github.com/moarvm/moarvm | IRC logs at irclog.perlgeek.de/moarvm/today
Set by moderator on 29 April 2018.
Geth MoarVM/release/2018.04.1: a08c4fd548 | (Samantha McVey)++ | 2 files
Enable TravisCI and Appveyor for 2018.04.1 release branch
00:33
00:43 travis-ci joined
travis-ci MoarVM build failed. Samantha McVey 'Enable TravisCI and Appveyor for 2018.04.1 release branch' 00:43
travis-ci.org/MoarVM/MoarVM/builds/372839860 github.com/MoarVM/MoarVM/compare/4...8c4fd54842
00:43 travis-ci left
Geth MoarVM/release/2018.04.1: ff5f9f1f38 | (Samantha McVey)++ | 2 files
Test 2018.04 version of nqp on CI
00:45
00:51 travis-ci joined
travis-ci MoarVM build errored. Samantha McVey 'Test 2018.04 version of nqp on CI' 00:51
travis-ci.org/MoarVM/MoarVM/builds/372841632 github.com/MoarVM/MoarVM/compare/a...5f9f1f38b1
00:51 travis-ci left
Geth MoarVM/release/2018.04.1: cf4a108d78 | (Samantha McVey)++ | 2 files
Test 2018.04 version of nqp on CI
00:53
01:00 Kaiepi joined 01:09 travis-ci joined
travis-ci MoarVM build passed. Samantha McVey 'Test 2018.04 version of nqp on CI' 01:09
travis-ci.org/MoarVM/MoarVM/builds/372842934 github.com/MoarVM/MoarVM/compare/f...4a108d785b
01:09 travis-ci left
Geth MoarVM/release/2018.04.1: 35e6693411 | (Samantha McVey)++ | docs/ChangeLog
Update ChangeLog for release
01:11
MoarVM/refs/tags/2018.04.1: e00dc3e53b | (Samantha McVey)++ | VERSION
Release 2018.04.1
01:16
01:38 FROGGS__ joined 01:56 ilbot3 joined
moderator github.com/moarvm/moarvm | IRC logs at irclog.perlgeek.de/moarvm/today
Geth MoarVM/master: 4 commits pushed by (Zoffix Znet)++, (Samantha McVey)++, (Aleks-Daniel Jakimenko-Aleksejev)++ 03:42
04:28 AlexDaniel joined
Geth MoarVM/release/2018.04.1: e00dc3e53b | (Samantha McVey)++ | VERSION
Release 2018.04.1
05:06
MoarVM: e00dc3e53b | (Samantha McVey)++ | VERSION
Release 2018.04.1
05:08
MoarVM: af60a86106 | (Aleks-Daniel Jakimenko-Aleksejev)++ | VERSION
Merge branch 'release/2018.04.1'
06:16 robertle joined 07:23 zakharyas joined 07:38 domidumont joined 07:44 domidumont joined 08:10 domidumont joined 08:18 zakharyas joined 10:01 domidumont joined 11:11 Ven`` joined 12:39 zakharyas joined 12:58 AlexDaniel joined 13:30 AlexDaniel joined 13:38 brrt joined
samcv jnthn: thoughts on naming on this github.com/samcv/MoarVM/blob/sipha...ps.c#L2893 13:41
specifically MVM_MAYBE_TO_LITTLE_ENDIAN_32. description of what it does is in the linked comment
brrt do we ever export the hash code? 13:42
do we make any guarantees whatsoever about it?
samcv brrt: no we don't 13:44
also i'd like to switch to 64 bit hashes instead of 32 bit, not sure if it's just as easy as changing the data type in uthash.h 13:46
i recall it using &'s to decide which bucket to put things in
brrt: that MVM_MAYBE_TO_LITTLE_ENDIAN is also used in the hashing implementation file as well, just to ensure we can show that things aren't broken on different endianess. 13:49
brrt why 64 bit hashes? 13:53
jnthn samcv: Why do we need the MVM_MAYBE_TO_LITTLE_ENDIAN_32 if we already guard it with #if defined(MVM_HASH_FORCE_LITTLE_ENDIAN? 13:55
brrt i don't know how, but i can no longer install moar using the mingw toolchain coming with strawbetrry perl
samcv jnthn: it's used many places not just for the hash secret 13:56
for the 32 bit codepoints, and also in the hashing code
jnthn ah, ok
I guess the name is OK
MVM_TO_LITTLE_ENDIAN_IF_FORCED is more explicit
samcv ah that sounds good 13:57
jnthn Well, more explicitly links it with MVM_HASH_FORCE_LITTLE_ENDIAN at least :)
13:58 Ven`` joined
samcv brrt: well siphash creates a 64 bit hash 14:13
brrt that seems valid enough 14:30
14:56 zakharyas joined 15:47 Ven`` joined 15:50 robertle joined
AlexDaniel hello wonderful #moarvm people! 16:07
Please take a look at this: irclog.perlgeek.de/perl6/2018-04-30#i_16114032
timotimo that looks like a patch we could totally apply right now, but we may want to actually CAS the result at the end so we know to delete the stuff we just created if someone else beat us to it 16:08
robertle I also wonder why this gets called twice, is that perhaps an indication of a bigger problem? 16:09
timotimo it's a benign race in case the bytecode only needs to be validated, not changed, and with your patch it's a benign race that just leaks a little bit of memory in case endian swapping is needed 16:11
16:44 Ven`` joined 17:31 zakharyas joined 18:05 Ven`` joined 18:33 Ven`` joined 19:40 evalable6 joined 20:15 brrt joined
brrt . 20:40
timotimo yo brrt 20:47
would the spesh slot value/address types for emit_c_call be super easy to do? 20:48
brrt yes 20:55
very 20:56
21:14 robertle joined
timotimo oh, i didn't look back for an answer 21:25
i'd appreciate if we had it :)
but really only for the three ops i recently added :D
21:30 brrt joined
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/04/30/...es-galore/ 21:43
brrt lizmat++ 21:51
21:52 jnthn joined
brrt unless objections, i'm going to hit the merge button on the spesh-debug-limit branch 23:01
23:22 notable6 joined, bloatable6 joined, evalable6 joined, nativecallable6 joined, benchable6 joined, committable6 joined, greppable6 joined, coverable6 joined, quotable6 joined, releasable6 joined, reportable6 joined, squashable6 joined, bisectable6 joined, unicodable6 joined, statisfiable6 joined 23:41 brrt joined