On Mon, Apr 26, 2021 at 05:38:28PM +0800, Moriis Ku wrote:
> From: Morris Ku <saumah@xxxxxxxxx>
>
> Sunix SDC Multi I/O board are different to Timedia's.
>
> This patch adds proper support for Sunix SDC Multi I/O board.
>
> Signed-off-by: Morris Ku<saumah@xxxxxxxxx>
Can you make your subject line look like:
[PATCH] serial: 8250_pci: add support for...
to match up the other patches for this driver?
> ---
> tty/serial/8250/8250_pci.c | 36 +++++++++++++++++++++++++++++++++++-
> 1 file changed, 35 insertions(+), 1 deletion(-)
>
> diff --git a/tty/serial/8250/8250_pci.c b/tty/serial/8250/8250_pci.c
> index 8814ff38..5758a98b 100644
> --- a/tty/serial/8250/8250_pci.c
> +++ b/tty/serial/8250/8250_pci.c
> @@ -1802,6 +1802,25 @@ pci_sunix_setup(struct serial_private *priv,
> return setup_port(priv, port, bar, offset, 0);
> }
>
> +static int
> +pci_sunix_sdc_setup(struct serial_private *priv,
Move this to be all on one line please.
> + const struct pciserial_board *board,
> + struct uart_8250_port *port, int idx)
Line up with the "(" or checkpatch will complain.
> +{
> + int bar;
> + int offset;
> +
> + port->port.flags |= UPF_FIXED_TYPE;
> + port->port.type = PORT_SUNIX;
> +
> + if (idx < 4) {
> + bar = 1
> + offset = idx * board->uart_offset;
> + }
So if idx is > 4, bar is unknown? That feels like a huge bug :(
> +
> + return setup_port(priv, port, bar, offset, 0);
Same for offset, looks like it could be used uninitialized as well.
The compiler didn't complain about this?
thanks,
greg k-h
