|
02:33
colomon joined
02:47
ilbot3 joined
02:50
colomon joined
03:12
vendethiel joined
04:15
colomon joined
06:53
colomon_ joined
07:02
domidumont joined
07:06
domidumont joined
07:24
FROGGS joined
07:54
zakharyas joined
09:48
leont joined
12:54
virtualsue joined
13:06
brrt joined
13:16
pochi joined
13:18
stmuk__ joined
13:20
Peter_R joined
|
|||
| hoelzro | o/ #moarvm | 14:08 | |
| has anyone run into a situation where turning tracing on only results in a handful of Op 159 (getcode) prints? | 14:09 | ||
| I'm tracing "my $sum = 0; for ^10 { $sum += $_ }" and all I see are two Op 159 | |||
| JimmyZ | m: my $sum = 0; for ^10 { $sum += $_ } | 14:13 | |
| camelia | ( no output ) | ||
| JimmyZ | spesh log ? or trace macro in interp.c? | 14:14 | |
| hoelzro | the latter | 14:28 | |
| MVM_TRACE | |||
| brrt | no idea... | 14:31 | |
| JimmyZ | hoelzro: I know | 14:32 | |
| you should turn off CGOTO | 14:33 | ||
| hoelzro | what's that? | ||
| JimmyZ | how do you enble TRACE? | 14:35 | |
| make tracing? | 14:36 | ||
| hoelzro | by stupidingly editing interp.c by hand =/ | ||
| JimmyZ | you should type `make tracing` | ||
| hoelzro tries | 14:37 | ||
| there we go | 14:39 | ||
| thanks JimmyZ! | |||
|
15:25
brrt joined
|
|||
| hoelzro | diakopter: yesterday you mentioned a branch that had some deserialization fixes; which branch is it? | 16:00 | |
| diakopter | hoelzro: cache_sc_idx | 16:32 | |
| hoelzro | diakopter: thanks! | ||
| diakopter | but it fails on windows last time we checked | ||
| hoelzro | oh, that's fine with me =) I just want to see if it fixes a bug I'm battling | ||
| diakopter | it just fixes caching during bootstrapping and makes spectesting loads faster | 16:33 | |
| hoelzro: feel free to merge master into it | 16:34 | ||
| nm I did it | 16:35 | ||
| dalek | Heuristic branch merge: pushed 140 commits to MoarVM/cache_sc_idx by diakopter | ||
| diakopter | hoelzro: see updated github.com/MoarVM/MoarVM/compare/cache_sc_idx | 16:36 | |
| hoelzro | thanks =) | ||
|
17:09
Peter_R joined
17:19
Peter_R joined
17:21
domidumont joined
17:26
camelia joined
17:36
FROGGS joined
|
|||
| diakopter | hoelzro: anything? | 18:08 | |
|
18:12
ilbot3 joined
|
|||
| lizmat | www.dagolden.com/index.php/2589/per...l-enemies/ # food for thought | 18:20 | |
|
18:51
vendethiel joined
|
|||
| hoelzro | diakopter: my bug sadly still occurs =( | 19:07 | |
| if anyone's curious; rt.perl.org/Ticket/Display.html?id=127089 | 19:11 | ||
| nwc10 | gosh. I'd forgotten how fast a Raspberry Pi v2 is | 19:53 | |
| it's about half the speed of x86_64 with ASAN | |||
| (on one core)(ie the setting) | |||
| diakopter should try it on the Zero | 20:03 | ||
| hoelzro: that's interesting | 20:04 | ||
| nwc10 | "might take longer". And I think you'll need to configure (more) swap as I think it peaked at over .75G of RAM | ||
| hoelzro | diakopter: what's more interesting is calling .map in item context fixes it | 20:05 | |
| hoelzro has to add that to the ticket | |||
| diakopter | m: gist.github.com/diakopter/7981dc5b1d0ff7598e53 | 20:08 | |
| camelia | rakudo-moar ed6ec7: OUTPUTĀ«beforeā¤afterā¤Ā» | ||
| diakopter | hoelzro: what am I doing wrong? ^ | ||
| hoelzro | diakopter: the sub has to be from another module | ||
| diakopter | oh | 20:10 | |
| hoelzro | I'll add that to the ticket too | 20:11 | |
|
20:38
leont joined
21:19
domidumont joined
21:26
thzre joined
|
|||
| hoelzro | the plot thickens: I think it's just map | 21:51 | |
| hmm, note() does it too | 22:00 | ||
| diakopter | hoelzro: sounds like it's sink context | 22:32 | |
| hoelzro: probably TimToady could diagnose it | 22:33 | ||
| hoelzro | diakopter: sink context and list | 22:34 | |
| it also depends on what's in the block | |||
| <...>.map({ $_ }) is fine | |||
| <...>.map({ $_ ~ 'suffix' }) is not | |||
| neither is .map({ $_ + 1 }) | |||
| diakopter | o_O | ||
| hoelzro | using note() and no map is also broken | 22:35 | |
| it's...a very interesting bug =) | |||
| diakopter | it might be something else also, but I feel strongly that it's an overeager optimization from a missing want or WANT | 22:36 | |
| hoelzro | it might be bisect-o-clock when I get home | 22:47 | |
| see if this ever worked | 22:48 | ||
| hoelzro relies too heavily on bisect for P6 work | |||
| lizmat decides to stash the setup-repositories work | 23:03 | ||
| it only gave about 8 msecs improvement, at the expense of not so nice code | 23:04 | ||
| :-( | |||
| hoelzro | lizmat++ # trying things | 23:05 | |
| lizmat | I thought the same code was also run at bare startup, but it wasn't | ||
| 8 msecs at a bare startup, I would quickly take | 23:06 | ||
| timotimo | hmm | 23:10 | |
| i would also take that, yeah | |||
| given a startup is currently like 100 msecs | |||
| hm, or 90 | 23:11 | ||
| lizmat | more like 110 for me, atm | 23:14 | |
| timotimo | mhm | ||
| diakopter | time perl -e 'print 1' # 2 ms | 23:15 | |
| lizmat | diakopter: try that with -MMoose :-) | 23:16 | |
| diakopter | yeah, I submitted a pull request to Moose that halves its startup time, but it was rejected for mostly bad reasons | 23:17 | |
| timotimo | i'm glad it got rejected, so perl6 looks better in comparison ... :P | 23:18 | |
| diakopter | well :P | ||
| timotimo | i'm hopeful we'll at least halve the bare startup time over the next half year | 23:20 | |
| diakopter | I'm quite hopeful it's doable | ||
| timotimo | and it's likely that our ram usage on bare startup can also go down a good chunk | 23:21 | |
| i don't actually know how we'll achieve that, though | |||
| diakopter | what's the ram usage now | ||
| timotimo | timo@schmetterling /p/a/ibm> time perl6 -e '' | 23:27 | |
| 0.08user 0.01system 0:00.09elapsed 100%CPU (0avgtext+0avgdata 65144maxresident)k | |||
| in other words "very bad" :) | |||
|
23:29
pochi_ joined
|
|||
| timotimo | i meant to build a separate setting that has a few classes missing to see how the ram usage does down from that | 23:30 | |
| to get a little measurement of how good lazy deserialization works | |||