Files
binutils-gdb/ld/scripttempl/elf64hppa.sc
H.J. Lu 9b854f169d ld: Add LTO and none-LTO output support for ld -r
Link with mixed IR/non-IR objects

* 2 kinds of object files
  o non-IR object file has
    * non-IR sections
  o IR object file has
    * IR sections
    * non-IR sections
    * The output of "ld -r" with mixed IR/non-IR objects should work with:
        o Compilers/linkers with IR support.
	o Compilers/linkers without IR support.
* Add the mixed object file which has
  o IR sections
  o non-IR sections:
    * Object codes from IR sections.
    * Object codes from non-IR object files.
  o Object-only section:
    * With section name ".gnu_object_only" and SHT_GNU_OBJECT_ONLY type
    on ELF:
    https://gitlab.com/x86-psABIs/Linux-ABI
    #define SHT_GNU_OBJECT_ONLY 0x6ffffff8	/* Object only */
    * Contain non-IR object file.
    * Input is discarded after link.
* Linker action:
  o Classify each input object file:
    * If there is a ".gnu_object_only" section, it is a mixed object file.
    * If there is a IR section, it is an IR object file.
    * Otherwise, it is a non-IR object file.
  o Relocatable non-IR link:
    * Prepare for an object-only output.
    * Prepare for a regular output.
    * For each mixed object file:
      * Add IR and non-IR sections to the regular output.
      * For object-only section:
	* Extract object only file.
	* Add it to the object-only output.
	* Discard object-only section.
    * For each IR object file:
      * Add IR and non-IR sections to the regular output.
    * For each non-IR object file:
      * Add non-IR sections to the regular output.
      * Add non-IR sections to the object-only output.
    * Final output:
      * If there are IR objects, non-IR objects and the object-only
      output isn't empty:
	* Put the object-only output into the object-only section.
	* Add the object-only section to the regular output.
	* Remove the object-only output.
  o Normal link and relocatable IR link:
    * Prepare for output.
    * IR link:
      * For each mixed object file:
	* Compile and add IR sections to the output.
	* Discard non-IR sections.
	* Object-only section:
	  * Extract object only file.
	  * Add it to the output.
	  * Discard object-only section.
      * For each IR object file:
        * Compile and add IR sections to the output.
	* Discard non-IR sections.
      * For each non-IR object file:
	* Add non-IR sections to the output.
    * Non-IR link:
      * For each mixed object file:
	* Add non-IR sections to the output.
	* Discard IR sections and object-only section.
      * For each IR object file:
	* Add non-IR sections to the output.
	* Discard IR sections.
      * For each non-IR object file:
	* Add non-IR sections to the output.

This is useful for Linux kernel build with LTO.

bfd/

	PR ld/12291
	PR ld/12430
	PR ld/13298
	* bfd.c (bfd_lto_object_type): Add lto_mixed_object.
	(bfd): Add object_only_section.
	(bfd_group_signature): New.
	* elf.c (special_sections_g): Add .gnu_object_only.
	* format.c: Include "plugin-api.h" and "plugin.h" if
	BFD_SUPPORTS_PLUGINS is defined.
	(bfd_set_lto_type): Set type to lto_mixed_object for
	GNU_OBJECT_ONLY_SECTION_NAME section.
	(bfd_check_format_matches): Don't check the plugin target twice
	if the plugin target is explicitly specified.
	* opncls.c (bfd_extract_object_only_section): New.
	* plugin.c (bfd_plugin_fake_text_section): New.
	(bfd_plugin_fake_data_section): Likewise.
	(bfd_plugin_fake_bss_section): Likewise.
	(bfd_plugin_fake_common_section): Likewise.
	(bfd_plugin_get_symbols_in_object_only): Likewise.
	* plugin.c (add_symbols): Call
	bfd_plugin_get_symbols_in_object_only and count
	plugin_data->object_only_nsyms.
	(bfd_plugin_get_symtab_upper_bound): Count
	plugin_data->object_only_nsyms.
	bfd_plugin_get_symbols_in_object_only and add symbols from
	object only section.
	(bfd_plugin_canonicalize_symtab): Remove fake_section,
	fake_data_section, fake_bss_section and fake_common_section.
	Set udata.p to NULL.  Use bfd_plugin_fake_text_section,
	bfd_plugin_fake_data_section, bfd_plugin_fake_bss_section and
	bfd_plugin_fake_common_section.
	Set udata.p to NULL.
	* plugin.h (plugin_data_struct): Add object_only_nsyms and
	object_only_syms.
	* section.c (GNU_OBJECT_ONLY_SECTION_NAME): New.
	* bfd-in2.h: Regenerated.

binutils/

	PR ld/12291
	PR ld/12430
	PR ld/13298
	* objcopy.c (group_signature): Removed.
	(is_strip_section): Replace group_signature with
	bfd_group_signature.
	(setup_section): Likewise.
	* readelf.c (get_os_specific_section_type_name): Handle
	SHT_GNU_OBJECT_ONLY.

gas/

	PR ld/12291
	PR ld/12430
	PR ld/13298
	* testsuite/gas/elf/section9.s: Add the .gnu_object_only test.
	* testsuite/gas/elf/section9.d: Updated.

include/

	PR ld/12291
	PR ld/12430
	PR ld/13298
	* elf/common.h (SHT_GNU_OBJECT_ONLY): New.

ld/

	PR ld/12291
	PR ld/12430
	PR ld/13298
	* ld.h (ld_config_type): Add emit_gnu_object_only and
	emitting_gnu_object_only.
	* ldelf.c (orphan_init_done): Make it file scope.
	(ldelf_place_orphan): Rename hold to orig_hold.  Initialize hold
	from orig_hold at run-time.
	(ldelf_finish): New.
	* ldelf.h (ldelf_finish): New.
	* ldexp.c (ldexp_init): Take a bfd_boolean argument to supprt
	object-only output.
	(ldexp_finish): Likewise.
	* ldexp.h (ldexp_init): Take a bfd_boolean argument.
	(ldexp_finish): Likewise.
	* ldfile.c (ldfile_try_open_bfd): Call
	cmdline_check_object_only_section.
	* ldlang.c: Include "ldwrite.h" and elf-bfd.h.
	* ldlang.c (cmdline_object_only_file_list): New.
	(cmdline_object_only_archive_list): Likewise.
	(cmdline_temp_object_only_list): Likewise.
	(cmdline_lists_init): Likewise.
	(cmdline_list_new): Likewise.
	(cmdline_list_append): Likewise.
	(print_cmdline_list): Likewise.
	(cmdline_on_object_only_archive_list_p): Likewise.
	(cmdline_object_only_list_append): Likewise.
	(cmdline_get_object_only_input_files): Likewise.
	(cmdline_arg): Likewise.
	(setup_section): Likewise.
	(copy_section): Likewise.
	(cmdline_fopen_temp): Likewise.
	(cmdline_add_object_only_section): Likewise.
	(cmdline_emit_object_only_section): Likewise.
	(cmdline_extract_object_only_section): Likewise.
	(cmdline_check_object_only_section): Likewise.
	(cmdline_remove_object_only_files): Likewise.
	(lang_init): Take a bfd_boolean argument to supprt object-only
	output.  Call cmdline_lists_init.
	(load_symbols): Call cmdline_on_object_only_archive_list_p
	to check if an archive member should be loaded.
	(lang_process): Handle object-only link.
	* ldlang.h (lang_init): Take a bfd_boolean argument.
	(cmdline_enum_type): New.
	(cmdline_header_type): Likewise.
	(cmdline_file_type): Likewise.
	(cmdline_bfd_type): Likewise.
	(cmdline_union_type): Likewise.
	(cmdline_list_type): Likewise.
	(cmdline_emit_object_only_section): Likewise.
	(cmdline_check_object_only_section): Likewise.
	(cmdline_remove_object_only_files): Likewise.
	* ldmain.c (main): Call xatexit with
	cmdline_remove_object_only_files.  Pass FALSE to lang_init,
	ldexp_init and ldexp_finish.  Use ld_parse_linker_script.
	Set link_info.output_bfd to NULL after close.  Call
	cmdline_emit_object_only_section if needed.
	(add_archive_element): Call cmdline_check_object_only_section.
	(ld_parse_linker_script): New.
	* ldmain.h (ld_parse_linker_script): New.
	* plugin.c (plugin_maybe_claim): Call
	cmdline_check_object_only_section on claimed IR files.
	* scripttempl/elf.sc: Also discard .gnu_object_only sections.
	* scripttempl/elf64hppa.sc: Likewise.
	* scripttempl/elfxtensa.sc: Likewise.
	* scripttempl/mep.sc: Likewise.
	* scripttempl/pe.sc: Likewise.
	* scripttempl/pep.sc: Likewise.
	* emultempl/aarch64elf.em (gld${EMULATION_NAME}_finish): Replace
	finish_default with ldelf_finish.
	* emultempl/alphaelf.em (alpha_finish): Likewise.
	* emultempl/avrelf.em (avr_finish): Likewise.
	* emultempl/elf.em (ld_${EMULATION_NAME}_emulation): Likewise.
	* emultempl/ppc32elf.em (ppc_finish): Likewise.
	* emultempl/ppc64elf.em (gld${EMULATION_NAME}_finish): Likewise.
	* emultempl/spuelf.em (gld${EMULATION_NAME}_finish): Likewise.
	* testsuite/ld-plugin/lto-10.out: New file.
	* testsuite/ld-plugin/lto-10a.c: Likewise.
	* testsuite/ld-plugin/lto-10b.c: Likewise.
	* testsuite/ld-plugin/lto-10r.d: Likewise.
	* testsuite/ld-plugin/lto-4.out: Likewise.
	* testsuite/ld-plugin/lto-4a.c: Likewise.
	* testsuite/ld-plugin/lto-4b.c: Likewise.
	* testsuite/ld-plugin/lto-4c.c: Likewise.
	* testsuite/ld-plugin/lto-4r-a.d: Likewise.
	* testsuite/ld-plugin/lto-4r-b.d: Likewise.
	* testsuite/ld-plugin/lto-4r-c.d: Likewise.
	* testsuite/ld-plugin/lto-4r-d.d: Likewise.
	* testsuite/ld-plugin/lto.exp (lto_link_tests): Prepare for
	"LTO 4[acd]", "lto-4r-[abcd]" and "LTO 10" tests.
	(lto_run_tests): Add "LTO 4[acd]" and "LTO 10" tests.
	Build liblto-4.a.  Run "lto-4r-[abcd]" tests.
	Run lto-10r and create tmpdir/lto-10.o.
	Add test for nm on mixed LTO/non-LTO object.

Signed-off-by: H.J. Lu <hjl.tools@gmail.com>
2025-01-14 05:22:04 +08:00

547 lines
22 KiB
Python

# Copyright (C) 2014-2025 Free Software Foundation, Inc.
#
# Copying and distribution of this file, with or without modification,
# are permitted in any medium without royalty provided the copyright
# notice and this notice are preserved.
#
# Unusual variables checked by this code:
# NOP - four byte opcode for no-op (defaults to 0)
# NO_SMALL_DATA - no .sbss/.sbss2/.sdata/.sdata2 sections if not
# empty.
# SMALL_DATA_CTOR - .ctors contains small data.
# SMALL_DATA_DTOR - .dtors contains small data.
# DATA_ADDR - if end-of-text-plus-one-page isn't right for data start
# INITIAL_READONLY_SECTIONS - at start of text segment
# OTHER_READONLY_SECTIONS - other than .text .init .rodata ...
# (e.g., .PARISC.milli)
# OTHER_TEXT_SECTIONS - these get put in .text when relocating
# OTHER_READWRITE_SECTIONS - other than .data .bss .ctors .sdata ...
# (e.g., .PARISC.global)
# OTHER_RELRO_SECTIONS - other than .data.rel.ro ...
# (e.g. PPC32 .fixup, .got[12])
# OTHER_BSS_SECTIONS - other than .bss .sbss ...
# ATTRS_SECTIONS - at the end
# OTHER_SECTIONS - at the end
# EXECUTABLE_SYMBOLS - symbols that must be defined for an
# executable (e.g., _DYNAMIC_LINK)
# TEXT_START_ADDR - the first byte of the text segment, after any
# headers.
# TEXT_BASE_ADDRESS - the first byte of the text segment.
# TEXT_START_SYMBOLS - symbols that appear at the start of the
# .text section.
# DATA_START_SYMBOLS - symbols that appear at the start of the
# .data section.
# DATA_END_SYMBOLS - symbols that appear at the end of the
# writeable data sections.
# OTHER_GOT_SYMBOLS - symbols defined just before .got.
# OTHER_GOT_SECTIONS - sections just after .got.
# OTHER_SDATA_SECTIONS - sections just after .sdata.
# OTHER_BSS_SYMBOLS - symbols that appear at the start of the
# .bss section besides __bss_start.
# DATA_PLT - .plt should be in data segment, not text segment.
# PLT_BEFORE_GOT - .plt just before .got when .plt is in data segement.
# BSS_PLT - .plt should be in bss segment
# NO_REL_RELOCS - Don't include .rel.* sections in script
# NO_RELA_RELOCS - Don't include .rela.* sections in script
# NON_ALLOC_DYN - Place dynamic sections after data segment.
# TEXT_DYNAMIC - .dynamic in text segment, not data segment.
# EMBEDDED - whether this is for an embedded system.
# SHLIB_TEXT_START_ADDR - if set, add to SIZEOF_HEADERS to set
# start address of shared library.
# INPUT_FILES - INPUT command of files to always include
# WRITABLE_RODATA - if set, the .rodata section should be writable
# INIT_START, INIT_END - statements just before and just after
# combination of .init sections.
# FINI_START, FINI_END - statements just before and just after
# combination of .fini sections.
# STACK_ADDR - start of a .stack section.
# OTHER_SYMBOLS - symbols to place right at the end of the script.
# ETEXT_NAME - name of a symbol for the end of the text section,
# normally etext.
# SEPARATE_GOTPLT - if set, .got.plt should be separate output section,
# so that .got can be in the RELRO area. It should be set to
# the number of bytes in the beginning of .got.plt which can be
# in the RELRO area as well.
# USER_LABEL_PREFIX - prefix to add to user-visible symbols.
#
# When adding sections, do note that the names of some sections are used
# when specifying the start address of the next.
#
# Many sections come in three flavours. There is the 'real' section,
# like ".data". Then there are the per-procedure or per-variable
# sections, generated by -ffunction-sections and -fdata-sections in GCC,
# and useful for --gc-sections, which for a variable "foo" might be
# ".data.foo". Then there are the linkonce sections, for which the linker
# eliminates duplicates, which are named like ".gnu.linkonce.d.foo".
# The exact correspondences are:
#
# Section Linkonce section
# .text .gnu.linkonce.t.foo
# .rodata .gnu.linkonce.r.foo
# .data .gnu.linkonce.d.foo
# .bss .gnu.linkonce.b.foo
# .sdata .gnu.linkonce.s.foo
# .sbss .gnu.linkonce.sb.foo
# .sdata2 .gnu.linkonce.s2.foo
# .sbss2 .gnu.linkonce.sb2.foo
# .debug_info .gnu.linkonce.wi.foo
# .tdata .gnu.linkonce.td.foo
# .tbss .gnu.linkonce.tb.foo
# .lrodata .gnu.linkonce.lr.foo
# .ldata .gnu.linkonce.l.foo
# .lbss .gnu.linkonce.lb.foo
#
# Each of these can also have corresponding .rel.* and .rela.* sections.
test -z "$ENTRY" && ENTRY=_start
test -z "${BIG_OUTPUT_FORMAT}" && BIG_OUTPUT_FORMAT=${OUTPUT_FORMAT}
test -z "${LITTLE_OUTPUT_FORMAT}" && LITTLE_OUTPUT_FORMAT=${OUTPUT_FORMAT}
if [ -z "$MACHINE" ]; then OUTPUT_ARCH=${ARCH}; else OUTPUT_ARCH=${ARCH}:${MACHINE}; fi
test -z "${ELFSIZE}" && ELFSIZE=32
test -z "${ALIGNMENT}" && ALIGNMENT="${ELFSIZE} / 8"
test "$LD_FLAG" = "N" && DATA_ADDR=.
test -z "${ETEXT_NAME}" && ETEXT_NAME=etext
test -n "$CREATE_SHLIB$CREATE_PIE" && test -n "$SHLIB_DATA_ADDR" && COMMONPAGESIZE=""
test -z "$CREATE_SHLIB$CREATE_PIE" && test -n "$DATA_ADDR" && COMMONPAGESIZE=""
test -n "$RELRO_NOW" && unset SEPARATE_GOTPLT
test -z "$ATTRS_SECTIONS" && ATTRS_SECTIONS=".gnu.attributes 0 : { KEEP (*(.gnu.attributes)) }"
DATA_SEGMENT_ALIGN="ALIGN(${SEGMENT_SIZE}) + (. & (${MAXPAGESIZE} - 1))"
DATA_SEGMENT_RELRO_END=""
DATA_SEGMENT_END=""
if test -n "${COMMONPAGESIZE}"; then
DATA_SEGMENT_ALIGN="ALIGN (${SEGMENT_SIZE}) - ((${MAXPAGESIZE} - .) & (${MAXPAGESIZE} - 1)); . = DATA_SEGMENT_ALIGN (${MAXPAGESIZE}, ${COMMONPAGESIZE})"
DATA_SEGMENT_END=". = DATA_SEGMENT_END (.);"
DATA_SEGMENT_RELRO_END=". = DATA_SEGMENT_RELRO_END (${SEPARATE_GOTPLT-0}, .);"
fi
if test -z "${INITIAL_READONLY_SECTIONS}${CREATE_SHLIB}"; then
INITIAL_READONLY_SECTIONS=".interp ${RELOCATING-0} : { *(.interp) }"
fi
if test -z "$PLT"; then
PLT=".plt ${RELOCATING-0} : { *(.plt) }"
fi
test -n "${DATA_PLT-${BSS_PLT-text}}" && TEXT_PLT=yes
if test -z "$GOT"; then
if test -z "$SEPARATE_GOTPLT"; then
GOT=".got ${RELOCATING-0} : {${RELOCATING+ *(.got.plt)} *(.got) }"
else
GOT=".got ${RELOCATING-0} : { *(.got) }"
GOTPLT=".got.plt ${RELOCATING-0} : { *(.got.plt) }"
fi
fi
DYNAMIC=".dynamic ${RELOCATING-0} : { *(.dynamic) }"
RODATA=".rodata ${RELOCATING-0} : { *(.rodata${RELOCATING+ .rodata.* .gnu.linkonce.r.*}) }"
DATARELRO=".data.rel.ro : { *(.data.rel.ro.local* .gnu.linkonce.d.rel.ro.local.*) *(.data.rel.ro .data.rel.ro.* .gnu.linkonce.d.rel.ro.*) }"
DISCARDED="/DISCARD/ : { *(.note.GNU-stack) *(.gnu_debuglink) *(.gnu.lto_*) *(.gnu_object_only) }"
if test -z "${NO_SMALL_DATA}"; then
SBSS=".sbss ${RELOCATING-0} :
{
${RELOCATING+${SBSS_START_SYMBOLS}}
${CREATE_SHLIB+*(.sbss2 .sbss2.* .gnu.linkonce.sb2.*)}
${RELOCATING+*(.dynsbss)}
*(.sbss${RELOCATING+ .sbss.* .gnu.linkonce.sb.*})
${RELOCATING+*(.scommon)}
${RELOCATING+${SBSS_END_SYMBOLS}}
}"
SBSS2=".sbss2 ${RELOCATING-0} : { *(.sbss2${RELOCATING+ .sbss2.* .gnu.linkonce.sb2.*}) }"
SDATA="/* We want the small data sections together, so single-instruction offsets
can access them all, and initialized data all before uninitialized, so
we can shorten the on-disk segment size. */
.sdata ${RELOCATING-0} :
{
${RELOCATING+${SDATA_START_SYMBOLS}}
${CREATE_SHLIB+*(.sdata2 .sdata2.* .gnu.linkonce.s2.*)}
*(.sdata${RELOCATING+ .sdata.* .gnu.linkonce.s.*})
}"
SDATA2=".sdata2 ${RELOCATING-0} :
{
${RELOCATING+${SDATA2_START_SYMBOLS}}
*(.sdata2${RELOCATING+ .sdata2.* .gnu.linkonce.s2.*})
}"
REL_SDATA=".rel.sdata ${RELOCATING-0} : { *(.rel.sdata${RELOCATING+ .rel.sdata.* .rel.gnu.linkonce.s.*}) }
.rela.sdata ${RELOCATING-0} : { *(.rela.sdata${RELOCATING+ .rela.sdata.* .rela.gnu.linkonce.s.*}) }"
REL_SBSS=".rel.sbss ${RELOCATING-0} : { *(.rel.sbss${RELOCATING+ .rel.sbss.* .rel.gnu.linkonce.sb.*}) }
.rela.sbss ${RELOCATING-0} : { *(.rela.sbss${RELOCATING+ .rela.sbss.* .rela.gnu.linkonce.sb.*}) }"
REL_SDATA2=".rel.sdata2 ${RELOCATING-0} : { *(.rel.sdata2${RELOCATING+ .rel.sdata2.* .rel.gnu.linkonce.s2.*}) }
.rela.sdata2 ${RELOCATING-0} : { *(.rela.sdata2${RELOCATING+ .rela.sdata2.* .rela.gnu.linkonce.s2.*}) }"
REL_SBSS2=".rel.sbss2 ${RELOCATING-0} : { *(.rel.sbss2${RELOCATING+ .rel.sbss2.* .rel.gnu.linkonce.sb2.*}) }
.rela.sbss2 ${RELOCATING-0} : { *(.rela.sbss2${RELOCATING+ .rela.sbss2.* .rela.gnu.linkonce.sb2.*}) }"
else
NO_SMALL_DATA=" "
fi
if test -z "${DATA_GOT}"; then
if test -n "${NO_SMALL_DATA}"; then
DATA_GOT=" "
fi
fi
if test -z "${SDATA_GOT}"; then
if test -z "${NO_SMALL_DATA}"; then
SDATA_GOT=" "
fi
fi
test -n "$SEPARATE_GOTPLT" && SEPARATE_GOTPLT=" "
test "${LARGE_SECTIONS}" = "yes" && REL_LARGE="
.rel.ldata ${RELOCATING-0} : { *(.rel.ldata${RELOCATING+ .rel.ldata.* .rel.gnu.linkonce.l.*}) }
.rela.ldata ${RELOCATING-0} : { *(.rela.ldata${RELOCATING+ .rela.ldata.* .rela.gnu.linkonce.l.*}) }
.rel.lbss ${RELOCATING-0} : { *(.rel.lbss${RELOCATING+ .rel.lbss.* .rel.gnu.linkonce.lb.*}) }
.rela.lbss ${RELOCATING-0} : { *(.rela.lbss${RELOCATING+ .rela.lbss.* .rela.gnu.linkonce.lb.*}) }
.rel.lrodata ${RELOCATING-0} : { *(.rel.lrodata${RELOCATING+ .rel.lrodata.* .rel.gnu.linkonce.lr.*}) }
.rela.lrodata ${RELOCATING-0} : { *(.rela.lrodata${RELOCATING+ .rela.lrodata.* .rela.gnu.linkonce.lr.*}) }"
test "${LARGE_SECTIONS}" = "yes" && OTHER_BSS_SECTIONS="
${OTHER_BSS_SECTIONS}
.lbss ${RELOCATING-0} :
{
*(.dynlbss)
*(.lbss${RELOCATING+ .lbss.* .gnu.linkonce.lb.*})
*(LARGE_COMMON)
}"
test "${LARGE_SECTIONS}" = "yes" && LARGE_SECTIONS="
.lrodata ${RELOCATING-0} ${RELOCATING+ALIGN(${MAXPAGESIZE}) + (. & (${MAXPAGESIZE} - 1))} :
{
*(.lrodata${RELOCATING+ .lrodata.* .gnu.linkonce.lr.*})
}
.ldata ${RELOCATING-0} ${RELOCATING+ALIGN(${MAXPAGESIZE}) + (. & (${MAXPAGESIZE} - 1))} :
{
*(.ldata${RELOCATING+ .ldata.* .gnu.linkonce.l.*})
${RELOCATING+. = ALIGN(. != 0 ? ${ALIGNMENT} : 1);}
}"
INIT_ARRAY=".init_array ${RELOCATING-0} :
{
${RELOCATING+${CREATE_SHLIB-PROVIDE_HIDDEN (${USER_LABEL_PREFIX}__init_array_start = .);}}
KEEP (*(SORT(.init_array.*)))
KEEP (*(.init_array))
${RELOCATING+${CREATE_SHLIB-PROVIDE_HIDDEN (${USER_LABEL_PREFIX}__init_array_end = .);}}
}"
FINI_ARRAY=".fini_array ${RELOCATING-0} :
{
${RELOCATING+${CREATE_SHLIB-PROVIDE_HIDDEN (${USER_LABEL_PREFIX}__fini_array_start = .);}}
KEEP (*(SORT(.fini_array.*)))
KEEP (*(.fini_array))
${RELOCATING+${CREATE_SHLIB-PROVIDE_HIDDEN (${USER_LABEL_PREFIX}__fini_array_end = .);}}
}"
CTOR=".ctors ${CONSTRUCTING-0} :
{
${CONSTRUCTING+${CTOR_START}}
/* gcc uses crtbegin.o to find the start of
the constructors, so we make sure it is
first. Because this is a wildcard, it
doesn't matter if the user does not
actually link against crtbegin.o; the
linker won't look for a file to match a
wildcard. The wildcard also means that it
doesn't matter which directory crtbegin.o
is in. */
KEEP (*crtbegin.o(.ctors))
KEEP (*crtbegin?.o(.ctors))
/* We don't want to include the .ctor section from
the crtend.o file until after the sorted ctors.
The .ctor section from the crtend file contains the
end of ctors marker and it must be last */
KEEP (*(EXCLUDE_FILE (*crtend.o *crtend?.o $OTHER_EXCLUDE_FILES) .ctors))
KEEP (*(SORT(.ctors.*)))
KEEP (*(.ctors))
${CONSTRUCTING+${CTOR_END}}
}"
DTOR=".dtors ${CONSTRUCTING-0} :
{
${CONSTRUCTING+${DTOR_START}}
KEEP (*crtbegin.o(.dtors))
KEEP (*crtbegin?.o(.dtors))
KEEP (*(EXCLUDE_FILE (*crtend.o *crtend?.o $OTHER_EXCLUDE_FILES) .dtors))
KEEP (*(SORT(.dtors.*)))
KEEP (*(.dtors))
${CONSTRUCTING+${DTOR_END}}
}"
STACK=".stack ${RELOCATING-0}${RELOCATING+${STACK_ADDR}} :
{
${RELOCATING+${USER_LABEL_PREFIX}_stack = .;}
*(.stack)
${RELOCATING+${STACK_SENTINEL}}
}"
TEXT_START_ADDR="SEGMENT_START(\"text-segment\", ${TEXT_START_ADDR})"
SHLIB_TEXT_START_ADDR="SEGMENT_START(\"text-segment\", ${SHLIB_TEXT_START_ADDR:-0})"
# If this is for an embedded system, don't add SIZEOF_HEADERS.
if [ -z "$EMBEDDED" ]; then
test -z "${TEXT_BASE_ADDRESS}" && TEXT_BASE_ADDRESS="${TEXT_START_ADDR} + SIZEOF_HEADERS"
else
test -z "${TEXT_BASE_ADDRESS}" && TEXT_BASE_ADDRESS="${TEXT_START_ADDR}"
fi
cat <<EOF
/* Copyright (C) 2014-2025 Free Software Foundation, Inc.
Copying and distribution of this script, with or without modification,
are permitted in any medium without royalty provided the copyright
notice and this notice are preserved. */
OUTPUT_FORMAT("${OUTPUT_FORMAT}", "${BIG_OUTPUT_FORMAT}",
"${LITTLE_OUTPUT_FORMAT}")
OUTPUT_ARCH(${OUTPUT_ARCH})
${RELOCATING+ENTRY(${ENTRY})}
${RELOCATING+${LIB_SEARCH_DIRS}}
${RELOCATING+${EXECUTABLE_SYMBOLS}}
${RELOCATING+${INPUT_FILES}}
${RELOCATING- /* For some reason, the Solaris linker makes bad executables
if gld -r is used and the intermediate file has sections starting
at non-zero addresses. Could be a Solaris ld bug, could be a GNU ld
bug. But for now assigning the zero vmas works. */}
SECTIONS
{
/* Read-only sections, merged into text segment: */
${CREATE_SHLIB-${CREATE_PIE-${RELOCATING+PROVIDE (__executable_start = ${TEXT_START_ADDR}); . = ${TEXT_BASE_ADDRESS};}}}
${CREATE_SHLIB+${RELOCATING+. = ${SHLIB_TEXT_START_ADDR} + SIZEOF_HEADERS;}}
${CREATE_PIE+${RELOCATING+. = ${SHLIB_TEXT_START_ADDR} + SIZEOF_HEADERS;}}
${INITIAL_READONLY_SECTIONS}
.note.gnu.build-id ${RELOCATING-0}: { *(.note.gnu.build-id) }
EOF
test -n "${RELOCATING+0}" || unset NON_ALLOC_DYN
test -z "${NON_ALLOC_DYN}" || TEXT_DYNAMIC=
cat > ldscripts/dyntmp.$$ <<EOF
${TEXT_DYNAMIC+${DYNAMIC}}
.hash ${RELOCATING-0} : { *(.hash) }
.gnu.hash ${RELOCATING-0} : { *(.gnu.hash) }
.dynsym ${RELOCATING-0} : { *(.dynsym) }
.dynstr ${RELOCATING-0} : { *(.dynstr) }
.gnu.version ${RELOCATING-0} : { *(.gnu.version) }
.gnu.version_d ${RELOCATING-0}: { *(.gnu.version_d) }
.gnu.version_r ${RELOCATING-0}: { *(.gnu.version_r) }
EOF
if [ "x$COMBRELOC" = x ]; then
COMBRELOCCAT="cat >> ldscripts/dyntmp.$$"
else
COMBRELOCCAT="cat > $COMBRELOC"
fi
eval $COMBRELOCCAT <<EOF
.rel.init ${RELOCATING-0} : { *(.rel.init) }
.rela.init ${RELOCATING-0} : { *(.rela.init) }
.rel.text ${RELOCATING-0} : { *(.rel.text${RELOCATING+ .rel.text.* .rel.gnu.linkonce.t.*}) }
.rela.text ${RELOCATING-0} : { *(.rela.text${RELOCATING+ .rela.text.* .rela.gnu.linkonce.t.*}) }
.rel.fini ${RELOCATING-0} : { *(.rel.fini) }
.rela.fini ${RELOCATING-0} : { *(.rela.fini) }
.rel.rodata ${RELOCATING-0} : { *(.rel.rodata${RELOCATING+ .rel.rodata.* .rel.gnu.linkonce.r.*}) }
.rela.rodata ${RELOCATING-0} : { *(.rela.rodata${RELOCATING+ .rela.rodata.* .rela.gnu.linkonce.r.*}) }
${OTHER_READONLY_RELOC_SECTIONS}
.rel.data.rel.ro ${RELOCATING-0} : { *(.rel.data.rel.ro${RELOCATING+ .rel.data.rel.ro.* .rel.gnu.linkonce.d.rel.ro.*}) }
.rela.data.rel.ro ${RELOCATING-0} : { *(.rela.data.rel.ro${RELOCATING+ .rela.data.rel.ro.* .rela.gnu.linkonce.d.rel.ro.*}) }
.rel.data ${RELOCATING-0} : { *(.rel.data${RELOCATING+ .rel.data.* .rel.gnu.linkonce.d.*}) }
.rela.data ${RELOCATING-0} : { *(.rela.data${RELOCATING+ .rela.data.* .rela.gnu.linkonce.d.*}) }
.rel.tdata ${RELOCATING-0} : { *(.rel.tdata${RELOCATING+ .rel.tdata.* .rel.gnu.linkonce.td.*}) }
.rela.tdata ${RELOCATING-0} : { *(.rela.tdata${RELOCATING+ .rela.tdata.* .rela.gnu.linkonce.td.*}) }
.rel.tbss ${RELOCATING-0} : { *(.rel.tbss${RELOCATING+ .rel.tbss.* .rel.gnu.linkonce.tb.*}) }
.rela.tbss ${RELOCATING-0} : { *(.rela.tbss${RELOCATING+ .rela.tbss.* .rela.gnu.linkonce.tb.*}) }
.rel.ctors ${RELOCATING-0} : { *(.rel.ctors) }
.rela.ctors ${RELOCATING-0} : { *(.rela.ctors) }
.rel.dtors ${RELOCATING-0} : { *(.rel.dtors) }
.rela.dtors ${RELOCATING-0} : { *(.rela.dtors) }
.rel.got ${RELOCATING-0} : { *(.rel.got) }
.rela.got ${RELOCATING-0} : { *(.rela.got) }
${OTHER_GOT_RELOC_SECTIONS}
${REL_SDATA}
${REL_SBSS}
${REL_SDATA2}
${REL_SBSS2}
.rel.bss ${RELOCATING-0} : { *(.rel.bss${RELOCATING+ .rel.bss.* .rel.gnu.linkonce.b.*}) }
.rela.bss ${RELOCATING-0} : { *(.rela.bss${RELOCATING+ .rela.bss.* .rela.gnu.linkonce.b.*}) }
${REL_LARGE}
EOF
if [ -n "$COMBRELOC" ]; then
cat >> ldscripts/dyntmp.$$ <<EOF
.rel.dyn ${RELOCATING-0} :
{
EOF
sed -e '/^[ ]*[{}][ ]*$/d;/:[ ]*$/d;/\.rela\./d;s/^.*: { *\(.*\)}$/ \1/' $COMBRELOC >> ldscripts/dyntmp.$$
cat >> ldscripts/dyntmp.$$ <<EOF
}
.rel.ifunc.dyn ${RELOCATING-0} :
{
*(.rel.ifunc.*)
}
.rela.dyn ${RELOCATING-0} :
{
EOF
sed -e '/^[ ]*[{}][ ]*$/d;/:[ ]*$/d;/\.rel\./d;s/^.*: { *\(.*\)}/ \1/' $COMBRELOC >> ldscripts/dyntmp.$$
cat >> ldscripts/dyntmp.$$ <<EOF
}
.rela.ifunc.dyn ${RELOCATING-0} :
{
*(.rela.ifunc.*)
}
EOF
fi
cat >> ldscripts/dyntmp.$$ <<EOF
.rel.plt ${RELOCATING-0} : { *(.rel.plt) }
.rela.plt ${RELOCATING-0} : { *(.rela.plt) }
${OTHER_PLT_RELOC_SECTIONS}
EOF
if test -z "${NON_ALLOC_DYN}"; then
if test -z "${NO_REL_RELOCS}${NO_RELA_RELOCS}"; then
cat ldscripts/dyntmp.$$
else
if test -z "${NO_REL_RELOCS}"; then
sed -e '/^[ ]*\.rela\.[^}]*$/,/}/d' -e '/^[ ]*\.rela\./d' ldscripts/dyntmp.$$
fi
if test -z "${NO_RELA_RELOCS}"; then
sed -e '/^[ ]*\.rel\.[^}]*$/,/}/d' -e '/^[ ]*\.rel\./d' ldscripts/dyntmp.$$
fi
fi
rm -f ldscripts/dyntmp.$$
fi
cat <<EOF
.init ${RELOCATING-0} :
{
${RELOCATING+${INIT_START}}
KEEP (*(SORT_NONE(.init)))
${RELOCATING+${INIT_END}}
} =${NOP-0}
${TEXT_PLT+${PLT}}
${TINY_READONLY_SECTION}
.text ${RELOCATING-0} :
{
${RELOCATING+${TEXT_START_SYMBOLS}}
*(.text .stub${RELOCATING+ .text.* .gnu.linkonce.t.*})
/* .gnu.warning sections are handled specially by elf.em. */
*(.gnu.warning)
${RELOCATING+${OTHER_TEXT_SECTIONS}}
} =${NOP-0}
.fini ${RELOCATING-0} :
{
${RELOCATING+${FINI_START}}
KEEP (*(SORT_NONE(.fini)))
${RELOCATING+${FINI_END}}
} =${NOP-0}
${RELOCATING+PROVIDE (__${ETEXT_NAME} = .);}
${RELOCATING+PROVIDE (_${ETEXT_NAME} = .);}
${RELOCATING+PROVIDE (${ETEXT_NAME} = .);}
${WRITABLE_RODATA-${RODATA}}
.rodata1 ${RELOCATING-0} : { *(.rodata1) }
${CREATE_SHLIB-${SDATA2}}
${CREATE_SHLIB-${SBSS2}}
${OTHER_READONLY_SECTIONS}
.eh_frame_hdr ${RELOCATING-0} : { *(.eh_frame_hdr) }
.eh_frame ${RELOCATING-0} : ONLY_IF_RO { KEEP (*(.eh_frame)) }
.gcc_except_table ${RELOCATING-0} : ONLY_IF_RO { *(.gcc_except_table${RELOCATING+ .gcc_except_table.*}) }
/* Adjust the address for the data segment. We want to adjust up to
the same address within the page on the next page up. */
${CREATE_SHLIB-${CREATE_PIE-${RELOCATING+. = ${DATA_ADDR-${DATA_SEGMENT_ALIGN}};}}}
${CREATE_SHLIB+${RELOCATING+. = ${SHLIB_DATA_ADDR-${DATA_SEGMENT_ALIGN}};}}
${CREATE_PIE+${RELOCATING+. = ${SHLIB_DATA_ADDR-${DATA_SEGMENT_ALIGN}};}}
/* Exception handling */
.eh_frame ${RELOCATING-0} : ONLY_IF_RW { KEEP (*(.eh_frame)) }
.gcc_except_table ${RELOCATING-0} : ONLY_IF_RW { *(.gcc_except_table${RELOCATING+ .gcc_except_table.*}) }
/* Thread Local Storage sections */
.tdata ${RELOCATING-0} : { *(.tdata${RELOCATING+ .tdata.* .gnu.linkonce.td.*}) }
.preinit_array ${RELOCATING-0} :
{
${RELOCATING+${CREATE_SHLIB-PROVIDE_HIDDEN (${USER_LABEL_PREFIX}__preinit_array_start = .);}}
KEEP (*(.preinit_array))
${RELOCATING+${CREATE_SHLIB-PROVIDE_HIDDEN (${USER_LABEL_PREFIX}__preinit_array_end = .);}}
}
${RELOCATING+${INIT_ARRAY}}
${RELOCATING+${FINI_ARRAY}}
${SMALL_DATA_CTOR-${RELOCATING+${CTOR}}}
${SMALL_DATA_DTOR-${RELOCATING+${DTOR}}}
.jcr ${RELOCATING-0} : { KEEP (*(.jcr)) }
${RELOCATING+${DATARELRO}}
${OTHER_RELRO_SECTIONS}
${TEXT_DYNAMIC-${DYNAMIC}}
${DATA_GOT+${RELRO_NOW+${GOT}}}
${DATA_GOT+${RELRO_NOW+${GOTPLT}}}
${DATA_GOT+${RELRO_NOW-${SEPARATE_GOTPLT+${GOT}}}}
${RELOCATING+${DATA_SEGMENT_RELRO_END}}
${DATA_GOT+${RELRO_NOW-${SEPARATE_GOTPLT-${GOT}}}}
${DATA_GOT+${RELRO_NOW-${GOTPLT}}}
${DATA_PLT+${PLT_BEFORE_GOT-${PLT}}}
.data ${RELOCATING-0} :
{
${RELOCATING+${DATA_START_SYMBOLS}}
*(.data${RELOCATING+ .data.* .gnu.linkonce.d.*})
${CONSTRUCTING+SORT(CONSTRUCTORS)}
}
.data1 ${RELOCATING-0} : { *(.data1) }
${WRITABLE_RODATA+${RODATA}}
${OTHER_READWRITE_SECTIONS}
${SMALL_DATA_CTOR+${RELOCATING+${CTOR}}}
${SMALL_DATA_DTOR+${RELOCATING+${DTOR}}}
${DATA_PLT+${PLT_BEFORE_GOT+${PLT}}}
${SDATA_GOT+${RELOCATING+${OTHER_GOT_SYMBOLS}}}
${SDATA_GOT+${GOT}}
${SDATA_GOT+${OTHER_GOT_SECTIONS}}
${SDATA}
${OTHER_SDATA_SECTIONS}
${RELOCATING+${DATA_END_SYMBOLS-${USER_LABEL_PREFIX}_edata = .; PROVIDE (${USER_LABEL_PREFIX}edata = .);}}
${RELOCATING+. = ALIGN(ALIGNOF(NEXT_SECTION));}
${RELOCATING+__bss_start = .;}
${RELOCATING+${OTHER_BSS_SYMBOLS}}
${SBSS}
${BSS_PLT+${PLT}}
.bss ${RELOCATING-0} :
{
${RELOCATING+*(.dynbss)}
*(.bss${RELOCATING+ .bss.* .gnu.linkonce.b.*})
${RELOCATING+*(COMMON)
/* Align here to ensure that the .bss section occupies space up to
_end. Align after .bss to ensure correct alignment even if the
.bss section disappears because there are no input sections.
FIXME: Why do we need it? When there is no .bss section, we do not
pad the .data section. */
. = ALIGN(. != 0 ? ${ALIGNMENT} : 1);}
}
${OTHER_BSS_SECTIONS}
${RELOCATING+${OTHER_BSS_END_SYMBOLS}}
${RELOCATING+. = ALIGN(${ALIGNMENT});}
${LARGE_SECTIONS}
${RELOCATING+. = ALIGN(${ALIGNMENT});}
${RELOCATING+${OTHER_END_SYMBOLS}}
${RELOCATING+${END_SYMBOLS-${USER_LABEL_PREFIX}_end = .; PROVIDE (${USER_LABEL_PREFIX}end = .);}}
${RELOCATING+${DATA_SEGMENT_END}}
${TINY_DATA_SECTION}
${TINY_BSS_SECTION}
${STACK_ADDR+${STACK}}
EOF
if test -n "${NON_ALLOC_DYN}"; then
if test -z "${NO_REL_RELOCS}${NO_RELA_RELOCS}"; then
cat ldscripts/dyntmp.$$
else
if test -z "${NO_REL_RELOCS}"; then
sed -e '/^[ ]*\.rela\.[^}]*$/,/}/d' -e '/^[ ]*\.rela\./d' ldscripts/dyntmp.$$
fi
if test -z "${NO_RELA_RELOCS}"; then
sed -e '/^[ ]*\.rel\.[^}]*$/,/}/d' -e '/^[ ]*\.rel\./d' ldscripts/dyntmp.$$
fi
fi
rm -f ldscripts/dyntmp.$$
fi
source_sh $srcdir/scripttempl/misc-sections.sc
source_sh $srcdir/scripttempl/DWARF.sc
cat <<EOF
${ATTRS_SECTIONS}
${OTHER_SECTIONS}
${RELOCATING+${OTHER_SYMBOLS}}
${RELOCATING+${DISCARDED}}
}
EOF