github.com/moarvm/moarvm | IRC logs at colabti.org/irclogger/irclogger_logs/moarvm
Set by AlexDaniel on 12 June 2018.
patrickb o/ 07:25
patrickb .tell timotimo For the record: --no-relocatable does not change the runner to be a shell script again. The things it changes are: 07:26
yoleaux patrickb: I'll pass your message to timotimo.
patrickb .tell timotimo - link libmoar with an absolute path. 07:27
yoleaux patrickb: I'll pass your message to timotimo.
patrickb .tell timotimo - Set NQP_HOME and PERL6_HOME to an absolute path and do not try to determine them at run time. 07:28
yoleaux patrickb: I'll pass your message to timotimo.
jnthn Kaiepi: Only MVMObject (though there are many kinds of this), MVMSTable, and sometimes MVMFrame (depending on if it's been heap-promoted) 07:39
yoleaux 14 Apr 2019 09:13Z <nine> jnthn: I only got that line number with --ll-exception in the first place.
jnthn Kaiepi: MVMThreadContext certainly can not be :)
patrickb jnthn: On master there are no shell runners left in the install directory. `perl6`, `perl6-m`, `perl6-debug-m` are all real executables. `nqp` is still a shell script. 07:53
jnthn: In the build directory things are different. In there shell scripts are still generated, because the binary runners don't work in the build dir (without fiddling). 07:54
jnthn patrickb: The shell scripts don't seem to entirely work either, though (without make install) 07:59
patrickb: They seem to pick up on parts of the installed version
patrickb jnthn: Oh! That's no good.
I'll have a look at that. 08:00
jnthn Thanks. It'll make life a bit hard if I manage to break things enough that `make install` fails - which, given some of the things I work on, I'm somewhat liable to do :) 08:09
timotimo afl was very quick to figure out how to segfault immediately in MVM_frame_initial_invoke by getting tc->cur_frame to become null somehow, and MVM_FRAME_IS_ON_CALLSTACK doesn't null pointer check 10:46
we could introduce an acronym "AFLHF" 10:47
american fuzzy lop hanging fruit
american fuzzy low hanging fruit - maybe
lizmat is c10fee65c839b7e60fb74 worth a NQP / Rakudo bump ? 11:22
timotimo jnthn ^^ 11:24
lizmat bumped 13:03
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2019/04/15/...hrodinger/ 13:26
ugexe lizmat: cpanel link is broke 18:49
lizmat ugexe++ # thanks. fixed 19:48
brrt lizmat++ weekly 20:11
Geth ¦ MoarVM: MasterDuke17 assigned to samcv Issue Vectorization warnings with clang 8.0.0 github.com/MoarVM/MoarVM/issues/1095 22:05
patrickz .tell jnthn I attempted a fix for the runners in the build dir not working. What I found out: They were dependent on CWD and only worked with CWD = build dir. The install path was and still is included as a lib dir in those runners, because they need NQP which is located in the same dir. I changed the include order though, so the files in the build dir are preferred. 23:58
yoleaux patrickz: I'll pass your message to jnthn.
patrickz .tell jnthn I put the changes in github.com/patzim/rakudo/tree/fix-build-runners
yoleaux patrickz: I'll pass your message to jnthn.

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!