[09/16] tty: serial: 8250_dma: Add a TX trigger workaround for AM33xx

Commit Message

Peter Hurley Sept. 11, 2014, 2:35 p.m. UTC

On 09/11/2014 08:50 AM, Sebastian Andrzej Siewior wrote:
> On 09/11/2014 02:32 PM, Peter Hurley wrote:
>> On 09/11/2014 07:42 AM, Sebastian Andrzej Siewior wrote:
>>> I also need a watchdog timer for TX since it seems that on omap3 the
>>> DMA engine suddenly forgets to continue with DMA…
>>
>> One difference I noticed between the omap driver and the 8250 driver is
>> the way modem status interrupts are handled.
>>
>> The omap driver only checks modem status for the UART_IIR_MSI interrupt type.
>> The 8250 driver checks modem status at every interrupt (other than NO_INT).
>>
>> I think the UART_MSR_DCTS bit always reflects that the CTS input has changed
>> between reads of the MSR _even if auto CTS is on_. So perhaps the hardware
>> is being stopped by uart_handle_cts_change() when auto CTS is on?
> 
> I doubt that. What I see from a timer debug is that the TX-FIFO level
> is at 0, the DMA transfer for say 1024 bytes start.
> The FIFO is filled to 64bytes and refilled so it doesn't drop below 50.
> At the time of the stall I see that the DMA engine has outstanding
> bytes which it should transfer and the TX FIFO is empty. If hardware
> flow control stops the transfer, I would expect that the DMA engine
> still fills the TX-FIFO until 64 and then waits. But it doesn't.
> Writing bytes into the FIFO leads to bytes beeing sent (and I see them
> on the other side) but the DMA transfer is still on hold. Canceling the
> DMA transfer and re-programming the remaining bytes transfers the
> remaining bytes.
> 
> The odd thing is that I only triggered it with "less file". It doesn't
> happen on regular console interaction or "cat large-file". And it only
> triggers on beagle board xm (omap34xx) and I wasn't able to reproduce
> it on am335x or dra7. The latter shares the same DMA engine as beagle
> board xm.
> 
> I remember also that I disabled the HW/SW float control just to make
> sure it is not it.

Ok.

I do need to find out if omap hardware sets UART_MSR_DCTS when auto CTS
is on. Would you mind running the debug patch below with HW flow control on?

Regards,
Peter Hurley

--- >% ---
Subject: [DEBUG] serial: does OMAP set UART_LSR_DCTS while autoCTS is on?

** debug patch only **

Signed-off-by: Peter Hurley <[e-mail]>
---
 drivers/tty/serial/serial_core.c | 3 +++
 1 file changed, 3 insertions(+)

Comments

Sebastian Andrzej Siewior Sept. 15, 2014, 5:01 p.m. UTC | #1

On 09/11/2014 04:35 PM, Peter Hurley wrote:
> I do need to find out if omap hardware sets UART_MSR_DCTS when auto CTS
> is on. Would you mind running the debug patch below with HW flow control on?

I didn't forget about this. However I told minicom to use hardware flow
control (and I see the driver set the HW bit) but I haven't seen that
uart_handle_cts_change() has been invoked at all. I'm going to check
two other boards and report then.

> Regards,
> Peter Hurley

Sebastian

Sebastian Andrzej Siewior Sept. 16, 2014, 4:55 p.m. UTC | #2

On 09/15/2014 07:01 PM, Sebastian Andrzej Siewior wrote:
> On 09/11/2014 04:35 PM, Peter Hurley wrote:
>> I do need to find out if omap hardware sets UART_MSR_DCTS when auto CTS
>> is on. Would you mind running the debug patch below with HW flow control on?
> 
> I didn't forget about this. However I told minicom to use hardware flow
> control (and I see the driver set the HW bit) but I haven't seen that
> uart_handle_cts_change() has been invoked at all. I'm going to check
> two other boards and report then.

No, I don't get into this at all function. So I connected my am335x-evm
with beagle board xm because both of them have an old fashion UART
connector (instead those uart-to-usb). Both configured with HW-Flow and
I haven't seen the function invoked but I saw "port->icount.overrun"
being incremented. This shouldn't happen. So I am a little puzzled here…

>> Regards,
>> Peter Hurley
> 

Sebastian

Peter Hurley Sept. 17, 2014, 12:20 p.m. UTC | #3

On 09/16/2014 12:55 PM, Sebastian Andrzej Siewior wrote:
> On 09/15/2014 07:01 PM, Sebastian Andrzej Siewior wrote:
>> On 09/11/2014 04:35 PM, Peter Hurley wrote:
>>> I do need to find out if omap hardware sets UART_MSR_DCTS when auto CTS
>>> is on. Would you mind running the debug patch below with HW flow control on?
>>
>> I didn't forget about this. However I told minicom to use hardware flow
>> control (and I see the driver set the HW bit) but I haven't seen that
>> uart_handle_cts_change() has been invoked at all. I'm going to check
>> two other boards and report then.
> 
> No, I don't get into this at all function.

Ok, good to know. Thanks for testing that.

> So I connected my am335x-evm
> with beagle board xm because both of them have an old fashion UART
> connector (instead those uart-to-usb). Both configured with HW-Flow and
> I haven't seen the function invoked but I saw "port->icount.overrun"
> being incremented. This shouldn't happen. So I am a little puzzled here…

Yeah, that's weird. Do you have a break-out box to confirm that RTS/CTS are
being driven?

Regards,
Peter Hurley

Sebastian Andrzej Siewior Sept. 17, 2014, 4:25 p.m. UTC | #4

On 09/17/2014 02:20 PM, Peter Hurley wrote:
>> So I connected my am335x-evm
>> with beagle board xm because both of them have an old fashion UART
>> connector (instead those uart-to-usb). Both configured with HW-Flow and
>> I haven't seen the function invoked but I saw "port->icount.overrun"
>> being incremented. This shouldn't happen. So I am a little puzzled here…
> 
> Yeah, that's weird. Do you have a break-out box to confirm that RTS/CTS are
> being driven?

Yeah, I've been thinking about that, too. I will be gone next week and
hopefully when I get back I will something around to test this.

> 
> Regards,
> Peter Hurley
> 

Sebastian

Sebastian Andrzej Siewior Sept. 29, 2014, 4:15 p.m. UTC | #5

* Peter Hurley | 2014-09-17 08:20:41 [-0400]:

>> So I connected my am335x-evm
>> with beagle board xm because both of them have an old fashion UART
>> connector (instead those uart-to-usb). Both configured with HW-Flow and
>> I haven't seen the function invoked but I saw "port->icount.overrun"
>> being incremented. This shouldn't happen. So I am a little puzzled here…
>
>Yeah, that's weird. Do you have a break-out box to confirm that RTS/CTS are
>being driven?

=>
- beagle board
  According to schematics the board has only RX and TX connected. No
  RTS/CTS

- am335x-evm
  The schematics say "DNI" next to a resistor on RTS/CTS. DNI stands
  most likely for "Do Not Install". With a scope it looks like the the
  wire behind the MAX is open.

- beagle bone black
  Only RX and TX are wired towards the USB2serial device.

In short: each device I have has RTS/CTS not working/connected and I
can't test HW handshake with HW available.

>Regards,
>Peter Hurley

Sebastian

Patch

diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c
index 5a78f69..1579a20 100644
--- a/drivers/tty/serial/serial_core.c
+++ b/drivers/tty/serial/serial_core.c
@@ -2783,6 +2783,9 @@  void uart_handle_cts_change(struct uart_port *uport, unsigned int status)
 	uport->icount.cts++;
 
 	if (tty_port_cts_enabled(port)) {
+
+		WARN_ON_ONCE(uport->flags & UPF_HARD_FLOW);
+
 		if (tty->hw_stopped) {
 			if (status) {
 				tty->hw_stopped = 0;