On 15 August 2016 at 12:34, Benjamin Herrenschmidt
<b...@kernel.crashing.org> wrote:
> On Mon, 2016-08-15 at 11:02 +0100, Peter Maydell wrote:
>> Some of these attributes are determined before the MMU
>> (eg "user", "requester ID"), and some of them are (for
>> a particular CPU architecture) determined by the MMU
>> from the translation tables ("secure"). From the point of
>> view of the CPU as a whole (or other transaction master)
>> they're all outputs, though.
>
> So you suggest I just print out all the fields as the output
> of translate along with the translated address ?Certainly I think the QMP protocol command should just return all the info (I think current best-practice is to implement new HMP commands as wrappers round an equivalent QMP command). At that point we get into UI design questions about whether we should just print everything for the HMP command output, or have it default to just-address with an option to print with the attributes. I don't have a strong opinion there. thanks -- PMM
