Re: [RFC PATCH v2 08/22] target/arm: Handle IS/FS in ISR_EL1 for NMI

[Jinjie Ruan]

On 2024/2/22 5:36, Richard Henderson wrote:
> On 2/21/24 03:08, Jinjie Ruan via wrote:
>> Add IS and FS bit in ISR_EL1 and handle the read according to whether the
>> NMI is IRQ or FIQ.
>>
>> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
>> ---
>>   target/arm/cpu.h    | 2 ++
>>   target/arm/helper.c | 9 +++++++++
>>   2 files changed, 11 insertions(+)
>>
>> diff --git a/target/arm/cpu.h b/target/arm/cpu.h
>> index 051e589e19..e2d07e3312 100644
>> --- a/target/arm/cpu.h
>> +++ b/target/arm/cpu.h
>> @@ -1476,6 +1476,8 @@ FIELD(CPTR_EL3, TCPAC, 31, 1)
>>   #define CPSR_N (1U << 31)
>>   #define CPSR_NZCV (CPSR_N | CPSR_Z | CPSR_C | CPSR_V)
>>   #define CPSR_AIF (CPSR_A | CPSR_I | CPSR_F)
>> +#define ISR_FS (1U << 9)
>> +#define ISR_IS (1U << 10)
>>     #define CPSR_IT (CPSR_IT_0_1 | CPSR_IT_2_7)
>>   #define CACHED_CPSR_BITS (CPSR_T | CPSR_AIF | CPSR_GE | CPSR_IT |
>> CPSR_Q \
>> diff --git a/target/arm/helper.c b/target/arm/helper.c
>> index 0bd7a87e51..62c8e5d611 100644
>> --- a/target/arm/helper.c
>> +++ b/target/arm/helper.c
>> @@ -2022,6 +2022,10 @@ static uint64_t isr_read(CPUARMState *env,
>> const ARMCPRegInfo *ri)
>>           if (cs->interrupt_request & CPU_INTERRUPT_HARD) {
>>               ret |= CPSR_I;
>>           }
>> +
>> +        if ((cs->interrupt_request & CPU_INTERRUPT_NMI) &&
>> env->nmi_is_irq) {
>> +            ret |= ISR_IS;
>> +        }
>>       }
>>         if (hcr_el2 & HCR_FMO) {
>> @@ -2032,6 +2036,11 @@ static uint64_t isr_read(CPUARMState *env,
>> const ARMCPRegInfo *ri)
>>           if (cs->interrupt_request & CPU_INTERRUPT_FIQ) {
>>               ret |= CPSR_F;
>>           }
>> +
>> +        if ((cs->interrupt_request & CPU_INTERRUPT_NMI) &&
>> +            (!env->nmi_is_irq)) {
>> +            ret |= ISR_FS;
>> +        }
>>       }
> 
> The external CPU_INTERRUPT_NMI will never signal FIQ.
> 
> With CPU_INTERRUPT_NMI, both CPSR_I and ISR_IS must be set.

Right! The IRQ and NMI signal are both set when it is a NMI.
> 
> Missing is the handling of HCRX_EL2.{VFNMI,VINMI} to signal vFIQ and
> vIRQ with superpriority.  Unless I missed it, I don't see HCRX_EL2
> adjusted for FEAT_NMI at all.

Sorry! The vNMI has not yet been implemented.

The HCRX_EL2.TALLINT has yet been handled.

> 
> 
> r~
>