??? 05/15/04 01:35 Read: times |
#70385 - RE: 51 with large SBUF? Responding to: ???'s previous message |
---> Design your UART datalink protocol to tolerate the loss of big data segments (i.e., detect when it happens and retry).
Yup, that's what it's doing now. Everything works, just has about a dozen retries for every successful transmission! In another system, I had the host wait for an acknowledge on _every_ byte. At 115200bps, it worked perfectly. I don't have much control over the design of the serial interface for this one. ---> Wow, interrupts disabled for ~26ms! No, the interrupts are usually enabled, just disabled during critical sections. With every RI there is a probability that it cannot be serviced before the next byte, so the longer packet, the worse off... I think it is amazing that with all the peripherals available on various 51 flavors, none seem to have a little FIFO on the UART. |
Topic | Author | Date |
51 with large SBUF? | 01/01/70 00:00 | |
51 without large SBUF! | 01/01/70 00:00 | |
RE: 51 without large SBUF! | 01/01/70 00:00 | |
RE: 51 without large SBUF! | 01/01/70 00:00 | |
RE: 51 with large SBUF? | 01/01/70 00:00 | |
RE: 51 with large SBUF? | 01/01/70 00:00 | |
RE: 51 with large SBUF? | 01/01/70 00:00 | |
RE: 51 with large SBUF? | 01/01/70 00:00 | |
RE: 51 with large SBUF?![]() | 01/01/70 00:00 |