Reland "Fix memory leak caused by field type in descriptor array."

test/mjsunit/mjsunit.status

 # Copyright 2012 the V8 project authors. All rights reserved.
 # Redistribution and use in source and binary forms, with or without
 # modification, are permitted provided that the following conditions are
 # met:
 #
 #     * Redistributions of source code must retain the above copyright
 #       notice, this list of conditions and the following disclaimer.
 #     * Redistributions in binary form must reproduce the above
 #       copyright notice, this list of conditions and the following
 #       disclaimer in the documentation and/or other materials provided
@@ skipping 225 matching lines @@
   'compiler/osr-regress-max-locals': [PASS, NO_VARIANTS],
   'math-floor-of-div': [PASS, NO_VARIANTS],
   'unicodelctest': [PASS, NO_VARIANTS],
   'unicodelctest-no-optimization': [PASS, NO_VARIANTS],
 
   # Too slow for gc stress.
   'asm/embenchen/box2d': [SKIP],
 
   # Issue 3723.
   'regress/regress-3717': [SKIP],
+  # Issue 3924.
+  'mjsunit/debug-clearbreakpointgroup': [SKIP],
+  # Issue 3969.
+  'mjsunit/debug-references': [SKIP],
 }],  # 'gc_stress == True'
 
 ##############################################################################
 ['byteorder == big', {
   # Emscripten requires little-endian, skip all tests on big endian platforms.
   'asm/embenchen/*': [SKIP],
 }],  # 'byteorder == big'
 
 ##############################################################################
 ['arch == arm64 or arch == android_arm64', {
@@ skipping 334 matching lines @@
 ['arch == ppc and simulator_run == True or arch == ppc64 and simulator_run == True', {
 
   # take too long with the simulator.
   'regress/regress-1132': [SKIP],
 
   # these two issues started after uplevel, believe they share common cause
   'debug-references': [SKIP],
   'mirror-object': [SKIP],
 }],  # 'arch == ppc and simulator_run == True'
 ]