github.com/moarvm/moarvm | IRC logs at irclog.perlgeek.de/moarvm/today
Set by moderator on 7 June 2013.
00:53 jlaire joined 01:12 benabik joined, arnsholt joined
benabik is looking through validation.c at diakopter's suggestion. 01:14
diakopter don't be too harsh.. that's ... early C for me. 01:16
.oO( as if even the recent stuff isn't early too... )
 01:17
benabik TODO says "validate any cu->strings index". It seems like it already does that? l. 130 01:18
diakopter: Haven't noticed anything wrong with the code other than "I have no idea what any of this is."
01:25 FROGGS joined
benabik coderef 1:1 static_frame? 01:33
Ah. Yes. 01:34
Hm. Well, I think I wrote the validation for literal code refs, but github seems to be down so I can't share. 01:42
Be back later and hopefully github will be too. 01:43
Well, pushed to github.com/Benabik/MoarVM/commit/a...2253b95d7a but github's slow/dead again. 01:59
02:32 arnsholt joined
diakopter benabik: s/lables/labels/ 02:32
benabik Hah, yes. 02:33
You know, I should probably compile code before pushing it. :-/
diakopter *giggle*
benabik There probably should be some tests for failed validation. :-/ 02:34
diakopter heh. 02:35
well..
I dunno.
I don't like tests.
.. and they don't like me.
benabik Tested now. 74a653a 02:38
I at least didn't break anything.
diakopter heh.
benabik Well, didn't break anything that make test tests. ;)
diakopter well, you modified that existing commit? 02:39
benabik Yeah. I'm a fan of amending and rebasing.
Until it's part of a major repository.
benabik wouldn't amend a commit in MoarVM/MoarVM 02:40
Looks like further validation work is, well, real work. 02:42
02:43 arnsholt joined
benabik Hm. The branch comment is interesting. 02:45
diakopter ? 02:46
which one
(it's probably my fault if it's interesting) 02:47
benabik "Also disallow anything that can branch from running between a prepargs and invoke"
diakopter oh. yeah.
well.
actually that's wrong :)
benabik No branches at all, or just nothing that escapes the area between the instructions?
diakopter the check actually needs to be prevent any branches *to* the region between a prepargs and its invoke 02:48
<- fail at writing todos
benabik Ah, full on branch analysis.
diakopter branching out of it is okay
(just can't come back to it) 02:49
benabik Need to propagate call backwards along jumps and ensure there's a prepargs before it? 02:50
diakopter eh, I think it's sufficient to just prevent any destination labels in the region. currently no compiler creates MAST that would violate that, afaik. 02:52
benabik Well, validators is generally not to protect against the friendly compilers. ;)
diakopter well, true.
well
my rule is stronger :P
(and vastly simpler) 02:53
02:53 arnsholt joined
benabik Yeah. I'm a little worried about corner cases, but I think it could work. 02:53
Basic rule: No jumps from outside a prep->call region to inside it 02:54
Yes?
diakopter I'd recommend building up a list of ranges as the bytecode is processed, then using the destination labels array to compare against them
yes
it's simply to prevent ints and nums values from being interpreted as pointer registers 02:55
benabik Okay, yeah, that works. Less complete, but perfectly sound. 02:56
diakopter hm, also, we need to verify that all the args are prepped for a given call
I don't remember if that's in the todo
benabik Um, no. 02:57
That's not terribly difficult to do if we forbid branching at all in the region.
diakopter they can't have stale data
k
so you'd have to use the MVMCallsite info
benabik If there's branching amongst the argument setting, then you have to start doing something like data-flow analysis to make sure all args are set along all paths. 03:00
And if we just forbid it now, we can loosen it later. :-D
sorear how do you handle foo(1, x ? bar() : baz(), 2) ?
diakopter yeah. I'm sure if a compiler wanted branches in the logic, it could just create distinct callsites.
but actually there's no benefit to such logic 03:01
sorear ?? !! rather :D
diakopter all teh args are already staged in registers
benabik if x { temp = bar() } else { temp = baz } ; foo( 1, temp, 2)
diakopter sorear: args are all evaluated first :P 03:02
oh, you're teasing.
mwahaha
sorear it was a serious question, but I don't know that much about moarvm yet 03:03
a serious stupid question
diakopter you're imagining optimization that isn't there. :) each expression's arg result gets its own register slot (not cpu register), so they're all ready before being staged into the args array 03:05
benabik Hm. No ops to set named args?
diakopter there are
oh.
sorear why have separate ops for staging and calling? if it were atomic the verifier's job would be simpler :D 03:06
diakopter sorear: I look forward to your optimization commits. :) We need a "how fast is sorear's branch yet?" graph 03:07
sorear heh 03:08
need to finish my jvm stuff first :D
diakopter sorear: in particular, I'm more-than-half-hoping you'll discover some brilliant technique to optimize p6 regexes to some form of automata, even dynamically emitted or something.
to find the eventual successful fate before actually creating all the cursors for such a thing. 03:09
also, keeping a cursor object pool could save some gc runs.
maybe not. 03:10
benabik Also: I don't see where argument names are stored. args.h says named args have two entries in the argument list, I don't see where that would be.
sorear someday I will have no choice but to learn how to write an escape analyzer
diakopter benabik: they're just inline
as their own args
num_pos defines the length of the positional extent 03:11
the rest will always be an even number
(arg_count - num_pos)/2 named args
benabik So *arg_flags is a pointer to a mix of call site entries and strings? 03:12
diakopter no
a ptr to a table of flags that define the type of arg in that position
benabik So the names aren't hanging off the call site, but something the call site is also attached to? 03:13
diakopter named/positional, optional/required
and type
they're inline the code
they're set as their own args, I think.
by the compiler
to MAST
sorear benabik: foo( "bar", $x, "quux", $y ) but with flags
diakopter er.
wait, I don't recall exactly 03:14
oh yeah.
benabik Ah, I was trying to see if MVMCallsite knew the names.
diakopter nope
it just knows how many 03:15
benabik So we can't tell if they set the right named args, just that they set enough?
Hm. Maybe I should just TODO anything with validating named args. 03:16
diakopter yeah, at least until someone [jnthn] with a working memory [jnthn] comes online [jnthn] 03:17
benabik Well, I very much doubt I'm going to get very far on this tonight given that I'm going to sleep soon. 03:19
sorear i have a working memory! it has 7±2 slots, most of which are full of $dayjob at the moment 03:20
benabik Mostly trying to sort out what needs to be done.
diakopter benabik: oh, could you add another todo:
optimization: cache all the malloc'd temporaries on the tc 03:21
benabik Huh?
diakopter (including allocated size of course)
all the tables malloc'd during a validation run
.... or am I imagining nonexistent things again 03:22
benabik Oh. tc = thread context. 03:23
03:24 arnsholt joined
benabik There's only one malloced thing so far. 03:24
diakopter benabik: it's annoying that a revised commit keeps its original timestamp
dalek arVM: 74a653a | benabik++ | src/core/validation.c:
Validate literal coderef

Based on bytecode.c:create_code_objects(), there is one coderef object per frame, so check the given number against the number of frames.		 03:25
arVM: ccbdedd | (Matthew Wilson)++ | src/core/validation.c:
Merge pull request #25 from Benabik/master

Validate coderef literal
diakopter (though perhaps irrationally annoying)
benabik Well, the commit timestamp changes but the author doesn't unless you explicitly reset it.
diakopter the timestamp didn't change in the github ui
benabik I don't know if github bothers to show the author date. 03:28
74a653a: AuthorDate: Tue Jun 11 21:39:02 2013 -0400, CommitDate: Tue Jun 11 22:38:19 2013 -0400 03:29
git doesn't bother to show it unless you try. --pretty=fuller for example. 03:30
So: ensure prepargs->invoke is linear. Ensure nobody jumps to it. Ensure all positional are set correctly. Do something magical with named args. 03:35
Probably should also bounds check getting arguments. 03:36
diakopter they're called params there 03:38
(meh)
benabik s/wrong names/right names/ while I'm at it. ;-)
diakopter params is right there 03:39
it's just non-intuitive at first glance
until you realize they're attached to the callee
benabik setting is args, getting is params?
diakopter yes
in this case.
setting is binding args to params. getting is retrieving args from params 03:40
params are the slots
args are the values
benabik Digging through interp.c has convinced me that the named param ops exist but the named arg ops don't. So validation will necessarily have to leave named ops as TODO. 03:41
diakopter they do exist
:P
I wrote tests (er, *a* test?) for them 03:42
benabik Hm. OIC. 03:43
diakopter param_Xn_X <- named
er.
now you got me confused. ;)
benabik And you set them with arg_s odd, arg_* even.
diakopter assuming there's an even number of positionals, yes
benabik Oh, well, yes. 03:44
diakopter brb
benabik Huh. A function call brings together an argument array, a call site, and a code object. The invoked code then needs to validate that the call site is what it needs. Interesting. 03:55
diakopter interesting? 03:56
benabik I had thought a callsite was a descriptor of a location that could be called. It's more a descriptor for a capture. 03:57
diakopter yeah, a call origin
benabik Ah. call site. A place that does a call. 03:58
So params just need to be validated against checkarity? Is checkarity required? 04:00
Can't statically check against the call site there because the call site is dynamic. 04:01
Hm. Looks like the param ops do the checking. I guess it should probably validate against a checkarity if present. 04:02
Hm. Although that heads into data flow stuff again. 04:03
diakopter benabik: yes, some degree of static analysis & optimization is indeed possible 04:12
each param fetch bounds checks wha 04:13
what's requested against what was actually sent
benabik diakopter: Is JIT territory really. If called with callsite A, then skip these checks and do these exact conversions instead of checking types.
Hard to do static analysis when the incoming data is dynamic. :-D 04:14
Anyway, I _really_ should sleep. 04:21
diakopter 'nite 04:28
04:47 arnsholt joined 05:53 chipdude joined 07:46 FROGGS joined
diakopter chipdude: howdy 07:48
decommute & 08:07
09:04 JimmyZ joined
JimmyZ jnthn: will MVMContainerSpec will be different from ContainerSpec in NQP? 09:05
jnthn That particular data structure won't be, nor will the op to set it up. Various aspects of its implementation besides that will be different, however. 09:08
sorear oh, chipdude == chip 09:12
howdy!
jnthn o/ chip
(from backlog) a callsite is a static description of the number/kind of arguments being passed (that is, it's caller side). In general, args = caller side things, params = callee side things. 09:14
jnthn has tried to stick to that naming everywhere; violators of that naming rule should be fixed... 09:15
diakopter hee 09:20
oh, you mean violations. ;)
jnthn oh!
.oO( prosecuters will be violated... )
 09:21
diakopter hee.
09:26 arnsholt joined
JimmyZ jnthn: thanks 09:30
diakopter anyone know liz' google account address (not necessarily gmail.com address)? 09:35
09:37 arnsholt joined
diakopter sorear: plz PM me your google account address so I can share docs with you and invite to hangouts and such 09:50
sorear: nm, found it. :) 09:54
sorear i have two 09:55
which one did you find
09:57 arnsholt joined
diakopter the one I've emailed before 09:58
10:24 cognominal joined 11:59 pmichaud joined, masak joined 12:32 tgt joined 13:01 birdwindupbird joined 14:02 tgt joined
benabik ~~ 14:30
14:32 JimmyZ joined 14:35 tgt joined, FROGGS joined
benabik Hm. I should store my todo list somewhere other than the backscroll. 14:41
jnthn If you put it there, it ends up in the IR clog for free... :)
benabik But I hate digging up clogs. Tend to be smelly and full of hair. 14:42
jnthn You'd think the IR would help with the smell...
JimmyZ sigh.. I got my container.c work lost after git clean -xdf :( 14:43
benabik Well, there's a reason you have to -f to make that happen. :-/
IR generally makes things smellier, UV is cleansing.
FROGGS JimmyZ: you have the editor still open? 14:45
JimmyZ FROGGS: Nope, I misreadjnthn++'s answer and then I did git clean -xdf, due to my poor english 14:47
jnthn gently suggests that JimmyZ will find it easier to learn "git stash" than jnthn will find it to learn Chinese :P 14:48
FROGGS :/
JimmyZ I did git stash, but I think before it i did git clean -xdf.. 14:49
dalek arVM/container: dcc5b46 | jimmy++ | src/ (4 files):
stub in container
15:04 bronco_creek joined
jnthn JimmyZ: github.com/MoarVM/MoarVM/commit/dc...7dc#L3R338 - plesae don't comment out code. If it's no longer relevant, just delete it. 15:15
Otherwise somebody will discover it months on and have no idea what the significance of it's commenting out is...
*its 15:16
JimmyZ jnthn: ok, will do it
masak is there an ANSI C version where using C++ comments like that doesn't trip up the C compiler? 15:17
jnthn Yes, MSVC I think :)
benabik C99?
masak ok, that sounds familiar.
jnthn um, wait
MSVC will trip up on those, iirc.
Should use /* ... */
benabik MSVC doesn't do C99? 15:18
jnthn benabik: No
benabik It's only 14 years later. :-/
PerlJam Does moarvm code target a specific version of C? 15:20
nwc10 I assume "what works" 15:23
you can't use // comments or have declarations after statements in MSVC 15:24
benabik Perhaps we should use -ansi for gcc to warn about these things. 15:25
Or at least -Werror= for particular things we care about. 15:26
jnthn That may help avoid accidental breakage, yes
PerlJam so ... C89? 15:27
benabik -ansi == -std=c89
PerlJam ok
nwc10 IIRC the only platform I found where you could tell gcc to *error* for non-C89 source was FreeBSD
PerlJam might want to put that in any docs for developers (if there are any; I haven't looked) 15:28
benabik Documentation good, getting the compiler to complain better.
nwc10: I would hope that -ansi causes it to complain.
And even if -ansi doesn't error decl-after-stmt, you can use -Werror=decl-after-stmt 15:29
nwc10 yes, you get warnings
for a new enough gcc.
it's worth doing, but there's a trade off between how far you can crank up gcc's grumpiness, and how fragile your build is on platforms you didn't test on 15:30
benabik Ensuring we don't accidentally break things on a major platform (MSVC) when coding on another major platform (fairly recent GCC on Linux) is probably worth rarer platforms needing to sort out their CFLAGS differently. 15:31
jnthn Well, it's possible to have a "perl Configure.pl --dev" which cranks them up for developers.
But not for those who just want to build.
nwc10 If I'm permitted to comment on something else, the current build system doesn't scale
you're adding a lot of redundant information for every *BSD variant, let alone every *nix like thing 15:32
benabik would rather it be default... Or at least default when there's a .git directory or something.
nwc10 probably a lot of duplication would be reduced by using Perl::OSType and bundling it if needed 15:33
[Coke] nwc10: e.g. when I added darwin, I bemoaned that to myself but erred on the side of "get the build working". ;) 15:34
benabik It's very repetitious now, but could be refactored with some way to specify the default gcc/clang options etc. 15:40
jnthn You can say that again! 15:44
benabik It's very repetitious now, but could be refactored with some way to specify the default gcc/clang options etc.
jnthn :P
benabik :D
jnthn Really, the build system needs love from somebody who knows enough about enough platforms to know how to factor it better.
nwc10 the suggestion about Perl::OSType is to have it default to good (or good enough) on the other >3 *BSD variants that no-one's tried yet 15:45
jnthn So far it's been a "make something work".
nwc10 I agree with "make something work"
benabik Although using Perl::OSType also raises a question... "Perl::OSType was first released with perl v5.13.9" "5.13.9 2011-Jan-20" 15:48
nwc10 bundle it. it's smaller than APR :-) 15:49
benabik OS X 10.7 only has 5.12.3
There's that.
15:51 tgt joined
benabik Huh. I have 5.12.4 and Perl::OSType. Apple's so strange sometimes. 15:53
[Coke] I have 5.12.4 and no Perl::OSType. 15:55
(on 10.8)
benabik Huh. Must have grabbed it with CPAN somewhere along the way. I thought I hadn't used CPAN on this machine.
16:01 tgt joined
dalek arVM/container: f53f66b | jimmy++ | src/ (3 files):
port more container code
 16:02
JimmyZ 'night 16:03
nwc10 I have a /usr/bin/perl and Apple haven't applied the CVE fix
BinGOs has a pending update for the build system 16:09
in fact I have bundled in Perl::OSType and I have done a bit of refactoring. 16:10
I need to do compiler preferences, then I'll submit a pull request.
benabik Good to know. I'll stop my quick and stupid refactor. :-D 16:12
BinGOs I just ran out of tuits last night.
benabik MVMConcatState unused? 16:13
BinGOs stares blankly. 16:14
diakopter benabik: heh, fossil. :) 16:37
benabik diakopter: Moar's new enough I can't tell the difference between fossil and hasn't been hooked up yet.
diakopter BinGOs: maybe add an option to specify clang over gcc 16:38
benabik +1
diakopter wonders if rurban would test it on cygwin
16:54 tgt joined 17:33 patspam joined 17:48 cognominal joined 17:58 lizmat joined
diakopter chipdude: ping 18:04
eternaleye ...Hm. I have clearly gone insane, because I am suddenly struck with a strange desire to add a DFHack plugin for MoarVM, so that I can automate Dwarf Fortress in NQP (for now) and P6 (in the future) 18:23
diakopter eternaleye: what's its interface? 18:25
eternaleye diakopter: Not entirely sure, but there are already lua and ruby plugins.
diakopter: It's embed rather than extend, from the perspective of the language. 18:26
diakopter ..maybe if I knew what Dwarf Fortress was.... my first guess is a miniaturized version of Team Fortress 18:27
eternaleye diakopter: Nope.
diakopter: More like Sim City meets Nethack meets Oregon Trail
timotimo_ oooh dwarf fortress 18:28
jnthn
.oO( Well, this is one way to get somebody to sort out the embedding support... )
diakopter speaking of. 18:29
eternaleye diakopter: Its scale is impressive. The game starts by generating a world procedurally, including soil erosion and orographic rain shadows. Then it starts simulating politics among civiliations that form, and war, and then you get to start playing.
It has *one* dev. 18:32
diakopter what made me think Mark Biggar had a hand in designing XS 18:35
18:38 benabik joined 19:01 tomyan joined 19:10 tgt joined 19:25 pmurias joined
pmurias has anyone tried to setup moarvm testing on travis? 19:25
nwc10 I'm unaware of anyone *trying*, but I'm not the most likely person to know 19:27
benabik dukeleto's the only person I know of to use travis.
pmurias is there a better alternative to travis? (me hasn't used travis but something like that seems like a good idea) 19:29
nwc10 I think that dukeleto had expressed some interest in doing it, but I don't know if he has the time. I supsect he's very busy keeping the world bootstrapped with decent coffee 19:32
benabik Not that I'm aware of? dukeleto just set it up for Parrot and I don't know anyone else who's used it. ISTR someone saying dukeleto was going to... what nwc10 said.
pmurias gives up on adding moarvm to travis as I'll need building parrot,nqp,etc... 19:46
diakopter I'm quite sure dukeleto will get to it at some point 19:47
19:59 tomyan joined
chipdude diakopter: g'day 20:25
I suppose most here have played with the Lua VM, but I'd like to mention that when I embedded it in an app the feature that I didn't expect and most appreciated was the debugging hook called every N opcodes 20:27
it allowed me to prevent runaway time usage. and of course the malloc hooks allowed memory caps. 20:28
I can speak to C99. There are four C99 features that we can get with MSVC: inline functions, // comments, <stdint.h>, and <inttypes.h>, 20:29
s/,$//
if we're willing to demand the use of MinGW or the like then we can go C99, else not. pity 20:30
otoh inline functions is probably the most important C99 feature, possibly tied with mixed-decls-and-statements
22:17 lizmat joined
diakopter jnthn: ping 22:53
jnthn diakopter: pongish 22:54
diakopter oops wrong window 22:56

Powered by Cro and the Raku® Programming Language.

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

Thank you!