drm/xe: fix xe_device_mem_access_get() races

It looks like there is at least one race here, given that the
pm_runtime_suspended() check looks to return false if we are in the
process of suspending the device (RPM_SUSPENDING vs RPM_SUSPENDED).  We
later also do xe_pm_runtime_get_if_active(), but since the device is
suspending or has now suspended, this doesn't do anything either.
Following from this we can potentially return from
xe_device_mem_access_get() with the device suspended or about to be,
leading to broken behaviour.

Attempt to fix this by always grabbing the runtime ref when our internal
ref transitions from 0 -> 1. The hard part is then dealing with the
runtime_pm callbacks also calling xe_device_mem_access_get() and
deadlocking, which the pm_runtime_suspended() check prevented.

v2:
 - ct->lock looks to be primed with fs_reclaim, so holding that and then
   allocating memory will cause lockdep to complain. Now that we
   unconditionally grab the mem_access.lock around mem_access_{get,put}, we
   need to change the ordering wrt to grabbing the ct->lock, since some of
   the runtime_pm routines can allocate memory (or at least that's what
   lockdep seems to suggest). Hopefully not a big deal.  It might be that
   there were already issues with this, just that the atomics where
   "hiding" the potential issues.
v3:
 - Use Thomas Hellström' idea with tracking the active task that is
   executing in the resume or suspend callback, in order to avoid
   recursive resume/suspend calls deadlocking on itself.
 - Split the ct->lock change.
v4:
 - Add smb_mb() around accessing the pm_callback_task for extra safety.
   (Thomas Hellström)
v5:
 - Clarify the kernel-doc for the mem_access.lock, given that it is quite
   strange in what it protects (data vs code). The real motivation is to
   aid lockdep. (Rodrigo Vivi)
v6:
 - Split out the lock change. We still want this as a lockdep aid but
   only for the xe_device_mem_access_get() path. Sticking a lock on the
   put() looks be a no-go, also the runtime_put() there is always async.
 - Now that the lock is gone move to atomics and rely on the pm code
   serialising multiple callers on the 0 -> 1 transition.
 - g2h_worker_func() looks to be the next issue, given that
   suspend-resume callbacks are using CT, so try to handle that.
v7:
 - Add xe_device_mem_access_get_if_ongoing(), and use it in
   g2h_worker_func().
v8 (Anshuman):
 - Just always grab the rpm, instead of just on the 0 -> 1 transition,
   which is a lot clearer and simplifies the code quite a bit.
v9:
 - Make sure we also adjust the CT fast-path with if-active.

Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/258
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Anshuman Gupta <anshuman.gupta@intel.com>
Acked-by: Anshuman Gupta <anshuman.gupta@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

1 files changed, 43 insertions, 25 deletions

diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index f336aec7085d..04b995aa848f 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -155,37 +155,65 @@ void xe_pm_runtime_fini(struct xe_device *xe)
 	pm_runtime_forbid(dev);
 }
 
+static void xe_pm_write_callback_task(struct xe_device *xe,
+				      struct task_struct *task)
+{
+	WRITE_ONCE(xe->pm_callback_task, task);
+
+	/*
+	 * Just in case it's somehow possible for our writes to be reordered to
+	 * the extent that something else re-uses the task written in
+	 * pm_callback_task. For example after returning from the callback, but
+	 * before the reordered write that resets pm_callback_task back to NULL.
+	 */
+	smp_mb(); /* pairs with xe_pm_read_callback_task */
+}
+
+struct task_struct *xe_pm_read_callback_task(struct xe_device *xe)
+{
+	smp_mb(); /* pairs with xe_pm_write_callback_task */
+
+	return READ_ONCE(xe->pm_callback_task);
+}
+
 int xe_pm_runtime_suspend(struct xe_device *xe)
 {
 	struct xe_gt *gt;
 	u8 id;
-	int err;
+	int err = 0;
 
-	if (xe->d3cold.allowed) {
-		if (xe_device_mem_access_ongoing(xe))
-			return -EBUSY;
+	if (xe->d3cold.allowed && xe_device_mem_access_ongoing(xe))
+		return -EBUSY;
+
+	/* Disable access_ongoing asserts and prevent recursive pm calls */
+	xe_pm_write_callback_task(xe, current);
 
+	if (xe->d3cold.allowed) {
 		err = xe_bo_evict_all(xe);
 		if (err)
-			return err;
+			goto out;
 	}
 
 	for_each_gt(gt, xe, id) {
 		err = xe_gt_suspend(gt);
 		if (err)
-			return err;
+			goto out;
 	}
 
 	xe_irq_suspend(xe);
-
-	return 0;
+out:
+	xe_pm_write_callback_task(xe, NULL);
+	return err;
 }
 
 int xe_pm_runtime_resume(struct xe_device *xe)
 {
 	struct xe_gt *gt;
 	u8 id;
-	int err;
+	int err = 0;
+
+	/* Disable access_ongoing asserts and prevent recursive pm calls */
+	xe_pm_write_callback_task(xe, current);
 
 	/*
 	 * It can be possible that xe has allowed d3cold but other pcie devices
@@ -199,7 +227,7 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 		for_each_gt(gt, xe, id) {
 			err = xe_pcode_init(gt);
 			if (err)
-				return err;
+				goto out;
 		}
 
 		/*
@@ -208,7 +236,7 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 		 */
 		err = xe_bo_restore_kernel(xe);
 		if (err)
-			return err;
+			goto out;
 	}
 
 	xe_irq_resume(xe);
@@ -219,10 +247,11 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 	if (xe->d3cold.allowed && xe->d3cold.power_lost) {
 		err = xe_bo_restore_user(xe);
 		if (err)
-			return err;
+			goto out;
 	}
-
-	return 0;
+out:
+	xe_pm_write_callback_task(xe, NULL);
+	return err;
 }
 
 int xe_pm_runtime_get(struct xe_device *xe)
@@ -236,19 +265,8 @@ int xe_pm_runtime_put(struct xe_device *xe)
 	return pm_runtime_put_autosuspend(xe->drm.dev);
 }
 
-/* Return true if resume operation happened and usage count was increased */
-bool xe_pm_runtime_resume_if_suspended(struct xe_device *xe)
-{
-	/* In case we are suspended we need to immediately wake up */
-	if (pm_runtime_suspended(xe->drm.dev))
-		return !pm_runtime_resume_and_get(xe->drm.dev);
-
-	return false;
-}
-
 int xe_pm_runtime_get_if_active(struct xe_device *xe)
 {
-	WARN_ON(pm_runtime_suspended(xe->drm.dev));
 	return pm_runtime_get_if_active(xe->drm.dev, true);
 }