kk: Propagate availability before we reset it in vkCmdResetQueryPool

Required to avoid availability write races. We could have an availability
update pending so adding the reset availability write to the same pool of
writes led to write races. Avoid this by flushing writes before reseting
queries.

Acked-by: Arcady Goldmints-Orlov <arcady@lunarg.com>
Signed-off-by: Aitor Camacho <aitor@lunarg.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/38957>

src/kosmickrisp/vulkan/kk_query_pool.c

@@ -254,10 +254,13 @@ kk_CmdResetQueryPool(VkCommandBuffer commandBuffer, VkQueryPool queryPool,
 {
    VK_FROM_HANDLE(kk_cmd_buffer, cmd, commandBuffer);
    VK_FROM_HANDLE(kk_query_pool, pool, queryPool);
+   /* Need to flush other availabilities just in case there is a reset after it
+    * was made available but the writes have not propagated yet. Need to avoid
+    * data rances in the writes. This is save to do sice vkCmdResetQueryPool
+    * cannot be called when a render pass is active. */
+   upload_queue_writes(cmd);
    emit_zero_queries(cmd, pool, firstQuery, queryCount, false);
-   /* If we are not mid encoder, just upload the writes */
-   if (cmd->encoder->main.last_used == KK_ENC_NONE)
-      upload_queue_writes(cmd);
+   upload_queue_writes(cmd);
 }
 
 VKAPI_ATTR void VKAPI_CALL