Still running. I need to go to bed, but assuming this works then tomorrow I think that I can start up the same VM in qemu and try again (I have never used qemu, so if someone can post an example use that would be great). If that then fails then there would be clear evidence that it's KVM (and of course it may still fail tonight inside VB). Andrew
On Thursday, 20 February 2014 21:31:16 UTC-3, andrew cooke wrote: > > > Wow; OK, sorry, I stand corrected. > > My VM (VB not KVM) is still running. Unfortunately it's a single CPU so > takes some time. > > Andrew > > On Thursday, 20 February 2014 20:57:34 UTC-3, Elliot Saba wrote: >> >> Actually, we have had issues in the past with VMs, particularly with >> CPU detection features. This looks to be a little different, but the >> KVM thing could be a factor, let's not rule that out yet. >> >> On Thu, Feb 20, 2014 at 3:46 PM, andrew cooke <[email protected]> wrote: >> > >> > this sounds incredibly unlikely. i have just started up an opensuse vm >> in >> > virtualbox and am compiling julia there. so far no errors. >> > >> > it is much more likely that you are doing something wrong than that >> julia >> > somehow can detect the difference between a "real" computer and a vm >> > (nothing personal - but people use vms all the time and they work just >> > fine). >> > >> > i did a git clone, followed by make (plus installing missing >> dependencies >> > and restarting make). what are you doing? >> > >> > andrew >> > >> > >> > >> > On Thursday, 20 February 2014 17:59:35 UTC-3, Ronan Chagas wrote: >> >> >> >> I have new info for this problem. >> >> >> >> I could reproduce locally the problem on the server when I tried to >> build >> >> julia in a KVM. >> >> Now the question: Why building inside a KVM fails and output >> >> "UndefVarError" whereas building in my host system works? >> >> >> >> Best regards, >> >> Ronan >> >
