I'm running...
rpmbuild -bb --noprep --with baseonly --without debuginfo --target=i686 kernel.spec
...and after several hours the device hangs somewhere after this:
+ popd ~/rpmbuild/BUILD/kernel-4.5.fc23/linux-4.6.0-0.rc5.git1.1.mdh.fc23.i686 + '[' '%{buildarch}' = noarch ']' + QA_CHECK_RPATHS=1 + case "${QA_CHECK_RPATHS:-}" in + /usr/lib/rpm/check-rpaths
After I restart the device, is there a way to get rpmbuild to resume?
Mike
On Sat, Apr 30, 2016 at 08:59:21AM -0400, Michael Hill wrote:
I'm running...
rpmbuild -bb --noprep --with baseonly --without debuginfo --target=i686 kernel.spec
...and after several hours the device hangs somewhere after this:
- popd
~/rpmbuild/BUILD/kernel-4.5.fc23/linux-4.6.0-0.rc5.git1.1.mdh.fc23.i686
- '[' '%{buildarch}' = noarch ']'
- QA_CHECK_RPATHS=1
- case "${QA_CHECK_RPATHS:-}" in
- /usr/lib/rpm/check-rpaths
After I restart the device, is there a way to get rpmbuild to resume?
people used to hack around this problem by modding the spec script to skip certain build steps based upon an arg passed as "RPM_COMMAND" shell env variable.
Typical snippet would be
%build
case "${RPM_COMMAND:-all}" in all|config) %configure ;; esac
case "${RPM_COMMAND:-all}" in all|config|build) make ;; esac
Of course this is considered a hack and you loose all guarantees that the result is a clean build.
Richard
On Sun, May 1, 2016 at 3:33 PM, Richard Z rz@linux-m68k.org wrote:
Of course this is considered a hack and you loose all guarantees that the result is a clean build.
Thanks, Richard. Perfect.
Mike
kernel@lists.fedoraproject.org