binutils-gdb

Commit Graph

Author	SHA1	Message	Date
Alan Modra	978c445051	readelf: move file related static vars to filedata The idea here is to get rid of a lot of file related static vars used to pass data around, in order to not have stale data about one object file persisting to the next one. * readelf.c (archive_file_offset, archive_file_size, dynamic_addr), (dynamic_size, dynamic_nent, dynamic_strings, dynamic_strings_length), (num_dynamic_syms, nbuckets, nchains, buckets, chains), (ngnubuckets, gnubuckets, gnuchains, mipsxlat, ngnuchains), (gnusymidx, dynamic_symbols, dynamic_syminfo, dynamic_syminfo_offset), (dynamic_syminfo_nent, program_interpreter, dynamic_info), (dynamic_info_DT_GNU_HASH, dynamic_info_DT_MIPS_XHASH, version_info), (dynamic_section, symtab_shndx_list, group_count, section_groups), (section_headers_groups): Move to struct filedata. Update use throughout file.	6 years ago
Alan Modra	6431e40968	readelf: cmdline data Don't use a struct filedata for cmdline, which only needs two of the filedata fields. * readelf.c (struct dump_data): New, used.. (cmdline): ..here, and.. (struct filedata): ..here. Adjust all uses. (request_dump_bynumber, request_dump, parse_args): Pass in a struct dump_data* rather than Filedata*. Adjust callers. (main): Don't set cmdline.file_name.	6 years ago
Alan Modra	a5e0be5cb2	readelf: segfault at readelf.c:12227 This is another one where not cleaning up sufficiently after processing one file can lead to errors when processing the next file. We have ngnuchains non-zero but gnuchains NULL in the following: off < ngnuchains && (gnuchains[off] & 1) == 0 * readelf.c (process_symbol_table): Clear ngnuchains, ngnubuckets and nbuckets.	6 years ago
Alan Modra	8ff66993e0	readelf: segfaults fuzzing multiple object files This patch is aimed at fixing a number of oss-fuzz segfaults that don't reproduce reliably with their current infrastructure, the problem being that one invocation of readelf is effectively being run on multiple object files. I believe that these segfaults could be reliably reproduced with just two fuzzed objects being presented to readelf, but those inputs are currently not identified by oss-fuzz. So there is some guesswork involved in this patch. The idea here is to clear stashed data such as symtab_shndx_list that is processed using section header info, at the same time that header info is cleared. * readelf.c (process_section_headers): Free dynamic symbols etc. earlier.	6 years ago
Alan Modra	ceb9bf11b9	readelf memory leaks This fixes two leaks found in the new code supporting display of dynamic symbols retrieved via dynamic tags. * readelf.c (get_num_dynamic_syms): Formatting. Don't return on error without freeing. (process_dynamic_section): Don't recreate dynamic symbols from dynamic tag info when the dynamic symbols have already been read via section headers.	6 years ago
Fredrik Strupe	e409955ddc	[PATCH v2] binutils: arm: Fix disassembly of conditional VDUPs. VDUP (neon) instructions can be conditional, but this is not taken into account in the current master. This commit fixes that by i) fixing the VDUP instruction masks and ii) adding logic for disassembling conditional neon instructions. opcodes * arm-dis.c (neon_opcodes): Fix VDUP instruction masks. (print_insn_neon): Support disassembly of conditional instructions. binutils* testsuite/binutils-all/arm/vdup-cond.d: New test for testing that conditional VDUP instructions are disassembled correctly. * testsuite/binutils-all/arm/vdup-cond.s: New file used by vdup-cond.d. * testsuite/binutils-all/arm/vdup-thumb.d: New test for testing that VDUP instructions (which are conditional in A32) can be disassembled in thumb mode. * testsuite/binutils-all/arm/vdup-cond.s: New file used by vdup-thumb.d.	6 years ago
Alan Modra	ca3f923f82	PR25840, Null pointer dereference in objdump PR 25840 * debug.c (debug_class_type_samep): Don't segfault on NULL type.	6 years ago
Alan Modra	937f661468	Extend objdump --no-addresses info * doc/binutils.texi: Mention --no-show-raw-insn in objdump --no-addresses description.	6 years ago
Alan Modra	b1bc1394df	objdump --no-addresses I find this useful when needing to compare compiler output, where the address of the instruction and the value of symbols results in unwanted differences. * objdump.c (no_addresses): New static var. (usage): Print help for --no-addresses. (long_options): Add --no-addresses entry. (objdump_print_addr_with_sym, objdump_print_addr): Omit symbol address. (disassemble_bytes): Don't print current line address, or reloc address. * doc/binutils.texi: Document objdump --no-addresses.	6 years ago
Alan Modra	001890e1f9	PR25822, Invalid read in process_symbol_table PR 25822 * readelf.c (get_num_dynamic_syms): Don't set num_of_syms when reading buckets or chains fails.	6 years ago
Alan Modra	b71d4fa7c3	readelf: zero static vars after freeing When readelf is processing more than one file, static bss vars won't start out as zero for the second file unless they are cleared. * readelf.c (process_symbol_table): Zero gnubuckets, gnuchains etc. after freeing.	6 years ago
Alan Modra	89246a0e79	readelf: increase size of static buffers Translated strings might be larger than the original. * readelf.c (get_group_flags): Translate text. (get_file_type, get_symbol_binding, get_symbol_type), (get_ppc64_symbol_other, get_symbol_other): Increase size of buffer.	6 years ago
Alan Modra	c98a4545dc	Re: readelf: Consolidate --syms --use-dynamic with --dyn-syms PR 25821 * readelf.c (get_num_dynamic_syms): Typo fix.	6 years ago
Alan Modra	645f43a80c	readelf memory leaks processing mips * readelf.c (process_mips_specific): Free eopt and iopt. Avoid possibility of overflow when checking number of conflicts.	6 years ago
H.J. Lu	10ca4b042d	readelf: Consolidate --syms --use-dynamic with --dyn-syms When reconstructing dynamic symbol table from the PT_DYNAMIC segment, compute dynamic symbol table size from hash table. For DT_HASH, the number of dynamic symbol table entries equals the number of chains. For DT_GNU_HASH/DT_MIPS_XHASH, only defined symbols with non-STB_LOCAL indings are in hash table. Since DT_GNU_HASH/DT_MIPS_XHASH place all symbols with STB_LOCAL binding before symbols with other bindings and all undefined symbols defined ones in dynamic symbol table, the highest symbol index in DT_GNU_HASH/DT_MIPS_XHASH is the highest dynamic symbol table index. Rewrite print_dynamic_symbol to dump dynamic symbol table for --dyn-syms and --syms --use-dynamic. binutils/ PR binutils/25707 * readelf.c (nbuckets): New. (nchains): Likewise. (buckets): Likewise. (chains): Likewise. (ngnubuckets): Likewise. (gnubuckets): Likewise. (gnuchains): Likewise. (mipsxlat): Likewise. (ngnuchains): Likewise. (gnusymidx): Likewise. (VALID_SYMBOL_NAME): Likewise. (VALID_DYNAMIC_NAME): Use it. (get_dynamic_data): Moved before process_dynamic_section. (get_num_dynamic_syms): New function. (process_dynamic_section): Use DT_SYMTAB, DT_SYMENT, DT_HASH, DT_GNU_HASH and DT_MIPS_XHASH to reconstruct dynamic symbol table. Use DT_STRTAB and DT_STRSZ to reconstruct dynamic string table. (get_symbol_index_type): Don't print "bad section index" when there is no section header. (print_dynamic_symbol): Rewrite. (process_symbol_table): Call print_dynamic_symbol to dump dynamic symbol table. ld/ PR binutils/25707 * testsuite/ld-arm/armthumb-lib.sym: Updated. * testsuite/ld-arm/farcall-mixed-app.sym: Likewise. * testsuite/ld-arm/farcall-mixed-app2.sym: Likewise. * testsuite/ld-arm/fdpic-main-m.sym: Likewise. * testsuite/ld-arm/fdpic-main.sym: Likewise. * testsuite/ld-arm/fdpic-shared-m.sym: Likewise. * testsuite/ld-arm/fdpic-shared.sym: Likewise. * testsuite/ld-arm/mixed-app.sym: Likewise. * testsuite/ld-arm/mixed-lib.sym: Likewise. * testsuite/ld-arm/preempt-app.sym: Likewise. * testsuite/ld-elf/hash.d: Likewise. * testsuite/ld-elf/pr13195.d: Likewise. * testsuite/ld-elfvsb/hidden2.d: Likewise. * testsuite/ld-mips-elf/hash2.d: Likewise.	6 years ago
Rainer Orth	bb89747721	ld: Disable ifunc tests on Solaris A couple of ld ifunc tests currently FAIL on 64-bit Solaris/x86: FAIL: ld-ifunc/ifunc-10-x86-64 FAIL: ld-ifunc/ifunc-11-x86-64 FAIL: ld-ifunc/ifunc-12-x86-64 FAIL: ld-ifunc/ifunc-13-x86-64 FAIL: ld-ifunc/ifunc-14a-x86-64 FAIL: ld-ifunc/ifunc-14b-x86-64 FAIL: ld-ifunc/ifunc-14c-x86-64 FAIL: ld-ifunc/ifunc-14d-x86-64 FAIL: ld-ifunc/ifunc-14e-x86-64 FAIL: ld-ifunc/ifunc-14f-x86-64 FAIL: ld-ifunc/ifunc-15-x86-64 FAIL: ld-ifunc/ifunc-17a-x86-64 FAIL: ld-ifunc/ifunc-17b-x86-64 FAIL: ld-ifunc/ifunc-2-local-x86-64-now FAIL: ld-ifunc/ifunc-2-local-x86-64 FAIL: ld-ifunc/ifunc-2-x86-64-now FAIL: ld-ifunc/ifunc-2-x86-64 FAIL: ld-ifunc/ifunc-20-x86-64 FAIL: ld-ifunc/pr17154-x86-64-now FAIL: ld-ifunc/pr17154-x86-64 For one, the actual error is weird: ./ld-new: target elf64-x86-64 not found failed with: <./ld-new: target elf64-x86-64 not found>, no expected output FAIL: ld-ifunc/ifunc-10-x86-64 although ld -V does report the elf_x86_64 emulation as supported: $ ./ld/ld-new -V GNU ld (GNU Binutils) 2.34.50.20200328 Supported emulations: elf_x86_64_sol2 elf_x86_64 [...] When using ld -m elf_x86_64_sol2 instead, one of the testcases links successfully. However, there's no point in pursuing this: Solaris does not support ifunc, as can be seen in <sys/elf.h>: /* * GNU/Linux specific symbol type not used by Solaris / #define STT_GNU_IFUNC 10 and never will, given that it has symbol capabilities as solution to effectively the same problem: http://www.linker-aliens.org/blogs/rie/entry/symbol_capabilitie/ Therefore this patch disables ifunc testing on Solaris completely by removing Solaris from binutils/testsuite/lib/binutils-common.exp (supports_gnu_osabi). The ifunc part is justified above. SHF_GNU_MBIND is in the OS-specific range and conflicts with #define SHF_SUNW_REALLOC 0x01000000 / internal: krtld realloc / While the comment suggests this might be relocatable without too much problems, the description of mbind (no formal spec AFAICS, just the comment in the binutils patch submission) strongly suggests that this isn't relevant to Solaris at all. Indirectly, clearing supports_gnu_osabi on Solaris disables supports_gnu_unique. Again, Solaris <sys/elf.h> has / * GNU/Linux specific binding not used by Solaris / #define STB_GNU_UNIQUE 10 so this seems the right thing to do. Afterwards, one can remove the explicit mentions of --solaris2 in quite a number of (but not all) the ld-ifunc dump file notarget lists. There's one fallout, though: two gas tests now XPASS because they are xfail'ed for !supports_gnu_osabi: XPASS: mbind sections 12 XPASS: mbind section contents 16 XPASS: mbind sections 16 XPASS: mbind section contents 16 To fix that, I've changed #xfail: ![supports_gnu_osabi] to notarget. Tested on x86_64-pc-solaris2.11, i386-pc-solaris2.11, x86_64-pc-linux-gnu, and i686-pc-linux-gnu. ld: * testsuite/ld-ifunc/ifunc-10-i386.d: Remove --solaris2* from notarget. * ifunc-11-i386.d: Likewise. * ifunc-12-i386.d: Likewise. * ifunc-13-i386.d: Likewise. * ifunc-14a-i386.d: Likewise. * ifunc-14b-i386.d: Likewise. * ifunc-14c-i386.d: Likewise. * ifunc-14d-i386.d: Likewise. * ifunc-14e-i386.d: Likewise. * ifunc-14f-i386.d: Likewise. * ifunc-15-i386.d: Likewise. * ifunc-16-i386-now.d: Likewise. * ifunc-16-i386.d: Likewise. * ifunc-17a-i386.d: Likewise. * ifunc-17b-i386.d: Likewise. * ifunc-18a-i386.d: Likewise. * ifunc-18b-i386.d: Likewise. * ifunc-19a-i386.d: Likewise. * ifunc-19b-i386.d: Likewise. * ifunc-2-i386-now.d: Likewise. * ifunc-2-i386.d: Likewise. * ifunc-2-local-i386-now.d: Likewise. * ifunc-2-local-i386.d: Likewise. * ifunc-20-i386.d: Likewise. * ifunc-21-i386.d: Likewise. * ifunc-22-i386.d: Likewise. * ifunc-5a-i386.d: Likewise. * ifunc-5a-local-i386.d: Likewise. * ifunc-5b-i386.d: Likewise. * ifunc-5b-local-i386.d: Likewise. * ifunc-5r-local-i386.d: Likewise. * ifunc-6a-i386.d: Likewise. * ifunc-6b-i386.d: Likewise. * ifunc-7a-i386.d: Likewise. * ifunc-7b-i386.d: Likewise. * ifunc-8-i386.d: Likewise. * ifunc-9-i386.d: Likewise. * pr17154-i386-now.d: Likewise. * pr17154-i386.d: Likewise. * ifunc-23a-x86.d: Remove notarget. * ifunc-24a-x86.d: Likewise. * ifunc-25a-x86.d: Likewise. gas: * testsuite/gas/elf/section12a.d: Use notarget instead of xfail. * testsuite/gas/elf/section12b.d: Likewise. * testsuite/gas/elf/section16a.d: Likewise. * testsuite/gas/elf/section16b.d: Likewise. binutils: * testsuite/lib/binutils-common.exp (supports_gnu_osabi): Don't enable on --solaris*.	6 years ago
Nick Clifton	00386881a3	Fix objcopy's --preserve-dates command line option so that it will work with PE format files. PR binutils/pr25662 bfd * libcoff-in.h (struct pe_tdata): Rename the insert_timestamp field to timestamp and make it an integer. * libcoff.h: Regenerate. * peXXigen.c (_bfd_XXi_only_swap_filehdr_out): Test the timestamp field in the pe_data structure rather than the insert_timestamp field. binutils* objcopy.c (copy_object): When copying PE format files set the timestamp field in the pe_data structure if the preserve_dates flag is set. * testsuite/binutils-all/objcopy.exp (objcopy_test) Use --preserve-dates in place of the -p option, in order to make its effect more obvious. ld * emultempl/pe.em (after_open): Replace initialisation of the insert_timestamp field in the pe_data structure with an initialisation of the timestamp field. * emultemp/pep.em: Likewise. * pe-dll.c (fill_edata): Use the timestamp field in the pe_data structure instead of the insert_timestamp field.	6 years ago
Alan Modra	0a6a8b596b	Re: Adjust objcopy_test Last patch didn't manage to xfail spu due to clear_xfail --elf. Clearing --elf dates back to a time when we had rather a lot more setup_xfail patterns, so limiting it to hppa--elf. Also, mips--irix ought to have been mips--irix* and I'm having second thoughts about xfailing mips and hiding what looks like a problem: If the mips target is supposed to emit names for local section symbols and does so for objcopy, why isn't it doing the same for ld? Also, lots more mips targets would be subject to this test failing. So I'm backing out those xfails and leaving it to someone more knowledgeable about mips. * testsuite/binutils-all/objcopy.exp (objcopy_test): Only clear_xfail hppa--elf. Revert mips xfails.	6 years ago
Alan Modra	6a81c59be7	Adjust objcopy_test xfails spu due to a note section getting a different vma, and some mips targets that give section symbols a name string. I added -p for the executable test in an attempt to fix all the pe target fails, but that doesn't preserve the date/time for some reason. * testsuite/binutils-all/objcopy.exp (objcopy_test): Move xfails from here to calls. Remove "m8--*" entry. Don't xfail tic54x but do xfail spu, mipstx39 and mips-sgi-irix for the executable test. Pass "-p" to objcopy for the executable test.	6 years ago
Jozef Lawrynowicz	1fafefd594	Add testcase for PR 25662 invalid sh_offset for section binutils/ChangeLog: 2020-03-27 Jozef Lawrynowicz <jozef.l@mittosystems.com> PR binutils/25662 * testsuite/binutils-all/objcopy.exp (objcopy_test): Add argument to specify whether an object file or executable should be built and tested. Change test names to report whether an object file or executable is being tested. * testsuite/binutils-all/pr25662.ld: New test. * testsuite/binutils-all/pr25662.s: New test.	6 years ago
Alan Modra	80e2a3b66e	Re: readelf looping in process_archive This patch fixes a leak of qualified_name caused by `4c83662712` and a double free introduced by `fd486f32d1`. Not breaking out of the loop results in an error: "failed to seek to next archive header". That's slightly better than silently preventing the possibility of endless loops. * readelf.c (process_archive): Don't double free qualified_name. Don't break out of loop with "negative" archive_file_size, just set file offset to max.	6 years ago
Alan Modra	4c83662712	readelf looping in process_archive With a crafted "negative" ar_hdr.ar_size it is possible to make readelf loop. This patch catches the overflow in a file offset calculation. * readelf.c (process_archive): Prevent endless loop.	6 years ago
H.J. Lu	1081065c51	bfd: Add a bfd_boolean argument to bfd_get_symbol_version_string We can't call _bfd_elf_get_symbol_version_name from nm.c since it isn't available for all target configurations. This patch add a bfd_boolean argument to bfd_get_symbol_version_string instead. bfd/ PR binutils/25708 * elf-bfd.h (_bfd_elf_get_symbol_version_name): Renamed to ... (_bfd_elf_get_symbol_version_string): This. * elf.c (_bfd_elf_get_symbol_version_name): Renamed to ... (_bfd_elf_get_symbol_version_string): This. (bfd_elf_print_symbol): Pass TRUE to _bfd_elf_get_symbol_version_string. * libbfd-in.h (_bfd_nosymbols_get_symbol_version_string): Add a bfd_boolean argument. * syms.c (_bfd_nosymbols_get_symbol_version_string): Likewise. * targets.c (_bfd_get_symbol_version_string): Likewise. (bfd_get_symbol_version_string): Likewise. * bfd-in2.h: Regenerated. binutils/ PR binutils/25708 * nm.c (print_symname): Replace _bfd_elf_get_symbol_version_name with bfd_get_symbol_version_string. (print_symbo): Pass TRUE to bfd_get_symbol_version_string. * objdump.c (objdump_print_symname): Likewise.	6 years ago
H.J. Lu	7e6e972f74	bfd: Display symbol version for nm -D Extend _bfd_elf_get_symbol_version_string for nm -D to display symbol version. _bfd_elf_get_symbol_version_name is added to avoid updating all XXX_get_symbol_version_string functions. bfd/ PR binutils/25708 * elf-bfd.h (_bfd_elf_get_symbol_version_name): New. * elf.c (_bfd_elf_get_symbol_version_name): New function. Based on the previous _bfd_elf_get_symbol_version_string. (_bfd_elf_get_symbol_version_string): Use it. binutils/ PR binutils/25708 * nm.c (SYM_NAME): Removed. (print_symname): Add a pointer to struct extended_symbol_info argument. Call _bfd_elf_get_symbol_version_name to get symbol version. (print_symdef_entry): Pass NULL to print_symname. (print_symbol_info_bsd): Update call to print_symname. (print_symbol_info_sysv): Likewise. (print_symbol_info_posix): Likewise. ld/ PR binutils/25708 * testsuite/ld-elf/pr25708.d: New file.	6 years ago
Alan Modra	5a814d6d96	readelf memory leak in process_mips_specific * readelf.c (process_mips_specific): Free iconf on error path.	6 years ago
Nick Clifton	0636b24556	Avoid trigger an assertion failure in the STABS parser by replacing the abort with an error return. PR 25714 * wrstabs.c (stab_pop_type): Replace assertion with error return. * write_stabs_in_sections_debugging_info: Likewise. * stab_enum_type: Likewise. * stab_modify_type: Likewise. * stab_struct_field: Likewise. * stab_end_struct_type: Likewise. * stab_start_class_type: Likewise. * stab_class_static_member: Likewise. * stab_class_baseclass: Likewise. * stab_class_start_method: Likewise. * stab_class_method_var: Likewise. * stab_class_end_method: Likewise. * stab_end_class_type: Likewise. * stab_typedef_type: Likewise. * stab_start_function: Likewise. * stab_end_block: Likewise. * stab_lineno: Likewise.	6 years ago
Nick Clifton	5844b4657a	Fix compile time warnings about the possibility of using uninitialised fields in the Elf_Internal_Chdr structure when building readelf. * readelf.c (get_compression_header): Add ATTRIBUTE_WARN_UNUSED_RESULT. (process_section_headers): Check the return value from get_compression_header. (dump_section_as_strings): Likewise. (dump_section_as_bytes): Likewise. (load_specific_debug_section): Likewise.	6 years ago
H.J. Lu	3d98c46092	plugin: Don't invoke LTO-wrapper Don't invoke LTO-wrapper since the LTO wrapper approach is not only slow but also unreliable. For GCC 10 or newer, LDPT_ADD_SYMBOLS_V2 will be used. bfd/ * configure.ac (HAVE_EXECUTABLE_SUFFIX): Removed. (EXECUTABLE_SUFFIX): Likewise. * config.in: Regenerated. * configure: Likewise. * plugin.c (bfd_plugin_close_and_cleanup): Defined as _bfd_generic_close_and_cleanup. (plugin_list_entry): Remove resolution_file, resolution_option, real_bfd, real_nsyms, real_syms, lto_nsyms, lto_syms, gcc, lto_wrapper, gcc_env and initialized, (need_lto_wrapper_p): Removed. (get_lto_wrapper): Likewise. (setup_lto_wrapper_env): Likewise. (register_all_symbols_read): Likewise. (egister_cleanup): Likewise. (get_symbols): Likewise. (add_input_file): Likewise. (bfd_plugin_set_program_name): Remove need_lto_wrapper. (add_symbols): Updated. (try_claim): Likewise. (try_load_plugin): Likewise. (bfd_plugin_canonicalize_symtab): Likewise. * plugin.h (bfd_plugin_set_program_name): Remove int argument. (plugin_data_struct): Remove real_bfd, real_nsyms and real_syms. binutils/ * ar.c (main): Update bfd_plugin_set_program_name call. * nm.c (main): Likewise. ld/ * testsuite/ld-plugin/lto.exp (lto_link_tests): Run PR ld/25355 test only for GCC 10 or newer.	6 years ago
Nick Clifton	f3a08f7778	Fix discrepancies in nm's --line-number output by adding support for the DW_AT_specification DWARF Attttribute. PR 25676 bfd * dwarf2.c (struct varinfo): Add unit_offset field to record the location of the varinfo in the unit's debug info data. Change the type of the stack field to a boolean. (lookup_var_by_offset): New function. Returns the varinfo structure for the variable described at the given offset in the unit's debug info. (scan_unit_for_symbols): Add support for variables which have the DW_AT_specification attribute. binutils* testsuite/binutils-all/dw4.s: New test source file. * testsuite/binutils-all/nm.exp: Run the new test.	6 years ago
H.J. Lu	ef3df11042	readelf.c: Use unsigned long to iterate over num_syms process_symbol_table () has unsigned long num_syms; ... for (si = 0, psym = symtab; si < num_syms; si++, psym++) We should use unsigned long to iterate over num_syms. * readelf.c (process_symbol_table): Use unsigned long for si.	6 years ago
Alan Modra	81a65eb338	Don't use bfd.h in binutils/elfcomm.c Better than warning about bfd types, just don't include bfd.h and warn against including the header again. * elfcomm.c: Don't include bfd.h or bucomm.h. (program_name): Declare. (process_archive_index_and_symbols): Replace bfd_boolean with int, and substitute FALSE and TRUE. (setup_archive, setup_nested_archive): Likewise. * elfcomm.h: Likewise.	6 years ago
Alan Modra	1cb7d8b1af	readelf leak in process_archive * readelf.c (process_archive): Always return via path freeing memory. Formatting.	6 years ago
Alan Modra	b966f55ffa	Invalid read in process_netbsd_elf_note * readelf.c (process_netbsd_elf_note): Validate descsz before accessing descdata. Formatting.	6 years ago
Alan Modra	645ba68105	setup_archive parameter mismatch * elfcomm.c (setup_archive): Make file_size an off_t. Comment. * elfcomm.h (setup_archive): Update prototype.	6 years ago
Alan Modra	28d1356774	asan: readelf leak in hppa_process_unwind This extracts code reading symbol tables into a common function that tidies up after errors. I've also changed an error reporting multiple string tables to an error on multiple symbol tables. * readelf.c (get_symbols): New function. (process_relocs, ia64_process_unwind, hppa_process_unwind), (arm_process_unwind, get_symbol_for_build_attribute): Use it.	7 years ago
Alan Modra	60e63c3e97	ubsan: shift exponent 70 is too large * unwind-ia64.c (unw_decode_uleb128): Prevent overlarge shifts. Detect shift overflows and check that terminating byte is found. Print an error on a bad uleb128.	7 years ago
Alan Modra	1bd6175ae7	Re: asan: more readelf leaks In git commit `fd486f32d1` I put some static variables used by get_symbol_for_build_attribute in a file scope ba_cache struct. This was to prevent leaks in get_symbol_for_build_attribute, and to tidy up before readelf exited. The patch wasn't quite right though. When readelf processes more than one file it was possible to double free arrays allocated in get_symbol_for_build_attribute. * readelf.c (process_file): Clean ba_cache.	7 years ago
Alan Modra	780f96aed2	readelf large memory allocation * elfcomm.h (setup_archive): Update prototype. * elfcomm.c (setup_archive): Add file_size parameter and sanity check longnames_size. (setup_nested_archive): Get file size and pass to setup_archive. * elfedit.c (process_archive): Likewise. * readelf.c (process_archive): Pass filedata->file_size to setup_archive.	7 years ago
Alan Modra	f761cb13a9	asan: readelf: memory leaks * readelf.c (dump_section_as_strings): Free memory on error exit. (dump_section_as_bytes, process_notes_at): Likewise. (get_build_id): Free enote.	7 years ago
Kamil Rytarowski	06d949ec31	Implement NT_NETBSDCORE_LWPSTATUS (NetBSD-Core) bfd/ChangeLog: * elf.c (elfcore_grok_netbsd_note): Add support for NT_NETBSDCORE_LWPSTATUS notes. binutils/ChangeLog: * readelf.c (get_netbsd_elfcore_note_type): Add support for NT_NETBSDCORE_LWPSTATUS notes. include/ChangeLog: * elf/common.h (NT_NETBSDCORE_LWPSTATUS): New define.	7 years ago
Alan Modra	fd486f32d1	asan: more readelf leaks * elfcomm.c (get_archive_member_name): Always return malloc'd string or NULL. * elfedit.c (process_archive): Tidy memory on all return paths. * readelf.c (process_archive): Likewise. (process_symbol_table): Likewise. (ba_cache): New, replacing .. (get_symbol_for_build_attribute): ..static vars here. Free strtab and symtab before loading new ones. Reject symtab without valid strtab in loop, breaking out of loop on valid symtab. (process_file): Free ba_cache symtab and strtab here, resetting ba_cache.	7 years ago
Alan Modra	8fb879cd16	asan: readelf leaks * readelf.c (process_section_headers): Don't just set filedata->section_headers NULL, free it first. Similarly for dynamic_symbols, dynamic_strings, dynamic_syminfo and symtab_shndx_list. Zero associated counts too. (process_object): Free symtab_shndx_list. (process_file): Free various allocated filedata tables.	7 years ago
Nick Clifton	5496f3c635	Add support for generating DWARF-5 format directory and file name tables from the assembler. PR 25611 PR 25614 * dwarf.h (DWARF2_Internal_LineInfo): Add li_address_size and li_segment_size fields. * dwarf.c (read_debug_line_header): Record the address size and segment selector size values (if present) in the lineinfo structure. (display_formatted_table): Warn if the format count is empty but the table itself is not empty. Display the format count and entry count at the start of the table dump. (display_debug_lines_raw): Display the address size and segement selector size fields, if present. * testsuite/binutils-all/dw5.W: Update expected output. gas * dwarf2dbg.c (DWARF2_FILE_TIME_NAME): Default to -1. (DWARF2_FILE_SIZE_NAME): Default to -1. (DWARF2_LINE_VERSION): Default to the current dwarf level or 3, whichever is higher. (DWARF2_LINE_MAX_OPS_PER_INSN): Provide a default value of 1. (NUM_MD5_BYTES): Define. (struct file entry): Add md5 field. (get_filenum): Delete and replace with... (get_basename): New function. (get_directory_table_entry): New function. (allocate_filenum): New function. (allocate_filename_to_slot): New function. (dwarf2_where): Use new functions. (dwarf2_directive_filename): Add support for extended .file pseudo-op. (dwarf2_directive_loc): Allow the use of file number zero with DWARF 5 or higher. (out_file_list): Rename to... (out_dir_and_file_list): Add DWARF 5 support. (out_debug_line): Emit extra values into the section header for DWARF 5. (out_debug_str): Allow for file 0 to be used with DWARF 5. * doc/as.texi (.file): Update the description of this pseudo-op. * testsuite/gas/elf-dwarf-5-file0.s: Add more lines. * testsuite/gas/elf-dwarf-5-file0.d: Update expected dump output. * testsuite/gas/lns/lns-diag-1.l: Update expected error message. * NEWS: Mention the new feature.	7 years ago
Alan Modra	9cc89dc0ac	PR25651, objcopy SIGSEGV in copy_object With the right set of options, the second block of code dealing with padding can see a different section count. So don't use the new count. Since I was editing those lines, I've also changed the code allocating arrays a little. array = malloc (n * sizeof (array)); for an array of ints is just better than array = malloc (n sizeof (int)); It's easier to write correctly in the first place and more robust against code changes that might modify the array element type. PR 25651 * objcopy.c (copy_object): Test "gaps" not gap_fill_set or pad_to_set on second block of code dealing with padding. Replace "c" with "num_sec" and don't recalculate number of sections on second block. Size arrays using sizeof (element) rather than sizeof (element type).	7 years ago
Alan Modra	41da082238	objdump disassembly of files without symbols ubsan complains about memcpy with a NULL src even when size is zero. * objdump.c (disassemble_section): Don't call qsort unless sym count is at least two. (disassemble_data): Don't call memcpy with NULL src.	7 years ago
Alan Modra	9509988993	PR25645, readelf segfault reading fuzzed alpha-vms binary PR 25645 * readelf.c (dump_ia64_vms_dynamic_fixups): Pass size and nmemb to get_data rather than multiplying. (dump_ia64_vms_dynamic_relocs): Likewise. (process_version_sections): Correct order of size and nmemb args in get_data call. (process_mips_specific): Likewise.	7 years ago
H.J. Lu	a15de1f599	readelf.c: Fix a typo in comments * readelf.c (get_dynamic_data): Replace "memory chekers" with "memory checkers" in comments.	7 years ago
Simon Marchi	9979ab6663	binutils: doc: move artifacts back to MAINTAINERCLEANFILES In commit `2b44a6a237` (" binutils: doc: make `make clean` clean more things"), I moved the doc build artifacts to MOSTLYCLEANFILES, which made them get removed by "make clean". Because generating binutils.info requires makeinfo, and we do not want to require makeinfo when building from the tarball, binutils.info should not get removed by "make clean" (otherwise, it won't be included in the tarball). And to be consistent with other projects (e.g. ld and gas), we also want to ship the built man pages in the tarball. This patch puts back all these in MAINTAINERCLEANFILES, so that they are bundled in the tarball, and only cleaned if you use "make maintainer-clean". Tested by building a source release and confirming they are present. binutils/ChangeLog: PR 25491 * doc/Makefile.am: Rename MOSTLYCLEANFILES to MAINTAINERCLEANFILES. * doc/Makefile.in: Re-generate.	7 years ago
Nick Clifton	a0dcf29705	Fix an abort triggered when objcopy is used to set the "share" section flag on an ELF section. binutils* objcopy.c (check_new_section_flags): New function. Reject the SEC_COFF_SHARED flag if the target is not a COFF binary. (copy_object): Call check_new_section_flags. (setup_section): Likewise. * doc/binutils.texi (objcopy): Add a note that the 'share' section flag cannot be applied to ELF binaries. bfd * elf.c (_bfd_elf_set_section_contents): Replace call to abort with error messages and failure return values.	7 years ago
Alan Modra	a9b90127e8	PR25637, objcopy : SIGSEGV in copy_object PR 25637 * objcopy.c (filter_symbols): Correct rem_leading_char logic.	7 years ago

1 2 3 4 5 ...

3291 Commits (279edac53db8fa6482ee3e305c9627f788fd2699)