From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> We're forgetting to unpin the last_context from the ggtt at GPU reset time. This leads to the vma pin_count leaking at every reset if the last context wasn't the ring default context. Further use of the same context will trigger the pin_count check in i915_gem_object_pin() and userspace will be faced with EBUSY as a result. This plaques kms_flip rather badly since it performs lots of resets, and every fd has its own default context these days. Fix the problem by properly unpinning the last context at reset. Testcase: igt/gem_ctx_exec/reset-pin-leak Signed-off-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> --- drivers/gpu/drm/i915/i915_gem_context.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index 3ffe308..e362c96 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -382,6 +382,9 @@ void i915_gem_context_reset(struct drm_device *dev) dctx->obj->active = 0; } + if (ring->last_context->obj && i == RCS) + i915_gem_object_ggtt_unpin(ring->last_context->obj); + i915_gem_context_unreference(ring->last_context); i915_gem_context_reference(dctx); ring->last_context = dctx; -- 1.8.5.5 _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
