Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2014-04-23

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

All times shown according to UTC.

Time Nick Message
00:05 tadzik joined #moarvm
00:09 jnthn timotimo: There will be a way, but I don't have a design I'm happy with yet. I have to figure it out for inlining. It'll get figured out when that's near the top of the task list.
00:09 jnthn timotimo: On the boxing instructions, yes, makes sense.
00:31 lizmat joined #moarvm
01:06 btyler joined #moarvm
01:23 FROGGS_ joined #moarvm
02:06 woosley joined #moarvm
06:52 zakharyas joined #moarvm
08:54 FROGGS joined #moarvm
10:06 cognominal joined #moarvm
13:30 dalek MoarVM/optimize_isreprid: a6205ef | (Timo Paulssen)++ | src/spesh/optimize.c:
13:30 dalek MoarVM/optimize_isreprid: try to optimize islist/isnum/... (not helpful yet)
13:30 dalek MoarVM/optimize_isreprid:
13:30 dalek MoarVM/optimize_isreprid: the problem is that islist and friends also include a null
13:30 dalek MoarVM/optimize_isreprid: check on their argument. At best, we could - if we only know
13:30 dalek MoarVM/optimize_isreprid: the type - turn these ops into a negated "isnull" op, but that
13:30 dalek MoarVM/optimize_isreprid: doesn't exist. So we have to wait for spesh's ability to
13:30 dalek MoarVM/optimize_isreprid: allocate new registers and add new operations on those.
13:30 dalek MoarVM/optimize_isreprid: review: https://github.com/MoarVM/MoarVM/commit/a6205ef0f6
13:39 dalek MoarVM: c34942c | (Timo Paulssen)++ | src/spesh/facts.c:
13:39 dalek MoarVM: boxing ops carry their types and result in concrete objects
13:39 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/c34942c797
13:39 timotimo this ought to get rid of a few decont ops, too.
13:40 timotimo they are not that expensive, but better is better.
13:40 moritz ... unless you spend more time rewriting the bytecode than executing the original bytecode
13:40 timotimo hehe.
13:41 timotimo in our experience, the specializer seems to be blazing fast
13:42 timotimo i'm not expecting us to gain good performance just by removing lots of decont ops :)
13:42 timotimo especially since currently we turn them into set ops instead of renaming the registers
13:42 timotimo which i would prefer if it weren't complicated to do :)
13:43 JimmyZ :)
13:44 btyler joined #moarvm
13:44 timotimo maybe we should introduce a KNOWN_NOT_NULL fact?
13:46 timotimo after object creation, we should know it's not null, for example
13:50 timotimo hmm. in theory, with a KNOWN_TYPE, we should be able to turn a getwhat into a wval or something like that; i wonder if that'd be more efficient?
13:50 timotimo (but if we do turn it into a wval, we also have its known object value available in the specializer)
13:55 timotimo gethow should be specializable, too, shouldn't it?
14:17 timotimo [Annotation: INS Deopt (idx 0 -> pc 170)]
14:18 dalek MoarVM: eadd2c8 | (Timo Paulssen)++ | src/spesh/dump.c:
14:18 dalek MoarVM: dump deopt instruction annotations, too.
14:18 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/eadd2c8f73
14:33 timotimo not sure what to do next. somehow i still don't feel up to the task of making nameds work properly :|
14:34 timotimo with interned callsites and spesh, that is
14:35 FROGGS timotimo: implement LABLEs \o/
14:35 FROGGS err, LABELs, even
14:38 timotimo oh, look at my wrist! i gotta go!
14:38 timotimo how about something easy that'll result in a dramatic speed improvement in common cases?
14:38 timotimo oh ... right ... those things aren't easy :|
14:38 moritz FROGGS: now you've scared him away :-)
14:39 FROGGS timotimo: these things are perhaps just wishes and dreams :o)
14:40 FROGGS at least I merged master into nqp/loop_labels, so if someone wants to play with it, nqp-p supports loop labels :o)
14:48 timotimo what's the exact significance of 512, 513 and 514?
14:51 FROGGS these are just IDs of exception handlers
14:51 FROGGS (next/last/redo)
14:51 moritz feels like they could go into pir::const thingy
14:52 FROGGS moritz: and then I can use such a const in pir code?
14:53 moritz FROGGS: in the NQP that generates the pir code
14:54 FROGGS moritz: well, then I can stick to my hash: https://github.com/perl6/nqp/compare/loop_labels#diff-900522366d019758be5b6ddbd59b29f1R552
14:55 moritz ah, right
15:26 FROGGS[mobile] joined #moarvm
16:06 FROGGS joined #moarvm
17:59 benabik joined #moarvm
18:58 zakharyas joined #moarvm
19:10 woolfy1 joined #moarvm
19:44 moritz FROGGS: https://rt.perl.org/Ticket/Display.html?id=121720 you recently fixed a bug with the same symptoms. HALP?
19:45 FROGGS moritz: last time it thought it never hit eof
19:49 * moritz was hoping for a patch :-)
19:50 woolfy joined #moarvm
19:51 FROGGS moritz: *g*
19:51 FROGGS let me take a sip on my beer first
19:53 FROGGS btw, working on loop labels again on nqp-m feels like wearing that old good jacket that fits so well...
19:53 FROGGS it feels right somehow
20:00 timotimo there's not as many weird gotchas everywhere :)
20:12 woolfy joined #moarvm
20:31 brrt joined #moarvm
20:31 brrt hey #moarvm
20:32 FROGGS ho brrt
20:33 brrt much progress for me to catch up with
20:34 jnthn well, not in Moar
20:34 jnthn Since weekend, anyway
20:39 timotimo aye, mostly cool supply methods
20:40 timotimo jnthn: comments about the things i wrote about spesh before?
20:42 brrt jnthn: https://github.com/MoarVM/MoarVM/blob/master/src/io/eventloop.c#L38 - shouldn't this be += ?
20:51 jnthn brrt: The code using the return value only cares about it in boolean sense.
20:51 brrt ok
20:51 brrt :-)
20:52 jnthn timotimo: Not at the moment...busy trying to write my talk and debug code for it... :)
20:52 jnthn Turns out we have some issue involving continuations when multiple threads are involved :(
20:54 FROGGS :{
20:54 brrt jnthn - btw, i'm sorry to say i won't make it to the dutch perl workshop this friday
20:55 jnthn brrt: ok, no worries
20:55 brrt its just a shame
20:55 jnthn aye
20:55 brrt hopefully you'll visit our pretty country another time
20:55 jnthn probably :)
20:55 brrt or, i'll go to sweden
20:56 jnthn or that :)
20:57 timotimo that's fair :)
20:57 timotimo (re: my spesh things)
21:19 * brrt is off to sleep again, see you tomorrow
21:19 brrt left #moarvm
21:27 FROGGS moritz: I have a patch
21:32 dalek MoarVM: a109e8d | (Tobias Leich)++ | src/io/syncfile.c:
21:32 dalek MoarVM: stat file descriptor in eof() when filename is unset
21:32 dalek MoarVM:
21:32 dalek MoarVM: This can happen when we get a file redirected, and therefor open it via
21:32 dalek MoarVM: MVM_file_handle_from_fd instead of MVM_file_open_fh.
21:32 dalek MoarVM: review: https://github.com/MoarVM/MoarVM/commit/a109e8d765
21:38 timotimo cool
21:43 timotimo what was the gsoc irc channel for perl again?
21:45 tadzik soc-help
21:45 timotimo oh
22:06 donaldh joined #moarvm
22:43 cognominal joined #moarvm
23:05 benabik joined #moarvm
23:41 benabik joined #moarvm
23:48 woolfy joined #moarvm
23:57 benabik joined #moarvm

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