On 4 January 2014 13:32, Peter Crosthwaite <peter.crosthwaite@xxxxxxxxxx> wrote:
> On Mon, Dec 23, 2013 at 8:49 AM, Peter Maydell <peter.maydell@xxxxxxxxxx> wrote:
>> +static const ARMCPRegInfo v8_cp_reginfo[] = {
>> + /* Minimal set of EL0-visible registers. This will need to be expanded
>> + * significantly for system emulation of AArch64 CPUs.
>> + */
>> + { .name = "NZCV", .state = ARM_CP_STATE_AA64,
>> + .opc0 = 3, .opc1 = 3, .opc2 = 0, .crn = 4, .crm = 2,
>> + .access = PL0_RW, .type = ARM_CP_NZCV },
>> + {. name = "FPCR", .state = ARM_CP_STATE_AA64,
>> + .opc0 = 3, .opc1 = 3, .opc2 = 0, .crn = 4, .crm = 4,
>> + .access = PL0_RW, .readfn = aa64_fpcr_read, .writefn = aa64_fpcr_write },
>> + {. name = "FPSR", .state = ARM_CP_STATE_AA64,
>> + .opc0 = 3, .opc1 = 3, .opc2 = 1, .crn = 4, .crm = 4,
>> + .access = PL0_RW, .readfn = aa64_fpsr_read, .writefn = aa64_fpsr_write },
>> + /* This claims a 32 byte cacheline size for icache and dcache, VIPT icache.
>> + * It will eventually need to have a CPU-specified reset value.
>> + */
>
> Not sure why this can't be solved short term as there is already the
> ctr field in cpu state struct as used by 32.
I hadn't noticed that the register was aliased to the CTR AArch32 register.
NB that we can't just share a single reginfo for AArch64 and AArch32
though -- the permissions are different since in AArch32 this is accessible
only in EL1 and above.
My aim here was mostly to set out a framework for how the system
register integration will work with system emulation, but just to
provide low-effort implementations of the EL0-visible regs for
the initial user-emulation patchset. (In the SuSE patchset these are
actually just all implemented as special cases inline in translate-a64.c,
so to some extent this array is just those special cases converted over
to reginfo.)
thanks
-- PMM
_______________________________________________
kvmarm mailing list
kvmarm@xxxxxxxxxxxxxxxxxxxxx
https://lists.cs.columbia.edu/cucslists/listinfo/kvmarm
