<html><head></head><body>Thanks, Greg. This is a great help. I'll have a look at it and let you know what I come up with.<br><br>Troy<br><br><div class="gmail_quote">On May 26, 2020 10:12:17 AM MST, Greg Hellings <greg.hellings@gmail.com> wrote:<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div dir="ltr"><div>I just wanted to point, again, to the build issues I've been having with Sword on PPC64LE architecture (e.g. the IBM POWER architecture, which is a supported platform for Fedora). The issue arises from mismatches in the definitions of 64-bit integers.</div><div><br></div><div>Diagnosing this might be tough on x86_64 hardware, so I've created a Vagrant box (at last - I've been struggling with this for months now) that you can run on a host that has qemu installed and is using the libvirt Vagrant driver. This is the default for Fedora/CentOS/RHEL systems.</div><div><br></div><div>It should be as simple, now, to get a POWER8 VM architecture running on the above systems as:</div><div><br></div><div>1) dnf install vagrant qemu-system-ppc64</div><div>2) cat < EOF > Vagrantfile</div><div>Vagrant.configure('2') do |config|<br> config.vm.box = 'greg-hellings/fedora-32-ppc64le'<br> config.vm.provider :libvirt do |lv, override|<br> lv.memory = '4096M'<br> lv.cpus = 4<br> end<br>end</div><div>EOF</div><div>3) vagrant up</div><div>4) vagrant ssh</div><div><br></div><div>Hopefully this helps you, Troy, in developing a patch that allows Sword to build on the ppc64le architecture. I'm currently building with a sizeable patch from Jaak that switches from __u64 to std::uint64_t throughout the engine so that we can still have Fedora packages. But it would be nice to see this addressed in mainline.<br></div><div><br></div><div>--Greg<br></div></div>
</blockquote></div><br>-- <br>Sent from my Android device with K-9 Mail. Please excuse my brevity.</body></html>