[perl #37345] [BUG] r9318: imcc label/op interference (only in test harness on win32)
Jerry Gay
# Please include the string: [perl #37345]
# in the subject line of all future correspondence about this issue.
# <URL: https://rt.perl.org/rt3/Ticket/Display.html?id=37345 >
D:\usr\local\parrot-bug\trunk>perl t/harness imcc/t/syn/clash.t
imcc/t/syn/clash....ok 7/15
imcc/t/syn/clash....NOK 8# Failed test (imcc/t/syn/clash.t at line 151)
# got: 'nok
# '
# expected: 'ok
# '
imcc/t/syn/clash....ok 15/15# Looks like you failed 1 test of 15.
imcc/t/syn/clash....dubious
Test returned status 1 (wstat 256, 0x100)
DIED. FAILED test 8
Failed 1/15 tests, 93.33% okay
Failed Test Stat Wstat Total Fail Failed List of Failed
-------------------------------------------------------------------------------
imcc/t/syn/clash.t 1 256 15 1 6.67% 8
Failed 1/1 test scripts, 0.00% okay. 1/15 subtests failed, 93.33% okay.
D:\usr\local\parrot-bug\trunk>parrot -t imcc/t/syn/clash_8.pir
0 branch 2
2 if I0, 17 - I0=0,
5 if I0, 14 - I0=0,
8 inc I0 - I0=0
10 unless I0, 9 - I0=1,
13 unless I0, 6 - I0=1,
16 print "ok\n"
ok
18 end
D:\usr\local\parrot-bug\trunk>
~jerry
Leopold Toetsch
>
> D:\usr\local\parrot-bug\trunk>parrot -t imcc/t/syn/clash_8.pir
> 0 branch 2
> 2 if I0, 17 - I0=0,
I see. The test doesn't initialize I0 before testing it.
I think we decided that Parrot doesn't provide pre-cleared I and N
registers, therefore the test should start with:
null I0
leo
Jerry Gay
test, and look for others that may suffer the same problem. i'll also
add a statement to this effect, if there's not one already.
~jerry