<ng0>do the guix contributors in here know about the ninja implementation in C99 and have any use for experimenting with it, if it's of any use ti reduce size of some packages? it's still in earlyy version stages but seems to work so far <ng0>i packaged it last week as devel/samurai for pkgsrc <oriansj`>janneke: well I can solve that tcc problem <oriansj`>but tcc really isn't a step that MesCC will want to support long term. We either need to extend MesCC to build GCC directly or tell me now and I'll figure out an alternative long term solution. <oriansj`>also there are some long term design changes that we will need to discuss janneke in MesCC; to ensure better performance, easier porting and easier debugging. ***oriansj` is now known as oriansj
<oriansj>janneke: also we need to work together to get mes-m2 into shape for MesCC and Guix <fosslinux>oriansj: how about bypassing mescc eventaully and jumping straight from M2-Planet to tcc? ***Server sets mode: +cnt
<janneke_>fosslinux: technically, everything is possible <janneke_>oriansj: yes, exactly. that's why i started the wip-merge-m2 effort, right when you prove-of-concept built mes-m2 with m2-planet ***janneke_ is now known as janneke
<janneke>we will probably need a second (and possibly third) merge effort when your boot and test work in mes-m2 has matured/been finished <janneke>i have not been looking at merging any of the big simplifications (reader, etc) you have been working on; that's a next (yet very important) step i would say ***ng0_ is now known as ng0
***ng0_ is now known as ng0
<janneke>it's going great, lots of progress; very nice r-b summit in marrakech <rain1>that's fun i will read up on the rb summit <rain1>ive been busy things have been hectic <janneke>you have been a great support, especially at times i was frustrated or feeling stuck <rain1>that's really sweet of you to say, im glad i was supportive <xentrac>i'd be very happy to have any feedback <xentrac>how about you? I'm glad to hear you're OK! <rain1>ive been caring for someone who got injured badly <xentrac>but thank goodness you can be there for them! <rain1>they hsould make a full recovery <xentrac>because a lot of times carers bear the brunt of that <rain1>you make me realize how lucky we really are <xentrac>life isn't long, but it is wonderful, despite being full of suffering <xentrac>appreciating each other makes it better <xentrac>when we can manage it! we can't always manage it <oriansj>xentrac: reading markdown/forth-assembling; I'd argue that when one is doing an assembler in FORTH shouldn't be doing a minimal assembler but rather as complete assembler as possible. <oriansj>and latest REPL tweak is up and slow_lisp is now merged with master in mes-m2 <pder>oriansj: great news! I am getting a failure with test007 in mes-m2 and was wondering if it is a known issue. It seems to fail if M2-Planet is bootstrapped from mescc-tools-seed. If M2-Planet is compiled with gcc, test 7 passes. <xentrac>oriansj: that is certainly plausible, but the question I was trying to answer was not how one should do an assembler in Forth so much as how one should do a minimal assembler <xentrac>(it may well be that the best way to kill a drug mule who's double-crossed you is with a Colt 1911, but that surely does not imply that if you have a Colt 1911, you should be using it to execute drug mules) <xentrac>that is, it's certainly possible that what I was trying to accomplish is ill-advised, but that wasn't the question I was considering in that essay so much as how to accomplish it