Re: [Qemu-devel] [PATCH V9 3/4] pvrdma: initial implementation

2018-02-02 Thread Dotan Barak via Qemu-devel 
Reviewed-by: Dotan Barak 

  From: Marcel Apfelbaum 
 To: qemu-devel@nongnu.org 
Cc: ehabk...@redhat.com; yuval.sh...@oracle.com; mar...@redhat.com; 
pbonz...@redhat.com; m...@redhat.com; coh...@redhat.com; dotan...@yahoo.com
 Sent: Thursday, February 1, 2018 10:55 PM
 Subject: [Qemu-devel] [PATCH V9 3/4] pvrdma: initial implementation
   
From: Yuval Shaia 

PVRDMA is the QEMU implementation of VMware's paravirtualized RDMA device.
It works with its Linux Kernel driver AS IS, no need for any special guest
modifications.

While it complies with the VMware device, it can also communicate with bare
metal RDMA-enabled machines and does not require an RDMA HCA in the host, it
can work with Soft-RoCE (rxe).

It does not require the whole guest RAM to be pinned allowing memory
over-commit and, even if not implemented yet, migration support will be
possible with some HW assistance.

Signed-off-by: Yuval Shaia 
Signed-off-by: Marcel Apfelbaum

Re: [Qemu-devel] [PATCH V9 3/4] pvrdma: initial implementation

2018-02-02 Thread Marcel Apfelbaum 
On 02/02/2018 14:08, Dotan Barak wrote:
> Reviewed-by: Dotan Barak 
> 

The Mellanox review for the RDMA code is very much appreciated!

Thanks Dotan, we know you put much effort into it and the V9
re-spin quality just went up :)
Marcel


> 
> *From:* Marcel Apfelbaum 
> *To:* qemu-devel@nongnu.org
> *Cc:* ehabk...@redhat.com; yuval.sh...@oracle.com; mar...@redhat.com; 
> pbonz...@redhat.com; m...@redhat.com;
> coh...@redhat.com; dotan...@yahoo.com
> *Sent:* Thursday, February 1, 2018 10:55 PM
> *Subject:* [Qemu-devel] [PATCH V9 3/4] pvrdma: initial implementation
> 
> From: Yuval Shaia mailto:yuval.sh...@oracle.com>>
> 
> PVRDMA is the QEMU implementation of VMware's paravirtualized RDMA device.
> It works with its Linux Kernel driver AS IS, no need for any special guest
> modifications.
> 
> While it complies with the VMware device, it can also communicate with bare
> metal RDMA-enabled machines and does not require an RDMA HCA in the host, it
> can work with Soft-RoCE (rxe).
> 
> It does not require the whole guest RAM to be pinned allowing memory
> over-commit and, even if not implemented yet, migration support will be
> possible with some HW assistance.
> 
> Signed-off-by: Yuval Shaia  >
> Signed-off-by: Marcel Apfelbaum mailto:mar...@redhat.com>>
> 
> 
>