On 09/12/11 19:42, Pekka Enberg wrote: > On Fri, Dec 9, 2011 at 8:55 AM, Matt Evans <matt@xxxxxxxxxx> wrote: >> Add a --hugetlbfs commandline option to give a path to hugetlbfs-map guest >> memory (down in kvm__arch_init()). For x86, guest memory is a normal >> ANON mmap() if this option is not provided, otherwise a hugetlbfs mmap. >> >> Signed-off-by: Matt Evans <matt@xxxxxxxxxx> > > Btw, why don't you want to use MADV_HUGEPAGE for this? You could just > do it unconditionally, no? Well, I'm manually mapping from hugetlbfs as currently* PPC KVM requires hugepages to back guest RAM and MADV_HUGEPAGE is just a hint, no? I also wanted things to work on kernels without transparent hugepages enabled. I think it's safer to do things explicitly, as if the user requests hugepages it's more transparent (I'm thinking benchmarking, etc.) to be definitely using hugepages. Cheers, Matt *: I know Paul's posted patches to implement smallpage support... so this will change in time. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
