Bruce,
Is there a memory location for BASIC to sample holding each Uart's last-character-received (so, also, to be able clear it) WITHOUT fetching characters from the RX buffer?
I'd like to be able to ignore messages coming in until a terminating character is received.
Of course, I can fetch them into a local buffer as received, but that's just more overhead code I'd like to avoid.
something like:
If (UART1Lastchr = LF) then
gosub fetch_the_Uart1_buffer
UART1Lastchr = -1
endif
...
go onward
...
Thanks,
Lloyd
RXD last character received
-
- Posts: 99
- Joined: Mon Apr 15, 2013 3:51 pm
- Location: NE Central FL
Re: RXD last character received
No, there is no equivalent of ungetc to a serial channel.
While in theory all that is in memory, there is no standard way to do that between boards, and firmware revisions (though I doubt there will be any firmware versions any time soon).
I've often done the equivalent of gets, and it is not really that much overhead.
While in theory all that is in memory, there is no standard way to do that between boards, and firmware revisions (though I doubt there will be any firmware versions any time soon).
I've often done the equivalent of gets, and it is not really that much overhead.
-
- Posts: 99
- Joined: Mon Apr 15, 2013 3:51 pm
- Location: NE Central FL
Re: RXD last character received
Thanks,
You're right that it's not a lot of additional overhead, but it's some, and although I have tons of code space left, arrays and matrices are bumping me up against my variables limits. I guess I'll have to start storing those in flash, rather than as variables.
I have not looked at all the memory maps of all your processors. If there is any common memory across models, a pointer in common RAM would allow such to be done across all firmware revisions and models of the board (but you knew that <G>). Just a thought for future revs. No big issue.
Lloyd
You're right that it's not a lot of additional overhead, but it's some, and although I have tons of code space left, arrays and matrices are bumping me up against my variables limits. I guess I'll have to start storing those in flash, rather than as variables.
I have not looked at all the memory maps of all your processors. If there is any common memory across models, a pointer in common RAM would allow such to be done across all firmware revisions and models of the board (but you knew that <G>). Just a thought for future revs. No big issue.
Lloyd
Re: RXD last character received
On the SuperPRO there is a 16K bank of SRAM at &H2007C000 that is not used by BASIC, mainly because it is not contiguous with the SRAM at &H10000000. You can access that through pointers.
One fault I have with NXP is that there is almost no consistency of memory maps between parts.
One fault I have with NXP is that there is almost no consistency of memory maps between parts.