>I have another question. I searched in FAQ and saw someone reported that >virtualbox could run faster when hardware VT-x was disabled. Not sure what >guest did they run. If software virtualization is incomplete, how can software >virtualization run faster than the hardware virtualization?
In my own experience, the typical desktop performance is about the same. VT-x on or off. Just some rare commands are not done correctly in software VT mode, so some OSes do not like it. (IBM OS/2, Mac OS X, OpenBSD, ancient versions of Linux such as Red Hat Linux 7, from year-2000, etc...) The "incomplete" part is only about the commands it supports, but if it works, it is super-fast. On Thu, Oct 24, 2013 at 12:02 PM, Knut St. Osmundsen <[email protected]> wrote: > Hi Louis, > > could you provide us with steps how to reproduce this, please? > Alternatively, a core dump might be helpful if it's difficult to reproduce > without your ARM app. > > Kind Regards, > bird. > > On 10/23/2013 7:35 AM, Louis Luo wrote: > > Hi VBox developers, > > We are experiencing some crashes when we run Android X86 with VBox on a > system without VT-x. The App we run contains ARM code, so we add libhoudini > to the VM (just like what AndroidVM does). When we start the App, VBox > crashes. If we turn on VT-x, it seems everything is just fine. Can anyone > explain what happened here? > > Thanks, > Louis > > > _______________________________________________ > vbox-dev mailing list > [email protected] > https://www.virtualbox.org/mailman/listinfo/vbox-dev > > > > _______________________________________________ > vbox-dev mailing list > [email protected] > https://www.virtualbox.org/mailman/listinfo/vbox-dev > -- -Alexey Eromenko "Technologov" _______________________________________________ vbox-dev mailing list [email protected] https://www.virtualbox.org/mailman/listinfo/vbox-dev
