I've updated https://wiki.linaro.org/MichaelHope/Sandbox/BuildFailures
to include the six new build failures since yesterday.
In my opinion, the most important are the two with compiler issues:
qemu-kvm: invalid assembler https://bugs.launchpad.net/gcc-linaro/+bug/605252
libtheora: ICE on 64 bit math https://bugs.launchpad.net/gcc-linaro/+bug/605255
These are next important as they're important packages that have some type of runtime failure:
apr: fails in the test phase https://bugs.launchpad.net/gcc-linaro/+bug/604862
axis2c: aborts https://bugs.launchpad.net/gcc-linaro/+bug/604864
firefox: segfaults when running a native tool during the build process https://bugs.launchpad.net/gcc-linaro/+bug/604874
-- Michael
Michael Hope michael.hope@linaro.org wrote:
libtheora: ICE on 64 bit math https://bugs.launchpad.net/gcc-linaro/+bug/605255
This is clearly a compiler bug; I've updated the LP report with a reduced test case.
Mit freundlichen Gruessen / Best Regards
Ulrich Weigand
-- Dr. Ulrich Weigand | Phone: +49-7031/16-3727 STSM, GNU compiler and toolchain for Linux on System z and Cell/B.E. IBM Deutschland Research & Development GmbH Vorsitzender des Aufsichtsrats: Martin Jetter | Geschäftsführung: Dirk Wittkopp Sitz der Gesellschaft: Böblingen | Registergericht: Amtsgericht Stuttgart, HRB 243294
Michael Hope wrote:
I've updated https://wiki.linaro.org/MichaelHope/Sandbox/BuildFailures
to include the six new build failures since yesterday.
Michael, I went through this list, and found some of them also appears in ubuntu build failures shown in, http://udd.debian.org/cgi-bin/ubuntu_ftbfs.cgi
Bug #605251 pyao fails to build from source with Linaro toolchain Bug #605248 ppp fails to build from source with Linaro toolchain Bug #604858 apt fails to build from source with Linaro toolchain Bug #604860 aptitude fails to build from source
So they are not linaro-gcc specific bugs. Shall we move them to somewhere else or change to "Invalid"?
On Fri, Jul 16, 2010, Yao Qi wrote:
I went through this list, and found some of them also appears in ubuntu build failures shown in, http://udd.debian.org/cgi-bin/ubuntu_ftbfs.cgi
Bug #605251 pyao fails to build from source with Linaro toolchain Bug #605248 ppp fails to build from source with Linaro toolchain Bug #604858 apt fails to build from source with Linaro toolchain Bug #604860 aptitude fails to build from source
So they are not linaro-gcc specific bugs. Shall we move them to somewhere else or change to "Invalid"?
So, we should ideally have done this check BEFORE filing it in gcc-linaro, since Matthias had reported a bunch of failures from http://udd.debian.org/cgi-bin/ubuntu_ftbfs.cgi against the Ubuntu packages.
The right workflow is:
check whether the same failure is on http://udd.debian.org/cgi-bin/ubuntu_ftbfs.cgi - if yes, this isn't Linaro specific, flag it as such in the wiki page, you're done; - else, file a gcc-linaro bug
Since Michael filed some bugs already, when looking at them check whether it's an issue in Ubuntu, if it isn't, investigate, otherwise check whether it's reported in Ubuntu already by going to the source package bugs page: https://bugs.launchpad.net/ubuntu/+source/foo if you find a bug, mark the Linaro GCC bug as a duplicate of the Ubuntu bug; if you don't, add an Ubuntu task to the Linaro GCC bug using "Also affects distribution", and mark the Linaro task as Invalid.
Thanks!
On Fri, Jul 16, 2010, Yao Qi wrote:
Bug #605251 pyao fails to build from source with Linaro toolchain
duplicate of open Ubuntu bug
Bug #605248 ppp fails to build from source with Linaro toolchain
duplicate of open Ubuntu bug
Bug #604858 apt fails to build from source with Linaro toolchain
duplicate of fixed Ubuntu bug
Bug #604860 aptitude fails to build from source
duplicate of fixed Ubuntu bug
Update wiki
linaro-toolchain@lists.linaro.org
-
Loïc Minier -
Michael Hope -
Ulrich Weigand -
Yao Qi