Re: [Qemu-devel] [RFC 1/5] memory: Define API for MemoryRegionOps to take attrs and return status

[Edgar E. Iglesias]

On Fri, Mar 27, 2015 at 02:16:53PM +0100, Paolo Bonzini wrote:
> 
> 
> On 27/03/2015 13:32, Edgar E. Iglesias wrote:
> >>> Is this related to masters relying on the memory frameworks magic
> >>> handling of unaliged accesses?
> >>
> >> Not necessarily, you can get the same just by doing a large write that
> >> spans multiple MemoryRegions.  See the loop in address_space_rw.
> > 
> > Right, this is another case of "magic" memory handling that allows masters
> > to issue unnatural transactions and rely on the memory framework to
> > split things up.
> > In these cases aren't the masters trading accuracy (including error
> > handling accuracy) for performance or model simplicity?
> 
> Yes.  There are no "natural" transactions beyond 32 or 64-bit accesses.
> 
> > It could maybe be useful to have a flag so masters can say one of the
> > following (could be encoded in the memattrs):
> > 1. Stop at first error and return.
> > 2. Keep going after errors and give me the OR result of all errors.
> 
> It could just be a length pointer in the same vein as
> address_space_map's.  If NULL, keep going.  If not NULL, stop and return.
> 

Good point, thanks.

Cheers,
Edgar