Charles, I asked around in #guile (irc.freenode.net), and someone says there are known incompatibilities between 1.8 and 2.0. Do you know if upstream (gschem) has already migrated to 2.0 and tested? Developers in #guile are usually responsive, when they're awake, so stop in. :)
Fang On Mon, 21 Jul 2014, Charles Lepple wrote: > On Jul 21, 2014, at 1:46 PM, David Fang <f...@csl.cornell.edu> wrote: > >> Charles, >> Is that an error during *building* guile20, or during use of guile20 in >> your package? I unfortunately don't have 10.9 for testing, but I can ask >> someone in #fink to test on 10.9 (and I thought I had before). > > I should have been more specific in the follow-up: guile20 (2.0.11-1) builds > fine, and geda-gaf links against it, but that error is at runtime when > starting gschem from geda-gaf. (Should be reproducible without any specific > data files, just change from guile18 to guile20 in geda-gaf.info, build, and > run gschem.) > >> Is vm.c already de-optimized in your guile20.info? > > I think so? > > The following evaluates to true (exit code 0): > > [[ ! "$(clang --version | head -n1 | cut -d- -f2 | cut -d')' -f1)" < > "503.0.38" ]] > > $ echo "$(clang --version | head -n1 | cut -d- -f2 | cut -d')' -f1)" > 503.0.40 > >> Fang >> >>> On Jul 8, 2014, at 7:25 AM, Charles Lepple <clep...@gmail.com> wrote: >>> >>>> On Jul 7, 2014, at 9:10 PM, David Fang <f...@csl.cornell.edu> wrote: >>>> >>>>> Charles, >>>>> >>>>> Can geda (and friends) use guile20 yet? >>>> >>>> Theoretically, yes. If I just replace guile18* with guile20 in the Fink >>>> info file, there is a strange error at runtime - we can take a look at >>>> that if you would like. >>> >>> Here is the guile20 error: >>> >>> Backtrace: >>> In ice-9/boot-9.scm: >>> 157: 1 [catch #t #<catch-closure 107600c60> ...] >>> In unknown file: >>> ?: 0 [apply-smob/1 #<catch-closure 107600c60>] >>> >>> ERROR: In procedure apply-smob/1: >>> ERROR: In procedure variable-ref: variable is unbound: #<variable 1076382f0 >>> value: #<undefined>> >>> >>> Unfortunately, searching for various parts of that error message just leads >>> me to the GDB documentation, which always has a real variable name instead >>> of a hex value. >>> >>>>> Fang >>>>> >>>>>> On 8/07/14 00:10, David Fang wrote: >>>>>>> Martin, and all, >>>>>>> Are there multiple places that need the CR fix? If this is an >>>>>>> isolated case, I could just apply that patch as a workaround. >>>>>> >>>>>> In the sources of geda-gaf, this seems the only place that needs fixing. >>>>>> With the patch to scheme_object.c, geda-gaf builds for me on 10.9. >>>>>> >>>>>> Similar errors could pop up in other packages depending on guile18, >>>>>> though, if they use guile-1.8-snarf. I wouldn't know how to fix it in >>>>>> guile18, short of replacing the relevant sed script in guile-1.8-snarf >>>>>> by the one from guile-2.0-snarf. >>>> >>>> I have a feeling that just replacing guile-1.8-snarf with guile-2.0-snarf >>>> will be the least-intrusive fix. I can try to rebuild gnucash with that >>>> hack, and maybe try some other packages that depend on guile-1.8, but I >>>> suspect a lot of them have silently broken with the clang update that >>>> Martin mentioned. >>>> >>> >>> Committed Martin's fix for now. Tested and validated on 10.9; let me know >>> if it breaks for another version. >>> >>> Thanks, >>> >>> >> >> -- >> David Fang >> http://www.csl.cornell.edu/~fang/ > > -- David Fang http://www.csl.cornell.edu/~fang/ ------------------------------------------------------------------------------ Want fast and easy access to all the code in your enterprise? Index and search up to 200,000 lines of code with a free copy of Black Duck Code Sight - the same software that powers the world's largest code search on Ohloh, the Black Duck Open Hub! Try it now. http://p.sf.net/sfu/bds _______________________________________________ Fink-users mailing list Fink-users@lists.sourceforge.net List archive: http://news.gmane.org/gmane.os.macosx.fink.user Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-users
