| Message ID | 156077793153.431407.3975189408068375970.stgit@bahia.lan |
|---|---|
| State | New |
| Headers | show |
| Series | xics: Don't call icp_kvm_realize() from icp_realize() | expand |
On 17/06/2019 15:25, Greg Kurz wrote: > kvm_kernel_irqchip is set by xics_kvm_init() which is always called after > icp_realize() since commit 3f777abc7107 "spapr/irq: add KVM support to the > 'dual' machine". I think we still need it for CPU hotplug. C. > This is thus dead code. Drop it. > > Signed-off-by: Greg Kurz <groug@kaod.org> > --- > hw/intc/xics.c | 8 -------- > 1 file changed, 8 deletions(-) > > diff --git a/hw/intc/xics.c b/hw/intc/xics.c > index 29f7d397813e..1dc3a0f12280 100644 > --- a/hw/intc/xics.c > +++ b/hw/intc/xics.c > @@ -351,14 +351,6 @@ static void icp_realize(DeviceState *dev, Error **errp) > return; > } > > - if (kvm_irqchip_in_kernel()) { > - icp_kvm_realize(dev, &err); > - if (err) { > - error_propagate(errp, err); > - return; > - } > - } > - > qemu_register_reset(icp_reset_handler, dev); > vmstate_register(NULL, icp->cs->cpu_index, &vmstate_icp_server, icp); > } >
On Mon, 17 Jun 2019 15:46:14 +0200 Cédric Le Goater <clg@kaod.org> wrote: > On 17/06/2019 15:25, Greg Kurz wrote: > > kvm_kernel_irqchip is set by xics_kvm_init() which is always called after > > icp_realize() since commit 3f777abc7107 "spapr/irq: add KVM support to the > > 'dual' machine". > > I think we still need it for CPU hotplug. > Yes you're right of course... a comment would probably help > C. > > > > This is thus dead code. Drop it. > > > > Signed-off-by: Greg Kurz <groug@kaod.org> > > --- > > hw/intc/xics.c | 8 -------- > > 1 file changed, 8 deletions(-) > > > > diff --git a/hw/intc/xics.c b/hw/intc/xics.c > > index 29f7d397813e..1dc3a0f12280 100644 > > --- a/hw/intc/xics.c > > +++ b/hw/intc/xics.c > > @@ -351,14 +351,6 @@ static void icp_realize(DeviceState *dev, Error **errp) > > return; > > } > > > > - if (kvm_irqchip_in_kernel()) { > > - icp_kvm_realize(dev, &err); > > - if (err) { > > - error_propagate(errp, err); > > - return; > > - } > > - } > > - > > qemu_register_reset(icp_reset_handler, dev); > > vmstate_register(NULL, icp->cs->cpu_index, &vmstate_icp_server, icp); > > } > > >
diff --git a/hw/intc/xics.c b/hw/intc/xics.c index 29f7d397813e..1dc3a0f12280 100644 --- a/hw/intc/xics.c +++ b/hw/intc/xics.c @@ -351,14 +351,6 @@ static void icp_realize(DeviceState *dev, Error **errp) return; } - if (kvm_irqchip_in_kernel()) { - icp_kvm_realize(dev, &err); - if (err) { - error_propagate(errp, err); - return; - } - } - qemu_register_reset(icp_reset_handler, dev); vmstate_register(NULL, icp->cs->cpu_index, &vmstate_icp_server, icp); }
kvm_kernel_irqchip is set by xics_kvm_init() which is always called after icp_realize() since commit 3f777abc7107 "spapr/irq: add KVM support to the 'dual' machine". This is thus dead code. Drop it. Signed-off-by: Greg Kurz <groug@kaod.org> --- hw/intc/xics.c | 8 -------- 1 file changed, 8 deletions(-)