Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2015-08-14

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

All times shown according to UTC.

Time Nick Message
00:46 ggoebel joined #moarvm
00:58 ShimmerFairy joined #moarvm
01:47 ilbot3 joined #moarvm
01:47 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
01:57 TEttinger feed it snowden leaks
05:00 harrow joined #moarvm
07:03 zakharyas joined #moarvm
07:10 lizmat joined #moarvm
08:23 mtj_ joined #moarvm
08:23 brrt joined #moarvm
08:34 brrt good moarning
08:36 jnthn dobre jitro :)
08:38 brrt iirc... jitro was late morning?
08:38 jnthn Yes :)
08:42 * jnthn starts typing "how to make atom package" and before he could get done google was suggesting "how to make atom bomb" o.O
08:43 brrt it's an interesting topic :-)
09:02 FROGGS jnthn: that probably just happens because you implemented DESTROY not too long ago :o)
09:03 * FROGGS .oO( $*W.^can: 'DESTROY' )
09:15 jnthn :)
09:42 Ven joined #moarvm
09:45 FROGGS_ joined #moarvm
12:57 brrt joined #moarvm
13:22 brrt ok, thinking about it properly, i concur that instruction selection comes after tile selection, and that it'd better be done in a preorder traversal order
13:26 brrt or something like that...
13:51 timotimo i still don't grok MVM_ASSIGN_REF
13:51 timotimo what i do understand is it's about serialization contexts
13:51 jnthn It's not :)
13:52 timotimo oh
13:52 jnthn It's about generational GC
13:52 timotimo that's sc_wb, then
13:52 timotimo okay; why do we need to reload objects when we hit a write barrier?
13:52 jnthn Right, this is just the normal write barrier
13:52 timotimo or is that only for when we have a p6opaque?
13:53 timotimo oooh
13:53 timotimo these stores and restores are just because registers get clobbered
13:53 brrt what
13:53 brrt which registers
13:53 timotimo that explains why it writes to rbp-0x*
13:53 timotimo TMP1 and TMP2 in the cases i'm trying to cargocult from :)
13:54 brrt ahaaah
13:54 timotimo no, TMP2 and TMP3
13:54 jnthn The write barrier might end up calling a C function
13:54 timotimo ah!
13:54 brrt what jnthn says
13:54 timotimo now there's an explanation that helps :)
13:54 brrt hit_wb calls a function, hence the values are spilt
13:54 brrt (temporaries)
13:54 timotimo i'm now finally going to implement setcodeobj via the old jit
13:55 brrt mvm_assign_ref is split in two, in a sense
14:07 timotimo yeah, check + hit
14:10 timotimo bleh!
14:10 timotimo setcodeobj never has a known type + correct reprid
14:16 timotimo it seems like that always comes from $!do
14:16 timotimo bbiab
14:32 camelia joined #moarvm
15:52 lizmat joined #moarvm
15:57 timotimo maybe i should add a little piece to spesh that checks if there's a guard for $!do having a known type so that that won't be thrown out before jit?
16:09 timotimo jnthn: we don't emit log instructions for the result of getattr_o; is there a good reason to keep it that way?
16:21 timotimo i was wrong
16:21 timotimo we actually do put in a log op for getattr_o and getattrs_o
16:22 timotimo darn, they log different types, but of course all have the same repr
16:24 FROGGS PR 233 is quite interesting
16:40 lizmat joined #moarvm
16:57 japhb FROGGS: Feels like it should come with a mini-benchmark though.  I would expect MVM{,_bigint}_radix are fast-path operations; it would be nice to avoid making them much slower (if indeed this has any significant effect).
17:10 timotimo perhaps a "known reprid" fact would be interesting?
19:59 synbot6 joined #moarvm
20:44 dalek MoarVM/jit_and_opt_setcodeobj: 0c3c8fb | timotimo++ | src/ (3 files):
20:44 dalek MoarVM/jit_and_opt_setcodeobj: jit and try to opt setcodeobj
20:44 dalek MoarVM/jit_and_opt_setcodeobj:
20:44 dalek MoarVM/jit_and_opt_setcodeobj: alas, we pretty much always log different type objects.
20:44 dalek MoarVM/jit_and_opt_setcodeobj: since the opt is only interested in REPR ID, maybe that
20:44 dalek MoarVM/jit_and_opt_setcodeobj: ought to become an entry in facts?
20:44 dalek MoarVM/jit_and_opt_setcodeobj: review: https://github.com/MoarVM/MoarVM/commit/0c3c8fb508
22:45 TEttinger joined #moarvm
22:49 notostraca joined #moarvm
23:44 notostraca joined #moarvm

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