Mesa (staging/20.1): radv: Fix threading issue with submission refcounts.

[GitLab Mirror]

Module: Mesa
Branch: staging/20.1
Commit: 5558c1c581bb94819587f735736a18cb072f3aa5
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=5558c1c581bb94819587f735736a18cb072f3aa5

Author: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Date:   Sat Aug 29 03:25:02 2020 +0200

radv: Fix threading issue with submission refcounts.

If decrement == 0 then:

- it isn't safe to access the submission
- even if it is, checking that the result of the atomic_sub is 0
  doesn't given an unique owner anymore.

So skip it. The submission always starts out with refcount >= 1,
so first one to decrement to 0 still get dibs on executing it.

Fixes: 4aa75bb3bdd "radv: Add wait-before-submit support for timelines."
Reviewed-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6478>
(cherry picked from commit 6b75262941b55960e2f73d93f85020fa6c9c2d2f)

---

 .pick_status.json            | 2 +-
 src/amd/vulkan/radv_device.c | 6 ++++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/.pick_status.json b/.pick_status.json
index 8696d3264d1..73f827f803a 100644
--- a/.pick_status.json
+++ b/.pick_status.json
@@ -229,7 +229,7 @@
         "description": "radv: Fix threading issue with submission refcounts.",
         "nominated": true,
         "nomination_type": 1,
-        "resolution": 0,
+        "resolution": 1,
         "master_sha": null,
         "because_sha": "4aa75bb3bdd195d4715ee8fae51bfb0c0fcd823b"
     },
diff --git a/src/amd/vulkan/radv_device.c b/src/amd/vulkan/radv_device.c
index 3fd6c86ef19..58367cbded2 100644
--- a/src/amd/vulkan/radv_device.c
+++ b/src/amd/vulkan/radv_device.c
@@ -4636,6 +4636,12 @@ radv_queue_enqueue_submission(struct radv_deferred_queue_submission *submission,
 	 * submitted, but if the queue was empty, we decrement ourselves as there is no previous
 	 * submission. */
 	uint32_t decrement = submission->wait_semaphore_count - wait_cnt + (is_first ? 1 : 0);
+
+	/* if decrement is zero, then we don't have a refcounted reference to the
+	 * submission anymore, so it is not safe to access the submission. */
+	if (!decrement)
+		return VK_SUCCESS;
+
 	if (__atomic_sub_fetch(&submission->submission_wait_count, decrement, __ATOMIC_ACQ_REL) == 0) {
 		list_addtail(&submission->processing_list, processing_list);
 	}