| ??? 10/05/07 12:43 Read: times |
#145458 - you MUST be wrong Responding to: ???'s previous message |
So the next IT0 is allowed to interrupt the current one.
Whatever means you use, if Atmel has not (as usual) screwed up, this will not be true. An interrupt interrupting itself would make total havoc in countless designs. I would not touch an Atmel with a ten foot pole, but maybe someone that does not have this adversity could confirm "you MUST be wrong" The external system is very fast so it happens that new edge is observed while the previous is still being serviced the cure for this is KISS (Keep ISRs Short and Simple). I GUESS what you are seeing is the fact that (if your ISR is coded correctly), while an interrupt is being serviced the request for the next will still be detected and immediately upon return from the ISR it will reenter, Erik |
| Topic | Author | Date |
| I tested changing prio level from the IT, it works | 01/01/70 00:00 | |
| you MUST be wrong | 01/01/70 00:00 | |
| For Erik | 01/01/70 00:00 | |
| redefine test | 01/01/70 00:00 | |
| what??? and comments | 01/01/70 00:00 | |
| what | 01/01/70 00:00 | |
| IT in derivatives | 01/01/70 00:00 | |
| Stored edges | 01/01/70 00:00 | |
| I'm still certain | 01/01/70 00:00 | |
| I don't see well what you mean. | 01/01/70 00:00 | |
| how about an answer | 01/01/70 00:00 | |
| you are right | 01/01/70 00:00 | |
just read \"the bible\" | 01/01/70 00:00 |