Re: [PATCH] 15/18 better pte invalidation

[Paul Mackerras <paulus@samba.org>]

Arnd Bergmann writes:

> 2) s/390 has an instruction called ipte (invalidate page table entry) that
>    is used to get rid of virtual pages. It does two things, first it set
>    the invalid bit in the pte and second it flushes the tlbs for this
>    page on all cpus. Very handy but it requires that the pte it should
>    flush is still valid. The introduction of establish_pte was a step
>    into the right direction but it is defined in mm/memory.c. We need
>    to be able to replace this function with a special s/390 variant.

It's now possible in 2.5 to go from a PTE pointer to the mm and
virtual address that it maps - see ptep_to_mm and ptep_to_address in
include/asm-generic/rmap.h.  I'm planning to use this to do MMU
hashtable management on PPC in set_pte et al. and thus avoid the
second walk of the Linux page tables that we currently do in
flush_tlb_* on PPC.  You could use those functions to let you use the
ipte instruction in set_pte et al. and then make the flush_tlb_*
functions be no-ops.

The ptep_to_mm and ptep_to_address functions came in with rmap, but
all they rely on is having the page->mapping and page->index fields
filled in for pagetable pages.  It should only be a minor addition to
the 2.4 MM system to set those fields, and it would be useful on ppc,
ppc64, s390 and s390x.

Paul.