Prop hub RAM fail - bad bit!
Peter Jakacki
Posts: 10,193
in Propeller 1
Never thought I'd see this but it seems I have a perfectly brand new Prop chip on a board that kept failing to accept some Tachyon code. It seems I have at least one hub memory location where there is a bad bit.
Perhaps I should run memory tests on all chips?
If I run this one-liner you can see the partial results where bit 2 ($04) never sets.
Perhaps I should run memory tests on all chips?
If I run this one-liner you can see the partial results where bit 2 ($04) never sets.
$100 0 DO CR I .BYTE SPACE I $5FDD C! $5FDD C@ .BYTE LOOP 00 00 01 01 02 02 03 03 04 00 05 01 06 02 07 03 08 08 09 09 0A 0A 0B 0B 0C 08 0D 09 0E 0A 0F 0B 10 10 11 11 12 12 13 13 14 10
Comments
That's what I was thinking since it's a brand new chip although I'm not worried about the chip itself but I do want to run some memory tests on this batch of prototypes to see if any of them have a problem.
In some cases failure analysis would have to be performed to determine the cause of the failure. Given that Parallax chips are often subjected to conditions exceeding the max handling/operating conditions (basing this on forum posts), they would need to determine when this would be appropriate.
Standard operating conditions of 3.3V with a 6MHz oscillator input x16 -> 96MHz but also tested these at 48MHz and still with the same problem. The fault is repeatable and consistent.
What is the date code?
Meanwhile Parallax can see if you uncovered a test hole - I don't believe that they use MEMBIST. If that's true and they load functional code to perform the memory testing, then perhaps they could effectively run it remotely with you.