From: Chris Wilson Date: Fri, 9 Sep 2016 13:12:01 +0000 (+0100) Subject: drm/i915: Serialise execbuf operation after a dma-buf reservation object X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=851ba2d6977d6ada31366ae0b791a69132f2c8a7;p=GitHub%2Fmoto-9609%2Fandroid_kernel_motorola_exynos9610.git drm/i915: Serialise execbuf operation after a dma-buf reservation object Now that we can wait upon fences before emitting the request, it becomes trivial to wait upon any implicit fence provided by the dma-buf reservation object. To protect against failure, we force any asynchronous waits on a foreign fence to timeout after 10s - so that a stall in another driver does not permanently cripple ourselves. Still unpleasant though! Testcase: igt/prime_vgem/fence-wait Signed-off-by: Chris Wilson Reviewed-by: John Harrison Reviewed-by: Joonas Lahtinen Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-21-chris@chris-wilson.co.uk --- diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index ccaf15ba4e32..33c85227643d 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -1131,6 +1131,7 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req, list_for_each_entry(vma, vmas, exec_list) { struct drm_i915_gem_object *obj = vma->obj; + struct reservation_object *resv; if (obj->flags & other_rings) { ret = i915_gem_request_await_object @@ -1139,6 +1140,16 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req, return ret; } + resv = i915_gem_object_get_dmabuf_resv(obj); + if (resv) { + ret = i915_sw_fence_await_reservation + (&req->submit, resv, &i915_fence_ops, + obj->base.pending_write_domain, 10*HZ, + GFP_KERNEL | __GFP_NOWARN); + if (ret < 0) + return ret; + } + if (obj->base.write_domain & I915_GEM_DOMAIN_CPU) i915_gem_clflush_object(obj, false); }