~mwhudson/pypy/imported-gc-jit-hack

Viewing all changes in revision 23277.

  • Committer: arigo
  • Date: 2009-11-05 10:46:09 UTC
  • Revision ID: svn-v4:fd0d7bf2-dfb6-0310-8d31-b7ecfe96aada:pypy/branch/gc-jit-hack:68987
This is the best result so far.  At least in this case I understand the
results: they seem to clearly show that a lot of the objects allocated
during jit-tracing actually survive.  The results are not good enough --
they say that jit-tracing takes only half the time, but the truth is
that it's always soon followed by a lengthy minor collection, so that
the total time (as well as the total GC time) end up not really
improved.  Closing this branch for now.

expand all expand all

Show diffs side-by-side

added added

removed removed

Lines of Context: