2e702c99c5
* configure: Regenerate. gold/ * nacl.cc: New file. * nacl.h: New file. * Makefile.am (CCFILES, HFILES): Add them. * Makefile.in: Regenerate. * i386.cc (Output_data_plt_i386_nacl): New class. (Output_data_plt_i386_nacl_exec): New class. (Output_data_plt_i386_nacl_dyn): New class. (Target_i386_nacl): New class. (Target_selector_i386_nacl): New class. (target_selector_i386): Use it instead of Target_selector_i386. * x86_64.cc (Output_data_plt_x86_64_nacl): New class. (Target_x86_64_nacl): New class. (Target_selector_x86_64_nacl): New class. (target_selector_x86_64, target_selector_x32): Use it instead of Target_selector_x86_64. * arm.cc (Output_data_plt_arm_nacl): New class. (Target_arm_nacl): New class. (Target_selector_arm_nacl): New class. (target_selector_arm, target_selector_armbe): Use it instead of Target_selector_arm. * target-select.cc (select_target): Take new Input_file* and off_t arguments, pass them on to recognize method of selector. * object.cc (make_elf_sized_object): Update caller. * parameters.cc (parameters_force_valid_target): Likewise. * incremental.cc (make_sized_incremental_binary): Likewise. * target-select.h: Update decl. (Target_selector::recognize): Take new Input_file* argument, pass it on to do_recognize. (Target_selector::do_recognize): Take new Input_file* argument. * freebsd.h (Target_selector_freebsd::do_recognize): Likewise. * powerpc.cc (Target_selector_powerpc::do_recognize): Likewise. * sparc.cc (Target_selector_sparc::do_recognize): Likewise. * testsuite/testfile.cc (Target_selector::do_recognize): Likewise. * target.h (Target::Target_info): New members isolate_execinstr and rosegment_gap. (Target::isolate_execinstr, Target::rosegment_gap): New methods. * arm.cc (Target_arm::arm_info): Update initializer. * i386.cc (Target_i386::i386_info): Likewise. * powerpc.cc (Target_powerpc::powerpc_info): Likewise. * sparc.cc (Target_sparc::sparc_info): Likewise. * x86_64.cc (Target_x86_64::x86_64_info): Likewise. * testsuite/testfile.cc (Target_test::test_target_info): Likewise. * layout.cc (Layout::attach_allocated_section_to_segment): Take new const Target* argument. If target->isolate_execinstr(), act like --rosegment. (Layout::find_first_load_seg): Take new const Target* argument; if target->isolate_execinstr(), reject PF_X segments. (Layout::relaxation_loop_body): Update caller. (Layout::set_segment_offsets): If target->isolate_execinstr(), reset file offset to zero when we hit LOAD_SEG, and then do a second loop over the segments before LOAD_SEG to reassign offsets after addresses have been determined. Handle target->rosegment_gap(). (Layout::attach_section_to_segment): Take new const Target* argument; pass it to attach_allocated_section_to_segment. (Layout::make_output_section): Update caller. (Layout::attach_sections_to_segments): Take new const Target* argument; pass it to attach_section_to_segment. * gold.cc (queue_middle_tasks): Update caller. * layout.h (Layout): Update method decls with new arguments. * arm.cc (Target_arm::Target_arm): Take optional argument for the Target_info pointer to use. (Target_arm::do_make_data_plt): New virtual method. (Target_arm::make_data_plt): New method that calls it. (Target_arm::make_plt_entry): Use it. (Output_data_plt_arm::Output_data_plt_arm): Take additional argument for the section alignment. (Output_data_plt_arm::do_first_plt_entry_offset): New abstract virtual method. (Output_data_plt_arm::first_plt_entry_offset): Call it. (Output_data_plt_arm::do_get_plt_entry_size): New abstract virtual method. (Output_data_plt_arm::get_plt_entry_size): Call it. (Output_data_plt_arm::do_fill_plt_entry): New abstract virtual method. (Output_data_plt_arm::fill_plt_entry): New method that calls it. (Output_data_plt_arm::do_fill_first_plt_entry): New abstract virtual method. (Output_data_plt_arm::fill_first_plt_entry): New method that calls it. (Output_data_plt_arm::set_final_data_size): Use get_plt_entry_size method instead of sizeof(plt_entry). (Output_data_plt_arm::add_entry): Likewise. Use first_plt_entry_offset method instead of sizeof(first_plt_entry). (Target_arm::first_plt_entry_offset): Call method on this->plt_ rather than static method. (Target_arm::plt_entry_size): Likewise. (Output_data_plt_arm::first_plt_entry, Output_data_plt_arm::plt_entry): Move to ... (Output_data_plt_arm_standard): ... here, new class. (Output_data_plt_arm::do_write): Move guts of PLT filling to... (Output_data_plt_arm_standard::do_fill_first_plt_entry): ... here ... (Output_data_plt_arm_standard::do_fill_plt_entry): ... and here. * x86_64.cc (Output_data_plt_x86_64::Output_data_plt_x86_64): Take additional argument for the PLT entry size. (Output_data_plt_x86_64::get_tlsdesc_plt_offset): Use get_plt_entry_size method rather than plt_entry_size variable. (Output_data_plt_x86_64::reserve_slot): Likewise. (Output_data_plt_x86_64::do_adjust_output_section): Likewise. (Output_data_plt_x86_64::add_entry): Likewise. (Output_data_plt_x86_64::add_local_ifunc_entry): Likewise. (Output_data_plt_x86_64::address_for_global): Likewise. (Output_data_plt_x86_64::address_for_local): Likewise. (Output_data_plt_x86_64::set_final_data_size): Likewise. (Output_data_plt_x86_64::first_plt_entry_offset): Likewise. Make method non-static. (Output_data_plt_x86_64::do_get_plt_entry_size): New abstract virtual method. (Output_data_plt_x86_64::get_plt_entry_size): Just call that. (Output_data_plt_x86_64::do_add_eh_frame): New abstract virtual method. (Output_data_plt_x86_64::add_eh_frame): New method to call it. (Output_data_plt_x86_64::do_fill_first_plt_entry): New abstract virtual method. (Output_data_plt_x86_64::fill_first_plt_entry): New method to call it. (Output_data_plt_x86_64::do_fill_plt_entry): New abstract virtual method. (Output_data_plt_x86_64::fill_plt_entry): New method to call it. (Output_data_plt_x86_64::do_fill_tlsdesc_entry): New abstract virtual method. (Output_data_plt_x86_64::fill_tlsdesc_entry): New method to call it. (Output_data_plt_x86_64::plt_entry_size) (Output_data_plt_x86_64::first_plt_entry) (Output_data_plt_x86_64::plt_entry) (Output_data_plt_x86_64::tlsdesc_plt_entry) (Output_data_plt_x86_64::plt_eh_frame_fde_size) (Output_data_plt_x86_64::plt_eh_frame_fde): Move to ... (Output_data_plt_x86_64_standard): ... here, new class. (Target_x86_64::Target_x86_64): Take optional argument for the Target_info pointer to use. (Target_x86_64::do_make_data_plt): New virtual method. (Target_x86_64::make_data_plt): New method to call it. (Target_x86_64::init_got_plt_for_update): Use that. Call this->plt_->add_eh_frame method here. (Output_data_plt_x86_64::init): Don't do add_eh_frame_for_plt here. (Target_x86_64::first_plt_entry_offset): Call method on this->plt_ rather than static method. (Target_x86_64::plt_entry_size): Likewise. (Output_data_plt_x86_64::do_write): Use get_plt_entry_size method rather than plt_entry_size variable. Move guts of PLT filling to... (Output_data_plt_x86_64_standard::do_fill_first_plt_entry): ... here ... (Output_data_plt_x86_64_standard::do_fill_plt_entry): ... and here ... (Output_data_plt_x86_64_standard::do_fill_tlsdesc_entry): ... and here. * i386.cc (Output_data_plt_i386::Output_data_plt_i386): Take additional argument for the section alignment. Don't do add_eh_frame_for_plt here. (Output_data_plt_i386::first_plt_entry_offset): Make the method non-static. Use get_plt_entry_size method rather than plt_entry_size variable. (Output_data_plt_i386::do_get_plt_entry_size): New abstract virtual method. (Output_data_plt_i386::get_plt_entry_size): Call it. (Output_data_plt_i386::do_add_eh_frame): New abstract virtual method. (Output_data_plt_i386::add_eh_frame): New method to call it. (Output_data_plt_i386::do_fill_first_plt_entry): New abstract virtual method. (Output_data_plt_i386::fill_first_plt_entry): New method to call it. (Output_data_plt_i386::do_fill_plt_entry): New abstract virtual method. (Output_data_plt_i386::fill_plt_entry): New method to call it. (Output_data_plt_i386::set_final_data_size): Use get_plt_entry_size method instead of plt_entry_size. (Output_data_plt_i386::plt_entry_size) (Output_data_plt_i386::plt_eh_frame_fde_size) (Output_data_plt_i386::plt_eh_frame_fde): Move to ... (Output_data_plt_i386_standard): ... here, new class. (Output_data_plt_i386_exec): New class. (Output_data_plt_i386::exec_first_plt_entry): Move to ... (Output_data_plt_i386_exec::first_plt_entry): ... here. (Output_data_plt_i386::exec_plt_entry): Move to ... (Output_data_plt_i386_exec::plt_entry): ... here. (Output_data_plt_i386_dyn): New class. (Output_data_plt_i386::first_plt_entry): Move to ... (Output_data_plt_i386_dyn::first_plt_entry): ... here. (Output_data_plt_i386::dyn_plt_entry): Move to ... (Output_data_plt_i386_dyn::plt_entry): ... here. (Target_i386::Target_i386): Take optional argument for the Target_info pointer to use. (Target_i386::do_make_data_plt): New virtual method. (Target_i386::make_data_plt): New method to call it. (Target_i386::make_plt_section): Use that. Call this->plt_->add_eh_frame method here. (Output_data_plt_i386::add_entry): Use get_plt_entry_size method rather than plt_entry_size variable. (Output_data_plt_i386::add_local_ifunc_entry): Likewise. (Output_data_plt_i386::address_for_local): Likewise. (Output_data_plt_i386::do_write): Likewise. Move guts of PLT filling to... (Output_data_plt_i386_exec::do_fill_first_plt_entry): ... here ... (Output_data_plt_i386_exec::do_fill_plt_entry): ... and here ... (Output_data_plt_i386_dyn::do_fill_first_plt_entry): ... and here ... (Output_data_plt_i386_dyn::do_fill_plt_entry): ... and here. Change-Id: Id24b95600489835ff5e860a39c147203d4380c2b |
||
---|---|---|
.. | ||
po | ||
testsuite | ||
aclocal.m4 | ||
archive.cc | ||
archive.h | ||
arm-reloc-property.cc | ||
arm-reloc-property.h | ||
arm-reloc.def | ||
arm.cc | ||
attributes.cc | ||
attributes.h | ||
binary.cc | ||
binary.h | ||
ChangeLog | ||
common.cc | ||
common.h | ||
compressed_output.cc | ||
compressed_output.h | ||
config.in | ||
configure | ||
configure.ac | ||
configure.tgt | ||
copy-relocs.cc | ||
copy-relocs.h | ||
cref.cc | ||
cref.h | ||
debug.h | ||
defstd.cc | ||
defstd.h | ||
descriptors.cc | ||
descriptors.h | ||
dirsearch.cc | ||
dirsearch.h | ||
dwarf_reader.cc | ||
dwarf_reader.h | ||
dynobj.cc | ||
dynobj.h | ||
ehframe.cc | ||
ehframe.h | ||
errors.cc | ||
errors.h | ||
expression.cc | ||
ffsll.c | ||
fileread.cc | ||
fileread.h | ||
freebsd.h | ||
ftruncate.c | ||
gc.cc | ||
gc.h | ||
gdb-index.cc | ||
gdb-index.h | ||
gold-threads.cc | ||
gold-threads.h | ||
gold.cc | ||
gold.h | ||
i386.cc | ||
icf.cc | ||
icf.h | ||
incremental-dump.cc | ||
incremental.cc | ||
incremental.h | ||
int_encoding.cc | ||
int_encoding.h | ||
layout.cc | ||
layout.h | ||
main.cc | ||
Makefile.am | ||
Makefile.in | ||
mapfile.cc | ||
mapfile.h | ||
merge.cc | ||
merge.h | ||
mremap.c | ||
nacl.cc | ||
nacl.h | ||
NEWS | ||
object.cc | ||
object.h | ||
options.cc | ||
options.h | ||
output.cc | ||
output.h | ||
parameters.cc | ||
parameters.h | ||
plugin.cc | ||
plugin.h | ||
powerpc.cc | ||
pread.c | ||
README | ||
readsyms.cc | ||
readsyms.h | ||
reduced_debug_output.cc | ||
reduced_debug_output.h | ||
reloc-types.h | ||
reloc.cc | ||
reloc.h | ||
resolve.cc | ||
script-c.h | ||
script-sections.cc | ||
script-sections.h | ||
script.cc | ||
script.h | ||
sparc.cc | ||
stringpool.cc | ||
stringpool.h | ||
symtab.cc | ||
symtab.h | ||
target-reloc.h | ||
target-select.cc | ||
target-select.h | ||
target.cc | ||
target.h | ||
timer.cc | ||
timer.h | ||
tls.h | ||
TODO | ||
token.h | ||
version.cc | ||
workqueue-internal.h | ||
workqueue-threads.cc | ||
workqueue.cc | ||
workqueue.h | ||
x86_64.cc | ||
yyscript.y |
gold is an ELF linker. It is intended to have complete support for ELF and to run as fast as possible on modern systems. For normal use it is a drop-in replacement for the older GNU linker. gold is part of the GNU binutils. See ../binutils/README for more general notes, including where to send bug reports. gold was originally developed at Google, and was contributed to the Free Software Foundation in March 2008. At Google it was designed by Ian Lance Taylor, with major contributions by Cary Coutant, Craig Silverstein, and Andrew Chatham. The existing GNU linker manual is intended to be accurate documentation for features which gold supports. gold supports most of the features of the GNU linker for ELF targets. Notable omissions--features of the GNU linker not currently supported in gold--are: * MRI compatible linker scripts * cross-reference reports (--cref) * various other minor options Notes on the code ================= These are some notes which may be helpful to people working on the source code of gold itself. gold is written in C++. It is a GNU program, and therefore follows the GNU formatting standards as modified for C++. Source documents in order of decreasing precedence: http://www.gnu.org/prep/standards/ http://gcc.gnu.org/onlinedocs/libstdc++/17_intro/C++STYLE http://www.zembu.com/eng/procs/c++style.html The linker is intended to have complete support for cross-compilation, while still supporting the normal case of native linking as fast as possible. In order to do this, many classes are actually templates whose parameter is the ELF file class (e.g., 32 bits or 64 bits). The C++ code is the same, but we don't pay the execution time cost of always using 64-bit integers if the target is 32 bits. Many of these class templates also have an endianness parameter: true for big-endian, false for little-endian. The linker is multi-threaded. The Task class represents a single unit of work. Task objects are stored on a single Workqueue object. Tasks communicate via Task_token objects. Task_token objects are only manipulated while holding the master Workqueue lock. Relatively few mutexes are used. Build requirements ================== The gold source code uses templates heavily. Building it requires a recent version of g++. g++ 4.0.3 and 4.1.3 are known to work. g++ 3.2, 3.4.3, and 4.1.2 are known to fail. The linker script parser uses features which are only in newer versions of bison. bison 2.3 is known to work. bison 1.26 is known to fail. If you are building gold from an official binutils release, the bison output should already be included.