> -----Original Message----- > From: Andrii Anisov [mailto:andrii.ani...@gmail.com] > Sent: 16 February 2017 16:22 > To: Paul Durrant <paul.durr...@citrix.com> > Cc: xen-de...@lists.xenproject.org; andrii_ani...@epam.com; Andrew > Cooper <andrew.coop...@citrix.com>; George Dunlap > <george.dun...@citrix.com>; Ian Jackson <ian.jack...@citrix.com>; > jbeul...@suse.com; konrad.w...@oracle.com; sstabell...@kernel.org; Tim > (Xen.org) <t...@xen.org>; Wei Liu <wei.l...@citrix.com> > Subject: Re: [RFC 2/6] rangeset_destroy() refactoring > > > What use are rangesets if the implementation doesn't control the list/tree? > How on earth would you implement an allocation function otherwise? > Just to be on the same page, my understanding of the rangesets is as > following: > > - Currently the `struct rangeset` is a list of `ranges`. This list > head is a `range_list` of `struct rangeset`. Currently `range_list` > manipulations are not protected by any locks. IMO this is the core > functionality of the rangeset. > > - Also there is another list head `rangeset_list` inside `struct > rangeset`. It is used to link a rangeset to an external list of > rangesets. This is protected by spinlocks now. IMO this functionality > is odd to the rangeset itself.
Ok. Thanks for the clarification. Yes, that second list_head does not strictly belong inside the rangeset structure itself. I guess it could live in a 'domain_rangeset' wrapper structure. Paul > > Sincerely, > Andrii Anisov. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel
