On 07/01/2016 12:13 AM, Benjamin Herrenschmidt wrote:
> On Thu, 2016-06-30 at 18:01 +0200, Cédric Le Goater wrote:
>> +static uint32_t ppc_hash64_pte_size_decode(PowerPCCPU *cpu, uint64_t
>> pte0,
>> + uint64_t pte1, uint32_t
>> slb_pshift)
>> {
>> -sw
On Fri, Jul 01, 2016 at 08:13:47AM +1000, Benjamin Herrenschmidt wrote:
> On Thu, 2016-06-30 at 18:01 +0200, Cédric Le Goater wrote:
> > +static uint32_t ppc_hash64_pte_size_decode(PowerPCCPU *cpu, uint64_t
> > pte0,
> > + uint64_t pte1, uint32_t
> > slb_ps
On Thu, 2016-06-30 at 18:01 +0200, Cédric Le Goater wrote:
> +static uint32_t ppc_hash64_pte_size_decode(PowerPCCPU *cpu, uint64_t
> pte0,
> + uint64_t pte1, uint32_t
> slb_pshift)
> {
> - switch (slb_pshift) {
> - case 12:
> - return 12;
> -
On 06/30/2016 12:56 PM, Anton Blanchard wrote:
> Hi,
>
>> From: Benjamin Herrenschmidt
>>
>> We were always advertising only 4K & 16M. Additionally the code wasn't
>> properly matching the page size with the PTE content, which meant we
>> could potentially hit an incorrect PTE if the guest used m
On Thu, 2016-06-30 at 20:56 +1000, Anton Blanchard wrote:
> Hi,
>
> > From: Benjamin Herrenschmidt
> >
> > We were always advertising only 4K & 16M. Additionally the code wasn't
> > properly matching the page size with the PTE content, which meant we
> > could potentially hit an incorrect PTE if
Hi,
> From: Benjamin Herrenschmidt
>
> We were always advertising only 4K & 16M. Additionally the code wasn't
> properly matching the page size with the PTE content, which meant we
> could potentially hit an incorrect PTE if the guest used multiple
> sizes.
>
> Finally, honor the CPU capabiliti
On Tue, Jun 28, 2016 at 08:48:34AM +0200, Cédric Le Goater wrote:
> From: Benjamin Herrenschmidt
>
> We were always advertising only 4K & 16M. Additionally the code wasn't
> properly matching the page size with the PTE content, which meant we
> could potentially hit an incorrect PTE if the guest
From: Benjamin Herrenschmidt
We were always advertising only 4K & 16M. Additionally the code wasn't
properly matching the page size with the PTE content, which meant we
could potentially hit an incorrect PTE if the guest used multiple sizes.
Finally, honor the CPU capabilities when decoding the
