[PATCH v9 10/39] drm/i915: Implement HDCP2.2 receiver authentication

C, Ramalingam ramalingam.c at intel.com
Wed Dec 19 15:48:20 UTC 2018


On 12/19/2018 9:05 PM, Daniel Vetter wrote:
> On Wed, Dec 19, 2018 at 08:35:48PM +0530, C, Ramalingam wrote:
>> On 12/19/2018 8:05 PM, Daniel Vetter wrote:
>>> On Thu, Dec 13, 2018 at 09:31:12AM +0530, Ramalingam C wrote:
>>>> +	if (!hdcp->is_repeater && shim->config_stream_type) {
>>>> +		/*
>>>> +		 * Errata for DP: As Stream type is used for encryption,
>>>> +		 * Receiver should be communicated with stream type for the
>>>> +		 * decryption of the content.
>>>> +		 * Repeater will be communicated with stream type as a
>>>> +		 * part of it's auth later in time.
>>>> +		 */
>>> I'm not following what you want to say with this comment, and haven't
>>> found anything in the hdcp2 dp spec about this either.
>> this is there in the errata of DP HDCP2.2 spec.
>> hdcp2.2 encryption algo involves the stream type as a parameter.
>> And as part of hdcp auth mentioned in base spec DP repeaters receive that details to decrypt the content.
>> But DP receivers dont get it. So errata adds the missing piece for decryption.
> Ok I found it, but the fake HDCP_2_2_ERRATA_DP_STREAM_TYPE define is kinda
> annoying. It doesn't exist in the spec, but we put it into the drm_hdcp.h
> header. Since you have a special ->config_stream_type hook for this
> anyway, and it's only needed for DP, please move that into the shim
> implementation.
>
> And then a follow-up patch to remove the fake #define from drm_hdcp.h.

This is defined to make the hdcp_shim interface to look common between DP and HDMI.
We can move this errata msg definition and its msg_id into intel_drv.h just for the intel's consumption.

-Ram

> -Daniel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20181219/6923c565/attachment.html>


More information about the dri-devel mailing list