qemu-ppc
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-ppc] [PATCH 2/2] spapr: Define NMI interface


From: Alexey Kardashevskiy
Subject: Re: [Qemu-ppc] [PATCH 2/2] spapr: Define NMI interface
Date: Fri, 28 Mar 2014 00:14:22 +1100
User-agent: Mozilla/5.0 (X11; Linux i686 on x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.4.0

On 03/27/2014 10:58 PM, Paolo Bonzini wrote:
> Il 27/03/2014 03:21, Alexey Kardashevskiy ha scritto:
>> This defines and makes use of an NMI interface in order to support
>> the "nmi" command.
>>
>> Signed-off-by: Alexey Kardashevskiy <address@hidden>
>> ---
>>  hw/ppc/spapr.c | 26 ++++++++++++++++++++++++++
>>  1 file changed, 26 insertions(+)
>>
>> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
>> index 62ddb4d..495fa88 100644
>> --- a/hw/ppc/spapr.c
>> +++ b/hw/ppc/spapr.c
>> @@ -48,6 +48,7 @@
>>  #include "hw/pci/pci.h"
>>  #include "hw/scsi/scsi.h"
>>  #include "hw/virtio/virtio-scsi.h"
>> +#include "hw/nmi.h"
>>
>>  #include "exec/address-spaces.h"
>>  #include "hw/usb.h"
>> @@ -1539,13 +1540,36 @@ static char *spapr_get_fw_dev_path(FWPathProvider
>> *p, BusState *bus,
>>      return NULL;
>>  }
>>
>> +static void spapr_do_nmi(void *arg)
>> +{
>> +    CPUState *cs = arg;
>> +    PowerPCCPU *cpu = POWERPC_CPU(cs);
>> +    CPUPPCState *env = &cpu->env;
>> +
>> +    cpu_synchronize_state(cs);
>> +    env->spr[SPR_SRR0] = env->nip;
>> +    env->spr[SPR_SRR1] = env->msr;
>> +    env->nip = 0x100;
>> +    env->msr = (1ULL << MSR_SF) | (1 << MSR_ME);
>> +    if (env->spr[SPR_LPCR] & LPCR_ILE) {
>> +        env->msr |= 1 << MSR_LE;
>> +    }
>> +}
> 
> 
> I think an interface isn't the right tool here.  You want a method in
> CPUClass, and you also should move the existing code for x86 and s390 to
> target-i386 and target-s390.


I can do that, no big deal (bit afraid there will be some third approach
then :) ), but

1) how many x86 CPUs/families are there to support? On ppc I'll add it for
POWER7/7+/8 families and I am ok.

2) what should "nmi" really do - deliver NMI to the current CPU (as s390
does) or on all CPUs (as x86 does)?



-- 
Alexey



reply via email to

[Prev in Thread] Current Thread [Next in Thread]