4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mark <dmarkh@xxxxxxxxxx>
commit 47ac76662ca9c5852fd353093f19de3ae85f2e66 upstream.
Revert commit ecb988a3b7985913d1f0112f66667cdd15e40711: tty: serial:
8250: 8250_core: NXP SC16C2552 workaround
The above commit causes userland application to no longer write
correctly its first write to a dumb terminal connected to /dev/ttyS0.
This commit seems to be the culprit. It's as though the TX FIFO is being
reset during that write. What should be displayed is:
PSW 80000000 INST 00000000 HALT
//
What is displayed is some variation of:
T 00000000 HAL//
Reverting this commit via this patch fixes my problem.
Signed-off-by: Mark Hounschell <dmarkh@xxxxxxxxxx>
Fixes: ecb988a3b798 ("tty: serial: 8250: 8250_core: NXP SC16C2552 workaround")
Cc: stable <stable@xxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/tty/serial/8250/8250_port.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
--- a/drivers/tty/serial/8250/8250_port.c
+++ b/drivers/tty/serial/8250/8250_port.c
@@ -83,8 +83,7 @@ static const struct serial8250_config ua
.name = "16550A",
.fifo_size = 16,
.tx_loadsz = 16,
- .fcr = UART_FCR_ENABLE_FIFO | UART_FCR_R_TRIG_10 |
- UART_FCR_CLEAR_RCVR | UART_FCR_CLEAR_XMIT,
+ .fcr = UART_FCR_ENABLE_FIFO | UART_FCR_R_TRIG_10,
.rxtrig_bytes = {1, 4, 8, 14},
.flags = UART_CAP_FIFO,
},
