[RFC PATCH 1/2] spi: Add multiple CS support for a single SPI device

Mahapatra, Amit Kumar amit.kumar-mahapatra at amd.com
Fri Jul 15 08:36:23 PDT 2022


Hello Mark,

> -----Original Message-----
> From: Mark Brown <broonie at kernel.org>
> Sent: Monday, July 11, 2022 8:23 PM
> To: Michal Simek <michal.simek at xilinx.com>
> Cc: Amit Kumar Mahapatra <amit.kumar-mahapatra at xilinx.com>; Krzysztof
> Kozlowski <krzysztof.kozlowski+dt at linaro.org>; p.yadav at ti.com;
> miquel.raynal at bootlin.com; richard at nod.at; vigneshr at ti.com;
> git at xilinx.com; linux-spi at vger.kernel.org; linux-arm-
> kernel at lists.infradead.org; linux-kernel at vger.kernel.org; michael at walle.cc;
> linux-mtd at lists.infradead.org
> Subject: Re: [RFC PATCH 1/2] spi: Add multiple CS support for a single SPI
> device
> 
> On Mon, Jul 11, 2022 at 02:47:54PM +0200, Michal Simek wrote:
> > On 6/9/22 13:54, Mark Brown wrote:
> > > On Mon, Jun 06, 2022 at 04:56:06PM +0530, Amit Kumar Mahapatra
> wrote:
> 
> > > > +	u32 cs[SPI_CS_CNT_MAX];
> > > > +	u8 idx;
> > > >   	/* Mode (clock phase/polarity/etc.) */
> > > >   	if (of_property_read_bool(nc, "spi-cpha"))
> 
> > > This is changing the DT binding but doesn't have any updates to the
> > > binding document.  The binding code also doesn't validate that we
> > > don't have too many chip selects.
> 
> > I would like to better understand your request here in connection to
> > change in the binding code for validation.
> > What exactly do you want to validate?
> > That child reg property is not bigger than num-cs in controller node?
> 
> If you are adding support for multiple chip selects in the driver then there
> must be some mechanism for expressing that in the bindings which I would
> expect to see appear as a change to the binding document.

Thank you for the clarification.
As per my understanding we would not be needing a new DT 
property for expressing multi chip select support in the driver.
We can use "num-cs" property of the spi controller for 
broadcasting multi chip select support of the driver.
If "num-cs" value is greater than 1 then the driver can support 
multiple chip selects.

Please let us know if you have any other approach in mind.

Regards,
Amit



More information about the linux-mtd mailing list