[Intel-gfx] [PATCH 6/6] drm/i915/gt: Always poison the kernel_context image before unparking

[Mika Kuoppala]

Chris Wilson <chris at chris-wilson.co.uk> writes:

> Keep scrubbing the kernel_context image with poison before we reset it
> in order to demonstrate that we will be resilient in the case where it
> is accidentally overwritten on idle.
>
> Suggested-by: Imre Deak <imre.deak at intel.com>
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Imre Deak <imre.deak at intel.com>
> ---
>  drivers/gpu/drm/i915/gt/intel_context_types.h |  2 ++
>  drivers/gpu/drm/i915/gt/intel_engine_pm.c     | 18 +++++++++++++++++-
>  drivers/gpu/drm/i915/gt/intel_lrc.c           |  4 ++--
>  3 files changed, 21 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h
> index 9527a659546c..ca1420fb8b53 100644
> --- a/drivers/gpu/drm/i915/gt/intel_context_types.h
> +++ b/drivers/gpu/drm/i915/gt/intel_context_types.h
> @@ -17,6 +17,8 @@
>  #include "intel_engine_types.h"
>  #include "intel_sseu.h"
>  
> +#define CONTEXT_REDZONE POISON_INUSE
> +
>  struct i915_gem_context;
>  struct i915_vma;
>  struct intel_context;
> diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
> index 1b9f73948f22..ea90ab3e396e 100644
> --- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c
> +++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
> @@ -37,8 +37,24 @@ static int __engine_unpark(struct intel_wakeref *wf)
>  
>  	/* Discard stale context state from across idling */
>  	ce = engine->kernel_context;
> -	if (ce)
> +	if (ce) {
> +		GEM_BUG_ON(test_bit(CONTEXT_VALID_BIT, &ce->flags));
> +
> +		/* First poison the image to verify we never fully trust it */
> +		if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) && ce->state) {
> +			struct drm_i915_gem_object *obj = ce->state->obj;
> +			int type = i915_coherent_map_type(engine->i915);
> +
> +			map = i915_gem_object_pin_map(obj, type);
> +			if (!IS_ERR(map)) {
> +				memset(map, CONTEXT_REDZONE, obj->base.size);

Just pondering if the dword granularity would suffice.

Reviewed-by: Mika Kuoppala <mika.kuoppala at linux.intel.com>

> +				i915_gem_object_flush_map(obj);
> +				i915_gem_object_unpin_map(obj);
> +			}
> +		}
> +
>  		ce->ops->reset(ce);
> +	}
>  
>  	if (engine->unpark)
>  		engine->unpark(engine);
> diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
> index 6d26d84812b6..f81e70cfd194 100644
> --- a/drivers/gpu/drm/i915/gt/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
> @@ -2409,7 +2409,7 @@ set_redzone(void *vaddr, const struct intel_engine_cs *engine)
>  
>  	vaddr += engine->context_size;
>  
> -	memset(vaddr, POISON_INUSE, I915_GTT_PAGE_SIZE);
> +	memset(vaddr, CONTEXT_REDZONE, I915_GTT_PAGE_SIZE);
>  }
>  
>  static void
> @@ -2420,7 +2420,7 @@ check_redzone(const void *vaddr, const struct intel_engine_cs *engine)
>  
>  	vaddr += engine->context_size;
>  
> -	if (memchr_inv(vaddr, POISON_INUSE, I915_GTT_PAGE_SIZE))
> +	if (memchr_inv(vaddr, CONTEXT_REDZONE, I915_GTT_PAGE_SIZE))
>  		dev_err_once(engine->i915->drm.dev,
>  			     "%s context redzone overwritten!\n",
>  			     engine->name);
> -- 
> 2.25.0.rc0
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx