On 26.12.2011, at 14:22, Avi Kivity wrote: > On 12/20/2011 11:21 AM, Paul Mackerras wrote: >> This moves gfn_to_memslot(), and the functions it calls, that is, >> search_memslots() and __gfn_to_memslot(), from kvm_main.c to kvm_host.h >> so that gfn_to_memslot() can be called from non-modular code even >> when KVM is a module. On powerpc, the Book3S HV style of KVM has >> code that is called from real mode which needs to call gfn_to_memslot() >> and thus needs this. (Module code is allocated in the vmalloc region, >> which can't be accessed in real mode.) >> >> With this, we can remove builtin_gfn_to_memslot() from book3s_hv_rm_mmu.c >> and thus eliminate a little bit of duplication. > > Those functions are too big to be inlined IMO. How about moving them to > another C file, and making it builtin for ppc? > > The only issue is what to call it. virt/kvm/builtin-for-ppc seems silly. Yeah - and it makes it pretty confusing to find the functions then. > Or we could move the implementation into a header file, with an extra __ > prefix, and have the C stubs call those inlines, so we have exactly on > instantiation. Your real mode code can then call the inlines. I like this version. That way everyone should be happy :) Alex -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
