[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <PH0PR12MB5481CAA3F57892FF7F05B004DCDF9@PH0PR12MB5481.namprd12.prod.outlook.com>
Date: Wed, 1 Jun 2022 18:58:50 +0000
From: Parav Pandit <parav@...dia.com>
To: Jason Wang <jasowang@...hat.com>
CC: "Michael S. Tsirkin" <mst@...hat.com>,
Eugenio PĂ©rez <eperezma@...hat.com>,
"kvm@...r.kernel.org" <kvm@...r.kernel.org>,
"virtualization@...ts.linux-foundation.org"
<virtualization@...ts.linux-foundation.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"martinh@...inx.com" <martinh@...inx.com>,
Stefano Garzarella <sgarzare@...hat.com>,
"martinpo@...inx.com" <martinpo@...inx.com>,
"lvivier@...hat.com" <lvivier@...hat.com>,
"pabloc@...inx.com" <pabloc@...inx.com>,
Eli Cohen <elic@...dia.com>,
Dan Carpenter <dan.carpenter@...cle.com>,
Xie Yongji <xieyongji@...edance.com>,
Christophe JAILLET <christophe.jaillet@...adoo.fr>,
Zhang Min <zhang.min9@....com.cn>,
Wu Zongyong <wuzongyong@...ux.alibaba.com>,
"lulu@...hat.com" <lulu@...hat.com>,
Zhu Lingshan <lingshan.zhu@...el.com>,
"Piotr.Uminski@...el.com" <Piotr.Uminski@...el.com>,
Si-Wei Liu <si-wei.liu@...cle.com>,
"ecree.xilinx@...il.com" <ecree.xilinx@...il.com>,
"gautam.dawar@....com" <gautam.dawar@....com>,
"habetsm.xilinx@...il.com" <habetsm.xilinx@...il.com>,
"tanuj.kamde@....com" <tanuj.kamde@....com>,
"hanand@...inx.com" <hanand@...inx.com>,
"dinang@...inx.com" <dinang@...inx.com>,
Longpeng <longpeng2@...wei.com>
Subject: RE: [PATCH v4 0/4] Implement vdpasim stop operation
> From: Jason Wang <jasowang@...hat.com>
> Sent: Tuesday, May 31, 2022 10:42 PM
>
> Well, the ability to query the virtqueue state was proposed as another
> feature (Eugenio, please correct me). This should be sufficient for making
> virtio-net to be live migrated.
>
The device is stopped, it won't answer to this special vq config done here.
Programming all of these using cfg registers doesn't scale for on-chip memory and for the speed.
Next would be to program hundreds of statistics of the 64 VQs through giant PCI config space register in some busy polling scheme.
I can clearly see how all these are inefficient for faster LM.
We need an efficient AQ to proceed with at minimum.
> https://lists.oasis-open.org/archives/virtio-comment/202103/msg00008.html
>
> > Once the device is stopped, its state cannot be queried further as device
> won't respond.
> > It has limited use case.
> > What we need is to stop non admin queue related portion of the device.
Powered by blists - more mailing lists