[RFC PATCH v3 5/5] arm/arm64: vgic-new: Implement KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO ioctl
Christoffer Dall
christoffer.dall at linaro.org
Tue Sep 6 12:20:32 PDT 2016
On Tue, Sep 06, 2016 at 07:42:19PM +0530, Vijay Kilari wrote:
> On Tue, Aug 30, 2016 at 7:30 PM, Christoffer Dall
> <christoffer.dall at linaro.org> wrote:
> >
> > On Wed, Aug 24, 2016 at 04:50:09PM +0530, vijay.kilari at gmail.com wrote:
> > > From: Vijaya Kumar K <Vijaya.Kumar at cavium.com>
> > > }
> > > diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c b/virt/kvm/arm/vgic/vgic-mmio-v3.c
> > > index 61abea0..fde1472 100644
> > > --- a/virt/kvm/arm/vgic/vgic-mmio-v3.c
> > > +++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c
> > > @@ -789,3 +789,22 @@ int vgic_v3_redist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
> > >
> > > return vgic_v3_uaccess(vcpu, dev, is_write, offset, val);
> > > }
> > > +
> > > +int vgic_v3_line_level_info_uaccess(struct kvm_vcpu *vcpu, bool is_write,
> > > + u32 intid, u32 *val)
> > > +{
> > > + unsigned int len = 4;
> > > + u8 buf[4];
> > > + int ret;
> > > +
> > > + if (is_write) {
> > > + vgic_data_host_to_mmio_bus(buf, len, *val);
> >
> > why do you involve the mmio bus in this?
>
> This is doing LE conversion. This is being done by vgic_uaccess function.
> IRC, This sys register access in not following vgic_uaccess path. Hence
> added it here.
See my e-mail on the other patch. I think this is all just legacy from
the time where we didn't distinguish between uaccess and mmio
dispatches.
> >
> >
> > > + ret = vgic_write_irq_line_level_info(vcpu, intid, len, buf);
> > > + } else {
> > > + ret = vgic_read_irq_line_level_info(vcpu, intid, len, buf);
> > > + if (!ret)
> > > + *val = vgic_data_mmio_bus_to_host(buf, len);
> > > + }
> > > +
> > > + return ret;
> > > +}
> > > diff --git a/virt/kvm/arm/vgic/vgic-mmio.c b/virt/kvm/arm/vgic/vgic-mmio.c
> > > index 38f2c75..74d0449 100644
> > > --- a/virt/kvm/arm/vgic/vgic-mmio.c
> > > +++ b/virt/kvm/arm/vgic/vgic-mmio.c
> > > @@ -391,6 +391,40 @@ void vgic_mmio_write_config(struct kvm_vcpu *vcpu,
> > > }
> > > }
> > >
> > > +int vgic_read_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid,
> > > + unsigned int len, void *val)
> > > +{
> > > + unsigned long data = 0;
> > > + int i;
> > > +
> > > + for (i = 0; i < len * 8; i++) {
> > > + struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
> > > +
> > > + if (irq->line_level)
> > > + data |= (1U << i);
> > > + }
> > > + vgic_data_host_to_mmio_bus(val, len, data);
> >
> > why???
>
> Same as above reason
> >
> >
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +int vgic_write_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid,
> > > + unsigned int len, const void *val)
> > > +{
> > > + int i;
> > > + unsigned long data = vgic_data_mmio_bus_to_host(val, len);
> >
> > why???
>
> Same as above reason
Thanks,
-Christoffer
More information about the linux-arm-kernel
mailing list