[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <blqs5xugab5kumljzrt26dvyhnun2pmpeu4y727xoz3vrponrt@bbxqnl55zhky>
Date: Thu, 20 Nov 2025 22:24:24 +0530
From: Manivannan Sadhasivam <mani@...nel.org>
To: Rob Herring <robh@...nel.org>
Cc: manivannan.sadhasivam@....qualcomm.com,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>, Jiri Slaby <jirislaby@...nel.org>,
Nathan Chancellor <nathan@...nel.org>, Nicolas Schier <nicolas.schier@...ux.dev>,
Hans de Goede <hansg@...nel.org>, Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>,
Mark Pearson <mpearson-lenovo@...ebb.ca>, "Derek J. Clark" <derekjohn.clark@...il.com>,
Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley <conor+dt@...nel.org>,
Marcel Holtmann <marcel@...tmann.org>, Luiz Augusto von Dentz <luiz.dentz@...il.com>,
Bartosz Golaszewski <brgl@...ev.pl>, linux-serial@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-kbuild@...r.kernel.org, platform-driver-x86@...r.kernel.org, linux-pci@...r.kernel.org,
devicetree@...r.kernel.org, linux-arm-msm@...r.kernel.org, linux-bluetooth@...r.kernel.org,
linux-pm@...r.kernel.org, Stephan Gerhold <stephan.gerhold@...aro.org>,
Dmitry Baryshkov <dmitry.baryshkov@....qualcomm.com>
Subject: Re: [PATCH 6/9] serdev: Skip registering serdev devices from DT is
external connector is used
On Thu, Nov 20, 2025 at 10:22:16AM -0600, Rob Herring wrote:
> On Wed, Nov 19, 2025 at 7:33 AM Manivannan Sadhasivam <mani@...nel.org> wrote:
> >
> > On Tue, Nov 18, 2025 at 07:03:51AM -0600, Rob Herring wrote:
> > > On Wed, Nov 12, 2025 at 8:45 AM Manivannan Sadhasivam via B4 Relay
> > > <devnull+manivannan.sadhasivam.oss.qualcomm.com@...nel.org> wrote:
> > > >
> > > > From: Manivannan Sadhasivam <manivannan.sadhasivam@....qualcomm.com>
> > > >
> > > > If an external connector like M.2 is connected to the serdev controller
> > > > in DT, then the serdev devices will be created dynamically by the connector
> > > > driver. So skip registering devices from DT node as there will be no
> > > > statically defined devices.
> > >
> > > You could still have statically defined devices. You are just choosing
> > > to probe them later from the connector driver.
> > >
> >
> > The point of coming up with the M.2 binding is to avoid hardcoding the devices
> > in DT. So static devices are ruled out IMO.
>
> Until you have any one of the reasons we have PCIe devices described
> even when in a standard slot. Take your pick. An ethernet adapter that
> omits an EEPROM for the MAC address.
>
Hmm, I didn't envision the need to have static and dynamic devices at the same
time. So you are suggesting to first look for the static devices and then look
for graph port at the end?
- Mani
--
மணிவண்ணன் சதாசிவம்
Powered by blists - more mailing lists