whiteknight/imcc_compreg_pmc branch

[James E Keenan]

Andrew Whitworth wrote:
> I've fixed several problems in this branch (with help from plobsing++
> and others). I would like to get as much help testing and exercising
> this branch as possible. Problematic architectures (I'm looking at
> your low-mem box for this, kid51!) and various HLLs. I would like to
> try to get this merged in before 3.3, but if there are problems
> obviously we will hold off on that.
>

I got one test failure in this branch on Darwin/PPC --  a failure which 
I do not get in master on this machine:

prove -v t/perl/Parrot_Test.t

not ok 66 - pir_error_output_like: todo

#   Failed test 'pir_error_output_like: todo'
#   at t/perl/Parrot_Test.t line 328.
# STDERR is:
# #   Failed (TODO) test 'pir_error_output_like: todo'
# #   at t/perl/Parrot_Test.t line 317.
# # Expected error but exited cleanly
# # Received:
# # foo
# #
# # Expected:
# # /bar/
# #
#
# not:
# #   Failed (TODO) test 'pir_error_output_like: todo'
# #   at t/perl/Parrot_Test.t line 317.
# # Expected error but exited cleanly
# # Received:
# # foo
# #
# # Expected:
# # /bar/
# #
#
# as expected

Failed 1/112 subtests
         (less 1 skipped subtest: 110 okay)

Test Summary Report
-------------------
t/perl/Parrot_Test.t (Wstat: 0 Tests: 112 Failed: 1)
   Failed test:  66
Files=1, Tests=112, 14 wallclock secs ( 0.20 usr  0.05 sys +  1.66 cusr 
  2.21 csys =  4.12 CPU)
Result: FAIL