Camelia, the Perl 6 bug

IRC log for #moarvm, 2013-10-20

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:32 ssutch joined #moarvm
02:17 benabik joined #moarvm
08:21 donaldh joined #moarvm
08:23 FROGGS nwc10: I get to the same error like you fwiw
08:24 FROGGS damn, the -ldl was an easy fix... I should have spotted it
08:55 dalek MoarVM: b85d5e6 | (Tobias Leich)++ | src/main.c:
08:55 dalek MoarVM: fix linebreaks in USAGE msg
08:55 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/b85d5e60d7
10:13 benabik joined #moarvm
11:18 dalek MoarVM/ext: 81c654e | moritz++ | src/strings/ops.c:
11:18 dalek MoarVM/ext: return -1 from index_s with start out of range
11:18 dalek MoarVM/ext:
11:18 dalek MoarVM/ext: this brings it in line with parrot and JVM, and easier for regex codegen
11:18 dalek MoarVM/ext: review: https://github.com/MoarVM/MoarVM/commit/81c654ed61
11:18 dalek MoarVM/ext: b85d5e6 | (Tobias Leich)++ | src/main.c:
11:18 dalek MoarVM/ext: fix linebreaks in USAGE msg
11:18 dalek MoarVM/ext: review: https://github.com/MoarVM/MoarVM/commit/b85d5e60d7
11:18 dalek MoarVM/ext: 0f32f5e | (Tobias Leich)++ | src/ (2 files):
11:18 dalek MoarVM/ext: Merge branch 'master' of github.com:MoarVM/MoarVM into ext
11:18 dalek MoarVM/ext: review: https://github.com/MoarVM/MoarVM/commit/0f32f5ec28
11:18 dalek MoarVM/ext: 28c47b6 | (Tobias Leich)++ | build/Makefile.in:
11:19 FROGGS dalek: that's all?
11:20 FROGGS it should state this too: "install config.h and libmoar" and "added --cflags and --libs"
11:21 FROGGS jnthn: I hope the flags are correct
11:57 colomon joined #moarvm
12:17 cognominal joined #moarvm
13:58 ggoebel10 joined #moarvm
14:03 grondilu joined #moarvm
14:07 ggoebel11 joined #moarvm
14:11 BenGoldberg joined #moarvm
17:26 cognominal joined #moarvm
17:50 timotimo joined #moarvm
17:51 FROGGS I have a problem regarding to http://irclog.perlgeek.de/m​oarvm/2013-10-16#i_7722788
17:52 FROGGS I thought it might be nice that the nqp::backendconfig op would return a proper hash of strings/arrays/hashes...
17:52 FROGGS but for that I have to turn %config hash in Configure.pl into C code
17:53 FROGGS so, should I create a backendcfg.c.in file, that gets filled by Configure? this would then provide a function that returns the hash
18:15 FROGGS I'll try this now with build/config.c.in
18:59 jnthn FROGGS: Something like that, yeah
19:00 FROGGS k, good to know
19:00 FROGGS jnthn: I will push to a branch, so it is reviewable
19:51 dalek joined #moarvm
20:00 ssutch joined #moarvm
20:00 FROGGS I believe the config thingeny works
20:02 jnthn k
20:03 FROGGS hmmm, do I need to create a stage0? it tells me here that the operation is unknown:
20:03 FROGGS class HLL::Backend::MoarVM {
20:03 FROGGS our %moar_config := nqp::backendconfig();
20:04 FROGGS if I dont use it it HLL::Compiler, then I can execute nqp::backendconfig()
20:07 jnthn Yeah, you can't use an nqp:: op if it's not known
20:17 FROGGS ahh damn, push_s strikes again
20:18 FROGGS ahh, wait
20:24 FROGGS jnthn: seems to work :o)
20:28 dalek MoarVM/config: c45df2e | (Tobias Leich)++ | / (11 files):
20:28 dalek MoarVM/config: added bakendconfig op
20:28 dalek MoarVM/config:
20:28 dalek MoarVM/config: Which return the hash $config from Configure.pl.
20:28 dalek MoarVM/config: review: https://github.com/MoarVM/MoarVM/commit/c45df2e404
20:32 FROGGS that is the generated C file: https://gist.github.com/FR​OGGS/6c57472990e44a8565d0
20:33 FROGGS nqp -e 'say( nqp::backendconfig()<version> )'
20:33 FROGGS 2013.10-42-g1e8ea80
20:34 FROGGS so, if that will be applied, I would rip out --cflags and --libs again
20:34 FROGGS --version can stay I'd say
20:39 jnthn I thought we were adding --show-oncig to dump the lot?
20:40 FROGGS jnthn: yes, this is what I'd do next
20:41 jnthn ok, but then we don't need the ones for specific flags
20:41 jnthn That cna just dump all of them
20:41 FROGGS exactly
20:41 FROGGS I would find and optional key to --show-config handy though
20:42 FROGGS dunno if rakudo would benefit from that
20:44 FROGGS jnthn: btw, --show-config would already work by adding three lines to HLL::Backend (after creating a new stage0)
20:54 jnthn aye
21:04 [Coke] is backendconfig basically the same thing as jvmgetconfig?
21:04 [Coke] (if so, can we unify the names?)
21:05 FROGGS same names for same things would be cool, yes :o)
21:07 FROGGS then we should wrap pir::getinterp__P()[pir::co​nst::IGLOBALS_CONFIG_HASH] into the new opname too, and we'd have a single thing in rakudo
21:07 diakopter .
21:07 diakopter <- is still alive
21:08 FROGGS :o)
21:11 FROGGS jnthn: what would be the name of choice? nqp::backendconfig? in nqp there is $!backend.config for example, so this would fit
21:12 FROGGS gnight
21:17 jnthn FROGGS: Seems reasonable
21:51 colomon joined #moarvm
22:22 [Coke] s/backend/vm/
22:22 [Coke] maybe.
22:22 [Coke] FROGGS++ jnthn++ either way.
22:27 d4l3k_ joined #moarvm
23:42 gshank joined #moarvm
23:56 colomon joined #moarvm

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary