Re: [Qemu-devel] [RFC PATCH v2 11/21] ppc/xive: push the EQ data in OS event queue

[Cédric Le Goater]

On 09/19/2017 09:45 AM, David Gibson wrote:
> On Mon, Sep 11, 2017 at 07:12:25PM +0200, Cédric Le Goater wrote:
>> If a triggered event is let through, the Event Queue data defined in
>> the associated IVE is pushed in the in-memory event queue. The latter
>> is a circular buffer provided by the OS using the H_INT_SET_QUEUE_CONFIG
>> hcall, one per target and priority couple. It is composed of Event
>> Queue entries which are 4 bytes long, the first bit being a
>> 'generation' bit and the 31 following bits the EQ Data field.
>>
>> The EQ Data field provides a way to set an invariant logical event
>> source number for an IRQ. It is set with the H_INT_SET_SOURCE_CONFIG
>> hcall.
>>
>> Notification of the CPU will be done in the following patch.
>>
>> Signed-off-by: Cédric Le Goater <address@hidden>
>> ---
>>  hw/intc/spapr_xive.c | 67 
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 67 insertions(+)
>>
>> diff --git a/hw/intc/spapr_xive.c b/hw/intc/spapr_xive.c
>> index 557a7e2535b5..4bc61cfda67a 100644
>> --- a/hw/intc/spapr_xive.c
>> +++ b/hw/intc/spapr_xive.c
>> @@ -175,9 +175,76 @@ static const MemoryRegionOps spapr_xive_tm_ops = {
>>      },
>>  };
>>  
>> +static void spapr_xive_eq_push(XiveEQ *eq, uint32_t data)
>> +{
>> +    uint64_t qaddr_base = (((uint64_t)(eq->w2 & 0x0fffffff)) << 32) | 
>> eq->w3;
>> +    uint32_t qsize = GETFIELD(EQ_W0_QSIZE, eq->w0);
>> +    uint32_t qindex = GETFIELD(EQ_W1_PAGE_OFF, eq->w1);
>> +    uint32_t qgen = GETFIELD(EQ_W1_GENERATION, eq->w1);
>> +
>> +    uint64_t qaddr = qaddr_base + (qindex << 2);
>> +    uint32_t qdata = cpu_to_be32((qgen << 31) | (data & 0x7fffffff));
>> +    uint32_t qentries = 1 << (qsize + 10);
>> +
>> +    if (dma_memory_write(&address_space_memory, qaddr, &qdata, 
>> sizeof(qdata))) {
>> +        qemu_log_mask(LOG_GUEST_ERROR, "%s: failed to write EQ data @0x%"
>> +                      HWADDR_PRIx "\n", __func__, qaddr);
>> +        return;
>> +    }
>> +
>> +    qindex = (qindex + 1) % qentries;
>> +    if (qindex == 0) {
>> +        qgen ^= 1;
>> +        eq->w1 = SETFIELD(EQ_W1_GENERATION, eq->w1, qgen);
>> +    }
>> +    eq->w1 = SETFIELD(EQ_W1_PAGE_OFF, eq->w1, qindex);
>> +}
>> +
>>  static void spapr_xive_irq(sPAPRXive *xive, int srcno)
>>  {
>> +    XiveIVE *ive;
>> +    XiveEQ *eq;
>> +    uint32_t eq_idx;
>> +    uint32_t priority;
>> +
>> +    ive = spapr_xive_get_ive(xive, srcno);
>> +    if (!ive || !(ive->w & IVE_VALID)) {
>> +        qemu_log_mask(LOG_GUEST_ERROR, "XIVE: invalid LISN %d\n", srcno);
>> +        return;
>> +    }
>> +
>> +    if (ive->w & IVE_MASKED) {
>> +        return;
>> +    }
>> +
>> +    /* Find our XiveEQ */
>> +    eq_idx = GETFIELD(IVE_EQ_INDEX, ive->w);
>> +    eq = spapr_xive_get_eq(xive, eq_idx);
>> +    if (!eq) {
>> +        qemu_log_mask(LOG_GUEST_ERROR, "XIVE: No EQ for LISN %d\n", srcno);
>> +        return;
>> +    }
>> +
>> +    if (eq->w0 & EQ_W0_ENQUEUE) {
>> +        spapr_xive_eq_push(eq, GETFIELD(IVE_EQ_DATA, ive->w));
>> +    } else {
>> +        qemu_log_mask(LOG_UNIMP, "XIVE: !ENQUEUE not implemented\n");
>> +    }
>> +
>> +    if (!(eq->w0 & EQ_W0_UCOND_NOTIFY)) {
>> +        qemu_log_mask(LOG_UNIMP, "XIVE: !UCOND_NOTIFY not implemented\n");
>> +    }
>> +
>> +    if (GETFIELD(EQ_W6_FORMAT_BIT, eq->w6) == 0) {
>> +        priority = GETFIELD(EQ_W7_F0_PRIORITY, eq->w7);
>>  
>> +        /* The EQ is masked. Can this happen ?  */
>> +        if (priority == 0xff) {
>> +            return;
> 
> How does the 8-bit priority field here interact with the 3-bit
> priority which selects which EQ to use?

priority OxFF is a special case kept for masking, see the hcall 
h_int_set_source_config. It should never reach the EQ lookup 
routines. So may be an assert would be better here.

C. 

> 
>> +        }
>> +    } else {
>> +        qemu_log_mask(LOG_UNIMP, "XIVE: w7 format1 not implemented\n");
>> +    }
>>  }
>>  
>>  /*
>