about summary refs log tree commit diff
Commit message (Collapse)AuthorAgeFilesLines
* Remove IS_IN_librtSiddhesh Poyarekar2014-11-2425-46/+73
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Replace with IS_IN (librt). Generated code unchanged on x86_64 * include/mqueue.h: Use IS_IN instead of IS_IN_librt. * nptl/pthreadP.h: Likewise. * sysdeps/unix/sysv/linux/aarch64/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/alpha/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/arm/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/hppa/nptl/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/i386/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/ia64/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/ia64/sysdep.h: Likewise. * sysdeps/unix/sysv/linux/m68k/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/microblaze/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/mips/mips64/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/mips/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/not-cancel.h: Likewise. * sysdeps/unix/sysv/linux/powerpc/powerpc32/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/powerpc/powerpc64/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/s390/s390-32/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/s390/s390-64/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/sh/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/sparc/sparc32/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/sparc/sparc64/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/tile/sysdep-cancel.h: Likewise. * sysdeps/unix/sysv/linux/tile/waitpid.S: Likewise. * sysdeps/unix/sysv/linux/x86_64/sysdep-cancel.h: Likewise.
* Remove IS_IN_libdlSiddhesh Poyarekar2014-11-2410-9/+19
| | | | | | | | | | | | | | Replace with IS_IN (libdl). No changes to generated code on x86_64. * dlfcn/dladdr.c: Use IS_IN. * dlfcn/dladdr1.c: Likewise. * dlfcn/dlclose.c: Likewise. * dlfcn/dlerror.c: Likewise. * dlfcn/dlinfo.c: Likewise. * dlfcn/dlmopen.c: Likewise. * dlfcn/dlopen.c: Likewise. * dlfcn/dlsym.c: Likewise. * dlfcn/dlvsym.c: Likewise.
* Remove IS_IN_nscdSiddhesh Poyarekar2014-11-245-5/+10
| | | | | | | | | Replace with IS_IN (nscd). Generated code unchanged on x86_64. * include/ifaddrs.h: Use IS_IN. * inet/check_pf.c: Likewise. * sysdeps/unix/sysv/linux/check_pf.c: Likewise. * nscd/Makefile (CPPFLAGS-nscd): Remove IS_IN_nscd.
* Remove IS_IN_ldconfigSiddhesh Poyarekar2014-11-244-3/+8
| | | | | | | | | Replace with IS_IN (ldconfig). No change in generated code. * elf/Makefile (CFLAGS-ldconfig.c): Remove definition of IS_IN_ldconfig. * sysdeps/unix/sysv/linux/x86_64/dl-procinfo.c: Use IS_IN. * sysdeps/unix/sysv/linux/x86_64/dl-procinfo.h: Likewise.
* Remove IS_IN_libcSiddhesh Poyarekar2014-11-243-5/+5
| | | | | | | | | Replace it with IS_IN (libc) and remove the one place that it is defined in. The generated code remains unchanged on x86_64. * include/shlib-compat.h [!NOT_IN_libc]: Remove. * nss/nss_files/files-parse.c (IS_IN_libc): Replace with IS_IN (libc).
* Define IN_MODULE for translation units that define NOT_IN_libcSiddhesh Poyarekar2014-11-242-0/+3
| | | | | | | | | Make sure that all instances where NOT_IN_libc is defined also defines IN_MODULE to facilitate removal NOT_IN_libc in future passes. Verified that the generated code is unchanged on x86_64. * elf/Makefile (libof-sotruss-lib): Set as extramodules.
* Remove IN_LIBSiddhesh Poyarekar2014-11-248-19/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Replace with IS_IN and IS_IN_LIB macros instead. This change results in a change in generated code, because it fixes a subtle bug. The bug was introduced when systemtap probes were added to lowlevellock.h, which resulted in stap-probe.h being included in a number of places. stap-probe.h always defines IN_LIB, which breaks a check in errno.h and netdb.h since they rely on that macro to decide whether to implement an internal version of a declaration or an external one. The components that see a code change due to this are: iconv_prog libmemusage.so libpcprofile.so libSegFault.so libutil.so.1 locale localedef nscd All other built components (i.e. libc, libpthread, etc.) remain unchanged by this on x86_64. * elf/Makefile (CPPFLAGS-.os): Remove IN_LIB. * elf/rtld-Rules (rtld-CPPFLAGS): Likewise. * extra-lib.mk (CPPFLAGS-$(lib)): Likewise. * include/libc-symbols.h (IS_IN_LIB): New macro. * include/errno.h: Use IS_IN_LIB instead of IN_LIB. * include/netdb.h: Likewise. * include/stap-probe.h: Remove all uses of IN_LIB.
* Use MODULE_NAME in stap-probe instead of IN_LIBSiddhesh Poyarekar2014-11-244-6/+21
| | | | | | | | | | | | | | Define MODULE_NAME in the build command and define IN_MODULE using MODULE_NAME. Verified that the generated code is unchanged on x86_64. * Makeconfig (module-cppflags-real): Define MODULE_NAME instead of IN_MODULE. * include/libc-symbols.h (IN_MODULE): Define using MODULE_NAME. (PASTE_NAME, PASTE_NAME1): New macros. * include/stap-probe.h (LIBC_PROBE_1): Use MODULE_NAME instead of IN_LIB. (STAP_PROBE_ASM): Likewise.
* hurd: Fix dlopening libraries from static programsSamuel Thibault2014-11-232-0/+10
| | | | | | | | | | | | | dlopening libraries from a static program would dlopen libc.so, which thus needs its own initialization, done in posixland_init, which was missing initializing RPCs so far. ChangeLog: 2014-11-23 Samuel Thibault <samuel.thibault@ens-lyon.org> * sysdeps/mach/hurd/i386/init-first.c (posixland_init): Call __mach_init in dlopened libc.
* arm: drop EABI checkMike Frysinger2014-11-223-27/+5
| | | | | | GCC marked OABI obsolete in 4.7 and dropped it in 4.8. So the number of people this is catching is shrinking every day. At this point, it's not terribly useful, so just drop it.
* NPTL: Fix pthread_create regression from default-sched.h refactoring.Roland McGrath2014-11-212-3/+15
|
* AArch64: Reformat inline-asm in elf_machine_load_addressAndrew Pinski2014-11-212-9/+15
| | | | | | | | | | | | | | | | | This patch reformats the inline-asm in elf_machine_load_address so it is easier to change only part of the inline-asm. That is using string concatenating instead of string continuation. Also document why this inline-asm works - it depends on the 32bit relocation being resolved at link time. ChangeLog: 2014-11-21 Will Newton <will.newton@linaro.org> Andrew Pinski <andrew.pinski@caviumnetworks.com> * sysdeps/aarch64/dl-machine.h (elf_machine_load_address): Refactor inline-asm. Also add comment.
* AArch64: Use ELF macros rather than Elf64 throughoutWill Newton2014-11-213-5/+11
| | | | | | | | | | | | | | | | | Using the macros for ELF types is required for adding ILP32 support. In the standard AArch64 configuration this makes no difference to the types used. ChangeLog: 2014-11-21 Will Newton <will.newton@linaro.org> Andrew Pinski <andrew.pinski@caviumnetworks.com> * sysdeps/aarch64/bits/link.h (la_aarch64_gnu_pltenter): Use ElfW macro instead of hardcoded Elf64 types. (la_aarch64_gnu_pltenter): Likewise. * sysdeps/aarch64/dl-machine.h (elf_machine_runtime_setup): Use ElfW(Addr).
* AArch64: Update relocations for ILP32Will Newton2014-11-213-9/+42
| | | | | | | | | | | | | | | | | | | | | | | | | | | The latest version of the binutils ELF header defines a new set of dynamic relocations for ILP32 and renames some to make the naming more uniform. ChangeLog: 2014-11-21 Will Newton <will.newton@linaro.org> Andrew Pinski <andrew.pinski@caviumnetworks.com> * elf/elf.h (R_AARCH64_P32_ABS32, R_AARCH64_P32_COPY, R_AARCH64_P32_GLOB_DAT, R_AARCH64_P32_JUMP_SLOT, R_AARCH64_P32_RELATIVE, R_AARCH64_P32_TLS_DTPMOD, R_AARCH64_P32_TLS_DTPREL, R_AARCH64_P32_TLS_TPREL, R_AARCH64_P32_TLSDESC, R_AARCH64_P32_IRELATIVE): Define. (R_AARCH64_TLS_DTPMOD64): Rename to .. (R_AARCH64_TLS_DTPMOD): This. (R_AARCH64_TLS_DTPREL64): Rename to ... (R_AARCH64_TLS_DTPREL): This. (R_AARCH64_TLS_TPREL64): Rename to ... (R_AARCH64_TLS_TPREL): This. * sysdeps/aarch64/dl-machine.h (elf_machine_type_class): Update R_AARCH64_TLS_DTPMOD64, R_AARCH64_TLS_DTPREL64, and R_AARCH64_TLS_TPREL64. (elf_machine_rela): Likewise.
* microblaze: 64b atomic operations are not supported.Torvald Riegel2014-11-212-1/+6
|
* BZ#16469: resolv: skip leading dot in domain to searchAlexandre Oliva2014-11-212-3/+20
| | | | | | | | | | | | | This should only happen if the domain to search is the root, represented as "." rather than by an empty string. Skipping it here prevents libc_res_nquerydomain from duplicating the trailing dot, which would cause the domain name compression to fail. for ChangeLog [BZ #16469] * resolv/res_query.c (__libc_res_nsearch): Skip leading dot in search domain names.
* BZ#16469: don't drop trailing dot in res_nquerydomain(..., name, NULL, ...)Alexandre Oliva2014-11-215-16/+85
| | | | | | | | | | | | | | | | If we drop it here, we will fail to detect a duplicate trailing dot later on. Retaining, OTOH, has no ill effects whatsoever, and it even saves us the trouble of copying the domain name minus the trailing dot, like we used to do. for ChangeLog [BZ #16469] * NEWS: Update. * resolv/res_query.c (__libc_res_nquerydomain): Retain trailing dot. * posix/tst-getaddrinfo5.c: New. * posix/Makefile (tests): Add it.
* BZ#14498: fix infinite loop in nss_db_getservbynameAlexandre Oliva2014-11-214-9/+24
| | | | | | | | | | | | | | | | nss_db uses nss_files code for services, but a continue on protocol mismatch that doesn't affect nss_files skipped the code that advanced to the next db entry. Any one of these changes would suffice to fix it, but fixing both makes them both safer to reuse elsewhere. for ChangeLog [BZ #14498] * NEWS: Fixed. * nss/nss_db/db-XXX.c (_nss_db_get##name##_r): Update hidx after parsing line but before break_if_match. * nss/nss_files/files-service (DB_LOOKUP): Don't "continue;" if there is a protocol mismatch.
* manual: fix addmntent's MT-Safety race annotationMa Shimiao2014-11-212-1/+6
| | | | | | | for ChangeLog * manual/sysinfo.texi (addmntent): It is actually MT-Safe, because the potential race is on the user-supplied stream.
* ctermid: return string literal, document MT-Safety pitfallAlexandre Oliva2014-11-213-11/+21
| | | | | | | | | | for ChangeLog * sysdeps/posix/ctermid.c (ctermid): Return a pointer to a string literal if not passed a buffer. * manual/job.texi (ctermid): Update reasoning, note deviation from posix, suggest mtasurace when not passed a buffer, for future non-preliminary safety notes.
* manual: cuserid is mtasurace if not passed a stringAlexandre Oliva2014-11-212-2/+9
| | | | | | | | for ChangeLog * manual/users.texi (cuserid): Fix MT-Safety note for the case of not passing it a buffer. Reported by Peng Haitao.
* Require check-safety.sh to pass; wish for check that all fns are documentedAlexandre Oliva2014-11-213-1/+13
| | | | | | | | | for ChangeLog * manual/Makefile ($(objpfx)stamp-summary): Require check-safety.sh to pass. * manual/check-safety.sh: Wish for verification that every @deftypefn and @deftypefun is followed by a @safety remark.
* NPTL: Use __libc_fatal in unwind.c.Roland McGrath2014-11-202-5/+4
|
* NPTL: Conditionalize asynchronous cancellation support on [SIGCANCEL].Roland McGrath2014-11-203-7/+24
|
* NPTL: Refactor scheduler setup in pthread_create.Roland McGrath2014-11-204-15/+91
|
* Use C11 atomics in pthread_once.Torvald Riegel2014-11-202-14/+14
|
* Add tests for C11-like atomic operations.Torvald Riegel2014-11-202-1/+133
|
* Add atomic operations similar to those provided by C11.Torvald Riegel2014-11-202-0/+227
|
* Add arch-specific configuration for C11 atomics support.Torvald Riegel2014-11-2022-0/+119
| | | | | | | | This sets __HAVE_64B_ATOMICS if provided. It also sets USE_ATOMIC_COMPILER_BUILTINS to true if the existing atomic ops use the __atomic* builtins (aarch64, mips partially) or if this has been tested (x86_64); otherwise, this is set to false so that C11 atomics will be based on the existing atomic operations.
* NPTL: Don't (re)validate sched_priority in pthread_create.Roland McGrath2014-11-192-20/+5
|
* Test that pthread_create diagnoses invalid scheduling parameters.Roland McGrath2014-11-193-1/+104
|
* CVE-2014-7817: wordexp fails to honour WRDE_NOCMD.Carlos O'Donell2014-11-194-13/+77
| | | | | | | | | | | | | | | | | | | | | | | | | | | The function wordexp() fails to properly handle the WRDE_NOCMD flag when processing arithmetic inputs in the form of "$((... ``))" where "..." can be anything valid. The backticks in the arithmetic epxression are evaluated by in a shell even if WRDE_NOCMD forbade command substitution. This allows an attacker to attempt to pass dangerous commands via constructs of the above form, and bypass the WRDE_NOCMD flag. This patch fixes this by checking for WRDE_NOCMD in exec_comm(), the only place that can execute a shell. All other checks for WRDE_NOCMD are superfluous and removed. We expand the testsuite and add 3 new regression tests of roughly the same form but with a couple of nested levels. On top of the 3 new tests we add fork validation to the WRDE_NOCMD testing. If any forks are detected during the execution of a wordexp() call with WRDE_NOCMD, the test is marked as failed. This is slightly heuristic since vfork might be used in the future, but it provides a higher level of assurance that no shells were executed as part of command substitution with WRDE_NOCMD in effect. In addition it doesn't require libpthread or libdl, instead we use the public implementation namespace function __register_atfork (already part of the public ABI for libpthread). Tested on x86_64 with no regressions.
* Auto-generate libc-modules.hSiddhesh Poyarekar2014-11-196-47/+91
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Remove libc-modules.h from the tree and auto-generate it from soversions.i and the list of modules in the built-modules variable defined in Makeconfig. Macros generated have increasing numbered values, with built-modules having lower values starting from 1, following which a separator value LIBS_BEGIN is added and then finally the library names from soversions.i are appended to the list. This allows us to conveniently differentiate between the versioned libraries and other built modules, which is needed in errno.h and netdb.h to decide whether to use an internal symbol or an external one. Verified that generated code remains unchanged on x86_64. * Makeconfig (built-modules): List non-library modules to be built. (module-cppflags): Include libc-modules.h for everything except shlib-versions.v.i. (CPPFLAGS): Use it. (before-compile): Add libc-modules.h. ($(common-objpfx)libc-modules.h, $(common-objpfx)libc-modules.stmp): New targets. (common-generated): Add libc-modules.h and libc-modules.stmp. ($(common-objpfx)Versions.v.i): Depend on libc-modules.h. * include/libc-symbols.h: Don't include libc-modules.h. * include/libc-modules.h: Remove file. * scripts/gen-libc-modules.awk: New script to generate libc-modules.h. * sysdeps/unix/Makefile ($(common-objpfx)sysd-syscalls): Depend on libc-modules.stmp.
* Fix -Wundef warning in SHLIB_COMPATSiddhesh Poyarekar2014-11-192-1/+3
| | | | | | | Replace the IS_IN_##lib with IS_IN(lib). Verified that the generated code remains the same. * include/shlib-compat.h (_SHLIB_COMPAT): Use IS_IN.
* Add new macro IN_MODULE to identify module in which source is builtSiddhesh Poyarekar2014-11-1918-28/+134
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current scheme to identify which module a translation unit is built in depends on defining multiple macros IS_IN_* and also defining NOT_IN_libc if we're building a non-libc module. In addition, there is an IN_LIB macro that does effectively the same thing, but for different modules (notably the systemtap probes). This macro scheme unifies both ideas to use just one macro IN_MODULE and assign it a value depending on the module it is being built into. If the module is not defined, it defaults to MODULE_libc. Patches that follow will replace uses of IS_IN_* variables with the IS_IN() macro. libc-symbols.h has been converted already to give an example of how such a transition will look. Verified that there are no relevant binary changes. One source change that will crop up repeatedly is that of nscd_stat, since it uses the build timestamp as a constant in its logic. * Makeconfig (in-module): Get value of libof set for the translation unit. (CPPFLAGS): Use $(in-module). * Makerules: Don't suffix routine names for nonlib. * include/libc-modules.h: New file. * include/libc-symbols.h: Include libc-modules.h (IS_IN): New macro to replace IS_IN_* macros. * elf/Makefile: Set libof-* for each routine. * elf/rtld-Rules: Likewise. * extra-modules.mk: Likewise. * iconv/Makefile: Likewise. * iconvdata/Makefile: Likewise. * locale/Makefile: Likewise. * malloc/Makefile: Likewise. * nss/Makefile: Likewise. * sysdeps/gnu/Makefile: Likewise. * sysdeps/ieee754/ldbl-opt/Makefile: Likewise. * sysdeps/unix/sysv/linux/Makefile: Likewise. * sysdeps/s390/s390-64/Makefile: Likewise. * nscd/Makefile: Set libof-* for each routine. Set CFLAGS and CPPFLAGS for nscd instead of nonlib.
* NPTL: Add stub createthread.cRoland McGrath2014-11-182-0/+49
|
* NPTL: Move Linux-specific createthread.c to sysdeps.Roland McGrath2014-11-182-0/+3
|
* NPTL: Refactor createthread.cRoland McGrath2014-11-183-169/+240
|
* Fix libm mpone, mptwo namespace (bug 17616).Joseph Myers2014-11-1810-34/+53
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | libm uses symbols mpone and mptwo for internal purposes. This patch moves them to the implementation namespace (__mpone and __mptwo). Tested for x86_64 (testsuite, and that installed stripped shared libraries are unchanged by the patch). [BZ #17616] * sysdeps/ieee754/dbl-64/mpa.c (mpone): Rename to __mpone. (mptwo): Rename to __mptwo. (__inv): Use __mptwo instead of mptwo. * sysdeps/ieee754/dbl-64/mpa.h (mpone): Rename to __mpone. (mptwo): Rename to __mptwo. * sysdeps/ieee754/dbl-64/mpatan.c (__mpatan): Use __mpone instead of mpone and __mptwo instead of mptwo. * sysdeps/ieee754/dbl-64/mpatan2.c (__mpatan2): Use __mpone instead of mpone. * sysdeps/ieee754/dbl-64/mpexp.c (__mpexp): Likewise. * sysdeps/ieee754/dbl-64/mplog.c (__mplog): Likewise. * sysdeps/ieee754/dbl-64/sincos32.c (__c32): Use __mpone instead of mpone and __mptwo instead of mptwo. (__mpranred): Use __mpone instead of mpone. * conform/Makefile (test-xfail-ISO/math.h/linknamespace): Remove variable. (test-xfail-ISO99/complex.h/linknamespace): Likewise. (test-xfail-ISO99/math.h/linknamespace): Likewise. (test-xfail-ISO99/tgmath.h/linknamespace): Likewise. (test-xfail-ISO11/complex.h/linknamespace): Likewise. (test-xfail-ISO11/math.h/linknamespace): Likewise. (test-xfail-ISO11/tgmath.h/linknamespace): Likewise. (test-xfail-XPG3/math.h/linknamespace): Likewise. (test-xfail-XPG4/math.h/linknamespace): Likewise. (test-xfail-POSIX/math.h/linknamespace): Likewise. (test-xfail-UNIX98/math.h/linknamespace): Likewise. (test-xfail-XOPEN2K/complex.h/linknamespace): Likewise. (test-xfail-XOPEN2K/math.h/linknamespace): Likewise. (test-xfail-XOPEN2K/tgmath.h/linknamespace): Likewise. (test-xfail-POSIX2008/complex.h/linknamespace): Likewise. (test-xfail-POSIX2008/math.h/linknamespace): Likewise. (test-xfail-POSIX2008/tgmath.h/linknamespace): Likewise. (test-xfail-XOPEN2K8/complex.h/linknamespace): Likewise. (test-xfail-XOPEN2K8/math.h/linknamespace): Likewise. (test-xfail-XOPEN2K8/tgmath.h/linknamespace): Likewise.
* Fix crossreference to nonexistent node BSD HandlerTom de Vries2014-11-182-1/+5
| | | | | | | 2014-11-18 Tom de Vries <tom@codesoucery.com> * manual/signal.texi (Primitives Interrupted by Signals): In section, replace BSD Handler xref with BSD Signal Handling.
* alpha: Fix soft-fp breakageRichard Henderson2014-11-1714-62/+95
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 5c0508a318af0a6c8be2a4cb2e3b76896218bf35 broke the Alpha port, as the extra parenthesis got in the way of some token pasting that we were doing in a redefined raw unpack macro. Avoid this situation in the future by not attempting to redefine a basic macro, but rather work from the outermost public interface. The compiler does in fact see through the added indirection. * sysdeps/alpha/soft-fp/local-soft-fp.h (_FP_UNPACK_RAW_2): Remove. (_FP_PACK_RAW_2): Remove. (AXP_DECL_RETURN_Q): Rename from FP_DECL_RETURN, use _FP_UNION_Q. (AXP_RETURN_Q): Rename from FP_RETURN, use _FP_UNION_Q. (AXP_UNPACK_RAW_Q, AXP_UNPACK_SEMIRAW_Q, AXP_UNPACK_Q): New. (AXP_PACK_RAW_Q, AXP_PACK_SEMIRAW_Q, AXP_PACK_Q): New. * sysdeps/alpha/soft-fp/ots_add.c (_OtsAddX): Update to match. * sysdeps/alpha/soft-fp/ots_cmp.c (internal_equality): Likewise. * sysdeps/alpha/soft-fp/ots_cmpe.c (internal_compare): Likewise. * sysdeps/alpha/soft-fp/ots_cvtqux.c (_OtsCvtQUX): Likewise. * sysdeps/alpha/soft-fp/ots_cvtqx.c (_OtsCvtQX): Likewise. * sysdeps/alpha/soft-fp/ots_cvttx.c (_OtsConvertFloatTX): Likewise. * sysdeps/alpha/soft-fp/ots_cvtxq.c (_OtsCvtXQ): Likewise. * sysdeps/alpha/soft-fp/ots_cvtxt.c (_OtsConvertFloatXT): Likewise. * sysdeps/alpha/soft-fp/ots_div.c (_OtsDivX): Likewise. * sysdeps/alpha/soft-fp/ots_mul.c (_OtsMulX): Likewise. * sysdeps/alpha/soft-fp/ots_nintxq.c (_OtsNintXQ): Likewise. * sysdeps/alpha/soft-fp/ots_sub.c (_OtsSubX): Likewise.
* Remove sigvec.Roland McGrath2014-11-1410-312/+209
|
* Remove x86_64 __GNUC_PREREQ (4, 6) conditional.Joseph Myers2014-11-142-10/+6
| | | | | | | | | | | | | | This patch removes a conditional on __GNUC_PREREQ (4, 6) in x86_64 code. Tested for x86_64 that installed shared libraries are unchanged by this patch. Committed (I think this file reasonably comes under math maintainership). * sysdeps/x86_64/fpu/dla.h [__FMA4__ && __GNUC_PREREQ (4, 6)] (DLA_FMS): Make definition conditional only on [__FMA4__]. [__FMA4__ && !__GNUC_PREREQ (4, 6)] (DLA_FMS): Remove conditional definition.
* Remove ARM __GNUC_PREREQ(4,4) conditionals.Joseph Myers2014-11-142-16/+8
| | | | | | | | | | | | | | | | This patch removes conditionals in ARM code on __GNUC_PREREQ(4,4), which were already obsolete even before the move from 4.4 to 4.6 as minimum GCC version for building glibc. Tested for ARM that installed shared libraries are unchanged by this patch. * sysdeps/arm/sysdep.h [PROF && __GNUC_PREREQ(4,4)] (CALL_MCOUNT): Make definition conditional only on [PROF]. [PROF && !__GNUC_PREREQ(4,4)] (CALL_MCOUNT): Remove conditional definition. [__GNUC_PREREQ(4,4)] (mcount): Make definition unconditional. [!__GNUC_PREREQ(4,4)] (mcount): Remove conditional definition.
* Only declare __sigpause in installed signal.h when necessary.Joseph Myers2014-11-144-5/+13
| | | | | | | | | | | | | | | | | | | | | | | This patch makes the installed signal.h declare __sigpause only when necessary (when a macro definition of sigpause makes use of __sigpause), rather than unconditionally. This fixes false positives in the linknamespace tests by making it visible to those tests that no use of ISO C functionality will actually bring in the definition of __sigpause and so bring in the other symbols defined in the same object. There is no bug filed in Bugzilla because this is fixing false positives rather than any user-visible bug. Tested for x86_64 (testsuite, and that installed stripped shared libraries are unchanged by this patch). * signal/signal.h (__sigpause): Only declare if [__USE_XOPEN && !__GNUC__]. * include/signal.h (__sigpause): Move declaration above call to libc_hidden_proto. * conform/Makefile (test-xfail-ISO/signal.h/linknamespace): Remove variable. (test-xfail-ISO99/signal.h/linknamespace): Likewise. (test-xfail-ISO11/signal.h/linknamespace): Likewise.
* Fix sparc build.David S. Miller2014-11-142-0/+6
| | | | | * sysdeps/sparc/sparc64/rtld-memcpy.c (NO_MEMPCPY_STPCPY_REDIRECT): Define before including <string/memcpy.c> and <string/mempcpy.c>.
* Require GCC 4.6 or later to build glibc.Joseph Myers2014-11-146-8/+25
| | | | | | | | | | | | | | | | | | | | | | | | As discussed in the thread starting at <https://sourceware.org/ml/libc-alpha/2014-10/msg00792.html>, and continuing into November, this patch increases the minimum GCC version for building glibc to 4.6 (there seemed to be no clear consensus for 4.7). In particular, this allows us to use #pragma GCC diagnostic for fine-grained warning control with -Werror (subject to establishing a suitable policy for that use). The documentation has a statement, as requested, about the most recent GCC version tested for building glibc, and I've updated <https://sourceware.org/glibc/wiki/Release> to refer to updating that statement. A NEWS entry is added for this change, although previous such changes didn't get them. Tested for x86_64 (testsuite, and that installed shared libraries are unchanged by this patch). * configure.ac (libc_cv_compiler_ok): Require GCC 4.6 or later. * configure: Regenerated. * manual/install.texi (Tools for Compilation): Document a requirement of GCC 4.6 or later and that GCC 4.9 is the newest compiler verified to work. * INSTALL: Regenerated.
* Fix build of C mempcpy and stpcpy.Joseph Myers2014-11-147-1/+22
| | | | | | | | | | | | | | | | | | | | | | This patch fixes the build of C mempcpy and stpcpy by disabling the redirection to __mempcpy and __stpcpy asm names if NO_MEMPCPY_STPCPY_REDIRECT is defined, and defining that macro in the relevant source files. Tested for powerpc32 that the build is fixed. * include/string.h [NO_MEMPCPY_STPCPY_REDIRECT] (mempcpy): Do not redeclare with asm name. [NO_MEMPCPY_STPCPY_REDIRECT] (stpcpy): Likewise. * string/mempcpy.c (NO_MEMPCPY_STPCPY_REDIRECT): Define before including <string.h>. * string/stpcpy.c (NO_MEMPCPY_STPCPY_REDIRECT): Likewise. * sysdeps/powerpc/powerpc32/power4/multiarch/mempcpy.c [!NOT_IN_libc] (NO_MEMPCPY_STPCPY_REDIRECT): Likewise. * sysdeps/powerpc/powerpc64/multiarch/mempcpy.c [!NOT_IN_libc] (NO_MEMPCPY_STPCPY_REDIRECT): Likewise. * sysdeps/powerpc/powerpc64/multiarch/stpcpy.c [SHARED && !NOT_IN_libc] (NO_MEMPCPY_STPCPY_REDIRECT): Likewise.
* Use prototype definition for __strtol.Joseph Myers2014-11-132-4/+3
| | | | | | | | | | This patch fixes those warnings by using a prototype definition for __strtol. Tested for x86_64 that stripped installed shared libraries are unchanged by this patch. * stdlib/strtol.c (__strtol): Use prototype definition.
* Fix strtoll / strtoull namespace for 32-bit (bug 17594).Joseph Myers2014-11-133-3/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For 32-bit platforms, strtoll and strtoull are strong symbols in libc, but they are not in ISO C90, and are brought in by references to __strtoll_internal / __strtoull_internal from scanf. (For 64-bit platforms, they are properly weak.) This patch makes them weak for 32-bit (it has a side-effect of making other symbols weak that don't need to be weak, such as strtol, but that's harmless). Tested for x86 (testsuite, and that the disassembly of installed shared libraries is unchanged by the patch). This fixes all 120 unXFAILed FAILs of the new linknamespace tests seen for x86 (in fact, there are now seven XPASSes of those tests for x86 XPASS: conform/POSIX2008/fcntl.h/linknamespace XPASS: conform/UNIX98/libgen.h/linknamespace XPASS: conform/XOPEN2K/fcntl.h/linknamespace XPASS: conform/XOPEN2K/libgen.h/linknamespace XPASS: conform/XOPEN2K8/fcntl.h/linknamespace XPASS: conform/XOPEN2K8/libgen.h/linknamespace XPASS: conform/XPG4/libgen.h/linknamespace so suggesting that the failures seen for those on x86_64 are in some way architecture-specific or 64-bit-specific). [BZ #17594] * stdlib/strtol.c (SYM__): New macro. (SYM__1): Likewise. (__strtol): Likewise. (strtol): Rename to __strtol and define as weak alias of __strtol. Use libc_hidden_weak.
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-11-26 06:39:32 +0000