LAUNCHXL2-570LC43: Time to toggle a GPIO pin

Durga Choudhury

Part Number: LAUNCHXL2-570LC43
Other Parts Discussed in Thread: TMS570LC4357

What is the nominal time to toggle a GPIO pin on the 570LC43 and what determines it (e.g. internal capacitance of the gates, inductance of the pins, something else). In my measurement, toggling a pin at the maximum possible rate in a while(1) loop, I am measuring a frequency ~ 1.33 MHz. At 300 MHz clock, this maps to a toggling time of > 100 processor cycles. Why does it take that long? For hard real time applications, triggering something from a GPIO line seems hard if it takes > 100 clocks to toggle it.

over 1 year ago

0 jagadish gundavarapu over 1 year ago

TI__Guru 69811 points

Hi Durga Choudhury,

Please look at below comment:

(+) GPIO toggling speed with RM48L952 CNCD - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

Thanks & regards,
Jagadish.

0 Durga Choudhury over 1 year ago in reply to jagadish gundavarapu

Intellectual 380 points

Hello Jagadish

Thanks for the clarification. I do have two follow up questions:

1. The thread you pointed out is for a different part, but I assume the same considerations apply for the part I am referring to also?

2. What is the recommended method to measure interrupt latency on these devices? Generally, firing a GPIO in the interrupt handler and measuring the time lapse between the interrupt trigger and response in an oscilloscope is the method, but given the excessive delay in firing a GPIO, it does not look like the best method for this chip.

Best regards

Durga

0 jagadish gundavarapu over 1 year ago in reply to Durga Choudhury

TI__Guru 69811 points

Hi Durga,

Durga Choudhury said:
1. The thread you pointed out is for a different part, but I assume the same considerations apply for the part I am referring to also?

Yes, same latency will be applicable for TMS570LC4357.

Durga Choudhury said:
2. What is the recommended method to measure interrupt latency on these devices? Generally, firing a GPIO in the interrupt handler and measuring the time lapse between the interrupt trigger and response in an oscilloscope is the method, but given the excessive delay in firing a GPIO, it does not look like the best method for this chip.

(+) TMS570LS3137 - GIO Interrupt latency - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

In above thread they mentioned that they measured using PMU unit, but exact procedure is not described. But the result they got was 47 cycles.

Thanks & Regards,
Jagadish.

0 Durga Choudhury over 1 year ago in reply to jagadish gundavarapu

Intellectual 380 points

OK, thanks a lot for your help, Jagadish.