• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 30 subscribers
  • Views 367 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

TMS570LS1227: HAL boot time from _c_int00 to main()

TMS570LS0914: SoC start-up time to c init

v01d
Expert 2035 points
Part Number: TMS570LS0914
Other Parts Discussed in Thread: HALCOGEN

I just got my TMS570 0914 chip, and, which checking / comparing boot times, it seems I observe approx 30 ms from power applied to the first statement(s) in the _c_ini00.   (*note*: before any setup or self tests, those all excluded from this time).

This seems surprisingly very long, and on a first check more than twice longer than 0432 TMS we have measured.

Is the TMS570 boot process described somewhere in full detail,  excluding the C init HalCoGen's code (this code is all available) , i.e,  everything before the jump to _c_int00 / or before the first resetEntry..?

Is there a ROM stage, which may be taking the time ..?

  • 0
    TI__Guru**** 197356 points

    Hi,

    After the VCCIO and VCC reach their minimum thresholds (3.0V and 1.14V), the nPORRST will keep asserted for at least 1ms. After the nPORRST is released, the internal HF LPO starts up, and validates the external oscillator, then eFUSE autoload and Flash pump and bank power-up. Please read the datasheet (6.3.1) about the cycles used for this power-up sequence.

    The number in the table doesn't include the delay from VCCIO/VCC power supply ICs. This part doesn't have ROM bootloader. 

  • 0 in reply to QJ Wang
    Expert 2035 points

    Ok, 

    So how do I calculate it?  Power-Up phases total says, from your ref section 3497 osc cycles,  and the cycles are of high-frequency oscillator. 

    Does that refer to the internal HF LPO..? Cause, cannot still make 30 ms from it even if I use 5.5Mhz as it's lower range number (typical says 9.6Mhz..).

    (And, my external crystal is 16Mhz much faster ...)

  • +1 in reply to v01d
    TI__Guru**** 197356 points

    Hi,

    The power-up sequence will not take 30ms. It only takes 3497 clock cycles (1032 LPO HF cycles, (3497-1032) external OSC cycles). HP LPO is 9.6MHz. 

    Please check the delay time from the power supply ICs.

  • 0 in reply to QJ Wang
    Expert 2035 points

    Ok, 

    I cannot see where the full 30 ms goes,  so for now will be closing as you suggesting it's our external issue.