| ??? 06/28/06 15:47 Read: times |
#119269 - exchange Responding to: ???'s previous message |
hi,
Jan Waclawek said:
in '51, you cannot mov one register into another.
Since this is correct I think Kai means just an idea. In case of strong assembler inside an interupt I would use XCH for this. So replacing mov r7,r6 ;save r6 mov r6,r5 ;save r5 mov r5,r4 ;save r4 mov r4,r3 ;save r3 mov r3,r2 ;save r2 mov a,p1 ;take a new sample mov r2,a ;store it in r2..with something like MOV R7,A MOV A,P1 XCH A,R2 XCH A,R3 XCH A,R4 XCH A,R5 XCH A,R6 XCH A,R7does speed up with correct way. P.S. Then I preffer to use a fixed register bank for the purpose. As a note: It is possible to use circle buffer with defined states` jumps as well (if free code space does allow it). "There are may ways to do same thing" even with assembler (= Regards, Oleg |
| Topic | Author | Date |
| Multiple switch debouncing | 01/01/70 00:00 | |
| Otherways | 01/01/70 00:00 | |
| Vertical counters | 01/01/70 00:00 | |
| vertical conters in C | 01/01/70 00:00 | |
| and here | 01/01/70 00:00 | |
| vertical counter with press detection | 01/01/70 00:00 | |
| debounce and denoise needed | 01/01/70 00:00 | |
| You are right of course, but... | 01/01/70 00:00 | |
| Reminds me... (off topic) | 01/01/70 00:00 | |
| XRL P2,A | 01/01/70 00:00 | |
| some hair-splitting | 01/01/70 00:00 | |
| exchange | 01/01/70 00:00 | |
| thanx Kai ... | 01/01/70 00:00 | |
| my way | 01/01/70 00:00 | |
| exactly! | 01/01/70 00:00 | |
| Spot on. | 01/01/70 00:00 | |
| Two samples enough? | 01/01/70 00:00 | |
| varies with design | 01/01/70 00:00 | |
| Conditional executing versus branching | 01/01/70 00:00 | |
| KEY RELEASE SUBROUTINE ... | 01/01/70 00:00 | |
| Not really... | 01/01/70 00:00 | |
| a TOTAL misconception or -very annoying | 01/01/70 00:00 | |
| He does not | 01/01/70 00:00 | |
| Beer o'clock | 01/01/70 00:00 | |
| Yes, good idea | 01/01/70 00:00 | |
| never did, never will | 01/01/70 00:00 | |
Agree | 01/01/70 00:00 |



