On 29.11.2023 10:35, Federico Serafini wrote: > I take this opportunity to inform that we are really close to the end > with Rule 8.3 for x86, this is the situation: > - do_multicall(), Stefano sent a patch; > - guest_walk_tables(), Andrew will take care of it; > - xenmem_add_to_physmap_one(), this is the last one. > > For the latter, I see you (x86) share the declaration with ARM, > where "gfn" is used for the last parameter instead of "gpfn". > Do you agree in changing the name in the definition from "gpfn" > to "gfn"?
Yes. > If you agree, do you have any suggestions on how to rename > the local variable "gfn"? Considering its exclusive use for the XENMAPSPACE_gmfn case, I'm inclined to suggest "gmfn", despite this being a term we generally try to get rid of. Maybe Andrew or Roger have a better suggestion. Along with renaming "gpfn" it would be nice (for consistency) to also rename "old_gpfn" at the same time. Jan
