Re: [PATCH] serial: 8250: Fix reporting real baudrate value in c_ospeed field
From: Greg Kroah-Hartman
Date: Mon Sep 27 2021 - 05:55:48 EST
On Mon, Sep 27, 2021 at 11:37:04AM +0200, Pali Rohár wrote:
> In most cases it is not possible to set exact baudrate value to hardware.
> So fix reporting real baudrate value which was set to hardware via c_ospeed
> termios field. It can be retrieved by ioctl(TCGETS2) from userspace.
> Real baudrate value is calculated from chosen hardware divisor and base
> clock. It is implemented in a new function serial8250_compute_baud_rate()
> which is inverse of serial8250_get_divisor() function.
> With this change is fixed also UART timeout value (it is updated via
> uart_update_timeout() function), which is calculated from the now fixed
> baudrate value too.
I can not parse this sentence, sorry. Can you try to rephrase it
> Signed-off-by: Pali Rohár <pali@xxxxxxxxxx>
> Cc: stable@xxxxxxxxxxxxxxx
> I have tested this change on device with 8250 compatible UART. I have not
> tested it on PORT_NPCM nor on UPF_MAGIC_MULTIPLIER hardware, as I do not
> have such.
> Tested device has 250 MHz base clock for 8250 UART. When I set baudrate to
> 115200, then ioctl(TCGETS2) reported B115200 CBAUD flag and value 114890 in
> c_ospeed field.
> This result is correct as HW for baudrate 115200 is using UART divisor 136,
> which results in correct reported baudrate: 250000000/(16*136) = 114890
So is this going to break all the userspace tools that set a baud rate
and then read it back and get a different number than what they thought
That feels very dangerous. Why does this matter?
> drivers/tty/serial/8250/8250_port.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
> diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c
> index 66374704747e..dc6900b2daa8 100644
> --- a/drivers/tty/serial/8250/8250_port.c
> +++ b/drivers/tty/serial/8250/8250_port.c
> @@ -2584,6 +2584,19 @@ static unsigned int serial8250_get_divisor(struct uart_port *port,
> return serial8250_do_get_divisor(port, baud, frac);
> +static unsigned int serial8250_compute_baud_rate(struct uart_port *port,
> + unsigned int quot)
> + if ((port->flags & UPF_MAGIC_MULTIPLIER) && quot == 0x8001)
> + return port->uartclk / 4;
> + else if ((port->flags & UPF_MAGIC_MULTIPLIER) && quot == 0x8002)
> + return port->uartclk / 8;
> + else if (port->type == PORT_NPCM)
> + return DIV_ROUND_CLOSEST(port->uartclk - 2 * (quot + 2), 16 * (quot + 2));
> + else
> + return DIV_ROUND_CLOSEST(port->uartclk, 16 * quot);
Where did these formulas come from?