On 23/11/12 00:59, Christoffer Dall wrote:
> From: Christoffer Dall <cdall@xxxxxxxxxxxxxxx>
>
> The set_pte_ext function orr'ed the RDONLY bit onto the PTEs, which is
> bit[7], which is HAP[1] and causes writable access to the pages.
>
> This was unfortunate.
>
> Cc: Nicolas Viennot <nviennot@xxxxxxxxxxxxxxx>
> Cc: Jeremy C. Andrus <jeremya@xxxxxxxxxxxxxxx>
> Signed-off-by: Christoffer Dall <c.dall@xxxxxxxxxxxxxxxxxxxxxx>
> ---
> arch/arm/kvm/mmu.c | 27 +++++++++++++++------------
> 1 file changed, 15 insertions(+), 12 deletions(-)
>
> diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c
> index 7b0e6e5..720bbd5 100644
> --- a/arch/arm/kvm/mmu.c
> +++ b/arch/arm/kvm/mmu.c
> @@ -41,6 +41,16 @@ static void kvm_tlb_flush_vmid(struct kvm *kvm)
> kvm_call_hyp(__kvm_tlb_flush_vmid, kvm);
> }
>
> +static void set_pte(pte_t *pte, pte_t new_pte)
> +{
> + *pte = new_pte;
One last nit: consider using "pte_val(*pte) = new_pte;" instead, so we
can still compile with STRICT_MM_TYPECHECKS.
M.
--
Jazz is not dead. It just smells funny...
_______________________________________________
kvmarm mailing list
kvmarm@xxxxxxxxxxxxxxxxxxxxx
https://lists.cs.columbia.edu/cucslists/listinfo/kvmarm
