|
Parrot 4.6.0 "Wild Parrots of Telegraph Hill" | parrot.org/ | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC Set by moderator on 18 July 2012. |
|||
|
00:12
dmalcolm joined
00:27
whiteknight joined
|
|||
| whiteknight | good evening, #parrot | 00:34 | |
|
02:21
kid51 joined
|
|||
| dalek | rrot: 938d2c2 | jkeenan++ | t/dynpmc/os.t: [codingstd] No cuddled elses. |
02:49 | |
| rrot: 89ce26c | jkeenan++ | / (8 files): Move Parrot_cpu_type function out of src/platform/misc.c. This way, it doesn't need to be repeated elsewhere. For symmetry, also remove it out of win32/misc.c and give it its own file there as well. Patch supplied by Andy Dougherty++. |
03:05 | ||
|
03:20
bluescreen joined
03:51
ingy joined
04:23
diakopter joined
|
|||
| diakopter | howdy | 04:23 | |
| sorear | hello diakopter. | ||
| diakopter | I was wondering whether m1 can run something like man-or-boy yet | 04:24 | |
| I was reading the parrot-dev archives and saw things mentioned | |||
| also, my $job has 400 quadcore solaris servers we don't know what to do with.. and some could be repurposed as *bsd. I'll ask around to see if I can't steal some public IPs and create a private network for you guys to use | 04:27 | ||
| so whiteknight can debug on them | 04:28 | ||
| anyone know about m1's capabilities? | 04:32 | ||
| benabik has started to wonder if we should port LCC or something. | 04:37 | ||
|
05:11
JimmyZ joined
|
|||
| JimmyZ | aloha, diakopter | 05:11 | |
| diakopter: I think m1 can't, m1 doesn't support pointer due to m0 | 05:13 | ||
|
06:09
fperrad joined
08:20
brrt joined
08:38
brrt left
09:20
Khisanth joined
09:25
brrt joined
09:41
brrt left,
brrt joined
|
|||
| brrt | hey folks, i'm not sure if i should report this as an issue or not | 09:41 | |
| but | |||
| if i instantiate two interpreters | |||
| and then start running code on the first one | |||
| (and those interpreters are not parents and children) | |||
| then i crash, hard | 09:42 | ||
| if they are parent / children | |||
| they do not crash | |||
| this is somewhat annoying as an embedding program must now keep track of at least one 'root' interpreter if more than one is required | |||
| thus, two different modules of a program cannot independently decide if they want an interpreter | 09:43 | ||
|
09:45
PacoAir joined
|
|||
| brrt | oh, i'm so unhappy i forgot my notebook now | 09:46 | |
| anyway | |||
| i'll report this as an issue | |||
| ... thinking about this is hard | 10:04 | ||
| its so much fun | 10:05 | ||
| but so hard | |||
| JimmyZ | :) | 10:07 | |
|
10:30
brrt joined
|
|||
| brrt | haha, scheduler issues once again | 10:32 | |
| brrt cries | |||
|
10:42
Timbus joined
|
|||
| dalek | kudo/inliner: 25d7c81 | jnthn++ | src/Perl6/Optimizer.pm: A further allomorphy tweak, fixing one more test that the inliner regressed. |
10:52 | |
| kudo/inliner: 72075f2 | jnthn++ | src/Perl6/Actions.pm: Don't consider things that use pseudo-packages to be inlinable. |
|||
|
10:55
whiteknight joined
11:01
rurban joined
11:07
bluescreen joined
11:11
JimmyZ joined
|
|||
| dalek | p: 6589656 | jnthn++ | src/QAST/Operations.nqp: Add a way for ops evaluating to native types to attach that information to nodes. |
12:08 | |
|
12:08
brrt joined
|
|||
| dalek | kudo/inliner: 4a658ee | jnthn++ | src/ (4 files): Fix return type check handling when inlining. |
12:09 | |
| kudo/inliner: 57697e2 | jnthn++ | src/Perl6/Actions.pm: Attach return types to nqp::op nodes. |
|||
| brrt | so, i'm here to report i've shot a scheduler-shaped hole in parrot again :-) | 12:10 | |
| ... and | 12:14 | ||
| i'm unhappy about it | |||
|
12:14
kid51 joined
12:17
kid51_ joined
12:50
brrt joined
|
|||
| dalek | kudo/inliner: 3bfb7a6 | jnthn++ | src/ (4 files): Very minimal implementation of the 'soft' pragma, but enough that we'll be able to use it to make wrap.t happy. |
12:58 | |
| kudo/nom: 5cd3686 | jnthn++ | src/Perl6/Ops.pm: Mark some Perl 6 ops as inlinable. |
13:09 | ||
| kudo/nom: 958a3ea | jnthn++ | src/ (2 files): We will keep around the QAST tree of inlinable things now, not some string thingy. |
|||
| kudo/nom: fdcee65 | jnthn++ | src/Perl6/Optimizer.pm: Tweak optimizer to look for inlining information properly (though not actually using it yet). |
|||
| kudo/nom: 44fef37 | jnthn++ | src/Perl6/Actions.pm: Update the inlinability analysis and start attaching inlining information to routines again. |
|||
| kudo/nom: 60117ea | jnthn++ | src/Perl6/Optimizer.pm: Implement inlining, now using nice AST splicing. Seems to basically work; handful of test failures that will need investigation. Also generated code not yet awesome. |
|||
| kudo/nom: 0c46ef0 | jnthn++ | src/Perl6/Optimizer.pm: D'oh, left behind debugging code. |
|||
| kudo/nom: 2659dd9 | jnthn++ | / (6 files): Merge branch 'nom' into inliner |
|||
| kudo/nom: 4c3f769 | jnthn++ | src/Perl6/ (2 files): Elimination of boxable_native, and better handling of literal allomorphy. |
|||
| kudo/nom: 25d7c81 | jnthn++ | src/Perl6/Optimizer.pm: A further allomorphy tweak, fixing one more test that the inliner regressed. |
|||
| kudo/nom: 72075f2 | jnthn++ | src/Perl6/Actions.pm: Don't consider things that use pseudo-packages to be inlinable. |
|||
| kudo/nom: 4a658ee | jnthn++ | src/ (4 files): Fix return type check handling when inlining. |
|||
| kudo/nom: 57697e2 | jnthn++ | src/Perl6/Actions.pm: Attach return types to nqp::op nodes. |
|||
| kudo/nom: 3bfb7a6 | jnthn++ | src/ (4 files): Very minimal implementation of the 'soft' pragma, but enough that we'll be able to use it to make wrap.t happy. |
|||
| kudo/nom: 550b999 | jnthn++ | src/ (9 files): Merge branch 'inliner' into nom |
|||
| brrt | i may have a fix for the interpreter crashing thing | 13:12 | |
| but i'll have to build and test it first | |||
| dalek | kudo/nom: 5f2af91 | jnthn++ | tools/build/NQP_REVISION: Bump NQP_REVISION to get improvements needed by the inliner. |
13:14 | |
| p: e401db5 | jnthn++ | src/6model/serialization.c: Re-enable re-use of strings on the string heap; it makes a fairly notable difference to the output size of CORE.setting compilation, and a small but certainly worthwhile difference to CORE.setting compilation time. Causes no issues here; hopefully not for anyone else now either. |
13:27 | ||
| kudo/nom: 44cb610 | jnthn++ | tools/build/NQP_REVISION: Get string sharing optimization for faster builds. |
13:31 | ||
| kudo/nom: 5e1b9a8 | jnthn++ | docs/ChangeLog: ChangeLog updates. |
|||
|
13:34
kid51_ joined
14:42
brrt joined
15:13
kid51_ joined
15:14
brrt1 joined
|
|||
| whiteknight | good morning, #parrot | 15:18 | |
| kid51_ | yo | 15:20 | |
| whiteknight | hello kid51_ | 15:23 | |
|
15:23
brrt joined
|
|||
| JimmyZ | good morning, whiteknight | 15:24 | |
| whiteknight | hello JimmyZ | ||
| JimmyZ | hi | ||
| brrt | good afternoon #parrot | 15:27 | |
|
15:34
brrt1 joined
16:21
kid51 joined
|
|||
| kid51 wishes his IRC connection was not dropping out on an hourly basis :-( | 16:21 | ||
| dalek | p/toqast: 2637cf1 | jnthn++ | src/NQPQ/World.pm: Create scopes as QAST::Block. |
16:22 | |
| p/toqast: 8ef56a3 | jnthn++ | src/NQPQ/Actions.pm: Update a couple other block creations to use QAST. |
|||
| p/toqast: 18cec82 | jnthn++ | src/NQPQ/ (2 files): Update comp_unit to start using QAST::CompUnit, and tackle some of the other bits of comp_unit updating. |
|||
| p/toqast: d579e84 | jnthn++ | src/NQPQ/World.pm: Update compile-time binding of lexicals. |
|||
| p/toqast: a05823a | jnthn++ | src/NQPQ/Actions.pm: Pull in CTXSAVE override (will update the default one in HLL::Actions after QAST migration). |
|||
| p/toqast: 476dd8e | jnthn++ | src/NQPQ/Actions.pm: Correct :hll copy-pasto. |
16:28 | ||
| p/toqast: 8440713 | jnthn++ | src/NQPQ/Actions.pm: Update argument list processing for QAST. |
|||
| p/toqast: 10f4753 | jnthn++ | src/NQPQ/Actions.pm: Update numeric literal handling. Now with no setting/regex lib loading, nqp::say(42) works. |
|||
|
16:54
tuxit joined
|
|||
| dalek | p/toqast: b9f1e11 | jnthn++ | src/NQPQ/Actions.pm: Add an EXPR that produces QAST. |
17:11 | |
| p/toqast: a5e4e27 | jnthn++ | src/NQPQ/Grammar.pm: Update most of the operators to use :op instead of :pirop/:pasttype. |
17:12 | ||
| p/toqast: c945388 | jnthn++ | src/NQPQ/Actions.pm: Couple of fixes so if/unless now work again. |
|||
| p/toqast: cd9fe1d | jnthn++ | src/QAST/Operations.nqp: Add an nqp::isfalse to go with nqp::istrue; NQP wants it. |
|||
| p/toqast: 375dfd0 | jnthn++ | src/NQPQ/ (2 files): Get setting/module loading to work again. |
|||
| p/toqast: 753dc29 | jnthn++ | src/NQPQ/Actions.pm: Get strings working. With this, we pass a few of the simple tests in t/nqp again. |
|||
|
18:40
whiteknight left
|
|||
| dalek | kudo/nom: 7e1b380 | jnthn++ | src/Perl6/Optimizer.pm: Don't lose line numbers when inlining; moritz++ for reporting. |
19:28 | |
|
19:40
benabik joined
20:04
diakopter left
20:20
awwaiid joined,
benabik joined
|
|||
| dalek | p/toqast: 729ce4d | jnthn++ | src/NQPQ/Actions.pm: Fix lexical variables. |
20:28 | |
| p/toqast: c115641 | jnthn++ | src/NQPQ/Actions.pm: Fix empty blocks. |
|||
| p/toqast: 0758db1 | jnthn++ | src/NQPQ/Grammar.pm: Spell defor correctly. |
|||
| p/toqast: 50bd8cc | jnthn++ | src/NQPQ/Actions.pm: Eliminate many uses of viviself. |
|||
| p/toqast: bfd58e5 | jnthn++ | src/NQPQ/ (2 files): A bunch more PAST::Var => QAST::Var, fixing simple sub declarations along the way. |
|||
| p/toqast: b1a033c | jnthn++ | src/NQPQ/Actions.pm: Eliminate multitype usage. |
20:44 | ||
| p/toqast: 8f9d295 | jnthn++ | src/NQPQ/Actions.pm: Toss hopefully unrequired namespace call. |
|||
| p/toqast: 263e374 | jnthn++ | src/NQPQ/World.pm: s/subid/cuid/ |
|||
| p/toqast: e8e11ea | jnthn++ | src/NQPQ/World.pm: Fix method setup, which gets basic classes working again. |
|||
| p/toqast: 24fb046 | jnthn++ | src/NQPQ/Actions.pm: Somewhat fix self, attribute access. |
20:56 | ||
|
21:05
lucian joined
21:15
benabik joined
21:32
benabik joined
21:50
contingencyplan joined
22:35
benabik joined
23:35
sivoais joined
23:45
TonyC joined
23:54
nopaste joined
|
|||