Parrot 6.1.0 "Black-collared Lovebird" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 18 February 2014.
00:35 kid51 joined
kid51 paste 00:46
'make' failed to complete in Parrot master: 00:56
thenceforward.net/parrot/771ca093a2...ld_failure
01:17 kid51 joined 01:30 Jimmy_ joined
Jimmy_ rurban: you may be interested with gist.github.com/Whiteknight/1198521 01:31
01:57 benabik joined 02:20 FROGGS_ joined 02:22 kid51 joined 06:28 awwaiid joined 06:36 noize joined 08:25 FROGGS joined 09:07 Chirag joined 09:39 denisboyun joined 10:24 noize joined 10:28 Chirag joined 13:00 bluescreen joined
dalek rrot: 5b3e6c0 | rurban++ | / (7 files):
[cage] fix --debugging and some of the latest SHIM/UNUSED decl

make headerizer and --debugging tested. TODO: UNUSED should be switched to SHIM in the decl.
13:10
13:26 travis-ci joined
travis-ci [travis-ci] parrot/parrot#980 (master - 5b3e6c0 : Reini Urban): The build failed. 13:26
[travis-ci] Change view : github.com/parrot/parrot/compare/7...3e6c054493
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20282185
13:26 travis-ci left
dalek rrot: c36a6d5 | rurban++ | config/gen/makefiles/root.in:
Makefile: fix testOx comments, replace testOp with testOc
14:34
rrot: 84807d1 | rurban++ | / (2 files):
fix clang optimizer increment issue GH #774

have to use volatile scratch registers in Parrot_Integer_increment_orig to avoid skipping overflow sementics: -1 => MAXINT fixes t/pmc/bigint.t test negate_min_integer
14:39
14:50 travis-ci joined
travis-ci [travis-ci] parrot/parrot#981 (master - c36a6d5 : Reini Urban): The build is still failing. 14:50
[travis-ci] Change view : github.com/parrot/parrot/compare/5...6a6d551fc2
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20287138
14:50 travis-ci left
FROGGS rurban: it is interesting to read your post about Parrot's JIT in contrast to this: lists.parrot.org/pipermail/parrot-d...07351.html 15:03
it is like an awesome artwork for one and not even worth mentioning for somebody else
that is pretty weird
rurban Yes, I was pretty angry then, Also when lobsinger did more damage 15:04
But the same happened in 2002 also. When Greenblatt came in and fixed their jit to be able call general ops. Even the old devs had no idea. 15:06
15:06 travis-ci joined
travis-ci [travis-ci] parrot/parrot#982 (master - 84807d1 : Reini Urban): The build was fixed. 15:06
[travis-ci] Change view : github.com/parrot/parrot/compare/c...807d1b7442
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20287339
15:06 travis-ci left
rurban "Implement all 1200 opcodes and their variants" is pure nonsense 15:07
5-10 would be enough
parrot and perl5 seems to have attracted pretty damaging developers over the years. But I blame the managers. They should have avoided this. Also the nci damage (e.g. removal of t for cstrings) 15:09
FROGGS that is what I like about rakudo/moarvm, the community is pretty healthy and the decisions which are made do make sense 15:11
but what can I say, I am here for just about 1.5years now :o)
15:12 FROGGS2 joined
rurban parrot and perl5 are much older, and the big difference is that the original devs are still leading 15:14
FROGGS2 hmmm, my impression is that you are the only dev nowadays 15:29
rurban looks like so. a good situation
But so far I still have to catch up undoing the damage done. I couldn't implement new features yet 15:30
signals, aio, semaphores, ...
6model
non-icu unicode from moarvm would also be nice to have 15:32
but I would prefer to have unicode as shared library, not builtin
15:33 FROGGS joined
dalek rrot: 10639a7 | rurban++ | / (6 files):
[cage] Revamp GH #957 pull request

Throw unknown key type exceptions in default case Remove some compiler warnings (tested even with clang++) Added even the strcmp case exception even if it makes not much sense. Harmonize "Wrong type of value from annotation" exception messages. Most work done by Paul Cochrane
In Makefile moved touch up to align the gnu replaceble testutils so that I can delete the superfluous slow $(PERL) -MExtUtils::Command -e variants manually more easily.
15:37
rrot/smoke-me/cached_findcclass-gh1027: e8e1377 | (Timo Paulssen)++ | src/string/encoding/shared.c:
cache iterators in encoding_find_*cclass

encoding_find_cclass and encoding_find_notcclass can now cache an iterator between calls, because there is at least one usage of the pattern "scan a whole string for newlines" in rakudo.
for utf8 files, like Actions.nqp and Grammar.nqp, it used to take 5s and 2s respectively, now takes 2.4s and 0.9s respectively after the patch.
  Benabik:
My only concern is that there's a chance of false positive on the cache
find_cclass called string GC'd new string allocated at same address find_cclass called
This seems unlikely, but heisenbug paths like this are really hard to track down if tripped over. We could add the cached iterator directly to the string, adding two words per string that we can try to reuse any time STRING_iter_skip is called. (Probably involves an API change: STRING_ITER_AT instead of STRING_ITER.)
And now that I think about it, caches can be very problematic in threaded environments.
See github.com/parrot/parrot/pull/1027
15:48
15:56 travis-ci joined
travis-ci [travis-ci] parrot/parrot#983 (master - 10639a7 : Reini Urban): The build passed. 15:56
[travis-ci] Change view : github.com/parrot/parrot/compare/8...639a7d0f4c
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20291297
15:56 travis-ci left 16:12 travis-ci joined
travis-ci [travis-ci] parrot/parrot#984 (smoke-me/cached_findcclass-gh1027 - e8e1377 : Timo Paulssen): The build failed. 16:12
[travis-ci] Change view : github.com/parrot/parrot/commit/e8e1377baba4
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20292126
16:12 travis-ci left
dalek rrot/smoke-me/cached_findcclass-gh1027: a60a9ec | (Timo Paulssen)++ | src/string/encoding/shared.c:
cache iterators in encoding_find_*cclass

encoding_find_cclass and encoding_find_notcclass can now cache an iterator between calls, because there is at least one usage of the pattern "scan a whole string for newlines" in rakudo.
for utf8 files, like Actions.nqp and Grammar.nqp, it used to take 5s and 2s respectively, now takes 2.4s and 0.9s respectively after the patch.
  Benabik:
My only concern is that there's a chance of false positive on the cache
find_cclass called string GC'd new string allocated at same address find_cclass called
This seems unlikely, but heisenbug paths like this are really hard to track down if tripped over. We could add the cached iterator directly to the string, adding two words per string that we can try to reuse any time STRING_iter_skip is called. (Probably involves an API change: STRING_ITER_AT instead of STRING_ITER.)
And now that I think about it, caches can be very problematic in threaded environments.
See github.com/parrot/parrot/pull/1027
16:25
rrot: 2713630 | (Timo Paulssen)++ | / (2 files):
cache iterators in encoding_find_*cclass

encoding_find_cclass and encoding_find_notcclass can now cache an iterator between calls, because there is at least one usage of the pattern "scan a whole string for newlines" in rakudo.
for utf8 files, like Actions.nqp and Grammar.nqp, it used to take 5s and 2s respectively, now takes 2.4s and 0.9s respectively after the patch.
  Benabik:
My only concern is that there's a chance of false positive on the cache
find_cclass called string GC'd new string allocated at same address find_cclass called
This seems unlikely, but heisenbug paths like this are really hard to track down if tripped over. We could add the cached iterator directly to the string, adding two words per string that we can try to reuse any time STRING_iter_skip is called. (Probably involves an API change: STRING_ITER_AT instead of STRING_ITER.)
And now that I think about it, caches can be very problematic in threaded environments.
See github.com/parrot/parrot/pull/1027
16:29
16:39 travis-ci joined
travis-ci [travis-ci] parrot/parrot#985 (smoke-me/cached_findcclass-gh1027 - a60a9ec : Timo Paulssen): The build has errored. 16:39
[travis-ci] Change view : github.com/parrot/parrot/compare/e...0a9ec80b4e
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20294653
16:39 travis-ci left 16:59 travis-ci joined
travis-ci [travis-ci] parrot/parrot#986 (master - 2713630 : Timo Paulssen): The build passed. 16:59
[travis-ci] Change view : github.com/parrot/parrot/compare/1...13630a9b41
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20294898
16:59 travis-ci left 17:27 benabik joined 17:54 Chirag joined 18:59 Chirag joined
dalek rrot/rurban/SHIM-gh1052: f744a20 | rurban++ | / (42 files):
[cage] change UNUSED(arg) to SHIM in the decl #1052

UNUSED just adds code to please the compiler unused warnings, but we should really declare the args as such with SHIM() We need UNUSED only in .pmc declaration because we have no control over SHIM, the pmc2c compiler does it for us there.
Exceptions: io vtable methods keep the 2nd handle arg. The C abi defines caller cleanup, so we could declare some of them with empty args, but for the sake of API sanity we accept at least 2 args.
Also do not return unused values, rather cast to (void) and strip PARROT_WARN_UNUSED_RESULT attributes on them.
20:09
rrot/rurban/SHIM-gh1052: 7d65604 | rurban++ | / (42 files):
[cage] change UNUSED(arg) to SHIM in the decl #1052

UNUSED just adds code to please the compiler unused warnings, but we should really declare the args as such with SHIM() We need UNUSED only in .pmc declaration because we have no control over SHIM, the pmc2c compiler does it for us there.
Also do not return unused values, rather cast to (void) and strip PARROT_WARN_UNUSED_RESULT attributes on them.
UNUSED Exceptions: io vtable methods keep the 2nd handle arg. The C abi defines caller cleanup, so we could declare some of them with empty args, but for the sake of API sanity we accept at least 2 args.
Headerizer does not grok #ifdef (yet), so we need to UNUSED some args.
20:12
rrot: d12ae0b | rurban++ | src/runcore/cores.c:
[docs] pod for the trace runcore
21:11
rrot: 9316fd9 | rurban++ | src/ (8 files):
[cage] annotate cachegrind results

clang--3-4++ =O3: valgrind --tool=cachegrind -- ./parrot -O2 examples/benchmarks/oo1.pir qcachegrind cachegrind.out.*
also fix a c++ cast error from latest src/string/encoding/shared.c patch
rurban done for today 21:13
21:27 travis-ci joined
travis-ci [travis-ci] parrot/parrot#987 (master - 9316fd9 : Reini Urban): The build passed. 21:27
[travis-ci] Change view : github.com/parrot/parrot/compare/2...16fd9749d6
[travis-ci] Build details : travis-ci.org/parrot/parrot/builds/20313940
21:27 travis-ci left 22:41 davidfetter joined 23:06 Psyche^ joined