Merge branch 'master' into distexprs2

StanfordLegion · Dec 20, 2021 · bf5c956 · bf5c956
2 parents cf4f5e7 + 088791e
commit bf5c956
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 1 deletion.
diff --git a/CHANGES.txt b/CHANGES.txt
@@ -1,6 +1,19 @@
 This file lists the major changes as they appear in the stable branch.  No
 attempt is made to keep this list accurate for the master branch.
 
+Version 21.12.0 (December ??, 2021)
+  * Realm
+     - Performance improvements for multi-dimensional copies, especially
+        inter-process transfers
+     - Support for loading CUDA driver (if present) at runtime instead of
+        link time, allowing same binary to be used on systems with and without
+        CUDA-capable GPUs (enabled with -DLegion_CUDA_DYNAMIC_LOAD=ON in
+        cmake build)
+  * Build
+     - Cmake allows control of max nodes (-DLegion_MAX_NUM_NODES=...) and
+        max processors/node (-DLegion_MAX_NUM_PROCS=...) supported by
+        Legion build
+
 Version 21.09.0 (September 28, 2021)
   * Realm
      - Numerous bug fixes in the `gasnetex` network layer

diff --git a/runtime/realm/gasnetex/gasnetex_internal.cc b/runtime/realm/gasnetex/gasnetex_internal.cc
@@ -1711,7 +1711,7 @@ namespace Realm {
 	      // no more completion replies, but do pushing if there are
 	      //  ready packets
 	      ncomps = 0;
-	      do_push = has_ready_packets;
+	      do_push = has_ready_packets || !put_head.load();
 	    }
 	  }
 	} else {