John, On 28 August 2012 15:31, John Mandereau <john.mander...@gmail.com> wrote: > 2012/8/28 Phil Holmes <m...@philholmes.net>: >> Yes - "it" is Gub. I might try getting it working on 64 bit once I've >> bedded down regularly running it in the VM. > > Running GUB inside a VM must slow it down a lot,
It depends. If you have V-T enabled CPUs on your PC and the Virtual Software (VirtualBox or linux's qemu-KVM in the 3.x kernels for example) is able to access the CPUs directly because of that - effectively the OS in the VM has direct access to the CPU as if it were installed on bare metal. There is 'some' loss but it's negligible on my system - I seem to recall about 120 second difference compiling native vs in a VM. SSDs do also help. E.g http://ark.intel.com/Products/VirtualizationTechnology Para-virtualization will slow things down but I don't have any figures, but even they are getting better. You may then ask why bother using a VM at all and just install and compile on your own OS. It's easy then to roll out cloned VMs or take snapshots and roll back to certain points - I used to use this a lot before we had patchy test scripts - I'd run make and make test-baseline, take a snapshot and the apply a patch and run make test. When that was done, roll back to the snapshot and apply next patch and run make ; make test without the need to do another make test-baseline. Took seconds to roll back, saved me 10 minutes a patch (and that's on a fast machine) and I could guarantee a clean make test each time. i am sure there are other methods that snapshots can be used for, as long as you have the disk space and reasonable amounts of RAM, I think it's a nice way to develop. Oh and it helps when testing new LilyDevs. James _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
