[Intel-gfx] [PATCH 0/8] HuC Loading Patches

[Arkadiusz Hiler]

On Wed, Nov 23, 2016 at 02:27:35PM -0800, Anusha Srivatsa wrote:
> These patches add HuC loading support. The userspace
> patches that check for a fully loaded HuC firmware and use
> it can be found at:
> https://lists.freedesktop.org/archives/libva/2016-September/004554.html
> https://lists.freedesktop.org/archives/libva/2016-September/004555.html
>  
> v2: rebased.
> v3: rebased. Changed the code following the review comments.
> v4: Added action_lock initialization fix provided by Arick
> (Hiler Arkadiusz) to the first patch in the series- Make the
> GuC fw loading helper functions general. 
> 
> Signed-off-by: Anusha Srivatsa <anusha.srivatsa at intel.com>
> 
> Anusha Srivatsa (2):
>   drm/i915/huc: Add BXT HuC Loading Support
>   drm/i915/HuC: Add KBL huC loading Support
> 
> Peter Antoine (6):
>   drm/i915/guc: Make the GuC fw loading helper functions general. Always
>         initialize action_lock
>   drm/i915/huc: Unified css_header struct for GuC and HuC
>   drm/i915/huc: Add HuC fw loading support
>   drm/i915/huc: Add debugfs for HuC loading status check
>   drm/i915/huc: Support HuC authentication

Bolting HuC authenticaiton to the guc_submission.c is not very elegant
solution. Yes, it uses the same mechanism, but HuC can be set up having
GuC submission turned off.

This approach already caused us issues with the 'action_lock'.

I do not think that is the place for the code anyway. Instead I have
whole patch series (currently in review) that does some GuC codebase
reorganization - it evolved from the mutex_init() moving patch you've
seen.

New files, intel_uc.{c,h} are introduced - those are suitable for
renamed intel_uc_* entities from:
 " drm/i915/guc: Make the GuC fw loading helper functions general"

The series also exposes intel_guc_{send,recv}() functions which you can
use for triggering the HuC authentication (rebranded host2guc_*).

Please take a look at "GuC code reorganization" series and share your
thoughs.

-- 
Cheers,
Arek