mirror of
https://github.com/KolibriOS/kolibrios.git
synced 2024-12-17 20:32:35 +03:00
19 lines
1000 B
NASM
19 lines
1000 B
NASM
|
.device ATmega8
|
|||
|
.org 0
.db 1,2
|
|||
|
.message "The previous line is ignored with avra-1.2.2 because .org 0 is terminated with CR only. "
|
|||
|
.message "This is line 5 but avra-1.2.2 shows line 4"
|
|||
|
.db "X%MINUTE%YEAR%" ; Take a look at this percent chars too : % % % %
|
|||
|
.db "%YEAR%HELLO%" ; Strange replacement, if one percent char is missing
|
|||
|
.db "%HOUR%:%MINUTE%%"
|
|||
|
.db "<22>C" ; Look at the special char. (Error in listing only. HEX-file was ok)
|
|||
|
; Additional warning : Don't use linux editors with UTF charset ! A single special char
|
|||
|
; (Code > 127 in codepage 850 e.g. german umlauts) could be an unvisible TWO bytes sequence
|
|||
|
; in UTF coding. To be on the save side never use chars with code > 127.
|
|||
|
; It's better to replace them by the code e.g. .db "M",129,"nchen" (german town 'Munich')
|
|||
|
ldi R16, ';' ; This is wrong with avra-1.2.2.
|
|||
|
ldi R16, 0x3b ; Should generate same code like above
|
|||
|
|
|||
|
; TODO :
|
|||
|
; ldi R16,high (11111) ; "high(" is OK, "high (" isn't. Same with other functions...
|
|||
|
|