Mantis Bug Tracker

View Issue Details Jump to Notes ] Print ]
IDProjectCategoryView StatusDate SubmittedLast Update
0000186FirmwareHardware Interfacepublic2011-06-18 21:042011-07-29 16:18
Assigned ToFred 
PlatformOSOS Version
Product Version 
Target Version0.4.0Fixed in Version0.2.0 
Summary0000186: Determine why PORTA pins go high during reset
DescriptionThe rest of the information on this topic exists in these threads:

Research: [^]
Comments: [^]
Steps To ReproduceAbsolutely anyone with a clue and a cpu (and a multimeter and a computer) can research this. It is not an effective use of my time to do this when there are key software tasks that require my attention. Furthermore I've put a bunch of time into this already. Lastly, I currently have no electrical equipment for diagnosing this situation.
Additional InformationUntil we know if these pins can be controlled in a predictable way, we can no reasonably assign them to the functions which we need them for. Thus future bit bang work for 0.4.0 depends on this issue and will be added as such.
TagsNo tags attached.
Issue TypeTask
Risk of Breakagevery low
Attached Files

- Relationships
child of 0000190assignedFred Select Permanent Pins For Core Functions 

-  Notes
User avatar (0000210)
chr1s (reporter)
2011-07-08 22:53

port A bit 6 is connected to the run/load switch on the TA demo board. The serial monitor code enables portA pullup on the PUCR to test state of this pin.

taken from serial monitor source:
;SwPort: equ PORTA ;pushbutton sw connected to port A
;Switch: equ PTA6 ;switch connected to PTA bit-6
;SwPullup: equ PUCR ;pullup enable for sw port
;mSwPullup: equ PUPAE ;Mask value to enable for sw port

bset SwPullup,mSwPullup ;enable pullup on force monitor sw

This could explain some of the behavior seen
User avatar (0000211)
sean94z (reporter)
2011-07-08 23:03

Anyone have a link handy to the full SM source?

Anyone want to give me a clue as to what to do? I'm thinking write a few lines to bang everything that's a potential problem to LOW.

Now where's my BDM!!!!

-thx sean
User avatar (0000212)
chr1s (reporter)
2011-07-08 23:03
edited on: 2011-07-08 23:06

serial monitor source code from TA [^]

User avatar (0000213)
sean94z (reporter)
2011-07-08 23:09

thx chris!

I really should get that BDM from abe. There is a chance I may brick my TAcard in doing something like this.
User avatar (0000214)
Fred (administrator)
2011-07-09 00:21

Sean, not possible to brick it. Either you have a BDM and can undo your doings, or you don't and can't screw with it in the first place.

If the TA card has a physical pullup on that pin, then we should rework it to not enable the pull up and solve a lot of headaches. Additionally, even if it doesn't (though I think that it does) users could add one easily.

Also, if reworking it, we can remove the check on the serial pin state and save all projects some hassle.

Such mods would need to be VERY carefully performed. It would be ideal if we could get to the point of generating an identical s19 from GCC asm/C sources in a new project, tag it as "TA" and then make some changes from there in a controlled way. If we can't do that, then we need to verify all of our manual binary changes very very very carefully, both by code inspection and by real world tests.
User avatar (0000215)
Fred (administrator)
2011-07-09 00:31 [^]

The original source from TA via their license agreement, as posted by Chris.
User avatar (0000216)
sean94z (reporter)
2011-07-09 00:42
edited on: 2011-07-09 16:00

That's what I mean when I said I better get the BDM. Right I was going to compile their src and check to make sure all is well then proceed from there. Seems many hw issues* can be solved by a few SM changes.

User avatar (0000229)
Fred (administrator)
2011-07-29 16:18

Verified! [^]

Copyright © 2000 - 2011 MantisBT Group
Powered by Mantis Bugtracker