On Mon, Aug 19, 2019 at 12:33:31PM -0400, Pavel Tatashin wrote: > On Mon, Aug 19, 2019 at 11:58 AM Mark Rutland <mark.rutland@xxxxxxx> wrote: > > On Fri, Aug 16, 2019 at 10:46:18PM -0400, Pavel Tatashin wrote: > > > trans_table_create_copy() and trans_table_map_page() are going to be > > > the basis for public interface of new subsystem that handles page > > > tables for cases which are between kernels: kexec, and hibernate. > > > > While the architecture uses the term 'translation table', in the kernel > > we generally use 'pgdir' or 'pgd' to refer to the tables, so please keep > > to that naming scheme. > > The idea is to have a unique name space for new subsystem of page > tables that are used between kernels: > between stage 1 and stage 2 kexec kernel, and similarly between > kernels during hibernate boot process. > > I picked: "trans_table" that stands for transitional page table: > meaning they are used only during transition between worlds. > > All public functions in this subsystem will have trans_table_* prefix, > and page directory will be named: "trans_table". If this is confusing, > I can either use a different prefix, or describe what "trans_table" > stand for in trans_table.h/.c Ok. I think that "trans_table" is unfortunately confusing, as it clashes with the architecture terminology, and differs from what we have elsewhere. I think that "trans_pgd" would be better, as that better aligns with what we have elsewhere, and avoids the ambiguity. Thanks, Mark. _______________________________________________ kexec mailing list kexec@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/kexec
