[dpdk-dev] [PATCH] fm10k: support XEN domain0

[He, Shaopeng]

Hi Stephen,

Do you see any problem of preventing your patch of "xen: allow choosing dom0 support at runtime" into upcoming 2.1 release? It will be better that fm10k's XEN support can base on your modifications, and I will rework this fm10k patch.

But in case your patch cannot make it into 2.1, to enable users to use fm10k with XEN domain0 in next release, should we enable this function by using current approach first?

Thanks,
--Shaopeng

-----Original Message-----
From: Liu, Jijiang 
Sent: Monday, May 18, 2015 10:23 AM
To: Stephen Hemminger; He, Shaopeng; Thomas Monjalon
Cc: dev at dpdk.org
Subject: RE: [dpdk-dev] [PATCH] fm10k: support XEN domain0

Hi guys,

> -----Original Message-----
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen Hemminger
> Sent: Saturday, May 16, 2015 7:58 AM
> To: He, Shaopeng
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] fm10k: support XEN domain0
> 
> On Fri, 15 May 2015 16:56:02 +0800
> Shaopeng He <shaopeng.he at intel.com> wrote:
> 
> > fm10k was failing to run in XEN domain0, as the physical memory for 
> > DMA should be allocated and translated in a different way for XEN 
> > domain0. So
> > rte_memzone_reserve_bounded() should be used for DMA memory 
> > allocation, and rte_mem_phy2mch() should be used for DMA memory 
> > address translation to support running fm10k PMD in XEN domain0.
> >
> > Signed-off-by: Shaopeng He <shaopeng.he at intel.com>
> 
> I agree with Thomas that this code has spread everywhere and should be 
> in a common spot.
> 
> Also, we discovered as part of the Xen net-front driver that it should 
> be a runtime determination, not a config option!

I also agree that it should be in a common spot.
But  it had better to apply the following Stephen's patch first. If so, Shaopeng just use the common function in the patch, which would be good.  
http://dpdk.org/ml/archives/dev/2015-March/014992.html