[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: arm release issues

On Mon, Aug 28, 2006 at 02:43:56PM -0400, Lennart Sorensen <lsorense@csclub.uwaterloo.ca> wrote:
> Well xulrunner seems to have an issue with javac based on this last part
> of the build log:

Waw, I must say I'm impressed that for once in a while it doesn't abort
ld because it takes too much time to do its job...

> Compiling Java interface classes
> find _javagen -name "*.java" > java.files
> /usr/lib/jvm/java-gcj/bin/javac -source 1.4 -classpath . -d . -sourcepath _javagen @java.files
> Exception in thread "main" java.lang.NoClassDefFoundError: org.eclipse.jdt.core.compiler.IProblem
>    <<No stacktrace available>>
> make[3]: *** [org/mozilla/xpcom/.class_done] Error 1
> make[3]: Leaving directory `/build/buildd/xulrunner-'
> make[2]: *** [tier_99] Error 2
> make[2]: Leaving directory `/build/buildd/xulrunner-'
> make[1]: *** [default] Error 2
> make[1]: Leaving directory `/build/buildd/xulrunner-'
> make: *** [build-stamp] Error 2

This is because of that:
Get:5 http://incoming.debian.org unstable/main ecj-bootstrap 3.2-1 [1101kB]
Get:6 http://incoming.debian.org unstable/main ecj-bootstrap-gcj 3.1.2-6 [2646kB]

ecj-bootstrap being arch:all, ecj-bootstrap-gcj arch:any.
ecj-bootstrap-gcj didn't build on arm because of:

gcj-4.1 \
        -O2 -g -Wl,-Bsymbolic -shared -fPIC -fjni -findirect-dispatch \
        -o build/dist/ecj.jar.so build/dist/ecj.jar
make: *** [build/stamp] Terminated
Build killed with signal 15 after 500 minutes of inactivity

What a surprise... that happens a lot with big packages on arm. Couldn't this
be somehow configured by source package ? So that the ones we know takes
a long time can work a longer time before being killed.

Anyways, a manual binNMU forcing the use of ecj-bootstrap 3.1.2-6 should
solve the issue.


Reply to: