gas/doc/c-z80.texi - binutils-gdb - Git at Google

 @c Copyright (C) 2011-2021 Free Software Foundation, Inc.
 @c This is part of the GAS manual.
 @c For copying conditions, see the file as.texinfo.

 @ifset GENERIC
 @page
 @node Z80-Dependent
 @chapter Z80 Dependent Features
 @end ifset


 @ifclear GENERIC
 @node Machine Dependencies
 @chapter Z80 Dependent Features
 @end ifclear

 @cindex Z80 support
 @menu
 * Z80 Options::              Options
 * Z80 Syntax::               Syntax
 * Z80 Floating Point::       Floating Point
 * Z80 Directives::           Z80 Machine Directives
 * Z80 Opcodes::              Opcodes
 @end menu

 @node Z80 Options
 @section Command-line Options
 @cindex Z80 options
 @cindex options for Z80
 @c man begin OPTIONS
 @table @gcctabopt

 @cindex @code{-march=} command-line option, Z80
 @item -march=@var{CPU}[-@var{EXT}@dots{}][+@var{EXT}@dots{}]
 This option specifies the target processor. The assembler will issue
 an error message if an attempt is made to assemble an instruction which
 will not execute on the target processor. The following processor names
 are recognized:
 @code{z80},
 @code{z180},
 @code{ez80},
 @code{gbz80},
 @code{z80n},
 @code{r800}.
 In addition to the basic instruction set, the assembler can be told to
 accept some extention mnemonics. For example,
 @code{-march=z180+sli+infc} extends @var{z180} with @var{SLI} instructions and
 @var{IN F,(C)}. The following extentions are currently supported:
 @code{full} (all known instructions),
 @code{adl} (ADL CPU mode by default, eZ80 only),
 @code{sli} (instruction known as @var{SLI}, @var{SLL} or @var{SL1}),
 @code{xyhl} (instructions with halves of index registers: @var{IXL}, @var{IXH},
 @var{IYL}, @var{IYH}),
 @code{xdcb} (instructions like @var{RotOp (II+d),R} and @var{BitOp n,(II+d),R}),
 @code{infc} (instruction @var{IN F,(C)} or @var{IN (C)}),
 @code{outc0} (instruction @var{OUT (C),0}).
 Note that rather than extending a basic instruction set, the extention
 mnemonics starting with @code{-} revoke the respective functionality:
 @code{-march=z80-full+xyhl} first removes all default extentions and adds
 support for index registers halves only.

 If this option is not specified then @code{-march=z80+xyhl+infc} is assumed.

 @cindex @code{-local-prefix} command-line option, Z80
 @item  -local-prefix=@var{prefix}
 Mark all labels with specified prefix as local. But such label can be
 marked global explicitly in the code. This option do not change default
 local label prefix @code{.L}, it is just adds new one.

 @cindex @code{-colonless} command-line option, Z80
 @item  -colonless
 Accept colonless labels. All symbols at line begin are treated as labels.

 @cindex @code{-sdcc} command-line option, Z80
 @item  -sdcc
 Accept assembler code produced by SDCC.

 @cindex @code{-fp-s} command-line option, Z80
 @item -fp-s=@var{FORMAT}
 Single precision floating point numbers format. Default: ieee754 (32 bit).

 @cindex @code{-fp-d} command-line option, Z80
 @item -fp-d=@var{FORMAT}
 Double precision floating point numbers format. Default: ieee754 (64 bit).
 @end table
 @c man end

 Floating point numbers formats.
 @table @option
 @item @code{ieee754}
 Single or double precision IEEE754 compatible format.

 @item @code{half}
 Half precision IEEE754 compatible format (16 bits).

 @item @code{single}
 Single precision IEEE754 compatible format (32 bits).

 @item @code{double}
 Double precision IEEE754 compatible format (64 bits).

 @item @code{zeda32}
 32 bit floating point format from z80float library by Zeda.

 @item @code{math48}
 48 bit floating point format from Math48 package by Anders Hejlsberg.
 @end table

 @cindex Z80 Syntax
 @node Z80 Syntax
 @section Syntax
 The assembler syntax closely follows the 'Z80 family CPU User Manual' by
 Zilog.
 In expressions a single @samp{=} may be used as ``is equal to''
 comparison operator.

 Suffices can be used to indicate the radix of integer constants;
 @samp{H} or @samp{h} for hexadecimal, @samp{D} or @samp{d} for decimal,
 @samp{Q}, @samp{O}, @samp{q} or @samp{o} for octal, and @samp{B} for
 binary.

 The suffix @samp{b} denotes a backreference to local label.

 @menu
 * Z80-Chars::                Special Characters
 * Z80-Regs::                 Register Names
 * Z80-Case::                 Case Sensitivity
 * Z80-Labels::               Labels
 @end menu

 @node Z80-Chars
 @subsection Special Characters

 @cindex line comment character, Z80
 @cindex Z80 line comment character
 The semicolon @samp{;} is the line comment character;

 If a @samp{#} appears as the first character of a line then the whole
 line is treated as a comment, but in this case the line could also be
 a logical line number directive (@pxref{Comments}) or a preprocessor
 control command (@pxref{Preprocessing}).

 @cindex line separator, Z80
 @cindex statement separator, Z80
 @cindex Z80 line separator
 The Z80 assembler does not support a line separator character.

 @cindex location counter, Z80
 @cindex hexadecimal prefix, Z80
 @cindex Z80 $
 The dollar sign @samp{$} can be used as a prefix for hexadecimal numbers
 and as a symbol denoting the current location counter.

 @cindex character escapes, Z80
 @cindex Z80, \
 A backslash @samp{\} is an ordinary character for the Z80 assembler.

 @cindex character constant, Z80
 @cindex single quote, Z80
 @cindex Z80 '
 The single quote @samp{'} must be followed by a closing quote. If there
 is one character in between, it is a character constant, otherwise it is
 a string constant.

 @node Z80-Regs
 @subsection Register Names
 @cindex Z80 registers
 @cindex register names, Z80

 The registers are referred to with the letters assigned to them by
 Zilog. In addition @command{@value{AS}} recognizes @samp{ixl} and
 @samp{ixh} as the least and most significant octet in @samp{ix}, and
 similarly @samp{iyl} and  @samp{iyh} as parts of @samp{iy}.

 @c The @samp{'} in @samp{ex af,af'} may be omitted.

 @node Z80-Case
 @subsection Case Sensitivity
 @cindex Z80, case sensitivity
 @cindex case sensitivity, Z80

 Upper and lower case are equivalent in register names, opcodes,
 condition codes  and assembler directives.
 The case of letters is significant in labels and symbol names. The case
 is also important to distinguish the suffix @samp{b} for a backward reference
 to a local label from the suffix @samp{B} for a number in binary notation.

 @node Z80-Labels
 @subsection Labels

 @cindex labels, Z80
 @cindex Z80 labels
 Labels started by @code{.L} acts as local labels. You may specify custom local
 label prefix by @code{-local-prefix} command-line option.
 Dollar, forward and backward local labels are supported. By default, all labels
 are followed by colon.
 Legacy code with colonless labels can be built with @code{-colonless}
 command-line option specified. In this case all tokens at line begin are treated
 as labels.

 @node Z80 Floating Point
 @section Floating Point
 @cindex floating point, Z80
 @cindex Z80 floating point
 Floating-point numbers of following types are supported:

 @table @option
 @item @code{ieee754}
 Supported half, single and double precision IEEE754 compatible numbers.

 @item @code{zeda32}
 32 bit floating point numbers from z80float library by Zeda.

 @item @code{math48}
 48 bit floating point numbers from Math48 package by Anders Hejlsberg.
 @end table

 @node Z80 Directives
 @section Z80 Assembler Directives
 @cindex Z80-only directives

 @command{@value{AS}} for the Z80 supports some additional directives for
 compatibility with other assemblers.

 These are the additional directives in @code{@value{AS}} for the Z80:

 @table @code
 @item @code{.assume ADL = @var{expression}}
 @cindex @code{.assume} directive, Z80
 Set ADL status for eZ80. Non-zero value enable compilation in ADL mode else
 used Z80 mode. ADL and Z80 mode produces incompatible object code. Mixing
 both of them within one binary may lead problems with disassembler.

 @item @code{db @var{expression}|@var{string}[,@var{expression}|@var{string}...]}
 @cindex @code{db} directive, Z80
 @itemx @code{defb @var{expression}|@var{string}[,@var{expression}|@var{string}...]}
 @cindex @code{defb} directive, Z80
 @itemx @code{defm @var{string}[,@var{string}...]}
 @cindex @code{defm} directive, Z80
 For each @var{string} the characters are copied to the object file, for
 each other @var{expression} the value is stored in one byte.
 A warning is issued in case of an overflow.
 Backslash symbol in the strings is generic symbol, it cannot be used as
 escape character.  @xref{Ascii,,@code{.ascii}}.

 @item @code{dw @var{expression}[,@var{expression}...]}
 @cindex @code{dw} directive, Z80
 @itemx @code{defw @var{expression}[,@var{expression}...]}
 @cindex @code{defw} directive, Z80
 For each @var{expression} the value is stored in two bytes, ignoring
 overflow.

 @item @code{d24 @var{expression}[,@var{expression}...]}
 @cindex @code{d24} directive, Z80
 @itemx @code{def24 @var{expression}[,@var{expression}...]}
 @cindex @code{def24} directive, Z80
 For each @var{expression} the value is stored in three bytes, ignoring
 overflow.

 @item @code{d32 @var{expression}[,@var{expression}...]}
 @cindex @code{d32} directive, Z80
 @itemx @code{def32 @var{expression}[,@var{expression}...]}
 @cindex @code{def32} directive, Z80
 For each @var{expression} the value is stored in four bytes, ignoring
 overflow.

 @item @code{ds @var{count}[, @var{value}]}
 @cindex @code{ds} directive, Z80
 @itemx @code{defs @var{count}[, @var{value}]}
 @cindex @code{defs} directive, Z80
 @c Synonyms for @code{ds.b},
 @c which should have been described elsewhere
 Fill @var{count} bytes in the object file with @var{value}, if
 @var{value} is omitted it defaults to zero.

 @item @code{@var{symbol} defl @var{expression}}
 @cindex @code{defl} directive, Z80
 The @code{defl} directive is like @code{.set} but with different
 syntax.  @xref{Set,,@code{.set}}.
 It set the value of @var{symbol} to @var{expression}. Symbols defined
 with @code{defl} are not protected from redefinition.

 @item @code{@var{symbol} equ @var{expression}}
 @cindex @code{equ} directive, Z80
 The @code{equ} directive is like @code{.equiv} but with different
 syntax.  @xref{Equiv, ,@code{.equiv}}.
 It set the value of @var{symbol} to @var{expression}. It is an error
 if @var{symbol} is already defined. Symbols defined with @code{equ}
 are not protected from redefinition.

 @item @code{psect @var{name}}
 @cindex @code{psect} directive, Z80
 A synonym for @code{.section}, no second argument should be given.
 @xref{Section,,@code{.section}}.

 @item @code{xdef @var{symbol}}
 @cindex @code{xdef} directive, Z80
 A synonym for @code{.global}, make @var{symbol} is visible to linker.
 @xref{Global,,@code{.global}}.

 @item @code{xref @var{name}}
 @cindex @code{xref} directive, Z80
 A synonym for @code{.extern} (@ref{Extern,,@code{.extern}}).
 @ignore

 The following attributes will possibly be recognized in the future
 @table @code
 @item abs
 The section is to be absolute. @code{@value{AS}} will issue an error
 message because it can not produce an absolute section.
 @item global
 The section is to be concatenated with other sections of the same name
 by the linker, this is the default.
 @item local
 The section is not global. @code{@value{AS}} will issue a warning if
 object file format is not soff.
 @item ovrld
 The section is to be overlapped with other sections of the same name by
 the linker. @code{@value{AS}} will issue an error message
 because it can not mark a section as such.
 @item pure
 The section is marked as read only.
 @end table
 @end ignore

 @end table

 @node Z80 Opcodes
 @section Opcodes
 In line with common practice, Z80 mnemonics are used for the Z80,
 Z80N, Z180, eZ80, Ascii R800 and the GameBoy Z80.

 In many instructions it is possible to use one of the half index
 registers (@samp{ixl},@samp{ixh},@samp{iyl},@samp{iyh}) in stead of an
 8-bit general purpose register. This yields instructions that are
 documented on the eZ80 and the R800, undocumented on the Z80 and
 unsupported on the Z180.
 Similarly @code{in f,(c)} is documented on the R800, undocumented on
 the Z80 and unsupported on the Z180 and the eZ80.

 The assembler also supports the following undocumented Z80-instructions,
 that have not been adopted in any other instruction set:
 @table @code
 @item out (c),0
 Sends zero to the port pointed to by register @code{C}.

 @item sli @var{m}
 Equivalent to @code{@var{m} = (@var{m}<<1)+1}, the operand @var{m} can
 be any operand that is valid for @samp{sla}. One can use @samp{sll} as a
 synonym for @samp{sli}.

 @item @var{op} (ix+@var{d}), @var{r}
 This is equivalent to

 @example
 ld @var{r}, (ix+@var{d})
 @var{op} @var{r}
 ld (ix+@var{d}), @var{r}
 @end example

 The operation @samp{@var{op}} may be any of @samp{res @var{b},},
 @samp{set @var{b},}, @samp{rl}, @samp{rlc}, @samp{rr}, @samp{rrc},
 @samp{sla}, @samp{sli}, @samp{sra} and @samp{srl}, and the register
 @samp{@var{r}} may be any of @samp{a}, @samp{b}, @samp{c}, @samp{d},
 @samp{e}, @samp{h} and @samp{l}.

 @item @var{op} (iy+@var{d}), @var{r}
 As above, but with @samp{iy} instead of @samp{ix}.
 @end table

 The web site at @uref{http://www.z80.info} is a good starting place to
 find more information on programming the Z80.

 You may enable or disable any of these instructions for any target CPU
 even this instruction is not supported by any real CPU of this type.
 Useful for custom CPU cores.
	@c Copyright (C) 2011-2021 Free Software Foundation, Inc.
	@c This is part of the GAS manual.
	@c For copying conditions, see the file as.texinfo.

	@ifset GENERIC
	@page
	@node Z80-Dependent
	@chapter Z80 Dependent Features
	@end ifset


	@ifclear GENERIC
	@node Machine Dependencies
	@chapter Z80 Dependent Features
	@end ifclear

	@cindex Z80 support
	@menu
	* Z80 Options:: Options
	* Z80 Syntax:: Syntax
	* Z80 Floating Point:: Floating Point
	* Z80 Directives:: Z80 Machine Directives
	* Z80 Opcodes:: Opcodes
	@end menu

	@node Z80 Options
	@section Command-line Options
	@cindex Z80 options
	@cindex options for Z80
	@c man begin OPTIONS
	@table @gcctabopt

	@cindex @code{-march=} command-line option, Z80
	@item -march=@var{CPU}[-@var{EXT}@dots{}][+@var{EXT}@dots{}]
	This option specifies the target processor. The assembler will issue
	an error message if an attempt is made to assemble an instruction which
	will not execute on the target processor. The following processor names
	are recognized:
	@code{z80},
	@code{z180},
	@code{ez80},
	@code{gbz80},
	@code{z80n},
	@code{r800}.
	In addition to the basic instruction set, the assembler can be told to
	accept some extention mnemonics. For example,
	@code{-march=z180+sli+infc} extends @var{z180} with @var{SLI} instructions and
	@var{IN F,(C)}. The following extentions are currently supported:
	@code{full} (all known instructions),
	@code{adl} (ADL CPU mode by default, eZ80 only),
	@code{sli} (instruction known as @var{SLI}, @var{SLL} or @var{SL1}),
	@code{xyhl} (instructions with halves of index registers: @var{IXL}, @var{IXH},
	@var{IYL}, @var{IYH}),
	@code{xdcb} (instructions like @var{RotOp (II+d),R} and @var{BitOp n,(II+d),R}),
	@code{infc} (instruction @var{IN F,(C)} or @var{IN (C)}),
	@code{outc0} (instruction @var{OUT (C),0}).
	Note that rather than extending a basic instruction set, the extention
	mnemonics starting with @code{-} revoke the respective functionality:
	@code{-march=z80-full+xyhl} first removes all default extentions and adds
	support for index registers halves only.

	If this option is not specified then @code{-march=z80+xyhl+infc} is assumed.

	@cindex @code{-local-prefix} command-line option, Z80
	@item -local-prefix=@var{prefix}
	Mark all labels with specified prefix as local. But such label can be
	marked global explicitly in the code. This option do not change default
	local label prefix @code{.L}, it is just adds new one.

	@cindex @code{-colonless} command-line option, Z80
	@item -colonless
	Accept colonless labels. All symbols at line begin are treated as labels.

	@cindex @code{-sdcc} command-line option, Z80
	@item -sdcc
	Accept assembler code produced by SDCC.

	@cindex @code{-fp-s} command-line option, Z80
	@item -fp-s=@var{FORMAT}
	Single precision floating point numbers format. Default: ieee754 (32 bit).

	@cindex @code{-fp-d} command-line option, Z80
	@item -fp-d=@var{FORMAT}
	Double precision floating point numbers format. Default: ieee754 (64 bit).
	@end table
	@c man end

	Floating point numbers formats.
	@table @option
	@item @code{ieee754}
	Single or double precision IEEE754 compatible format.

	@item @code{half}
	Half precision IEEE754 compatible format (16 bits).

	@item @code{single}
	Single precision IEEE754 compatible format (32 bits).

	@item @code{double}
	Double precision IEEE754 compatible format (64 bits).

	@item @code{zeda32}
	32 bit floating point format from z80float library by Zeda.

	@item @code{math48}
	48 bit floating point format from Math48 package by Anders Hejlsberg.
	@end table

	@cindex Z80 Syntax
	@node Z80 Syntax
	@section Syntax
	The assembler syntax closely follows the 'Z80 family CPU User Manual' by
	Zilog.
	In expressions a single @samp{=} may be used as ``is equal to''
	comparison operator.

	Suffices can be used to indicate the radix of integer constants;
	@samp{H} or @samp{h} for hexadecimal, @samp{D} or @samp{d} for decimal,
	@samp{Q}, @samp{O}, @samp{q} or @samp{o} for octal, and @samp{B} for
	binary.

	The suffix @samp{b} denotes a backreference to local label.

	@menu
	* Z80-Chars:: Special Characters
	* Z80-Regs:: Register Names
	* Z80-Case:: Case Sensitivity
	* Z80-Labels:: Labels
	@end menu

	@node Z80-Chars
	@subsection Special Characters

	@cindex line comment character, Z80
	@cindex Z80 line comment character
	The semicolon @samp{;} is the line comment character;

	If a @samp{#} appears as the first character of a line then the whole
	line is treated as a comment, but in this case the line could also be
	a logical line number directive (@pxref{Comments}) or a preprocessor
	control command (@pxref{Preprocessing}).

	@cindex line separator, Z80
	@cindex statement separator, Z80
	@cindex Z80 line separator
	The Z80 assembler does not support a line separator character.

	@cindex location counter, Z80
	@cindex hexadecimal prefix, Z80
	@cindex Z80 $
	The dollar sign @samp{$} can be used as a prefix for hexadecimal numbers
	and as a symbol denoting the current location counter.

	@cindex character escapes, Z80
	@cindex Z80, \
	A backslash @samp{\} is an ordinary character for the Z80 assembler.

	@cindex character constant, Z80
	@cindex single quote, Z80
	@cindex Z80 '
	The single quote @samp{'} must be followed by a closing quote. If there
	is one character in between, it is a character constant, otherwise it is
	a string constant.

	@node Z80-Regs
	@subsection Register Names
	@cindex Z80 registers
	@cindex register names, Z80

	The registers are referred to with the letters assigned to them by
	Zilog. In addition @command{@value{AS}} recognizes @samp{ixl} and
	@samp{ixh} as the least and most significant octet in @samp{ix}, and
	similarly @samp{iyl} and @samp{iyh} as parts of @samp{iy}.

	@c The @samp{'} in @samp{ex af,af'} may be omitted.

	@node Z80-Case
	@subsection Case Sensitivity
	@cindex Z80, case sensitivity
	@cindex case sensitivity, Z80

	Upper and lower case are equivalent in register names, opcodes,
	condition codes and assembler directives.
	The case of letters is significant in labels and symbol names. The case
	is also important to distinguish the suffix @samp{b} for a backward reference
	to a local label from the suffix @samp{B} for a number in binary notation.

	@node Z80-Labels
	@subsection Labels

	@cindex labels, Z80
	@cindex Z80 labels
	Labels started by @code{.L} acts as local labels. You may specify custom local
	label prefix by @code{-local-prefix} command-line option.
	Dollar, forward and backward local labels are supported. By default, all labels
	are followed by colon.
	Legacy code with colonless labels can be built with @code{-colonless}
	command-line option specified. In this case all tokens at line begin are treated
	as labels.

	@node Z80 Floating Point
	@section Floating Point
	@cindex floating point, Z80
	@cindex Z80 floating point
	Floating-point numbers of following types are supported:

	@table @option
	@item @code{ieee754}
	Supported half, single and double precision IEEE754 compatible numbers.

	@item @code{zeda32}
	32 bit floating point numbers from z80float library by Zeda.

	@item @code{math48}
	48 bit floating point numbers from Math48 package by Anders Hejlsberg.
	@end table

	@node Z80 Directives
	@section Z80 Assembler Directives
	@cindex Z80-only directives

	@command{@value{AS}} for the Z80 supports some additional directives for
	compatibility with other assemblers.

	These are the additional directives in @code{@value{AS}} for the Z80:

	@table @code
	@item @code{.assume ADL = @var{expression}}
	@cindex @code{.assume} directive, Z80
	Set ADL status for eZ80. Non-zero value enable compilation in ADL mode else
	used Z80 mode. ADL and Z80 mode produces incompatible object code. Mixing
	both of them within one binary may lead problems with disassembler.

	@item @code{db @var{expression}\|@var{string}[,@var{expression}\|@var{string}...]}
	@cindex @code{db} directive, Z80
	@itemx @code{defb @var{expression}\|@var{string}[,@var{expression}\|@var{string}...]}
	@cindex @code{defb} directive, Z80
	@itemx @code{defm @var{string}[,@var{string}...]}
	@cindex @code{defm} directive, Z80
	For each @var{string} the characters are copied to the object file, for
	each other @var{expression} the value is stored in one byte.
	A warning is issued in case of an overflow.
	Backslash symbol in the strings is generic symbol, it cannot be used as
	escape character. @xref{Ascii,,@code{.ascii}}.

	@item @code{dw @var{expression}[,@var{expression}...]}
	@cindex @code{dw} directive, Z80
	@itemx @code{defw @var{expression}[,@var{expression}...]}
	@cindex @code{defw} directive, Z80
	For each @var{expression} the value is stored in two bytes, ignoring
	overflow.

	@item @code{d24 @var{expression}[,@var{expression}...]}
	@cindex @code{d24} directive, Z80
	@itemx @code{def24 @var{expression}[,@var{expression}...]}
	@cindex @code{def24} directive, Z80
	For each @var{expression} the value is stored in three bytes, ignoring
	overflow.

	@item @code{d32 @var{expression}[,@var{expression}...]}
	@cindex @code{d32} directive, Z80
	@itemx @code{def32 @var{expression}[,@var{expression}...]}
	@cindex @code{def32} directive, Z80
	For each @var{expression} the value is stored in four bytes, ignoring
	overflow.

	@item @code{ds @var{count}[, @var{value}]}
	@cindex @code{ds} directive, Z80
	@itemx @code{defs @var{count}[, @var{value}]}
	@cindex @code{defs} directive, Z80
	@c Synonyms for @code{ds.b},
	@c which should have been described elsewhere
	Fill @var{count} bytes in the object file with @var{value}, if
	@var{value} is omitted it defaults to zero.

	@item @code{@var{symbol} defl @var{expression}}
	@cindex @code{defl} directive, Z80
	The @code{defl} directive is like @code{.set} but with different
	syntax. @xref{Set,,@code{.set}}.
	It set the value of @var{symbol} to @var{expression}. Symbols defined
	with @code{defl} are not protected from redefinition.

	@item @code{@var{symbol} equ @var{expression}}
	@cindex @code{equ} directive, Z80
	The @code{equ} directive is like @code{.equiv} but with different
	syntax. @xref{Equiv, ,@code{.equiv}}.
	It set the value of @var{symbol} to @var{expression}. It is an error
	if @var{symbol} is already defined. Symbols defined with @code{equ}
	are not protected from redefinition.

	@item @code{psect @var{name}}
	@cindex @code{psect} directive, Z80
	A synonym for @code{.section}, no second argument should be given.
	@xref{Section,,@code{.section}}.

	@item @code{xdef @var{symbol}}
	@cindex @code{xdef} directive, Z80
	A synonym for @code{.global}, make @var{symbol} is visible to linker.
	@xref{Global,,@code{.global}}.

	@item @code{xref @var{name}}
	@cindex @code{xref} directive, Z80
	A synonym for @code{.extern} (@ref{Extern,,@code{.extern}}).
	@ignore

	The following attributes will possibly be recognized in the future
	@table @code
	@item abs
	The section is to be absolute. @code{@value{AS}} will issue an error
	message because it can not produce an absolute section.
	@item global
	The section is to be concatenated with other sections of the same name
	by the linker, this is the default.
	@item local
	The section is not global. @code{@value{AS}} will issue a warning if
	object file format is not soff.
	@item ovrld
	The section is to be overlapped with other sections of the same name by
	the linker. @code{@value{AS}} will issue an error message
	because it can not mark a section as such.
	@item pure
	The section is marked as read only.
	@end table
	@end ignore

	@end table

	@node Z80 Opcodes
	@section Opcodes
	In line with common practice, Z80 mnemonics are used for the Z80,
	Z80N, Z180, eZ80, Ascii R800 and the GameBoy Z80.

	In many instructions it is possible to use one of the half index
	registers (@samp{ixl},@samp{ixh},@samp{iyl},@samp{iyh}) in stead of an
	8-bit general purpose register. This yields instructions that are
	documented on the eZ80 and the R800, undocumented on the Z80 and
	unsupported on the Z180.
	Similarly @code{in f,(c)} is documented on the R800, undocumented on
	the Z80 and unsupported on the Z180 and the eZ80.

	The assembler also supports the following undocumented Z80-instructions,
	that have not been adopted in any other instruction set:
	@table @code
	@item out (c),0
	Sends zero to the port pointed to by register @code{C}.

	@item sli @var{m}
	Equivalent to @code{@var{m} = (@var{m}<<1)+1}, the operand @var{m} can
	be any operand that is valid for @samp{sla}. One can use @samp{sll} as a
	synonym for @samp{sli}.

	@item @var{op} (ix+@var{d}), @var{r}
	This is equivalent to

	@example
	ld @var{r}, (ix+@var{d})
	@var{op} @var{r}
	ld (ix+@var{d}), @var{r}
	@end example

	The operation @samp{@var{op}} may be any of @samp{res @var{b},},
	@samp{set @var{b},}, @samp{rl}, @samp{rlc}, @samp{rr}, @samp{rrc},
	@samp{sla}, @samp{sli}, @samp{sra} and @samp{srl}, and the register
	@samp{@var{r}} may be any of @samp{a}, @samp{b}, @samp{c}, @samp{d},
	@samp{e}, @samp{h} and @samp{l}.

	@item @var{op} (iy+@var{d}), @var{r}
	As above, but with @samp{iy} instead of @samp{ix}.
	@end table

	The web site at @uref{http://www.z80.info} is a good starting place to
	find more information on programming the Z80.

	You may enable or disable any of these instructions for any target CPU
	even this instruction is not supported by any real CPU of this type.
	Useful for custom CPU cores.