On Tue, Aug 20, 2013 at 04:19:27PM -0500, Andrew McNabb wrote: > On Tue, Aug 20, 2013 at 10:13:10PM +0100, Peter Robinson wrote: > > > > In terms of testing or debugging probably the easiest way is to spin up an > > image under qemu emulation, it's not the fastest but it works pretty well. > > If there's anything else I can help with feel free to ask. > > > > http://fedoraproject.org/wiki/Architectures/ARM/F19/Installation#For_Versatile_Express_Emulation_with_QEMU > > That's very good to know about, and it makes tracking this down seem > less hopeless. I tried tracking down the problem using an ARM virtual machine using these instructions, but it was too slow in helping track down the failed tests (in fact, it had many more failed tests). As a "temporary" measure, I then changed python-pexpect to no longer be a noarch package. It looks like this will have a ripple effect, requiring python-ipython (an important package) to lose noarch. At present, the upstream maintainer and myself have no way to track down the problem because no one has ARM hardware running Fedora. So it could be a very long time before this is fixed. Since ARM is a primary architecture but ARM hardware running Fedora is so rare, it would be nice if there were some way to give upstream maintainers access to ARM machines to track down bugs. -- Andrew McNabb http://www.mcnabbs.org/andrew/ PGP Fingerprint: 8A17 B57C 6879 1863 DE55 8012 AB4D 6098 8826 6868 -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
