> Am 07.05.2014 um 11:40 schrieb Peter Maydell <peter.mayd...@linaro.org>: > >> On 7 May 2014 10:37, Alexander Graf <ag...@suse.de> wrote: >> >> >>>> Am 07.05.2014 um 11:26 schrieb Peter Maydell <peter.mayd...@linaro.org>: >>>> >>>> On 7 May 2014 10:09, Alexander Graf <ag...@suse.de> wrote: >>>> I don't think we should overengineer hacks for legacy virtio. >>> >>> Agreed. So what's our final conclusion: virtio endianness >>> is the endianness of the guest kernel at the point where >>> it triggers a reset of the virtio device, yes? >> >> The interrupt endianness for book3s PPC. Since that's an arch >> specific thing I think we should just make the determination >> mechanism arch dependent and list it in the spec. >> >> Booke for example would be vastly different since there is no >> global LE flag - it's a bit in the TLB entries. > > Sure, but I think we should also state the general principle > we're aiming to implement with the arch specific detail, > so that people figuring out what a new arch should do > have a guide to follow.
The general principle is "Try to find the Linux guest compile time endianness" :). Alex