??? 11/08/06 17:10 Read: times |
#127600 - More hints Responding to: ???'s previous message |
You also have to think about what happens at boot time. When the processor first starts running, it has to check to see whether or not an intact application is available. (There might not be, either because nobody has loaded one yet, or because a previous attempt to load one was interrupted before it finished.)
If the boot code discovers an application, it can jump to it and start running. Otherwise, it has to go into the "wait for download" mode and hope somebody is around to send an updated app. There is also the really nasty problem of redirecting interrupts. There is some information on this on the Silicon Labs forum. Try "boot loader interrupt vector" (without the quotes) in the search box there to get started. I would say this falls into the "not for the faint of heart" category. Try something simple and small to get started, and then expand on that one tiny step at a time until you get to where you need to be. -- Russ |
Topic | Author | Date |
Bootloader Guidance | 01/01/70 00:00 | |
As I understand it | 01/01/70 00:00 | |
Can I do this? | 01/01/70 00:00 | |
you are (partially) right | 01/01/70 00:00 | |
FLIP is not a possibility | 01/01/70 00:00 | |
reliability is relatively easy when done right | 01/01/70 00:00 | |
FLIP still possibility | 01/01/70 00:00 | |
Would FLIP/BatchISP manage RS485? | 01/01/70 00:00 | |
On WinCE?![]() | 01/01/70 00:00 | |
More hints | 01/01/70 00:00 |