We need to ensure that the kernel context is using the permanently pinned HWSP so that we can always submit a pm request from any context. By construction, the engine->kernel_context should only be using the engine->status_page.vma so let's assert that is still true when we have to submit a request for parking the engine.
Signed-off-by: Chris Wilson <[email protected]> Cc: Tvrtko Ursulin <[email protected]> --- drivers/gpu/drm/i915/gt/intel_engine_pm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c index d0a1078ef632..8ec3eecf3e39 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c @@ -142,6 +142,7 @@ static bool switch_to_kernel_context(struct intel_engine_cs *engine) return true; GEM_BUG_ON(!intel_context_is_barrier(ce)); + GEM_BUG_ON(ce->timeline->hwsp_ggtt != engine->status_page.vma); /* Already inside the kernel context, safe to power down. */ if (engine->wakeref_serial == engine->serial) -- 2.20.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
