drm/i915: Make the stale cached active node available for any timeline

Rather than require the next timeline after idling to match the MRU
before idling, reset the index on the node and allow it to match the
first request. However, this requires cmpxchg(u64) and so is not trivial
on 32b, so for compatibility we just fallback to keeping the cached node
pointing to the MRU timeline.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Thomas Hellström <thomas.hellstrom@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200731085015.32368-5-chris@chris-wilson.co.uk
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c
index e63578c..89c34a6 100644 (file)
--- a/drivers/gpu/drm/i915/i915_active.c
+++ b/drivers/gpu/drm/i915/i915_active.c
@@ -157,6 +157,10 @@ __active_retire(struct i915_active *ref)
                rb_link_node(&ref->cache->node, NULL, &ref->tree.rb_node);
                rb_insert_color(&ref->cache->node, &ref->tree);
                GEM_BUG_ON(ref->tree.rb_node != &ref->cache->node);
+
+               /* Make the cached node available for reuse with any timeline */
+               if (IS_ENABLED(CONFIG_64BIT))
+                       ref->cache->timeline = 0; /* needs cmpxchg(u64) */
        }
 
        spin_unlock_irqrestore(&ref->tree_lock, flags);
@@ -235,6 +239,8 @@ static struct active_node *__active_lookup(struct i915_active *ref, u64 idx)
 {
        struct active_node *it;
 
+       GEM_BUG_ON(idx == 0); /* 0 is the unordered timeline, rsvd for cache */
+
        /*
         * We track the most recently used timeline to skip a rbtree search
         * for the common case, under typical loads we never need the rbtree
@@ -243,8 +249,28 @@ static struct active_node *__active_lookup(struct i915_active *ref, u64 idx)
         * current timeline.
         */
        it = READ_ONCE(ref->cache);
-       if (it && it->timeline == idx)
-               return it;
+       if (it) {
+               u64 cached = READ_ONCE(it->timeline);
+
+               /* Once claimed, this slot will only belong to this idx */
+               if (cached == idx)
+                       return it;
+
+#ifdef CONFIG_64BIT /* for cmpxchg(u64) */
+               /*
+                * An unclaimed cache [.timeline=0] can only be claimed once.
+                *
+                * If the value is already non-zero, some other thread has
+                * claimed the cache and we know that is does not match our
+                * idx. If, and only if, the timeline is currently zero is it
+                * worth competing to claim it atomically for ourselves (for
+                * only the winner of that race will cmpxchg return the old
+                * value of 0).
+                */
+               if (!cached && !cmpxchg(&it->timeline, 0, idx))
+                       return it;
+#endif
+       }
 
        BUILD_BUG_ON(offsetof(typeof(*it), node));