Re: [PATCH v3 1/6] PCI: rockchip: Create individual folder for rockchip drivers

From: Lorenzo Pieralisi <lorenzo.pieralisi-5wv7dgnIgG8@public.gmane.org>
To: Shawn Lin <shawn.lin-TNX95d0MmH7DzftRWevZcw@public.gmane.org>
Cc: devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-rockchip-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
	Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
	Bjorn Helgaas <helgaas-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
	Greg Kroah-Hartman
	<gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org>,
	Bjorn Helgaas <bhelgaas-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
Subject: Re: [PATCH v3 1/6] PCI: rockchip: Create individual folder for rockchip drivers
Date: Thu, 22 Mar 2018 11:30:00 +0000	[thread overview]
Message-ID: <20180322113000.GA24976@e107981-ln.cambridge.arm.com> (raw)
In-Reply-To: <7275bd77-016a-2729-482d-2855703d9b56-TNX95d0MmH7DzftRWevZcw@public.gmane.org>

On Thu, Mar 22, 2018 at 09:03:58AM +0800, Shawn Lin wrote:

[...]

> How about renaming drivers/pci/host to drivers/pci/controller,
> and then put all pcie-rockchip* files in there, then we don't
> need add seperate directory containning trivial Kconfig, Makefile.

Yes we can do that but not at -rc6 tail end.

> >>Certainly, I could do that beased on your suggestion, but I would like
> >>to know if this's the final agreement both you and Lorenzo reached. :)
> >
> >Or we can add all EP drivers into drivers/pci/endpoint but then the problem
> >is how to share code between host and endpoint directories and it starts
> >feeling like we are going round in circles.
> 
> I would prefer to rename it to endpoint-framwrok or whatever to clearly
> states it is for framework but not the drivers. All the drivers
> whichever for host or endpoint should be in controller directory.
> 
> >
> >I can merge this series as-is as long as we will rework the directory
> >structure according to this discussion (given that's already late -rc6
> >so directory moves/renames can be disrupting - if feasible at all).
> 
> This sounds super great to me to make this series into -next for some
> days before hitting into v4.17. Then we could simultaneously head up to
> see how we could move forward with this.

Send a last version asap and I will try to merge it for this cycle,
keeping in mind that what we discussed above must be implemented as
soon as a new cycle starts.

Thanks,
Lorenzo