Re: [dpdk-dev] [PATCH] eventdev: event device to contain rte device holder

[Nipun Gupta <nipun.gupta@nxp.com>]

Hi Stephen,

> -----Original Message-----
> From: Stephen Hemminger [mailto:stephen@networkplumber.org]
> Sent: Saturday, February 18, 2017 1:00
> To: Nipun Gupta <nipun.gupta@nxp.com>
> Cc: dev@dpdk.org; Hemant Agrawal <hemant.agrawal@nxp.com>;
> jerin.jacob@caviumnetworks.com; bruce.richardson@intel.com;
> gage.eads@intel.com; harry.van.haaren@intel.com
> Subject: Re: [dpdk-dev] [PATCH] eventdev: event device to contain rte device
> holder
> 
> On Thu, 16 Feb 2017 16:22:29 +0530
> Nipun Gupta <nipun.gupta@nxp.com> wrote:
> 
> > Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
> >
> > rte_device is a generic device which is available to the applications
> > and EAL. This patch replaces rte_pci_device in 'struct rte_eventdev'
> > and in 'struct rte_event_dev_info' with common rte_device.
> > ---
> >  drivers/event/skeleton/skeleton_eventdev.c | 2 +-
> >  lib/librte_eventdev/rte_eventdev.c         | 6 +++---
> >  lib/librte_eventdev/rte_eventdev.h         | 6 +++---
> >  3 files changed, 7 insertions(+), 7 deletions(-)
> >
> > diff --git a/drivers/event/skeleton/skeleton_eventdev.c
> b/drivers/event/skeleton/skeleton_eventdev.c
> > index dee0faf..770dce3 100644
> > --- a/drivers/event/skeleton/skeleton_eventdev.c
> > +++ b/drivers/event/skeleton/skeleton_eventdev.c
> > @@ -383,7 +383,7 @@
> >  	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> >  		return 0;
> >
> > -	pci_dev = eventdev->pci_dev;
> > +	pci_dev = RTE_DEV_TO_PCI(eventdev->dev);
> 
> How will this work when there are more than just PCI devices?
> For example, upcoming patches will add rte_vmbus_device.  There is no
> run time type checking in C.

This is within the device driver. AFAIU, a driver will be based on a single bus and
it will get its own device by calling the right conversion function.
So, if a new bus like vmbus is there shall be a corresponding RTE_DEV_TO_xyDEV
to get the 'devices on vmbus' from 'rte_device'.

Thanks,
Nipun