Date:   Mon, 6 Mar 2023 18:31:32 -0600
From:   Rob Herring <robh@kernel.org>
To:     Vincent Whitchurch <vincent.whitchurch@axis.com>
Cc:     Johannes Berg <johannes@sipsolutions.net>,
        Richard Weinberger <richard@nod.at>,
        Anton Ivanov <anton.ivanov@cambridgegreys.com>,
        devicetree@vger.kernel.org,
        "linux-um@lists.infradead.org" <linux-um@lists.infradead.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        kernel <kernel@axis.com>
Subject: Re: [PATCH] virt-pci: add platform bus support
Message-ID: <20230307003132.GA6366-robh@kernel.org>
References: <20230127-uml-pci-platform-v1-1-ec6b45d2829f@axis.com>
 <d3f6d627290bb1a6a1fcfdfd5fad915578453e02.camel@sipsolutions.net>
 <Y+t6qXBxLqf/+eQM@axis.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Y+t6qXBxLqf/+eQM@axis.com>
Precedence: bulk

On Tue, Feb 14, 2023 at 01:12:25PM +0100, Vincent Whitchurch wrote:
> On Mon, Feb 13, 2023 at 06:54:49PM +0100, Johannes Berg wrote:
> > On Fri, 2023-01-27 at 15:30 +0100, Vincent Whitchurch wrote:
> > > My first approach to getting platform drivers working on UML was by
> > > adding a minimal PCI-to-platform bridge driver, which worked without
> > > modifications to virt-pci, but that got shot down:
> > > 
> > >  https://lore.kernel.org/lkml/20230120-simple-mfd-pci-v1-1-c46b3d6601ef@axis.com/
> > 
> > Reading through that ... OK that isn't fun either :-)
> > 
> > Sounds like there's a use case for something else though, but the PCI
> > IDs issue also makes that thorny.
> 
> Yes, Greg was initially totally opposed to the idea of putting platform
> devices under PCI devices, but in his latest email he seemed to
> allow it in some cases.  It's still unclear if he'd be OK with a
> "virtual PCI-to-platform bridge" though.  And yes, adding platform
> devices support like in this patch removes one layer and also eliminates
> the disadvantage of having to wait for user space to specify a PCI ID
> for the bridge device.

Like I said in that thread, we have multiple usecases needing something 
similar for non-discoverable MMIO devices behind a PCI device. And I 
convinced Greg a platform device was okay, so please continue that path.

I'm adding you to the thread of other usecases.

Rob