EXTI Interrupt Issue on STM32

[liudeqiang_2008]

Hello,

I will first describe the conclusion:
EXTI interrupts can simulate running on STM32F411CE, but cannot simulate running on STM32F303CB.

What makes me strange is that on F411CE, after writing the program, you need to manually switch the "Input Wear Bias" in "Interrupt Properties" to any other item (Pull up or Pull down), debug Run, and then can simulate interrupts normally. Switching back to the default 'No bias' can also run normally.

But, on F303CB, even the above methods cannot make the interrupt simulation run normally.

In order to describe this issue more clearly, I have recorded a video of the operation from the beginning to the end.
Video link: https://www.youtube.com/watch?v=tN7XQ6M2NtU

I have also uploaded the programs for F411CE and F303CB to the attachment.


Please be patient and take a look, and hope it can be fixed it asap.
PS. I can not try all the stm32 chips.

Thanks

Br, Deqiang

[medelec35]

Hello.
Thank you for letting us know,
I will see if I can sort the external interrupts simulation out.

[medelec35]

Hello.
All the EXI interrupts should simulate.
I have fixed a bug that causes an error when you compile using EXI2.
The cause is Touch Sensing Controller (TSC) shares the interrupt vector with EXTI 2 so the register name is different.

If you browse to this location using File Explorer, it's hidden by default so just paste the link into your address bar.

Code: Select all

%ProgramData%\MatrixTSL\FlowcodeV10\FCD\PIC\

Replace the original file with the attached version.
If Flowcode is already open, then you will need to reload your project.
If you run any component updates, then the fcdx you added will be overwritten.

If you are happy with the changes, I will get the fcdx file push on the update system.

[liudeqiang_2008]

Hello.
All the EXI interrupts should simulate.
I have fixed a bug that causes an error when you compile using EXI2.
The cause is Touch Sensing Controller (TSC) shares the interrupt vector with EXTI 2 so the register name is different.

If you browse to this location using File Explorer, it's hidden by default so just paste the link into your address bar.

Code: Select all

%ProgramData%\MatrixTSL\FlowcodeV10\FCD\PIC\

Replace the original file with the attached version.
If Flowcode is already open, then you will need to reload your project.
If you run any component updates, then the fcdx you added will be overwritten.

If you are happy with the changes, I will get the fcdx file push on the update system.

Hello Martin,

Thank you for your excellent support, the simulation on F303CB working well after replacing the fcdx file.
I am not in the lab today so I haven't tested downloading to the chip. Tomorrow I will test it and feedback to you.

Thanks and have a nice day!

Br, Deqiang

[liudeqiang_2008]

One more question:
How to set the priority of interrupts？

[liudeqiang_2008]

Hello.
All the EXI interrupts should simulate.
I have fixed a bug that causes an error when you compile using EXI2.
The cause is Touch Sensing Controller (TSC) shares the interrupt vector with EXTI 2 so the register name is different.

If you browse to this location using File Explorer, it's hidden by default so just paste the link into your address bar.

Code: Select all

%ProgramData%\MatrixTSL\FlowcodeV10\FCD\PIC\

Replace the original file with the attached version.
If Flowcode is already open, then you will need to reload your project.
If you run any component updates, then the fcdx you added will be overwritten.

If you are happy with the changes, I will get the fcdx file push on the update system.

Hello Martin,

Keep you updated:
I have tested downloading a simple interrupt FC program that simulates OK to a STM32F303CB board, looks working good on physical board !
Think a fcdx file push is needed for all related stm32 chips.

Thank you.

Br, Deqiang

[medelec35]

Hi Deqiang.
A push has been arranged for all affected devices, not just STM32F303CB.
Thanks for keeping me updated.

[liudeqiang_2008]

Hi Deqiang.
A push has been arranged for all affected devices, not just STM32F303CB.
Thanks for keeping me updated.

Hello ,

Could you please look the TIMER Interrupt function ?
Board: STM32F303CBT6
I have set the MCU configure to HSE, so the system clock should be 72MHz, the APB1 timer clocks should also be 72MHz (After measuring looks like TIMER interrupt speed is 36MHz )
The strange thing I met, the timer trigger output is not stable, after downloading with auto restart the output looks ok. But if I restart the power, the output frequency getting slowly, from 1Hz to 0.2Hz. I test 2 different boards with same MCU, same issue, I can not find the reason.
If I keep switch on and off, will have few chance the output goes back to 1Hz, but most time not.

If I add more than 1 timmer interrupts, let they working together, they will be fully out of control.

I have put the test program and 1Hz timer interrupt output measuring result in attachment.

If I was wrong ,please let me know.

Br, Deqiang

[LeighM]

Could you try a test running on HSI?
(To see if it's a problem with the crystal)

Might also be worth changing the main loop to give another (LED) output toggle 500ms off and on, to check what the system clock is doing.

[liudeqiang_2008]

Could you try a test running on HSI?
(To see if it's a problem with the crystal)

Might also be worth changing the main loop to give another (LED) output toggle 500ms off and on, to check what the system clock is doing.

Hello,

Thank you for your suggestion.
I agree with you, and I have tested, the HSI is working. I want to know why HSE not working.
Because, I use stm32Cubemx & C programming, the HSE is working well. So the crystal should be no problem.

Really make me confused.

Br, Deqiang