Test failures in t/run/exit.t

[doughera]

For revision r43946, I'm getting 4 test failures in t/run/exit.t on OpenSolaris with Sun's compiler. Unfortunately, the message from the failed tests is rather less than helpful. It claims that '0' ne '0', which is quite clearly false. I have no idea what to actually make of that message. Perhaps it makes sense to someone intimately familiar with the tests in question, but to me, it makes no sense.

$ prove -v t/run/exit.t
t/run/exit....1..10
ok 1 - STDERR & STDOUT open: normal exit
#     Failed test (t/run/exit.t at line 45)
#     '0'
#         ne
#     '0'
not ok 2 - STDERR & STDOUT open: abnormal exit
ok 3 - STDERR closed: normal exit
#     Failed test (t/run/exit.t at line 45)
#     '0'
#         ne
#     '0'
not ok 4 - STDERR closed: abnormal exit
ok 5 - STDERR & STDOUT closed: normal exit
#     Failed test (t/run/exit.t at line 45)
#     '0'
#         ne
#     '0'
not ok 6 - STDERR & STDOUT closed: abnormal exit
ok 7 - STDOUT closed: normal exit
#     Failed test (t/run/exit.t at line 45)
#     '0'
#         ne
#     '0'
not ok 8 - STDOUT closed: abnormal exit
error:imcc:syntax error, unexpected IDENTIFIER ('Parse')
        in file '/tmp/jIaHID0ADa/test.pir' line 1
ok 9 - parrot returns error on parse failure
#
ok 10 - parrot doesn't create outfile on parse failure
#
# Looks like you failed 4 tests of 10.
dubious
        Test returned status 4 (wstat 1024, 0x400)
DIED. FAILED tests 2, 4, 6, 8
        Failed 4/10 tests, 60.00% okay
Failed Test  Stat Wstat Total Fail  Failed  List of Failed
-------------------------------------------------------------------------------
t/run/exit.t    4  1024    10    4  40.00%  2 4 6 8
Failed 1/1 test scripts, 0.00% okay. 4/10 subtests failed, 60.00% okay.

[whiteknight]

For the record, this error does not occur in OpenSolaris on x86 with GCC. I'm going to retry with Sun's compiler next.

[whiteknight]

As another data point, this error does not appear in OpenSolaris on x86 with SunCC either. Without access to Sparc hardware, I'm not able to test this one any further.

[doughera]

Replying to whiteknight:

As another data point, this error does not appear in OpenSolaris on x86 with SunCC either. Without access to Sparc hardware, I'm not able to test this one any further.

Actually it did occur in OpenSolaris on x86, with both gcc and Sun's cc. That's what I tested on. I suspect you didn't test the same revision.

In any case, something subsequent to r43946 appears to have fixed the problem, as I no longer see that message. (The message is still very confusing, in my opinion, but that's a separate matter.) I'll close the ticket.