Re: zap_page_range(): TLB flush race

["Stephen C. Tweedie" <sct@redhat.com>]

Hi,

On Mon, Apr 10, 2000 at 04:12:18PM -0700, David S. Miller wrote:
>    On Sun, Apr 09, 2000 at 12:37:05AM +0100, Alan Cox wrote:
>    > 
>    > Basically establish_pte() has to be architecture specific, as some processors
>    > need different orders either to avoid races or to handle cpu specific
>    > limitations.
> 
>    What exactly do different architectures need which set_pte() doesn't 
>    already allow them to do magic in?  
> 
> Doing a properly synchronized PTE update and Cache/TLB flush when the
> mapping can exist on multiple processors is not most efficiently done
> if we take some generic setup.

OK, I'm sure there are optimisation issues, but I was worried about
correctness problems from what Alan said.

--Stephen
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux.eu.org/Linux-MM/