|
|
|
|
|
|||
|
|||
|
|
|
|
|
|
|
|
|
 |
 |
|
|
|||
| Red Hat Documentation |
 |
The linker always uses a linker script. If you do not supply one yourself, the linker will use a default script that is compiled into the linker executable. You can use the --verbose command line option to display the default linker script. Certain command line options, such as -r or -N, will affect the default linker script. You may supply your own linker script by using the -T command line option. When you do this, your linker script will replace the default linker script. You may also
use linker scripts implicitly by naming them as input files to the linker,
as though they were files to be linked. If the linker opens a file, which
it can not recognize as an object file or as an archive file, it will try
to read it as a linker script. If the file can not be parsed as a linker
script, the linker will report an error. An implicit linker script will
not replace the default linker script. Typically an implicit linker script
would contain only the INPUT, GROUP, or VERSION
commands.
We need to define some basic
concepts and vocabulary in order to describe the linker script language. The linker
combines input files into a single output file. The output file and each
input file are in a special data format known as an object file format .
Each file is called an object file .
The output file is often called an executable ,
but for our purposes we will also call it an object file. Each object file
has, among other things, a list of sections. We sometimes refer to a section
in an input file as an input section ;
similarly, a section in the output file is an output section .
Each section
in an object file has a name and a size. Most sections also have an associated
block of data, known as the section contents .
A section may be marked as loadable ,
meaning that the contents should be loaded into memory when the output
file is run. A section with no contents may be allocatable ,
which means that an area in memory should be set aside, but nothing in
particular should be loaded there (in some cases this memory must be zeroed
out). A section, which is neither loadable nor allocatable, typically contains
some sort of debugging information.
Every loadable
or allocatable output section has two addresses. The first is the VMA ,
or virtual memory address . This is the
address the section will have when the output file is run. The second is
the LMA , or load memory address .
This is the address at which the section will be loaded. In most cases
the two addresses will be the same. An example of when they might be different
is when a data section is loaded into ROM, and then copied into RAM when
the program starts up (this technique is often used to initialize global
variables in a ROM based system). In this case the ROM address would be
the LMA, and the RAM address would be
the VMA.
Every object
file also has a list of symbols , known
as the symbol table . A symbol may be
defined or undefined. Each symbol has a name, and each defined symbol has
an address, among other information. If you compile a C or C++ program
into an object file, you will get a defined symbol for every defined function
and global or static variable. Every undefined function or global variable,
which is referenced in the input file, will become an undefined symbol.
You can see the symbols in an object file by using the nm
program, or by using the objdump
program with the -t option.
Linker scripts are text files. You write a linker script as a series of commands. Each command is either a keyword, possibly followed by arguments or an assignment to a symbol. You may separate commands using semicolons. Whitespace is generally ignored. Strings such
as file or format names can normally be entered directly. If the file name
contains a character such as a comma, which would otherwise serve to separate
file names, you may put the file name in double quotes. There is no way
to use a double quote character in a file name.
You may include
comments in linker scripts just as in C, delimited by /*
and */. As in C, comments are syntactically equivalent to whitespace.
Many linker scripts are fairly simple. The simplest possible linker script has just one command: SECTIONS. You use the SECTIONS command to describe the memory layout of the output file. The SECTIONS command is a powerful command. Here we will describe a simple use of it. Lets assume your program consists only of code, initialized data, and uninitialized data. These will be in the .text , .data, and .bss sections, respectively. Lets assume further that these are the only sections, which appear in your input files. For this example,
lets say that the code should be loaded at address 0x10000 , and
that the data should start at address 0x8000000. The following
linker script will do this function.
SECTIONS {
. = 0x10000; .text : { *(.text) }
. = 0x8000000; .data : { *(.data) }
.bss : { *(.bss) }
} You write the
SECTIONS command as the keyword SECTIONS, followed by
a series of symbol assignments and output section descriptions enclosed
in curly braces. The first line in the above example sets the special symbol
., which is the location counter. If you do not specify the address
of an output section in some other way (other ways are described later),
the address is set from the current value of the location counter. The
location counter is then incremented by the size of the output section.
The second line defines an output section, .text. The colon is
required syntax, which may be ignored for now. Within the curly braces
after the output section name, you list the names of the input sections,
which should be placed into this output section. The * is a wildcard
which matches any file name. The expression *(.text)
means all .text input sections
in all input files.
Since the location
counter is 0x10000 when the output section .text is defined,
the linker will set the address of the .text
section in the output file to be 0x10000. The remaining lines
define the .data and .bss sections in the output file.
The .data output section will be at address 0x8000000
. When the .bss output section is defined, the value of the location
counter will be 0x8000000 plus the size of the ` .data
output section. The effect is that the .bss
output section will follow immediately after the .data
output section in memory.
In the following documentation, the discussion describes the simple linker script commands. See also Command line options for ld and BFD.
The first instruction
to execute in a program is called the entry point.
You can use the ENTRY linker script command to set the entry point.
The argument is a symbol name:
ENTRY (symbol) There are several
ways to set the entry point. The linker will set the entry point by trying
each of the following methods in order, and stopping when one of them succeeds:
Several linker script commands deal with files. See also Command line options for ld and BFD.
INPUT (file file ...)
GROUP (file file ...)
A couple of linker script commands deal with object file formats. See also Command line options for ld and BFD. You
can use OUTPUT_FORMAT with three arguments to use different formats
based on the -EB and -EL command line options. This permits
the linker script to set the output format based on the desired endianness.
If neither -EB nor -EL is used, then the output format
will be the first argument, DEFAULT . If -EB is used,
the output format will be the second argument, BIG. If -EL
is used, the output format will be the third argument, LITTLE.
For example, the default linker script for the MIPS ELF target uses the
following command:
The following options are a few other linker scripts commands. See also Command line options for ld and BFD.
You may assign a value
to a symbol in a linker script. This will define the symbol as a global
symbol.
floating_point = 0; SECTIONS {
.text : {
*(.text) _etext = .; } _bdata = (. + 3) & ~ 4; .data : { *(.data) }
} In the previous
example, the floating_point symbol will be defined as zero. The
_etext symbol will be defined as the address following the last
.text input section. The _bdata symbol will be defined
as the address following the .text output section aligned upward
to a 4 byte boundary.
In some cases, it is desirable for a linker script to define a symbol only if it is referenced and is not defined by any object included in the link. For example, traditional linkers defined the symbol, etext. However, ANSI C requires that the user be able to use etext as a function name without encountering an error. The PROVIDE keyword may be used to define a symbol, such as etext, only if it is referenced but not defined. The syntax is PROVIDE( symbol
= expression).
SECTIONS {
.text : {
*(.text) _etext = .; PROVIDE(etext = .); } } In the previous
example, if the program defines _etext, the linker will give a
multiple definition error. If, on the other hand, the program defines etext,
the linker will silently use the definition in the program. If the program
references etext but does not define it, the linker will use the
definition in the linker script.
The SECTIONS command tells the linker how to map input sections into output sections, and how to place the output sections in memory. The format of the SECTIONS command is: SECTIONS {
sections - command sections - command ... }
The ENTRY
command and symbol assignments are permitted inside the SECTIONS
command for convenience in using the location counter in those commands.
This can also make the linker script easier to understand because you can
use those commands at meaningful points in the layout of the output file.
See Output section description
and Overlay description.
If you do not
use a SECTIONS command in your linker script, the linker will
place each input section into an identically named output section in the
order that the sections are first encountered in the input files. If all
input sections are present in the first file, for example, the order of
sections in the output file will match the order in the first input file.
The first section will be at address-zero.
SECTION [address] [(type)] : [AT(LMA)] {
output-sections-command output-sections-command ... } [>region] [: phdr : phdr ...] [=fillexp] Most output
sections do not use most of the optional section attributes. The whitespace
around SECTION is required, so that the section name is unambiguous.
The colon and the curly braces are also required. The line breaks and other
white space are optional. The expression, LMA, that follows
the AT keyword specifies the load address of the section. This
feature is designed to make it easy to build a ROM image; see also Output
section LMA.
The name of
the output section is section.
section must meet the constraints of your output format.
In formats which only support a limited number of sections, such as a.out,
the name must be one of the names supported by the format (a.out,
for example, allows only .text, .data or .bss).
If the output format supports any number of sections, but with numbers
and not names (as is the case for Oasys), the name should be supplied as
a quoted numeric string . A section name
may consist of any sequence of characters, but a name, which contains any
unusual characters such as commas, must be quoted. The output section name
/DISCARD/ is special. See Output
section discarding.
The address is an expression for the VMA (the virtual memory address) of the output section. If you do not provide address, the linker will set it based on REGION if present, or otherwise based on the current value of the location counter. If you provide
address, the address of the output section will be set
to precisely that specification. If you provide neither address
nor region, then the address of the output section will
be set to the current value of the location counter aligned to the alignment
requirements of the output section. The alignment requirement of the output
section is the strictest alignment of any input section contained within
the output section.
The first will
set the address of the .text output section to the current value
of the location counter. The second will set it to the current value of
the location counter aligned to the strictest alignment of a .text
input section.
The address
may be an arbitrary expression. See Expressions
in linker scripts. For example, if you want to align the section on
a 0x10 byte boundary, so that the lowest four bits of the section address
are zero, you could use something like the following declaration.
.text ALIGN(0x10) : { *(.text) } This works
because ALIGN returns the current location counter aligned upward
to the specified value.
The most common
output section command is an input section description .
The input section description is the most basic linker script operation.
You use output sections to tell the linker how to lay out your program
in memory. You use input section descriptions to tell the linker how to
map the input files into your memory layout.
An input section
description consists of a file name optionally
followed by a list of section names in parentheses. The file name and the
section name may be wildcard patterns, which we describe; see Input
section wildcard patterns. The most common input section description
is to include all input sections with a particular name in the output section.
For example, to include all input .text sections, you would write:
*(.text) The difference
between these is the order in which the .text and .rdata
input sections will appear in the output section. In the first example,
they will be intermingled. In the second example, all .text input
sections will appear first, followed by all .rdata input sections.
You can specify
a file name to include sections from a particular file. You would do this
if one or more of your files contain special data that needs to be at a
particular location in memory. For example, use the following input.
data.o(.data) If you use
a file name without a list of sections, then all sections in the input
file will be included in the output section. This is not commonly done,
but it may by useful on occasion. For example, use the following input.
data.o When you use
a file name, which does not contain any wild card characters, the linker
will first see if you also specified the file name on the linker command
line or in an INPUT command. If you did not, the linker will attempt
to open the file as an input file, as though it appeared on the command
line. This differs from an INPUT command, because the linker will
not search for the file in the archive search path.
In an input
section description, either the file name or the section name or both may
be wildcard patterns. The file name of *
seen in many examples is a simple wildcard pattern for the file name. The
wildcard patterns are like those used by the Unix shell.
When a file
name is matched with a wildcard, the wildcard characters will not match
a / character (used to separate directory names on Unix). A pattern
consisting of a single * character is an exception; it will always
match any file name, whether it contains a / or not. In a section
name, the wildcard characters will match a / character.
File name wildcard patterns only match files which are explicitly specified on the command line or in an INPUT command. The linker does not search directories to expand wildcards. If a file name matches more than one wildcard pattern, or if a file name appears explicitly and is also matched by a wildcard pattern, the linker will use the first match in the linker script. For example, the following input uses a sequence of input section descriptions that is probably in error, because the data.o rule will not be used.
This example
shows how wildcard patterns might be used to partition files. This linker
script directs the linker to place all .text sections in .text
and all .bss sections in .bss. The linker will place
the .data section from all files beginning with an upper case
character in .DATA; for all other files, the linker will place
the .data section in .data.
SECTIONS { .text : { *(.text) }
.DATA : { [A-Z]*(.data) }
.data : { *(.data) }
.bss : { *(.bss) }
} A special notation is needed for common symbols because, in many object file formats, common symbols do not have a particular input section. The linker treats common symbols as though they are in an input section named COMMON. You may use
file names with the COMMON section just as with any other input
sections. You can use this to place common symbols from a particular input
file in one section while common symbols from other input files are placed
in another section.
In most cases,
common symbols in input files will be placed in the .bss section
in the output file. For example, use the following input.
.bss { *(.bss) *(COMMON) } Some object
file formats have more than one type of common symbol. For example, the
MIPS ELF object file format distinguishes standard common symbols and small
common symbols. In this case, the linker will use a different special section
name for other types of common symbols. In the case of MIPS ELF, the linker
uses COMMON for standard common symbols and .scommon
for small common symbols. This permits you to map the different types of
common symbols into memory at different locations.
You will sometimes
see [COMMON] in old linker scripts. This notation is now considered
obsolete. It is equivalent to *(COMMON).
The following example is a complete linker script that tells the linker to read all of the sections from file, all.o, and place them at the start of output section, outputa, which starts at a location, 0x10000 . All of section .input1 from file, foo.o, follows immediately, in the same output section. All of section .input2 from foo.o goes into output section outputb, followed by section .input1 from foo1.o. All of the remaining .input1 and .input2 sections from any files are written to output section outputc . SECTIONS { outputa 0x10000 : {
all.o foo.o (.input1) } outputb : {
foo.o (.input2) foo1.o (.input1) } outputc : {
*(.input1) *(.input2) } } You can include
explicit bytes of data in an output section by using BYTE, SHORT,
LONG, QUAD, or SQUAD as an output section command.
Each keyword is followed by an expression in parentheses providing the
value to store; see Expressions
in linker scripts. The value of the expression is stored at the current
value of the location counter.
The BYTE
, SHORT, LONG, and QUAD commands store one,
two, four, and eight bytes (respectively). After storing the bytes, the
location counter is incremented by the number of bytes stored. For example,
this will store the byte 1 followed by the four byte value of the symbol,
addr:
BYTE(1) LONG(addr) When using
a 64-bit host or target, QUAD and SQUAD are the same;
they both store an 8-byte, or 64-bit, value. When both host and target
are 32 bits, an expression is computed as 32 bits. In this case QUAD
stores a 32-bit value zero extended to 64 bits, and SQUAD stores
a 32-bit value sign extended to 64 bits.
If the object
file format of the output file has an explicit endianness, which is the
normal case, the value will be stored in that endianness. When the object
file format does not have an explicit endianness, as is true of, for example,
S-records, the value will be stored in the endianness of the first input
object file.
You may use
the FILL command to set the fill pattern for the current section.
It is followed by an expression in parentheses. Any otherwise unspecified
regions of memory within the section (for example, gaps left due to the
required alignment of input sections) are filled with the two least significant
bytes of the expression, repeated as necessary. A FILL statement
covers memory locations after the point at which it occurs in the section
definition; by including more than one FILL statement, you can
have different fill patterns in different parts of an output section.
FILL(0x9090) The FILL
command is similar to the =fillexp output section attribute
(see Output section fill); but it only
affects the part of the section following the FILL command, rather
than the entire section. If both are used, the FILL command takes
precedence.
When linking, using the a.out object file format, the linker uses an unusual set construct to support C++ global constructors and destructors. When linking object file formats, which do not support arbitrary sections, such as ECOFF and XCOFF, the linker will automatically recognize C++ global constructors and destructors by name. For these object file formats, the CONSTRUCTORS command tells the linker to place constructor information in the output section where the CONSTRUCTORS command appears. The CONSTRUCTORS command is ignored for other object file formats. The
__CTOR_LIST__ symbol marks the start of the global constructors,
and the __DTOR_LIST symbol marks the end. The first word in the
list is the number of entries, followed by the address of each constructor
or destructor, followed by a zero word. The compiler must arrange to actually
run the code. For these object file formats, GNU C++ normally calls constructors
from a subroutine, __main;
a call to __main is automatically inserted into the startup code
for main. GNU C++ normally runs destructors either by using atexit,
or directly from the function, exit.
__CTOR_LIST__ = .; LONG((__CTOR_END__ - __CTOR_LIST__) / 4 - 2) *(.ctors) LONG(0) __CTOR_END__ = .; __DTOR_LIST__ = .; LONG((__DTOR_END__ - __DTOR_LIST__) / 4 - 2) *(.dtors) LONG(0) __DTOR_END__ = .; The linker will not create output section which do not have any contents. This is for convenience when referring to input sections that may or may not be present in any of the input files. For example, the following input will only create a .foo section in the output file if there is a .foo section in at least one input file. .foo { *(.foo) } If you use
anything other than an input section description as an output section command,
such as a symbol assignment, then the output section will always be created,
even if there are no matching input sections.
The special
output section name, /DISCARD/ may be used to discard input sections.
Any input sections which are assigned to an output section named /DISCARD/
are not included in the output file.
We showed with Output section description that the full description of an output section looked like the following input. SECTION [address] [(type)] : [AT(LMA)] {
output-sections-command output-sections-command ... } [> region ] [: phdr : phdr ...] [=fillexp] We’ve already
described section, address, and output-sections-command.
In the following discussion, we will describe the remaining section attributes.
Each output
section may have a type. The type is a keyword in parentheses. The following
types are defined:
The
linker normally sets the attributes of an output section, based on the
input sections, which map into it. You can override this by using the section
type. For example, in the script sample below, the ROM section
is addressed at memory location 0 and does not need to be loaded
when the program is run. The contents of the ROM section will
appear in the linker output file as usual.
SECTIONS { ROM 0 (NOLOAD) : { ... }
... } Every section has a virtual address (VMA) and a load address (LMA); see Basic linker script concepts. The address expression that, may appear in an output section description sets the VMA. The linker will normally set the LMA equal to the VMA. You can change that by using the AT keyword. The expression, LMA, that follows the AT keyword specifies the load address of the section. This feature is designed to make it easy to build a ROM image. For example, the following linker script creates three output sections: one called .text, which starts at 0x1000, one called .mdata, which is loaded at the end of the .textsection even though its VMA is 0x2000, and one called .bssto hold uninitialized data at address, 0x3000. The symbol _data is defined with the value, 0x2000, which shows that the location counter holds the VMA value, not the LMA value. SECTIONS {
.text 0x1000 : { *(.text) _etext = . ; }
.mdata 0x2000 : AT ( ADDR (.text) + SIZEOF (.text) ) { _data = . ; *(.data); _edata = . ; }
.bss 0x3000 : { _bstart = . ; *(.bss) *(COMMON) ; _bend = . ;}
} The run-time
initialization code for use with a program generated with this linker script
would include something like the following, to copy the initialized data
from the ROM image to its runtime address. Notice how this code takes advantage
of the symbols defined by the linker script.
extern char _etext, _data, _edata, _bstart, _bend; char *src = &_etext; char *dst = &_data; /* ROM has data at end of text; copy it. */ while (dst < &_edata) {
*dst++ = *src++; } /* Zero bss */ for (dst = &_bstart; dst< &_bend; dst++) *dst = 0; You can assign
a section to a previously defined region of memory by using >region,
where region is the defined region for the section. The
following example shows usage.
MEMORY { rom : ORIGIN = 0x1000, LENGTH = 0x1000 } SECTIONS { ROM : { *(.text) } >rom }
You can assign
a section to a previously defined program segment by using :phdr.
If a section is assigned to one or more segments, then all subsequent allocated
sections will be assigned to those segments as well, unless they use an
explicitly :phdr modifier. To prevent a section from being
assigned to a segment when it would normally default to one, use :NONE.
See PHDRS command.
PHDRS { text PT_LOAD ; } SECTIONS { .text : { *(.text) } :text }
You can set
the fill pattern for an entire section by using =fillexp.
fillexp is an expression; see Expressions
in linker scripts. Any otherwise unspecified regions of memory within
the output section (for example, gaps left due to the required alignment
of input sections) will be filled with the two least significant bytes
of the value, repeated as necessary.
You can also
change the fill value with a FILL command in the output section
commands. See Output section data.
SECTIONS { .text : { *(.text) } =0x9090 } An overlay description provides an easy way to describe sections, which are to be loaded as part of a single memory image but are to be run at the same memory address. At run time, some sort of overlay manager will copy the overlaid sections in and out of the runtime memory address as required, perhaps by simply manipulating addressing bits. This approach can be useful, for example, when a certain region of memory is faster than another region of memory. Overlays are
described using the OVERLAY command. The OVERLAY command
is used within a SECTIONS command, like an output section description.
The full syntax of the OVERLAY command is as follows:
OVERLAY [ start ] : [NOCROSSREFS] [AT (ldaddr)] {
secname1 {
output-section-command output-section-command ... } [:PHDR...] [=FILL] secname2 {
output-section-command output-section-command ... } [:phdr ...] [=fill] ... } [>region ] [:phdr ...] [=fill] Everything
is optional except OVERLAY
(a keyword), and each section must have a name (secname1 and secname2
in the previous examples input). The section definitions within the OVERLAY
construct are identical to those within the general SECTIONS
construct, except that no addresses and no memory regions may be defined
for sections within an OVERLAY. See SECTIONS
command.
The sections
are all defined with the same starting address. The load addresses of the
sections are arranged, so that they are consecutive in memory, starting
at the load address used for the OVERLAY
as a whole (as with normal section definitions. The load address is optional,
and defaults to the start address. The start address is also optional,
and defaults to the current value of the location counter).
If the NOCROSSREFS
keyword is used, and there any references among the sections, the linker
will report an error. Since the sections all run at the same address, it
normally does not make sense for one section to refer directly to another.
For each section
within the OVERLAY, the linker automatically defines two symbols.
The symbol, __load_start_ secname, is defined as the
starting load address of the section. The symbol, __load_stop_
secname, is defined as the final load address of the section.
Any characters within secname that are not legal within C identifiers
are removed. C (or assembler) code may use these symbols to move the overlaid
sections around as necessary.
At the end
of the overlay, the value of the location counter is set to the start address
of the overlay plus the size of the largest section.
OVERLAY 0x1000 : AT (0x4000) {
.text0 { o1/*.o(.text) }
.text1 { o2/*.o(.text) }
} This will define
both .text0 and .text1 to start at address 0x1000.
.text0 will be loaded at address 0x4000, and .text1will
be loaded immediately after .text0. The following symbols will
be defined: __load_start_text0, __load_stop_text0 , __load_start_text1,
and __load_stop_text1.
extern char __load_start_text1, __load_stop_text1; memcpy ((char *) 0x1000, &__load_start_text1, &__load_stop_text1 - &__load_start_text1); Note that the
OVERLAY command is just syntactic sugar, since everything it does
can be done using the more basic commands. The previous example could have
been input identically as follows.
.text0 0x1000 : AT (0x4000) { o1/*.o(.text) } __load_start_text0 = LOADADDR (.text0); __load_stop_text0 = LOADADDR (.text0) + SIZEOF (.text0); .text1 0x1000 : AT (0x4000 + SIZEOF (.text0)) { o2/*.o(.text) }
__load_start_text1 = LOADADDR (.text1); __load_stop_text1 = LOADADDR (.text1) + SIZEOF (.text1); . = 0x1000 + MAX (SIZEOF (.text0), SIZEOF (.text1)); The linkers default configuration permits allocation of all available memory. You can override this by using the MEMORY command. The MEMORY
command describes the location and size of blocks of memory in the target.
You can use it to describe which memory regions may be used by the linker,
and which memory regions it must avoid. You can then assign sections to
particular memory regions. The linker will set section addresses based
on the memory regions, and will warn about regions that become too full.
The linker will not shuffle sections around to fit into the available regions.
A linker script
may contain at most one use of the MEMORY command. However, you
can define as many blocks of memory within it as you wish. The syntax is:
MEMORY {
name [( attr )] : ORIGIN = origin , LENGTH = len ... } The name
is a name used in the linker script to refer to the region. The region
name has no meaning outside of the linker script. Region names are stored
in a separate name space, and will not conflict with symbol names, file
names, or section names. Each memory region must have a distinct name.
The attr
string is an optional list of attributes that specify whether to use
a particular memory region for an input section, which is not explicitly
mapped in the linker script. If you do not specify an output section for
some input section, the linker will create an output section with the same
name as the input section. If you define region attributes, the linker
will use them to select the memory region for the output section that it
creates. See SECTIONS command.
If an unmapped
section matches any of the listed attributes other than !, it
will be placed in the memory region. The ! attribute reverses
this test, so that an unmapped section will be placed in the memory region
only if it does not match any of the listed attributes.
The ORIGIN
is an expression for the start address of the memory region. The expression
must evaluate to a constant before memory allocation is performed, which
means that you may not use any section relative symbols. The ORIGIN
keyword may be abbreviated to org or o (but not, for
example, to ORG).
The len
is an expression for the size in bytes of the memory region. As with the
origin expression, the expression must evaluate to a constant before memory
allocation is performed. The LENGTH keyword may be abbreviated
to len or l.
In the following
example, we specify that there are two memory regions available for allocation:
one starting at 0 for 256 kilobytes, and the other starting at
0x40000000 for four megabytes. The linker will place into the
rom memory region every section, which is not explicitly mapped
into a memory region, and is either read-only or executable. The linker
will place other sections, which are not explicitly mapped into a memory
region into the ram memory region.
MEMORY {
rom (rx) : ORIGIN = 0, LENGTH = 256K ram (!rx) : org = 0x40000000, l = 4M } If you have
defined a memory region named mem, you can direct the linker to
place specific output sections into that memory region by using the >region
output section attribute. If no address was specified for the output section,
the linker will set the address to the next available address within the
memory region. If the combined output sections directed to a memory region
are too large for the region, the linker will issue an error message. See
Output section region.
The ELF object file format uses program headers, also knows as segments . The program headers describe how the program should be loaded into memory. You can print them out by using the objdump program with the -p option. When you run
an ELF program on a native ELF system, the system loader reads the program
headers in order to figure out how to load the program. This will only
work if the program headers are set correctly. This documentation does
not describe the details of how the system loader interprets program headers;
for more information, see the ELF ABI.
The linker
will create reasonable program headers by default. However, in some cases,
you may need to specify the program headers more precisely. You may use
the PHDRS command for this purpose. When the linker sees the PHDRS
command in the linker script, it will not create any program headers other
than the ones specified.
The linker
only pays attention to the PHDRS command when generating an ELF
output file. In other cases, the linker will simply ignore PHDRS.
{
name type [ FILEHDR ] [ PHDRS ] [ AT (address) ] [ FLAGS (flags) ] ; } The name is
used only for reference in the SECTIONS command of the linker
script. It is not put into the output file. Program header names are stored
in a separate name space, and will not conflict with symbol names, file
names, or section names. Each program header must have a distinct name.
Certain program
header types describe segments of memory, which the system loader will
load from the file. In the linker script, you specify the contents of these
segments by placing allocatable output sections in the segments. You use
the : phdr output section attribute to place a section
in a particular segment. See Output section
phdr.
It is normal
to put certain sections in more than one segment. This merely implies that
one segment of memory contains another. You may repeat : phdr,
using it once for each segment which should contain the section.
If you place
a section in one or more segments using : phdr, then the linker
will place all subsequent allocatable sections which do not specify :
phdr in the same segments. This is for convenience, since generally
a whole set of contiguous sections will be placed in a single segment.
To prevent a section from being assigned to a segment when it would normally
default to one, use :NONE.
You may use
the FILEHDR and PHDRS keywords appear after the program
header type to further describe the contents of the segment. The FILEHDR
keyword means that the segment should include the ELF file header. The
PHDRS keyword means that the segment should include the ELF program
headers themselves.
You can specify
that a segment should be loaded at a particular address in memory by using
an AT expression. This is identical to the AT command
used as an output section attribute. The AT command for a program
header, overrides the output section attribute. See Output
section, LMA.
The linker
will normally set the segment flags based on the sections, which comprise
the segment. You may use the FLAGS keyword to specify the segment
flags, explicitly. The value of flags must be an integer. It is used to
set the p_flags field of the program header.
The following
example shows the script for PHDRS. This shows a typical set of
program headers used on a native ELF system.
PHDRS {
headers PT_PHDR PHDRS ; interp PT_INTERP ; text PT_LOAD FILEHDR PHDRS ; data PT_LOAD ; dynamic PT_DYNAMIC ; } SECTIONS {
. = SIZEOF_HEADERS; .interp : { *(.interp) } :text :interp
.text : { *(.text) } :text
.rodata : { *(.rodata) } /* defaults to :text */
... . = . + 0x1000; /* move to a new page in memory */ .data : { *(.data) } :data
.dynamic : { *(.dynamic) } :data :dynamic
... } The linker supports symbol versions when using ELF. Symbol versions are only useful when using shared libraries. The dynamic linker can use symbol versions to select a specific version of a function when it runs a program that may have been linked against an earlier version of the shared library. You can include
a version script directly in the main linker script, or you can supply
the version script as an implicit linker script. You can also use the --version-script
linker option.
The format
of the version script commands is identical to that used by Suns linker
in Solaris 2.5. The version script defines a tree of version nodes. You
specify the node names and interdependencies in the version script. You
can specify which symbols are bound to which version nodes, and you can
reduce a specified set of symbols to local scope so that they are not globally
visible outside of the shared library.
VERS_1.1 { global: foo1; local: old*; original*; new*; }; VERS_1.2 {
foo2; } VERS_1.1; VERS_2.0 {
bar1; bar2; } VERS_1.2; This example
version script defines three version nodes. The first version node defined
is VERS_1.1; it has no other dependencies. The script binds the
symbol, foo1, to VERS_1.1. It reduces a number
of symbols to local scope so that they are not visible outside of the shared
library.
Next, the version
script defines node, VERS_1.2. This node depends upon VERS_1.1
. The script binds the symbol, foo2, to the version node,
VERS_1.2.
Finally, the
version script defines node, VERS_2.0. This node depends upon
VERS_1.2. The script binds the symbols, bar1 and bar2,
to the version node, VERS_2.0.
When the linker
finds a symbol defined in a library, which is not specifically bound to
a version node, it will effectively bind it to an unspecified base version
of the library. You can bind all otherwise unspecified symbols to a given
version node by using global: * somewhere in the version script.
The names of
the version nodes have no specific meaning other than what they might suggest
to the person reading them. The 2.0 version could just as well
have appeared in between 1.1 and 1.2. However, this would
be a confusing way to write a version script.
When you link
an application against a shared library that has versioned symbols, the
application itself knows which version of each symbol it requires, and
it also knows which version nodes it needs from each shared library it
is linked against. Thus at runtime, the dynamic loader can make a quick
check to make sure that the libraries you have linked against do in fact
supply all of the version nodes that the application will need to resolve
all of the dynamic symbols. In this way it is possible for the dynamic
linker to know with certainty that all external symbols that it needs will
be resolvable without having to search for each symbol reference.
The symbol
versioning is in effect a much more sophisticated way of doing minor version
checking that SunOS does. The fundamental problem that is being addressed
here is that typically references to external functions are bound on an
as-needed basis, and are not all bound when the application starts up.
If a shared library is out of date, a required interface may be missing;
when the application tries to use that interface, it may suddenly and unexpectedly
fail. With symbol versioning, the user will get a warning when they start
their program if the libraries being used with the application are too
old.
There are several
GNU extensions to Suns versioning approach. The first of these is the ability
to bind a symbol to a version node in the source file where the symbol
is defined instead of in the versioning script. This was done mainly to
reduce the burden on the library maintainer. You can do this by putting
something like this in the C source file:
__asm__(".symver original_foo,foo@VERS_1.1"); This renames
the function, original_foo, to be an alias for foo bound
to the version node VERS_1.1. The local: directive can
be used to prevent the symbol, original_foo, from being exported.
The second
GNU extension is to allow multiple versions of the same function to appear
in a given, shared library. In this way you can make an incompatible change
to an interface without increasing the major version number of the shared
library, while still allowing applications linked against the old interface
to continue to function.
To do this,
you must use multiple .symver directives in the source file. The
following example shows the usage.
__asm__(".symver original_foo,foo@"); __asm__(".symver old_foo,foo@VERS_1.1");
__asm__(".symver old_foo1,foo@VERS_1.2");
__asm__(".symver new_foo,foo@@VERS_2.0");
In this example,
foo@ represents the symbol, foo, bound to the unspecified
base version of the symbol. The source file that contains this example
would define four C functions: original_foo, old_foo
, old_foo1, and new_foo.
When you have
multiple definitions of a given symbol, there needs to be some way to specify
a default version to which external references to this symbol will be bound.
You can do this with the foo@@VERS_2.0 type of .symver
directive. You can only declare one version of a symbol as the default
in this manner; otherwise you would effectively have multiple definitions
of the same symbol.
If you wish
to bind a reference to a specific version of the symbol within the shared
library, you can use the aliases of convenience (i.e., old_foo
), or you can use the .symver directive to bind specifically to
an external version of the function in question.
The syntax for expressions in the linker script language is identical to that of C expressions. All expressions are evaluated as integers. All expressions are evaluated in the same size, which is 32 bits if both the host and target are 32 bits, and is otherwise 64 bits. You can use and set symbol values in expressions. The linker defines several special purpose builtin functions for use in expressions.
All constants are integers. As in C, the linker considers an integer beginning with 0 to be octal, and an integer beginning with 0x or 0X to be hexadecimal. In addition,
you can use the suffixes K and M to scale a constant
by 1024, or 1024*1024, respectively. For example, the
following example declarations all refer to the same quantity.
_fourk_1 = 4K; _fourk_2 = 4096; _fourk_3 = 0x1000; Unless quoted, symbol names start with a letter, underscore, or period and may include letters, digits, underscores, periods, and hyphens. Unquoted symbol names must not conflict with any keywords. You can specify a symbol, which contains odd characters or has the same name as a keyword by surrounding the symbol name in double quotes. "SECTION" = 9; "with a space" = "also with a space" + 10; Since symbols
can contain many non-alphabetic characters, it is safest to delimit symbols
with spaces. For example, A-B is one symbol, whereas A - B
is an expression involving subtraction.
The special linker dot (.) variable always contains the current output location counter. Since the . symbol always refers to a location in an output section, it may only appear in an expression within a SECTIONS command. The . symbol may appear anywhere that an ordinary symbol is allowed in an expression. Assigning a
value to . symbol will cause the location counter to be moved.
This may be used to create holes in the output section. The location counter
may never be moved backwards.
SECTIONS {
output : {
file1(.text) . = . + 1000; file2(.text) . += 1000; file3(.text) } = 0x1234; } In the previous
example, the .text section from file1 is located at the
beginning of the output section, output. It is followed by a 1000
byte gap. Then the .text section from file2 appears,
also with a 1000 byte gap following before the .text section from
file3. The notation, = 0x1234, specifies data to write
in the gaps.
The linker recognizes the
standard C set of arithmetic operators, with the standard bindings and
precedence levels; see Table
1: Arithmetic operators with precedence levels and bindings associations.
The linker evaluates expressions lazily. It only computes the value of an expression when absolutely necessary. The linker
needs some information, such as the value of the start address of the first
section, and the origins and lengths of memory regions, in order to do
any linking at all. These values are computed as soon as possible when
the linker reads in the linker script.
However, other
values (such as symbol values) are not known or needed until after storage
allocation. Such values are evaluated later, when other information (such
as the sizes of output sections) is available for use in the symbol assignment
expression.
The sizes of
sections cannot be known until after allocation, so assignments dependent
upon these are not performed until after allocation.
Some expressions,
such as those depending upon the location counter ., must be evaluated
during section allocation.
If the result
of an expression is required, but the value is not available, then an error
results. For example, a script, like the following, will cause the error
message, non constant expression for initial address:
SECTIONS {
.text 9+this_isnt_constant : { *(.text) }
} When the linker evaluates an expression, the result is either absolute or relative to some section. A relative expression is expressed as a fixed offset from the base of a section. The position
of the expression within the linker script determines whether it is absolute
or relative. An expression, which appears within an output section definition,
is relative to the base of the output section. An expression, which appears
elsewhere, will be absolute.
A symbol set
to a relative expression will be relocatable if you request relocatable
output using the -r option. That means that a further link operation
may change the value of the symbol. The symbols section will be the section
of the relative expression.
A symbol set
to an absolute expression will retain the same value through any further
link operation. The symbol will be absolute, and will not have any particular
associated section.
You can use
the builtin function ABSOLUTE to force an expression to be absolute
when it would otherwise be relative. For example, to create an absolute
symbol set to the address of the end of the output section .data
:
SECTIONS {
.data : { *(.data) _edata = ABSOLUTE(.); }
} The linker script language
includes a number of builtin functions for use in linker script expressions.
ADDR( section
)
.output1 : {
start_of_output_1 = ABSOLUTE(.); ... } .output : {
symbol_1 = ADDR(.output1); symbol_2 = start_of_output_1; } ... } ALIGN(exp)
(. + exp - 1) & ~( exp - 1)ALIGN doesnt change the value of the location counter, it just does arithmetic on it. The following example aligns the output .data section to the next 0x2000 byte boundary after the preceding section and sets a variable within the section to the next 0x8000 boundary after the input sections: SECTIONS { ... .data ALIGN(0x2000): {
*(.data) variable = ALIGN(0x8000); } ... }The first use of ALIGN in this example specifies the location of a section because it is used as the optional ADDRESS attribute of a section definition. The second use of ALIGN is to define the value of a symbol. The builtin function, NEXT, is closely related to ALIGN. See Output section address. BLOCK(exp)
DEFINED(symbol)
SECTIONS{ ... .text : {
begin = DEFINED(begin) ? begin : . ; ... } ... } LOADADDR(section)
SIZEOF( section
)
SECTIONS{ ... .output {
.start = . ; ... |
|
|
© 2001 Red Hat, Inc. All rights reserved. About Red Hat : Legal statement : Privacy statement : Y2K statement : Contact Red Hat |