MSPM0G3507: UART DMA interrupt issue

Hi Brandon,

Thank you for your quick response.

As you have mentioned, I have updated the handler as you can see in the below image.

However, after adding this caused a DMA interrupt stuck in the very first UART transaction as I have shown in the below image.

Also, I do not know if it is related, but in my application, I have also configured ADC, PWM, and SysTick interrupts. Is any of these interrupts can cause problems in UART DMA ?

Kindly help me out on this.

Regards,

Virendra Dalal

Einfochips

Hi Virendra,

Please also comment out the call to: DL_UART_disableDMATransmitEvent(DEBUG_UART_INST); 

We are effectively moving this portion of the workaround to the interrupt handler so it occurs quickly without being preempted, so you should comment it out here. 

VIRENDRA DALAL said:

Also, I do not know if it is related, but in my application, I have also configured ADC, PWM, and SysTick interrupts. Is any of these interrupts can cause problems in UART DMA ?

Its possible, are you using the DMA in conjunction with any of these peripherals? That would be the first place I'd look. 

Best Regards,
Brandon Fisher

Hi Brandon,

Thank you for the quick response.

I have tried commenting out "DL_UART_disableDMATransmitEvent(DEBUG_UART_INST); ". But, the issue remains the same.

Brandon Fisher94 said:

VIRENDRA DALAL said:

Also, I do not know if it is related, but in my application, I have also configured ADC, PWM, and SysTick interrupts. Is any of these interrupts can cause problems in UART DMA ?

Its possible, are you using the DMA in conjunction with any of these peripherals? That would be the first place I'd look. 

No, I am not using DMA in conjunction with any of the peripherals.

For your reference, I am attaching screenshots for the UART configuration in syscfg file.

Regards,

Virendra Dalal

Einfochips

Hi Virendra,

I don't see anything obviously wrong here. Have you changed anything else from the example about UART like the Baud rate or clock settings?

In my setup the example is running just fine with this configuration.

Does the UART fail at a consistent place each time or does it seem to be random? 

Best Regards,
Brandon Fisher

Hi Brandon,

Brandon Fisher94 said:

Have you changed anything else from the example about UART like the Baud rate or clock settings?

No, I have exact same configurations as the UART DMA example. An example project is running fine, but my firmware application is not working.  

Brandon Fisher94 said:

Does the UART fail at a consistent place each time or does it seem to be random? 

Yes, it does fail at the same place. My firmware application basically first initializes all the peripherals and then waits for the button interrupt. I printed a welcome message on UART after initialization is done, and that works fine. But when I get the button interrupt, I start taking ADC samples and print these samples over UART. On this first sample print, it always gets stuck.

I have also tried changing the FIFO levels as shown in the below snapshot.

I have also tried the same configuration as suggested in the "MSPM0_EarlySamplesFunctionalDifferences_v1p0" document.

Is there any other UART configuration that I am missing? Or can you suggest any other way?

Thanks and Regards,

Virendra Dalal

Einfochips

Hi Brandon,

Brandon Fisher94 said:

When you start capturing ADC samples, how are you then printing the message out to UART? 

I think there may be a hint in how the ADC is triggering the UART since the failure occurs here. 

Do you call the UART_console_write function from inside your ADC done ISR? What does this structure look like?

Yes, you are right. I mistakenly printed one line in ADC ISR. I can now print all the ADC samples without any issues.

Thanks for your continuous support.

I do have one doubt regarding the PWM APIs. I have configured PWM0 for two channels as you can see in the below image:

Is there any way to stop one PWM channel? I have tried to set the first PWM channel capture compare value to 0, but does not seem to work. I need to use the "DL_TimerG_stopCounter" method, which stops the whole PWM group. 

Can you suggest me any way of doing this?

Thanks and Regards,
Virendra Dalal

Einfochips

Hi Virendra,

VIRENDRA DALAL said:

Yes, you are right. I mistakenly printed one line in ADC ISR. I can now print all the ADC samples without any issues.

Thanks for your continuous support.

Great, glad to hear the UART output is working now.

VIRENDRA DALAL said:

Is there any way to stop one PWM channel? I have tried to set the first PWM channel capture compare value to 0, but does not seem to work. I need to use the "DL_TimerG_stopCounter" method, which stops the whole PWM group. 

Are you wanting the PWM pin to go low or high when idle/disconnected?

You should be able to set the duty cycle to 0%, by setting your CCP value to the load value. 

Alternatively, and I should note that I haven't actually tried this method before, you could also disconnect the TIMx channel from the relevant pin using the IOMUX. In that mode it could be left as high z or reconnected as a GPIO and driven whichever way you want.

Best Regards,
Brandon Fisher

Hi Brandon,

I did manage to stop the individual PWM channel. I can correctly run both of the PWM channels independently. The issue was I have set the Max CCP value to 16000 which is basically a 0% duty cycle. In order to get it to work, I need to set this value to 15999. 

Thank you for your all efforts and support.

Thanks and Regards,

Virendra Dalal

Einfochips