Hi,
On Thu, May 19, 2022 at 02:40:38PM +0100, Will Deacon wrote:
> If we fail to allocate the 'supported_cpus' cpumask in kvm_arch_init_vm()
> then be sure to return -ENOMEM instead of success (0) on the failure
> path.
>
> Signed-off-by: Will Deacon <will@xxxxxxxxxx>
> ---
> arch/arm64/kvm/arm.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
> index 523bc934fe2f..775b52871b51 100644
> --- a/arch/arm64/kvm/arm.c
> +++ b/arch/arm64/kvm/arm.c
> @@ -146,8 +146,10 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
> if (ret)
> goto out_free_stage2_pgd;
>
> - if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL))
> + if (!zalloc_cpumask_var(&kvm->arch.supported_cpus, GFP_KERNEL)) {
> + ret = -ENOMEM;
> goto out_free_stage2_pgd;
> + }
> cpumask_copy(kvm->arch.supported_cpus, cpu_possible_mask);
>
> kvm_vgic_early_init(kvm);
Thank you for the fix:
Reviewed-by: Alexandru Elisei <alexandru.elisei@xxxxxxx>
This can go in independent of the series. I can send it after rc1 if you
prefer to focus on something else.
Thanks,
Alex
_______________________________________________
kvmarm mailing list
kvmarm@xxxxxxxxxxxxxxxxxxxxx
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
