Re: [Qemu-devel] [PATCH v4 03/16] register: Add Memory API glue

[Alistair Francis]

On Fri, Feb 26, 2016 at 9:25 AM, Alex Bennée <address@hidden> wrote:
>
> Alistair Francis <address@hidden> writes:
>
>> From: Peter Crosthwaite <address@hidden>
>>
>> Add memory io handlers that glue the register API to the memory API.
>> Just translation functions at this stage. Although it does allow for
>> devices to be created without all-in-one mmio r/w handlers.
>>
>> Signed-off-by: Peter Crosthwaite <address@hidden>
>> Signed-off-by: Alistair Francis <address@hidden>
>> ---
>>
>>  hw/core/register.c    | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
>>  include/hw/register.h | 31 +++++++++++++++++++++++++++++++
>>  2 files changed, 79 insertions(+)
>>
>> diff --git a/hw/core/register.c b/hw/core/register.c
>> index 7e47df5..9cd50c8 100644
>> --- a/hw/core/register.c
>> +++ b/hw/core/register.c
>> @@ -150,3 +150,51 @@ void register_reset(RegisterInfo *reg)
>>
>>      register_write_val(reg, reg->access->reset);
>>  }
>> +
>> +static inline void register_write_memory(void *opaque, hwaddr addr,
>> +                                         uint64_t value, unsigned size, 
>> bool be)
>> +{
>> +    RegisterInfo *reg = opaque;
>> +    uint64_t we = ~0;
>> +    int shift = 0;
>> +
>> +    if (reg->data_size != size) {
>> +        we = (size == 8) ? ~0ull : (1ull << size * 8) - 1;
>> +        shift = 8 * (be ? reg->data_size - size - addr : addr);
>> +    }
>
> What happens if the user writes too large a value at once to the
> register? I haven't attempted to decode the shift magic going on here
> but perhaps the handling would be clearer if there was a:
>
> if (size > reg->data_size) {
>    ..deal with it..
> } else if (size < data_size ) {
>   ..do other magic..
> }

Good point. I think I have tidied this up.

>
>> +
>> +    assert(size + addr <= reg->data_size);
>
> Why are we asserting expected input conditions after we've done stuff?

I am not sure why this is here. Removed.

>
>> +    register_write(reg, value << shift, we << shift);
>> +}
>> +
>> +void register_write_memory_be(void *opaque, hwaddr addr, uint64_t value,
>> +                              unsigned size)
>> +{
>> +    register_write_memory(opaque, addr, value, size, true);
>> +}
>> +
>> +
>> +void register_write_memory_le(void *opaque, hwaddr addr, uint64_t value,
>> +                              unsigned size)
>> +{
>> +    register_write_memory(opaque, addr, value, size, false);
>> +}
>> +
>> +static inline uint64_t register_read_memory(void *opaque, hwaddr addr,
>> +                                            unsigned size, bool be)
>> +{
>> +    RegisterInfo *reg = opaque;
>> +    int shift = 8 * (be ? reg->data_size - size - addr : addr);
>> +
>
> Well we never have to deal with an over/undersized read? I suspect the
> magic above might not correctly represent some hardware when presented
> with such a thing on the bus.

I think I have fixed this up.

>
>> +    return register_read(reg) >> shift;
>> +}
>> +
>> +uint64_t register_read_memory_be(void *opaque, hwaddr addr, unsigned size)
>> +{
>> +    return register_read_memory(opaque, addr, size, true);
>> +}
>> +
>> +uint64_t register_read_memory_le(void *opaque, hwaddr addr, unsigned size)
>> +{
>> +    return register_read_memory(opaque, addr, size, false);
>> +}
>> diff --git a/include/hw/register.h b/include/hw/register.h
>> index 444239c..9aa9cfc 100644
>> --- a/include/hw/register.h
>> +++ b/include/hw/register.h
>> @@ -69,9 +69,14 @@ struct RegisterAccessInfo {
>>   * @prefix: String prefix for log and debug messages
>>   *
>>   * @opaque: Opaque data for the register
>> + *
>> + * @mem: optional Memory region for the register
>>   */
>>
>>  struct RegisterInfo {
>> +    /* <private> */
>> +    MemoryRegion mem;
>> +
>
> This seems unconnected with adding the helpers. Should it come with the
> original definition or when it actually gets used?

I think it should be added in this patch (although in the next version
it is added in a different struct) as this is starting to add the
memory infrastructure.

Thanks,

Alistair

>
>>      /* <public> */
>>      void *data;
>>      int data_size;
>> @@ -108,4 +113,30 @@ uint64_t register_read(RegisterInfo *reg);
>>
>>  void register_reset(RegisterInfo *reg);
>>
>> +/**
>> + * Memory API MMIO write handler that will write to a Register API register.
>> + *  _be for big endian variant and _le for little endian.
>> + * @opaque: RegisterInfo to write to
>> + * @addr: Address to write
>> + * @value: Value to write
>> + * @size: Number of bytes to write
>> + */
>> +
>> +void register_write_memory_be(void *opaque, hwaddr addr, uint64_t value,
>> +                              unsigned size);
>> +void register_write_memory_le(void *opaque, hwaddr addr, uint64_t value,
>> +                              unsigned size);
>> +
>> +/**
>> + * Memory API MMIO read handler that will read from a Register API register.
>> + *  _be for big endian variant and _le for little endian.
>> + * @opaque: RegisterInfo to read from
>> + * @addr: Address to read
>> + * @size: Number of bytes to read
>> + * returns: Value read from register
>> + */
>> +
>> +uint64_t register_read_memory_be(void *opaque, hwaddr addr, unsigned size);
>> +uint64_t register_read_memory_le(void *opaque, hwaddr addr, unsigned size);
>> +
>>  #endif
>
>
> --
> Alex Bennée
>