Index

Show enters and exits. Hide enters and exits.

15:31:54boyscoutAdd NUM2OFFT, fixes #348 - 3e3c412 - Dirkjan Bussink
15:31:54boyscoutDon't refer to non existant 0 entries in the profile graph - cac1547 - Dirkjan Bussink
15:41:13boyscoutCI: rubinius: cac1547 successful: 3441 files, 13543 examples, 41064 expectations, 0 failures, 0 errors
17:03:58jeremyevansI'm trying to get 1.0.1 compiling on OpenBSD 4.7-current, but I'm getting errors. Build log available at http://code.jeremyevans.net/build.log, patches just to get this far at http://code.jeremyevans.net/patches.diff
17:11:14Dammnice openbsd + rubinus :)
17:15:05jeremyevansDamm: It would be nice if I could get it to work :) I'm most of the way there, I think, but I don't understand why I'm getting the errors at the end of the log.
17:16:56Dammjeremyevans: I'm dealing with trying to find a happy medium with instances with gcc 4.1 and rubinus
17:16:57Damm:/
17:17:52Dammjeremyevans: and unfortunately yeah I agree with the errors that's new to me
17:18:09Dammjeremyevans: is that with LLVM perhaps?
17:18:49Dammnope it disabled it
17:19:16Dammhasn't ran openbsd in a recent enough version in 2 years to really give more then that
17:19:55jeremyevansDamm: This is using amd64 with gcc-4.2.1
17:20:19Dammjeremyevans: hrm I use only 4.2.4+
17:20:20Dammto touch
17:23:22jeremyevansDamm: OpenBSD amd64 just switched to 4.2.1 from 3.3.5, since 4.2.1 is the last GPLv2 version
17:23:52Dammjeremyevans: well good news is you have better luck with 4.2.1 then 3.3.5
17:24:09jeremyevansDamm: I figured I would. I didn't even really try with 3.3.5
17:26:54Dammyeah i wouldn't
17:27:01Dammyou'd have better chances of swimming to England from NY
17:36:51jeremyevansDamm: I'm going to give it a shot now with LLVM and some optimizations turned off
17:37:37Dammgood luck
17:43:25jeremyevansDamm: http://github.com/evanphx/rubinius/issues/issue/75
17:43:52jeremyevansDamm: I just have to figure out why it is stripping vm/vm
17:44:44JudsonHas anyone tried to build rbx for Gentoo?
17:53:02DammJudson: it works here
17:53:22DammI have 1.0.1 on Gentoo 2010 with gcc 4.3.4
18:01:36JudsonOkay. I'm trying to build an ebuild for it, and the specs are failing. Things like IO::Sockets are being opened when they should fail, and chroots are failing.
18:01:48JudsonI suspect it has to do with the ebuild sandbox
18:46:04dbussinkJudson: running stuff as root?
18:54:41jeremyevansI'm having the same issue as http://github.com/evanphx/rubinius/issues/issue/75, but vm/vm is not stripped: vm/vm: ELF 64-bit LSB executable, x86-64, version 1, for OpenBSD, dynamically linked (uses shared libs), not stripped
18:59:11JudsonYes, tests are running as root
18:59:16JudsonHas to for ebuild
19:00:23dbussinkJudson: because it's know there are some failures when running it as root
19:00:44Judsondbussink: Aha.
19:00:59dbussinkjeremyevans: it could also be that openbsd is missing some stuff that rbx can't bind to in the ffi
19:01:11dbussinkjeremyevans: do some functions work with the ffi or do they all fail?
19:01:51jeremyevansdbussink: Well, I can't even start the vm, so I'm not sure
19:02:07jeremyevansdbussink: OpenBSD has a libffi port, but rbx looks like it uses it's own
19:02:50dbussinkjeremyevans: yeah it uses it's own ffi
19:03:22dbussinkbut getting the bits right for the dlopen stuff etc. can be tricky, so it's either libffi not setup propery at all or some methods being unavailable
19:03:45dbussinkjeremyevans: does it give an error on malloc?
19:03:51dbussinkjeremyevans: btw, at railsconf or not?
19:04:10jeremyevansdbussink: Not at railsconf
19:04:27dbussinkjeremyevans: ah ok, otherwise you might have been able to talk to evan about it
19:04:34dbussinkbut i'm heading out for the closing keynote
19:04:36jeremyevansdbussink: Is there some way I could test for a malloc error
19:05:06jeremyevansdbussink: No rush, I'm just trying to get it setup so I can start testing Sequel with it
19:05:18Judsondbussink: Well, is there an rbx opinion as to how to deal with that?
19:05:38Judsondbussink: I'm up at the front on the left. Any chance we could meet up for 5 minutes?
19:05:46jeremyevansdbussink: FWIW, it doesn't look that OpenBSD has an patches in it's ffi port that would be related to this error
19:05:49dbussinkJudson: for now it's just ignore the failures ;)
19:06:13Judsondbussink: Okay: I can patch the Rakefile to just not run the rpsecs :)
19:50:21DammJudson, trying to compile rubinus in the past as root has left some undesirebles
19:50:24Dammit's best to avoid it if you can
19:50:49Dammremembers the rm -f /dev/null bug
19:53:00JudsonHrm.
19:53:18JudsonTo do installs on Gentoo, it's kind of a pain.
19:53:30JudsonMaybe I could su down to a system user
20:14:29JamamoogaSo is anyone actually making progress on windows support? Im a compiler writer myself & am willing to contribute if help is needed.
20:20:09slavaJamamooga: what compiler?
20:22:13kronos_vanoJamamooga, Here is a bunch of forks. Rubinius already can be compiled on windows, but it doesn't run)
20:22:52JamamoogaIm making a language currently called Jamzam, im at the early stages right now. It is based on termite scheme, smalltalk, haskell, & erlang. Im writing the first version in Ruby on IronRuby.
20:23:33slavascheme, smalltalk, haskell and erlang, eh?
20:24:02JamamoogaYes. not just scheme, termite scheme is like erlang.
20:24:25slavaI know
20:26:36JamamoogaLike Rubinius it is in the compilable, but not runable stage on windows. I recently switched back to windows after about 10 years of linux since I couldlnt get my games to work right with Wine or virtualbox.
20:36:09JamamoogaI will be releasing Jamzam under the CC0 license when it is a bit more usable.
21:04:38jvoorhisevan: have you had a chance to check out lldb yet?