[Intel-gfx] [PATCH v5] drm/i915: Introduce concept of a sub-platform

[Chris Wilson]

Quoting Tvrtko Ursulin (2019-03-15 17:09:28)
> diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
> index 3d8020888604..e3360a31f8d3 100644
> --- a/drivers/gpu/drm/i915/i915_gpu_error.c
> +++ b/drivers/gpu/drm/i915/i915_gpu_error.c
> @@ -677,6 +677,7 @@ static void __err_print_to_sgl(struct drm_i915_error_state_buf *m,
>         err_printf(m, "Reset count: %u\n", error->reset_count);
>         err_printf(m, "Suspend count: %u\n", error->suspend_count);
>         err_printf(m, "Platform: %s\n", intel_platform_name(error->device_info.platform));
> +       err_printf(m, "Subplatform: 0x%x\n", intel_subplatform(m->i915));

We didn't take a copy of the runtime_info? Ww should have done at least,
that's even more volatile than device_info.

It'll do for now, but would be nicer if RUNTIME_INFO(i915) wasn't baked
so hard into the subplatform :)

> +void intel_device_info_subplatform_init(struct drm_i915_private *i915)
> +{
> +       const unsigned int pi =
> +               __platform_mask_index(i915, INTEL_INFO(i915)->platform);
> +       const unsigned int pb =
> +               __platform_mask_bit(i915, INTEL_INFO(i915)->platform);
> +       u16 devid = INTEL_DEVID(i915);
> +       u32 mask = 0;
> +
> +       RUNTIME_INFO(i915)->platform_mask[pi] = BIT(pb);
> +
> +       if (IS_PINEVIEW(i915)) {
> +               if (devid == 0xa001)
> +                       mask |= BIT(INTEL_SUBPLATFORM_PINEVIEW_G);
> +               else if (devid == 0xa011)
> +                       mask |= BIT(INTEL_SUBPLATFORM_PINEVIEW_M);

I take it, leaving is-mobile conversion for the display gurus?
-Chris