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

[Mark Brown]

On Mon, Jun 06, 2022 at 04:56:06PM +0530, Amit Kumar Mahapatra wrote:

> ---
>  drivers/spi/spi-zynqmp-gqspi.c | 30 ++++++++++++++++++++++++++----
>  drivers/spi/spi.c              | 10 +++++++---
>  include/linux/spi/spi.h        | 10 +++++++++-
>  3 files changed, 42 insertions(+), 8 deletions(-)

Please split the core and driver support into separate patches, they are
separate things.

> --- a/drivers/spi/spi.c
> +++ b/drivers/spi/spi.c
> @@ -2082,6 +2082,8 @@ static int of_spi_parse_dt(struct spi_controller *ctlr, struct spi_device *spi,
>  {
>  	u32 value;
>  	int rc;
> +	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.

> +	/* Bit mask of the chipselect(s) that the driver
> +	 * need to use form the chipselect array.
> +	 */
> +	u8			cs_index_mask : 2;

Why make this a bitfield?  

I'm also not seeing anything here that checks that the driver supports
multiple chip selects - it seems like something that's going to cause
issues and we should probably have something to handle that situation.

Attachment: signature.asc
Description: PGP signature