On Thu, Nov 29, 2018 at 04:29:05PM +0100, Andrew Jones wrote:
> Reported-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> Signed-off-by: Andrew Jones <drjones@xxxxxxxxxx>
> ---
> tools/testing/selftests/kvm/lib/ucall.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tools/testing/selftests/kvm/lib/ucall.c b/tools/testing/selftests/kvm/lib/ucall.c
> index 4777f9bb5194..e28bdd6b546c 100644
> --- a/tools/testing/selftests/kvm/lib/ucall.c
> +++ b/tools/testing/selftests/kvm/lib/ucall.c
> @@ -58,7 +58,7 @@ void ucall_init(struct kvm_vm *vm, ucall_type_t type, void *arg)
> start = 1ul << (vm->va_bits * 2 / 3);
> end = 1ul << vm->va_bits;
> step = 1ul << (vm->va_bits / 6);
> - for (gpa = start; gpa >= 0; gpa -= step) {
> + for (gpa = start; gpa > 0; gpa -= step) {
> if (ucall_mmio_init(vm, gpa & ~(vm->page_size - 1)))
> return;
> }
Ugh... sorry for this junk. I was obviously way too hasty "fixing" this. I
forgot to consider that the virtual address space is most likely *not* a
multiple of the calculated step. I'll send a v2 that doesn't suck in a
moment.
drew
