In order for bos to retire eventually, a request must be sent down the
ring. This is expected, for example, by occlusion queries for which mesa
will wait upon (whilst running glean) before issuing more batches and so
the normal activity upon the ring is suspended and we need to emit a
request to clear the idle ring.
Reported-by: Jinjin, Wang <jinjin.wang@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=30380
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* use this buffer rather sooner than later, so issuing the required
* flush earlier is beneficial.
*/
- if (obj->write_domain & I915_GEM_GPU_DOMAINS)
+ if (obj->write_domain & I915_GEM_GPU_DOMAINS) {
i915_gem_flush_ring(dev, file_priv,
obj_priv->ring,
0, obj->write_domain);
+ } else if (obj_priv->ring->outstanding_lazy_request) {
+ /* This ring is not being cleared by active usage,
+ * so emit a request to do so.
+ */
+ u32 seqno = i915_add_request(dev,
+ NULL, NULL,
+ obj_priv->ring);
+ if (seqno == 0)
+ ret = -ENOMEM;
+ }
/* Update the active list for the hardware's current position.
* Otherwise this only updates on a delayed timer or when irqs