Re: patch to trim page table cache

Chris Wedgwood (chris@cybernet.co.nz)
Mon, 20 Jul 1998 10:35:38 +1200


On Sun, Jul 19, 1998 at 04:23:46AM +0200, Ove Ewerlid wrote:

> Please consider adding the ability to optionally disable global IPI
> delivery for locked CPU's. A CPU, running hard realtime algorithms that
> fit in L1, may not need these.

Good point. Is anyone actually doing anything cute along these lines, real
time signal processing perhaps?

> Thus, only invalidate the TLB when the CPU is exiting from the locked
> mode.

If some kind of TLB invalidation is required while the CPU is locked, when
it exists, the entire TLB will have to be dumped, keeping track only of the
pages which have changed would be too expensive?

-cw

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html