Re: [PATCH V2] serial: exar: Preserve FCTR[5] bit in pci_xr17v35x_setup()

["Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>]

[-- Attachment #1: Type: text/plain, Size: 2037 bytes --]

On Mon, 8 Apr 2024, Ilpo Järvinen wrote:

> On Mon, 8 Apr 2024, Matthew Howell wrote:
> 
> > On Wed, 2024-02-21 at 16:16 -0500, Matthew Howell wrote:
> > > Allows the use of the EN485 hardware pin by preserving the value of
> > > FCTR[5] in pci_xr17v35x_setup().
> > > 
> > > Per the XR17V35X datasheet, the EN485 hardware pin works by setting
> > > FCTR[5] when the pin is active. pci_xr17v35x_setup() prevented the use
> > > of EN485 because it overwrote the FCTR register.
> > > 
> > > Signed-off-by: Matthew Howell <matthew.howell@sealevel.com>
> > > ---
> > > V1 -> V2
> > > Fixed wordwrap in diff
> > > 
> > > diff --git a/drivers/tty/serial/8250/8250_exar.c b/drivers/tty/serial/8250/8250_exar.c
> > > index 23366f868..97711606f 100644
> > > --- a/drivers/tty/serial/8250/8250_exar.c
> > > +++ b/drivers/tty/serial/8250/8250_exar.c
> > > @@ -596,6 +596,7 @@ pci_xr17v35x_setup(struct exar8250 *priv, struct pci_dev *pcidev,
> > >  	unsigned int baud = 7812500;
> > >  	u8 __iomem *p;
> > >  	int ret;
> > > +	u8 en485mask;
> > >  
> > >  	port->port.uartclk = baud * 16;
> > >  	port->port.rs485_config = platform->rs485_config;
> > > @@ -618,7 +619,8 @@ pci_xr17v35x_setup(struct exar8250 *priv, struct pci_dev *pcidev,
> > >  	p = port->port.membase;
> > >  
> > >  	writeb(0x00, p + UART_EXAR_8XMODE);
> > > -	writeb(UART_FCTR_EXAR_TRGD, p + UART_EXAR_FCTR);
> > > +	en485mask = readb(p + UART_EXAR_FCTR) & UART_FCTR_EXAR_485;
> > > +	writeb(UART_FCTR_EXAR_TRGD | en485mask, p + UART_EXAR_FCTR);
> > >  	writeb(128, p + UART_EXAR_TXTRG);
> > >  	writeb(128, p + UART_EXAR_RXTRG);
> 
> Why you need to read rs485 state from the register? It should be available 
> in ->rs485.flags & SER_RS485_ENABLED.
> 
> pci_fastcom335_setup() seems to have the same problem? Path small part 

I meant "That small part ..."

> seems to be common code anyway which should be moved into helper, only the 
> trigger threshold seems to differ which can be given in a parameter.
> 
> 

-- 
 i.