1 Practical tests to be performed to ensure no regression prior to each release :
4 1. Ensure sync is lost with an extra pulse, and one too few
7 1. Increase and decrease of pulse width to extremes both below max pulse width threshold and above.
8 2. Increase and decrease of timing both when it does and doesn't affect overlap of schedule with or when pulse width is longer than cycle.
9 3. Increase and decrease of RPM such that pulse widths do and do not overlap, and both.
10 4. All above tests with and without staged injection occurring, ensure staged does not stick on.
13 1. Monitor other functions during above tests. They should continue to operate smoothly despite interrupt load.
14 Such functions include :
21 grep -rn "~0x[0-9][0-9];" src
22 This should be done after make and before clean to ensure that no macros etc expand into notted 8 bit literals.
24 grep -h "#define" src | sort > ~/defines.sorted
25 This will allow you to easily see the duplications. There is probably an automated way to test for duplicate lines, but this works
28 Drop each type of packet through and ensure the correct response. There should be a packet type for each and every code path created.