[Intel-gfx] [PATCH 05/18] drm/i915: Only sanitize GEM from late suspend

[Chris Wilson]

During testing we encounter a conflict between SUSPEND_TEST_DEVICES and
disabling reset (gem_eio/suspend). This results in the device continuing
on without being reset, but since it has gone through HW sanitization to
account for the suspend/resume cycle, we have to assume the device has
been reset to its defaults. A simple way around this is to skip the
sanitize phase for SUSPEND_TEST_DEVICES by moving it to suspend-late.

Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/i915_drv.c |  6 +++++-
 drivers/gpu/drm/i915/i915_drv.h |  1 +
 drivers/gpu/drm/i915/i915_gem.c | 20 ++++++++++++--------
 3 files changed, 18 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 9d6ac7f44812..be8137b90a51 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -636,6 +636,8 @@ static const struct vga_switcheroo_client_ops i915_switcheroo_ops = {
 
 static void i915_gem_fini(struct drm_i915_private *dev_priv)
 {
+	i915_gem_suspend_late(dev_priv);
+
 	/* Flush any outstanding unpin_work. */
 	i915_gem_drain_workqueue(dev_priv);
 
@@ -1605,7 +1607,6 @@ static int i915_drm_suspend(struct drm_device *dev)
 	opregion_target_state = suspend_to_idle(dev_priv) ? PCI_D1 : PCI_D3cold;
 	intel_opregion_notify_adapter(dev_priv, opregion_target_state);
 
-	intel_uncore_suspend(dev_priv);
 	intel_opregion_unregister(dev_priv);
 
 	intel_fbdev_set_suspend(dev, FBINFO_STATE_SUSPENDED, true);
@@ -1627,7 +1628,10 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation)
 
 	disable_rpm_wakeref_asserts(dev_priv);
 
+	i915_gem_suspend_late(dev_priv);
+
 	intel_display_set_init_power(dev_priv, false);
+	intel_uncore_suspend(dev_priv);
 
 	/*
 	 * In case of firmware assisted context save/restore don't manually
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 487922f88b76..4257ffc1bae1 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -3169,6 +3169,7 @@ void i915_gem_cleanup_engines(struct drm_i915_private *dev_priv);
 int i915_gem_wait_for_idle(struct drm_i915_private *dev_priv,
 			   unsigned int flags);
 int __must_check i915_gem_suspend(struct drm_i915_private *dev_priv);
+void i915_gem_suspend_late(struct drm_i915_private *dev_priv);
 void i915_gem_resume(struct drm_i915_private *dev_priv);
 int i915_gem_fault(struct vm_fault *vmf);
 int i915_gem_object_wait(struct drm_i915_gem_object *obj,
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 5a7e0b388ad0..b0980b3186e7 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -5051,6 +5051,17 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv)
 	if (WARN_ON(!intel_engines_are_idle(dev_priv)))
 		i915_gem_set_wedged(dev_priv); /* no hope, discard everything */
 
+	intel_runtime_pm_put(dev_priv);
+	return 0;
+
+err_unlock:
+	mutex_unlock(&dev->struct_mutex);
+	intel_runtime_pm_put(dev_priv);
+	return ret;
+}
+
+void i915_gem_suspend_late(struct drm_i915_private *dev_priv)
+{
 	/*
 	 * Neither the BIOS, ourselves or any other kernel
 	 * expects the system to be in execlists mode on startup,
@@ -5070,16 +5081,9 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv)
 	 * machines is a good idea, we don't - just in case it leaves the
 	 * machine in an unusable condition.
 	 */
+
 	intel_uc_sanitize(dev_priv);
 	i915_gem_sanitize(dev_priv);
-
-	intel_runtime_pm_put(dev_priv);
-	return 0;
-
-err_unlock:
-	mutex_unlock(&dev->struct_mutex);
-	intel_runtime_pm_put(dev_priv);
-	return ret;
 }
 
 void i915_gem_resume(struct drm_i915_private *i915)
-- 
2.17.0