Re: [Qemu-devel] [PATCH 2/2] hw/arm/vexpress: add SP810 to the vexpress

[Aggeler Fabian]

On 17 Jul 2014, at 01:29, Peter Crosthwaite <address@hidden> wrote:

> On Thu, Jul 17, 2014 at 1:05 AM, Alex Bennée <address@hidden> wrote:
>> 
>> Fabian Aggeler writes:
>> 
>>> The SP810, which is present in the Versatile Express motherboards,
>>> allows to set the timing reference to either REFCLK or TIMCLK.
>>> QEMU currently sets the SP804 timer to 1MHz by default. To reflect
>>> this, we set the TimerEn0Sel, TimerEn1Sel, TimerEn2Sel, and
>>> TimerEn3Sel of the system control register (SCCTRL) to TIMCLK (1).
>>> 
>>> Signed-off-by: Fabian Aggeler <address@hidden>
>>> ---
>>> hw/arm/vexpress.c | 11 +++++++++--
>>> 1 file changed, 9 insertions(+), 2 deletions(-)
>>> 
>>> diff --git a/hw/arm/vexpress.c b/hw/arm/vexpress.c
>>> index a88732c..b96c3fd 100644
>>> --- a/hw/arm/vexpress.c
>>> +++ b/hw/arm/vexpress.c
>>> @@ -512,7 +512,7 @@ static pflash_t *ve_pflash_cfi01_register(hwaddr base, 
>>> const char *name,
>>> static void vexpress_common_init(VEDBoardInfo *daughterboard,
>>>                                  MachineState *machine)
>>> {
>>> -    DeviceState *dev, *sysctl, *pl041;
>>> +    DeviceState *dev, *sysctl, *pl041, *sp810;
>>>     qemu_irq pic[64];
>>>     uint32_t sys_id;
>>>     DriveInfo *dinfo;
>>> @@ -575,7 +575,14 @@ static void vexpress_common_init(VEDBoardInfo 
>>> *daughterboard,
>>>     qdev_init_nofail(sysctl);
>>>     sysbus_mmio_map(SYS_BUS_DEVICE(sysctl), 0, map[VE_SYSREGS]);
>>> 
>>> -    /* VE_SP810: not modelled */
>>> +    /* VE_SP810 */
>>> +    sp810 = qdev_create(NULL, "arm_sp810");
> 
> Move the the type definition macro to header as well.
> 
> Regards,
> Peter

Thanks for your comments. I addressed them in v2 which I am going to send 
shortly.

Best,
Fabian

> 
>>> +    /* SP804 is already running at 1MHz (TIMCLK) so SCCTRL TimerEnXSel=1 */
>>> +    qdev_prop_set_uint32(sp810, "sc-ctrl", (1 << 15) | (1 << 17)
>>> +                                          | (1 << 19) | (1 << 21));
>> <snip>
>> 
>> Could the #defines in the first patch be moved into a header and used
>> here rather than manually  setting these bits?
>> 
>> --
>> Alex Bennée