Luigi, On Tue, Sep 09, 2014 at 07:59:42PM +0200, Luigi Rizzo wrote: L> My point is preserving support for out of tree modules, L> and spares (or spare accessors, or the ABI you mention below; L> something that gets you quickly a vendor specific pointer L> from an opaque ifnet) were useful for that. L> L> I think the removal of spares should have happened together L> with the commit of the new mechanism. If it is ready now, L> let's move with it and be done with this discussion. L> L> If not, I would like to bring back the pspares L> with a big note summarizing this discussion, L> and remove then when the new mechanism is ready. L> If i read correctly your comment below about L> the "properly named placeholder" you seem to be ok with that ? It would be absolutely okay if you commit right now a properly named placeholder for your new subsystem, that you work on right now. With the proper name no one will unintentionally hijack it. Would this be a satisfiable solution for you? The suggested ABI mechanism is still under discussion and development. -- Totus tuus, Glebius.Received on Wed Sep 10 2014 - 04:37:42 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:52 UTC