NUC472: Workaround for unknown error with power-down #10436
Conversation
On NUC472, on wake-up from power-down mode, we may meet hard fault or some other unknown error. Before its cause is found, we enter idle mode instead for a workaround. To simulate power-down mode with idle mode, we also disable us_ticker during power-down period.
|
@ccli8, thank you for your changes. |
@0xc0170 what do you mean by "startup"? |
Sorry, wake-up I meant |
|
@0xc0170 This issue was checked before, but conclude nothing. Currently, we can only use the workaround. The side effect is power consumption. |
|
I am just after: "Where is this issue being tracked" so we do not forget about it. |
@0xc0170 It is tracked internally. |
| @@ -49,9 +56,11 @@ void hal_deepsleep(void) | |||
| } | |||
| #endif | |||
|
|
|||
| CLK_DisableModuleClock(nu_us_ticker_modinit()->clkidx); | |||
There was a problem hiding this comment.
How about free us_ticker instead here? So that you don't risk receiving unwanted interrupt?
There was a problem hiding this comment.
@Ronny-Liu The sequence of us_ticker_free() > CLK_Idle() > us_ticker_init() in hal_deepsleep() will reset us_ticker counter. Some tests like mbed-os-tests-mbed_hal-sleep and mbed-os-tests-mbed_hal-sleep_manager require us_ticker counter to continue on wake-up from power-down. Besides, hal_deepsleep() is called in interrupt-disabled context by sleep_manager_sleep_auto(), so the unwanted interrupt can be avoided.
|
ci started |
Test run: SUCCESSSummary: 11 of 11 test jobs passed |
Description
On NUC472, on wake-up from power-down mode, we may meet hard fault or some other unknown error. Before its cause is found, this PR tries to give a workaround by entering idle mode instead. To simulate power-down mode with idle mode, us_ticker clock is also disabled during power-down period.
Pull request type