drm/i915: Keep engine alive as we retire the context

Though we pin the context first before taking the pm wakeref, during
retire we need to unpin before dropping the pm wakeref (breaking the
"natural" onion). During the unpin, we may need to attach a cleanup
operation on to the engine wakeref, ergo we want to keep the engine
awake until after the unpin.

v2: Push the engine wakeref into the barrier so we keep the onion unwind
ordering in the request itself

Fixes: ce476c80b8bf ("drm/i915: Keep contexts pinned until after the next kernel context switch")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190618074153.16055-1-chris@chris-wilson.co.uk

diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c
index 2d019ac..293e5bc 100644 (file)
--- a/drivers/gpu/drm/i915/i915_active.c
+++ b/drivers/gpu/drm/i915/i915_active.c
@@ -4,6 +4,8 @@
  * Copyright © 2019 Intel Corporation
  */
 
+#include "gt/intel_engine_pm.h"
+
 #include "i915_drv.h"
 #include "i915_active.h"
 #include "i915_globals.h"
@@ -268,8 +270,9 @@ int i915_active_acquire_preallocate_barrier(struct i915_active *ref,
                                            struct intel_engine_cs *engine)
 {
        struct drm_i915_private *i915 = engine->i915;
+       struct llist_node *pos, *next;
        unsigned long tmp;
-       int err = 0;
+       int err;
 
        GEM_BUG_ON(!engine->mask);
        for_each_engine_masked(engine, i915, engine->mask, tmp) {
@@ -279,7 +282,7 @@ int i915_active_acquire_preallocate_barrier(struct i915_active *ref,
                node = kmem_cache_alloc(global.slab_cache, GFP_KERNEL);
                if (unlikely(!node)) {
                        err = -ENOMEM;
-                       break;
+                       goto unwind;
                }
 
                i915_active_request_init(&node->base,
@@ -288,10 +291,24 @@ int i915_active_acquire_preallocate_barrier(struct i915_active *ref,
                node->ref = ref;
                ref->count++;
 
+               intel_engine_pm_get(engine);
                llist_add((struct llist_node *)&node->base.link,
                          &ref->barriers);
        }
 
+       return 0;
+
+unwind:
+       llist_for_each_safe(pos, next, llist_del_all(&ref->barriers)) {
+               struct active_node *node;
+
+               node = container_of((struct list_head *)pos,
+                                   typeof(*node), base.link);
+               engine = (void *)rcu_access_pointer(node->base.request);
+
+               intel_engine_pm_put(engine);
+               kmem_cache_free(global.slab_cache, node);
+       }
        return err;
 }
 
@@ -328,6 +345,7 @@ void i915_active_acquire_barrier(struct i915_active *ref)
 
                llist_add((struct llist_node *)&node->base.link,
                          &engine->barrier_tasks);
+               intel_engine_pm_put(engine);
        }
        i915_active_release(ref);
 }