Re: [RFC PATCH 08/11] asm-generic/tlb: Track freeing of page-table directories in struct mmu_gather

From: Nicholas Piggin
Date: Mon Aug 27 2018 - 00:45:14 EST

On Fri, 24 Aug 2018 16:52:43 +0100
Will Deacon <will.deacon@xxxxxxx> wrote:

> From: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Some architectures require different TLB invalidation instructions
> depending on whether it is only the last-level of page table being
> changed, or whether there are also changes to the intermediate
> (directory) entries higher up the tree.
> Add a new bit to the flags bitfield in struct mmu_gather so that the
> architecture code can operate accordingly if it's the intermediate
> levels being invalidated.
> Signed-off-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Signed-off-by: Will Deacon <will.deacon@xxxxxxx>

powerpc should be able to move right over to using this rather
than keeping the bit in need_flush_all.

powerpc may be able to use the unmap granule thing to improve
its page size dependent flushes, but it might prefer to go
a different way and track start-end for different page sizes.
I wonder how much of that stuff should go into generic code,
and whether it should instead go into a struct arch_mmu_gather.