On Sun, Feb 9, 2014 at 9:52 PM, Peter Maydell <peter.maydell@xxxxxxxxxx> wrote:
> On 9 February 2014 02:15, Peter Crosthwaite
> <peter.crosthwaite@xxxxxxxxxx> wrote:
>> On Sat, Feb 1, 2014 at 1:45 AM, Peter Maydell <peter.maydell@xxxxxxxxxx> wrote:
>>> - { .name = "CCSIDR", .cp = 15, .crn = 0, .crm = 0, .opc1 = 1, .opc2 = 0,
>>> + { .name = "CCSIDR", .state = ARM_CP_STATE_BOTH,
>>> + .opc0 = 3, .crn = 0, .crm = 0, .opc1 = 1, .opc2 = 0,
>>
>> Why is the .cp field lost in the conversion to STATE_BOTH?
>
> Because 64 bit sysregs don't have a concept of coprocessor. STATE_BOTH
> means "this is a 64 bit sysreg with a cp15 encoding in the obvious place".
> Anything other than cp15 needs split definitions anyway, so if we required
> the .cp to be specified it would always be '.cp = 15'. Essentially this is
> attempting to keep the length of cpreg definitions down by abbreviating
> parts where there isn't actually any choice.
>
Thanks,
Reviewed-by: Peter Crosthwaite <peter.crosthwaite@xxxxxxxxxx>
> thanks
> -- PMM
>
_______________________________________________
kvmarm mailing list
kvmarm@xxxxxxxxxxxxxxxxxxxxx
https://lists.cs.columbia.edu/cucslists/listinfo/kvmarm
