Re: [Qemu-devel] [RFC v2 00/20] Memory API

[Avi Kivity <avi@redhat.com>]

On 06/27/2011 06:37 PM, Anthony Liguori wrote:
> On 06/27/2011 08:21 AM, Avi Kivity wrote:
>> As expected, this is taking longer than expected, so I'm releasing 
>> something
>> less complete than I'd have liked.  Not even all of the PC machine is
>> converted, but the difficult parts are (cirrus).  It appears to work 
>> well.
>>
>> The major change compared to v1 is the introduction of
>> memory_region_init_alias(), which defines a memory region in terms of 
>> another.
>> With the current API, the ability to alias is provided by address 
>> arithmetic
>> on ram_addr_t:
>>
>>    ram_addr = qemu_ram_alloc(...);
>>    cpu_register_physical_memory(..., ram_addr, size, ...);
>>    /* alias: */
>>    cpu_register_physical_memory(..., ram_addr + offset, another_size, 
>> ...);
>>
>> With the new API, you have to create an alias:
>>
>>    memory_region_init_ram(&mem, ...);
>>    memory_region_register_subregion(...,&mem);
>>    /* alias: */
>>    memory_region_init_alias(&alias, ...,&mem, offset, another_size);
>>    memory_region_register_subregion(...,&alias);
>
> What's the rationale for explicit aliasing verses registering the same 
> region to two different address spaces?

Two:

- an alias is usually just a portion of the original region (vga 
windows; the two main RAM sections below and above 4GB); so we need an 
object to hold the differences
- easier for the implementation (a MemoryRegion object can be in just 
one child list instead of many)

While I can work around the implementation difficulties, there's no 
point since the alias is usually not identical to the origin, so we need 
a new object anyway.

(ram_addr_t works around that by carrying the offset in the handle, and 
letting the caller carry the size).

-- 
error compiling committee.c: too many arguments to function