[ofa-general] Re: [RFC] OpenSM vendor layer

[Sasha Khapyorsky]

On 09:04 Mon 09 Feb     , Ira Weiny wrote:
> > 
> > Actually, although more disruptive, it might be cleaner (and safer in
> > the long run) to add to the vendor API. There could be additional osm
> > vendor APIs for pkeys and gids and these could return some suitable
> > IB_ error from ib_types in vendor layers where they are unimplemented.
> > IB_UNSUPPORTED looks good to me. I'm likely to head down this approach
> > unless I hear otherwise.
> 
> This sounds more reasonable to me, better to suffer now than later...

I don't see how it is "safer" in the long run than just extending.

Adding new APIs now will require adding this to another vendor
implementations as well (without actual possibility to test :( ).
Extending osm_vendor_get_all_port_attr() only requires fixing port_array
initializations (I guess it is 3-5 places in total in opensm and ibutils
trees) and with other vendor implementation will work automatically as
"unsupported" - no pkey table will be returned.

I'm not yet saying that following this approach we are opening way for
adding various new "doesn't make sense" API call for each port/whatever
attribute.... :)

Sasha