Re: [PATCH 0/2] ksm: take dirty bit as reference to avoid volatilepages

From: Izik Eidus
Date: Mon Mar 28 2011 - 11:23:40 EST


On 03/28/2011 04:14 PM, Nai Xia wrote:
Currently, ksm uses page checksum to detect volatile pages. Izik Eidus
suggested that we could use pte dirty bit to optimize. This patch series
adds this new logic.


Hi,

One small note:
When kvm will use ksm on intel cpu with extended page tables support, the cpu won`t track
dirty bit, therefore the calc_hash() logic should be used in such cases
(untill intel will fadd this support in their cpus)...

Moreover I think that even though that AMD nested page tables does update dirty bit, you still need
to sync it with the host page table using mmu notifiers ?

(Not that on regular application use case of ksm any of this should be an issue)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/