base,sim: implement a faster mutex for single thread case

This change applies an atomic variable to check if we really need to
obtain a mutex, and uses a condition variable to notify.

See about 5% improvement in the simulation speed.

Change-Id: I7e165987dcb587b27fae90978b9b3fde6f5563ef
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/34915
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Richard Cooper <richard.cooper@arm.com>
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>

src/base/SConscript

@@ -73,6 +73,7 @@ Source('trace.cc')
 GTest('trie.test', 'trie.test.cc')
 Source('types.cc')
 GTest('types.test', 'types.test.cc', 'types.cc')
+GTest('uncontended_mutex.test', 'uncontended_mutex.test.cc')
 
 Source('stats/group.cc')
 Source('stats/text.cc')