IRC channel logs

2019-08-14.log

back to list of logs

***scs is now known as Guest20797
***vagrantc_ is now known as vagrantc
***scs is now known as Guest26796
***scs is now known as Guest84414
***scs is now known as Guest97805
***scs is now known as Guest30145
***scs is now known as Guest87698
<OriansJ>morning;
<OriansJ>The AMD64 bootstrap (cc_x86.M1) is now less than 1000 lines to completion (with various testing and minor fixes of course) and then I'll be doing cc_amd64.M1 as a string update (The diffs will be beautiful)
<OriansJ>Then I get to update mescc-tools-seed to be multi-arch and then the ARMv7l bootstrap goes onto the weekend burner and getting mes-m2 running is finally priority (This is going to get "interesting")
***scs is now known as Guest18605
***mniip_ is now known as mniip
<xentrac>congratulations, OriansJ!
<dongcarl>OriansJ: congratulations
<Hagfish>sounds great!
<janneke>OriansJ: wow, beautiful!
*vagrantc wonders what vagrantc missed
<vagrantc>i need to build mes soon on Debian to check what outstanding issues are still there; all the build-dependencies are in Debian now
<vagrantc>i remember one issue was file clashes, such as bin/ar ... which work fine in a guix context but not in Debian
<janneke>vagrantc: ah, i'm still --slowly-- working on a mes-0.20 and renaming bin/ar is one of the things I did
<janneke>vagrantc: iwbn if you could try the wip-0.20 branch some time
<vagrantc>janneke: sure
*vagrantc isn't sure when... but will give it a whirl
<janneke>vagrantc: <OriansJ> The AMD64 bootstrap (cc_x86.M1) is now less than 1000 lines to completion (with various testing and minor fixes of course) and then I'll be doing cc_amd64.M1 as a string update (The diffs will be beautiful)
<janneke>vagrantc: <OriansJ> Then I get to update mescc-tools-seed to be multi-arch and then the ARMv7l bootstrap goes onto the weekend burner and getting mes-m2 running is finally priority (This is going to get "interesting")
<vagrantc>exciting times :)
<janneke>yeah
<vagrantc>janneke: ugh. lots of conflicts between wip and wip-0.20 ...
<vagrantc>last version i tested was based on wip branch
<vagrantc>ah, merging the other direction seemed to work fine ...
<janneke>eek...let's hope that works -- do you have many modifications outside of debian/ left?
<vagrantc>zero modifications outside debian/
<vagrantc>so should be fine
<vagrantc>i also used: git merge --strategy=ours
<vagrantc>from the wip-0.20 branch ...
<janneke>ah, very good
<vagrantc>apparently "don't know when" is now :)
<janneke>if not now, then when :)
<vagrantc>my last build took ~31 minutes ... but i probably won't be able to check back in till later.
<vagrantc>running test suite...
<vagrantc>XFAIL is expected failure?
<janneke>yes, PASS, XFAIL are "okay"
<janneke>FAIL and XPASS are bad news
<vagrantc>janneke: it built, with no major issues
<vagrantc>i'll need to update debian/copyright being the main issue
<janneke>vagrantc: that's very good news, thanks for checking that
<vagrantc>janneke: there's one issue that i'm not clear on regarding debugging symbols
<vagrantc>janneke: https://lintian.debian.org/tags/debug-file-with-no-debug-symbols.html
<janneke>we can have a 0.20 release that works on debian and as a basis for stage0/mes-m2/further reduced guix builds
<janneke>vagrantc: is that a gcc-built file, or could it be a mescc+mescc-tools built file?
*vagrantc shrugs :)
<janneke>if it's gcc-built, we may (still) have a problem passing the `-g', mescc+mescc-tools will certainly not produce conforming debugging in executables (there is only very rudimentary gdb support, atm)
<janneke>yeah, i don't see it as a big problem, but i'll give it a look, thanks
<vagrantc>might be possible to just exclude it or something if it's not a "real" issue
<janneke>OK -- me needs to go for a moment