test failures in -r runcore

[whiteknight]

Ubuntu 8.10, x86-64.

I'm getting two test failures in the -r runcore:

Failed Test       Stat Wstat Total Fail  List of Failed
-------------------------------------------------------
t/op/pushaction.t    1   256     7    1  6
t/pmc/sub.t          1   256    64    1  63

t/pmc/sub..............................NOK 63/64                             
#   Failed test ':outer with identical sub names'
#   at t/pmc/sub.t line 1457.
# Exited with error code: 1
# Received:
# ABC::outer
# ABC::inner
# Null PMC in say
# current instr.: 'parrot;ABC;inner' pc 47 (/home/andrew/Projects/parrot/t/pmc/sub_63.pir:20)
# called from Sub 'parrot;ABC;outer' pc 37 (/home/andrew/Projects/parrot/t/pmc/sub_63.pir:14)
# called from Sub 'main' pc 9 (/home/andrew/Projects/parrot/t/pmc/sub_63.pir:3)
# 
# Expected:
# ABC::outer
# ABC::inner
# ABC lex
# DEF::outer
# DEF::inner
# DEF lex
# 
t/pmc/sub..............................ok 64/64# Looks like you failed 1 test of 64.
t/pmc/sub..............................dubious                               
	Test returned status 1 (wstat 256, 0x100)
DIED. FAILED test 63
	Failed 1/64 tests, 98.44% okay

t/op/pushaction........................NOK 6/7                               
#   Failed test 'pushaction as closure'
#   at t/op/pushaction.t line 100.
# Exited with error code: 1
# Received:
# main
# at popmark, flag = 0
# a = Null PMC in say
# current instr.: 'exit_handler' pc 32 (/home/andrew/Projects/parrot/t/op/pushaction_6.pir:22)
# called from Sub 'main' pc 15 (/home/andrew/Projects/parrot/t/op/pushaction_6.pir:11)
# 
# Expected:
# main
# at popmark, flag = 0
# a = 42
# 
# Looks like you failed 1 test of 7.
t/op/pushaction........................dubious                               
	Test returned status 1 (wstat 256, 0x100)
DIED. FAILED test 6
	Failed 1/7 tests, 85.71% okay

I assume this all has to do with the opcode problems of x86-64 systems.

[jkeenan]

Replying to whiteknight:

Ubuntu 8.10, x86-64. I'm getting two test failures in the -r runcore: [snip] I assume this all has to do with the opcode problems of x86-64 systems.

Not necessarily. On Darwin/PPC yesterday (reported on mailing list today), I got the same test failures with:

perl t/harness --gc-debug --running-make-test -r --runcore-tests
Test Summary Report
-------------------
t/op/pushaction                    (Wstat: 256 Tests: 7 Failed: 1)
  Failed test:  6
  Non-zero exit status: 1
t/pmc/sub                          (Wstat: 256 Tests: 64 Failed: 1)
  Failed test:  63
  Non-zero exit status: 1
Files=252, Tests=8415, 4681 wallclock secs (10.58 usr  5.50 sys + 597.24 cusr 291.74 csys = 905.06 CPU)
Result: FAIL

kid51

[jkeenan]

See also Allison's report of these two tests failing on Darwin/Intel in TT #419.

[rg]

Hmm I thought lex not working with -r was a known feature. There's a whole set of test skipped because of it, most notably t/op/lexicals.t, which says:

skip_all => 'lexicals not thawed properly from PBC, RT #60652'

I think this is simply a case of missing skips.

[rg]

TODO failing tests

[rg]

If you're agreeing with me that RT #60652 applies, please apply the patch I've just attached. Note that it requires the change from TT #403 to map -r to --run-pbc.

[allison]

Applied in r37350, thanks! Resolving this ticket, but retaining a reference to it in the parent RT #60652.