🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
00:03 reportable6 left, reportable6 joined
Geth rakudo: tony-o++ created pull request #4399:
Fix for #4394
00:27 linkable6 left 00:28 linkable6 joined
tonyo can someone not on m1 macos arch verify that make test passes with that PR ^ ? 02:23
03:13 linkable6 left, evalable6 left
raydiak tonyo: looks good on x64 linux 03:13
03:15 evalable6 joined 03:16 linkable6 joined 05:00 evalable6 left, linkable6 left 05:01 evalable6 joined 05:03 linkable6 joined 06:02 reportable6 left 06:05 reportable6 joined 07:21 japhb left 08:41 reportable6 left, squashable6 left, linkable6 left, greppable6 left, coverable6 left, unicodable6 left, statisfiable6 left, quotable6 left, evalable6 left, bloatable6 left, shareable6 left, bisectable6 left, notable6 left, releasable6 left, benchable6 left, nativecallable6 left, tellable6 left, committable6 left, sourceable6 left, evalable6 joined, unicodable6 joined, releasable6 joined 08:42 linkable6 joined, nativecallable6 joined, notable6 joined, squashable6 joined, statisfiable6 joined, sourceable6 joined, greppable6 joined 08:43 committable6 joined, coverable6 joined, tellable6 joined, quotable6 joined, benchable6 joined, reportable6 joined, bloatable6 joined, bisectable6 joined, shareable6 joined 09:44 Kaiepi left
Geth rakudo: usev6++ created pull request #4400:
[JVM] Remove special case for native arrays on JVM
09:48 Kaiepi joined 09:57 sena_kun joined
lizmat m: use Test; is-deeply Mu, Mu 10:40
camelia Type check failed in binding to parameter '<anon>'; expected Any but got Mu (Mu)
in sub _is_deeply at /home/camelia/rakudo-m-inst-1/share/perl6/core/sources/2D4A7CA10695CD2B374573413D261A660E282E2A (Test) line 694
in sub is-deeply at /home/cam…
lizmat meh
Geth rakudo: 28ff12a8b8 | (Christian Bartolomäus)++ | 5 files
[JVM] Remove special case for native arrays on JVM

It's now possible to use native arrays in the setting. Fixed with github.com/Raku/nqp/pull/720 and
11:42 linkable6 left, evalable6 left, linkable6 joined 11:43 evalable6 joined 12:02 reportable6 left 12:05 reportable6 joined 13:09 Altai-man joined 13:16 b2gills left 13:18 sena_kun left 13:31 b2gills joined
bartolin (problem on JVM with large nqp::hash) I dropped to bed yesterday evening, but I've created an issue now: github.com/Raku/nqp/issues/727 14:20
I don't understand the details, but AFAICS splitting the data into smaller hashes wouldn't help. 14:22
14:37 evalable6 left, linkable6 left 14:40 evalable6 joined, linkable6 joined
bartolin adding entries from a file to a hash works fine even for large numbers 14:49
raydiak I didn't just mean multiple hashes, but also appending each one in a separate routine, each of which would get called in series to build up the final hash. no idea if it would help or not though, and it'd be an ugly hack
bartolin I'm not sure, but I'd think something like that could work 14:51
raydiak and you mentioned that larges hashes work in rakudo-j, just not nqp-j? that's...interesting... 14:53
bartolin yes, I was surprised by that, too. 14:56
I just tried with a hash with 25_000 entries. It wasn't really fast, but it didn't blow up. With 50_000 entries I got a 'Class file too large!', though 15:01
bartolin is adding that to github.com/Raku/nqp/issues/727 15:05
raydiak hm 15:07
bartolin heh, moarvm is also unhappy with 50_000 entries (Too many arguments (100001) in flattening array, only 65535 allowed.)
I mean, rakudo-m 15:08
nine Yeah, you know....enough is enough :D
raydiak arguments? is it implying that the list is passed as a bunch of arguments to a function somewhere? 15:10
either way, fwiw, we're only talking about 1449 hash entries, 23396 bytes for the entires themselves as written in the code. no idea how that translates by the time jvm sees it though 15:15
s/23396/23369/ 15:16
bartolin btw, it's really not important, since it's only the jvm backend. Please don't waste your time there ;) I mostly wanted to document what I've seen so far. 15:18
raydiak of course. this is more academic curiosity. I already just added the if jvm/!jvm bits and the old implementation into the PR 15:20
bartolin yeah, saw that. raydiak++ (I don't feel competent to review that change for Moar, though.) 15:23
raydiak no worries, I appreciate the time and advice you've given on it 15:26
tonyo nine: added a comment to that PR, the disassembler helps see the problem a lot more clearly 16:08
but also now i have .o objects everywhere .. like .DS_Stores 16:09
18:02 reportable6 left 18:04 reportable6 joined
lizmat m: dd Mu eqv Mu 19:50
camelia Type check failed in binding to parameter '<anon>'; expected Any but got Mu (Mu)
in block <unit> at <tmp> line 1
20:48 squashable6 left 20:50 squashable6 joined 21:41 hankache joined 22:02 hankache left