Section are used to organize code and data of a program on the binary level.
The (compiler-generated) assembly code assigns code, data and other entities like debug information to so called input sections. These sections serve as input to the linker, which bundles similar sections together to output sections like .text and .data according to rules defined in the linker description file.
The final ELF binary is then used by programming tools like avrdude, simulators, debuggers and other programs, for example programs from the GNU Binutils family like avr-size, avr-objdump and avr-readelf.
Sections may have extra properties like section alignment, section flags, section type and rules to locate them or to assign them to memory regions.
Concepts
Named Sections
Named sections are sections that can be referred to by their name. The name and other properties can be provided with the .section directive like in
.section name, "flags", @type
or with the .pushsection directive, which directs the assembler to assemble the following code into the named section.
An example of a section that is not referred to by its name is the COMMON section. In order to put an object in that section, special directives like .comm name,size or .lcomm name,size have to be used.
Directives like .text are basically the same like .section .text, where the assembler assumes appropriate section flags and type; same for directives .data and .bss.
Section Flags
The section flags can be specified with the .section and .pushsection directives, see section type for an example. Section flags of output sections can be specified in the linker description file, and the linker implements heuristics to determine the section flags of output sections from the various input section that go into it.
Section Flags
| Flag | Meaning |
a | The section will be allocated, i.e. it occupies space on the target hardware |
w | The section contains data that can be written at run-time. Sections that only contain read-only entities don't have the w flag set |
x | The section contains executable code, though the section may also contain non-executable objects |
M | A mergeable section |
S | A string section |
G | A section group, like used with comdat objects |
The last three flags are listed for completeness. They are used by the compiler, for example for header-only C++ modules and to ensure that multiplle instanciations of the same template in different compilaton units does occur at most once in the executable file.
Section Type
The section type can be specified with the .section and .pushsection directives, like in
.section .text.myfunc,"ax",@progbits
.pushsection ".data.myvar", "a", "progbits"
On ELF level, the section type is stored in the section header like Elf32_Shdr.sh_type = SHT_PROGBITS.
Section Types
| Type | Meaning |
@progbits | The section contains data that will be loaded to the target, like objects in the .text and .data sections. |
@nobits | The section does not contain data that needs to be transferred to the target device, like data in the .bss and .noinit sections. The section still occupies space on the target. |
@note | The section is a note, like for example the .note.gnu.avr.deviceinfo section. |
Section Alignment
The alignment of a section is the maximum over the alignments of the objects in the section.
Subsections
Subsections are compartments of named sections and are introduced with the .subsection directive. Subsections are located in order of increasing index in their input section. The default subsection after switching to a new section is subsection 0.
- Note
- A common misconception is that a section like .text.module.func were a subsection of .text.module. This is not the case. These two sections are independent, and there is no subset relation. The sections may have different flags and type, and they may be assigned to different output sections.
Orphan Sections
Orphan sections are sections that are not mentioned in the linker description file. When an input section is orphan, then the GNU linker implicitly generates an output section of the same name. The linker implements various heuristics to determine sections flags, section type and location of orphaned sections. One use of orphan sections is to locate code to a fixed address.
Like for any other output section, the start address can be specified by means of linking with -Wl,--section-start,secname=address
LMA: Load Memory Address
The LMA of an object is the address where a loader like avrdude puts the object when the binary is being uploaded to the target device.
VMA: Virtual Memory Address
The VMA is the address of an object as used by the running program.
VMA and LMA may be different: Suppose a small ATmega8 program with executable code that extends from byte address 0x0 to 0x20f, and one variable my_var in static strorage. The default linker script puts the content of the .data output section after the .text output section and into the text segment. The startup code then copies my_data from its LMA location beginning at 0x210 to its VMA location beginning at 0x800060, because C/C++ requires that all data in static storage must have been initialized when main is entered.
The internal SRAM of ATmega8 starts at RAM address 0x60, which is offset by 0x800000 in order to linearize the address space (VMA 0x60 is a flash address). The AVR program only ever uses the lower 16 bits of VMAs in static storage so that the offset of 0x800000 is masked out. But code like "LDI r24,hh8(my_data)" actually sets R24 to 0x80 and reveals that my_data is an object located in RAM.
The Linker Script: Building Blocks
The linker description file is the central hub to channel functions and static storage objects of a program to the various memory spaces and address ranges of a device.
Input Sections and Output Sections
Input sections are sections that are inputs to the linker. Functions and static variables but also additional notes and debug information are assigned to different input sections by means of assembler directives like .section or .text. The linker takes all these sections and assigns them to output sections as specified in the linker script.
Output sections are defind in the linker description file. Contrary to the unlimited number of input sections a program can come up with, there is only a handfull of output sections like .text and .data, that roughly correspond to the memory spaces of the target device.
One step in the final link is to locate the sections, that is the linker/locator determines at which memory location to put the output sections, and how to arrange the many input sections within their assigned output section. Locating means that the linker assigns Load Memory Addresses — addresses as used by a loader like avrdude — and Virtual Memory Addresses, which are the addresses as used by the running program.
While it is possible to directly assign LMAs and VMAs to output sections in the linker script, the default linker scripts provided by Binutils assign memory regions (aka. memory segments) to the output sections. This has some advantages like a linker script that is easier to maintain. An output sections can be assigned to more than one memory region. For example, non-zero data in static storage (.data) goes to
- the
data region (VMA), because such variables occupy RAM which has to be allocated
- the
text region (LMA), because the initializers for such data has to be kept in some non-volatile memory (program ROM), so that the startup code can initialize that data so that the variables have their expected initial values when main() is entered.
The SECTIONS{} portion of a linker script models the input and output section, and it assignes the output section to the memory regions defined in the MEMORY{} part.
Memory Regions
The memory regions defined in the default linker script model and correspond to the different kinds of memories of a device.
Memory Regions of the Default Linker Script
| Region | Virtual
Address1 | Flags | Purpose |
text | 02 | rx | Executable code, vector table, data in PROGMEM, __flash and __memx, startup code, linker stubs, initializers for .data |
data | 0x8000002 | rw | Data in static storage |
| rodata3 | 0xa000002 | r | Read-only data in static storage |
eeprom | 0x810000 | rw | EEPROM data |
fuse | 0x820000 | rw | Fuse bytes |
lock | 0x830000 | rw | Lock bytes |
signature | 0x840000 | rw | Device signature |
user_signatures | 0x850000 | rw | User signature |
Notes
- The VMAs for regions other than
text are offset in order to linearize the non-linear memory address space of the AVR Harvard architecture. The target code only ever uses the lower 16 bits of the VMA to access objects in non-text regions.
- The addresses for regions
text, data and rodata are actually defined as symbols like __TEXT_REGION_ORIGIN__, so that they can be adjusted by means of, say -Wl,--defsym,__DATA_REGION_ORIGIN__=0x800060. Same applies for the lengths of all the regions, which is __NAME_REGION_LENGTH__ for region name.
- The
rodata region is only present in the avrxmega2_flmap and avrxmega4_flmap emulations, which is the case for Binutils since v2.42 for the AVR64 and AVR128 devices without -mrodata-in-ram.
Output Sections of the Default Linker Script
This section describes the various output sections defined in the default linker description files.
Output Sections and Memory Regions
| Output | Purpose | Memory Region |
| Section | | LMA | VMA |
| .text | Executable code, data in progmem | text | text |
| .data | Non-zero data in static storage | text | data |
| .bss | Zero data in static storage | — | data |
| .noinit | Non-initialized data in static storage | — | data |
| .rodata1 | Read-only data in static storage | text | LMA + offset3 |
| .rodata2 | Read-only data in static storage | 0x8000 * __flmap4 | rodata |
| .eeprom | Data in EEPROM | Note5 | eeprom |
| .fuse | Fuse bytes | fuse |
| .lock | Lock bytes | lock |
| .signature | Signature bytes | signature |
| User signature bytes | user_signatures |
Notes
- On avrxmega3 and avrtiny devices.
- On AVR64 and AVR128 devices without
-mrodata-in-ram.
- With an offset
__RODATA_PM_OFFSET__ of 0x4000 or 0x8000 depending on the device.
- The value of symbol
__flmap defaults to the last 32 KiB block of program memory, see the GCC v14 release notes.
- The LMA actually equals the VMA, but is unused. The flash loader like avrdude knows where to put the data,
The .text Output Section
The .text output section contains the actual machine instructions which make up the program, but also additional code like jump tables and lookup tables placed in program memory with the PROGMEM attribute.
The .text output section contains the input sections described below. Input sections that are not used by the tools are omitted. A * wildcard stands for any sequence of characters, including empty ones, that are valid in a section name.
.vectors The .vectors sections contains the interrupt vector table which consists of jumps to weakly defined labels: To __init for the first entry at index 0, and to __vector_N for the entry at index N ≥ 1. The default value for __vector_N is __bad_interrupt, which jumps to weakly defined __vector_default, which jumps to __vectors, which is the start of the .vectors section.
Implementing an interrupt service ruotine (ISR) is performed with the help of the ISR macro in C/C++ code.
.progmem.data .progmem.data.* .progmem.gcc.* This section is used for read-only data declared with attribute PROGMEM, and for data in address-space __flash.
The compiler assumes that the .progmem sectons are located in the lower 64 KiB of program memory. When it does not fit in the lower 64 KiB block, then the program reads garbage except pgm_read_*_far is used. In that case however, code can be located in the .progmemx section which does not require to be located in the lower program memory.
- .trampolines
- Linker stubs for indirect jumps and calls on devices with more than 128 KiB of program memory. This section must be located in the same 128 KiB block like the interrupt vector table. For some background on linker stubs, see the GCC documentation on EIND.
.text .text.* - Executable code. This is where almost all of the executable code of an application will go.
.ctors .dtors - Tables with addresses of static constructors and destructors, like C++ static constructors and functions declared with attribute
constructor.
- The
.initN Sections
These sections are used to hold the startup code from reset up through the start of main().
The .initN sections are executed in order from 0 to 9: The code from one init section falls through to the next higher init section. This is the reason for why code in these sections must be naked (more precisely, it must not contain return instructions), and why code in these sections must never be called explicitly.
When several modules put code in the same init section, the order of execuation is not specified.
The .initN Sections
| Section | Performs | Hosted By | Symbol1 |
| .init0 | Weakly defines the __init label which is the jump target of the first vector in the interrupt vector table. When the user defines the __init() function, it will be jumped to instead. | AVR-LibC2 | |
| .init1 | Unused | — |
| .init2 |
- Clears
__zero_reg__ - Initializes the stack pointer to the value of weak symbol __stack, which has a default value of RAMEND as defined in avr/io.h - Initializes EIND to hh8(pm(__vectors)) on devices that have it - Initializes RAMPX, RAMPY, RAMPZ and RAMPD on devices that have all of them AVR-LibC .init3 Initializes the NVMCTRLB.FLMAP bit-field on devices that have it, except when -mrodata-in-ram is specified AVR-LibC __do_flmap_init .init4 Initializes data in static storage: Initializes \ref sec_dot_data .data"</tt> and clears
<tt>\ref sec_dot_bss ".bss"</tt>
<td>libgcc
<td>\c __do_copy_data<br>\c __do_clear_bss
</tr>
<tr>
<td><tt>.init5</tt>
<td>Unused
<td>—
<td>
</tr>
<tr>
<td><tt>.init6</tt>
<td>Run static C++ constructors and functions defined with
<code>__attribute__((constructor))</code>.
<td>libgcc
<td>\c __do_global_ctors
</tr>
<tr>
<td><tt>.init7</tt>
<td>Unused
<td>—
<td rowspan="3">
</tr>
<tr>
<td><tt>.init8</tt>
<td>Unused
<td>—
</tr>
<tr>
<td><tt>.init9</tt>
<td>Calls \c main and then jumps to \c #exit
<td>AVR-LibC
</tr>
</table>
Notes
-# Code in the <tt>.init3</tt>, <tt>.init4</tt> and <tt>.init6</tt>
sections is optional; it will only be present when there is something
to do. This will be tracked by the compiler — or has to be tracked by
the assembly programmer — which pulls in the code from the respective
library by means of the mentioned symbols, e.g. by linking
with <tt>-Wl,-u,__do_flmap_init</tt> or by means of
@code
.global __do_copy_data
\endcode
Conversely, when the respective code is not desired for some reason, the
symbol can be satisfied by defining it with, say,
<tt>-Wl,\--defsym,__do_copy_data=0</tt> so that the code is not pulled in
any more.
-# The code is provided by
<a href="https://github.com/avrdudes/avr-libc/blob/main/crt1/gcrt1.S"
><tt>gcrt1.S</tt></a>.
</dd>
<dt>@anchor sec_dot_fini The <code>.fini<em>N</em></code> Sections</dt>
<dd>
Shutdown code.
These sections are used to hold the exit code executed after return from
main() or a call to exit().
The <tt>.fini<em>N</em></tt> sections are executed in descending order
from 9 to 0 in a fallthrough manner.
<table>
<caption>The <tt>.fini<em>N</em></tt> Sections</caption>
<tr>
<th>Section
<th>Performs
<th>Hosted By
<th>Symbol
</tr>
<tr>
<td><tt>.fini9</tt>
<td>Defines \c _exit and weakly defines the \c #exit label
<td>libgcc
<td rowspan="3">
</tr>
<tr>
<td><tt>.fini8</tt>
<td>Run functions registered with \c atexit()
<td>AVR-LibC
</tr>
<tr>
<td><tt>.fini7</tt>
<td>Unused
<td>—
</tr>
<tr>
<td><tt>.fini6</tt>
<td>Run static C++ destructors and functions defined with
<code>__attribute__((destructor))</code>
<td>libgcc
<td>\c __do_global_dtors
</tr>
<tr>
<td><tt>.fini5...1</tt>
<td>Unused
<td>—
<td rowspan="2">
</tr>
<tr>
<td><tt>.fini0</tt>
<td>Globally disables interrupts and enters an infinite loop to
label \c __stop_program
<td>libgcc
</tr>
</table>
It is unlikely that ordinary code uses the fini sections.
When there are no static destructors and \c atexit() is not used, then
the respective code is not pulled in form the libraries, and the
fini code just consumes four bytes: a \c CLI and a RJMP to itself.
Common use cases of fini code is when running the GCC test suite where
it reduces fallout, and in simulators to determine (un)orderly
termination of a simulated program.
</dd>
<dt>@anchor sec_dot_progmemx <tt>.progmemx.*</tt></dt>
<dd>
Read-only data in program memory without the requirement that it
must reside in the lower 64 KiB. The compiler uses this section
for data in the named address-space \c __memx.
Data can be accessed with \ref avr_pgmspace "pgm_read_*_far" when it is
not in a named address-space:
@code
#include <avr/pgmspace.h>
const __memx int array1[] = { 1, 4, 9, 16, 25, 36 };
PROGMEM_FAR
const int array2[] = { 2, 3, 5, 7, 11, 13, 17 };
int add (uint8_t id1, uint8_t id2)
{
uint_farptr_t p_array2 = pgm_get_far_address (array2);
int val2 = pgm_read_int_far (p_array2 + sizeof(int) * id2);
return val2 + array1[id1];
}
\endcode
</dd>
<dt>@anchor sec_dot_jumptables <tt>.jumptables*</tt></dt>
<dd>
Used to place jump tables in some cases.
</dd>
</dl>
@subsection sec_dot_data The .data Output Section
This section contains data in static storage which has an initializer
that is not all zeroes.
This includes the following input sections:
<dl>
<dt><tt>.data*</tt></dt>
<dd>
Read-write data
</dd>
<dt><tt>.rodata*</tt></dt>
<dd>
Read-only data. These input sections are only included on devices
that host read-only data in RAM.
</dd>
</dl>
It is possible to tell the linker the SRAM address of the beginning of the
<tt>.data</tt> section. This is accomplished by linking with
@code
avr-gcc ... -Tdata addr -Wl,--defsym,__DATA_REGION_START__=addr
\endcode
Note that <tt>addr</tt> must be \ref harvard_arch "offset" by adding
0x800000 the to real SRAM address so that the linker knows that the address is
in the SRAM memory segment. Thus, if you want the <tt>.data</tt> section
to start at \c 0x1100, pass \c 0x801100 as the address to the linker.
\note
When using \c malloc() in the application (which could even happen inside
library calls), \ref malloc_extram "additional adjustments" are required.
@subsection sec_dot_bss The .bss Output Section
Data in static storage that will be zeroed by the startup code.
This are data objects without explicit initializer, and data objects
with initializers that are all zeroes.
Input sections are <tt>.bss*</tt> and \c COMMON. Common symbols are
defined with directives <tt>.comm</tt> or <tt>.lcomm</tt>.
@subsection sec_dot_noinit The .noinit Output Section
Data objects in static storage that should not be initialized by the
startup code. As the C/C++ standard requires that <em>all</em> data
in static storage is initialized — which includes data without explicit
initializer, which will be initialized to all zeroes — such objects
have to be put into section <tt>.noinit</tt> by hand:
@code
__attribute__ ((section (".noinit")))
int foo;
\endcode
The only input section in this output section is <tt>.noinit</tt>.
Only data without initializer can be put in this section.
@subsection sec_dot_rodata The .rodata Output Section
This section contains read-only data in static storage from
<tt>.rodata*</tt> input sections. This output section is only
present for devices where read-only data remains in program memory,
which are the devices where (parts of) the program memory are visible in the
RAM address space. This is currently the case for the emulations
\c avrtiny, \c avrxmega3, \c avrxmega2_flmap and \c avrxmega4_flmap.
@subsection sec_dot_eeprom The .eeprom Output Section
This is where EEPROM variables are stored, for example variables declared
with the \c #EEMEM attribute.
The only input section (pattern) is <tt>.eeprom*</tt>.
@subsection sec_dot_fuse The .fuse, .lock and .signature Output Sections
These sections contain fuse bytes, lock bytes and device signature bytes,
respectively. The respective input section patterns are
<tt>.fuse*</tt> <tt>.lock*</tt> and <tt>.signature*</tt>.
@subsection sec_dot_note The .note.gnu.avr.deviceinfo Section
This section is actually <em>not mentioned</em> in the default linker
script, which means it is an \ref sec_orphan "orphan section"
and hence the respective output section is implicit.
The startup code from AVR-LibC puts device information in that section
to be picked up by simulators or tools like avr-size, avr-objdump,
avr-readelf, etc,
The section is contained in the ELF file but not loaded onto the target.
Source of the device specific information are the device header file
and compiler builtin macros. The layout conforms to the standard
<a href="https://man7.org/linux/man-pages/man5/elf.5.html"
>ELF note section</a> layout and is laid out as follows.
@code
#include <elf.h>
typedef struct
{
Elf32_Word n_namesz; /* AVR_NOTE_NAME_LEN */
Elf32_Word n_descsz; /* size of avr_desc */
Elf32_Word n_type; /* 1 - the only AVR note type */
} Elf32_Nhdr;
#define AVR_NOTE_NAME_LEN 4
struct note_gnu_avr_deviceinfo
{
Elf32_Nhdr nhdr;
char note_name[AVR_NOTE_NAME_LEN]; /* = "AVR\0" */
struct
{
Elf32_Word flash_start;
Elf32_Word flash_size;
Elf32_Word sram_start;
Elf32_Word sram_size;
Elf32_Word eeprom_start;
Elf32_Word eeprom_size;
Elf32_Word offset_table_size;
/* Offset table containing byte offsets into
string table that immediately follows it.
index 0: Device name byte offset */
Elf32_Off offset_table[1];
/* Standard ELF string table.
index 0 : NULL
index 1 : Device name
index 2 : NULL */
char strtab[2 + strlen(__AVR_DEVICE_NAME__)];
} avr_desc;
};
\endcode
The contents of this section can be displayed with
- <tt>avr-objdump -P avr-deviceinfo <em>file</em></tt>, which is supported since
Binutils v2.43.
- <tt>avr-readelf -n <em>file</em></tt>, which displays all notes.
@section sec_symbols Symbols in the Default Linker Script
Most of the symbols like \c main are defined in the code of the
application, but some symbols are defined in the default linker script:
<dl>
<dt><tt>__<em>name</em>_REGION_ORIGIN__</tt></dt>
<dd>
Describes the physical properties of memory region <tt><em>name</em></tt>,
where <tt><em>name</em></tt> is one of \c TEXT or \c DATA.
The address is a VMA and offset at explained above.
<br>
The linker script only supplies a default for the symbol values when
they have not been defined by other means, like for example
in the startup code or by <tt>\ref opt_defsym "--defsym"</tt>.
For example, to let the code start at address \c 0x100, one can link
with
@verbatim
avr-gcc ... -Ttext=0x100 -Wl,--defsym,__TEXT_REGION_ORIGIN__=0x100
\endverbatim
</dd>
<dt><tt>__<em>name</em>_REGION_LENGTH__</tt></dt>
<dd>
Describes the physical properties of memory region <tt><em>name</em></tt>,
where <tt><em>name</em></tt> is one of:
\c TEXT, \c DATA, \c EEPROM, \c LOCK, \c FUSE, \c SIGNATURE or
\c USER_SIGNATURE.
<br>
Only a default is supplied when the symbol is not yet defined by
other means. Most of these symbols are
\ref dot_weak "weakly" defined in the startup code.
</dd>
<dt><tt>__data_start</tt></dt>
<dt><tt>__data_end</tt></dt>
<dd>
Start and (one past the) end \ref sec_vma "VMA" address of the
<tt>\ref sec_dot_data ".data"</tt> section in RAM.
</dd>
<dt><tt>__data_load_start</tt></dt>
<dt><tt>__data_load_end</tt></dt>
<dd>
Start and (one past the) end \ref sec_lma "LMA" address of the
<tt>\ref sec_dot_data ".data"</tt> section initializers located
in program memory.
Used together with the VMA addresses above by the
\ref sec_dot_init "startup code" to copy data initializers from
program memory to RAM.
</dd>
<dt><tt>__bss_start</tt></dt>
<dt><tt>__bss_end</tt></dt>
<dd>
Start and (one past the) end VMA address of the
<tt>\ref sec_dot_bss ".bss"</tt> section.
The startup code clears this part of the RAM.
</dd>
<dt><tt>__rodata_start</tt></dt>
<dt><tt>__rodata_end</tt></dt>
<dt><tt>__rodata_load_start</tt></dt>
<dt><tt>__rodata_load_end</tt></dt>
<dd>
Start and (one past the) end VMA resp. LMA address of the \c \.rodata
output section. These symbols are only defined when \c \.rodata is not
output to the \c text region, which is the case for emulations
\c avrxmega2_flmap and \c avrxmega4_flmap.
</dd>
<dt><tt>__heap_start</tt></dt>
<dd>
One past the last object located in static storage. Immediately
follows the <tt>\ref sec_dot_noinit ".noinit"</tt> section
(which immediately follows <tt>\ref sec_dot_bss ".bss"</tt>,
which immediately follows <tt>\ref sec_dot_data ".data"</tt>).
Used by \ref malloc "malloc()" and friends.
</dd>
</dl>
Code that computes a checksum over all relevant code and data in program
memory has to consider:
- The range from the beginning of the \c \.text section
(address 0x0 in the default layout) up to \c __data_load_start.
- For emulations that have the \c rodata
\ref sec_memory_regions "memory region",
the range from \c __rodata_load_start to \c __rodata_load_end has also to be
taken into account.
@section sec_size Output Sections and Code Size
The \c avr-size program (part of Binutils), coming from a Unix background,
doesn't account for the <code>\ref sec_dot_data ".data"</code> initialization
space added to the <code>\ref sec_dot_text ".text"</code> section,
so in order to know how much flash the final program will consume, one
needs to add the values for both, <code>.text</code> and <code>.data</code>
(but not <code>\ref sec_dot_bss ".bss"</code>), while the amount of
pre-allocated SRAM is the sum of <code>.data</code> and <code>.bss</code>.
Memory usage and free memory can also be displayed with
@verbatim
avr-objdump -P mem-usage code.elf
\endverbatim
@section sec_using Using Sections
@subsection c_sections In C/C++ Code
The following example shows how to read and reset the \c MCUCR special
function register on ATmega328.
This SFR holds to reset source like "watchdog reset" or "external reset",
and should be read early, prior to the initialization of RAM and
execution of static constructors which may take some time. This means the
code has to be placed prior to <tt>\ref sec_dot_init ".init4"</tt>
which initializes static storage, but after <tt>.init2</tt> which
initializes <tt>\ref zero_reg "__zero_reg__"</tt>.
As the code runs prior to the initialization of static storage,
variable \c mcucr must be placed in section
<tt>\ref sec_dot_noinit ".noinit"</tt> so that it won't be overridden
by that part of the startup code:
@code
#include <avr/io.h>
__attribute__((section(".noinit")))
uint8_t mcucr;
__attribute__((used, unused, naked, section(".init3")))
static void read_MCUCR (void)
{
mcucr = MCUCR;
MCUCR = 0;
}
\endcode
- The \c used attribute tells the compiler that the function
is used although it is never called.
- The \c unused attribute tells the compiler that it is fine that
the function is unused, and silences respective diagnostics
about the seemingly unused functions.
- The \c naked attribute is required because the code is located in an
init section. The function <em>must not have a \c RET statement</em>
because the function is never called.
According to the GCC documentation, the only code supported in naked
functions is inline assembly, but the code above is simple enough so
that GCC can deal with it.
@subsection asm_sections In Assembly Code
Example:
@code
#include <avr/io.h>
.section .init3,"ax",@progbits
lds r0, MCUCR
.pushsection .noinit,"a",@nobits
mcucr:
.type mcucr, @object
.size mcucr, 1
.space 1
.popsection ; Proceed with .init3
sts mcucr, r0
sts MCUCR, __zero_reg__ ; Initialized in .init2
.text
.global main
.type main, @function
lds r24, mcucr
clr r25
rjmp putchar
.size main, .-main
\endcode
- The <tt>"ax"</tt> \ref sec_flags "flags" tells that the sections is
<b>a</b>llocatable (consumes space on the target hardware) and is
e<b>x</b>ecutable.
- The <tt>\@progbits</tt> \ref sec_type "type"
tells that the section contains bits that have to be uploaded to the
target hardware.
For more detais, see the see the gas user manual on the
<a href="https://sourceware.org/binutils/docs/as/Section.html#ELF-Version" >.section
directive. |
1.15.0