01:16 pyrimidine joined 02:12 diakopter joined 02:24 pyrimidine joined 02:40 MasterDuke joined 02:48 ilbot3 joined 03:26 pyrimidine joined 03:36 geekosaur joined
samcv there's a nqp test harness? 03:49
japhb samcv: Yes, pmurias wrote most of it to ensure his nqp-js work was on the right track. 03:57
samcv nice
japhb Oh, actually, that would be nqp test *suite*. I haven't actually checked what the NQP 'prove' or equivalent is written in.
samcv weird still finds everything perfectly, but i cannot run perl 6 or compile nqp, and there's a hang on the spectest 04:04
let me push it to my repo 04:05
github.com/MoarVM/MoarVM/compare/m...g?expand=1 04:15
diakopter samcv: do you have a guard that "h_pos + 1" is valid 04:21
I mean, I dunno how that could cause a hang 04:22
but stil
samcv let me add a print and make sure thatn isn't happening 04:25
yeah it seems to be pretty rare, but I think that's the issue 04:26
at least i saw it shortly before the hang
diakopter crosses fingers
diakopter fondly remembers writing "brute force for now. horrible, yes. halp." 04:27
samcv uh nope 04:28
fixed it and now not getting any printout there, but it's hanging 04:29
diakopter well, it's not valid to skip along += ngraphs
samcv I changed the while loop to while (h_pos < hgraphs - 1)
diakopter change it to increment only 1 04:30
why would it be okay to += ngraphs
samcv still hang diakopter 04:31
diakopter oh, because it's ==2, yeah, I suppose
samcv yep
MasterDuke samcv: did you check out my boyer-moore branch at all?
yoleaux2 23 Jan 2017 10:31Z <lizmat> MasterDuke: seems like a logical continuation of the other Nd work you've done :-)
samcv i tad 04:32
MasterDuke it worked fine for me (no problems), but it wasn't faster
and i don't know why
diakopter: if you wrote that original code maybe you could take a look at what i tried? 04:33
github.com/MasterDuke17/MoarVM/tre..._for_index
samcv MasterDuke, can you link me it again?
i'll look a little closer
diakopter samcv: could you confirm that it's looping in your while loop?
MasterDuke: is yours working?
samcv yeah gonna add some prints 04:34
diakopter samcv: (or is it looping as the result of some bad result of that function)
MasterDuke if i remember correctly, it works in that it gives the same results as the original code, it just isn't any faster
sure there's some overhead from building the tables, but i would have expected to see some difference at some combination of needle/haystack size 04:35
diakopter well, how many times is that MVM_string_index called
samcv yeah it seems to be stuck 04:37
loopy h_pos 1 hgraphs 10
loopy h_pos 3 hgraphs 10
both those lines just printed on loop
MasterDuke diakopter: was that question to me?
samcv so it's looping twice and then.. starting over? or someplace is calling it again and a again? 04:38
should add in a print for when it exits the loop
diakopter MasterDuke: sure :)
anyone
MasterDuke diakopter: just once per nqp::index call right? but i think i tried doing many in a loop and the total time was no faster 04:39
diakopter just curious for the motive of optimization (or, how much one could theoretically speed it up)
MasterDuke but i believe also used for literal regexes
e.g., /foobar/
samcv no, it's exiting the loop
diakopter well I meant in a long-running program such as p6 setting compilation, how many times is it called
samcv a huge no
diakopter samcv: ok, so it's returning a wrong result 04:40
MasterDuke i've profiled the compile of rakudo, but moarvm ops don't show up in a profile at that level
diakopter yeah you'd have to use something fancy :D
MasterDuke i thought i once perf recorded a rakudo compile, but i'm not sure if i gisted the results 04:41
samcv ok yea'h it is
10 in the haystack it fails somehow. will fix that
curious what part of the code just uh. rechecking forever
if it can't find something
diakopter samcv: yes I'm curious about that too 04:42
MasterDuke diakopter: gist.github.com/MasterDuke17/4b9dc...ac0b2556cd 04:44
also running a new one now 04:46
diakopter 0.16% 0.07% moar libmoar.so [.] MVM_string_index 04:47
so the max you could speed up that run, by optimizing that function infinitely, is only that fraction 04:48
thereabouts
samcv yeah that's what i thought, i can't change the while loop to while (h_pos < hgraphs - 1) 04:49
MasterDuke of the rakudo compile yeah, but other workloads should be helped more
samcv gonna set it back then run prove and check the logs see which it hangs
SUCC 1 h_pos -2134653648 hgraphs -2122111808
XD
succ means success 04:50
diakopter heh.
samcv well. at least it passed one of the checks. i numbered them
diakopter that's a very long string
do you have that much negative memory?
samcv well it's alternating uhm 04:51
between two things, one where it succeeds and one which fails
alternating
the failing one is the odd numbers
maybe it's just a print error idfk 04:52
or maybe it's the same one and only failing in part of the print's
oh i forgot to put anything after the fprintf (variables) 04:53
wtf it looks like the values get corrupted? 04:55
loopy h_pos 1 hgraphs 10 04:56
loopy h_pos 3 hgraphs 10
SUCC 1 h_pos 530666800 hgraphs 543208640
exiting loop succ 1.1 h_pos 3 hgraphs 10
fprintf(stderr, "SUCC 1 h_pos %i hgraphs %i\n", h_pos, hgraphs); # corruption
then below itc
fprintf(stderr, "exiting loop succ 1.1 h_pos %i hgraphs %i\n", h_pos, hgraphs)
and prints the correct values
04:57 MasterDuke joined
diakopter MasterDuke: true, so then profile those other workloads.. :) 04:57
samcv time to try making them all
%li
diakopter lol at "idfk" 04:58
MasterDuke for the heck of it i'll perf record a spectest run, see what that looks like
diakopter well that'll be a TON of invocations 04:59
process executions
MasterDuke if anyone is interested, here's a perf report of building rakudo with target=parse. gist.github.com/MasterDuke17/aff44...6d1d0ba537 05:02
samcv ok diakopter it does fine on all the things i've tested it with 05:24
but it's still hanging
diakopter samcv: push the changes? 05:26
samcv diakopter, pushed 05:32
diakopter "h_pos + 1 < hgraphs - 1" 05:34
I think you meant <=
samcv that's what it says? 05:36
line 245? or elsewhere?
diakopter yes, 245, I think you meant to write <= instead of <
samcv yeah that seems so then 05:37
diakopter I think the check needs to be "h_pos + 1 <= hgraphs - 1" or "h_pos + 1 < hgraphs"
I still can't conceive of why that would make something hang though XD 05:38
samcv yeah
in the test it ust goes over the same string segment again and again
some 10 long string 05:39
i tested all positions of abcdefg.. for hgraphs 10
but yeah shouldn't hang
should give me some message like, confused at ...
diakopter so did that fix it 05:42
samcv i really doubt it 05:43
MasterDuke fyi, just doing an nqp::index() in a loop gives ~40% time in MVM_string_substrings_equal_nocheck and ~10% time in MVM_string_index
samcv yeah no
hangggg 05:44
MasterDuke which isn't all that surprising
samcv it tests the same string again and again
MasterDuke anyway, i'm off to sleep, later all... 05:45
samcv kk
diakopter try removing the return -1 line 05:47
to let it fall back to the other thing
sleep& 05:48
samcv k night
i guess i could attempt to print out the needle and the haystack as their unicode cp's... 05:56
unless there's an easy way to dump a string?
06:31 pyrimidi_ joined 07:13 domidumont joined 07:20 domidumont joined 08:23 zakharyas joined
timotimo the easiest thing is to just encode it to utf8, print that, and free the string again 09:45
09:52 brrt joined 10:04 domidumont joined 10:43 brrt joined 13:02 pyrimidine joined 13:08 pyrimidine joined 13:39 pyrimidine joined 14:06 pyrimidine joined 14:22 ggoebel joined 14:57 pyrimidine joined 15:14 pyrimidine joined 15:32 dalek joined 15:34 synopsebot6 joined 15:52 brrt joined
brrt ohai #moarvm 15:54
timotimo greetings brrt 15:55
brrt \o timotimo
how are things
timotimo they're okay 15:56
yesterday i tried some simple stretches and yoga poses some random artist on twitter recommended if you regularly suffer from back pain
i don't really have back pain often, but i thought i'd give it a go anyway
now i have muscle soreness all over my back and shoulders %)
i'm looking forward to getting a little involved with the "easier" parts of the jit soon-ish :3 15:57
i'd like to hear how you are today, too. i'm going to be AFK for ~half an hour first, though
brrt ehm, i'm kind of busy, as usual 15:58
unfortunately, i might add :-P
i figured out that i need to add a 'barrier stack 15:59
or rather
barrier queue to the JIT
jnthn has completed his very busy part of the week and looks forward to digging into Moar/Perl 6 stuff again tomorrow. :)
brokenchicken \o/
brrt i.e. a list of all those instructions that cause a flush of (a subset of) registers
brrt envies jnthn
16:08 ggoebel joined 16:27 brrt joined
brrt so, know how i've been talking about solving the 3-operand to two-operand conversion process 16:36
that also needs to be done for the indirect ops 16:37
to give you an example: suppose I have something intended as:
rcx = rdx + [rcx + 5];
well, i can't very well overwrite rcx with rdx if i expect that to work 16:38
timotimo good point 16:39
i hadn't considered indirect addressing :S
brrt utlimately, the limitation is that linear scan operates on live ranges, not on tiles, and so it can't enforce per-tile constraints very well 16:40
timotimo let's add another pass! :) 16:41
brrt hmmm
yes, but
where 16:42
before, or after linear scan
timotimo .o( why not both? )
wasn't actually a serious suggestion
brrt i know :-)
the 'lets' add another pass' probably has some truth to it, but i'm not sure how
so, to recap 16:44
this is only relevant to binary ops
this is only relevant when the output register would overwrite the *second* input register 16:45
or actually, 'a secondary', because there may be more than one 16:46
17:23 domidumont joined 18:20 domidumont joined 18:38 domidumont joined 18:50 FROGGS joined 19:38 domidumont joined 20:25 zakharyas joined 20:26 pyrimidine joined 21:28 pyrimidine joined
MasterDuke is there a fast way to know/find out in moar if a string is all ascii chars? 23:12
timotimo only whatever i do in ops.c or the encoding .c files i think? 23:15
MasterDuke ops.c is ~10k lines, what should i search for? 23:18

Powered by Cro and the Raku® Programming Language.

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

Thank you!