[Intel-gfx] [PATCH 01/43] drm: hdcp2.2 authentication msg definitions

[Ramalingam C]

On Thursday 15 February 2018 01:10 AM, Jani Nikula wrote:
> On Wed, 14 Feb 2018, "Winkler, Tomas" <tomas.winkler at intel.com> wrote:
>>> This patch defines the hdcp2.2 protocol messages for the
>>> HDCP2.2 authentication.
>>>
>>> Signed-off-by: Ramalingam C <ramalingam.c at intel.com>
>>> ---
>>>   include/drm/drm_hdcp.h | 226
>>> +++++++++++++++++++++++++++++++++++++++++++++++++
>>>   1 file changed, 226 insertions(+)
>>>
>>> diff --git a/include/drm/drm_hdcp.h b/include/drm/drm_hdcp.h index
>>> 562fa7df2637..9661c700cebb 100644
>>> --- a/include/drm/drm_hdcp.h
>>> +++ b/include/drm/drm_hdcp.h
>>> @@ -38,4 +38,230 @@
>>>   #define DRM_HDCP_DDC_BSTATUS			0x41
>>>   #define DRM_HDCP_DDC_KSV_FIFO			0x43
>>>
>>> +#define DRM_HDCP_1_4_SRM_ID			0x8
>>> +#define DRM_HDCP_1_4_VRL_LENGTH_SIZE		3
>>> +#define DRM_HDCP_1_4_DCP_SIG_SIZE		40
>>> +
>>> +struct cp_srm_header {
>>> +	struct {
>>> +		uint8_t reserved_hi:4;
>>> +		uint8_t srm_id:4;
>>> +		uint8_t reserved_lo;
>>> +	} spec_indicator;
>> Do you really want to work with bit fields?  I mean in all the all structures.
> We *can't* use bitfields in drm core for (un)marshalling. They depend on
> endianness. (Thanks to folks on #dri-devel for confirming.) We use them
> at places in i915 where we can be pretty sure about running on
> little-endian machines, but that doesn't hold here.
>
> Packed structs are fine otherwise though, just not bitfields.
Thanks Jani. I will rework on that.

--Ram
>
> BR,
> Jani.
>