Parrot 3.0.0 Released | parrot.org | Log: irclog.perlgeek.de/parrot/today | Goals: Fix ipv6-related failures | Test imcc_interfaces and annotations-tree branches
Set by moderator on 31 January 2011.
cotto_work It seems to be pretty close to running. 00:00
whiteknight ok
00:00 benabik is now known as benabik_away 00:03 davidfetter left 00:13 nwellnhof left 00:20 kid51 is now known as kid51_at_dinner
whiteknight imcc 00:23
hate
imcc
hate
imcc
hate
cotto_work decommutes 00:28
dukeleto whiteknight: i would like to say thank you for hacking on IMCC 00:33
whiteknight :)
dukeleto whiteknight: what's up?
whiteknight I'm fixing some IMCC-related stuff in parrot-instrument 00:34
some functions are differnt after my first round of interface cleanups
dukeleto whiteknight: ah
whiteknight here's a question: I have a function that takes a C const char ** and returns a ResizableStringArray. Where should that function live? 00:35
src/pmc.c? src/string/api.c? src/utils.c? 00:36
sorear is it intended to be used only by embedders? 00:37
whiteknight by anybody. Its going to be a general-purpose function
it will be used by the embedding API, but also by extenders
specifically, I also need it in the parrot-instrument project
maybe src/extend.c
actually, there are several other _box_ routines in src/pmc.c 00:39
00:39 contingencyplan joined
dukeleto whiteknight: what does the function do? 00:44
whiteknight: merely converts between those datastructures, or anything else? 00:45
whiteknight it takes a const char** array of strings from C, and wraps them up into a ResizableStringArray
no, that's it
dukeleto whiteknight: hmmm
whiteknight: kind of seems string API-ish
whiteknight it's an encapsulation of some logic used by the embedding API now, but which I want to be able to use for an extension 00:46
00:46 bacek left
dalek rrot: 240d5b7 | Whiteknight++ | src/ (2 files):
Add a new function to execute a PackFile* as a program
 00:49
rrot: 0efb5bd | Whiteknight++ | src/packfile/api.c:
+docs and a TODO note
rrot: 619f4f4 | Whiteknight++ | / (5 files):
pull the guts of Parrot_api_wrap_string_array out into it's own utility function, so we can reuse that logic from elsewhere
rrot/whiteknight/imcc_compreg_pmc: 5eb65ad | Whiteknight++ | / (4 files):
Merge branch 'whiteknight/imcc_compreg_pmc' of github.com:parrot/parrot into whiteknight/imcc_compreg_pmc
ttbot Parrot 619f4f4e i386-linux-thread-multi make error tt.taptinder.org/file/cmdout/5568.txt (tt.taptinder.org/buildstatus/pr-Parrot) 00:51
whiteknight yeah, damnit
dukeleto slightly OT: can someone inform me what the most common ways are to run tests for Python projects? 00:52
i am finding this hard to search for
whiteknight step 1: Identify the code you want to test 00:53
step 2: Mash your face into the keyboard
dukeleto basically, which are the most common ways to run tests for python projects?
whiteknight: :)
whiteknight step3: ....
step 4: profit!
dukeleto whiteknight: i am talking about : what is the equivalent to "./Build test" or "make test" in the python world
dalek rrot: e2dbb1a | Whiteknight++ | src/p (2 files):
oops fixes.
rrot-instrument: 60a1f4a | cotto++ | src/dynpmc/instrument.pmc:
update some more function calls; still won't build
 00:56
rrot-instrument: 03eb7dd | cotto++ | src/dynpmc/instrument.pmc:
make some code less wrong, though still wrong
rrot-instrument: bdce676 | Whiteknight++ | src/dynpmc/instrument.pmc:
fix IMCC invocation using new extending functions
whiteknight msg cotto: I fixed up the IMCC invocation stuff. Requires newest Parrot. Build fails now building instrumentop.pmc. I'm looking into it, but outside my area of expertise 00:57
aloha OK. I'll deliver the message.
00:58 bacek joined
whiteknight plobsing: ping 00:58
01:00 kid51_at_dinner is now known as kid51
whiteknight actually, nevemrind 01:02
kid51: ready whenever
cotto ~~ 01:21
whiteknight, awesome 01:22
axesome, even
good to see that kid51 caught you
01:30 contingencyplan left
dalek nxed: r786 | NotFound++ | trunk/examples/Xlib.winxed:
update example Xlib to more recent idioms and capabilities
 01:43
01:55 kid51_ joined 01:58 kid51 left 02:06 ligne joined
whiteknight who stole my Parrot_pcc_get_constants function? 02:08
and by stole, I mean deleted. And by "my", I mean Parrot's
plobsing dead code tells no tales 02:14
ligne git tells all
whiteknight Ive got one more build failure in parrot-instrument involving that function
of course, it could just be hiding more failures in later files 02:15
plobsing: Can you help me translate this code to something newer:
key = (Parrot_pcc_get_constants(attr->interp, cc)[arg]).u.key;
The .u.key Part can go away, I know
attr->interp->code->const_table->pmc.constants[arg]? 02:16
plobsing Parrot_pcc_get_pmc_constant 02:17
constant type is strongly enforced now
no more tagged union where we don't even check the tag half the time 02:18
cotto brave new world 02:30
whiteknight cotto: parrot-instrument is building 02:31
fails tests miserably, but builds 02:32
I'll push in a minute, I'm exploring some more problems 02:35
cotto thanks! 02:48
that's definite progress
dalek rrot-instrument: 114fe27 | Whiteknight++ | src/dynpmc/ (2 files):
necessary changes so that parrot-instrument builds. All tests fail, though the failures look trivial
 02:51
rrot-instrument: 7f481cf | Whiteknight++ | t/instrument (7 files):
update test files to shebang installed parrot, not a local one
rrot-instrument: ea9473a | Whiteknight++ | / (2 files):
fix the library search path in one of the tests. Added diagnostics information if we cannot find the dynpmclib. At the moment, the tests cannot find it.
whiteknight you're going to have to look at it, because these failures are driving me crazy
it can't find the .so file
t/instrument.t is the only test file that I've updaed the paths for
cotto ok 02:52
I'm up for a tag team
kid51_ I'm just about ready to eliminate a 1163-line file! 02:53
02:53 kid51_ is now known as kid51
whiteknight kid51++ 02:53
aka kid52
kid51 is back where he started in Parrot 4 years ago: pmc2c.pl
cotto I see some warnings that probably indicate something suspicious.
kid51 cautions that he's not eliminating 1163 lines of code :-) 02:54
whiteknight forget zee warnings. Full speed ahead!
kid51 has discovered it can be moved into another file
cotto um, no
/usr/local/parrot-git/include/3.0.0-devel/parrot/api.h:757: note: expected ā€˜Parrot_PMCā€™ but argument is of type ā€˜Parrot_Interpā€™
;)
I'm pretty sure that's not going to work. 02:55
plobsing rethrow weirdness: "exception type: Exception - Not a throwable object"
exception isn't throwable?
whiteknight cotto: What's that on, the imcc_run_api function?
cotto yup 02:56
whiteknight cotto: which file is that in? 02:57
cotto instrument.pmc
I'll probably have to take off soon, but I'm looking forward to working on this (if there's anything left to do). 02:58
whiteknight fudge
cotto sure, thanks
whiteknight let me looksie 02:59
dalek rrot-instrument: 6f65c85 | Whiteknight++ | src/ (2 files):
fix bad parameter type error when calling imcc_run_api
 03:01
whiteknight still can't load instrument_group 03:03
this is starting to become infuriating 03:04
cotto I'll have to come back later tonight. 03:05
whiteknight Parrot_dlopen is returning null 03:10
blah. I don't have the tolerance for these kinds of bugs 03:12
I'm staring at the file. I can *see* that it is in the correct place. GDB is telling me that the path is correct.
so....fuggedaboudit
goodnight
03:13 whiteknight left
dalek Heuristic branch merge: pushed 84 commits to parrot/tt1988_pmcemitter by jkeenan 03:16
rrot/tt1988_pmcemitter: 39a55b7 | jkeenan++ | lib/Parrot/Pmc2c/PMC.pm:
Eliminate commented-out code.
 03:22
03:34 Andy joined
dalek rrot/tt1988_pmcemitter: 786b906 | jkeenan++ | lib/Parrot/Pmc2c/UtilFunctions.pm:
Remove ltrim() and rtrim() -- never used anywhere.
 04:25
04:25 kid51 left 04:31 contingencyplan joined
dalek m-eta-wink-kzd: 2c31120 | plobsing++ | t/harness:
output test filename from harness to show progress
 04:48
m-eta-wink-kzd: a2828dd | plobsing++ | / (5 files):
use rethrow where appropriate

rethrow maintains the original throw location, allowing more insight for debugging.
  OMeta.fail must be wrapped as the exception payload because exceptions cannot be safely re-used.
m-eta-wink-kzd: 420cc37 | plobsing++ | t/tutorial.Ī©Ī·:
update test for new exception convention
04:57 theory left 05:43 Andy left 05:58 rurban_ joined 06:01 rurban left, rurban_ is now known as rurban 07:25 fperrad joined 08:07 particle left 08:11 jsut joined 08:16 jsut_ left 09:07 ligne left 10:37 contingencyplan left
dalek umage: 696955d | fperrad++ | plumage/metadata/winxed.json:
[winxed] add instruction for clean & test
 10:43
umage: 0bb1d58 | fperrad++ | src/lib/Plumage/Project.nqp:
add uninstall_make,

see github.com/parrot/plumage/issues#issue/4
umage: 4e6dc61 | fperrad++ | plumage/metadata/ohm-eta.json:
[ohm-eta] add instruction for uninstall
11:13 ligne joined
dalek nxed: r787 | NotFound++ | trunk/winxedst1.winxed:
some more special cases for null
 11:43
nxed: r788 | NotFound++ | trunk/examples/fly.winxed:
delete a debug helper statement commited by mistake
12:01 kid51 joined 12:11 lucian joined 12:28 SingAlong joined
dalek rrot/tt1988_pmcemitter: 0624a16 | jkeenan++ | / (2 files):
Parrot::Pmc2c::PMC::ParrotClass and Parrot::Pmc2c::ComposedMethod were never used. Remove them.
 12:29
12:34 ligne left 12:53 SingAlong left 12:54 Coke left
dalek rrot: 84d22ce | mikehh++ | src/p (2 files):
fix pod syntax and fix to conform to 80 char line limit for t/codingstd/c_function_docs.t
 12:57
12:59 akashmanohar joined
dalek p-rx/nom: 6a63e74 | jonathan++ | src/metamodel/ (2 files):
Support .^name on KnowHOWs.
 12:59
p-rx/nom: 0b5765d | jonathan++ | src/NQP/ (2 files):
Stop treating meta-object names as strings, and actually reference the meta-objects, just as we do with Actions classes.
nxed: r789 | NotFound++ | trunk/winxedst1.winxed:
better handling of predef arguments
 13:00
13:02 whiteknight joined 13:06 Coke joined 13:07 lucian_ joined
whiteknight good morning, #parrot 13:07
akashmanohar is parrot going to be one of the organisations for GSoC 2011?
whiteknight I hope so 13:09
akashmanohar: why, you interested in becoming a student?
akashmanohar whiteknight: ya
whiteknight awesome. What kinds of projects are you interested in?
akashmanohar whiteknight: I've not dipped my feet into compilers or VMs. But my final year project which I'm doing currently is a distributed language (written in ruby). 13:10
13:11 lucian left
whiteknight akashmanohar: that's cool. Compilers can be a little tricky because of all the theory you need to understand. But VMs aren't too special 13:11
akashmanohar: what languages do you know? Ruby obviously 13:12
akashmanohar whiteknight: did some C, python and scheme.
I do webdev. So ruby, php have been the languages I've used the most. 13:13
whiteknight okay
akashmanohar I would learn anything else if it's required tho.
whiteknight We have a Ruby compiler called Cardinal, but it isn't too active right now
akashmanohar whiteknight: sw33t. Is that written in PCT or PIR? 13:14
kid51 whiteknight: Generally true, but I've seen some commits to cardinal in the past few days
whiteknight github.com/parrot/cardinal
The parser is written in NQP, but it looks like most of the runtime is written in PIR 13:15
Parrot's object model was deemed to be insufficient to finish the Ruby runtime
akashmanohar any parrot libraries ideas (I read that parrot libraries can be shared cross-language)
whiteknight akashmanohar: We need lots of libraries. There are relatively few right now 13:16
akashmanohar Or is there any page for suggested contributions to parrot? (like some sort of TODO page). That might give ideas
whiteknight: oh thats nice. would be happy to give them a shot. 13:17
whiteknight We have a list of big project ideas, though it's a little out of date: trac.parrot.org/parrot/wiki/BigProjectIdeas
akashmanohar whiteknight: anything I need to go thru after the PCT tutorial to be able to write cross-language libraries?
whiteknight no
What kinds of projects interest you, off the top of your head? 13:20
akashmanohar whiteknight: writing libraries. Right now the memcached interface sounds interesting to start off with. Found that on the wiki page 13:21
whiteknight Yeah, a memcached interface library would be nice
it's not really a "big" project. Could probably be only a few hundred lines or PIR, less than that if you used a higher language like NQP
but it would have big impact, I think 13:22
akashmanohar whiteknight: ya but sounds like a nice start to me since I have no experience with parrot.
whiteknight yes, that's very true. Do you have prior experience with memcached?
akashmanohar whiteknight: ya used it with my web apps on appengine.
whiteknight oh nice
akashmanohar whiteknight: they have this memcache API for web apps that they expose thru python (havent tried the java version of appengine). 13:23
whiteknight ok
akashmanohar whiteknight: I'll finish the PCT, a small library and then get back to you on my progress. 13:24
*pct tutorial
whiteknight that would be awesome.
We haven't started fielding too many GSoC project ideas yet. So we are probably going to start putting together a list of things we would like to have
mikehh t/src/embed/api.t - Failed test: 2 13:25
whiteknight that won't be a definitive list, but should give you plenty of ideas
akashmanohar whiteknight: ya. probably thats would be the place for others to start too.
probably title it "Ideas for contributions" to generalize it.
whiteknight yeah 13:26
We haven't been officially accepted into GSoC this year yet, so I need to make sure that happens first before putting together a big list 13:27
I suspect we will
akashmanohar whiteknight: well even if not I would be interested in contributing whatever's in my capicity.
whiteknight that's awesome to hear
If you have any questions while you work, don't hesitate to ask us 13:28
akashmanohar sure. thanks!
whiteknight we're very helpful, and we don't bite
akashmanohar haha :)
nopaste "mikehh" at 192.168.1.3 pasted "t/src/embed/api.t - Failed test: 2" (36 lines) at nopaste.snit.ch/30631 13:32
13:51 nwellnhof joined 14:00 rurban_ joined 14:01 kid51 left 14:03 rurban left, rurban_ is now known as rurban
dalek rrot: 4ef11b1 | fperrad++ | t/pmc/exception.t:
[t] use diag() which produces correct TAP comment
 14:05
rrot: 3718798 | fperrad++ | t/harness.pir:
[harness.pir] add missing files
14:11 kid51 joined 14:27 lucian_ left
dalek rrot: 2310551 | nwellnhof++ | src/embed/api.c:
Fix assertion failure
 14:28
p-rx/nom: 594f186 | jonathan++ | / (3 files):
Start to stub out role meta-objects. All roles are implicitly parametric because a mention of $?CLASS is generic. The base here could be built up to do parametric roles in NQP, but that's not on the cards at the momnet; of course, Rakudo's meta-objects will carry a full implementation of those.
 14:30
14:31 lucian joined
pmichaud good morning, #parrot 14:48
dalek nxed: r790 | NotFound++ | trunk/examples/fly.winxed:
some improvements and coding style updates in example fly
 14:50
pmichaud This morning I woke up with a question: TT #451 says that the packfile structure is 'experimental'. Does that mean that the packfile format is unstable -- i.e., it doesn't follow the deprecation policy?
More to the point, if a compiler writes creates its own packfiles instead of using imcc to do it, are those guaranteed to work across supported releases?
s/writes// 14:51
14:52 NotFound_b joined
dalek rrot: fd01a85 | nwellnhof++ | t/pmc/ (2 files):
[t] Set AI_PASSIVE on server sockets
 14:55
15:00 baest left
nwellnhof pmichaud: the packfile format changes pretty often. afaik, there are no provisions to support multiple packfile versions in a single release. 15:00
pmichaud Any idea when it might stabilize? If we're all being told to move away from PIR/IMCC, it would be good to have something stable as a replacement. 15:01
nwellnhof no idea. 15:03
pmichaud I'll post a message to parrot-dev 15:05
mikehh nwellnhof: thanks for fixin' that, I was tied up and didn't have time to investigate 15:17
nwellnhof mikehh: no problem, i got bitten by it, too 15:18
whiteknight the bytecode format itself is unstable, but the API to create and read them in will become stable 15:30
the actual file format is really an internal implementation detail
pmichaud whiteknight: maybe an answer to parrot-dev, then 15:31
whiteknight ok
pmichaud is there an API to create and read them, ooc?
I'm only aware of IMCC.
dalek nxed: r791 | NotFound++ | trunk/examples/fly.winxed:
improve maniobrability in example fly
 15:45
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#7609) fulltest) at 3_0_0-468-gfd01a85 - Ubuntu 10.10 i386 (g++-4.5) 16:06
16:12 ambs joined 16:16 Patterner left, Psyche^ joined, Psyche^ is now known as Patterner
whiteknight there is currently no API for it. Which is why we haven't pulled the plug on IMCC/PIR yet. We are going to have a good, stable alternative in place for at least one supported release before we take anything away 16:20
that's the policy, and there's no reason to go any faster in this case
the single most important thing is properly encapsulating IMCC to make Parrot more language-agnostic 16:26
16:27 dngor left
dalek rrot/nwellnhof/unicode_dynpmcs: 9184def | nwellnhof++ | / (4 files):
Implement Unicode can_read, can_write, can_execute
 16:27
rrot/nwellnhof/unicode_dynpmcs: 2f1552d | nwellnhof++ | / (4 files):
Implement Unicode rename
rrot/nwellnhof/unicode_dynpmcs: 3df82d6 | nwellnhof++ | / (4 files):
Implement Unicode chmod and chroot

OS dynpmc is now free of platform-specific code
rrot/nwellnhof/unicode_dynpmcs: 026e280 | nwellnhof++ | / (2 files):
Win32 fixes
rrot/nwellnhof/unicode_dynpmcs: d8ebfbc | nwellnhof++ | t/dynpmc/os.t:
[t] Remove remaining tests for system-specific error messages
rrot/nwellnhof/unicode_dynpmcs: 702a4b5 | nwellnhof++ | src/dynpmc/file.pmc:
Remove some platform-specific code frpm File dynpmc
 16:34
sorear whiteknight: Packfile PMCs aren't a supported API then? 16:35
whiteknight sorear: not yet. They need serious work
what we will likely do is offer something like those in a compiler-builder library
16:40 dngor joined 16:41 akashmanohar left 16:43 nwellnhof_ joined 16:47 nwellnhof left, nwellnhof_ is now known as nwellnhof 16:50 theory joined
dalek nxed: r792 | NotFound++ | trunk/winxedst1.winxed:
new arguments lacked optimization phase in stage 1, fixed
 16:51
pmichaud if someone is looking for a task that would be immediately useful, TT #1994 needs some love 17:17
afk, lunch & errands 17:18
17:21 benabik_away is now known as benabik
kid51 ~~ 17:22
benabik What is ~~ for? Seen several people doing it. 17:23
kid51 waves (hello) 17:24
sorear I am now online
plobsing message bots, please nag me 17:25
kid51 Also, if anyone has stored private messages for me via aloha, that will cause aloha to burp
benabik A multi-purpose short-cut. I approve. :-D
17:32 theory left, contingencyplan joined
benabik What's the default GC? ms or ms2? (Or something else entire?) 17:36
17:38 kid51 left
dalek rrot: 5755f0f | bacek++ | / (2 files):
Add statistic functions to Pointer_Array
 17:39
rrot: 425b544 | bacek++ | src/gc/gc_ms2.c:
Return more stats from GC MS2. Closes of #1994.

There is a lie, big lie and statistics. We are lying about TOTAL_PMCS by upper bounding it to C<objects> allocated capacity.
rrot: d740315 | bacek++ | t/op/gc.t:
Add test for ACTIVE_PMCS and TOTAL_PMCS in GC.
benabik Hah. In the time it took me to figure out the problem, bacek++ solved it. 17:40
bacek benabik, I wrote this code :) 17:41
Pointer_Array and MS2 17:42
benabik I suppose that might give you a headstart on understanding it. :-D
dalek Heuristic branch merge: pushed 25 commits to parrot/generational_gc by bacek 17:50
TT #1994 closed by bacek++: interpinfo TOTAL_PMCS and ACTIVE_PMCS no longer work in 3.0.0 17:51
TT #1994: trac.parrot.org/parrot/ticket/1994
18:11 theory joined 18:12 theory_ joined, theory left, theory_ is now known as theory 18:27 ambs left 18:28 ambs joined
dalek p-rx/nom: 8f6ca5a | jonathan++ | / (2 files):
Add an NQPModuleHOW.
 18:44
p-rx/nom: ae3ec52 | jonathan++ | src/NQP/ (2 files):
A little cleanup and consistency improvement in the actions to eliminate some bits left over from the 6model transition, and in prep for roles additions.
18:49 ambs left 19:07 theory left, theory joined
dalek nxed: r793 | NotFound++ | trunk/winxedst1.winxed:
refactor common part of two WinxedHLL compile methods and add new target type
 19:25
19:43 TonyC left, nopaste left 19:44 theory left
cotto I love seeing gsoc_instrument being resuscitated. 20:14
whiteknight who's resuscitating it? 20:18
cotto heh 20:25
whiteknight I gave up last night after those library problems
cotto you made some progress though
whiteknight yeah. it builds. I just can't test it 20:26
cotto I
'm playing with it now.
20:29 fperrad left
cotto I'm really glad I asked khairul to put his build scripts in the branch. 20:30
20:30 benabik is now known as benabik_away 20:44 vmspb joined
cotto whiteknight, ping 20:46
whiteknight pong
cotto whiteknight, gsoc_instrument needs to know some internal details about the gc, specifically the GC_Subsystem struct. What's the best way to get it that info?
whiteknight cotto: telepathy 20:47
cotto non-fictitious techniques would be preferable
it currently wants access to gc_private.h 20:48
that worked fine when it lived in trunk (now master), but not so much now
whiteknight it had a local copy of gc_private.h in there 20:51
src/dynext/
cotto so it does
yuck
it "works", but yuck
whiteknight I removed some stuff out of there that was causing problems
what works? 20:52
cotto I mean that having a copy of gc_private.h works
21:17 sheant joined 21:26 theory joined
dalek TT #1997 created by ulyQSJjKHC0++: The way to obtain Qualified Business Cards Printed In the greatest ... 21:27
TT #1997: trac.parrot.org/parrot/ticket/1997
plobsing anyone know how to baleet spam tickets? 21:32
whiteknight it's an admin function, I think 21:33
cotto I can do it.
boom 21:34
plobsing that's the oddest spam I've ever seen. no mention of where one should buy the product, just why one would want such a product. 21:36
cotto catchy username though 21:37
21:48 preflex left 21:56 NotFound_b left
dalek nxed: r794 | NotFound++ | trunk/examples/fly.winxed:
visual improvement in example fly
 22:00
TT #1998 created by aOcgKMmPwXeCjZ0++: Information about how to obtain Successful Business Cards Personalised In ...
TT #1998: trac.parrot.org/parrot/ticket/1998
22:01 rurban_ joined
dalek p-rx/nom: 50264f9 | jonathan++ | src/ (4 files):
Implement does so roles can be composed. Based on the meta-object, do code generation a little differently for bodies of parametric types. Have NQPClassHOW instantiate the role with the correct type object and add it and the generic version to its roles done list. .^does works as a result. Also update type check cache publisher to know about roles. Missing: all the rest of composition.
 22:01
22:03 rurban left, rurban_ is now known as rurban
plobsing nopaste? 22:12
cotto whiteknight, ping 22:13
whiteknight, unping
aloha, nopaste? 22:14
aloha cotto: nopaste is is nopaste.snit.ch (works with the script in $_PARROT/tools/dev/nopaste.pl)
plobsing gist.github.com/812857 < primitive lisp implemented in Ī©Ī· 22:16
22:24 nwellnhof left
dalek p-rx/nom: ea83082 | jonathan++ | src/metamodel/how/NQP (2 files):
.^add_method, .^add_multi_method, .^attributes, .^add_parent and (todo) .^add_role for parametric and concrete role meta-objects. Should mean we're just a composer away from working roles in NQP.
 22:30
nxed: r795 | NotFound++ | trunk/winxedst1.winxed:
quick fix to operator * in stage 1
 22:33
rrot-instrument: 1ffa1a7 | cotto++ | tools/gen_ (2 files):
copy some helpful scripts from the parrot.git branch
 22:38
rrot-instrument: 0042dfc | cotto++ | src/dynpmc/instrumentclass.pmc:
remove some obsolete VTABLE slots from InstrumentClass
rrot-instrument: 5af485b | cotto++ | src/dynpmc/instrument.pmc:
add a cast to get rid of a compiler warning
rrot-instrument: e691fac | cotto++ | src/dynpmc/instrumentop.pmc:
update function name
rrot-instrument: 89ee2bc | cotto++ | src/dynpmc/instrumentruncore.pmc:
another function name update
cotto It builds without warnings and segfaults when instantiating an Instrument. I love progress. 22:42
NotFound Looks like nopaste is down? 22:58
plobsing yeah, had to use gist in stead
23:00 preflex joined
NotFound Playing with keys. Is this supported? gist.github.com/812897 23:01
plobsing not sure. my opinion is it should not be. key access like that makes register allocation near-impossible. 23:04
cotto That'd require imcc to pick the k variant at runtime. 23:05
NotFound Then how k, not kc, can be used?
cotto well, potentially
either that or get smarter
NotFound If k cannot/should not be used we are generating a lot of unuseful opcode variants. 23:08
23:11 TonyC joined, nopaste joined
NotFound Looks like [$P0] generates code using $P0 as key, not a key containing $P0 23:26
PIR always amazes me.
plobsing: yes, looks like registar allocation is near-impossible because of set $P0, some_key 23:32
sorear keys haven't been abolished yet? 23:36
whiteknight cotto: you can instantiate an Instrument? that is good progress
sorear: no, but thanks for pointing out our lack of progress in that area 23:37
cotto whiteknight, I can until it segfaults 23:38
whiteknight whatever. Segfaults are fine. I get them all the time in my imcc branch
cotto ships it 23:39
whiteknight you debugging this, or you want me to? 23:42
cotto I'm digging in, but I won't refuse a second pair of eyes. 23:43
whiteknight supervised->code is not being set to anything 23:46
it's null
that's probably a result of the IMCC work
got it 23:49
dalek rrot-instrument: 7331328 | Whiteknight++ | src/ (2 files):
fix a segfault, the ->code member isn't set automatically
 23:50
whiteknight yay. t/instrument.t passes again
cotto nice
whiteknight++ 23:51
now I get to dig into distutils to figure out how to merge pbc files 23:55
whiteknight slowly...
dalek rrot-instrument: 0c042a7 | Whiteknight++ | t/ (7 files):
add in a new directory that is needed for temporary files in the tests
23:55 sheant left
cotto at least it's documented 23:58
fperrad++ for that (and fperrad++ for writing it in the first place)
whiteknight fperrad++ for a lot of things
cotto I hate the bus number, but it's worked out surprisingly well so far. 23:59

Powered by Cro and the RakuĀ® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!