Andrea Arcangeli wrote:
On Wed, Apr 15, 2009 at 05:43:03PM -0700, Jeremy Fitzhardinge wrote:
Shouldn't that be kmap_atomic's job anyway? Otherwise it would be hard to
No because those are full noops in no-highmem kernels. I commented in
other email why I think it's safe thanks to the wrprotect + smp tlb
flush of the userland PTE.
I think Andrew's query was about data cache synchronization in
architectures with virtually indexed d-cache. On x86 it's a non-issue,
but on architectures for which it is an issue, I assume kmap_atomic does
any necessary cache flushes, as it does tlb flushes on x86 (which may be
none at all, if no mapping actually happens).
J
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
