[PATCH] serial: 8250_omap: Implementation of Errata i2310

[PATCH] serial: 8250_omap: Implementation of Errata i2310

[Udit Kumar]

As per Errata i2310[0], Erroneous timeout can be triggered,
if this Erroneous interrupt is not cleared then it may leads
to strom of interrupts, therefore apply Errata i2310 solution.

[0] https://www.ti.com/lit/pdf/sprz536 page 23

Signed-off-by: Udit Kumar <u-kumar1@ti.com>
---
 drivers/tty/serial/8250/8250_omap.c | 25 ++++++++++++++++++++-----
 1 file changed, 20 insertions(+), 5 deletions(-)

diff --git a/drivers/tty/serial/8250/8250_omap.c b/drivers/tty/serial/8250/8250_omap.c
index 170639d12b2a..38eb639f78d3 100644
--- a/drivers/tty/serial/8250/8250_omap.c
+++ b/drivers/tty/serial/8250/8250_omap.c
@@ -115,6 +115,10 @@
 /* RX FIFO occupancy indicator */
 #define UART_OMAP_RX_LVL		0x19
 
+/* Timeout Low and High */
+#define UART_OMAP_TO_L                 0x26
+#define UART_OMAP_TO_H                 0x27
+
 /*
  * Copy of the genpd flags for the console.
  * Only used if console suspend is disabled
@@ -663,13 +667,24 @@ static irqreturn_t omap8250_irq(int irq, void *dev_id)
 
 	/*
 	 * On K3 SoCs, it is observed that RX TIMEOUT is signalled after
-	 * FIFO has been drained, in which case a dummy read of RX FIFO
-	 * is required to clear RX TIMEOUT condition.
+	 * FIFO has been drained or erroneously.
+	 * So apply solution of Errata i2310 as mentioned in
+	 * https://www.ti.com/lit/pdf/sprz536
 	 */
 	if (priv->habit & UART_RX_TIMEOUT_QUIRK &&
-	    (iir & UART_IIR_RX_TIMEOUT) == UART_IIR_RX_TIMEOUT &&
-	    serial_port_in(port, UART_OMAP_RX_LVL) == 0) {
-		serial_port_in(port, UART_RX);
+		(iir & UART_IIR_RX_TIMEOUT) == UART_IIR_RX_TIMEOUT) {
+		unsigned char efr2, timeout_h, timeout_l;
+
+		efr2 = serial_in(up, UART_OMAP_EFR2);
+		timeout_h = serial_in(up, UART_OMAP_TO_H);
+		timeout_l = serial_in(up, UART_OMAP_TO_L);
+		serial_out(up, UART_OMAP_TO_H, 0xFF);
+		serial_out(up, UART_OMAP_TO_L, 0xFF);
+		serial_out(up, UART_OMAP_EFR2, 0x1);
+		serial_in(up, UART_IIR);
+		serial_out(up, UART_OMAP_EFR2, efr2);
+		serial_out(up, UART_OMAP_TO_H, timeout_h);
+		serial_out(up, UART_OMAP_TO_L, timeout_l);
 	}
 
 	/* Stop processing interrupts on input overrun */
-- 
2.34.1

Re: [PATCH] serial: 8250_omap: Implementation of Errata i2310

[Greg KH]

On Fri, Jun 14, 2024 at 11:43:14AM +0530, Udit Kumar wrote:
> As per Errata i2310[0], Erroneous timeout can be triggered,
> if this Erroneous interrupt is not cleared then it may leads
> to strom of interrupts, therefore apply Errata i2310 solution.
> 
> [0] https://www.ti.com/lit/pdf/sprz536 page 23
> 
> Signed-off-by: Udit Kumar <u-kumar1@ti.com>
> ---
>  drivers/tty/serial/8250/8250_omap.c | 25 ++++++++++++++++++++-----
>  1 file changed, 20 insertions(+), 5 deletions(-)

Does this need to go to older/stable kernels?  If so, how far back?

thanks,

greg k-h

Re: [PATCH] serial: 8250_omap: Implementation of Errata i2310

[Vignesh Raghavendra]

On 14/06/24 12:16, Greg KH wrote:
> On Fri, Jun 14, 2024 at 11:43:14AM +0530, Udit Kumar wrote:
>> As per Errata i2310[0], Erroneous timeout can be triggered,
>> if this Erroneous interrupt is not cleared then it may leads
>> to strom of interrupts, therefore apply Errata i2310 solution.
>>

s/strom/storm

>> [0] https://www.ti.com/lit/pdf/sprz536 page 23
>>
>> Signed-off-by: Udit Kumar <u-kumar1@ti.com>
>> ---
>>  drivers/tty/serial/8250/8250_omap.c | 25 ++++++++++++++++++++-----
>>  1 file changed, 20 insertions(+), 5 deletions(-)
> 
> Does this need to go to older/stable kernels?  If so, how far back?
> 

Yes, this seems to be an updated workaround to existing errata.
Please add below Fixes tag when you respin

Fixes: b67e830d38fa ("serial: 8250: 8250_omap: Fix possible interrupt storm on K3 SoCs")


> thanks,
> 
> greg k-h

-- 
Regards
Vignesh

Re: [PATCH] serial: 8250_omap: Implementation of Errata i2310

[Ilpo Järvinen]

On Fri, 14 Jun 2024, Udit Kumar wrote:

> As per Errata i2310[0], Erroneous timeout can be triggered,
> if this Erroneous interrupt is not cleared then it may leads
> to strom of interrupts, therefore apply Errata i2310 solution.
> 
> [0] https://www.ti.com/lit/pdf/sprz536 page 23
> 
> Signed-off-by: Udit Kumar <u-kumar1@ti.com>
> ---
>  drivers/tty/serial/8250/8250_omap.c | 25 ++++++++++++++++++++-----
>  1 file changed, 20 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/tty/serial/8250/8250_omap.c b/drivers/tty/serial/8250/8250_omap.c
> index 170639d12b2a..38eb639f78d3 100644
> --- a/drivers/tty/serial/8250/8250_omap.c
> +++ b/drivers/tty/serial/8250/8250_omap.c
> @@ -115,6 +115,10 @@
>  /* RX FIFO occupancy indicator */
>  #define UART_OMAP_RX_LVL		0x19
>  
> +/* Timeout Low and High */
> +#define UART_OMAP_TO_L                 0x26
> +#define UART_OMAP_TO_H                 0x27
> +
>  /*
>   * Copy of the genpd flags for the console.
>   * Only used if console suspend is disabled
> @@ -663,13 +667,24 @@ static irqreturn_t omap8250_irq(int irq, void *dev_id)
>  
>  	/*
>  	 * On K3 SoCs, it is observed that RX TIMEOUT is signalled after
> -	 * FIFO has been drained, in which case a dummy read of RX FIFO
> -	 * is required to clear RX TIMEOUT condition.
> +	 * FIFO has been drained or erroneously.
> +	 * So apply solution of Errata i2310 as mentioned in
> +	 * https://www.ti.com/lit/pdf/sprz536
>  	 */
>  	if (priv->habit & UART_RX_TIMEOUT_QUIRK &&
> -	    (iir & UART_IIR_RX_TIMEOUT) == UART_IIR_RX_TIMEOUT &&
> -	    serial_port_in(port, UART_OMAP_RX_LVL) == 0) {
> -		serial_port_in(port, UART_RX);
> +		(iir & UART_IIR_RX_TIMEOUT) == UART_IIR_RX_TIMEOUT) {
> +		unsigned char efr2, timeout_h, timeout_l;
> +
> +		efr2 = serial_in(up, UART_OMAP_EFR2);
> +		timeout_h = serial_in(up, UART_OMAP_TO_H);
> +		timeout_l = serial_in(up, UART_OMAP_TO_L);
> +		serial_out(up, UART_OMAP_TO_H, 0xFF);
> +		serial_out(up, UART_OMAP_TO_L, 0xFF);
> +		serial_out(up, UART_OMAP_EFR2, 0x1);

Eh, this doesn't match the workaround in the errata???

Also, don't use literals but name the bits with defines (for the correct 
bit there's probably a pre-existing define but it's not named as good as 
it could be, I'd say it should be named as 
UART_OMAP_EFR2_TIMEOUT_PERIODIC).

> +		serial_in(up, UART_IIR);
> +		serial_out(up, UART_OMAP_EFR2, efr2);
> +		serial_out(up, UART_OMAP_TO_H, timeout_h);
> +		serial_out(up, UART_OMAP_TO_L, timeout_l);

-- 
 i.

Re: [PATCH] serial: 8250_omap: Implementation of Errata i2310

[Kumar, Udit]

On 6/14/2024 1:58 PM, Ilpo Järvinen wrote:
> On Fri, 14 Jun 2024, Udit Kumar wrote:
>
>> As per Errata i2310[0], Erroneous timeout can be triggered,
>> if this Erroneous interrupt is not cleared then it may leads
>> to strom of interrupts, therefore apply Errata i2310 solution.
>>
>> [0] https://www.ti.com/lit/pdf/sprz536 page 23
>>
>> Signed-off-by: Udit Kumar <u-kumar1@ti.com>
>> ---
>>   drivers/tty/serial/8250/8250_omap.c | 25 ++++++++++++++++++++-----
>>   1 file changed, 20 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/tty/serial/8250/8250_omap.c b/drivers/tty/serial/8250/8250_omap.c
>> index 170639d12b2a..38eb639f78d3 100644
>> --- a/drivers/tty/serial/8250/8250_omap.c
>> +++ b/drivers/tty/serial/8250/8250_omap.c
>> @@ -115,6 +115,10 @@
>>   /* RX FIFO occupancy indicator */
>>   #define UART_OMAP_RX_LVL		0x19
>>   
>> +/* Timeout Low and High */
>> +#define UART_OMAP_TO_L                 0x26
>> +#define UART_OMAP_TO_H                 0x27
>> +
>>   /*
>>    * Copy of the genpd flags for the console.
>>    * Only used if console suspend is disabled
>> @@ -663,13 +667,24 @@ static irqreturn_t omap8250_irq(int irq, void *dev_id)
>>   
>>   	/*
>>   	 * On K3 SoCs, it is observed that RX TIMEOUT is signalled after
>> -	 * FIFO has been drained, in which case a dummy read of RX FIFO
>> -	 * is required to clear RX TIMEOUT condition.
>> +	 * FIFO has been drained or erroneously.
>> +	 * So apply solution of Errata i2310 as mentioned in
>> +	 * https://www.ti.com/lit/pdf/sprz536
>>   	 */
>>   	if (priv->habit & UART_RX_TIMEOUT_QUIRK &&
>> -	    (iir & UART_IIR_RX_TIMEOUT) == UART_IIR_RX_TIMEOUT &&
>> -	    serial_port_in(port, UART_OMAP_RX_LVL) == 0) {
>> -		serial_port_in(port, UART_RX);
>> +		(iir & UART_IIR_RX_TIMEOUT) == UART_IIR_RX_TIMEOUT) {
>> +		unsigned char efr2, timeout_h, timeout_l;
>> +
>> +		efr2 = serial_in(up, UART_OMAP_EFR2);
>> +		timeout_h = serial_in(up, UART_OMAP_TO_H);
>> +		timeout_l = serial_in(up, UART_OMAP_TO_L);
>> +		serial_out(up, UART_OMAP_TO_H, 0xFF);
>> +		serial_out(up, UART_OMAP_TO_L, 0xFF);
>> +		serial_out(up, UART_OMAP_EFR2, 0x1);
> Eh, this doesn't match the workaround in the errata???
>
> Also, don't use literals but name the bits with defines (for the correct
> bit there's probably a pre-existing define but it's not named as good as
> it could be, I'd say it should be named as
> UART_OMAP_EFR2_TIMEOUT_PERIODIC).


Thanks, will address in v2.


>
>> +		serial_in(up, UART_IIR);
>> +		serial_out(up, UART_OMAP_EFR2, efr2);
>> +		serial_out(up, UART_OMAP_TO_H, timeout_h);
>> +		serial_out(up, UART_OMAP_TO_L, timeout_l);