[01:14] *** dalek joined [01:40] *** PerlJam joined [01:40] *** dalek joined [01:45] *** [Coke] joined [02:23] aha, aha. two instructions directly in sequence grabbing the value from one spesh slot, then the next one and putting them into the very same register [02:23] without kicking one of the getspeshslot instructions out [07:33] *** FROGGS joined [08:13] *** zakharyas joined [08:46] *** woolfy left [08:53] http://paste.scsys.co.uk/438182 -- SAN barfs on t/spec/S17-lowlevel/thread.t [08:53] no, I don't know what changed, but the stack trace starts in get_lexotic_for_handler_idx [09:19] *** kjs_ joined [09:51] *** kjs_ joined [11:01] oh, i didn't even see hoelzro's patch up there [11:02] MoarVM: 811b90c | (Timo Paulssen)++ | src/jit/graph.c: [11:02] MoarVM: jitting write_fhs lets us jit many print methods/subs [11:02] MoarVM: review: https://github.com/MoarVM/MoarVM/commit/811b90cf55 [11:25] ohh nice! [11:25] timotimo++ [11:26] doesn't seem to win us terribly much [11:26] we may want to have a specialization for dynamic variables like $*OUT [11:26] it seems like we're spending quite some time looking up $*OUT over and over [11:26] or $*EXECUTABLE [11:27] maybe [11:27] I dunno [12:49] *** JimmyZ joined [12:49] timotimo: ping [13:01] hmm, It would be nice to have docs/opitmize.md [13:21] hey JimmyZ [13:21] hello [13:22] what kind of stuff would you like to see in that document? [13:24] I think the basic structure, and a bit examples about the idea of optimization like optimize_coerce [13:24] so it should focus on spesh? [13:24] and how to debug ... [13:25] yeah [13:27] speaking of optimize_coerce, is it trying to change coerce_in to const_n and let the eliminate_dead_ins function remove the pre const_i instruction? [13:28] and before coerce_in, it always is const_i? [13:30] of course not [13:30] but there are many cases when a const_i is followed by a coerce_in [13:30] because our code-gen does that in a not so clever way [13:31] that's why it looks for "KNOWN_VALUE [13:31] " [13:36] what's about the first question? [13:40] ah, yes, the const_i instruction should be removed by remove_:ead_ins [13:41] nice [13:41] well, hopefully :) [14:14] * [Coke] wonders if nwc10 is seeing any of the failures coke & lizmat are getting with moar, jit or nonjit. [14:14] <[Coke]> (coke & lizmat are on os x) [14:30] MoarVM: 26656ba | (Timo Paulssen)++ | src/spesh/dump.c: [14:30] MoarVM: for logging purposes, a BB's address is a good unique ID [14:30] MoarVM: review: https://github.com/MoarVM/MoarVM/commit/26656ba004 [14:32] [Coke]: no. only the one I reported. Everything else passes on "my" machine [14:32] x86_64 CentOS [14:35] <[Coke]> shame. :) [14:51] MoarVM: 5dd519d | (Timo Paulssen)++ | src/spesh/ (3 files): [14:51] MoarVM: dump a spesh graph's callsite (if any) [14:51] MoarVM: review: https://github.com/MoarVM/MoarVM/commit/5dd519d051 [15:08] timotimo: I thinks it should use MVM_string_utf8_encode_C_string instead of MVM_string_utf8_encode [15:09] oh? i stole that from a few lines below in that same file %) [15:12] that may be wrongish, since MVM_string_utf8_encode returns non-null-end strings, and call strlen ( in append function ) to the non-null-end strings may ... [15:18] https://imgur.com/N1sbOYa [15:21] what are the colors for? [15:21] readability [15:21] making the same register pop out visually [15:21] ahh, well, if you say so :o) [15:21] ahh [15:22] it's far from perfect [15:22] i'd really appreciate feedback [15:22] i'm not sold on the "greyscale for version number" idea either [15:23] that's nice! [15:25] https://imgur.com/hFNJ0VV - maybe a bit better [15:26] that's better, aye [15:27] https://imgur.com/KQskiOx - wow. [15:29] wow, but why there is a background color in the right shape :) [15:30] hm? [15:30] you mean outside the triangle? [15:30] yeah [15:31] version of the register [15:31] follows a greyscale [15:33] *** vendethiel joined [15:33] It a bit hurt my eye ... and others are awesome :P [15:34] the background part :P [15:34] MoarVM: e2264ff | (Timo Paulssen)++ | tools/colorize_spesh.p6: [15:34] MoarVM: tool to colorize registers in spesh logs [15:34] MoarVM: review: https://github.com/MoarVM/MoarVM/commit/e2264ffdda [15:34] feel free to play around with different "hashing algorithms" :P [15:38] ;) [15:45] now i want a tool that turns the SSA form into a graphviz graph [15:57] that will be much more awesome with dotted arrow [15:57] ;) [15:58] dotted ... arrow? [15:58] to guide the next version [16:00] ah, yes [16:00] these would go through operators [16:01] something like git lg [16:01] :P [16:02] i wonder if i can tell graphviz to connect two nodes as "messily" as necessary [16:03] .oO(Perl6::Cario...) [16:06] doesn't seem to exist [16:59] <[Coke]> moar/moar-jit still the failingest backends. [18:14] *** FROGGS joined [19:22] MoarVM: c0053cf | (Timo Paulssen)++ | tools/ (2 files): [19:22] MoarVM: let update_ops.p6 also write a perl6 module with ops data [19:22] MoarVM: review: https://github.com/MoarVM/MoarVM/commit/c0053cfd61 [20:17] timotimo: what are you going to do with MAST::Ops? [21:27] FROGGS: i need to know if the first argument of an op is a write or read register [21:27] or a write/read [21:27] (like in the inc op, iirc) [21:37] ahh [22:58] *** lizmat joined [23:00] *** woolfy joined [23:04] *** woolfy joined [23:09] *** woolfy joined [23:11] *** woolfy1 joined [23:51] *** xiaomiao joined