2017-08-01 11:30:40

by Sean Young

Subject: [PATCH v4.13 REGRESSION] Revert "serial: Delete dead code for CIR serial ports"

This reverts commit 1104321a7b3bb670dc614ffa7958c553e7b3b836.

The code is not dead at all and breaks winbond-cir.

Serial: 8250/16550 driver, 32 ports, IRQ sharing enabled
00:02: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A
00:03: ttyS1 at I/O 0x2f8 (irq = 3, base_baud = 115200) is a CIR port
lirc lirc0: lirc_dev: driver ir-lirc-codec (winbond-cir) registered at minor = 0
winbond-cir 00:03: Region 0x2f8-0x2ff already in use!
winbond-cir: probe of 00:03 failed with error -16

Signed-off-by: Sean Young <[email protected]>
---
drivers/tty/serial/8250/8250_core.c | 23 +++++++++++++++++------
1 file changed, 17 insertions(+), 6 deletions(-)

diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c
index b5def356af63..1aab3010fbfa 100644
--- a/drivers/tty/serial/8250/8250_core.c
+++ b/drivers/tty/serial/8250/8250_core.c
@@ -1043,13 +1043,24 @@ int serial8250_register_8250_port(struct uart_8250_port *up)
if (up->dl_write)
uart->dl_write = up->dl_write;

- if (serial8250_isa_config != NULL)
- serial8250_isa_config(0, &uart->port,
- &uart->capabilities);
+ if (uart->port.type != PORT_8250_CIR) {
+ if (serial8250_isa_config != NULL)
+ serial8250_isa_config(0, &uart->port,
+ &uart->capabilities);
+
+ ret = uart_add_one_port(&serial8250_reg,
+ &uart->port);
+ if (ret == 0)
+ ret = uart->port.line;
+ } else {
+ dev_info(uart->port.dev,
+ "skipping CIR port at 0x%lx / 0x%llx, IRQ %d\n",
+ uart->port.iobase,
+ (unsigned long long)uart->port.mapbase,
+ uart->port.irq);

- ret = uart_add_one_port(&serial8250_reg, &uart->port);
- if (ret == 0)
- ret = uart->port.line;
+ ret = 0;
+ }
}
mutex_unlock(&serial_mutex);

--
2.13.3

2017-08-01 15:57:46

by Matthias Brugger

Subject: Re: [PATCH v4.13 REGRESSION] Revert "serial: Delete dead code for CIR serial ports"

On 08/01/2017 01:24 PM, Sean Young wrote:
> This reverts commit 1104321a7b3bb670dc614ffa7958c553e7b3b836.
>
> The code is not dead at all and breaks winbond-cir.
>
> Serial: 8250/16550 driver, 32 ports, IRQ sharing enabled
> 00:02: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A
> 00:03: ttyS1 at I/O 0x2f8 (irq = 3, base_baud = 115200) is a CIR port
> lirc lirc0: lirc_dev: driver ir-lirc-codec (winbond-cir) registered at minor = 0
> winbond-cir 00:03: Region 0x2f8-0x2ff already in use!
> winbond-cir: probe of 00:03 failed with error -16
>

You are right, uart->port.type gets overwritten, so it is definitely not dead
code. That was an oversight from me, sorry for the inconvenience.

Reviewed-by: Matthias Brugger <[email protected]>

> Signed-off-by: Sean Young <[email protected]>
> ---
> drivers/tty/serial/8250/8250_core.c | 23 +++++++++++++++++------
> 1 file changed, 17 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c
> index b5def356af63..1aab3010fbfa 100644
> --- a/drivers/tty/serial/8250/8250_core.c
> +++ b/drivers/tty/serial/8250/8250_core.c
> @@ -1043,13 +1043,24 @@ int serial8250_register_8250_port(struct uart_8250_port *up)
> if (up->dl_write)
> uart->dl_write = up->dl_write;
>
> - if (serial8250_isa_config != NULL)
> - serial8250_isa_config(0, &uart->port,
> - &uart->capabilities);
> + if (uart->port.type != PORT_8250_CIR) {
> + if (serial8250_isa_config != NULL)
> + serial8250_isa_config(0, &uart->port,
> + &uart->capabilities);
> +
> + ret = uart_add_one_port(&serial8250_reg,
> + &uart->port);
> + if (ret == 0)
> + ret = uart->port.line;
> + } else {
> + dev_info(uart->port.dev,
> + "skipping CIR port at 0x%lx / 0x%llx, IRQ %d\n",
> + uart->port.iobase,
> + (unsigned long long)uart->port.mapbase,
> + uart->port.irq);
>
> - ret = uart_add_one_port(&serial8250_reg, &uart->port);
> - if (ret == 0)
> - ret = uart->port.line;
> + ret = 0;
> + }
> }
> mutex_unlock(&serial_mutex);
>
>