[Intel-gfx] [PATCH] drm/i915/guc: GuC suspend path cleanup

[Daniele Ceraolo Spurio]

On 3/21/19 10:14 AM, Sujaritha Sundaresan wrote:
> Adding a call to intel_uc_suspend in i915_gem_suspend, which
> is a common point for the suspend/resume and hibernate paths.
> This fixes an unbalanced call that causes issues with the CTB
> register/deregister.
> 
> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio at intel.com>
> Cc: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Michal Wajdeczko <michal.wajdeczko at intel.com>
> Signed-off-by: Sujaritha Sundaresan <sujaritha.sundaresan at intel.com>
> ---
>   drivers/gpu/drm/i915/i915_gem.c | 3 +++
>   1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index 1a684b7e8c09..980855ebdeda 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -4371,6 +4371,9 @@ void i915_gem_suspend(struct drm_i915_private *i915)
>   	 */
>   	switch_to_kernel_context_sync(i915, i915->gt.active_engines);
>   
> +	if (!__i915_wedged(&i915->gpu_error))
> +		intel_uc_suspend(i915);

We can probably call intel_uc_suspend unconditionally, because it 
internally checks for INTEL_UC_FIRMWARE_SUCCESS, which should be false 
if we wedged due to the uc_reset_prepare we added to the wedge function.

Daniele

> +
>   	mutex_unlock(&i915->drm.struct_mutex);
>   	i915_reset_flush(i915);
>   
>