2012-10-03 12:28:14

by Alexey Brodkin

Subject: [PATCH v3] serial/8250/8250_early: Prevent rounding error in uartclk

Modify divisor to select the nearest baud rate divider rather than the
lowest. It minimizes baud rate errors especially on low UART clock
frequencies.

For example, if uartclk is 33000000 and baud is 115200 the ratio is
about 17.9 The current code selects 17 (5% error) but should select 18
(0.5% error).

This 5% error in baud rate leads to garbage on receiving end, while 0.5%
doesn't.

The issue showed up when using the stock 8250 driver for
Synopsys DW UART. This was on a FPGA with ~12MHz UART clock.
When we enabled early serial, we saw garbage which was narrowed down
to the rounding error.

So the bug had been latent and it only showed up with such low cock rates.

Signed-off-by: Alexey Brodkin <[email protected]>
---
drivers/tty/serial/8250/8250_early.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/tty/serial/8250/8250_early.c b/drivers/tty/serial/8250/8250_early.c
index eaafb98..843a150 100644
--- a/drivers/tty/serial/8250/8250_early.c
+++ b/drivers/tty/serial/8250/8250_early.c
@@ -140,7 +140,7 @@ static void __init init_port(struct early_serial8250_device *device)
serial_out(port, UART_FCR, 0); /* no fifo */
serial_out(port, UART_MCR, 0x3); /* DTR + RTS */

- divisor = port->uartclk / (16 * device->baud);
+ divisor = DIV_ROUND_CLOSEST(port->uartclk, 16 * device->baud);
c = serial_in(port, UART_LCR);
serial_out(port, UART_LCR, c | UART_LCR_DLAB);
serial_out(port, UART_DLL, divisor & 0xff);
--
1.7.9.5

2012-10-03 12:36:09

by Alan

Subject: Re: [PATCH v3] serial/8250/8250_early: Prevent rounding error in uartclk

> So the bug had been latent and it only showed up with such low cock
> rates.

clock ?

Perhaps Greg can tweak the description as it goes in - otherwise all
fine with me

2012-10-03 12:53:21

by Alexey Brodkin

Subject: Re: [PATCH v3] serial/8250/8250_early: Prevent rounding error in uartclk

On 03.10.2012 16:37, Alan Cox wrote:
>> So the bug had been latent and it only showed up with such low cock
>> rates.
>
> clock ?
>
> Perhaps Greg can tweak the description as it goes in - otherwise all
> fine with me
>
Sorry for this.
Essentially I meant "clock".
May do another respin if needed.

2012-10-05 16:21:07

by Greg Kroah-Hartman

Subject: Re: [PATCH v3] serial/8250/8250_early: Prevent rounding error in uartclk

On Wed, Oct 03, 2012 at 04:53:06PM +0400, Alexey Brodkin wrote:
> On 03.10.2012 16:37, Alan Cox wrote:
> >>So the bug had been latent and it only showed up with such low cock
> >>rates.
> >
> >clock ?
> >
> >Perhaps Greg can tweak the description as it goes in - otherwise all
> >fine with me
> >
> Sorry for this.
> Essentially I meant "clock".
> May do another respin if needed.

Don't worry, I can edit it :)

greg k-h