Re: [PATCH 5/9] KVM: selftests: Align memory region addresses to 1M on s390x

2019-05-24 Thread Thomas Huth
On 24/05/2019 10.29, Christian Borntraeger wrote: > > > On 23.05.19 19:40, Andrew Jones wrote: >> On Thu, May 23, 2019 at 06:43:05PM +0200, Thomas Huth wrote: >>> On s390x, there is a constraint that memory regions have to be aligned >>> to 1M (or running the VM will fail). Introduce a new "align

Re: Re: [PATCH 5/9] KVM: selftests: Align memory region addresses to 1M on s390x

2019-05-24 Thread Christian Borntraeger
On 23.05.19 19:40, Andrew Jones wrote: > On Thu, May 23, 2019 at 06:43:05PM +0200, Thomas Huth wrote: >> On s390x, there is a constraint that memory regions have to be aligned >> to 1M (or running the VM will fail). Introduce a new "alignment" variable >> in the vm_userspace_mem_region_add() fun

Re: [PATCH 5/9] KVM: selftests: Align memory region addresses to 1M on s390x

2019-05-23 Thread Andrew Jones
On Thu, May 23, 2019 at 06:43:05PM +0200, Thomas Huth wrote: > On s390x, there is a constraint that memory regions have to be aligned > to 1M (or running the VM will fail). Introduce a new "alignment" variable > in the vm_userspace_mem_region_add() function which now can be used for > both, huge pa

[PATCH 5/9] KVM: selftests: Align memory region addresses to 1M on s390x

2019-05-23 Thread Thomas Huth
On s390x, there is a constraint that memory regions have to be aligned to 1M (or running the VM will fail). Introduce a new "alignment" variable in the vm_userspace_mem_region_add() function which now can be used for both, huge page and s390x alignment requirements. Signed-off-by: Thomas Huth ---