Welcome to the main channel on the development of MoarVM, a virtual machine for NQP and Rakudo (moarvm.org). This channel is being logged for historical purposes.
Set by lizmat on 24 May 2021.
MasterDuke timo: some more info: gist.github.com/MasterDuke17/1dd6e...cded9902af 00:58
trying to dump bytecode kills gdb though 01:04
Geth MoarVM: MasterDuke17++ created pull request #1920:
Only look for lexicals in frames that have any		 02:40
MasterDuke ^^^ unrelated to questions about MVMP6intBody 02:41
lizmat: it looks like github.com/rakudo/rakudo/commit/3f...308cc27440 broke t/02-rakudo/12-proto-arity-count.t 03:04
weird that's it not on all ci jobs though... 03:05
oh right, now rakudo has separate build and test jobs 03:13
Geth MoarVM/main: c153f9f70b | (Daniel Green)++ | 2 files
Only look for lexicals in frames that have any

The change in frame_walker.c got rid of 37m calls to MVM_get_lexical_by_name() when compiling CORE.c.		 03:15
MoarVM/main: 23885cac29 | MasterDuke17++ (committed using GitHub Web editor) | 2 files
Merge pull request #1920 from MasterDuke17/only_look_for_lexicals_in_frames_that_have_them

Only look for lexicals in frames that have any
MasterDuke all the fails for my pr were caused by t/02-rakudo/12-proto-arity-count.t, so it seemed safe to merge 03:16
timo MasterDuke, can you put printf's or something in P6int.c's mk_storage_spec to tell you what the ALIGNOF results in for the different sizes? 14:09
i get an alignof of 4 for int32 14:10
lizmat MasterDuke: re c153f9f70b is it supposed to make things faster? I don't see it :-( and possibly the opposite 15:14
linkable6 (2025-03-04) github.com/MoarVM/MoarVM/commit/c153f9f70b Only look for lexicals in frames that have any
tellable6 lizmat, I'll pass your message to MasterDuke
MasterDuke timo: gist updated 22:27
tellable6 2025-03-05T15:14:14Z #moarvm <lizmat> MasterDuke: re c153f9f70b is it supposed to make things faster? I don't see it :-( and possibly the opposite
linkable6 (2025-03-04) github.com/MoarVM/MoarVM/commit/c153f9f70b Only look for lexicals in frames that have any
MasterDuke lizmat: i definitely saw slightly faster rakudo compilation
what did you see slower?
and i don't really know how it could make anything slower 22:29
lizmat well, if the check for absence would take more than the check for 0 elements, I guess ? 22:42

Powered by Cro and the Raku® Programming Language.

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

Thank you!