[Intel-gfx] [PATCH v3 3/6] drm/i915/huc: Add HuC fw loading support

[Dave Gordon]

On 06/07/16 15:24, Peter Antoine wrote:
> The HuC loading process is similar to GuC. The intel_uc_fw_fetch()
> is used for both cases.
>
> HuC loading needs to be before GuC loading. The WOPCM setting must
> be done early before loading any of them.
>
> v2: rebased on-top of drm-intel-nightly.
>     removed if(HAS_GUC()) before the guc call. (D.Gordon)
>     update huc_version number of format.
> v3: rebased to drm-intel-nightly, changed the file name format to
>     match the one in the huc package.
>     Changed dev->dev_provate to to_i915()
>
> Signed-off-by: Alex Dai <yu.dai at intel.com>
> Signed-off-by: Peter Antoine <peter.antoine at intel.com>
> ---
>  drivers/gpu/drm/i915/Makefile           |   1 +
>  drivers/gpu/drm/i915/i915_drv.c         |   3 +
>  drivers/gpu/drm/i915/i915_drv.h         |   3 +
>  drivers/gpu/drm/i915/i915_guc_reg.h     |   3 +
>  drivers/gpu/drm/i915/intel_guc.h        |   1 +
>  drivers/gpu/drm/i915/intel_guc_loader.c |  26 ++--
>  drivers/gpu/drm/i915/intel_huc.h        |  44 ++++++
>  drivers/gpu/drm/i915/intel_huc_loader.c | 267 ++++++++++++++++++++++++++++++++
>  8 files changed, 336 insertions(+), 12 deletions(-)
>  create mode 100644 drivers/gpu/drm/i915/intel_huc.h
>  create mode 100644 drivers/gpu/drm/i915/intel_huc_loader.c

I found another problem, which at root is because this was *copied* from 
the GuC loading code some time ago, rather than being implemented in a 
single generic function. The GuC code has been changed since it was 
cloned, but of course the corresponding updates are missing from the HuC 
version.

[snip]

> +
> +	/* Start the DMA */
> +	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(HUC_UKERNEL | START_DMA));
> +
> +	/* Wait for DMA to finish */
> +	ret = wait_for_atomic((I915_READ(DMA_CTRL) & START_DMA) == 0, 50);

Can't use wait_for_atomic() here. The GuC version now reads:

         /*
          * Wait for the DMA to complete & the GuC to start up.
          * NB: Docs recommend not using the interrupt for completion.
          * Measurements indicate this should take no more than 20ms,
	 * so a timeout here indicates that the GuC has is unusable.
          * (Higher levels of the driver will attempt to fall back to
          * execlist mode if this happens.)
          */
         ret = wait_for(guc_ucode_response(dev_priv, &status), 100);

.Dave.