Add `Mapper` methods to map `PageRange` to `PhysFrameRange`

[chbaker0]

Hey, first I want to say this crate is great. It removes a lot of pain from starting with OS dev.

I ran into a minor annoyance: mapping a large range of memory is pretty slow. In my code I'm trying to identity map all the machine's physical memory. It takes a few seconds in a release build and quite a bit longer in a debug build. I suspect a lot of this is from traversing the same tables over and over when repeatedly calling map_to.

I think methods for mapping entire ranges could be faster: instead of a loop that traverses down all levels each iteration, there could be loops at each level.

Is this a reasonable request? I'm interested in writing up a PR if it's something you'd accept.

[josephlr]

This honestly seems quite reasonable (especially if it gives a huge speedup). Algorithmically, I'm not sure about the best way to go about things, but improvements can definitely be made.

My preferred way to do this would be to have a new method: map_range_with_table_flags which has a default implementation based on map_to_with_table_flags, but is overridden by mappers to be more efficient.

Then the following methods can be provided:

• map_range
• identity_map_range

EDIT: For a short-term workaround, using larger 2M/1G pages is a way to very quickly identity map the entire space.

[phil-opp]

My preferred way to do this would be to have a new method: map_range_with_table_flags which has a default implementation based on map_to_with_table_flags, but is overridden by mappers to be more efficient.

Sounds reasonable to me. This method would then behave like the Mapper::map method introduced in #136 and allocate the target frames from the frame allocator, right?