On Tue, 2012-07-24 at 15:09 -0400, Andrew Kane wrote:
+static int sched_arinc653_domain_set(libxl__gc *gc, uint32_t domid,
+ const libxl_domain_sched_params *scinfo)
+{
+ // Currently, the ARINC 653 scheduler does not take any domain-specific
+ // configuration, so we simply return success.
I think using C (/* */) style for comment is highly recommended, if not
required. :-)
Oops. That's what I get for trusting the editor with comments. =)
:-)
Our thought was to define this following the structure that exists for the
other schedulers, both for consistency and to facilitate future work
on the ARINC 653 scheduler.
Yeah, I got that, and it's not bad thinking actually.
Thinking a bit more about this, right below
libxl_domain_sched_params_set() (in libxl.c) there is another function
called libxl_domain_sched_params_get(), doing pretty much the same
thing, although of course it retrieves instead of setting.
Shouldn't you be doing something similar to that too?
If/when we actually need domain-specific configuration like this,
it would only involve changes in the sched_arinc653_domain_set
function, and wouldn't require any changes to
libxl_domain_sched_params_set.
If the preference is to hold off on implementing a
sched_arinc653_domain_set function until there's actually something
to put in it, I'm happy to change it. =)
It's mostly a matter of taste I guess.
The way I pointed is my preference, but I really don't care that much.
If you send a patch with proper commenting (and perhaps dealing with the
*_get() path), I'll ack it no matter if you have those empty functions
or not... Which will then mean it'll be up to Goerge and Ian (added to
the Cc list) to decide what they like better. :-)
