commit 5ae2c9335a0c4eeaf4c67a52d066f16aad1aa907
author Mathieu Chartier <mathieuc@google.com> Fri Mar 28 16:22:20 2014 -0700
committer Mathieu Chartier <mathieuc@google.com> Fri Mar 28 16:24:55 2014 -0700
tree f3d6a0d2a5e67ac5bcfe984f196465078ba9d7a5
parent 17b3fb82db845bae4713ca9cbe2f0c9a2dd2abc2

Fix non concurrent mark sweep ergonomics.

Previously we would continue to do sticky GC until the sticky GC did
not free enough memory for the allocation, this was excessive since it
could do one sticky GC per allocation. The new logic uses the next GC
type before trying all the GCs in the plan.

Before memalloc benchmark (non concurrent mark sweep):
Total time spent in GC: 11.212701s
Score: 7790

After:
Total time spent in GC: 9.422676s
Score: 6870

Change-Id: Iba75b70ea825ef3fd4b3e064d4f12c2fe5a3b176