Re: [Qemu-devel] [PATCH] hw/arm/virt: add 2.8 machine type

2016-10-04 Thread Andrew Jones 
On Tue, Oct 04, 2016 at 02:32:36PM +0100, Peter Maydell wrote:
> On 23 September 2016 at 15:41, Andrew Jones  wrote:
> > Signed-off-by: Andrew Jones 
> >
> > ---
> >
> > This will conflict with Wei's PMU property series (which is why I
> > asked him to add it to that series...), but anyway, it'll be a minor
> > resolution to make.
> > ---
> 
> There wasn't a textual conflict but it (slightly belatedly)
> occured to me that perhaps the "support KVM GICv3 ITS" patches
> which went in to master in the same pullreq as this might
> need some tweaks so the ITS doesn't get used in the 2.7
> machine. Would you or Eric A like to have a look?
>

Good call. We should avoid guest visible changes, even when
the changes complete already exposed devices. I'll send a
patch.

Thanks,
drew

Re: [Qemu-devel] [PATCH] hw/arm/virt: add 2.8 machine type

2016-10-04 Thread Peter Maydell 
On 23 September 2016 at 15:41, Andrew Jones  wrote:
> Signed-off-by: Andrew Jones 
>
> ---
>
> This will conflict with Wei's PMU property series (which is why I
> asked him to add it to that series...), but anyway, it'll be a minor
> resolution to make.
> ---

There wasn't a textual conflict but it (slightly belatedly)
occured to me that perhaps the "support KVM GICv3 ITS" patches
which went in to master in the same pullreq as this might
need some tweaks so the ITS doesn't get used in the 2.7
machine. Would you or Eric A like to have a look?

thanks
-- PMM

Re: [Qemu-devel] [PATCH] hw/arm/virt: add 2.8 machine type

2016-09-29 Thread Peter Maydell 
On 23 September 2016 at 07:41, Andrew Jones  wrote:
> Signed-off-by: Andrew Jones 
>
> ---
>
> This will conflict with Wei's PMU property series (which is why I
> asked him to add it to that series...), but anyway, it'll be a minor
> resolution to make.
> ---



Applied to target-arm.next, thanks.

-- PMM

[Qemu-devel] [PATCH] hw/arm/virt: add 2.8 machine type

2016-09-23 Thread Andrew Jones 
Signed-off-by: Andrew Jones 

---

This will conflict with Wei's PMU property series (which is why I
asked him to add it to that series...), but anyway, it'll be a minor
resolution to make.
---
 hw/arm/virt.c | 19 +--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index a193b5a95b3f..2ead58db8ff9 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -1479,7 +1479,7 @@ static void machvirt_machine_init(void)
 }
 type_init(machvirt_machine_init);
 
-static void virt_2_7_instance_init(Object *obj)
+static void virt_2_8_instance_init(Object *obj)
 {
 VirtMachineState *vms = VIRT_MACHINE(obj);
 
@@ -1512,10 +1512,25 @@ static void virt_2_7_instance_init(Object *obj)
 "Valid values are 2, 3 and host", NULL);
 }
 
+static void virt_machine_2_8_options(MachineClass *mc)
+{
+}
+DEFINE_VIRT_MACHINE_AS_LATEST(2, 8)
+
+#define VIRT_COMPAT_2_7 \
+HW_COMPAT_2_7
+
+static void virt_2_7_instance_init(Object *obj)
+{
+virt_2_8_instance_init(obj);
+}
+
 static void virt_machine_2_7_options(MachineClass *mc)
 {
+virt_machine_2_8_options(mc);
+SET_MACHINE_COMPAT(mc, VIRT_COMPAT_2_7);
 }
-DEFINE_VIRT_MACHINE_AS_LATEST(2, 7)
+DEFINE_VIRT_MACHINE(2, 7)
 
 #define VIRT_COMPAT_2_6 \
 HW_COMPAT_2_6
-- 
2.7.4