github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
01:10 sena_kun joined 01:11 rbt left, rbt joined, Altai-man_ left 02:20 rbt left 03:09 Altai-man_ joined 03:12 sena_kun left 05:10 sena_kun joined 05:11 Altai-man_ left 06:00 Altai-man_ joined 06:03 sena_kun left
nwc10 good *, #moarvm 06:33
07:58 zakharyas joined 08:01 sena_kun joined 08:03 Altai-man_ left 08:05 patrickb joined 09:07 squashable6 left, squashable6 joined 09:30 zakharyas left 09:32 zakharyas joined 09:34 nwc10_ joined, nwc10 left 09:35 nwc10_ is now known as nwc10 09:37 zakharyas left 09:40 zakharyas joined 10:01 Altai-man_ joined 10:03 sena_kun left
Geth MoarVM: 23dfde978c | (Patrick Böker)++ | build/setup.pm
Explicitly state we use gnu99 on GCC

This fixes building on older versions of GCC (< 5.0) that default to gnu89. Using `-std=c99` instead of `-std=gnu99` breaks the build of libuv because some pthread declarations are then left out. So it seems sensible to go with the GCC default of using the `gnu99` variant instead of the `c99` variant. ... (7 more lines)
11:09
MoarVM: bf6af07dee | (Patrick Böker)++ (committed using GitHub Web editor) | build/setup.pm
Merge pull request #1318 from patrickbkr/fix-gcc4

Explicitly state we use gnu99 on GCC
11:14 zakharyas left
lizmat I guess we need a bump now? 11:23
Altai-man_ lizmat, can try to see if the issue resolved.
lizmat ok, will do 11:24
12:01 sena_kun joined 12:03 Altai-man_ left 12:08 brrt joined
dogbert17 oh, it's brrt, hello there 12:29
nwc10 good *, brrt 12:30
good *, dogbert17
dogbert17 hello nwc10, what's the state of your beer fridge? 12:31
dogbert17 temperature is slowly creeping towards 30 degrees centigrade here 12:32
dogbert17 could use a beer but don't have any atm :( 12:33
12:33 timotimo left 12:35 timotimo joined
nwc10 it has beer. Although I realised last night that most of the beer was in the crate near it, and not enough in the fridge itself 12:35
timotimo hewwo bwwt 12:36
dogbert17 so it's not very cold then :)
hello timotimo
timotimo temperature is like 28 centigrade, will be 30 the next two days
dogbert17 Yeah, just picked up a cheap fan which I can use if things get out of hand 12:38
my apartment is on the top floor and can get uncomfartably hot during the summer 12:40
timotimo oh no
top floor is terrible
dogbert17 almost 30 degrees inside atm
timotimo disgusting
dogbert17 yes indeed
timotimo i am maybe too sensitive to heat, but damn, it just kills me
dogbert17 I don't like it either 12:41
I mean what's wrong with say 22-24 degrees 12:42
timotimo aye 12:44
brrt hi nwc10, dogbeter, timotimo 12:50
*dogbert17
dogbert17 brrt, is it extremely hot where you live as well today? 12:55
lizmat: stresstest looks fine 12:57
13:13 AlexDaniel joined, AlexDaniel left, AlexDaniel joined 13:15 zakharyas joined 13:18 greppable6 left 13:19 sena_kun left, sena_kun joined 13:21 zakharyas left, zakharyas1 joined 14:01 Altai-man_ joined 14:03 sena_kun left 14:14 notable6 left, coverable6 left, evalable6 left, benchable6 left, statisfiable6 left, nativecallable6 left, reportable6 left, quotable6 left, bloatable6 left, linkable6 left, committable6 left, tellable6 left, sourceable6 left, releasable6 left, shareable6 left, bisectable6 left, squashable6 left 14:19 coverable6 joined, reportable6 joined, quotable6 joined 14:20 notable6 joined, bisectable6 joined, unicodable6 joined, sourceable6 joined, benchable6 joined, evalable6 joined, shareable6 joined, bloatable6 joined 14:21 greppable6 joined, squashable6 joined, releasable6 joined, linkable6 joined, nativecallable6 joined, statisfiable6 joined 14:22 committable6 joined, tellable6 joined 14:52 brrt left 16:02 sena_kun joined 16:03 Altai-man_ left 16:48 sivoais left 17:00 sivoais joined 17:01 zakharyas1 left
lizmat dogbert17: any opinions about performance of the match-refactor branch in stresstest ? 17:12
17:13 MasterDuke left 17:21 brrt joined
brrt dogbert17: it's nearly 30 degrees C, which is warm, but not nearly as warm as last year 17:27
which still might happen, midn
18:01 Altai-man_ joined 18:03 sena_kun left 18:50 brrt left 19:02 brrt joined
brrt on the topic of bombshells.... www.perl.com/article/announcing-perl-7/ 19:02
[Coke] I don't know about "bombshell", I mean we pretty much set them up for that. 19:05
... that said, first I heard was post-sawyer presention
brrt hehe true 19:07
I read it and I *think* it is all sane and good 19:08
preferable to sticking with perl5.xxxxx
nwc10 it was a very well kept secret. In that, it remained an unknown unknown
19:30 brrt left 19:41 sivoais left 19:44 sivoais joined 20:01 sena_kun joined 20:03 Altai-man_ left
timotimo i wonder what the next step for the jit currently is 20:21
nwc10 ARM. Ready for next year's macs? 20:22
<ducks> 20:23
jnthn Teach the expression JIT about dispatch :P 20:24
20:29 sivoais left
timotimo wouldn't spesh be code-genning most stuff so the jit would just see guards and simple ops? 20:40
20:41 sivoais joined 20:42 MasterDuke joined
jnthn Only in the case the callsite was linked... 20:42
Or rewritable
Dunno if we'll have some that ain't 20:43
(other than unlinked, of course)
That said, one option for handling a dispatch that was never linked is to always turn it into a deopt
That way, you get rid of unreachable codepaths
uh, unreached
timotimo so we cut every instruction after the unconditional deopt and end up with a shorter frame instead? 20:51
jnthn Yeah 21:01
22:01 Altai-man_ joined 22:03 sena_kun left 22:25 patrickb left 23:04 squashable6 left 23:07 squashable6 joined