Re: [Qemu-devel] [Qemu-ppc] [PATCH v9 1/6] ppc: spapr: Handle "ibm, nmi-register" and "ibm, nmi-interlock" RTAS calls

[Aravinda Prasad]

On Tuesday 04 June 2019 08:20 PM, Greg Kurz wrote:
> On Tue, 4 Jun 2019 11:38:31 +0530
> Aravinda Prasad <address@hidden> wrote:
> 
>> On Monday 03 June 2019 04:47 PM, Greg Kurz wrote:
>>> On Mon, 3 Jun 2019 12:12:43 +0200
>>> Greg Kurz <address@hidden> wrote:
>>>   
>>>> On Wed, 29 May 2019 11:10:14 +0530
>>>> Aravinda Prasad <address@hidden> wrote:
>>>>  
>>>>> This patch adds support in QEMU to handle "ibm,nmi-register"
>>>>> and "ibm,nmi-interlock" RTAS calls.
>>>>>
>>>>> The machine check notification address is saved when the
>>>>> OS issues "ibm,nmi-register" RTAS call.
>>>>>
>>>>> This patch also handles the case when multiple processors
>>>>> experience machine check at or about the same time by
>>>>> handling "ibm,nmi-interlock" call. In such cases, as per
>>>>> PAPR, subsequent processors serialize waiting for the first
>>>>> processor to issue the "ibm,nmi-interlock" call. The second
>>>>> processor that also received a machine check error waits
>>>>> till the first processor is done reading the error log.
>>>>> The first processor issues "ibm,nmi-interlock" call
>>>>> when the error log is consumed. This patch implements the
>>>>> releasing part of the error-log while subsequent patch
>>>>> (which builds error log) handles the locking part.
>>>>>
>>>>> Signed-off-by: Aravinda Prasad <address@hidden>
>>>>> Reviewed-by: David Gibson <address@hidden>
>>>>> ---    
>>>>
>>>> The code looks okay but it still seems wrong to advertise the RTAS
>>>> calls to the guest that early in the series. The linux kernel in
>>>> the guest will assume FWNMI is functional, which isn't true until
>>>> patch 6 (yes, migration is part of the feature, it should be
>>>> supported upfront, not fixed afterwards).
>>>>
>>>> It doesn't help much to introduce the RTAS calls early and to
>>>> modify them in the other patches. I'd rather see the rest of
>>>> the code first and a final patch that introduces the fully
>>>> functional RTAS calls and calls spapr_rtas_register().
>>>>  
>>>
>>> Thinking again, you should introduce the "fwnmi-mce" spapr capability in
>>> its own patch first, default to "off" and and have the last patch in the
>>> series to switch the default to "on" for newer machine types only.
>>>
>>> This patch should then only register the RTAS calls if "fwnmi-mcr" is set
>>> to "on".
>>>
>>> This should address the fact that we don't want to expose a partially
>>> implemented FWNMI feature to the guest, and we don't want to support
>>> FWNMI at all with older machine types for the sake of compatibility.  
>>
>> When you say "expose a partially implemented FWNMI feature to the
>> guest", do you mean while debugging/bisect we may end up with exposing
>> the partially implemented FWNMI feature? Otherwise it is expected that
> 
> Yes, we don't want to break someone else's bisect.

ok.

> 
>> QEMU runs with all the 6 patches.
>>
>> If that is the case, I will have the rtas nmi register functionality as
>> the last patch in the series. This way we don't have to have spapr cap
>> turned off first and later turned on. However, as mentioned earlier
>> (when David raised the same concern), use of guest_machine_check_addr
>> may look odd at other patches as it is set only during rtas nmi register.
>>
> 
> Why odd ?

see below

> 
>> Or else, as a workaround, I can return RTAS_OUT_NOT_SUPPORTED for rtas
>> nmi register till the entire functionality is implemented and only in
>> the last patch in the series I will return RTAS_OUT_SUCCESS. This will
>> ensure that we have a logical connection between the patches and the
>> partially implemented fwnmi is not exposed to the guest kernel.
>>
> 
> Not exactly true. FWNMI would be exposed to the guest in the device tree
> and the guest kernel would _just_ fail to set the fwnmi_active global:
> 
>       if (0 == rtas_call(ibm_nmi_register, 2, 1, NULL, system_reset_addr,
>                               machine_check_addr))
>               fwnmi_active = 1;

Sorry for the confusion. I thought the suggestion was to introduce
rtas_ibm_nmi_register() call later in the series, but now I see that I
can still have rtas_ibm_nmi_register(), but have a final patch that
calls spapr_rtas_register().


> 
>> Regards,
>> Aravinda
>>
>>
>>
>>
>>>   
>>>>>  hw/ppc/spapr.c         |    7 +++++
>>>>>  hw/ppc/spapr_rtas.c    |   65 
>>>>> ++++++++++++++++++++++++++++++++++++++++++++++++
>>>>>  include/hw/ppc/spapr.h |    9 ++++++-
>>>>>  3 files changed, 80 insertions(+), 1 deletion(-)
>>>>>
>>>>> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
>>>>> index e2b33e5..fae28a9 100644
>>>>> --- a/hw/ppc/spapr.c
>>>>> +++ b/hw/ppc/spapr.c
>>>>> @@ -1808,6 +1808,11 @@ static void spapr_machine_reset(void)
>>>>>      first_ppc_cpu->env.gpr[5] = 0;
>>>>>  
>>>>>      spapr->cas_reboot = false;
>>>>> +
>>>>> +    spapr->guest_machine_check_addr = -1;
>>>>> +
>>>>> +    /* Signal all vCPUs waiting on this condition */
>>>>> +    qemu_cond_broadcast(&spapr->mc_delivery_cond);
>>>>>  }
>>>>>  
>>>>>  static void spapr_create_nvram(SpaprMachineState *spapr)
>>>>> @@ -3072,6 +3077,8 @@ static void spapr_machine_init(MachineState 
>>>>> *machine)
>>>>>  
>>>>>          kvmppc_spapr_enable_inkernel_multitce();
>>>>>      }
>>>>> +
>>>>> +    qemu_cond_init(&spapr->mc_delivery_cond);
>>>>>  }
>>>>>  
>>>>>  static int spapr_kvm_type(MachineState *machine, const char *vm_type)
>>>>> diff --git a/hw/ppc/spapr_rtas.c b/hw/ppc/spapr_rtas.c
>>>>> index 5bc1a93..e7509cf 100644
>>>>> --- a/hw/ppc/spapr_rtas.c
>>>>> +++ b/hw/ppc/spapr_rtas.c
>>>>> @@ -352,6 +352,38 @@ static void rtas_get_power_level(PowerPCCPU *cpu, 
>>>>> SpaprMachineState *spapr,
>>>>>      rtas_st(rets, 1, 100);
>>>>>  }
>>>>>  
>>>>> +static void rtas_ibm_nmi_register(PowerPCCPU *cpu,
>>>>> +                                  SpaprMachineState *spapr,
>>>>> +                                  uint32_t token, uint32_t nargs,
>>>>> +                                  target_ulong args,
>>>>> +                                  uint32_t nret, target_ulong rets)
>>>>> +{
>>>>> +    hwaddr rtas_addr = spapr_get_rtas_addr();
>>>>> +
>>>>> +    if (!rtas_addr) {
>>>>> +        rtas_st(rets, 0, RTAS_OUT_NOT_SUPPORTED);
>>>>> +        return;
>>>>> +    }
>>>>> +
>>>>> +    spapr->guest_machine_check_addr = rtas_ld(args, 1);
>>>>> +    rtas_st(rets, 0, RTAS_OUT_SUCCESS);
>>>>> +}
>>>>> +
>>>>> +static void rtas_ibm_nmi_interlock(PowerPCCPU *cpu,
>>>>> +                                   SpaprMachineState *spapr,
>>>>> +                                   uint32_t token, uint32_t nargs,
>>>>> +                                   target_ulong args,
>>>>> +                                   uint32_t nret, target_ulong rets)
>>>>> +{
>>>>> +    if (spapr->guest_machine_check_addr == -1) {
>>>>> +        /* NMI register not called */
>>>>> +        rtas_st(rets, 0, RTAS_OUT_PARAM_ERROR);
>>>>> +    } else {
>>>>> +        qemu_cond_signal(&spapr->mc_delivery_cond);
>>>>> +        rtas_st(rets, 0, RTAS_OUT_SUCCESS);
>>>>> +    }
>>>>> +}
>>>>> +
>>>>>  static struct rtas_call {
>>>>>      const char *name;
>>>>>      spapr_rtas_fn fn;
>>>>> @@ -470,6 +502,35 @@ void spapr_load_rtas(SpaprMachineState *spapr, void 
>>>>> *fdt, hwaddr addr)
>>>>>      }
>>>>>  }
>>>>>  
>>>>> +hwaddr spapr_get_rtas_addr(void)
>>>>> +{
>>>>> +    SpaprMachineState *spapr = SPAPR_MACHINE(qdev_get_machine());
>>>>> +    int rtas_node;
>>>>> +    const struct fdt_property *rtas_addr_prop;
>>>>> +    void *fdt = spapr->fdt_blob;
>>>>> +    uint32_t rtas_addr;
>>>>> +
>>>>> +    /* fetch rtas addr from fdt */
>>>>> +    rtas_node = fdt_path_offset(fdt, "/rtas");
>>>>> +    if (rtas_node == 0) {
>>>>> +        return 0;
>>>>> +    }
>>>>> +
>>>>> +    rtas_addr_prop = fdt_get_property(fdt, rtas_node, "linux,rtas-base", 
>>>>> NULL);
>>>>> +    if (!rtas_addr_prop) {
>>>>> +        return 0;
>>>>> +    }
>>>>> +
>>>>> +    /*
>>>>> +     * We assume that the OS called RTAS instantiate-rtas, but some other
>>>>> +     * OS might call RTAS instantiate-rtas-64 instead. This fine as of 
>>>>> now
>>>>> +     * as SLOF only supports 32-bit variant.
>>>>> +     */
>>>>> +    rtas_addr = fdt32_to_cpu(*(uint32_t *)rtas_addr_prop->data);
>>>>> +    return (hwaddr)rtas_addr;
>>>>> +}
>>>>> +
>>>>> +
>>>>>  static void core_rtas_register_types(void)
>>>>>  {
>>>>>      spapr_rtas_register(RTAS_DISPLAY_CHARACTER, "display-character",
>>>>> @@ -493,6 +554,10 @@ static void core_rtas_register_types(void)
>>>>>                          rtas_set_power_level);
>>>>>      spapr_rtas_register(RTAS_GET_POWER_LEVEL, "get-power-level",
>>>>>                          rtas_get_power_level);
>>>>> +    spapr_rtas_register(RTAS_IBM_NMI_REGISTER, "ibm,nmi-register",
>>>>> +                        rtas_ibm_nmi_register);
>>>>> +    spapr_rtas_register(RTAS_IBM_NMI_INTERLOCK, "ibm,nmi-interlock",
>>>>> +                        rtas_ibm_nmi_interlock);
>>>>>  }
>>>>>  
>>>>>  type_init(core_rtas_register_types)
>>>>> diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h
>>>>> index 4f5becf..9dc5e30 100644
>>>>> --- a/include/hw/ppc/spapr.h
>>>>> +++ b/include/hw/ppc/spapr.h
>>>>> @@ -188,6 +188,10 @@ struct SpaprMachineState {
>>>>>       * occurs during the unplug process. */
>>>>>      QTAILQ_HEAD(, SpaprDimmState) pending_dimm_unplugs;
>>>>>  
>>>>> +    /* State related to "ibm,nmi-register" and "ibm,nmi-interlock" calls 
>>>>> */
>>>>> +    target_ulong guest_machine_check_addr;
>>>>> +    QemuCond mc_delivery_cond;
>>>>> +
>>>>>      /*< public >*/
>>>>>      char *kvm_type;This means it isn't related to XIVE it to set
>>>>>      char *host_model;
>>>>> @@ -624,8 +628,10 @@ target_ulong spapr_hypercall(PowerPCCPU *cpu, 
>>>>> target_ulong opcode,
>>>>>  #define RTAS_IBM_CREATE_PE_DMA_WINDOW           (RTAS_TOKEN_BASE + 0x27)
>>>>>  #define RTAS_IBM_REMOVE_PE_DMA_WINDOW           (RTAS_TOKEN_BASE + 0x28)
>>>>>  #define RTAS_IBM_RESET_PE_DMA_WINDOW            (RTAS_TOKEN_BASE + 0x29)
>>>>> +#define RTAS_IBM_NMI_REGISTER                   (RTAS_TOKEN_BASE + 0x2A)
>>>>> +#define RTAS_IBM_NMI_INTERLOCK                  (RTAS_TOKEN_BASE + 0x2B)
>>>>>  
>>>>> -#define RTAS_TOKEN_MAX                          (RTAS_TOKEN_BASE + 0x2A)
>>>>> +#define RTAS_TOKEN_MAX                          (RTAS_TOKEN_BASE + 0x2C)
>>>>>  
>>>>>  /* RTAS ibm,get-system-parameter token values */
>>>>>  #define RTAS_SYSPARM_SPLPAR_CHARACTERISTICS      20
>>>>> @@ -876,4 +882,5 @@ void spapr_check_pagesize(SpaprMachineState *spapr, 
>>>>> hwaddr pagesize,
>>>>>  #define SPAPR_OV5_XIVE_BOTH     0x80 /* Only to advertise on the 
>>>>> platform */
>>>>>  
>>>>>  void spapr_set_all_lpcrs(target_ulong value, target_ulong mask);
>>>>> +uint64_t spapr_get_rtas_addr(void);
>>>>>  #endif /* HW_SPAPR_H */
>>>>>     
>>>>
>>>>  
>>>   
>>
> 

-- 
Regards,
Aravinda