[PATCH 20/19] ARM: LPAE: Invalidate the TLB before freeing the PMD

[Catalin Marinas]

On Wed, 2011-05-11 at 15:00 +0100, Russell King - ARM Linux wrote:
> On Wed, May 11, 2011 at 02:40:49PM +0100, Catalin Marinas wrote:
> > On Wed, 2011-05-11 at 11:54 +0100, Russell King - ARM Linux wrote:
> > > One of the points about the shootdown interface is that it batches things
> > > up.  So what's wrong with:
> > >
> > > static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmdp,
> > >         unsigned long addr)
> > > {
> > > #ifdef CONFIG_ARM_LPAE
> > >         tlb_add_flush(tlb, addr);
> > >         tlb_remove_page(tlb, virt_to_page(pmdp));
> > > #endif
> > > }
> > >
> > > and leave the tlb invalidate and actual page freeing to the batching code
> > > to deal with?
> >
> > There isn't a big overhead with my initial code as a pmd covers 1GB and
> > we only have 1 or 2 pmds per process that we can free.
> >
> > Is there any room for optimising the mmu_gather range? I think this only
> > matters for case 1 in your tlb_flush() comment - unmapping a page range
> > with a few pages in one pmd and a few other pages in the next pmd we get
> > over 1GB range when we actually only need to flush the TLB for a few
> > pages.
> 
> One of the points is to keep the code as similar to other architectures
> so that the folk who are working on consolidating this stuff across other
> architectures don't have to wonder why ARM is _unnecessarily_ doing things
> differently.
> 
Actually Peter Zijlstra's proposal uses a tlb_track_range() function
which takes start and end range arguments.

But I'm fine with your variant for now.

-- 
Catalin