15 messages in com.xensource.lists.xen-ia64-develRe: [Xen-ia64-devel] [RFC] grant tabl...| From | Sent On | Attachments |
|---|---|---|
| Tian, Kevin | 27 Mar 2006 23:23 | |
| Tian, Kevin | 27 Mar 2006 23:32 | |
| Isaku Yamahata | 28 Mar 2006 01:49 | |
| Tian, Kevin | 28 Mar 2006 03:45 | |
| Isaku Yamahata | 28 Mar 2006 04:26 | |
| Tian, Kevin | 28 Mar 2006 05:03 | |
| Tristan Gingold | 26 Apr 2006 04:44 | |
| Isaku Yamahata | 26 Apr 2006 19:31 | |
| Tristan Gingold | 27 Apr 2006 00:34 | |
| Isaku Yamahata | 27 Apr 2006 01:40 | |
| Tristan Gingold | 27 Apr 2006 02:19 | |
| Isaku Yamahata | 27 Apr 2006 03:10 | |
| Tristan Gingold | 27 Apr 2006 04:01 | |
| Isaku Yamahata | 27 Apr 2006 04:18 | |
| Tristan Gingold | 27 Apr 2006 04:34 |
| Subject: | Re: [Xen-ia64-devel] [RFC] grant table API clean up and performancetuning memo![]() |
|---|---|
| From: | Tristan Gingold (Tris...@bull.net) |
| Date: | 04/27/2006 02:19:41 AM |
| List: | com.xensource.lists.xen-ia64-devel |
Le Jeudi 27 Avril 2006 10:40, Isaku Yamahata a écrit :
On Thu, Apr 27, 2006 at 09:34:54AM +0200, Tristan Gingold wrote:
Le Jeudi 27 Avril 2006 04:31, Isaku Yamahata a écrit : [...]
Two questions: It seems your mechanism still need to trust the domains. So we need to add checks for itc/ptc.
However, can a granted page be mapped several times ? If so, tracking is necessary and has a cost.
Grant table API allows it. Xen tracks it by the counter(act->pin). However current xenLinux maps one-time per a granted page, I believe.
Even if it is mapped only once in guest physical memory, it may still be mapped many times in virtual memory. True ?
Yes, it's true. Usually only __va()'ed address is used.
That's certainly true for vbd. For vnif, if the buffer is allocated from the slab, it may then be wrong. For balloon, this is not true.
I am trying to understand what can be done to avoid global vtlb flush.
Tracking tlb insert is a possible solution, but has an high cost.
Enforcing one insert policy is an easier solution: each page granted may only be mapped once in virtual memory. But this doesn't work for transfer.
I also suppose ballooning is not very frequent and thus global vtlb flush can be used when page are removed. Is it correct ?
Tristan.
_______________________________________________ Xen-ia64-devel mailing list Xen-...@lists.xensource.com http://lists.xensource.com/xen-ia64-devel




