HAL bug in LPTIM interrupt.

[Andres Cao]

Posted on November 27, 2017 at 14:20

Hi everyone,

I'm trying to clear my head around interrupts with an STM32l0x3. I'm not using HAL functions, rather I'm trying to understand the logic behind all this.

This becomes a little bit messier since the NVIC registers are not described in the Reference Manual, but in the Programming Manual. 

Anyway, working on this I found that for the EXTI the reference manual says:

For the direct interrupt lines: the interrupt is enabled by default in the interrupt mask register and there is no corresponding pending bit in the pending register.

The LPTIM interrupt is a direct line, hence this text applies.  LPTIM is interrupt number 29. 

PIF29 doesn't exist, and bit 23:31 should be kept at reset value as stated.

However, at the end of  ''HAL_LPTIM_IRQHandler(&hlptim1);'' we can find  ''__HAL_LPTIM_WAKEUPTIMER_EXTI_CLEAR_FLAG();''

This macro translates to ''EXTI->PR = LPTIM_EXTI_LINE_WAKEUPTIMER_EVENT''

and #define LPTIM_EXTI_LINE_WAKEUPTIMER_EVENT ((uint32_t)EXTI_IMR_IM29)

So, this means that EXTI_PR 29th bit is being set to 1. First of all, as I understand, this has no effect at all. Second, it goes against the reference manual.

Is my understanding correct?

Cheers,

Andrés

[Amel NASRI]

Posted on November 30, 2017 at 14:59

Hi

Cao.Diego_Andres

‌,

To clear up the ambiguity, I confirm that the RM0367 is correct.

I raised internally the error you highlighted regarding the definition of both macros '__HAL_LPTIM_WAKEUPTIMER_EXTI_GET_FLAG' and '__HAL_LPTIM_WAKEUPTIMER_EXTI_CLEAR_FLAG' where EXTI_PR is wrongly used.

-Amel

[Amel NASRI]

Hello,

Please note that the raised error was fixed in STM32CubeL0 V1.11.0.

-Amel