* Zcal interface */
>>>> +	void (*dw_zcal_reset)(void __iomem *regs);
>>>> +	bool (*dw_zcal_done)(void __iomem *regs);
>>>> +	void __iomem *dw_zcal_arg;
>>>> +};
>>>> +
>>>> +#endif /* __DW_HDMI_RX_PDATA_H__ */
>>> I now have support for HDCP 1.4 in this driver. Can you send me
>>> the patches about HDCP that you mentioned a while ago?
>> This is what I have:
>>
>> https://urldefense.proofpoint.com/v2/url?u=https-3A__git.linuxtv.org_hverkuil_media-5Ftree.git_log_-3Fh-3Dhdcp&d=DwICaQ&c=DPL6_X_6JkXFx7AXWqB0tg&r=WHDsc6kcWAl4i96Vm5hJ_19IJiuxx_p_Rzo2g-uHDKw&m=89GFcWI869VtRcbl-8t_R_cPeWiZlxloSapfdhTf4vg&s=MUyGTv9NeittcGKYdkUd6gfL55UKG3aR6vqx9MDJ-Bw&e= 
>>
>> This is very old and somewhat messy.
>>
>> It uses ioctls for the bksv's, but I wonder if array/compound controls
>> wouldn't be more appropriate (those didn't exist when this was written
>> originally).
>>
>> It also needs to be checked against HDCP 2 so it can support that as well
> That's HDCP 2.2, of course.
>
>> (or at least be easily extended for that).
> Just a follow-up: I would really appreciated it if someone (you?) could get
> this finalized. The code in the branch above worked for us, but was never
> actually used in any product. It was internal test code only. It also is
> HDCP 1.4 only.

I saw the code. I can send a RFC but I'm very limited now in
terms of what I can test with this. The only available setup I
have is a receiver so I can't test transmitters neither
repeaters. As for HDCP 2.2 my implementation is *very* custom.

So, as I only have a receiver we are not very reactive. I mean I
can send events to userspace saying that auth
started/finished/failed but thats it because the transmitter
takes the lead for HDCP 1.4 and 2.2.

Also, as for what I submitted, I think I have to find another way
to do this because I'm using a "flavor" of the controller where
HDCP 1.4 keys must be programmed by SW. Right now they are passed
by pdata but these are secret keys so it doesn't make any sense
because manufacturers will not share their keys and they are
mostly embedded directly in the HW. Short story: my hdcp 1.4 code
does not make much sense.

>
> For a proper API we should think about HDCP 1.4 and 2.2 support, and also
> look at how this works for DisplayPort. Hmm, looks like DP supports HDCP 1.3
> and 2.2. I'm not sure what the differences are (if any) between 1.3 and 1.4.

Never used DP so I don't know if there are differences.

>
> Regards,
>
> 	Hans

Best regards,
Jose Miguel Abreu