Is MWU in Softdevice?

RE: Is MWU in Softdevice?

Susheel Nuguru — Mon, 23 Mar 2020 14:28:48 GMT

if you are not using the softdevice and you are not enabling MWU anywhere in the application, then you do not need to apply workaround for Errata 75.

RE: Is MWU in Softdevice?

hmolesworth — Fri, 20 Mar 2020 22:15:47 GMT

You don't; just the other workaround. I use both though and am hoping for any feedback from other users. I hadn't anticipated having to look so closely at this.

RE: Is MWU in Softdevice?

Justin — Fri, 20 Mar 2020 22:07:08 GMT

If we are not using softdevice and not using MWU why do we need to add the workaround?

RE: Is MWU in Softdevice?

hmolesworth — Fri, 20 Mar 2020 22:03:58 GMT

For a more general case where the Softdevice is/isn't installed and user code does/doesn't use the MWU I am using the following:

// Errata 75: MWU: Increased current consumption
//  This has to be handled by turning off MWU region 0 which is used in SoftDevice
//  see https://infocenter.nordicsemi.com/index.jsp?topic=%2Ferrata_nRF52832_EngB%2FERR%2FnRF52832%2FEngineeringB%2Flatest%2Fanomaly_832_75.html
#define MWU_ACCESS_WATCH_MASK (0x0F0000FF)

The code replacing the WFE() is as follows:

  // Errata 220: CPU: RAM is not ready when written - Disable IRQ while using WFE
  // Enable SEVONPEND to disable interrupts so the internal events that generate the interrupt cause wakeuup in __WFE context and not in interrupt context
  // Before: ENABLE_WAKEUP_SOURCE -> __WFE -> WAKEUP_SOURCE_ISR -> CONTINUE_FROM_ISR  next line of __WFE
  // After:  ENABLE_WAKEUP_SOURCE -> SEVONPEND -> DISABLE_INTERRUPTS -> __WFE -> WAKEUP inside __WFE -> ENABLE_interrupts -> WAKEUP_SOURCE_ISR
  //
  // Errata 75: MWU: Increased current consumption
  // This has to be handled by turning off MWU but it is used in SoftDevice
  // see https://infocenter.nordicsemi.com/index.jsp?topic=%2Ferrata_nRF52832_EngB%2FERR%2FnRF52832%2FEngineeringB%2Flatest%2Fanomaly_832_75.html
  //
  // Errata 220: Enable SEVONPEND
  SCB->SCR |= SCB_SCR_SEVONPEND_Msk;
  __disable_irq();
  // Errata 75: MWU Disable
  uint32_t MWU_AccessWatchMask = NRF_MWU->REGIONEN & MWU_ACCESS_WATCH_MASK;
  // Handle MNU if any areas are enabled
  if (MWU_AccessWatchMask)
  {
     NRF_MWU->REGIONENCLR = MWU_AccessWatchMask; // Disable write access watch in region[0] and PREGION[0]
     __WFE();
     __NOP(); __NOP(); __NOP(); __NOP();
     // Errata 75: MWU Enable
     NRF_MWU->REGIONENSET = MWU_AccessWatchMask;
  }
  else
  {
     __WFE();
     __NOP(); __NOP(); __NOP(); __NOP();
  }
  __enable_irq();

This code is required in nrf_pwr_mgmt.c, main.c and nrf_soc.c to cover our use cases with both BLE SoftDevice and without. Feel free to provide constructive feedback .. this is very hard to test!

RE: Is MWU in Softdevice?

Susheel Nuguru — Fri, 20 Mar 2020 07:36:14 GMT

I am sorry Justin,

It is my fault that I did not think about Errata 75 when suggesting you to apply the workaround for Errata 220.

You are right, we need to apply the workaround of Errata 75 too into the workaround of Errata-220.

So the required and combined workaround for 220 and 75 becomes like below

SCB->SCR |= SCB_SCR_SEVONPEND_Msk;
__disable_irq();

/* MWU Disable */
NRF_MWU->REGIONENCLR
    = ((MWU_REGIONENCLR_RGN0WA_Clear << MWU_REGIONENCLR_RGN0WA_Pos) 
     | (MWU_REGIONENCLR_PRGN0WA_Clear << MWU_REGIONENCLR_PRGN0WA_Pos));

__WFE()
__nop();__nop();__nop();__nop();

/* MWU Enable */
NRF_MWU->REGIONENSET
    = ((MWU_REGIONENSET_RGN0WA_Set << MWU_REGIONENSET_RGN0WA_Pos) 
     | (MWU_REGIONENSET_PRGN0WA_Set << MWU_REGIONENSET_PRGN0WA_Pos));

__enable_irq();

RE: Is MWU in Softdevice?

Susheel Nuguru — Fri, 20 Mar 2020 07:00:02 GMT

Sorry for the delayed reply, I think we need to do, I am asking the experts to confirm this.

RE: Is MWU in Softdevice?

Justin — Tue, 17 Mar 2020 14:38:19 GMT

Hi Susheel,

In this ticket you asked us to use __WFE instead of sd_app_evt as a workaround for Errata 220.

Please suggest what needs to be done in this case?

RE: Is MWU in Softdevice?

Susheel Nuguru — Tue, 17 Mar 2020 13:30:20 GMT

If you use sd_app_evt_wait to sleep, then the workaround is applied inside this sleep function. But if you use __WFE instead of sd_app_evt_wait, then when you go to sleep then MWU remains enabled by the softdevice and the ERRATA issue exists keeping higher power consumption in the sleep.

Is there any specific reason you do not use sd_app_evt_wait() to go to sleep?