|
From: | Frederic Konrad |
Subject: | Re: [Qemu-devel] [RFC PATCH V6 15/18] cpu: introduce tlb_flush*_all. |
Date: | Fri, 26 Jun 2015 18:35:42 +0200 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.5.0 |
On 26/06/2015 18:31, Paolo Bonzini wrote:
On 26/06/2015 18:30, Frederic Konrad wrote:Yes this is not the case as I implemented it. The rest of the TB will be executed before the tlb_flush work really happen. The old version did this, was slow and was a mess (if two VCPUs want to tlb_flush at the same time and an other tlb_flush_page.. it becomes tricky..)Have you tried implementing the solution based on cpu->halted?
You mean based on cpu_has_work?Yes it was a little painfull (eg: it required cpu to be halted.. but maybe it's what you
were suggesting?)
I think it's not really terrible if the other VCPU execute some stuff before doing the tlb_flush.? So the solution would be only to cut the TranslationBlock after instruction which require a tlb_flush?Yes, this is required too. Paolo
[Prev in Thread] | Current Thread | [Next in Thread] |