rgbds/man/rgbasm.5

'\" e
.\"
.\" SPDX-License-Identifier: MIT
.\"
.Dd June 30, 2025
.Dt RGBASM 5
.Os
.Sh NAME
.Nm rgbasm
.Nd language documentation
.Sh DESCRIPTION
This is the full description of the assembly language used by
.Xr rgbasm 1 .
For the full description of instructions in the machine language supported by the Game Boy CPU, see
.Xr gbz80 7 .
.Pp
It is advisable to have some familiarity with the Game Boy hardware before reading this document.
RGBDS is specifically targeted at the Game Boy, and thus a lot of its features tie directly to its concepts.
This document is not intended to be a Game Boy hardware reference.
.Pp
Generally,
.Dq the linker
will refer to
.Xr rgblink 1 ,
but any program that processes RGBDS object files (described in
.Xr rgbds 5 )
can be used in its place.
.Sh SYNTAX
The syntax is line-based, just as in any other assembler.
Each line may have components in this order:
.Pp
.Dl Oo Ar directive Oc Oo ;\  Ns Ar comment Oc
.Dl Oo Ar label : Oc Oo Ar instruction Oo :: Ar instruction ... Oc Oc Oo ;\  Ns Ar comment Oc
.Pp
Directives are commands to the assembler itself, such as
.Ic PRINTLN ,
.Ic SECTION ,
or
.Ic OPT .
.Pp
Labels tie a name to a specific location within a section (see
.Sx Labels
below).
.Pp
Instructions are assembled into Game Boy opcodes.
Multiple instructions on one line can be separated by double colons
.Ql :: .
.Pp
The available instructions are documented in
.Xr gbz80 7 .
.Pp
Note that where an instruction requires an 8-bit register
.Ar r8 ,
.Nm
can interpret
.Ic HIGH Ns Pq Ar r16
as the top 8-bit register of the given
.Ar r16 ,
for example,
.Ic HIGH Ns Pq Ic HL
for
.Ic H ;
and
.Ic LOW Ns Pq Ar r16
as the bottom one, for example,
.Ic LOW Ns Pq Ic HL
for
.Ic L
(except for
.Ic LOW Ns Pq Ic AF ,
since
.Ic F
is not a valid register).
.Pp
Note also that where an instruction requires a condition code
.Ar cc ,
.Nm
can interpret
.Ic ! Ns Ar cc
as the opposite condition code; for example,
.Ic !nz
for
.Ic z .
.Pp
All reserved keywords (directives, register names, etc.) are case-insensitive;
all identifiers (labels and other symbol names) are case-sensitive.
.Pp
Comments are used to give humans information about the code, such as explanations.
The assembler
.Em always
ignores comments and their contents.
.Pp
There are two kinds of comments, inline and block.
Inline comments are anything that follows a semicolon
.Ql \&;
not inside a string, until the end of the line.
Block comments, beginning with
.Ql /*
and ending with
.Ql */ ,
can be split across multiple lines, or occur in the middle of an expression.
.Pp
An example demonstrating these syntax features:
.Bd -literal -offset indent
SECTION "My Code", ROM0\ \ ;\ a directive
MyFunction:\ \ \ \ \ \ \ \ \ \ \ \ \ \ ;\ a label
    push hl\ \ \ \ \ \ \ \ \ \ \ \ \ \ ;\ an instruction
    /* ...and multiple instructions,
       with mixed case */
    ld a, [hli] :: LD H, [HL] :: Ld l, a
    pop /*wait for it*/ hl
    ret
.Ed
.Pp
Sometimes lines can be too long and it may be necessary to split them.
To do so, put a backslash at the end of the line:
.Bd -literal -offset indent
    DB 1, 2, 3,\ \e
       4, 5, 6,\ \e\ ;\ Put it before any comments
       7, 8, 9
    DB "Hello,\ \e\ \ ;\ Space before the \e is included
world!"\ \ \ \ \ \ \ \ \ \ \ ;\ Any leading space is included
.Ed
.Ss Symbol interpolation
A funky feature is writing a symbol between
.Ql {braces} ,
called
.Dq symbol interpolation .
This will paste the symbol's contents as if they were part of the source file.
If it is a string symbol, its characters are simply inserted as-is.
If it is a numeric symbol, its value is converted to hexadecimal notation with a dollar sign
.Sq $
prepended.
.Pp
Symbol interpolations can be nested, too!
.Bd -literal -offset indent
DEF topic EQUS "life, the universe, and \e"everything\e""
DEF meaning EQUS "answer"
;\ Defines answer = 42
DEF {meaning} = 42
;\ Prints "The answer to life, the universe, and "everything" is $2A"
PRINTLN "The {meaning} to {topic} is {{meaning}}"
PURGE topic, meaning, {meaning}
.Ed
.Pp
Symbols can be
.Em interpolated
even in the contexts that disable automatic
.Em expansion
of string constants:
.Ql name
will be expanded in all of
.Ql DEF({name}) ,
.Ql DEF {name} EQU/=/EQUS/etc ... ,
.Ql PURGE {name} ,
and
.Ql MACRO {name} ,
but, for example, won't be in
.Ql DEF(name) .
.Pp
It's possible to change the way symbols are printed by specifying a print format like so:
.Ql {fmt:symbol} .
The
.Ql fmt
specifier consists of these parts:
.Ql <sign><exact><align><pad><width><frac><prec><type> .
These parts are:
.Bl -column "<exact>"
.It Sy Part Ta Sy Meaning
.It Ql <sign> Ta May be
.Ql +
or
.Ql \  .
If specified, prints this character in front of non-negative numbers.
.It Ql <exact> Ta May be
.Ql # .
If specified, prints the value in an "exact" format: with a base prefix for non-decimal integer types
.Pq So $ Sc , So & Sc , or So % Sc ;
with a
.Ql q
precision suffix for fixed-point numbers; or with
.Ql \e
escape characters for strings.
.It Ql <align> Ta May be
.Ql - .
If specified, aligns left instead of right.
.It Ql <pad> Ta May be
.Ql 0 .
If specified, pads right-aligned numbers with zeros instead of spaces.
.It Ql <width> Ta May be one or more
.Ql 0
\[en]
.Ql 9 .
If specified, pads the value to this width, right-aligned with spaces by default.
.It Ql <frac> Ta May be
.Ql \&.
followed by one or more
.Ql 0
\[en]
.Ql 9 .
If specified, prints this many fractional digits of a fixed-point number.
Defaults to 5 digits, maximum 255 digits.
.It Ql <prec> Ta May be
.Ql q
followed by one or more
.Ql 0
\[en]
.Ql 9 .
If specified, prints a fixed-point number at this precision.
Defaults to the current
.Fl Q
option.
.It Ql <type> Ta Specifies the type of value.
.El
.Pp
All the format specifier parts are optional except the
.Ql <type> .
Valid print types are:
.Bl -column -offset indent "Type" "Lowercase hexadecimal" "Example"
.It Sy Type Ta Sy Format Ta Sy Example
.It Ql d Ta Signed decimal Ta -42
.It Ql u Ta Unsigned decimal Ta 42
.It Ql x Ta Lowercase hexadecimal Ta 2a
.It Ql X Ta Uppercase hexadecimal Ta 2A
.It Ql b Ta Binary Ta 101010
.It Ql o Ta Octal Ta 52
.It Ql f Ta Fixed-point Ta 1234.56789
.It Ql s Ta String Ta string contents
.El
.Pp
Examples:
.Bd -literal -offset indent
SECTION "Test", ROM0[2]
X:                 ;\ This works with labels **whose address is known**
DEF Y = 3          ;\ This also works with variables
DEF SUM EQU X + Y  ;\ And likewise with numeric constants
; Prints "%0010 + $3 == 5"
PRINTLN "{#05b:X} + {#x:Y} == {d:SUM}"

rsset 32
DEF PERCENT rb 1   ;\ Same with offset constants
DEF VALUE = 20
DEF RESULT = MUL(20.0, 0.32)
; Prints "32% of 20 = 6.40"
PRINTLN "{d:PERCENT}% of {d:VALUE} = {f:RESULT}"

DEF WHO EQUS STRLWR("WORLD")
; Prints "Hello world!"
PRINTLN "Hello {s:WHO}!"
.Ed
.Pp
Although, for these examples,
.Ic STRFMT
would be more appropriate; see
.Sx String expressions
below.
.Sh EXPRESSIONS
An expression can be composed of many things.
Numeric expressions are always evaluated using signed 32-bit math.
Zero is considered to be the only "false" number, all non-zero numbers (including negative) are "true".
.Pp
An expression is said to be "constant" if
.Nm
knows its value.
This is generally always the case, unless a label is involved, as explained in the
.Sx SYMBOLS
section.
However, some operators can be constant even with non-constant operands, as explained in
.Sx Operators
below.
.Pp
The instructions in the macro-language generally require constant expressions.
.Ss Numeric formats
There are a number of numeric formats.
.Bl -column -offset indent "Precise fixed-point" "Possible prefixes"
.It Sy Format type Ta Sy Possible prefixes Ta Sy Accepted characters
.It Decimal Ta none Ta 0123456789
.It Hexadecimal Ta Li $ , 0x , 0X Ta 0123456789ABCDEF
.It Octal Ta Li & , 0o , 0O Ta 01234567
.It Binary Ta Li % , 0b , 0B Ta 01
.It Fixed-point Ta none Ta 01234.56789
.It Precise fixed-point Ta none Ta 12.34q8
.It Character constant Ta none Ta \(dqABYZ\(dq
.It Game Boy graphics Ta Li \` Ta 0123
.El
.Pp
Underscores are also accepted in numbers, except at the beginning of one.
This can be useful for grouping digits, like
.Ql 123_456
or
.Ql %1100_1001 .
.Pp
The "character constant" form yields the value the character maps to in the current charmap.
For example, by default
.Pq refer to Xr ascii 7
.Sq \(dqA\(dq
yields 65.
See
.Sx Character maps
for information on charmaps.
.Pp
The last one, Game Boy graphics, is quite interesting and useful.
After the backtick, 8 digits between 0 and 3 are expected, corresponding to pixel values.
The resulting value is the two bytes of tile data that would produce that row of pixels.
For example,
.Sq \`01012323
is equivalent to
.Sq $0F55 .
.Pp
You can also use symbols, which are implicitly replaced with their value.
.Ss Operators
You can use these operators in numeric expressions (listed from highest to lowest precedence):
.Bl -column -offset indent "!= == <= >= < >"
.It Sy Operator Ta Sy Meaning
.It Li \&( \&) Ta Grouping
.It Li FUNC() Ta Built-in function call
.It Li ** Ta Exponentiation
.It Li + - ~ \&! Ta Unary plus, minus (negation), complement (bitwise negation), and Boolean negation
.It Li * / % Ta Multiplication, division, and modulo (remainder)
.It Li << >> >>> Ta Bit shifts (left, sign-extended right, zero-extended right)
.It Li & \&| ^ Ta Bitwise AND/OR/XOR
.It Li + - Ta Addition and subtraction
.It Li == != < > <= >= Ta Comparisons
.It Li && Ta Boolean AND
.It Li || Ta Boolean OR
.El
.Pp
.Sq **
raises a number to a non-negative power. It is the only
.Em right-associative
operator, meaning that
.Ql p ** q ** r
is equal to
.Ql p ** (q ** r) ,
not
.Ql (p ** q) ** r .
All other binary operators are left-associative.
.Pp
.Sq ~
complements a value by inverting all 32 of its bits.
.Pp
.Sq %
is used to get the remainder of the corresponding division, so that
.Ql x / y * y + x % y == x
is always true.
The result has the same sign as the divisor.
This makes
.Ql x % y
equal to
.Ql (x + y) % y
or
.Ql (x - y) % y .
.Pp
Shifting works by shifting all bits in the left operand either left
.Pq Sq <<
or right
.Pq Sq >>
by the right operand's amount.
When shifting left, all newly-inserted bits are reset; when shifting right, they are copies of the original most significant bit instead.
This makes
.Sq a << b
and
.Sq a >> b
equivalent to multiplying and dividing by 2 to the power of b, respectively.
.Pp
Comparison operators return 0 if the comparison is false, and 1 otherwise.
.Pp
Unlike in many other languages, and for technical reasons,
.Nm
still evaluates both operands of
.Sq &&
and
.Sq || .
.Pp
The operators
.Sq &&
and
.Sq &
with a zero constant as either operand will be constant 0, and
.Sq ||
with a non-zero constant as either operand will be constant 1, even if the other operand is non-constant.
.Pp
.Sq \&!
returns 1 if the operand was 0, and 0 otherwise.
Even a non-constant operand with any non-zero bits will return 0.
.Ss Integer functions
Besides operators, there are also some functions which have more specialized uses.
.Bl -column "BITWIDTH(n)"
.It Sy Name Ta Sy Operation
.It Fn HIGH n Ta Equivalent to Ql Po Ns Ar n No & $FF00 Pc >> 8 .
.It Fn LOW n Ta Equivalent to Ql Ar n No & $FF .
.EQ
delim $$
.EN
.It Fn BITWIDTH n Ta Returns the number of bits necessary to represent
.Ar n .
Some useful formulas:
.Ic BITWIDTH Ns ( Ar n Ns )\ \-\ 1
equals $\[lf] log sub 2 ( n ) \[rf]$,
.Ic BITWIDTH Ns Pq Ar n Ns \ \-\ 1
equals $\[lc] log sub 2 ( n ) \[rc]$, and
.No 32\ \-\  Ns Ic BITWIDTH Ns Pq Ar n
equals $roman clz ( n )$.
.It Fn TZCOUNT n Ta Returns $roman ctz ( n )$, the count of trailing zero bits at the end of the binary representation of
.Ar n .
.El
.EQ
delim off
.EN
.Ss Fixed-point expressions
Fixed-point numbers are technically just integers, but conceptually they have a decimal point at a fixed location (hence the name).
This gives them increased precision, at the cost of a smaller range, while remaining far cheaper to manipulate than floating-point numbers (which
.Nm
does not support).
.Pp
The default precision of all fixed-point numbers is 16 bits, meaning the lower 16 bits are used for the fractional part; so they count in 65536ths of 1.0.
This precision can be changed with the
.Fl Q
command-line option, and/or by
.Ic OPT Q
.Pq see Sx Changing options while assembling .
An individual fixed-point literal can specify its own precision, overriding the current default, by appending a
.Dq q
followed by the number of fractional bits: for example,
.Ql 1234.5q8
is equal to $0004d2_80
.EQ
delim $$
.EN
($= 1234.5 * 2 sup 8$).
.Pp
Since fixed-point values are still just integers, you can use them in normal integer expressions.
You can easily truncate a fixed-point number into an integer by shifting it right by the number of fractional bits.
It follows that you can convert an integer to a fixed-point number by shifting it left that same amount.
.Pp
Note that the current number of fractional bits can be computed as
.Ic TZCOUNT Ns Pq 1.0 .
.Pp
The following functions are designed to operate with fixed-point numbers:
.Bl -column -offset indent "ATAN2(y, x)"
.It Sy Name Ta Sy Operation
.It Fn DIV x y Ta Fixed-point division
.It Fn MUL x y Ta Fixed-point multiplication
.It Fn FMOD x y Ta Fixed-point modulo
.It Fn POW x y Ta $x sup y$
.It Fn LOG x y Ta Logarithm of $x$ to the base $y$
.It Fn ROUND x Ta Round $x$ to the nearest integer
.It Fn CEIL x Ta Round $x$ up to the nearest integer
.It Fn FLOOR x Ta Round $x$ down to the nearest integer
.It Fn SIN x Ta Sine of $x$
.It Fn COS x Ta Cosine of $x$
.It Fn TAN x Ta Tangent of $x$
.It Fn ASIN x Ta Inverse sine of $x$
.It Fn ACOS x Ta Inverse cosine of $x$
.It Fn ATAN x Ta Inverse tangent of $x$
.It Fn ATAN2 y x Ta Angle between $( x , y )$ and $( 1 , 0 )$
.El
.EQ
delim off
.EN
.Pp
There are no functions for fixed-point addition and subtraction, because the
.Sq +
and
.Sq -
operators can add and subtract pairs of fixed-point operands.
.Bd -ragged -offset indent
Note that some operators or functions are meaningful when combining integers and fixed-point values.
For example,
.Ql 2.0 * 3
is equivalent to
.Ql MUL(2.0, 3.0) ,
and
.Ql 6.0 / 2
is equivalent to
.Ql DIV(6.0, 2.0) .
Be careful and think about what the operations mean when doing this sort of thing.
.Ed
.Pp
All of these fixed-point functions can take an optional final argument, which is the precision to use for that one operation.
For example,
.Ql MUL(6.0q8, 7.0q8, 8)
will evaluate to
.Ql 42.0q8
no matter what value is set as the current
.Cm Q
option.
.Nm
.Em does not check precisions for consistency ,
so nonsensical input like
.Ql MUL(4.2q8, 6.9q12, 16)
will produce a nonsensical (but technically correct) result:
.Dq garbage in, garbage out .
.Pp
The
.Ic FMOD
function
is used to get the remainder of the corresponding fixed-point division, so that
.Ql MUL(DIV(x, y), y) + FMOD(x, y) == x
is always true.
The result has the same sign as the
.Em dividend ;
this is the opposite of how the integer modulo operator
.Sq %
works!
.Pp
The trigonometry functions
.Pq Ic SIN , Ic COS , Ic TAN , No etc
are defined in terms of a circle divided into 1.0
.Dq turns
.EQ
delim $$
.EN
(equal to $2 pi$ radians, or 360 degrees).
.EQ
delim off
.EN
.Pp
These functions are useful for automatic generation of various tables.
For example:
.Bd -literal -offset indent
; Generate a table of 128 sine values
; from sin(0.0) to sin(0.5) excluded,
; with amplitude scaled from [-1.0, 1.0] to [0.0, 128.0].
FOR angle, 0.0, 0.5, 0.5 / 128
    db MUL(SIN(angle) + 1.0, 128.0 / 2) >> 16
ENDR
.Ed
.Ss String expressions
The most basic string expression is any number of characters contained in double quotes
.Pq Ql \&"for instance" .
The backslash character
.Ql \e
is special in that it causes the character following it to be
.Dq escaped ,
meaning that it is treated differently from normal.
There are a number of escape sequences you can use within a string:
.Bl -column -offset indent "Sequence"
.It Sy Sequence Ta Sy Meaning
.It Ql \e\e Ta Backslash Pq escapes the escape character itself
.It Ql \e" Ta Double quote Pq does not terminate the string
.It Ql \e{ Ta Open curly brace Pq does not start interpolation
.It Ql \e} Ta Close curly brace Pq does not end interpolation
.It Ql \en Ta Newline Pq ASCII $0A
.It Ql \er Ta Carriage return Pq ASCII $0D
.It Ql \et Ta Tab Pq ASCII $09
.It Ql \e0 Ta Null Pq ASCII $00
.El
.Pp
Multi-line strings are contained in triple quotes
.Pq Ql \&"\&"\&"for instance""" .
Escape sequences work the same way in multi-line strings; however, literal newline characters will be included as-is, without needing to escape them with
.Ql \er
or
.Ql \en .
.Pp
Raw strings are prefixed by a hash
.Sq # .
Inside them, backslashes and braces are treated like regular characters, so they will not be expanded as macro arguments, interpolated symbols, or escape sequences.
For example, the raw string
.Ql #"\et\e1{s}\e"
is equivalent to the regular string
.Ql \&"\e\et\e\e1\e{s}\e\e" .
(Note that this prevents raw strings from including the double quote character.)
Raw strings also may be contained in triple quotes for them to be multi-line, so they can include literal newline or quote characters (although still not three quotes in a row).
.Pp
You can use the
.Sq ++
operator to concatenate two strings.
.Ql \&"str" ++ \&"ing"
is equivalent to
.Ql \&"string" ,
or to
.Ql STRCAT("str", \&"ing") .
.Pp
The following functions operate on string expressions, and return strings themselves.
.Bl -column "STRSLICE(str, start, stop)"
.It Sy Name Ta Sy Operation
.It Fn STRCAT strs... Ta Concatenates Ar strs .
.It Fn STRUPR str Ta Returns Ar str No with all ASCII letters
.Pq Ql a-z
in uppercase.
.It Fn STRLWR str Ta Returns Ar str No with all ASCII letters
.Pq Ql A-Z
in lowercase.
.It Fn STRSLICE str start stop Ta Returns a substring of Ar str No starting at Ar start No and ending at Ar stop No (exclusive). If Ar stop No is not specified, the substring continues to the end of Ar str .
.It Fn STRRPL str old new Ta Returns Ar str No with each non-overlapping occurrence of the substring Ar old No replaced with Ar new .
.It Fn STRFMT fmt args... Ta Returns the string Ar fmt No with each
.Ql %spec
pattern replaced by interpolating the format
.Ar spec
.Pq using the same syntax as Sx Symbol interpolation
with its corresponding argument in
.Ar args
.Pq So %% Sc is replaced by the So % Sc character .
.It Fn STRCHAR str idx Ta Returns the substring of Ar str No for the charmap entry at Ar idx No with the current charmap . Pq Ar idx No counts charmap entries, not characters.
.It Fn REVCHAR vals... Ta Returns the string that is mapped to Ar vals No with the current charmap. If there is no unique charmap entry for Ar vals Ns , an error occurs.
.El
.Pp
The following functions operate on string expressions, but return integers.
.Bl -column "STRRFIND(str, sub)"
.It Sy Name Ta Sy Operation
.It Fn STRLEN str Ta Returns the number of characters in Ar str .
.It Fn STRCMP str1 str2 Ta Compares Ar str1 No and Ar str2 No according to ASCII ordering of their characters. Returns -1 if Ar str1 No is lower than Ar str2 Ns , 1 if Ar str1 No is greater than Ar str2 Ns , or 0 if they match.
.It Fn STRFIND str sub Ta Returns the first index of Ar sub No in Ar str Ns , or -1 if it's not present.
.It Fn STRRFIND str sub Ta Returns the last index of Ar sub No in Ar str Ns , or -1 if it's not present.
.It Fn INCHARMAP str Ta Returns 1 if Ar str No has an entry in the current charmap, or 0 otherwise.
.It Fn CHARLEN str Ta Returns the number of charmap entries in Ar str No with the current charmap.
.It Fn CHARCMP str1 str2 Ta Compares Ar str1 No and Ar str2 No according to their charmap entry values with the current charmap. Returns -1 if Ar str1 No is lower than Ar str2 Ns , 1 if Ar str1 No is greater than Ar str2 Ns , or 0 if they match.
.It Fn CHARSIZE char Ta Returns how many values are in the charmap entry for Ar char No with the current charmap.
.It Fn CHARVAL char idx Ta Returns the value at Ar idx No of the charmap entry for Ar char .
.El
.Pp
Note that indexes count starting from 0 at the beginning, or from -1 at the end.
The characters of a string are counted by
.Ql STRLEN ;
the charmap entries of a string are counted by
.Ql CHARLEN ;
and the values of a charmap entry are counted by
.Ql CHARSIZE .
.Pp
The following legacy functions are similar to other functions that operate on string expressions, but for historical reasons, they count starting from
.Em position 1 ,
not from index 0!
(Position -1 still counts from the end.)
.Bl -column "STRSUB(str, pos, len)"
.It Sy Name Ta Sy Operation
.It Fn STRSUB str pos len Ta Returns a substring of Ar str No starting at Ar pos No and Ar len No characters long. If Ar len No is not specified, the substring continues to the end of Ar str No .
.It Fn STRIN str sub Ta Returns the first position of Ar sub No in Ar str Ns , or 0 if it's not present.
.It Fn STRRIN str sub Ta Returns the last position of Ar sub No in Ar str Ns , or 0 if it's not present.
.It Fn CHARSUB str pos Ta Returns the substring of Ar str No for the charmap entry at Ar pos No with the current charmap . Pq Ar pos No counts charmap entries, not characters.
.El
.Ss Character maps
When writing text strings that are meant to be displayed on the Game Boy, the character encoding in the ROM may need to be different than the source file encoding.
For example, the tiles used for uppercase letters may be placed starting at tile index 128, which differs from ASCII starting at 65.
.Pp
Character maps allow mapping strings to arbitrary sequences of numbers:
.Bd -literal -offset indent
CHARMAP "A", 42
CHARMAP ":)", 39
CHARMAP "<br>", 13, 10
CHARMAP "&euro;", $20ac
.Ed
.Pp
This would result in
.Ql db \(dqAmen :)<br>\(dq
being equivalent to
.Ql db 42, 109, 101, 110, 32, 39, 13, 10 ,
and
.Ql dw \(dq25&euro;\(dq
being equivalent to
.Ql dw 50, 53, $20ac .
.Pp
Any characters in a string without defined mappings will be copied directly, using the source file's encoding of characters to bytes.
.Pp
It is possible to create multiple character maps and then switch between them as desired.
This can be used to encode debug information in ASCII and use a different encoding for other purposes, for example.
Initially, there is one character map called
.Sq main
and it is automatically selected as the current character map from the beginning.
There is also a character map stack that can be used to save and restore which character map is currently active.
.Bl -column "NEWCHARMAP name, basename"
.It Sy Command Ta Sy Meaning
.It Ic NEWCHARMAP Ar name Ta Creates a new, empty character map called Ar name No and switches to it .
.It Ic NEWCHARMAP Ar name , basename Ta Creates a new character map called Ar name , No copied from character map Ar basename , No and switches to it .
.It Ic SETCHARMAP Ar name Ta Switch to character map Ar name .
.It Ic PUSHC Ta Push the current character map onto the stack.
.It Ic PUSHC Ar name Ta Push the current character map onto the stack and switch to character map Ar name .
.It Ic POPC Ta Pop a character map off the stack and switch to it.
.El
.Pp
.Sy Note :
Modifications to a character map take effect immediately from that point onward.
.Ss Other functions
There are a few other functions that do things beyond numeric or string operations:
.Bl -column "SECTION(symbol)"
.It Sy Name Ta Sy Operation
.It Fn DEF symbol Ta Returns 1 if
.Ar symbol
has been defined, 0 otherwise.
String constants are not expanded within the parentheses.
.It Fn ISCONST arg Ta Returns 1 if Ar arg Ap s value is known by RGBASM (e.g. if it can be an argument to
.Ic IF ) ,
or 0 if only RGBLINK can compute its value.
.It Fn BANK arg Ta Returns a bank number.
If
.Ar arg
is the symbol
.Ic @ ,
this function returns the bank of the current section.
If
.Ar arg
is a string, it returns the bank of the section that has that name.
If
.Ar arg
is a label, it returns the bank number the label is in.
The result may be constant if
.Nm
is able to compute it.
.It Fn SECTION symbol Ta Returns the name of the section that
.Ar symbol
is in.
.Ar symbol
must have been defined already.
.It Fn SIZEOF arg Ta If
.Ar arg
is a string, this function returns the size of the section named
.Ar arg .
If
.Ar arg
is a section type keyword, it returns the size of that section type.
The result is not constant, since only RGBLINK can compute its value.
.It Fn STARTOF arg Ta If
.Ar arg
is a string, this function returns the starting address of the section named
.Ar arg .
If
.Ar arg
is a section type keyword, it returns the starting address of that section type.
The result is not constant, since only RGBLINK can compute its value.
.El
.Sh SECTIONS
Before you can start writing code, you must define a section.
This tells the assembler what kind of information follows and, if it is code, where to put it.
.Pp
.Dl SECTION Ar name , type
.Dl SECTION Ar name , type , options
.Dl SECTION Ar name , type Ns Bo Ar addr Bc
.Dl SECTION Ar name , type Ns Bo Ar addr Bc , Ar options
.Pp
.Ar name
is a string enclosed in double quotes, and can be a new name or the name of an existing section.
If the type doesn't match, an error occurs.
All other sections must have a unique name, even in different source files, or the linker will treat it as an error.
.Pp
Possible section
.Ar type Ns s
are as follows:
.Bl -tag -width Ds
.It Ic ROM0
A ROM section.
.Ar addr
can range from
.Ad $0000
to
.Ad $3FFF ,
or
.Ad $0000
to
.Ad $7FFF
if tiny ROM mode is enabled in the linker.
.It Ic ROMX
A banked ROM section.
.Ar addr
can range from
.Ad $4000
to
.Ad $7FFF .
.Ar bank
can range from 1 to 511.
Becomes an alias for
.Ic ROM0
if tiny ROM mode is enabled in the linker.
.It Ic VRAM
A banked video RAM section.
.Ar addr
can range from
.Ad $8000
to
.Ad $9FFF .
.Ar bank
can be 0 or 1, but bank 1 is unavailable if DMG mode is enabled in the linker.
.It Ic SRAM
A banked external (save) RAM section.
.Ar addr
can range from
.Ad $A000
to
.Ad $BFFF .
.Ar bank
can range from 0 to 15.
.It Ic WRAM0
A general-purpose RAM section.
.Ar addr
can range from
.Ad $C000
to
.Ad $CFFF ,
or
.Ad $C000
to
.Ad $DFFF
if WRAM0 mode is enabled in the linker.
.It Ic WRAMX
A banked general-purpose RAM section.
.Ar addr
can range from
.Ad $D000
to
.Ad $DFFF .
.Ar bank
can range from 1 to 7.
Becomes an alias for
.Ic WRAM0
if WRAM0 mode is enabled in the linker.
.It Ic OAM
An object attribute RAM section.
.Ar addr
can range from
.Ad $FE00
to
.Ad $FE9F .
.It Ic HRAM
A high RAM section.
.Ar addr
can range from
.Ad $FF80
to
.Ad $FFFE .
.El
.Pp
Since RGBDS produces ROMs, code and data can only be placed in
.Ic ROM0
and
.Ic ROMX
sections.
To put some in RAM, have it stored in ROM, and copy it to RAM.
.Pp
.Ar option Ns s are comma-separated and may include:
.Bl -tag -width Ds
.It Ic BANK Ns Bq Ar bank
Specify which
.Ar bank
for the linker to place the section in.
See above for possible values for
.Ar bank ,
depending on
.Ar type .
.It Ic ALIGN Ns Bq Ar align , offset
Place the section at an address whose
.Ar align
least-significant bits are equal to
.Ar offset .
Note that
.Ic ALIGN Ns Bq Ar align
is a shorthand for
.Ic ALIGN Ns Bq Ar align , No 0 .
This option can be used with
.Bq Ar addr ,
as long as they don't contradict each other.
It's also possible to request alignment in the middle of a section; see
.Sx Requesting alignment
below.
.El
.Pp
If
.Bq Ar addr
is not specified, the section is considered
.Dq floating ;
the linker will automatically calculate an appropriate address for the section.
Similarly, if
.Ic BANK Ns Bq Ar bank
is not specified, the linker will automatically find a bank with enough space.
.Pp
Sections can also be placed by using a linker script file.
The format is described in
.Xr rgblink 5 .
They allow the user to place floating sections in the desired bank in the order specified in the script.
This is useful if the sections can't be placed at an address manually because the size may change, but they have to be together.
.Pp
Section examples:
.Bl -item
.It
.Bd -literal -offset indent
SECTION "Cool Stuff", ROMX
.Ed
.Pp
This switches to the section called
.Dq CoolStuff ,
creating it if it doesn't already exist.
It can end up in any ROM bank.
Code and data may follow.
.It
If it is needed, the the base address of the section can be specified:
.Bd -literal -offset indent
SECTION "Cool Stuff", ROMX[$4567]
.Ed
.It
An example with a fixed bank:
.Bd -literal -offset indent
SECTION "Cool Stuff", ROMX[$4567], BANK[3]
.Ed
.It
And if you want to force only the section's bank, and not its position within the bank, that's also possible:
.Bd -literal -offset indent
SECTION "Cool Stuff", ROMX, BANK[7]
.Ed
.It
Alignment examples:
The first one could be useful for defining an OAM buffer to be DMA'd, since it must be aligned to 256 bytes.
The second could also be appropriate for GBC HDMA, or for an optimized copy code that requires alignment.
.Bd -literal -offset indent
SECTION "OAM Data", WRAM0, ALIGN[8] ;\ align to 256 bytes
SECTION "VRAM Data", ROMX, BANK[2], ALIGN[4] ;\ align to 16 bytes
.Ed
.El
.Pp
The current section can be ended without starting a new section by using
.Ic ENDSECTION .
This directive will clear the section context, so you can no longer write code until you start another section.
It can be useful to avoid accidentally defining code or data in the wrong section.
.Ss Section stack
.Ic POPS
and
.Ic PUSHS
provide the interface to the section stack.
The number of entries in the stack is limited only by the amount of memory in your machine.
.Pp
.Ic PUSHS
will push the current section context on the section stack.
.Ic POPS
can then later be used to restore it.
Useful for defining sections in included files when you don't want to override the section context at the point the file was included.
.Pp
.Ic PUSHS
can also take the same arguments as
.Ic SECTION ,
in order to push the current section context and define a new section at the same time:
.Bd -literal -offset indent
SECTION "Code", ROM0
Function:
    ld a, 42
    PUSHS "Variables", WRAM0
        wAnswer: db
    POPS
    ld [wAnswer], a
.Ed
.Ss RAM code
Sometimes you want to have some code in RAM.
But then you can't simply put it in a RAM section, you have to store it in ROM and copy it to RAM at some point.
.Pp
This means the code (or data) will not be stored in the place it gets executed.
Luckily,
.Ic LOAD
blocks are the perfect solution to that.
Here's an example of how to use them:
.Bd -literal -offset indent
SECTION "LOAD example", ROMX
CopyCode:
    ld de, RAMCode
    ld hl, RAMLocation
    ld c, RAMCode.end - RAMCode
\&.loop
    ld a, [de]
    inc de
    ld [hli], a
    dec c
    jr nz, .loop
    ret

RAMCode:
  LOAD "RAM code", WRAM0
RAMLocation:
    ld hl, .string
    ld de, $9864
\&.copy
    ld a, [hli]
    ld [de], a
    inc de
    and a
    jr nz, .copy
    ret

\&.string
    db "Hello World!\e0"
  ENDL
\&.end
.Ed
.Pp
A
.Ic LOAD
block feels similar to a
.Ic SECTION
declaration because it creates a new one.
All data and code generated within such a block is placed in the current section like usual, but all labels are created as if they were placed in this newly-created section.
.Pp
In the example above, all of the code and data will end up in the
.Dq LOAD example
section.
You will notice the
.Sq RAMCode
and
.Sq RAMLocation
labels.
The former is situated in ROM, where the code is stored, the latter in RAM, where the code will be loaded.
.Pp
You cannot nest
.Ic LOAD
blocks, nor can you change or stop the current section within them.
.Pp
The current
.Ic LOAD
block can be ended by using
.Ic ENDL .
This directive is only necessary if you want to resume writing code in its containing ROM section.
Any of
.Ic LOAD , SECTION , ENDSECTION ,
or
.Ic POPS
will end the current
.Ic LOAD
block before performing its own function.
.Pp
.Ic LOAD
blocks can use the
.Ic UNION
or
.Ic FRAGMENT
modifiers as described in
.Sx Unionized sections
below.
.Ss Unionized sections
When you're tight on RAM, you may want to define overlapping static memory allocations, as explained in the
.Sx Allocating overlapping spaces in RAM
section.
However, a
.Ic UNION
only works within a single file, so it can't be used e.g. to define temporary variables across several files, all of which use the same statically allocated memory.
Unionized sections solve this problem.
To declare an unionized section, add a
.Ic UNION
keyword after the
.Ic SECTION
one; the declaration is otherwise not different.
Unionized sections follow some different rules from normal sections:
.Bl -bullet -offset indent
.It
The same unionized section (i.e. having the same name) can be declared several times per
.Nm
invocation, and across several invocations.
Different declarations are treated and merged identically whether within the same invocation, or different ones.
.It
If one section has been declared as unionized, all sections with the same name must be declared unionized as well.
.It
All declarations must have the same type.
For example, even if
.Xr rgblink 1 Ap s
.Fl w
flag is used,
.Ic WRAM0
and
.Ic WRAMX
types are still considered different.
.It
Different constraints (alignment, bank, etc.) can be specified for each unionized section declaration, but they must all be compatible.
For example, alignment must be compatible with any fixed address, all specified banks must be the same, etc.
.It
Unionized sections cannot have type
.Ic ROM0
or
.Ic ROMX .
.El
.Pp
Different declarations of the same unionized section are not appended, but instead overlaid on top of each other, just like
.Sx Allocating overlapping spaces in RAM .
Similarly, the size of an unionized section is the largest of all its declarations.
.Ss Section fragments
Section fragments are sections with a small twist: when several of the same name are encountered, they are concatenated instead of producing an error.
This works within the same file (paralleling the behavior "plain" sections has in previous versions), but also across object files.
To declare an section fragment, add a
.Ic FRAGMENT
keyword after the
.Ic SECTION
one; the declaration is otherwise not different.
However, similarly to
.Sx Unionized sections ,
some rules must be followed:
.Bl -bullet -offset indent
.It
If one section has been declared as fragment, all sections with the same name must be declared fragments as well.
.It
All declarations must have the same type.
For example, even if
.Xr rgblink 1 Ap s
.Fl w
flag is used,
.Ic WRAM0
and
.Ic WRAMX
types are still considered different.
.It
Different constraints (alignment, bank, etc.) can be specified for each section fragment declaration, but they must all be compatible.
For example, alignment must be compatible with any fixed address, all specified banks must be the same, etc.
.It
A section fragment may not be unionized; after all, that wouldn't make much sense.
.El
.Pp
When RGBASM merges two fragments, the one encountered later is appended to the one encountered earlier.
.Pp
When RGBLINK merges two fragments, the one whose file was specified last is appended to the one whose file was specified first.
For example, assuming
.Ql bar.o ,
.Ql baz.o ,
and
.Ql foo.o
all contain a fragment with the same name, the command
.Dl rgblink -o rom.gb baz.o foo.o bar.o
would produce the fragment from
.Ql baz.o
first, followed by the one from
.Ql foo.o ,
and the one from
.Ql bar.o
last.
.Ss Fragment literals
Fragment literals are useful for short blocks of code or data that are only referenced once.
They are section fragments created by surrounding instructions or directives with
.Ql [[
double brackets
.Ql ]] ,
without a separate
.Ic SECTION FRAGMENT
declaration.
.Pp
The content of a fragment literal becomes a
.Ic SECTION FRAGMENT ,
sharing the same name and bank as its parent ROM section, but without any other constraints.
The parent section also becomes a
.Ic FRAGMENT
if it was not one already, so that it can be merged with its fragment literals.
RGBLINK merges the fragments in no particular order.
.Pp
A fragment literal can take the place of any 16-bit integer constant
.Ql n16
from the
.Xr gbz80 7
documentation, as well as a
.Ic DW
item.
The fragment literal then evaluates to its starting address.
For example, you can
.Ic CALL
or
.Ic JP
to a fragment literal.
.Pp
This code using named labels:
.Bd -literal -offset indent
DataTable:
    dw First
    dw Second
    dw Third
First:  db 1
Second: db 4
Third:  db 9
Routine:
    push hl
    ld hl, Left
    jr z, .got_it
    ld hl, Right
\&.got_it
    call .print
    pop hl
    ret
\&.print:
    ld de, $1003
    ld bc, STARTOF(VRAM)
    jp Print
Left:  db "left\e0"
Right: db "right\e0"
.Ed
.Pp
is equivalent to this code using fragment literals:
.Bd -literal -offset indent
DataTable:
    dw [[ db 1 ]]
    dw [[ db 4 ]]
    dw [[ db 9 ]]
Routine:
    push hl
    ld hl, [[ db "left\e0" ]]
    jr z, .got_it
    ld hl, [[ db "right\e0" ]]
\&.got_it
    call [[
        ld de, $1003
        ld bc, STARTOF(VRAM)
        jp Print
    ]]
    pop hl
    ret
.Ed
.Pp
The difference is that the example using fragment literals does not declare a particular order for its pieces.
.Pp
Fragment literals can be arbitrarily nested, so extreme use cases are
.Em technically
possible.
This code using named labels:
.Bd -literal -offset indent
dw FortyTwo
FortyTwo:
    call Sub1
    jr Sub2
Sub1:
    ld a, [Twenty]
    ret
Twenty: db 20
Sub2:
    jp Sub3
Sub3:
    call Sub1
    inc a
    add a
    ret
.Ed
.Pp
is equivalent to this code using fragment literals:
.Bd -literal -offset indent
dw [[
    call [[
        Sub1: ld a, [ [[db 20]] ] :: ret
    ]]
    jr [[
        jp [[ call Sub1 :: inc a :: add a :: ret ]]
    ]]
]]
.Ed
.Sh SYMBOLS
RGBDS supports several types of symbols:
.Bl -hang
.It Sy Label
Numeric symbol designating a memory location.
May or may not have a value known at assembly time.
.It Sy Constant
Numeric symbol whose value has to be known at assembly time.
.It Sy Macro
A block of
.Nm
code that can be invoked later.
.It Sy String
A text string that can be expanded later, similarly to a macro.
.El
.Pp
Symbol names can contain ASCII letters, numbers, underscores
.Sq _ ,
hashes
.Sq # ,
dollar signs
.Sq $ ,
and at signs
.Sq @ .
However, they must begin with either a letter or an underscore.
Additionally, label names can contain up to a single dot
.Ql \&. ,
which may not be the first character.
.Pp
A symbol cannot have the same name as a reserved keyword, unless its name is a
.Dq raw identifier
prefixed by a hash
.Sq # .
For example,
.Ql #load
denotes a symbol named
.Ql load ,
and
.Ql #LOAD
denotes a different symbol named
.Ql LOAD ;
in both cases the
.Sq #
prevents them from being treated as the keyword
.Ic LOAD .
.Ss Labels
One of the assembler's main tasks is to keep track of addresses for you, so you can work with meaningful names instead of
.Dq magic
numbers.
Labels enable just that: a label ties a name to a specific location within a section.
A label resolves to a bank and address, determined at the same time as its parent section's (see further in this section).
.Pp
A label is defined by writing its name at the beginning of a line, followed by one or two colons, without any whitespace between the label name and the colon(s).
Declaring a label (global or local) with two colons
.Ql ::
will define and
.Ic EXPORT
it at the same time.
(See
.Sx Exporting and importing symbols
below).
When defining a local label, the colon can be omitted, and
.Nm
will act as if there was only one.
.Pp
A label is said to be
.Em local
if its name contains a dot
.Ql \&. ;
otherwise, it is said to be
.Em global
(not to be mistaken with
.Dq exported ,
explained in
.Sx Exporting and importing symbols
below).
More than one dot in label names is not allowed.
.Pp
For convenience, local labels can use a shorthand syntax: when a symbol name starting with a dot is found (for example, inside an expression, or when declaring a label), then the current
.Dq label scope
is implicitly prepended.
.Pp
Defining a global label sets it as the current
.Dq label scope ,
until the next global label definition, or the end of the current section.
.Pp
Here are some examples of label definitions:
.Bd -literal -offset indent
GlobalLabel:
AnotherGlobal:
\&.locallabel ;\ This defines "AnotherGlobal.locallabel"
\&.another_local:
AnotherGlobal.with_another_local:
ThisWillBeExported:: ;\ Note the two colons
ThisWillBeExported.too::
.Ed
.Pp
In a numeric expression, a label evaluates to its address in memory.
.Po To obtain its bank, use the
.Ql BANK()
function described in
.Sx Other functions
.Pc .
For example, given the following,
.Ql ld de, vPlayerTiles
would be equivalent to
.Ql ld de, $80C0
assuming the section ends up at
.Ad $80C0 :
.Bd -literal -offset indent
SECTION "Player tiles", VRAM
vPlayerTiles:
    ds 6 * 16
\&.end
.Ed
.Pp
A label's location (and thus value) is usually not determined until the linking stage, so labels usually cannot be used as constants.
However, if the section in which the label is defined has a fixed base address, its value is known at assembly time.
.Pp
Also, while
.Nm
obviously can compute the difference between two labels if both are constant, it is also able to compute the difference between two non-constant labels if they both belong to the same section, such as
.Ql PlayerTiles
and
.Ql PlayerTiles.end
above.
.Ss Anonymous labels
Anonymous labels are useful for short blocks of code.
They are defined like normal labels, but without a name before the colon.
Anonymous labels are independent of label scoping, so defining one does not change the scoped label, and referencing one is not affected by the current scoped label.
.Pp
Anonymous labels are referenced using a colon
.Ql \&:
followed by pluses
.Ql +
or minuses
.Ql - .
Thus
.Ic :+
references the next one after the expression,
.Ic :++
the one after that;
.Ic :-
references the one before the expression;
and so on.
.Bd -literal -offset indent
    ld hl, :++
:   ld a, [hli] ; referenced by "jr nz"
    ldh [c], a
    dec c
    jr nz, :-
    ret

:   ; referenced by "ld hl"
    dw $7FFF, $1061, $03E0, $58A5
.Ed
.Ss Variables
An equal sign
.Sq =
is used to define mutable numeric symbols.
Unlike the other symbols described below, variables can be redefined.
This is useful for internal symbols in macros, for counters, etc.
.Bd -literal -offset indent
DEF ARRAY_SIZE EQU 4
DEF COUNT = 2
DEF COUNT = 3
DEF COUNT = ARRAY_SIZE + COUNT
DEF COUNT *= 2
;\ COUNT now has the value 14
.Ed
.Pp
Note that colons
.Ql \&:
following the name are not allowed.
.Pp
Variables can be conveniently redefined by compound assignment operators like in C:
.Bl -column -offset indent "*= /= %="
.It Sy Operator Ta Sy Meaning
.It Li += -= Ta Compound plus/minus
.It Li *= /= %= Ta Compound multiply/divide/modulo
.It Li <<= >>= Ta Compound shift left/right
.It Li &= \&|= ^= Ta Compound and/or/xor
.El
.Pp
Examples:
.Bd -literal -offset indent
DEF x = 10
DEF x += 1    ; x == 11
DEF y = x - 1 ; y == 10
DEF y *= 2    ; y == 20
DEF y >>= 1   ; y == 10
DEF x ^= y    ; x == 1
.Ed
.Pp
Declaring a variable with
.Ic EXPORT DEF
or
.Ic EXPORT REDEF
will define and
.Ic EXPORT
it at the same time.
(See
.Sx Exporting and importing symbols
below).
.Ss Numeric constants
.Ic EQU
is used to define numeric constant symbols.
Unlike
.Sq =
above, constants defined this way cannot be redefined.
These constants can be used for unchanging values such as properties of the hardware.
.Bd -literal -offset indent
def SCREEN_WIDTH  equ 160 ;\ In pixels
def SCREEN_HEIGHT equ 144
.Ed
.Pp
Note that colons
.Ql \&:
following the name are not allowed.
.Pp
If you
.Em really
need to, the
.Ic REDEF
keyword will define or redefine a numeric constant symbol.
(It can also be used for variables, although it's not necessary since they are mutable.)
This can be used, for example, to update a constant using a macro, without making it mutable in general.
.Bd -literal -offset indent
    def NUM_ITEMS equ 0
MACRO add_item
    redef NUM_ITEMS equ NUM_ITEMS + 1
    def ITEM_{02x:NUM_ITEMS} equ \e1
ENDM
    add_item 1
    add_item 4
    add_item 9
    add_item 16
    assert NUM_ITEMS == 4
    assert ITEM_04 == 16
.Ed
.Pp
Declaring a numeric constant with
.Ic EXPORT DEF
or
.Ic EXPORT REDEF
will define and
.Ic EXPORT
it at the same time.
(See
.Sx Exporting and importing symbols
below).
.Ss Offset constants
The RS group of commands is a handy way of defining structure offsets:
.Bd -literal -offset indent
               RSRESET
DEF str_pStuff RW   1
DEF str_tData  RB   256
DEF str_bCount RB   1
DEF str_SIZEOF RB   0
.Ed
.Pp
The example defines four constants as if by:
.Bd -literal -offset indent
DEF str_pStuff EQU 0
DEF str_tData  EQU 2
DEF str_bCount EQU 258
DEF str_SIZEOF EQU 259
.Ed
.Pp
There are five commands in the RS group of commands:
.Bl -column "DEF name RB constexpr"
.It Sy Command Ta Sy Meaning
.It Ic RSRESET Ta Equivalent to Ql RSSET 0 .
.It Ic RSSET Ar constexpr Ta Sets the Ic _RS No counter to Ar constexpr .
.It Ic DEF Ar name Ic RB Ar constexpr Ta Sets Ar name No to Ic _RS No and then adds Ar constexpr No to Ic _RS .
.It Ic DEF Ar name Ic RW Ar constexpr Ta Sets Ar name No to Ic _RS No and then adds Ar constexpr No * 2 to Ic _RS .
.It Ic DEF Ar name Ic RL Ar constexpr Ta Sets Ar name No to Ic _RS No and then adds Ar constexpr No * 4 to Ic _RS .
.El
.Pp
If the
.Ar constexpr
argument to
.Ic RB , RW ,
or
.Ic RL
is omitted, it's assumed to be 1.
.Pp
Note that colons
.Ql \&:
following the name are not allowed.
.Pp
Declaring an offset constant with
.Ic EXPORT DEF
will define and
.Ic EXPORT
it at the same time.
(See
.Sx Exporting and importing symbols
below).
.Ss String constants
.Ic EQUS
is used to define string constant symbols.
Wherever the assembler reads a string constant, it gets
.Em expanded :
the symbol's name is replaced with its contents, similarly to
.Ic #define
in the C programming language.
This expansion is disabled in a few contexts:
.Ql DEF(name) ,
.Ql DEF name EQU/=/EQUS/etc ... ,
.Ql PURGE name ,
and
.Ql MACRO name
will not expand string constants in their names.
Expansion is also disabled if the string constant's name is a raw identifier prefixed by a hash
.Sq # .
.Bd -literal -offset indent
DEF COUNTREG EQUS "[hl+]"
    ld a, COUNTREG

DEF PLAYER_NAME EQUS "\e"John\e""
    db PLAYER_NAME
.Ed
.Pp
This will be interpreted as:
.Bd -literal -offset indent
    ld a, [hl+]
    db "John"
.Ed
.Pp
String constants can also be used to define small one-line macros:
.Bd -literal -offset indent
DEF pusha EQUS "push af\enpush bc\enpush de\enpush hl\en"
.Ed
.Pp
Note that colons
.Ql \&:
following the name are not allowed.
.Pp
String constants, like numeric constants, cannot be redefined.
However, the
.Ic REDEF
keyword will define or redefine a string constant symbol.
For example:
.Bd -literal -offset indent
DEF s EQUS "Hello, "
REDEF s EQUS "{s}world!"
; prints "Hello, world!"
PRINTLN "{s}\en"
.Ed
.Pp
String constants can't be exported or imported.
.Pp
.Sy Important note :
When a string constant is expanded, its expansion may contain another string constant, which will be expanded as well, and may be recursive.
If this creates an infinite loop,
.Nm
will error out once a certain depth is
reached (see the
.Fl r
command-line option in
.Xr rgbasm 1 ) .
The same problem can occur if the expansion of a string constant invokes a macro, which itself expands.
.Ss Macros
One of the best features of an assembler is the ability to write macros for it.
Macros can be called with arguments, and can react depending on input using
.Ic IF
constructs.
.Bd -literal -offset indent
MACRO my_macro
    ld a, 80
    call MyFunc
ENDM
.Ed
.Pp
The example above defines
.Ql my_macro
as a new macro.
String constants are not expanded within the name of the macro.
.Pp
Macros can't be exported or imported.
.Pp
Nesting macro definitions is not possible, so this won't work:
.Bd -literal -offset indent
MACRO outer
    MACRO inner
        PRINTLN "Hello!"
    ENDM ; this actually ends the 'outer' macro...
ENDM     ; ...and then this is a syntax error!
.Ed
.Pp
But you can work around this limitation using
.Ic EQUS ,
so this will work:
.Bd -literal -offset indent
MACRO outer
    DEF definition EQUS "MACRO inner\enPRINTLN \e"Hello!\e"\enENDM"
    definition
    PURGE definition
ENDM
.Ed
.Pp
More about how to define and invoke macros is described in
.Sx THE MACRO LANGUAGE
below.
.Ss Exporting and importing symbols
Importing and exporting of symbols is a feature that is very useful when your project spans many source files and, for example, you need to jump to a routine defined in another file.
.Pp
Exporting of symbols has to be done manually, importing is done automatically if
.Nm
finds a symbol it does not know about.
.Pp
The following will cause
.Ar symbol1 , symbol2
and so on to be accessible to other files during the link process:
.Dl Ic EXPORT Ar symbol1 Bq , Ar symbol2 , No ...
.Pp
For example, if you have the following three files:
.Pp
.Ql a.asm :
.Bd -literal -offset indent -compact
SECTION "a", WRAM0
LabelA:
.Ed
.Pp
.Ql b.asm :
.Bd -literal -offset indent -compact
SECTION "b", WRAM0
ExportedLabelB1::
ExportedLabelB2:
    EXPORT ExportedLabelB2
.Ed
.Pp
.Ql c.asm :
.Bd -literal -offset indent -compact
SECTION "C", ROM0[0]
    dw LabelA
    dw ExportedLabelB1
    dw ExportedLabelB2
.Ed
.Pp
Then
.Ql c.asm
can use
.Ql ExportedLabelB1
and
.Ql ExportedLabelB2 ,
but not
.Ql LabelA ,
so linking them together will fail:
.Bd -literal -offset indent
$ rgbasm -o a.o a.asm
$ rgbasm -o b.o b.asm
$ rgbasm -o c.o c.asm
$ rgblink a.o b.o c.o
error: c.asm(2): Unknown symbol "LabelA"
Linking failed with 1 error
.Ed
.Pp
Note also that only exported symbols will appear in symbol and map files produced by
.Xr rgblink 1 .
.Ss Purging symbols
.Ic PURGE
allows you to completely remove a symbol from the symbol table, as if it had never been defined.
Be
.Em very
careful when purging symbols, especially labels, because it could result in unpredictable errors if something depends on the missing symbol (for example, expressions the linker needs to calculate).
.Bd -literal -offset indent
DEF Kamikaze EQUS "I don't want to live anymore"
AOLer: DB "Me too lol"
PURGE Kamikaze, AOLer
ASSERT !DEF(Kamikaze) && !DEF(AOLer)
.Ed
.Pp
String constants are not expanded within the symbol names.
.Ss Predeclared symbols
The following symbols are defined by the assembler:
.Bl -column -offset indent "__ISO_8601_LOCAL__" "EQUS"
.It Sy Name Ta Sy Type Ta Sy Contents
.It Dv @ Ta Ic EQU Ta PC value (essentially, the current memory address)
.It Dv . Ta Ic EQUS Ta The current global label scope
.It Dv .. Ta Ic EQUS Ta The current local label scope
.It Dv _RS Ta Ic = Ta _RS Counter
.It Dv _NARG Ta Ic EQU Ta Number of arguments passed to macro, updated by Ic SHIFT
.It Dv __DATE__ Ta Ic EQUS Ta Today's date
.It Dv __TIME__ Ta Ic EQUS Ta The current time
.It Dv __ISO_8601_LOCAL__ Ta Ic EQUS Ta ISO 8601 timestamp (local)
.It Dv __ISO_8601_UTC__ Ta Ic EQUS Ta ISO 8601 timestamp (UTC)
.It Dv __UTC_YEAR__ Ta Ic EQU Ta Today's year
.It Dv __UTC_MONTH__ Ta Ic EQU Ta Today's month number, 1\[en]12
.It Dv __UTC_DAY__ Ta Ic EQU Ta Today's day of the month, 1\[en]31
.It Dv __UTC_HOUR__ Ta Ic EQU Ta Current hour, 0\[en]23
.It Dv __UTC_MINUTE__ Ta Ic EQU Ta Current minute, 0\[en]59
.It Dv __UTC_SECOND__ Ta Ic EQU Ta Current second, 0\[en]59
.It Dv __RGBDS_MAJOR__ Ta Ic EQU Ta Major version number of RGBDS
.It Dv __RGBDS_MINOR__ Ta Ic EQU Ta Minor version number of RGBDS
.It Dv __RGBDS_PATCH__ Ta Ic EQU Ta Patch version number of RGBDS
.It Dv __RGBDS_RC__ Ta Ic EQU Ta Release candidate ID of RGBDS, not defined for final releases
.It Dv __RGBDS_VERSION__ Ta Ic EQUS Ta Version of RGBDS, as printed by Ql rgbasm --version
.El
.Pp
The current time values will be taken from the
.Dv SOURCE_DATE_EPOCH
environment variable if that is defined as a UNIX timestamp.
Refer to the spec at
.Lk https://reproducible-builds.org/docs/source-date-epoch/ reproducible-builds.org .
.Sh DEFINING DATA
.Ss Defining constant data in ROM
.Ic DB
defines a list of bytes that will be stored in the final image.
Ideal for tables and text.
.Bd -literal -offset indent
DB 1,2,3,4,"This is a string"
.Ed
.Pp
Alternatively, you can use
.Ic DW
to store a list of words (16-bit) or
.Ic DL
to store a list of double-words/longs (32-bit).
Both of these write their data in little-endian byte order; for example,
.Ql dw $CAFE
is equivalent to
.Ql db $FE, $CA
and not
.Ql db $CA, $FE .
.Pp
Strings are handled a little specially: they first undergo charmap conversion (see
.Sx Character maps ) ,
then each resulting character is output individually.
For example, under the default charmap, the following two lines are identical:
.Bd -literal -offset indent
DW "Hello!"
DW "H", "e", "l", "l", "o", "!"
.Ed
.Pp
If you do not want this special handling, enclose the string in parentheses.
.Pp
.Ic DS
can also be used to fill a region of memory with some repeated values.
For example:
.Bd -literal -offset indent
; outputs 3 bytes: $AA, $AA, $AA
DS 3, $AA
; outputs 7 bytes: $BB, $CC, $BB, $CC, $BB, $CC, $BB
DS 7, $BB, $CC
.Ed
.Pp
You can also use
.Ic DB , DW
and
.Ic DL
without arguments.
This works exactly like
.Ic DS 1 , DS 2
and
.Ic DS 4
respectively.
Consequently, no-argument
.Ic DB , DW
and
.Ic DL
can be used in a
.Ic WRAM0
/
.Ic WRAMX
/
.Ic HRAM
/
.Ic VRAM
/
.Ic SRAM
section.
.Ss Including binary data files
You probably have some graphics, level data, etc. you'd like to include.
Use
.Ic INCBIN
to include a raw binary file as it is.
If the file isn't found in the current directory, the include-path list passed to
.Xr rgbasm 1
(see the
.Fl I
option) on the command line will be searched.
.Bd -literal -offset indent
INCBIN "titlepic.bin"
INCBIN "sprites/hero.bin"
.Ed
.Pp
You can also include only part of a file with
.Ic INCBIN .
The example below includes 256 bytes from data.bin, starting from byte 78.
.Bd -literal -offset indent
INCBIN "data.bin", 78, 256
.Ed
.Pp
The length argument is optional.
If only the start position is specified, the bytes from the start position until the end of the file will be included.
.Ss Statically allocating space in RAM
.Ic DS
statically allocates a number of empty bytes.
This is the preferred method of allocating space in a RAM section.
You can also use
.Ic DB , DW
and
.Ic DL
without any arguments instead (see
.Sx Defining constant data in ROM
below).
.Bd -literal -offset indent
DS 42 ;\ Allocates 42 bytes
.Ed
.Pp
Empty space in RAM sections will not be initialized.
In ROM sections, it will be filled with the value passed to the
.Fl p
command-line option, except when using overlays with
.Fl O .
.Pp
Instead of an exact number of bytes, you can specify
.Ic ALIGN Ns Bq Ar align , offset
to allocate however many bytes are required to align the subsequent data.
Thus,
.Sq Ic DS ALIGN Ns Bo Ar align , offset Bc , No ...
is equivalent to
.Sq Ic DS Ar n , No ...
followed by
.Sq Ic ALIGN Ns Bq Ar align , offset ,
where
.Ar n
is the minimum value needed to satisfy the
.Ic ALIGN
constraint (see
.Sx Requesting alignment
below).
Note that
.Ic ALIGN Ns Bq Ar align
is a shorthand for
.Ic ALIGN Ns Bq Ar align , No 0 .
.Ss Allocating overlapping spaces in RAM
Unions allow multiple static memory allocations to overlap, like unions in C.
This does not increase the amount of memory available, but allows re-using the same memory region for different purposes.
.Pp
A union starts with a
.Ic UNION
keyword, and ends at the corresponding
.Ic ENDU
keyword.
.Ic NEXTU
separates each block of allocations, and you may use it as many times within a union as necessary.
.Bd -literal -offset indent
; Let's say PC == $C0DE here
UNION
    ; Here, PC == $C0DE
    wName:: ds 10
    ; Now, PC == $C0E8
    wNickname:: ds 10
    ; PC == $C0F2
NEXTU
    ; PC is back to $C0DE
    wHealth:: dw
    ; PC == $C0E0
    wLives:: db
    ; PC == $C0E1
    ds 7
    ; PC == $C0E8
    wBonus:: db
    ; PC == $C0E9
NEXTU
    ; PC is back to $C0DE again
    wVideoBuffer: ds 16
    ; PC == $C0EE
ENDU
; Afterward, PC == $C0F2
.Ed
.Pp
In the example above,
.Sq wName , wHealth ,
and
.Sq wVideoBuffer
all have the same value; so do
.Sq wNickname
and
.Sq wBonus .
Thus, keep in mind that
.Ql ld [wHealth], a
assembles to the exact same thing as
.Ql ld [wName], a .
.Pp
This whole union's total size is 20 bytes, the size of the largest block (the first one, containing
.Sq wName
and
.Sq wNickname ) .
.Pp
Unions may be nested, with each inner union's size being determined as above, and affecting its outer union like any other allocation.
.Pp
Unions may be used in any section, but they may only contain space-allocating directives like
.Ic DS
(see
.Sx Statically allocating space in RAM ) .
.Ss Requesting alignment
While
.Ic ALIGN
as presented in
.Sx SECTIONS
is often useful as-is, sometimes you instead want a particular piece of data (or code) in the middle of the section to be aligned.
This is made easier through the use of mid-section
.Ic ALIGN Ar align , offset .
It will retroactively alter the section's attributes to ensure that the location the
.Ic ALIGN
directive is at, has its
.Ar align
lower bits equal to
.Ar offset .
.Pp
If the constraint cannot be met (for example because the section is fixed at an incompatible address), an error is produced.
Note that
.Ic ALIGN Ar align
is a shorthand for
.Ic ALIGN Ar align , No 0 .
.Pp
There may be times when you don't just want to specify an alignment constraint at the current location, but also skip ahead until the constraint can be satisfied.
In that case, you can use
.Ic DS ALIGN Ns Bq Ar align , offset
to allocate however many bytes are required to align the subsequent data.
.Pp
If the constraint cannot be met by skipping any amount of space, an error is produced.
Note that
.Ic ALIGN Ns Bq Ar align
is a shorthand for
.Ic ALIGN Ns Bq Ar align , No 0 .
.Sh THE MACRO LANGUAGE
.Ss Invoking macros
A macro is invoked by using its name at the beginning of a line, like a directive, followed by any comma-separated arguments.
.Bd -literal -offset indent
    add a, b
    ld sp, hl
    my_macro      ;\ This will be expanded
    sub a, 87
    my_macro 42   ;\ So will this
    ret c
    my_macro 1, 2 ;\ And this
.Ed
.Pp
After
.Nm
has read the macro invocation line, it will expand the body of the macro (the lines between
.Ic MACRO
and
.Ic ENDM )
in its place.
.Pp
.Sy Important note :
When a macro body is expanded, its expansion may contain another macro invocation, which will be expanded as well, and may be recursive.
If this creates an infinite loop,
.Nm
will error out once a certain depth is
reached (see the
.Fl r
command-line option in
.Xr rgbasm 1 ) .
The same problem can occur if the expansion of a macro then expands a string constant, which itself expands.
.Pp
It's possible to pass arguments to macros as well!
.Bd -literal -offset indent
MACRO lb
    ld \e1, (\e2) << 8 | (\e3)
ENDM
    lb hl, 20, 18        ; Expands to "ld hl, ((20) << 8) | (18)"
    lb de, 3 + 1, NUM**2 ; Expands to "ld de, ((3 + 1) << 8) | (NUM**2)"
.Ed
.Pp
You expand the arguments inside the macro body by using the escape sequences
.Ic \e1
through
.Ic \e9 , \e1
being the first argument,
.Ic \e2
being the second, and so on. Since there are only nine digits, you can only use the first nine macro arguments that way.
To use the rest, you put the argument number in angle brackets, like
.Ic \e<10> .
.Pp
This bracketed syntax supports decimal numbers and numeric symbols, where negative values count from the last argument.
For example,
.Ql \e<_NARG>
or
.Ql \e<-1>
will get the last argument.
.Pp
Other macro arguments and symbol interpolations will also be expanded inside the angle brackets.
For example, if
.Ql \e1
is
.Ql 13 ,
then
.Ql \e<\e1>
inside the macro body will expand to
.Ql \e<13> .
Or if
.Ql DEF v10 = 42
and
.Ql DEF x = 10 ,
then
.Ql \e<v{d:x}>
will expand to
.Ql \e<42> .
.Pp
Macro arguments are passed as string constants, although there's no need to enclose them in quotes.
Thus, arguments are not evaluated as expressions, but instead are expanded directly inside the macro body.
This means that they support all the escape sequences of strings (see
.Sx String expressions
above), as well as some of their own:
.Bl -column -offset indent "Sequence"
.It Sy Sequence Ta Sy Meaning
.It Ql \e, Ta Comma Pq does not terminate the argument
.It Ql \e( Ta Open parenthesis Pq does not start enclosing argument contents
.It Ql \e) Ta Close parenthesis Pq does not end enclosing argument contents
.El
.Pp
Line continuations work as usual inside macros or lists of macro arguments.
However, some characters need to be escaped, as in the following example:
.Bd -literal -offset indent
MACRO PrintMacro1
    PRINTLN STRCAT(\e1)
ENDM
    PrintMacro1 "Hello "\e, \e
                "world"
MACRO PrintMacro2
    PRINT \e1
ENDM
    PrintMacro2 STRCAT("Hello ", \e
                       "world\en")
.Ed
.Pp
The comma in
.Ql PrintMacro1
needs to be escaped to prevent it from starting another macro argument.
The comma in
.Ql PrintMacro2
does not need escaping because it is inside parentheses, similar to macro arguments in the C programming language.
The backslash in
.Ql \en
also does not need escaping because quoted string literals work as usual inside macro arguments.
.Pp
Since macro arguments are expanded directly, it's often a good idea to put parentheses around them if they're meant as part of a numeric expression.
For instance, consider the following:
.Bd -literal -offset indent
MACRO print_double
    PRINTLN \e1 * 3
ENDM
    print_double 1 + 2
.Ed
.Pp
The body will expand to
.Ql PRINTLN 1 + 2 * 3 ,
which will print 7 and not 9 as you might have expected.
.Pp
The
.Ic SHIFT
directive is only available inside macro bodies.
It shifts the argument numbers by one to the left, so what was
.Ic \e2
is now
.Ic \e1 ,
what was
.Ic \e3
is now
.Ic \e2 ,
and so forth.
(What was
.Ic \e1
is no longer accessible, so
.Dv _NARG
is decreased by 1.)
.Pp
.Ic SHIFT
can also take an integer parameter to shift that many times instead of once.
A negative parameter will shift the arguments to the right, which can regain access to previously shifted ones.
.Pp
.Ic SHIFT
is especially useful in
.Ic REPT
loops to iterate over different arguments, evaluating the same loop body each time.
.Pp
There are some escape sequences which are only valid inside the body of a macro:
.Bl -column -offset indent "Sequence"
.It Sy Sequence Ta Sy Meaning
.It So \e1 Sc \[en] So \e9 Sc Ta The 1st\[en]9th macro argument
.It Ql \e<...> Ta Further macro arguments
.It Ql \e# Ta All Dv _NARG No macro arguments, separated by commas
.It Ql \e@ Ta Unique symbol name affix Pq see below
.El
.Pp
The
.Ic \e@
escape sequence is often useful in macros which define symbols.
Suppose your macro expands to a loop of assembly code:
.Bd -literal -offset indent
MACRO loop_c_times
    xor a, a
\&.loop
    ld [hl+], a
    dec c
    jr nz, .loop
ENDM
.Ed
.Pp
If you use this macro more than once in the same label scope, it will define
.Ql \&.loop
twice, which is an error.
To work around this problem, you can use
.Ic \e@
as a label suffix:
.Bd -literal -offset indent
MACRO loop_c_times_fixed
    xor a, a
\&.loop\e@
    ld [hl+], a
    dec c
    jr nz, .loop\e@
ENDM
.Ed
.Pp
This will expand to a different value in each invocation, similar to
.Ic gensym
in the Lisp programming language.
.Pp
.Ic \e@
also works in
.Ic REPT
blocks, expanding to a different value in each iteration.
.Ss Automatically repeating blocks of code
Suppose you want to unroll a time-consuming loop without copy-pasting it.
.Ic REPT
is here for that purpose.
Everything between
.Ic REPT
and the matching
.Ic ENDR
will be repeated a number of times just as if you had done a copy/paste operation yourself.
The following example will assemble
.Ql add a, c
four times:
.Bd -literal -offset indent
REPT 4
    add a, c
ENDR
.Ed
.Pp
You can also use
.Ic REPT
to generate tables on the fly:
.Bd -literal -offset indent
; Generate a table of square values from 0**2 = 0 to 100**2 = 10000
DEF x = 0
REPT 101
    dw x * x
    DEF x += 1
ENDR
.Ed
.Pp
As in macros, you can also use the escape sequence
.Ic \e@ .
.Ic REPT
blocks can be nested.
.Pp
A common pattern is to repeat a block for each value in some range.
.Ic FOR
is simpler than
.Ic REPT
for that purpose.
Everything between
.Ic FOR
and the matching
.Ic ENDR
will be repeated for each value of a given symbol.
String constants are not expanded within the symbol name.
For example, this code will produce a table of squared values from 0 to 255:
.Bd -literal -offset indent
FOR N, 256
    dw N * N
ENDR
.Ed
.Pp
It acts just as if you had done:
.Bd -literal -offset indent
DEF N = 0
    dw N * N
DEF N = 1
    dw N * N
DEF N = 2
    dw N * N
; ...
DEF N = 255
    dw N * N
DEF N = 256
.Ed
.Pp
You can customize the range of
.Ic FOR
values, similarly to the
.Ql range
function in the Python programming language:
.Bl -column "FOR V, start, stop, step"
.It Sy Code Ta Sy Range
.It Ic FOR Ar V , stop Ta Ar V No increments from 0 to Ar stop
.It Ic FOR Ar V , start , stop Ta Ar V No increments from Ar start No to Ar stop
.It Ic FOR Ar V , start , stop , step Ta Ar V No goes from Ar start No to Ar stop No by Ar step
.El
.Pp
The
.Ic FOR
value will be updated by
.Ar step
until it reaches or exceeds
.Ar stop ,
i.e. it covers the half-open range from
.Ar start
(inclusive) to
.Ar stop
(exclusive).
The variable
.Ar V
will be assigned this value at the beginning of each new iteration; any changes made to it within the
.Ic FOR
loop's body will be overwritten.
So the symbol
.Ar V
need not be already defined before any iterations of the
.Ic FOR
loop, but it must be a variable
.Pq Sx Variables
if so.
For example:
.Bd -literal -offset indent
FOR V, 4, 25, 5
    PRINT "{d:V} "
    DEF V *= 2
ENDR
    PRINTLN "done {d:V}"
.Ed
.Pp
This will print:
.Bd -literal -offset indent
4 9 14 19 24 done 29
.Ed
.Pp
Just like with
.Ic REPT
blocks, you can use the escape sequence
.Ic \e@
inside of
.Ic FOR
blocks, and they can be nested.
.Pp
You can stop a repeating block with the
.Ic BREAK
command.
A
.Ic BREAK
inside of a
.Ic REPT
or
.Ic FOR
block will interrupt the current iteration and not repeat any more.
It will continue running code after the block's
.Ic ENDR .
For example:
.Bd -literal -offset indent
FOR V, 1, 100
    PRINT "{d:V}"
    IF V == 5
        PRINT " stop! "
        BREAK
    ENDC
    PRINT ", "
ENDR
    PRINTLN "done {d:V}"
.Ed
.Pp
This will print:
.Bd -literal -offset indent
1, 2, 3, 4, 5 stop! done 5
.Ed
.Ss Conditionally assembling blocks of code
The four commands
.Ic IF , ELIF , ELSE ,
and
.Ic ENDC
let you have
.Nm
skip over parts of your code depending on a condition.
This is a powerful feature commonly used in macros.
.Bd -literal -offset indent
IF NUM < 0
  PRINTLN "NUM < 0"
ELIF NUM == 0
  PRINTLN "NUM == 0"
ELSE
  PRINTLN "NUM > 0"
ENDC
.Ed
.Pp
The
.Ic ELIF
(standing for "else if") and
.Ic ELSE
blocks are optional.
.Ic IF
/
.Ic ELIF
/
.Ic ELSE
/
.Ic ENDC
blocks can be nested.
.Pp
Note that if an
.Ic ELSE
block is found before an
.Ic ELIF
block, the
.Ic ELIF
block will be ignored.
All
.Ic ELIF
blocks must go before the
.Ic ELSE
block.
Also, if there is more than one
.Ic ELSE
block, all of them but the first one are ignored.
.Ss Including other source files
Use
.Ic INCLUDE
to process another assembler file and then return to the current file when done.
If the file isn't found in the current directory, the include path list (see the
.Fl I
option in
.Xr rgbasm 1 )
will be searched.
You may nest
.Ic INCLUDE
calls infinitely (or until you run out of memory, whichever comes first).
.Bd -literal -offset indent
    INCLUDE "irq.inc"
.Ed
.Pp
You may also implicitly
.Ic INCLUDE
a file before the source file with the
.Fl P
option of
.Xr rgbasm 1 .
.Ss Printing things during assembly
The
.Ic PRINT
and
.Ic PRINTLN
commands print text and values to the standard output.
Useful for debugging macros, or wherever you may feel the need to tell yourself some important information.
.Bd -literal -offset indent
PRINT "Hello world!\en"
PRINTLN "Hello world!"
PRINT _NARG, " arguments\en"
PRINTLN "sum: ", 2+3, " product: ", 2*3
PRINTLN STRFMT("E = %f", 2.718)
.Ed
.Bl -inset
.It Ic PRINT
prints out each of its comma-separated arguments.
Numbers are printed as unsigned uppercase hexadecimal with a leading
.Sq $ .
For different formats, use
.Ic STRFMT .
.It Ic PRINTLN
prints out each of its comma-separated arguments, if any, followed by a newline
.Pq Ql \en .
.El
.Ss Aborting the assembly process
.Ic FAIL
and
.Ic WARN
can be used to print errors and warnings respectively during the assembly process.
This is especially useful for macros that get an invalid argument.
.Ic FAIL
and
.Ic WARN
take a string as the only argument and they will print this string out as a normal error with a line number.
.Pp
.Ic FAIL
stops assembling immediately while
.Ic WARN
shows the message but continues afterwards.
.Pp
If you need to ensure some assumption is correct when compiling, you can use
.Ic ASSERT
and
.Ic STATIC_ASSERT .
Syntax examples are given below:
.Bd -literal -offset indent
Function:
    xor a
ASSERT LOW(MyByte) == 0
    ld h, HIGH(MyByte)
    ld l, a
    ld a, [hli]
; You can also indent this!
    ASSERT BANK(OtherFunction) == BANK(Function)
    call OtherFunction
; Lowercase also works
    ld hl, FirstByte
    ld a, [hli]
assert FirstByte + 1 == SecondByte
    ld b, [hl]
    ret
\&.end
    ; If you specify one, a message will be printed
    STATIC_ASSERT .end - Function < 256, "Function is too large!"
.Ed
.Pp
First, the difference between
.Ic ASSERT
and
.Ic STATIC_ASSERT
is that the former is evaluated by RGBASM if it can, otherwise by RGBLINK; but the latter is only ever evaluated by RGBASM.
If RGBASM cannot compute the value of the argument to
.Ic STATIC_ASSERT ,
it will produce an error.
.Pp
Second, as shown above, a string can be optionally added at the end, to give insight into what the assertion is checking.
.Pp
Finally, you can add one of
.Ic WARN , FAIL
or
.Ic FATAL
as the first optional argument to either
.Ic ASSERT
or
.Ic STATIC_ASSERT .
If the assertion fails,
.Ic WARN
will cause a simple warning (controlled by
.Xr rgbasm 1
flag
.Fl Wassert )
to be emitted;
.Ic FAIL
(the default) will cause a non-fatal error; and
.Ic FATAL
immediately aborts.
.Sh MISCELLANEOUS
.Ss Changing options while assembling
.Ic OPT
can be used to change some of the options during assembling from within the source, instead of defining them on the command-line.
.Pq See Xr rgbasm 1 .
.Pp
.Ic OPT
takes a comma-separated list of options as its argument:
.Bd -literal -offset indent
PUSHO
    OPT g.oOX, Wdiv       ; acts like command-line -g.oOX -Wdiv
    DW `..ooOOXX          ; uses the graphics constant characters from OPT g
    PRINTLN $80000000/-1  ; prints a warning about division
POPO
    DW `00112233          ; uses the default graphics constant characters
    PRINTLN $80000000/-1  ; no warning by default
.Ed
.Pp
.Ic OPT
can modify the options
.Cm b , g , p , Q , r ,
and
.Cm W .
.Pp
.Ic POPO
and
.Ic PUSHO
provide the interface to the option stack.
.Ic PUSHO
will push the current set of options on the option stack.
.Ic POPO
can then later be used to restore them.
Useful if you want to change some options in an include file and you don't want to destroy the options set by the program that included your file.
The stack's number of entries is limited only by the amount of memory in your machine.
.Pp
.Ic PUSHO
can also take a comma-separated list of options, to push the current set and apply the argument set at the same time:
.Bd -literal -offset indent
PUSHO b.X, g.oOX
    DB %..XXXX..
    DW `..ooOOXX
POPO
.Ed
.Sh SEE ALSO
.Xr rgbasm 1 ,
.Xr rgblink 1 ,
.Xr rgblink 5 ,
.Xr rgbfix 1 ,
.Xr rgbgfx 1 ,
.Xr gbz80 7 ,
.Xr rgbasm-old 5 ,
.Xr rgbds 5 ,
.Xr rgbds 7
.Sh HISTORY
.Xr rgbasm 1
was originally written by
.An Carsten S\(/orensen
as part of the ASMotor package, and was later repackaged in RGBDS by
.An Justin Lloyd .
It is now maintained by a number of contributors at
.Lk https://github.com/gbdev/rgbds .