On 12/14/2011 11:26 AM, Peter Maydell wrote:
On 14 December 2011 17:19, Anthony Liguori<anth...@codemonkey.ws> wrote:
If you have a better suggestion, I'm all for it.
Shrug. I'm not sure what problem you're solving here.
To introduce the infrastructure such that we can have high quality internal
documentation such that I can use that infrastructure with QEMU Object Model.
The bits
of QEMU that are the worst to understand are the ones which aren't
documented at all,
What are the bits that aren't documented at all? An easy answer to that would
be "anything that isn't present on http://wiki.qemu.org/docs-internal";
How does someone help documented the undocumented things? Send a patch adding
gtk-doc style documentation to a file and then once it's merged it will
automagically appear on qemu.org.
I think that goes a long way to solving the problem.
and the bits where the general structure and
underlying assumptions aren't documented. I'm happy to look in a
header file for specific function API info.
I really prefer reading it in the form of a reference document. It also helps
enforce a certain amount of consistency in the format of the documentation which
is hard to do if there isn't some sort of tool running against it.
Regards,
Anthony Liguori
-- PMM
