Dec 16, 2019
01:32 AM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dec 16, 2019
01:32 AM
The SafeTLib (535) has been configured and built for proprietary TC275 based hardware.
There are 56 tests in total, distributed over all three cores. The only test not run is Lockstep on core 1, as it's not currently enabled (Not my decision, don't ask).
All tests are passing, apart from the Input Output Monitor (IOM) Test, which reports a 0x221502 (IOMTST_ALARMNOTSETERR).
I have:
I can't get this one test to pass, does anyone have any idea what I'm doing wrong?
There are 56 tests in total, distributed over all three cores. The only test not run is Lockstep on core 1, as it's not currently enabled (Not my decision, don't ask).
All tests are passing, apart from the Input Output Monitor (IOM) Test, which reports a 0x221502 (IOMTST_ALARMNOTSETERR).
I have:
- Read the SafeTLib User Manual IOM test section at least five times.
- Ensured that the AoU are met (22K pull-down).
- Debugged the test and checked that the pin read back state toggles.
- Removed the 22K pull down.
- Tried the test on all three cores.
- Tried different pins.
- Tried different LAMs.
- Checked the Release Notes.
- Checked the Addendum.
- Compared with existing projects.
I can't get this one test to pass, does anyone have any idea what I'm doing wrong?
2 Replies
Dec 16, 2019
08:25 AM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dec 16, 2019
08:25 AM
A 22K pull-down conflicts with this part of the Release Notes Addendum:
With a 22K pull-down vs. the weak internal pull devices, it might have taken a few more clock cycles to reach a high state. It might look okay in the debugger, but at full speed, it might not work.
Have you tried it on a pin that's simply open? Which pin and LAM are you using?
GPIO pin used for IomTst shall not be connected to external circuitry which forces the pin to a permanent high/low state during the test
With a 22K pull-down vs. the weak internal pull devices, it might have taken a few more clock cycles to reach a high state. It might look okay in the debugger, but at full speed, it might not work.
Have you tried it on a pin that's simply open? Which pin and LAM are you using?
Dec 16, 2019
08:41 AM
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dec 16, 2019
08:41 AM
I asked a hardware engineer to physically remove the resistor from the PCB and now the test is passing.
Thanks.
Thanks.
This widget could not be displayed.