summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Correct improper-inclusion check in bits/libio-ldbl.h.Zack Weinberg2017-12-312-1/+11
| | | | | | | | | | | The patch which moved libio.h proper into the bits directory also changed the name of its guard macro, and I neglected to check whether anything depended on that name. It turns out that there is a conditionally-used bits header that looks at it; this broke the libgcc build on at least sparc64-*-* and sparcv9-*-*. * libio/bits/libio-ldbl.h: Correct check for improper inclusion. Add own multiple include guard.
* elf: Check for empty tokens before dynamic string token expansion [BZ #22625]Aurelien Jarno2017-12-303-16/+47
| | | | | | | | | | | | | | | | | | | | | | | | | The fillin_rpath function in elf/dl-load.c loops over each RPATH or RUNPATH tokens and interprets empty tokens as the current directory ("./"). In practice the check for empty token is done *after* the dynamic string token expansion. The expansion process can return an empty string for the $ORIGIN token if __libc_enable_secure is set or if the path of the binary can not be determined (/proc not mounted). Fix that by moving the check for empty tokens before the dynamic string token expansion. In addition, check for NULL pointer or empty strings return by expand_dynamic_string_token. The above changes highlighted a bug in decompose_rpath, an empty array is represented by the first element being NULL at the fillin_rpath level, but by using a -1 pointer in decompose_rpath and other functions. Changelog: [BZ #22625] * elf/dl-load.c (fillin_rpath): Check for empty tokens before dynamic string token expansion. Check for NULL pointer or empty string possibly returned by expand_dynamic_string_token. (decompose_rpath): Check for empty path after dynamic string token expansion.
* linux/x86: sync sys/ptrace.h with Linux 4.14 [BZ #22433]Dmitry V. Levin2017-12-292-0/+197
| | | | | | | | | | Clone generic linux/sys/ptrace.h, add the following x86-specific __ptrace_request constants: PTRACE_GET_THREAD_AREA, PTRACE_SET_THREAD_AREA, PTRACE_ARCH_PRCTL, PTRACE_SYSEMU, PTRACE_SYSEMU_SINGLESTEP, and PTRACE_SINGLEBLOCK. [BZ #22433] * sysdeps/unix/sysv/linux/x86/sys/ptrace.h: New file.
* linux: update sys/ptrace.h commentsDmitry V. Levin2017-12-298-25/+60
| | | | | | | | | | | | | * sysdeps/unix/sysv/linux/sys/ptrace.h (__ptrace_request): Add comments about PTRACE_PEEKSIGINFO, PTRACE_GETSIGMASK, PTRACE_SETSIGMASK, and PTRACE_SECCOMP_GET_FILTER. Update comments about PTRACE_SINGLESTEP and PTRACE_SYSCALL. * sysdeps/unix/sysv/linux/aarch64/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/ia64/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/powerpc/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/s390/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/sparc/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/tile/sys/ptrace.h: Likewise.
* linux: move a shared part of sys/ptrace.h to bits/ptrace-shared.hDmitry V. Levin2017-12-2910-383/+109
| | | | | | | | | | | | | | | | | | | | | Move a shared part of sys/ptrace.h which is the same on all architectures to a separate file. * sysdeps/unix/sysv/linux/sys/ptrace.h: Include <bits/ptrace-shared.h>. (__ptrace_setoptions, __ptrace_eventcodes, __ptrace_peeksiginfo_args, __ptrace_peeksiginfo_flags, ptrace): Move to ... * sysdeps/unix/sysv/linux/bits/ptrace-shared.h: ... new file. * sysdeps/unix/sysv/linux/Makefile (sysdep_headers): Add bits/ptrace-shared.h. * sysdeps/unix/sysv/linux/aarch64/sys/ptrace.h: Include <bits/ptrace-shared.h>. (__ptrace_setoptions, __ptrace_eventcodes, __ptrace_peeksiginfo_args, __ptrace_peeksiginfo_flags, ptrace): Remove. * sysdeps/unix/sysv/linux/ia64/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/powerpc/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/s390/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/sparc/sys/ptrace.h: Likewise. * sysdeps/unix/sysv/linux/tile/sys/ptrace.h: Likewise.
* scandir: fix wrong assumption about errno [BZ #17804]Aurelien Jarno2017-12-292-5/+14
| | | | | | | | | | | | | | | malloc and realloc may set errno to ENOMEM even if they are successful. The scandir code wrongly assume that they do not change errno, this causes scandir to fail with ENOMEM even if malloc succeed. The code already handles that readdir might set errno by calling __set_errno (0) to clear the error. Move that part at the end of the loop to also take malloc and realloc into account. Changelog: [BZ #17804] * dirent/scandir-tail.c (SCANDIR_TAIL): Move __set_errno (0) at the end of the loop. Improve comments.
* manual: clarify errno value on success [BZ #22615]Aurelien Jarno2017-12-292-14/+20
| | | | | | | | | | | | | The current glibc manual is ambiguous about the errno value on success and suggests that it is left unchanged. Some functions might and sometimes do change the errno value, however they never set it to 0. This patch from Zack Weinberg clarifies this section of the manual. Changelog: [BZ #22615] * manual/errno.texi (Checking for Errors): Explicitly say that errno might be set on success.
* tst-realloc: do not check for errno on success [BZ #22611]Aurelien Jarno2017-12-292-4/+6
| | | | | | | | | | | POSIX explicitly says that applications should check errno only after failure, so the errno value can be clobbered on success as long as it is not set to zero. Changelog: [BZ #22611] * malloc/tst-realloc.c (do_test): Remove the test checking that errno is unchanged on success.
* elf: remove redundant is_path argumentDmitry V. Levin2017-12-275-24/+27
| | | | | | | | | | | is_path argument is no longer used and could be safely removed. * elf/dl-dst.h (DL_DST_COUNT): Remove is_path argument, all callers updated. * elf/dl-load.c (is_dst, _dl_dst_count, _dl_dst_substitute, expand_dynamic_string_token): Likewise. * sysdeps/generic/ldsodefs.h (_dl_dst_count, _dl_dst_substitute): Remove is_path argument.
* elf: remove redundant code from is_dstDmitry V. Levin2017-12-272-5/+6
| | | | | | | | | | | | | | | | | is_dst is called either by _dl_dst_count or by _dl_dst_substitute. _dl_dst_count is called by DL_DST_COUNT only. DL_DST_COUNT is called either by expand_dst with is_path == 0 or by expand_dynamic_string_token. _dl_dst_substitute is called either from expand_dst with is_path == 0 or from expand_dynamic_string_token. The latter function is called either from _dl_map_object with is_path == 0 or from fillin_rpath with is_path == 1 and name containing no ':'. In any case (is_path && name[i] == ':') is always false and all code depending on it can be safely removed. * elf/dl-load.c (is_dst): Remove checks that is_path is set and name contains ':', and all code depending on these checks.
* elf: remove redundant code from _dl_dst_substituteDmitry V. Levin2017-12-272-20/+6
| | | | | | | | | | | | | | There are just two users of _dl_dst_substitute: one is expand_dst that sets is_path argument to 0, another one is expand_dynamic_string_token. The latter function also has just two users: one is _dl_map_object that sets is_path argument to 0, another one is fillin_rpath that sets is_path argument to 1 and name argument contains no ':'. In any case (is_path && name[i] == ':') is always false and all code depending on it can be safely removed. * elf/dl-load.c (_dl_dst_substitute): Remove checks that is_path is set and name contains ':', and all code depending on these checks.
* Deprecate external use of libio.h and _G_config.h.Zack Weinberg2017-12-2421-524/+626
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | libio.h was originally the header for a set of supported GNU extensions, but they have not been maintained as such in many years, they are now standing in the way of improvements to stdio, and we don't think there are any remaining external users. _G_config.h was never intended for public use, but predates the bits convention. Move both of these headers into the bits directory and provide stubs at top level which issue deprecation warnings. The contents of (bits/)libio.h and (bits/)_G_config.h are still exposed to external software via stdio.h; changing that requires more complex surgery than I have time to attempt right now. * libio/libio.h, libio/_G_config.h: New stub headers which issue a deprecation warning and then include <bits/libio.h>, <bits/_G_config.h> respectively. * libio/libio.h: Rename the original version of this file to libio/bits/libio.h. Error out if not included by stdio.h or the stub libio.h. * include/libio.h: Move to include/bits. Forward to libio/bits/libio.h. * sysdeps/generic/_G_config.h: Move to top-level bits/. Error out if not included by bits/libio.h or the stub _G_config.h. * sysdeps/unix/sysv/linux/_G_config.h: Move to sysdeps/unix/sysv/linux/bits. Error out if not included by bits/libio.h or the stub _G_config.h. * libio/stdio.h: Include bits/libio.h, not libio.h. * libio/Makefile: Install bits/libio.h and bits/_G_config.h as well as libio.h and _G_config.h. * csu/init.c, libio/fmemopen.c, libio/iolibio.h, libio/oldfmemopen.c * libio/strfile.h, stdio-common/vfscanf.c * sysdeps/pthread/flockfile.c, sysdeps/pthread/funlockfile.c Include stdio.h, not _G_config.h nor libio.h. * libio/iofgetpos.c: Also rename fgetpos64 out of the way. * libio/iofsetpos.c: Also rename fsetpos64 out of the way. * scripts/check-installed-headers.sh: Skip libio.h and _G_config.h.
* getrandom: fix comment [BZ #22347]Dmitry V. Levin2017-12-233-4/+10
| | | | | * stdlib/getrandom.c (getrandom): Fix comment. * sysdeps/unix/sysv/linux/getrandom.c (getrandom): Likewise.
* manual: fix a typo in strtoul description [BZ #21161]Aurelien Jarno2017-12-232-1/+4
| | | | | | | | Typo reported by Vincent Lefèvre: 'retrict' -> 'restrict'. Changelog: [BZ #21161] * manual/arith.texi (strtoul): Fix a typo.
* manual: fix finite description [BZ #22596]Aurelien Jarno2017-12-232-2/+8
| | | | | | | | | | That way it matches the standard and the behaviour of the finite function. Changelog: [BZ #22596] * manual/arith.texi (finite): Fix the description of the return value.
* Avoid gcc warnings on cygwinEric Blake2017-12-223-1/+10
| | | | | | * posix/regex_internal.c (re_string_reconstruct) [!RE_ENABLE_I18N]: * posix/regexec.c (check_arrival_add_next_nodes) [!RE_ENABLE_I18N]: Avoid unused variable.
* copy_file_range: New function to copy file dataFlorian Weimer2017-12-2244-3/+1340
| | | | | The semantics are based on the Linux system call, but a very close emulation in user space is provided.
* Add aarch64 disable-multi-arch variant to build-many-glibcs.pySzabolcs Nagy2017-12-212-1/+8
| | | | | | | aarch64 has several ifuncs now so test it without multiarch support separately. * scripts/build-many-glibcs.py (Context.add_all_configs): Add disable-multi-arch variant to aarch64-linux-gnu.
* Update miscellaneous files from upstream sources.Joseph Myers2017-12-206-179/+240
| | | | | | | | | | | | | | This patch updates various miscellaneous files from their upstream sources. Tested for x86_64, including "make pdf". * manual/texinfo.tex: Update to version 2017-12-18.20 with trailing whitespace removed. * scripts/config.guess: Update to version 2017-12-17. * scripts/config.sub: Update to version 2017-11-23. * scripts/install-sh: Update to version 2017-09-23.17. * scripts/move-if-change: Update to version 2017-09-13 06:45.
* Simplify tile assembly definitionsAdhemerval Zanella2017-12-2017-219/+236
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With tilepro removal, the uppercase instruction are not anymore required to be defines as potentially macros. This is a mechanical change done by the following shell script: --- INSNS="LD LD4U ST ST4 BNEZ BEQZ BEQZT BGTZ CMPEQI CMPEQ CMOVEQZ CMOVNEZ" FILES=$(find sysdeps/tile sysdeps/unix/sysv/linux/tile -iname *.S) for insn in $INSNS; do repl=$(echo $insn | tr '[:upper:]' '[:lower:]') sed -i 's/\b'$insn'\b/'$repl'/g' $FILES done --- Checked with a build for tilegx-linux-gnu and tilegx-linux-gnu-32 with and without the patch, there is no difference in generated binary with a dissassemble. * sysdeps/tile/__longjmp.S (__longjmp): Use lowercase instructions. * sysdeps/tile/__tls_get_addr.S (__tls_get_addr): Likewise. * sysdeps/tile/_mcount.S (__mcount): Likewise. * sysdeps/tile/crti.S (_init, _fini): Likewise. * sysdeps/tile/crtn.S: Likewise. * sysdeps/tile/dl-start.S (_start): Likewise. * sysdeps/tile/dl-trampoline.S: Likewise. * sysdeps/tile/setjmp.S (__sigsetjmp): Likewise. * sysdeps/tile/start.S (_start): Likewise. * sysdeps/unix/sysv/linux/tile/clone.S (_clone): Likewise. * sysdeps/unix/sysv/linux/tile/getcontext.S (__getcontext): Likewise. * sysdeps/unix/sysv/linux/tile/ioctl.S (__ioctl): Likewise. * sysdeps/unix/sysv/linux/tile/setcontext.S (__setcontext): Likewise. * sysdeps/unix/sysv/linux/tile/swapcontext.S (__swapcontext): Likewise. * sysdeps/unix/sysv/linux/tile/syscall.S (syscall): Likewise. * sysdeps/unix/sysv/linux/tile/vfork.S (__vfork): Likewise.
* Simplify tilegx sysdeps folderAdhemerval Zanella2017-12-2068-121/+114
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With tilepro support removal we can now simplify internal tile support by moving the directory structure to avoid the unnecessary directory levels in tile/tilegx both on generic and linux folders. Checked with a build for tilegx-linux-gnu and tilegx-linux-gnu-32 with and without the patch, there is no difference in generated binary with a dissassemble. * stdlib/bug-getcontext.c (do_test): Remove tilepro mention in comment. * sysdeps/tile/preconfigure: Remove tilegx folder. * sysdeps/tile/tilegx/Implies: Move definitions to ... * sysdeps/tile/Implies: ... here. * sysdeps/tile/tilegx/Makefile: Move rules to ... * sysdeps/tile/Makefile: ... here. * sysdeps/tile/tilegx/atomic-machine.h: Move definitions to ... * sysdeps/tile/atomic-machine.h: ... here. Add include guards. * sysdeps/tile/tilegx/bits/wordsize.h: Move to ... * sysdeps/tile/bits/wordsize.h: ... here. * sysdeps/tile/tilegx/*: Move to ... * sysdeps/tile/*: ... here. * sysdeps/tile/tilegx/tilegx32/Implies: Move to ... * sysdeps/tile/tilegx32/Implies: ... here. * sysdeps/tile/tilegx/tilegx64/Implies: Move to ... * sysdeps/tile/tilegx64/Implies: ... here. * sysdeps/unix/sysv/linux/tile/tilegx/Makefile: Move definitions to ... * sysdeps/unix/sysv/linux/tile/Makefile: ... here. * sysdeps/unix/sysv/linux/tile/tilegx/*: Move to ... * sysdeps/unix/sysv/linux/tile/*: ... here. * sysdeps/unix/sysv/linux/tile/tilegx/tilegx32/*: Move to ... * sysdeps/unix/sysv/linux/tile/tilegx32/*: ... here. * sysdeps/unix/sysv/linux/tile/tilegx/tilegx64/*: Move to ... * sysdeps/unix/sysv/linux/tile/tilegx64/*: ... here.
* Remove tilepro-*-linux-gnu supportAdhemerval Zanella2017-12-2053-5106/+98
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As from previous discussions [1] this patch removes tileprox-*-linux-gnu support from GLIBC. This patch is a straigthfoward one, which just remove tilepro specific implementation and configurations (no sysdep simplfication or reorganization is done). * README: Remove tilepro-*-linux-gnu from supported architecture. * scripts/build-many-glibcs.py: Likewise. * sysdeps/tile/__tls_get_addr.S (__tls_get_addr): Likewise. * sysdeps/tile/crti.S (PREINIT_FUNCTION): Likewise. * sysdeps/tile/dl-machine.h (ELF_MACHINE_NAME, elf_machine_matches_host, elf_machine_dynamic, elf_machine_load_address, elf_machine_runtime_setup, reloc_howto howto, elf_machine_rela): Likewise * sysdeps/tile/dl-start.S (_start): Likewise. * sysdeps/tile/memcmp.c (DBLALIGN, REVBYTES): Likewise. * sysdeps/tile/memcopy.h (MEMCPY_OK_FOR_FWD_MEMMOVE, MEMCPY_OK_FOR_FWD_MEMMOVE, op_t): Likewise. * sysdeps/tile/nptl/pthread_spin_lock.c (TNS, CMPTNS): Likewise. * sysdeps/tile/nptl/pthread_spin_trylock.c (TNS): Likewise. * sysdeps/tile/nptl/pthread_spin_unlock.c (pthread_spin_unlock): Likewise. * sysdeps/tile/nptl/tls.h (DB_THREAD_SELF): Likewise. * sysdeps/tile/preconfigure: Likewise. * sysdeps/tile/stackguard-macros.h (STACK_CHK_GUARD, POINTER_CHK_GUARD): Likewise. * sysdeps/tile/stackinfo.h (__stackinfo_sub): Likewise. * sysdeps/tile/start.S (_start): Likewise. * sysdeps/tile/tls-macros.h (TLS_GD_OFFSET, TLS_IE_OFFSET, _TLS_LE): Likewise. * sysdeps/tile/sysdep.h (REGSIZE): Likewise. (LD, LD4U, ST, ST4, BNEZ, BEQZ, BEQZT, BGTZ, CMPEQI, CMPEQ, CMOVEQZ, CMOVNEZ): Remove. * sysdeps/unix/sysv/linux/tile/bits/environments.h (__ILP32_OFF32_CFLAGS, __ILP32_OFFBIG_CFLAGS, __ILP32_OFF32_LDFLAGS, __ILP32_OFFBIG_LDFLAGS, __LP64_OFF64_CFLAGS, __LP64_OFF64_LDFLAGS): Likewise. * sysdeps/tile/wordcopy.c (DBLALIGN): Likewise. * sysdeps/tile/tilepro/Implies: Remove file. * sysdeps/tile/tilepro/atomic-machine.h: Likewise. * sysdeps/tile/tilepro/bits/wordsize.h: Likewise. * sysdeps/tile/tilepro/memchr.c: Likewise. * sysdeps/tile/tilepro/memcpy.S: Likewise. * sysdeps/tile/tilepro/memset.c: Likewise. * sysdeps/tile/tilepro/memusage.h: Likewise. * sysdeps/tile/tilepro/rawmemchr.c: Likewise. * sysdeps/tile/tilepro/strchr.c: Likewise. * sysdeps/tile/tilepro/strchrnul.c: Likewise. * sysdeps/tile/tilepro/strlen.c: Likewise. * sysdeps/tile/tilepro/strrchr.c: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/Implies: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/c++-types.data: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/jmp_buf-macros.h: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/ld.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/ldconfig.h: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libBrokenLocale.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libanl.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libc.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libcrypt.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libdl.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libm.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libnsl.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libpthread.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libresolv.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/librt.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libthread_db.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/libutil.abilist: Likewise. * sysdeps/unix/sysv/linux/tile/tilepro/register-dump.h: Likewise. * sysdeps/unix/sysv/linux/tile/sysconf.c (linux_sysconf): Remove tilepro mention in comment. [1] https://sourceware.org/ml/libc-alpha/2017-12/msg00038.html
* nptl: Consolidate pthread_{timed,try}join{_np}Adhemerval Zanella2017-12-209-230/+145
| | | | | | | | | | | | | | | | | | | | | | | | | | This patch consolidates the pthread_join and gnu extensions to avoid code duplication. The function pthread_join, pthread_tryjoin_np, and pthread_timedjoin_np are now based on pthread_timedjoin_ex. It also fixes some inconsistencies on ESRCH, EINVAL, EDEADLK handling (where each implementation differs from each other) and also on clenup handler (which now always use a CAS). Checked on i686-linux-gnu and x86_64-linux-gnu. * nptl/pthreadP.h (__pthread_timedjoin_np): Define. * nptl/pthread_join.c (pthread_join): Use __pthread_timedjoin_np. * nptl/pthread_tryjoin.c (pthread_tryjoin): Likewise. * nptl/pthread_timedjoin.c (cleanup): Use CAS on argument setting. (pthread_timedjoin_np): Define internal symbol and common code from pthread_join. * sysdeps/unix/sysv/linux/i386/lowlevellock.h (__lll_timedwait_tid): Remove superflous checks. * sysdeps/unix/sysv/linux/x86_64/lowlevellock.h (__lll_timedwait_tid): Likewise. Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* aarch64: update libm-test-ulpsSzabolcs Nagy2017-12-202-20/+24
| | | | * sysdeps/aarch64/libm-test-ulps: Update.
* aarch64: fix memset with --disable-multi-archAdhemerval Zanella2017-12-202-0/+8
| | | | * sysdeps/aarch64/memset.S (MEMSET): Define.
* nptl: Implement pthread_self in libc.so [BZ #22635]Florian Weimer2017-12-207-11/+45
| | | | | | | | | | | | | | | | | All binaries use TLS and thus need a properly set up TCB, so we can simply return its address directly, instead of forwarding to the libpthread implementation from libc. For versioned symbols, the dynamic linker checks that the soname matches the name supplied by the link editor, so a compatibility symbol in libpthread is needed. To avoid linking against the libpthread function in all cases, we would have to bump the symbol version of libpthread in libc.so and supply a compat symbol. This commit does not do that because the function implementation is so small, so the overhead by two active copies of the same function might well be smaller than the increase in symbol table size.
* regex: Fix spelling in comments.Arnold Robbins2017-12-195-7/+14
| | | | | | | Fix the spelling in various comments throughout the regex implementation. These changes are also present in gnulib and will be integrated there also, see: https://sourceware.org/ml/libc-alpha/2017-12/msg00688.html
* Provide a C++ version of iseqsig (bug 22377)Gabriel F. T. Gomes2017-12-195-5/+198
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | In C++ mode, __MATH_TG cannot be used for defining iseqsig, because __MATH_TG relies on __builtin_types_compatible_p, which is a C-only builtin. This is true when float128 is provided as an ABI-distinct type from long double. Moreover, the comparison macros from ISO C take two floating-point arguments, which need not have the same type. Choosing what underlying function to call requires evaluating the formats of the arguments, then selecting which is wider. The macro __MATH_EVAL_FMT2 provides this information, however, only the type of the macro expansion is relevant (actually evaluating the expression would be incorrect). This patch provides a C++ version of iseqsig, in which only the type of __MATH_EVAL_FMT2 (__typeof or decltype) is used as a template parameter for __iseqsig_type. This function calls the appropriate underlying function. Tested for powerpc64le and x86_64. [BZ #22377] * math/Makefile [C++] (tests): Add test for iseqsig. * math/math.h [C++] (iseqsig): New implementation, which does not rely on __MATH_TG/__builtin_types_compatible_p. * math/test-math-iseqsig.cc: New file. * sysdeps/powerpc/powerpc64le/Makefile (CFLAGS-test-math-iseqsig.cc): New variable.
* elf: remove redundant __libc_enable_secure check from fillin_rpathDmitry V. Levin2017-12-192-30/+9
| | | | | | | | | | | | | | | | There are just two users of fillin_rpath: one is decompose_rpath that sets check_trusted argument to 0, another one is _dl_init_paths that sets check_trusted argument to __libc_enable_secure and invokes fillin_rpath only when LD_LIBRARY_PATH is non-empty. Starting with commit glibc-2.25.90-512-gf6110a8fee2ca36f8e2d2abecf3cba9fa7b8ea7d, LD_LIBRARY_PATH is ignored for __libc_enable_secure executables, so check_trusted argument of fillin_rpath is always zero. * elf/dl-load.c (is_trusted_path): Remove. (fillin_rpath): Remove check_trusted argument and its use, all callers changed.
* Replece LDFLAGS-* = $(no-pie-ldflag) with tst-*-no-pie = yes [BZ #22630]H.J. Lu2017-12-194-5/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After commit 9d7a3741c9e59eba87fb3ca6b9f979befce07826 Author: H.J. Lu <hjl.tools@gmail.com> Date: Fri Dec 15 16:59:33 2017 -0800 Add --enable-static-pie configure option to build static PIE [BZ #19574] and commit 00c714df398b63934540d95ce3792596f7a94a6c Author: H.J. Lu <hjl.tools@gmail.com> Date: Mon Dec 18 12:24:26 2017 -0800 Pass -no-pie to GCC only if GCC defaults to PIE [BZ #22614] $(no-pie-ldflag) is no longer effective since no-pie-ldflag is defined to -no-pie only if GCC defaults to PIE. When --enable-static-pie is used to configure glibc build and GCC doesn't default to PIE. no-pie-ldflag is undefined and these tests: elf/Makefile:LDFLAGS-tst-dlopen-aout = $(no-pie-ldflag) elf/Makefile:LDFLAGS-tst-prelink = $(no-pie-ldflag) elf/Makefile:LDFLAGS-tst-main1 = $(no-pie-ldflag) gmon/Makefile:LDFLAGS-tst-gmon := $(no-pie-ldflag) may fail to link. This patch replaces "-pie" with $(if $($(@F)-no-pie),$(no-pie-ldflag),-pie) and repleces LDFLAGS-* = $(no-pie-ldflag) with tst-*-no-pie = yes so that tst-dlopen-aout, tst-prelink, tst-main1 and tst-gmon are always built as non-PIE, with and without --enable-static-pie, regardless if GCC defaults to PIE or non-PIE. Tested with build-many-glibcs.py without --enable-static-pie as well as with --enable-static-pie for x86_64, x32 and i686. [BZ #22630] * Makeconfig (link-pie-before-libc): Replace -pie with $(if $($(@F)-no-pie),$(no-pie-ldflag),-pie). * elf/Makefile (LDFLAGS-tst-dlopen-aout): Removed. (tst-dlopen-aout-no-pie): New. (LDFLAGS-tst-prelink): Removed. (tst-prelink-no-pie): New. (LDFLAGS-tst-main1): Removed. (tst-main1-no-pie): New. * gmon/Makefile (LDFLAGS-tst-gmon): Removed. (tst-gmon-no-pie): New.
* Document that --enable-static-pie implies PIEH.J. Lu2017-12-193-2/+11
| | | | | | | | | | | | | | To build static PIE, all .o files are compiled with -fPIE. Since --enable-static-pie is designed to provide additional security hardening benefits, it also implies that glibc programs and tests are created as dynamic position independent executables (PIE) by default for better security hardening. Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> * manual/install.texi: Document that --enable-static-pie implies PIE. * INSTALL: Regenerated.
* Avoid signed shift overflow in pow (bug 21309).Bernd Edlinger2017-12-192-1/+8
| | | | | | | | | | | | | | As noted in bug 21309, dbl-64/e_pow.c contains signed int shifts that, although the shift count is in the range [0, 31], shift bits into and beyond the sign bit and so are undefined in ISO C. Although this is defined in GNU C, this patch from the bug cleans up the code to avoid those shifts. Tested for x86_64. [BZ #21309] * sysdeps/ieee754/dbl-64/e_pow.c (checkint): Make m and n unsigned.
* Revert exp reimplementation (causes test failures).Joseph Myers2017-12-1920-485/+376
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Revert: 2017-12-19 Joseph Myers <joseph@codesourcery.com> * sysdeps/x86_64/fpu/libm-test-ulps: Update. 2017-12-19 Patrick McGehearty <patrick.mcgehearty@oracle.com> * sysdeps/ieee754/dbl-64/e_exp.c: Include <math-svid-compat.h> and <errno.h>. Include "eexp.tbl". (half): New constant. (one): Likewise. (__ieee754_exp): Rewrite. (__slowexp): Remove prototype. * sysdeps/ieee754/dbl-64/eexp.tbl: New file. * sysdeps/ieee754/dbl-64/slowexp.c: Remove file. * sysdeps/i386/fpu/slowexp.c: Likewise. * sysdeps/ia64/fpu/slowexp.c: Likewise. * sysdeps/m68k/m680x0/fpu/slowexp.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-avx.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-fma.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-fma4.c: Likewise. * sysdeps/generic/math_private.h (__slowexp): Remove prototype. * sysdeps/ieee754/dbl-64/e_pow.c: Remove mention of slowexp.c in comment. * sysdeps/powerpc/power4/fpu/Makefile [$(subdir) = math] (CPPFLAGS-slowexp.c): Remove variable. * sysdeps/x86_64/fpu/multiarch/Makefile (libm-sysdep_routines): Remove slowexp-fma, slowexp-fma4 and slowexp-avx. (CFLAGS-slowexp-fma.c): Remove variable. (CFLAGS-slowexp-fma4.c): Likewise. (CFLAGS-slowexp-avx.c): Likewise. * sysdeps/x86_64/fpu/multiarch/e_exp-avx.c (__slowexp): Do not define as macro. * sysdeps/x86_64/fpu/multiarch/e_exp-fma.c (__slowexp): Likewise. * sysdeps/x86_64/fpu/multiarch/e_exp-fma4.c (__slowexp): Likewise. * math/Makefile (type-double-routines): Remove slowexp. * manual/probes.texi (slowexp_p6): Remove. (slowexp_p32): Likewise.
* glob: Silence warning about void pointer arithmeticAdhemerval Zanella2017-12-192-1/+5
| | | | | | | | | | | Sync with gnulib 0e14f025d2. Checked on x86_64-linux-gnu. * lib/glob.c (glob): Use a 'char *', not a 'void *', in pointer arithmetic. Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* glob: pacify fuzzer for mempcpyAdhemerval Zanella2017-12-192-2/+9
| | | | | | | | | | | | Problem reported by Tim Rühsen [1]. Sync with gnulib 0e14f025d2. [1] https://lists.gnu.org/archive/html/bug-gnulib/2017-10/msg00054.html Checked on x86_64-linux-gnu. * lib/glob.c (glob): Do not pass NULL to mempcpy. Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Update x86_64 libm-test-ulps.Joseph Myers2017-12-192-8/+14
| | | | * sysdeps/x86_64/fpu/libm-test-ulps: Update.
* Improve __ieee754_exp() performance by greater than 5x on sparc/x86.Patrick McGehearty2017-12-1919-326/+509
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These changes will be active for all platforms that don't provide their own exp() routines. They will also be active for ieee754 versions of ccos, ccosh, cosh, csin, csinh, sinh, exp10, gamma, and erf. Typical performance gains is typically around 5x when measured on Sparc s7 for common values between exp(1) and exp(40). Using the glibc perf tests on sparc, sparc (nsec) x86 (nsec) old new old new max 17629 395 5173 144 min 399 54 15 13 mean 5317 200 1349 23 The extreme max times for the old (ieee754) exp are due to the multiprecision computation in the old algorithm when the true value is very near 0.5 ulp away from an value representable in double precision. The new algorithm does not take special measures for those cases. The current glibc exp perf tests overrepresent those values. Informal testing suggests approximately one in 200 cases might invoke the high cost computation. The performance advantage of the new algorithm for other values is still large but not as large as indicated by the chart above. Glibc correctness tests for exp() and expf() were run. Within the test suite 3 input values were found to cause 1 bit differences (ulp) when "FE_TONEAREST" rounding mode is set. No differences in exp() were seen for the tested values for the other rounding modes. Typical example: exp(-0x1.760cd2p+0) (-1.46113312244415283203125) new code: 2.31973271630014299393707e-01 0x1.db14cd799387ap-3 old code: 2.31973271630014271638132e-01 0x1.db14cd7993879p-3 exp = 2.31973271630014285508337 (high precision) Old delta: off by 0.49 ulp New delta: off by 0.51 ulp In addition, because ieee754_exp() is used by other routines, cexp() showed test results with very small imaginary input values where the imaginary portion of the result was off by 3 ulp when in upward rounding mode, but not in the other rounding modes. For x86, tgamma showed a few values where the ulp increased to 6 (max ulp for tgamma is 5). Sparc tgamma did not show these failures. I presume the tgamma differences are due to compiler optimization differences within the gamma function.The gamma function is known to be difficult to compute accurately. * sysdeps/ieee754/dbl-64/e_exp.c: Include <math-svid-compat.h> and <errno.h>. Include "eexp.tbl". (half): New constant. (one): Likewise. (__ieee754_exp): Rewrite. (__slowexp): Remove prototype. * sysdeps/ieee754/dbl-64/eexp.tbl: New file. * sysdeps/ieee754/dbl-64/slowexp.c: Remove file. * sysdeps/i386/fpu/slowexp.c: Likewise. * sysdeps/ia64/fpu/slowexp.c: Likewise. * sysdeps/m68k/m680x0/fpu/slowexp.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-avx.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-fma.c: Likewise. * sysdeps/x86_64/fpu/multiarch/slowexp-fma4.c: Likewise. * sysdeps/generic/math_private.h (__slowexp): Remove prototype. * sysdeps/ieee754/dbl-64/e_pow.c: Remove mention of slowexp.c in comment. * sysdeps/powerpc/power4/fpu/Makefile [$(subdir) = math] (CPPFLAGS-slowexp.c): Remove variable. * sysdeps/x86_64/fpu/multiarch/Makefile (libm-sysdep_routines): Remove slowexp-fma, slowexp-fma4 and slowexp-avx. (CFLAGS-slowexp-fma.c): Remove variable. (CFLAGS-slowexp-fma4.c): Likewise. (CFLAGS-slowexp-avx.c): Likewise. * sysdeps/x86_64/fpu/multiarch/e_exp-avx.c (__slowexp): Do not define as macro. * sysdeps/x86_64/fpu/multiarch/e_exp-fma.c (__slowexp): Likewise. * sysdeps/x86_64/fpu/multiarch/e_exp-fma4.c (__slowexp): Likewise. * math/Makefile (type-double-routines): Remove slowexp. * manual/probes.texi (slowexp_p6): Remove. (slowexp_p32): Likewise.
* ia64: Fix memchr for large input sizes (BZ #22603)Adhemerval Zanella2017-12-192-0/+11
| | | | | | | | | | | | | | | | | | | | | | Current optimized ia64 memchr uses a strategy to check for last address by adding the input one with expected size. However it does not take care for possible overflow. It was triggered by 3038145ca23 where default rawmemchr now uses memchr (p, c, (size_t)-1). This patch fixes it by implement a satured addition where overflows sets the maximum pointer size to UINTPTR_MAX. Checked on ia64-linux-gnu where it fixes both stratcliff and test-rawmemchr failures. Adhemerval Zanella <adhemerval.zanella@linaro.org> James Clarke <jrtc27@jrtc27.com> [BZ #22603] * sysdeps/ia64/memchr.S (__memchr): Avoid overflow in pointer addition.
* sh: Fix clone exit return code (BZ #22605)Adhemerval Zanella2017-12-192-0/+7
| | | | | | | | | | | | | | | | Since 3f823e87cc (Call exit directly in clone (BZ #21512)) SH clone implementation fails to set the exit code resulting in the failures: FAIL: nptl/tst-align-clone FAIL: nptl/tst-getpid1 This patch fixes the both testcases. [BZ #22605] * sysdeps/unix/sysv/linux/sh/clone.S (__clone): Fix exit return code. Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* x86: Add feature_1 to tcbhead_t [BZ #22563]H.J. Lu2017-12-195-1/+19
| | | | | | | | | | | | | | | | | | On x86, padding in struct __jmp_buf_tag is used for shadow stack pointer to support Shadow Stack in Intel Control-flow Enforcemen Technology. cancel_jmp_buf has been updated to include saved_mask so that it is as large as struct __jmp_buf_tag. We must suport the old cancel_jmp_buf in existing binaries. Since symbol versioning doesn't work on cancel_jmp_buf, feature_1 is added to tcbhead_t so that setjmp and longjmp can check if shadow stack is enabled. NB: Shadow stack is enabled only if all modules are shadow stack enabled. [BZ #22563] * sysdeps/i386/nptl/tcb-offsets.sym (FEATURE_1_OFFSET): New. * sysdeps/i386/nptl/tls.h (tcbhead_t): Add feature_1. * sysdeps/x86_64/nptl/tcb-offsets.sym (FEATURE_1_OFFSET): New. * sysdeps/x86_64/nptl/tls.h (tcbhead_t): Rename __glibc_unused1 to feature_1.
* Linux/x86: Update cancel_jmp_buf to match __jmp_buf_tag [BZ #22563]H.J. Lu2017-12-199-11/+143
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On x86, padding in struct __jmp_buf_tag is used for shadow stack pointer to support shadow stack in Intel Control-flow Enforcemen Technology. Since the cancel_jmp_buf array is passed to setjmp and longjmp by casting it to pointer to struct __jmp_buf_tag, it should be as large as struct __jmp_buf_tag. Otherwise when shadow stack is enabled, setjmp and longjmp will write and read beyond cancel_jmp_buf when saving and restoring shadow stack pointer. This patch adds bits/types/__cancel_jmp_buf_tag.h to define struct __cancel_jmp_buf_tag so that Linux/x86 can add saved_mask to cancel_jmp_buf. Tested natively on i386, x86_64 and x32. Tested hppa-linux-gnu with build-many-glibcs.py. [BZ #22563] * bits/types/__cancel_jmp_buf_tag.h: New file. * sysdeps/unix/sysv/linux/x86/bits/types/__cancel_jmp_buf_tag.h * sysdeps/unix/sysv/linux/x86/pthreaddef.h: Likewise. * sysdeps/unix/sysv/linux/x86/nptl/pthreadP.h: Likewise. * nptl/Makefile (headers): Add bits/types/__cancel_jmp_buf_tag.h. * nptl/descr.h [NEED_SAVED_MASK_IN_CANCEL_JMP_BUF] (pthread_unwind_buf): Add saved_mask to cancel_jmp_buf. * sysdeps/nptl/pthread.h: Include <bits/types/__cancel_jmp_buf_tag.h>. (__pthread_unwind_buf_t): Use struct __cancel_jmp_buf_tag with __cancel_jmp_buf. * sysdeps/unix/sysv/linux/hppa/pthread.h: Likewise.
* Add --enable-static-pie variants to x86_64, x32 and i686H.J. Lu2017-12-182-0/+14
| | | | | | | | | | | | Since the default GCC and binutils versions used by build-many-glibcs.py, which are GCC 7 branch and binutils 2.29 branch, support static PIE on x86_64, x32 and i686, this patch adds --enable-static-pie glibc variants to x86_64, x32 and i686 to get some coverage for static PIE. Tested with build-many-glibcs.py. * scripts/build-many-glibcs.py (Context.add_all_configs): Add --enable-static-pie variants to x86_64, x32 and i686.
* Fix m68k bits/mathinline.h attributes (bug 22631).Joseph Myers2017-12-193-68/+94
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | m68k bits/mathinline.h declares various functions with const attributes. These are inappropriate for functions that have results depending on the rounding mode; the machine-independent bits/mathcalls.h only uses const attributes for a very few functions with no rounding mode dependence, and the m68k header should do likewise. GCC uses pure for such functions with -frounding-math, resulting in GCC mainline warning for conflicts with between the header and the built-in attributes and glibc failing to build for m68k with GCC mainline. This patch fixes the attributes to avoid using const except when bits/mathcalls.h does so. (There are a few functions where maybe bits/mathcalls.h could do so but doesn't, but keeping the headers in sync in this regard seems to be the safe approach.) Tested compilation with build-many-glibcs.py with GCC mainline. [BZ #22631] * sysdeps/m68k/m680x0/fpu/bits/mathinline.h (__m81_defun): Add argument for attrubutes. All callers changed. (__inline_mathop1): Likewise. All callers changed. (__inline_mathop): Likewise. All callers changed. [__USE_MISC] (scalbn): Use __inline_forward instead of __inline_forward_c. [__USE_ISOC99] (scalbln): Likewise. [__USE_ISOC99] (nearbyint): Likewise. [__USE_ISOC99] (lrint): Likewise. [__USE_MISC] (scalbnf): Likewise. [__USE_ISOC99] (scalblnf): Likewise. [__USE_ISOC99] (nearbyintf): Likewise. [__USE_ISOC99] (lrintf): Likewise. [__USE_MISC] (scalbnl): Likewise. [__USE_ISOC99] (scalblnl): Likewise. [__USE_ISOC99] (nearbyintl): Likewise. [__USE_ISOC99] (lrintl): Likewise. * sysdeps/m68k/m680x0/fpu/mathimpl.h: All callers of __inline_mathop and __m81_defun changed.
* Fix build-many-glibcs.py arm-linux-gnueabihf builds with mainline GCC.Joseph Myers2017-12-192-5/+11
| | | | | | | | | | | | | | | | | My fix to make the arm-linux-gnueabihf build-many-glibcs.py builds actually use the hard-float ABI as intended showed up another issue when building with mainline GCC: GCC now determines an FPU based on the selected CPU or architecture and gives an error for -mfloat-abi=hard when the CPU does not imply a choice of FPU. This patch fixes all the affected configurations to specify a suitable --with-cpu, --with-fpu or -mfpu option explicitly to avoid that error from GCC. Tested the relevant configurations with build-many-glibcs.py with mainline GCC. * scripts/build-many-glibcs.py (Context.add_all_configs): Specify CPU or FPU for ARM hard-float configurations.
* Disable -Wrestrict for two nptl/tst-attr3.c tests.Joseph Myers2017-12-182-0/+22
| | | | | | | | | | | | | | | | | | | | nptl/tst-attr3 fails to build with GCC mainline because of (deliberate) aliasing between the second (attributes) and fourth (argument to thread start routine) arguments to pthread_create. Although both those arguments are restrict-qualified in POSIX, pthread_create does not actually dereference its fourth argument; it's an opaque pointer passed to the thread start routine. Thus, the aliasing is actually valid in this case, and it's deliberate in the test. So this patch makes the test disable -Wrestrict for the two pthread_create calls in question. (-Wrestrict was added in GCC 7, hence the __GNUC_PREREQ conditions, but the particular warning in question is new in GCC 8.) Tested compilation with build-many-glibcs.py for aarch64-linux-gnu. * nptl/tst-attr3.c: Include <libc-diag.h>. (do_test) [__GNUC_PREREQ (7, 0)]: Ignore -Wrestrict for two tests.
* Fix truncation warnings in posix/tst-glob_symlinks.c.Joseph Myers2017-12-182-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | The test posix/tst-glob_symlinks.c fails to build with GCC mainline: tst-glob_symlinks.c: In function 'do_test': tst-glob_symlinks.c:124:30: error: 'snprintf' output may be truncated before the last format character [-Werror=format-truncation=] snprintf (buf, sizeof buf, "%s?", dangling_link); ^~~~~ tst-glob_symlinks.c:124:3: note: 'snprintf' output between 2 and 4097 bytes into a destination of size 4096 snprintf (buf, sizeof buf, "%s?", dangling_link); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ tst-glob_symlinks.c:128:30: error: 'snprintf' output may be truncated before the last format character [-Werror=format-truncation=] snprintf (buf, sizeof buf, "%s*", dangling_link); ^~~~~ tst-glob_symlinks.c:128:3: note: 'snprintf' output between 2 and 4097 bytes into a destination of size 4096 snprintf (buf, sizeof buf, "%s*", dangling_link); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This patch fixes the test to avoid such truncation warnings by increasing the buffer in question by one byte, to ensure it can hold any possible result of %s? or %s* formats where %s comes from a buffer of size PATH_MAX. Tested compilation with build-many-glibcs.py for aarch64-linux-gnu. * posix/tst-glob_symlinks.c (do_test): Increase size of buf.
* Disable strncat test array-bounds warnings for GCC 8.Joseph Myers2017-12-182-2/+11
| | | | | | | | | | | | | | | | Some strncat tests fail to build with GCC 8 because of -Warray-bounds warnings. These tests are deliberately test over-large size arguments passed to strncat, and already disable -Wstringop-overflow warnings, but now the warnings for these tests come under -Warray-bounds so that option needs disabling for them as well, which this patch does (with an update on the comments; the DIAG_IGNORE_NEEDS_COMMENT call for -Warray-bounds doesn't need to be conditional itself, because that option is supported by all versions of GCC that can build glibc). Tested compilation with build-many-glibcs.py for aarch64-linux-gnu. * string/tester.c (test_strncat): Also disable -Warray-bounds warnings for two tests.
* Pass -no-pie to GCC only if GCC defaults to PIE [BZ #22614]H.J. Lu2017-12-185-9/+28
| | | | | | | | | | | | | | | | | | | | After --enable-static-pie is added to configure, libc_cv_pie_default is set to yes when either --enable-static-pie is used to configure glibc or GCC defaults to PIE. We should set no-pie-ldflag to -no-pie, which is supported on GCC 6 and later, only if GCC defaults to PIE, not when --enable-static-pie is used to configure glibc. Tested on x32 with --enable-static-pie using GCC 5 and without --enable-static-pie using GCC 7. [BZ #22614] * Makeconfig (no-pie-ldflag): Set to -no-pie only if $(cc-pie-default) == yes. * config.make.in (cc-pie-default): New. * configure.ac (libc_cv_pie_default): Renamed to ... (libc_cv_cc_pie_default): This. (libc_cv_pie_default): Set to $libc_cv_cc_pie_default. * configure: Regenerated.
* ld.so: Examine GLRO to detect inactive loader [BZ #20204]Florian Weimer2017-12-1815-16/+61
| | | | | | | | | | | | GLRO (_rtld_global_ro) is read-only after initialization and can therefore not be patched at run time, unlike the hook table addresses and their contents, so this is a desirable hardening feature. The hooks are only needed if ld.so has not been initialized, and this happens only after static dlopen (dlmopen uses a single ld.so object across all namespaces). Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* Fix nscd readlink argument aliasing (bug 22446).Joseph Myers2017-12-182-2/+9
| | | | | | | | | | | | | | | | Current GCC mainline detects that nscd calls readlink with the same buffer for both input and output, which is not valid (those arguments are both restrict-qualified in POSIX). This patch makes it use a separate buffer for readlink's input (with a size that is sufficient to avoid truncation, so there should be no problems with warnings about possible truncation, though not strictly minimal, but much smaller than the buffer for output) to avoid this problem. Tested compilation for aarch64-linux-gnu with build-many-glibcs.py. [BZ #22446] * nscd/connections.c (handle_request) [SO_PEERCRED]: Use separate buffers for readlink input and output.
View Lorry Controller Status