Re: Properly implement flush_dcache_page in 2.4? (Or is it possible?)

[Jun Sun]

On Fri, May 30, 2003 at 07:09:29PM +0100, Russell King wrote:
> On Fri, May 30, 2003 at 10:32:54AM -0700, Jun Sun wrote:
> > So my question is: how other CPU arches with the same problem
> > implement flush_dcache_page()?  Flushing the whole cache? Or
> > have a broken implementation and pretend it is OK?  :)
> 
> See __flush_dcache_page() in arch/arm/mm/fault-armv.c in 2.5.70.
>

Is this routine tested to be working?  At least passing a page
index as a full virtual address to flush_cache_page() looks suspicious.

In addition, I am not sure if the vma struct will show up in the
"shared" list _if_ the page is only mapped in one user process and
in kernel (for example, those pages you obtain through get_user_pages()
call).

I am not familiar with 2.5 kernel.  I was under impression that reverse
page mapping might provide an easy solution to this problem.

Jun