00:07 reportable6 left 00:08 reportable6 joined 01:08 linkable6 left, evalable6 left 01:10 linkable6 joined, evalable6 joined 01:40 MasterDuke joined 02:00 vrurg joined 02:04 vrurg left 04:01 vrurg joined 04:06 vrurg left 04:25 vrurg joined 05:11 vrurg left, vrurg joined 05:12 vrurg left, vrurg joined 05:17 vrurg left 06:07 reportable6 left 06:09 reportable6 joined 06:31 nebuchadnezzar left 08:18 nebuchadnezzar joined
Nicholas [* GOOD *] 08:20
08:26 sena_kun joined 09:23 nine left, camelia left, nine joined 09:29 camelia joined 10:09 Altai-man joined 10:30 frost joined
lizmat m: my uint $a = 2; my uint $b = -1; say $a; say $b; say $a > $b # this feels... wrong ? nine 11:51
camelia 2
18446744073709551615
True
lizmat wrong in the sense that apparently the uint is interpreted as an int in the expression >
jnthn Looks wrong to me, yes 11:52
11:53 vrurg joined
lizmat looks to be wrong at nqp level: 11:55
m: my uint $a = 2; my uint $b = -1; say $a; say $b; use nqp; say nqp::isgt_i($a,$b)
camelia 2
18446744073709551615
1
11:57 vrurg left
lizmat I probably should make that a MoarVM issue, right ? 12:00
12:07 reportable6 left 12:08 reportable6 joined
nine The nqp::isgt_i($a,$b) result here is correct. You are asking it to interpret the values as (signed) ints. What you want is nqp::isgt_u which doesn't exist yet (but is trivial to implement) 12:13
12:15 frost left
lizmat I'm afraid I'm not tall enough to reach that low hanging fruit 12:22
I will need something to lift me for that
do I just start adding ops to interp.c ? 12:29
jnthn lizmat: 1. Add to src/core/oplist, see the instructions at the top for where to put it. 2) run tools/update_ops.raku. 3) add implementtation in interp.c. 4) Add a mapping in NQP (src/vm/moar/QAST/QASTOperationsMAST.nqp or similar) 12:34
(All names from memory, so probably at least one is wrong :))
lizmat thanks, I'll have a stab at it 12:37
m: use nqp; dd nqp::uname 12:43
camelia BOOTStrArray("Linux", "5.3.18-59.10-default", "#1 SMP Fri Jun 25 12:36:56 UTC 2021 (6856d31)", "x86_64")
lizmat interesting
wonder what that does on Win
ok, will look at this in more depth once I get the next release of App::Rak out 12:48
which should be soon :-)
meanwhile, I'll use 0x7fff_ffff_ffff_ffff for the largest positive value :-) 12:49
[Coke] * :watch: [Sep-23] - Inna Bembenok's last day! 12:51
DAMMITR
BOOTStrArray("Windows_NT", "10.0.19044", "Windows 10 Enterprise", "x86_64") 12:52
Do I have like 2 different cut and paste buffers? so annoying.
lizmat: ^^
lizmat [Coke]++ 12:53
patrickb o/ 13:50
I issued a GitHub App installation request for the MoarVM project. Can someone with permissions to do so grant that request? That app is part of the Rakudo CI bot I'm working on. I don't plan to change the current CI workflow yet, but having the permission allows preparing things and running noninvasive tests. 13:57
lizmat patrickb: fwiw, I've just okd this for rakudo / raku 14:04
patrickb thanks! 14:11
lizmat: Can you somehow give the app access to the rakudo/nqp repos only? 14:13
lizmat you mean rakudo/rakudo and raku/nqp ? 14:14
patrickb: afaics, it is already limited to rakudo/rakudo and raku/nqp 14:18
[Coke] I read his send as rakudo&nqp, aye. 14:23
patrickb OK. Then it's already fine in those two repos / projects. 14:31
So MoarVM/MoarVM is left. 14:32
lizmat patrickb: I don't have sufficient rights for that, I think only jnthn does 14:34
14:44 [Coke]_ joined 14:45 [Coke] left
jnthn patrickb: Wehre should I find the installation request? I don't see it 14:58
15:05 [Coke]_ is now known as [Coke] 15:38 Altai-man left 15:39 Altai-man joined
patrickb I don't know. lizmat: Where have you seen the request? 16:09
lizmat in my mail box
jnthn I saw requests for Raku and rakudo orgs in my mailbox, but not a MoarVM one
lizmat Subject: [Github] RakudoCIBot is requesting updated permissions 16:10
patrickb Ah, then I'll see if I can re-do the MoarVM request. 16:13
Maybe I have just re done a request... 16:14
16:22 [Coke]_ joined
jnthn Think it's approved 16:22
16:24 [Coke] left 16:25 [Coke]_ is now known as [Coke] 16:29 [Coke]_ joined
patrickb Looks good on my side as well. Thanks! 16:29
16:31 [Coke] left 16:32 [Coke]_ is now known as [Coke] 17:10 Altai-man left
jnthn Welcome :) 17:14
17:26 japhb left 17:33 japhb joined 17:59 [Coke]_ joined 18:00 [Coke] left 18:05 [Coke] joined 18:06 reportable6 left 18:07 [Coke]_ left, reportable6 joined
MasterDuke interesting, for all the nqp::is(lt|le|gt|ge)_ there are _i, _I, _n, and _s variants. but for nqp::is(eq|ne)_ there's only _i, _I, _n 18:13
tellable6 2022-09-22T06:51:08Z #moarvm <nine> MasterDuke: yes it is
MasterDuke doh. ignore that 18:17
18:19 [Coke]_ joined 18:22 [Coke] left 18:24 [Coke]_ left, [Coke] joined 18:35 [Coke]_ joined 18:38 [Coke] left 18:40 [Coke] joined 18:43 [Coke]_ left 18:45 [Coke]_ joined 18:48 [Coke] left 19:03 [Coke]_ is now known as [Coke] 19:36 [Coke]_ joined 19:37 [Coke] left 19:43 [Coke] joined 19:45 [Coke]_ left 20:18 nebuchadnezzar left 21:14 [Coke] left 21:18 [Coke] joined 21:22 sena_kun left 22:52 [Coke] left 22:55 [Coke] joined 23:00 [Coke] left, [Coke] joined