MasterDuke i just jitted param_op_(i|o), and now INTERPOLATE doesn't bail according to the jit log, but it still doesn't show as jitted in a profile 01:03
timotimo: still around? 01:04
01:08 mst joined 01:52 ilbot3 joined
Geth MoarVM: MasterDuke17++ created pull request #660:
JIT param_op_(i|o)
02:45
02:46 MasterDuke joined 03:35 travis-ci joined
travis-ci MoarVM build failed. Samantha McVey 'Merge pull request #652 from MasterDuke17/add_error_string_to_failed_to_resolve_exception 03:35
travis-ci.org/MoarVM/MoarVM/builds/267683413 github.com/MoarVM/MoarVM/compare/1...b339553405
03:35 travis-ci left
MasterDuke ^^^ looks like it just got caught in between the nqp/rakudo bumps 03:40
03:45 bloatable6 joined, benchable6 joined, bisectable6 joined, coverable6 joined, committable6 joined, evalable6 joined, unicodable6 joined, releasable6 joined, statisfiable6 joined 05:38 travis-ci joined
travis-ci MoarVM build failed. Samantha McVey 'Merge pull request #648 from samcv/knuth_morris_pratt 05:38
travis-ci.org/MoarVM/MoarVM/builds/267766807 github.com/MoarVM/MoarVM/compare/7...1a25634797
05:38 travis-ci left 06:00 brrt joined
Geth MoarVM/jit-legacy-cleanup: 786a756252 | (Bart Wiegmans)++ | 7 files
Wrap dasm_State in MVMJitComiler structure

This changes function signatures, but is not very exciting otherwise.
06:35
MoarVM/jit-legacy-cleanup: a6b7a27e98 | (Bart Wiegmans)++ | 2 files
Take compiler initialization and destruction out

This cleans up error handling when we can't allocate an executable page.
06:47 domidumont joined
timotimo MasterDuke: the op ops are for when the parameter is optional, so the .i64 value could just be bogus 06:47
probably zeroed out memory?
i'm on my phone, so can't look at the code
oh 06:48
it has to kbow whether to jump or not
which it does if the parameter was not passed i believe 06:49
06:53 domidumont joined
Geth MoarVM/jit_nativecall: 7109ee5229 | (Stefan Seifert)++ | 5 files
JIT native calling functions which return integers
07:01
07:17 brrt joined
brrt yeah, the optional ones are trickier to JIT 07:18
did anyone read the 'writing portable makefiles' thingy 07:31
now i know how to make rules that work on multiple makes
that one time i learned something from hacker news 07:33
O M G it actually 07:38
moritz my problem is that whenever I need a slightly non-trivial Makefile feature, it turns out to be non-standard 07:48
brrt oh, thats true 07:50
but
when do you actually need them?
or, what features do you need?
moritz typically stuff like running a command, and capturing the output in a variable 07:57
or anything else that is more than simply invoking a command, and is typically trivial to do in shell, but very nasty or impossible in cross-platform Makefile 07:58
brrt right, that kind of think 08:02
*thing
that's hard yes
08:18 zakharyas joined 08:37 leont joined 09:12 lizmat joined 09:58 AlexDaniel joined 11:28 domidumont joined 11:50 zakharyas joined 13:02 domidumont1 joined 13:26 leont joined 13:33 domidumont joined 14:14 leont joined 14:19 geekosaur joined 14:52 mst joined 15:13 brrt joined
Geth MoarVM/jit-legacy-cleanup: 5 commits pushed by (Bart Wiegmans)++ 15:53
brrt i'm going to have to refactor this refactor at some point 15:56
because it became a mess :-P
16:05 AlexDaniel joined
timotimo we could in theory write a version of param_op_i that takes a pointer where it stores the "should jump" bit 16:08
16:16 geekosaur joined 16:20 robertle joined 16:36 dogbert11 joined
[Coke] c:\users\user\rakudo\nqp\moarvm\src\strings\parse_num.c(265) : warning C4715: 'parse_simple_number': not all control paths return a value 16:57
16:59 domidumont joined 17:03 lizmat joined 17:19 brrt joined
brrt .tell timotimo that's way overthinking it 17:47
yoleaux brrt: I'll pass your message to timotimo.
brrt you only need a flag to check the branch
samcv good *
brrt ohai samcv
i'm like leaving in 5s though :-)
saw you merged the KMP algorithm
samcv yep 17:50
17:52 geekosaur joined 17:54 AlexDaniel joined 18:13 brrt joined 18:23 geekosaur joined 18:32 geekosaur joined
samcv .ask brrt can I get the slides from YAPC-EU? 19:02
yoleaux samcv: I'll pass your message to brrt.
19:05 travis-ci joined
travis-ci MoarVM build failed. Samantha McVey 'Merge pull request #652 from MasterDuke17/add_error_string_to_failed_to_resolve_exception 19:05
travis-ci.org/MoarVM/MoarVM/builds/267683413 github.com/MoarVM/MoarVM/compare/1...b339553405
19:05 travis-ci left 20:30 lizmat_ joined 22:08 dogbert11 joined 22:25 leont joined
MasterDuke timotimo, brrt: so my PR needs some serious rework (not unexpected)? 22:31
Geth MoarVM: c322b96758 | (Samantha McVey)++ | 2 files
Add up to date string documentation
22:35
MoarVM: 95f6152dee | (Samantha McVey)++ | docs/strings.markdown
Remove old strings.markdown documentation file
22:43
23:33 travis-ci joined
travis-ci MoarVM build failed. Samantha McVey 'Add up to date string documentation' 23:33
travis-ci.org/MoarVM/MoarVM/builds/268165469 github.com/MoarVM/MoarVM/compare/1...22b96758dc
23:33 travis-ci left
samcv not sure why some of the builds are failing 23:47
only the clang ones, but it works fine for me, i'm building with clang 23:48
MasterDuke segfault in the nqp build 23:51
/tmp/moar/bin/moar --libpath=src/vm/moar/stage0 src/vm/moar/stage0/nqp.moarvm --bootstrap --module-path=gen/moar/stage1 --setting-path=gen/moar/stage1--setting=NQPCORE --no-regex-lib --target=mbc --output=gen/moar/stage1/NQPHLL.moarvm gen/moar/stage1/NQPHLL.nqp 23:52