1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * Unicode 10.0.0 Support: Character encoding, character type info, and
13 transliteration tables are all updated to Unicode 10.0.0, using
14 generator scripts contributed by Mike FABIAN (Red Hat).
16 * Collation of Hungarian has been overhauled and is now consistent with "The
17 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
20 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
22 - The GNU C Library will now detect when /etc/resolv.conf has been
23 modified and reload the changed configuration. The new resolver option
24 “no-reload” (RES_NORELOAD) disables this behavior.
26 - The GNU C Library now supports an arbitrary number of search domains
27 (configured using the “search” directive in /etc/resolv.conf);
28 previously, there was a hard limit of six domains. For backward
29 compatibility, applications that directly modify the ‘_res’ global
30 object are still limited to six search domains.
32 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
33 Library will now randomly pick a name server from the configuration as a
34 starting point. (Previously, the second name server was always used.)
36 * The tunables feature is now enabled by default. This allows users to tweak
37 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
39 * New function reallocarray, which resizes an allocated block (like realloc)
40 to the product of two sizes, with a guaranteed clean failure upon integer
41 overflow in the multiplication. Originally from OpenBSD, contributed by
42 Dennis Wölfing and Rüdiger Sonderfeld.
44 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
45 These are extended versions of preadv and pwritev, respectively, taking an
46 additional flags argument. The set of supported flags depends on the
47 running kernel; full support currently requires kernel 4.7 or later.
49 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
50 create a new session ID for the spawned process. This feature is
51 scheduled to be added to the next major revision of POSIX; for the time
52 being, it is available under _GNU_SOURCE.
54 * errno.h is now safe to use from C-preprocessed assembly language on all
55 supported operating systems. In this context, it will only define the
56 Exxxx constants, as preprocessor macros expanding to integer literals.
58 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
59 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
60 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
61 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
63 To compile programs that use this feature, the compiler must support
64 128-bit floating point with the type name _Float128 (as defined by TS
65 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
66 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
67 must be defined to make the new interfaces visible.
69 The new functions and macros correspond to those present for other
70 floating-point types (except for a few obsolescent interfaces not
71 supported for the new type), with F128 or f128 suffixes; for example,
72 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
73 printf or scanf formats for the new type; the strfromf128 and strtof128
74 interfaces should be used instead.
76 Deprecated and removed features, and other changes affecting compatibility:
78 * The synchronization that pthread_spin_unlock performs has been changed to
79 now be equivalent to a C11 atomic store with release memory order to the
80 spin lock's memory location. Previously, several (but not all)
81 architectures used stronger synchronization (e.g., containing what is
82 often called a full barrier). This change can improve performance, but
83 may affect odd fringe uses of spin locks that depend on the previous
84 behavior (e.g., using spin locks as atomic variables to try to implement
85 Dekker's mutual exclusion algorithm).
87 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
90 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
91 will only be built and installed when the GNU C Library is configured with
92 --enable-obsolete-rpc. This allows alternative RPC implementations, such
93 as TIRPC or rpcsvc-proto, to be used.
95 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
96 libnss_compat, are deprecated, and will not be built or installed by
99 The NIS(+) support library, libnsl, is also deprecated. By default, a
100 compatibility shared library will be built and installed, but not headers
101 or development libraries. Only a few NIS-related programs require this
102 library. (In particular, the GNU C Library has never required programs
103 that use 'gethostbyname' to be linked with libnsl.)
105 Replacement implementations based on TIRPC, which additionally support
106 IPv6, are available from <https://github.com/thkukuk/>. The configure
107 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
108 name service modules, to be built and installed.
110 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
111 support is enabled, the configured recursive resolver must support EDNS.
112 (Responding to EDNS-enabled queries with responses which are not
113 EDNS-enabled is fine, but FORMERR responses are not.)
115 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
116 servers have not supported this opcode for a long time.
118 * The _res_opcodes variable has been removed from libresolv. It had been
119 exported by accident.
121 * <string.h> no longer includes inline versions of any string functions,
122 as this kind of optimization is better done by the compiler. The macros
123 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
125 * The nonstandard header <xlocale.h> has been removed. Most programs should
126 use <locale.h> instead. If you have a specific need for the definition of
127 locale_t with no other declarations, please contact
128 libc-alpha@sourceware.org and explain.
130 * The obsolete header <sys/ultrasound.h> has been removed.
132 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
134 * The obsolete function cfree has been removed. Applications should use
137 * The stack_t type no longer has the name struct sigaltstack. This changes
138 the C++ name mangling for interfaces involving this type.
140 * The ucontext_t type no longer has the name struct ucontext. This changes
141 the C++ name mangling for interfaces involving this type.
143 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
144 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
145 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
146 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
147 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
148 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
149 fpregset_t type no longer has the name struct fpu. This changes the C++
150 name mangling for interfaces involving those types.
152 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
153 synced with the kernel:
155 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
156 are not supported on this architecture and have been removed.
158 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
159 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
160 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
162 Programs that assume the GET/SETREGS ptrace requests are universally
163 available will now fail to build, instead of malfunctioning at runtime.
165 Changes to build and runtime requirements:
167 * Linux kernel 3.2 or later is required at runtime, on all architectures
168 supported by that kernel. (This is a change from version 2.25 only for
171 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
173 * On most architectures, GCC 4.9 or later is required to build the GNU C
174 Library. On powerpc64le, GCC 6.2 or later is required.
176 Older GCC versions and non-GNU compilers are still supported when
177 compiling programs that use the GNU C Library. (We do not know exactly
178 how old, and some GNU extensions to C may be _de facto_ required. If you
179 are interested in helping us make this statement less vague, please
180 contact libc-alpha@sourceware.org.)
182 Security related changes:
184 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
185 to avoid fragmentation-based spoofing attacks.
187 The following bugs are resolved with this release:
189 [The release manager will add the list generated by
190 scripts/list-fixed-bugs.py just before the release.]
195 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
196 24731-2:2010, is supported to enable declarations of functions from that
197 TR. Note that not all functions from that TR are supported by the GNU C
200 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
201 18661-1:2014, is supported to enable declarations of functions and macros
202 from that TS. Note that not all features from that TS are supported by
205 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
206 18661-4:2015, is supported to enable declarations of functions and macros
207 from that TS. Note that most features from that TS are not supported by
210 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
211 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
212 Since the GNU C Library defaults to a much newer revision of POSIX, this
213 will only affect programs that specifically request an old conformance
214 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
215 see a change in the visible declarations, but a program compiled with
216 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
219 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
220 defined by all multithreaded code, but glibc has not required this for
223 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
224 means that in a future release, the macros “major”, “minor”, and “makedev”
225 will only be available from <sys/sysmacros.h>.
227 These macros are not part of POSIX nor XSI, and their names frequently
228 collide with user code; see for instance glibc bug 19239 and Red Hat bug
229 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
230 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
233 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
234 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
235 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
237 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
238 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
239 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
240 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
241 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
242 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
243 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
244 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
245 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
246 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
247 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
250 * New <math.h> features are added from TS 18661-1:2014:
252 - Signaling NaN macros: SNANF, SNAN, SNANL.
254 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
255 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
256 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
258 - llogb functions: the llogb, llogbf and llogbl functions, and the
259 FP_LLOGB0 and FP_LLOGBNAN macros.
261 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
264 - Comparison macros: iseqsig.
266 - Classification macros: iscanonical, issubnormal, iszero.
268 - Total order functions: totalorder, totalorderf, totalorderl,
269 totalordermag, totalordermagf, totalordermagl.
271 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
273 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
274 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
276 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
277 are added to libc. They convert a floating-point number into string.
279 * Most of glibc can now be built with the stack smashing protector enabled.
280 It is recommended to build glibc with --enable-stack-protector=strong.
281 Implemented by Nick Alcock (Oracle).
283 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
284 intended to be used instead of memset() to erase sensitive data after use;
285 the compiler will not optimize out calls to explicit_bzero even if they
286 are "unnecessary" (in the sense that no _correct_ program can observe the
287 effects of the memory clear).
289 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
290 to float instead of double. This does not affect the ABI of any libraries
291 that are part of the GNU C Library, but may affect the ABI of other
292 libraries that use this type in their interfaces.
294 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
295 float_t and double_t types are now defined to long double instead of float
296 and double. These options are not the default, and this does not affect
297 the ABI of any libraries that are part of the GNU C Library, but it may
298 affect the ABI of other libraries that use this type in their interfaces,
299 if they are compiled or used with those options.
301 * The getentropy and getrandom functions, and the <sys/random.h> header file
304 * The buffer size for byte-oriented stdio streams is now limited to 8192
305 bytes by default. Previously, on Linux, the default buffer size on most
306 file systems was 4096 bytes (and thus remains unchanged), except on
307 network file systems, where the buffer size was unpredictable and could be
308 as large as several megabytes.
310 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
311 for the Linux quota interface which predates kernel version 2.4.22 has
314 * The malloc_get_state and malloc_set_state functions have been removed.
315 Already-existing binaries that dynamically link to these functions will
316 get a hidden implementation in which malloc_get_state is a stub. As far
317 as we know, these functions are used only by GNU Emacs and this change
318 will not adversely affect already-built Emacs executables. Any undumped
319 Emacs executables, which normally exist only during an Emacs build, should
320 be rebuilt by re-running “./configure; make” in the Emacs build tree.
322 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
323 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
324 “no-ip6-dotint” had already been the default, and support for the
325 “ip6-dotint” option was removed from the Internet in 2006.
327 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
328 flag from <resolv.h> have been removed. The option relied on a
329 backwards-incompatible DNS extension which was never deployed on the
332 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
333 RES_BLAST defined in the <resolv.h> header file have been deprecated.
334 They were already unimplemented.
336 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
337 _res.flags are deprecated. The flag was standardized in RFC 2133, but
338 removed again from the IETF name lookup interface specification in RFC
339 2553. Applications should use getaddrinfo instead.
341 * DNSSEC-related declarations and definitions have been removed from the
342 <arpa/nameser.h> header file, and libresolv will no longer attempt to
343 decode the data part of DNSSEC record types. Previous versions of glibc
344 only implemented minimal support for the previous version of DNSSEC, which
345 is incompatible with the currently deployed version.
347 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
348 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
349 <arpa/nameser.h> header file because the distinction between RR types and
350 meta-RR types is not officially standardized, subject to revision, and
351 thus not suitable for encoding in a macro.
353 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
354 and rhook members of the res_state type in <resolv.h> have been removed.
355 The glibc stub resolver did not support these hooks, but the header file
356 did not reflect that.
358 * For multi-arch support it is recommended to use a GCC which has
359 been built with support for GNU indirect functions. This ensures
360 that correct debugging information is generated for functions
361 selected by IFUNC resolvers. This support can either be enabled by
362 configuring GCC with '--enable-gnu-indirect-function', or by
363 enabling it by default by setting 'default_gnu_indirect_function'
364 variable for a particular architecture in the GCC source file
367 * GDB pretty printers have been added for mutex and condition variable
368 structures in POSIX Threads. When installed and loaded in gdb these pretty
369 printers show various pthread variables in human-readable form when read
370 using the 'print' or 'display' commands in gdb.
372 * Tunables feature added to allow tweaking of the runtime for an application
373 program. This feature can be enabled with the '--enable-tunables' configure
374 flag. The GNU C Library manual has details on usage and README.tunables has
375 instructions on adding new tunables to the library.
377 * A new version of condition variables functions have been implemented in
378 the NPTL implementation of POSIX Threads to provide stronger ordering
381 * A new version of pthread_rwlock functions have been implemented to use a more
382 scalable algorithm primarily through not using a critical section anymore to
385 Security related changes:
387 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
388 have been created with makecontext could fail to terminate due to a
389 missing .cantunwind annotation. This has been observed to lead to a hang
390 (denial of service) in some Go applications compiled with gccgo. Reported
391 by Andreas Schwab. (CVE-2016-6323)
393 * The DNS stub resolver functions would crash due to a NULL pointer
394 dereference when processing a query with a valid DNS question type which
395 was used internally in the implementation. The stub resolver now uses a
396 question type which is outside the range of valid question type values.
399 The following bugs are resolved with this release:
401 [4099] stdio: Overly agressive caching by stream i/o functions.
402 [7065] build: Support building glibc with -fstack-protector or -fstack-
404 [9842] localedata: en_CA: incorrect date format
405 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
406 before it started waiting
407 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
408 [15765] nptl: sem_open is wrongly a cancellation point
409 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
410 [16458] libc: endian.h and netinet/in.h byte order macros return values of
412 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
413 library linked with pthread
414 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
416 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
417 [17252] libc: getrandom and getentropy syscall
418 [17730] malloc: thread-local storage is sometimes improperly free()'d
419 after being __libc_memalign()'d
420 [18241] stdio: failed fseek on memstream does not set errno and can fail
422 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
424 [18463] nptl: pthread_cond_broadcast issue when surrounded by
425 PTHREAD_PRIO_INHERIT mutex on ARM
426 [18784] network: res_query and related function crash for special record
427 type queries (CVE-2015-5180)
428 [19380] math: strtod does not raise "inexact"
429 [19387] string: Integer overflow in memchr
430 [19390] string: Integer overflow in strncat
431 [19398] build: linknamespace tests fail with massively parallel build
432 [19402] nptl: Deadlock with robust shared mutex and asynchronous
434 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
435 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
436 [19514] libc: [PATCH] Fix spelling errors in spelling
437 "implement"/"implementation" in several places
438 [19582] network: Deprecate RES_USE_INET6
439 [19673] manual: clog10 docs appear to be erroneous
440 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
441 causes a segmentation fault
442 [19826] libc: invalid pointer returned from __tls_get_addr with static
444 [20016] network: resolv: Remove hooks support from the API
445 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
447 [20033] math: [x86_64] vectorized math function don't call the __finite
449 [20116] nptl: use after free in pthread_create
450 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
451 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
453 [20311] nptl: please install proc_service.h
454 [20366] libc: Compilation errors in installed headers in strict-compliance
456 [20370] malloc: malloc: Arena free list management is still racy
457 (incorrect fix in bug 19243)
458 [20386] libc: assert (X = 0) does not result in GCC warning
459 [20432] malloc: malloc: Minimize interface required for interposition
460 [20435] libc: Missing unwind info in __startcontext causes infinite loop
461 in _Unwind_Backtrace (CVE-2016-6323)
462 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
463 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
465 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
466 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
467 [20477] network: resolv: incorrect double-checked locking related to
469 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
471 [20495] math: x86_64 performance degradation due to AVX/SSE transition
473 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
474 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
475 cause transition penalty
476 [20517] math: sparcv9 missing fdiml compat symbol
477 [20524] manual: strverscmp is inconsistent
478 [20525] libc: <sys/quota.h> should be based on kernel headers
479 [20539] math: GCC 7 -static -lm fails to link at -O3
480 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
482 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
484 [20591] network: Remove obsolete DNSSEC support
485 [20592] network: DNS resource record type classification macros in
486 <arpa/nameser.h> are incorrect
487 [20593] network: Update DNS RR type definitions
488 [20611] network: getaddrinfo accepts invalid numeric scope IDs
489 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
491 [20629] network: libresolv: Remove support for bitstring labels
493 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
494 assertion: "maxidx >= 2"
495 [20660] math: [arm] Use VSQRT
496 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
497 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
498 [20689] libc: Test for FMA should also check for AVX.
499 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
500 [20715] math: iszero macro breaks existing code
501 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
502 [20728] libc: powerpc: Missing TOC stub in clone
503 [20729] build: glibc-2.24 fails to build for i486 with -Os
504 [20750] build: Build fails with default PIE enabled gcc-6.x
505 [20768] math: [alpha] sqrt fegetenv namespace
506 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
508 [20787] math: float_t is defined as float incorrectly on x86_64 even if
509 __FLT_EVAL_METHOD__ is 2
510 [20790] network: rpcgen buffer overrun in get_prog_declaration
511 [20822] nptl: race condition in __lll_unlock_elision on powerpc
512 [20829] libc: crypt snprintf namespace
513 [20847] libc: tst-vfork3 failure
514 [20855] math: Default bits/mathdef.h has inappropriate float_t
515 [20859] math: [sh4] FP_ILOGB0 invalid
516 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
517 U+20AC), not same as GBK
518 [20915] dynamic-link: global-dynamic TLS broken on aarch64
519 [20916] math: pow handling of sNaN arguments
520 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
521 [20919] math: Bad pow (qNaN, 0) result with -lieee
522 [20940] math: hypot sNaN handling
523 [20947] math: fmax, fmin sNaN handling
524 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
525 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
527 [20971] string: powerpc64/power7 memchr overflows internal pointer check
528 [20973] nptl: robust mutexes: Lost wake-ups
529 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
530 [20978] nis: strlen on null pointer in nss_nisplus
531 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
533 [21014] string: i686 memchr overflows internal pointer check
534 [21019] libc: [mips] n32 lseek incorrect on overflow
535 [21022] libc: [microblaze] __backtrace get_frame_size namespace
536 [21026] libc: [MIPS] readahead syscall is broken on n64
537 [21028] math: Fallback fesetexceptflag should always succeed
538 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
539 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
541 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
542 [21061] librt: [microblaze] librt lost clock_* exports
543 [21073] libc: tunables: insecure environment variables passed to
544 subprocesses with AT_SECURE
545 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
549 * The minimum Linux kernel version that this version of the GNU C Library
550 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
551 kernel version 2.6.32 or later suffices (on architectures that already
552 required kernel versions more recent than 3.2, those requirements remain
553 unchanged). Linux 3.2 or later kernel headers are required on all
556 * The pap_AN locale has been deleted. This has been deprecated for a long
557 time. It has been replaced by pap_AW & pap_CW, both of which have long
558 been included in previous releases.
560 * The readdir_r and readdir64_r functions have been deprecated. It is
561 recommended to use readdir and readdir64 instead.
563 * The type “union wait” has been removed. It was deprecated in the early
564 1990s and never part of POSIX. Application code should use the int type
565 instead of “union wait”.
567 * A new NSS action is added to facilitate large distributed system
568 administration. The action, MERGE, allows remote user stores like LDAP
569 to be merged into local user stores like /etc/groups in order to provide
570 easy to use, updated, and managed sets of merged credentials. The new
571 action can be used by configuring it in /etc/nsswitch.conf:
572 group: files [SUCCESS=merge] nis
573 Implemented by Stephen Gallagher (Red Hat).
575 * The deprecated __malloc_initialize_hook variable has been removed from the
578 * The long unused localedef --old-style option has been removed. It hasn't
579 done anything in over 16 years. Scripts using this option can safely
582 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
583 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
584 return the next representable value in the direction of positive infinity
585 and the nextdown functions return the next representable value in the
586 direction of negative infinity. These are currently enabled as GNU
589 Security related changes:
591 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
592 could result in a stack overflow when getnetbyname was called with an
593 overly long name. (CVE-2016-3075)
595 * Previously, getaddrinfo copied large amounts of address data to the stack,
596 even after the fix for CVE-2013-4458 has been applied, potentially
597 resulting in a stack overflow. getaddrinfo now uses a heap allocation
598 instead. Reported by Michael Petlan. (CVE-2016-3706)
600 * The glob function suffered from a stack-based buffer overflow when it was
601 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
602 Reported by Alexander Cherepanov. (CVE-2016-1234)
604 * The Sun RPC UDP client could exhaust all available stack space when
605 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
606 alloca plugin for GCC. (CVE-2016-4429)
608 * The IPv6 name server management code in libresolv could result in a memory
609 leak for each thread which is created, performs a failing naming lookup,
610 and exits. Over time, this could result in a denial of service due to
611 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
613 The following bugs are resolved with this release:
615 [1170] localedata: ne_NP: update Nepali locale definition file
616 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
618 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
619 [6796] math: fdim() does not set errno on overflow
620 [10354] libc: posix_spawn should use vfork() in more cases than presently
621 [11213] localedata: localedata: add copyright disclaimer to locale files
622 [12143] localedata: chr_US: new Cherokee locale
623 [12450] localedata: sgs_LT: new locale
624 [12676] localedata: ln_CD: new locale
625 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
627 [13304] math: fma, fmaf, fmal produce wrong results
628 [14259] build: --localedir arg to configure is ignored
629 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
630 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
631 [14934] localedata: es_CL: wrong first weekday chilean locale
632 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
634 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
636 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
637 [15368] nptl: raise() is not async-signal-safe
638 [15479] math: ceil, floor, round and trunc raise inexact exception
639 [15578] localedata: kk_KZ: various updates
640 [16003] localedata: pap_AN: punt old locale
641 [16137] localedata: iw_IL: punt old locale
642 [16190] localedata: eo: new esperanto locale
643 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
644 [16742] malloc: race condition: pthread_atfork() called before first
645 malloc() results in unexpected locking behaviour/deadlocks
646 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
648 [16983] localedata: postal_fmt does not allow %l and %n modifiers
649 [17565] localedata: pt_PT: wrong (work-)week start
650 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
651 [17950] build: Build fails with -msse
652 [18205] localedata: be_BY*: wrong first_weekday and first_workday
653 [18433] libc: posix_spawn does not return correctly upon failure to
655 [18453] localedata: charmaps/IBM875: incorrect codes
656 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
658 [18896] localedata: he_IL: improvements for currency
659 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
661 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
663 [19056] libc: Deprecate readdir_r
664 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
666 [19198] localedata: nl_NL: small improvements for Dutch locales
667 [19257] network: Per-thread memory leak in __res_vinit with IPv6
668 nameservers (CVE-2016-5417)
669 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
671 [19400] locale: Language missing in "iso-639.def", trivial fix in
673 [19431] malloc: Deadlock between fflush, getdelim, and fork
674 [19505] libc: Incorrect file descriptor validity checks in
675 posix_spawn_file_actions_add{open,close,dup2}
676 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
678 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
679 `intl/{gettextP,loadinfo}.h'
680 [19534] libc: execle, execlp may use malloc
681 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
682 [19573] network: res_nclose and __res_maybe_init disagree about name
683 server initialization, breaking Hesiod
684 [19575] localedata: Status of GB18030 tables
685 [19581] localedata: sr_* date_fmt string contains additional newline
686 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
688 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
689 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
690 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
691 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
693 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
695 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
697 [19610] dynamic-link: ldconfig -X removes stale symbolic links
698 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
699 [19633] locale: strfmon_l applies global locale to number formatting
700 [19642] network: Memory leak in getnameinfo
701 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
702 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
704 [19654] math: [x86_64] Need testcase for BZ #19590 fix
705 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
707 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
708 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
710 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
712 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
714 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
715 pointers and lengths in error-case.
716 [19727] locale: Converting from/to UTF-xx with iconv() does not always
717 report errors on UTF-16 surrogates values.
718 [19755] nscd: nscd assertion failure in gc
719 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
720 [19759] libc: mempcpy shouldn't be inlined
721 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
722 [19765] libc: s390 needs an optimized mempcpy
723 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
724 NAME_MAX limit assumption (CVE-2016-1234)
725 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
726 [19787] network: Missing and incorrect truncation checks in getnameinfo
727 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
729 [19791] network: Assertion failure in res_query.c with un-connectable name
731 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
732 [19822] math: libm.so install clobbers old version
733 [19825] network: resolv: send_vc can return uninitialized data in second
734 response to getaddrinfo
735 [19830] network: nss_dns: should check RDATA length against buffer length
736 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
737 confronted with A/AAAA records of invalid size
738 [19837] nss: nss_db: No retries for some long lines with a larger buffer
739 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
740 [19853] stdio: Printing IBM long double in decimal with high precision is
742 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
743 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
744 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
746 [19865] network: Assertion failure or memory leak in
747 _nss_dns_getcanonname_r
748 [19868] network: nss_dns: netent code does not skip over non-PTR records
749 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
751 [19881] string: Improve x86-64 memset
752 [19907] string: Incorrect memcpy tests
753 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
755 [19925] libc: termios.h XCASE namespace
756 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
757 [19929] libc: limits.h NL_NMAX namespace
758 [19931] stdio: Memory leak in vfprintf
759 [19957] libc: clone(CLONE_VM) access invalid parent memory
760 [19963] localedata: en_IL: New locale
761 [19989] stdio: stdio.h cuserid namespace
762 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
764 [19996] locale: langinfo.h nl_langinfo_l namespace
765 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
766 [20010] network: getaddrinfo: Stack overflow in hostent translation
768 [20012] stdio: libio: fmemopen append mode failure
769 [20014] stdio: stdio.h namespace for pre-threads POSIX
770 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
771 [20023] libc: fcntl.h timespec namespace
772 [20024] math: [x86_64] vectorized sincos trashes the stack
773 [20031] network: nss_hesiod: Heap overflow in get_txt_records
774 [20041] time: sys/time.h timespec namespace
775 [20043] libc: unistd.h missing cuserid for UNIX98 and before
776 [20044] libc: unistd.h missing pthread_atfork for UNIX98
777 [20051] libc: ttyslot in wrong header under wrong conditions
778 [20054] libc: gethostname not declared for XPG4
779 [20055] libc: termios.h missing tcgetsid for XPG4
780 [20072] dynamic-link: x86 init_cpu_features is called twice in static
782 [20073] libc: sys/stat.h fchmod namespace
783 [20074] libc: stdlib.h rand_r namespace
784 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
785 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
787 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
788 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
790 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
791 [20119] libc: Wrong mask for processors level type from CPUID
792 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
793 AS not supporting AVX512
794 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
796 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
797 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
799 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
800 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
801 [20178] libc: posix_spawn{p} should not call exit
802 [20191] stdio: libio: vtables hardening
803 [20195] string: FMA4 detection requires CPUID execution with register
805 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
806 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
808 [20212] math: acos (sNaN) returns sNaN
809 [20213] math: asin (sNaN) returns sNaN
810 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
811 [20218] math: [i386] asinhl (sNaN) returns sNaN
812 [20219] math: [i386] atanhl (sNaN) returns sNaN
813 [20222] stdio: fopencookie: Mangle function pointers
814 [20224] math: [i386] cbrtl (sNaN) returns sNaN
815 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
816 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
818 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
819 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
820 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
821 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
822 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
823 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
824 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
825 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
826 [20240] math: modf (sNaN) returns sNaN
827 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
828 [20250] math: frexp (sNaN) returns sNaN
829 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
830 [20255] math: [i386] fdim, fdimf return with excess range and precision /
832 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
833 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
834 subscript is below array bounds [-Werror=array-bounds]
835 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
836 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
838 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
839 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
841 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
843 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
844 [20316] localedata: id_ID: Februari instead of Pebruari
845 [20327] string: POWER8 strcasecmp returns incorrect result
846 [20347] math: Failure: Test: j0_downward (0xap+0)
847 [20348] libc: FAIL: misc/tst-preadvwritev64
848 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
849 [20350] libc: There is no test for p{read,write}64
850 [20357] math: Incorrect cos result for 1.5174239687223976
851 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
855 * Unicode 8.0.0 Support: Character encoding, character type info, and
856 transliteration tables are all updated to Unicode 8.0.0, using new
857 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
858 These updates cause user visible changes, such as the fixes for bugs
859 89, 16061, and 18568.
861 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
862 kernel-internal CPU set size. This means that requests that change the
863 CPU affinity which failed before (for example, an all-ones CPU mask) will
864 now succeed. Applications that need to determine the effective CPU
865 affinities need to call sched_getaffinity or pthread_getaffinity_np after
866 setting it because the kernel can adjust it (and the previous size check
867 would not detect this in the majority of cases).
869 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
870 the following new symbols are used: fts64_children, fts64_close,
871 fts64_open, fts64_read and fts64_set.
873 * getaddrinfo now detects certain invalid responses on an internal netlink
874 socket. If such responses are received, an affected process will
875 terminate with an error message of "Unexpected error <number> on netlink
876 descriptor <number>" or "Unexpected netlink response of size <number> on
877 descriptor <number>". The most likely cause for these errors is a
878 multi-threaded application which erroneously closes and reuses the netlink
879 file descriptor while it is used by getaddrinfo.
881 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
882 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
883 unnecessary serialization of memory allocation requests across threads.
884 The defect is now corrected. Users should see a substantial increase in
885 the concurent throughput of allocation requests for applications which
886 trigger this bug. Affected applications typically create and
887 destroy threads frequently. (Bug 19048 was reported and analyzed by
890 * There is now a --disable-timezone-tools configure option for disabling the
891 building and installing of the timezone related utilities (zic, zdump, and
892 tzselect). This is useful for people who build the timezone data and code
893 independent of the GNU C Library.
895 * The obsolete header <regexp.h> has been removed. Programs that require
896 this header must be updated to use <regex.h> instead.
898 * The obsolete functions bdflush, create_module, get_kernel_syms,
899 query_module and uselib are no longer available to newly linked binaries;
900 the header <sys/kdaemon.h> has been removed. These functions and header
901 were specific to systems using the Linux kernel and could not usefully be
902 used with the GNU C Library on systems with version 2.6 or later of the
905 * Optimized string, wcsmbs and memory functions for IBM z13.
906 Implemented by Stefan Liebler.
908 * Newly linked programs that define a variable called signgam will no longer
909 have it set by the lgamma, lgammaf and lgammal functions. Programs that
910 require signgam to be set by those functions must ensure that they use the
911 variable provided by the GNU C Library and declared in <math.h>, without
912 defining their own copy.
914 * The minimum GCC version that can be used to build this version of the GNU
915 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
916 still be used to compile programs using the GNU C Library.
918 Security related changes:
920 * An out-of-bounds value in a broken-out struct tm argument to strftime no
921 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
923 * The LD_POINTER_GUARD environment variable can no longer be used to disable
924 the pointer guard feature. It is always enabled. Previously,
925 LD_POINTER_GUARD could be used to disable security hardening in binaries
926 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
929 * An integer overflow in hcreate and hcreate_r could lead to an
930 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
932 * The catopen function no longer has unbounded stack usage. Reported by
935 * The nan, nanf and nanl functions no longer have unbounded stack usage
936 depending on the length of the string passed as an argument to the
937 functions. Reported by Joseph Myers. (CVE-2014-9761)
939 * A stack-based buffer overflow was found in libresolv when invoked from
940 libnss_dns, allowing specially crafted DNS responses to seize control
941 of execution flow in the DNS client. The buffer overflow occurs in
942 the functions send_dg (send datagram) and send_vc (send TCP) for the
943 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
944 family. The use of AF_UNSPEC triggers the low-level resolver code to
945 send out two parallel queries for A and AAAA. A mismanagement of the
946 buffers used for those queries could result in the response of a query
947 writing beyond the alloca allocated buffer created by
948 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
949 the overflow. Thanks to the Google Security Team and Red Hat for
950 reporting the security impact of this issue, and Robert Holiday of
951 Ciena for reporting the related bug 18665. (CVE-2015-7547)
953 The following bugs are resolved with this release:
955 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
956 [887] math: Math library function "logb" and "nextafter" inconsistent
957 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
958 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
959 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
960 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
962 [4404] localedata: German translation of "Alarm clock" is misleading
963 [6799] math: nextafter() and nexttoward() doen't set errno on
964 overflow/underflow errors
965 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
967 [10432] nis: _nss_nis_setnetgrent assertion failure
968 [11460] libc: fts has no LFS support
969 [12926] network: getaddrinfo()/make_request() may spin forever
970 [13065] nptl: Race condition in pthread barriers
971 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
972 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
974 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
975 [14912] libc: Rename non-installed bits/*.h headers
976 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
977 [15367] math: Let gcc use __builtin_isinf
978 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
979 [15421] math: lgamma wrongly sets signgam for ISO C
980 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
981 FE_INVALID with argument out of range
982 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
984 [15786] dynamic-link: ifunc resolver functions can smash function
986 [15918] math: Unnecessary check for equality in hypotf()
987 [16061] localedata: Review / update transliteration data
988 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
990 [16141] time: strptime %z offset restriction
991 [16171] math: drem should be alias of remainder
992 [16296] math: fegetround is pure?
993 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
994 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
996 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
997 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
999 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1000 "Exception "Inexact" set" on ppc32
1001 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1002 [16517] math: Missing underflow exception from tanf/tan/tanl
1003 [16519] math: Missing underflow exception from sinhf
1004 [16520] math: Missing underflow exception from tanhf
1005 [16521] math: Missing underflow exception from exp2
1006 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1008 [16734] stdio: fopen calls mmap to allocate its buffer
1009 [16961] math: nan function incorrect handling of bad sequences
1010 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1011 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1012 [16985] locale: localedef: confusing error message when opening output
1014 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1015 [17197] locale: Redundant shift character in iconv conversion output at
1017 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1018 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1019 [17250] dynamic-link: static linking breaks nss loading
1020 (getaddrinfo/getpwnam/etc...)
1021 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1023 [17441] math: isnan() should use __builtin_isnan() in GCC
1024 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1025 (related to lock elision)
1026 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1027 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1028 [17887] time: strptime should be able to parse "+01:00" style timezones
1029 [17905] libc: catopen() Multiple unbounded stack allocations
1031 [18084] libc: backtrace (..., 0) dumps core on x86
1032 [18086] libc: nice() sets errno to 0 on success
1033 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1034 is too large (CVE-2015-8778)
1035 [18251] dynamic-link: SONAME missing when audit modules provides path
1036 [18265] libc: add attributes for wchar string and memory functions
1037 [18370] math: csqrt missing underflows
1038 [18421] libc: [hppa] read-only segment has dynamic relocations
1039 [18472] libc: Obsolete syscall wrappers should be compat symbols
1040 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1041 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1042 [18525] localedata: Remove locale timezone information
1043 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1044 [18568] localedata: Update locale data to Unicode 8.0
1045 [18589] locale: sort-test.sh fails at random
1046 [18595] math: ctan, ctanh missing underflows
1047 [18604] libc: assert macro-expands its argument
1048 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1049 contains a vector instruction exception.
1050 [18611] math: j1, jn missing errno setting on underflow
1051 [18618] localedata: sync Chechen locale definitions with other *_RU
1053 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1054 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1055 [18665] network: In send_dg, the recvfrom function is NOT always using the
1056 buffer size of a newly created buffer (CVE-2015-7547)
1057 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1059 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1061 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1062 [18699] math: tilegx cproj() for various complex infinities does not yield
1064 [18724] libc: Harden put*ent functions against data injection
1065 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1066 [18755] build: build errors with -DNDEBUG
1067 [18757] stdio: fmemopen fails to set errno on failure
1068 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1070 [18781] libc: openat64 lacks O_LARGEFILE
1071 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1072 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1073 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1074 [18790] math: [ldbl-128ibm] tanhl inaccurate
1075 [18795] libc: stpncpy fortification misses buffer lengths that are
1076 statically too large
1077 [18796] build: build fails for --disable-mathvec
1078 [18803] math: hypot missing underflows
1079 [18820] stdio: fmemopen may leak memory on failure
1080 [18823] math: csqrt spurious underflows
1081 [18824] math: fma spurious underflows
1082 [18825] math: pow missing underflows
1083 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1084 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1085 [18870] build: sem_open.c fails to compile with missing symbol
1087 [18872] stdio: Fix memory leak in printf_positional
1088 [18873] libc: posix_fallocate overflow check ineffective
1089 [18875] math: Excess precision leads incorrect libm
1090 [18877] libc: arm: mmap offset regression
1091 [18887] libc: memory corruption when using getmntent on blank lines
1092 [18918] localedata: hu_HU: change time to HH:MM:SS format
1093 [18921] libc: Regression: extraneous stat() and fstat() performed by
1095 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1096 binaries (CVE-2015-8777)
1097 [18951] math: tgamma missing underflows
1098 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1100 [18953] localedata: lt_LT: change currency symbol to the euro
1101 [18956] math: powf inaccuracy
1102 [18961] math: [i386] exp missing underflows
1103 [18966] math: [i386] exp10 missing underflows
1104 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1105 [18969] build: multiple string test failures due to missing locale
1107 [18970] libc: Reference of pthread_setcancelstate in libc.a
1108 [18977] math: float / long double Bessel functions not in XSI POSIX
1109 [18980] math: i386 libm functions return with excess range and precision
1110 [18981] math: i386 scalb*, ldexp return with excess range and precision
1111 [18982] stdio: va_list and vprintf
1112 [18985] time: Passing out of range data to strftime() causes a segfault
1114 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1115 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1116 [19012] locale: iconv_open leaks memory on error path
1117 [19016] math: clog, clog10 inaccuracy
1118 [19018] nptl: Mangle function pointers in tls_dtor_list
1119 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1120 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1121 [19048] malloc: malloc: arena free list can become cyclic, increasing
1123 [19049] math: [powerpc] erfc incorrect zero sign
1124 [19050] math: [powerpc] log* incorrect zero sign
1125 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1126 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1127 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1128 [19074] network: Data race in _res_hconf_reorder_addrs
1129 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1130 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1131 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1133 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1135 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1136 [19086] manual: posix_fallocate64 documented argument order is wrong.
1137 [19088] math: lround, llround missing exceptions close to overflow
1139 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1140 [19095] math: dbl-64 lrint incorrect for 64-bit long
1141 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1142 [19124] dynamic-link: ld.so failed to build with older assmebler
1143 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1144 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1145 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1146 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1147 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1148 pthread_setaffinity_np
1149 [19156] math: [ldbl-128] j0l spurious underflows
1150 [19164] nptl: tst-getcpu fails with many possible CPUs
1151 [19168] math: math/test-ildoubl and math/test-ldouble failure
1152 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1153 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1155 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1156 (FE_NOMASK_ENV) do not clear SSE exceptions
1157 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1158 _int_new_arena/reused_arena
1159 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1160 [19201] math: dbl-64 remainder incorrect sign of zero result
1161 [19205] math: bits/math-finite.h conditions do not match math.h and
1163 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1164 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1165 for C99-based standards
1166 [19212] libc: features.h not -Wundef clean
1167 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1169 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1170 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1171 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1173 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1174 "inexact" exceptions
1175 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1177 [19242] libc: strtol incorrect in Turkish locales
1178 [19243] malloc: reused_arena can pick an arena on the free list, leading
1179 to an assertion failure and reference count corruption
1180 [19253] time: tzset() ineffective when temporary TZ did not include DST
1182 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1183 [19270] math: [hppa] Shared libm missing __isnanl
1184 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1185 MAP_HUGETLB and MAP_STACK defines
1186 [19313] nptl: Wrong __cpu_mask for x32
1187 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1188 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1189 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1190 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1191 [19363] time: x32: times() return value wrongly truncates/sign extends
1193 [19367] dynamic-link: Improve branch prediction on Silvermont
1194 [19369] network: Default domain name not reset by res_ninit when "search"
1195 / "domain" entry is removed from resolv.conf
1196 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1198 [19385] localedata: bg_BG: time separator should be colon, not comma
1199 [19408] libc: linux personality syscall wrapper may erroneously return an
1200 error on 32-bit architectures
1201 [19415] libc: dladdr returns wrong names on hppa
1202 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1203 IBM905, IBM907, and IBM909
1204 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1205 [19443] build: build failures with -DDEBUG
1206 [19451] build: Make check fails on test-double-vlen2
1207 [19462] libc: Glibc failed to build with -Os
1208 [19465] math: Wrong code with -Os
1209 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1211 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1213 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1214 [19486] math: S390: Math tests fail with "Exception Inexact set".
1215 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1216 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1218 [19590] math: Fail to build shared objects that use libmvec.so functions.
1222 * The following bugs are resolved with this release:
1224 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1225 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1226 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1227 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1228 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1229 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1230 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1231 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1232 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1233 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1234 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1235 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1236 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1237 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1238 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1239 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1240 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1241 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1242 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1243 18657, 18676, 18694, 18696, 18887.
1245 * Cache information can be queried via sysconf() function on s390 e.g. with
1246 _SC_LEVEL1_ICACHE_SIZE as argument.
1248 * A buffer overflow in gethostbyname_r and related functions performing DNS
1249 requests has been fixed. If the NSS functions were called with a
1250 misaligned buffer, the buffer length change due to pointer alignment was
1251 not taken into account. This could result in application crashes or,
1252 potentially arbitrary code execution, using crafted, but syntactically
1253 valid DNS responses. (CVE-2015-1781)
1255 * The time zone file parser has been made more robust against crafted time
1256 zone files, avoiding heap buffer overflows related to the processing of
1257 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1258 large time zone data files. Overly long time zone specifiers in the TZ
1259 variable no longer result in stack overflows and crashes.
1261 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1262 for LD and GD on x86 and x86-64, has been implemented. You will need
1263 binutils-2.24 or later to enable this optimization.
1265 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1266 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1267 Hat). These updates cause user visible changes, such as the fix for bug
1270 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1271 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1272 condition in some applications.
1274 * Added vector math library named libmvec with the following vectorized x86_64
1275 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1277 The library can be disabled with --disable-mathvec. Use of the functions is
1278 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1279 Shared library libmvec.so is linked in as needed when using -lm (no need to
1280 specify -lmvec explicitly for not static builds).
1281 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1283 * A new fmemopen implementation has been added with the goal of POSIX
1284 compliance. The new implementation fixes the following long-standing
1285 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1286 old implementation is still present for use be by existing binaries.
1288 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1289 releases. It has been fixed to match 2.19 and older, but binaries built
1290 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1292 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1293 Contributed by Roland McGrath (Google).
1295 * The header <regexp.h> is deprecated, and will be removed in a future
1296 release. Use of this header will trigger a deprecation warning.
1297 Application developers should update their code to use <regex.h> instead.
1299 This header was formerly part of SUS, but was deprecated in 1994 and
1300 removed from the standard in 2001. Also, the glibc implementation
1301 leaks memory. See BZ#18681 for more details.
1305 * The following bugs are resolved with this release:
1307 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1308 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1309 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1310 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1311 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1312 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1313 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1314 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1315 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1316 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1319 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1320 too little memory for the to-be-scanned arguments and overflow the
1321 allocated buffer. The implementation now correctly computes the required
1322 buffer size when using malloc, and switches to malloc from alloca as
1325 * A new semaphore algorithm has been implemented in generic C code for all
1326 machines. Previous custom assembly implementations of semaphore were
1327 difficult to reason about or ensure that they were safe. The new version
1328 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1329 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1330 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1332 * Port to Altera Nios II has been contributed by Mentor Graphics.
1334 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1335 implementations for powerpc64/powerpc64le.
1336 Implemented by Adhemerval Zanella (IBM).
1338 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1339 and powerpc64le. This may improve lock scaling of existing programs on
1340 HTM capable systems. The lock elision code is only enabled with
1341 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1342 powerpc will issue a transaction abort on every syscall to avoid side
1343 effects being visible outside transactions.
1345 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1346 AArch64. Contributed by ARM Ltd.
1348 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1350 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1351 under certain input conditions resulting in the execution of a shell for
1352 command substitution when the application did not request it. The
1353 implementation now checks WRDE_NOCMD immediately before executing the
1354 shell and returns the error WRDE_CMDSUB as expected.
1356 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1357 processing format strings with a large number of format specifiers.
1359 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1360 infinite loop if the DNS response contained a PTR record of an unexpected
1363 * The minimum GCC version that can be used to build this version of the GNU
1364 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1365 still be used to compile programs using the GNU C Library.
1367 * The GNU C Library is now built with -Werror by default. This can be
1368 disabled by configuring with --disable-werror.
1370 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1372 * The obsolete sigvec function has been removed. This was the original
1373 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1374 programs have been using instead for about 25 years. Of course, ABI
1375 compatibility for old binaries using sigvec remains intact.
1377 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1378 with newer versions of bison.
1380 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1381 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1382 registers overlay two consecutive single-precision registers. MIPS32R2
1383 introduced a new FPU mode (FR=1) where double-precision registers extend the
1384 corresponding single-precision registers which is incompatible with the
1385 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1386 require the use of FR=1 making a transition necessary. New o32 ABI
1387 extensions enable users to migrate over time from the original o32 ABI
1388 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1389 tracks the ABI of any loaded object and verifies that new objects are
1390 compatible. Mode transitions will also be requested as required and
1391 unsupportable objects will be rejected. The ABI checks include both soft and
1392 hard float ABIs for o32, n32 and n64.
1394 GCC 5 with GNU binutils 2.25 onwards:
1395 It is strongly recommended that all o32 system libraries are built using the
1396 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1397 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1398 compiler using --with-fp-32=xx to set this by default.
1402 * The following bugs are resolved with this release:
1404 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1405 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1406 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1407 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1408 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1409 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1410 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1411 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1412 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1413 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1414 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1415 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1416 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1417 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1418 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1420 * Reverted change of ABI data structures for s390 and s390x:
1421 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1422 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1423 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1424 are preserved pointing straight to the same implementation as the old ones.
1425 Given that, new callers will simply provide a too-big buffer to these
1426 functions. Any applications/libraries out there that embed jmp_buf or
1427 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1428 against 2.19 headers will have to rebuilt again. This is necessary in any
1429 case to revert the breakage in their ABI caused by the glibc change.
1431 * Support for file description locks is added to systems running the
1432 Linux kernel. The standard file locking interfaces are extended to
1433 operate on file descriptions, not file descriptors, via the use of
1434 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1435 are associated with an open file instead of a process.
1437 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1439 * The minimum Linux kernel version that this version of the GNU C Library
1440 can be used with is 2.6.32.
1442 * Running the testsuite no longer terminates as soon as a test fails.
1443 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1444 with PASS or FAIL lines for individual tests. A summary of the results is
1445 printed, including a list of failing lists, and "make check" exits with
1446 error status if there were any unexpected failures. "make check
1447 stop-on-test-failure=y" may be used to keep the old behavior.
1449 * The am33 port, which had not worked for several years, has been removed
1452 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1453 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1454 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1455 conflicted with POSIX (support for which was removed in 2.19), the
1456 interfaces those macros enabled remain available when compiling with
1457 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1458 test macros defined.
1460 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1462 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1463 This may improve lock scaling of existing programs on TX capable systems.
1464 The lock elision code is only built with --enable-lock-elision=yes and
1465 then requires a GCC version supporting the TX builtins. With lock elision
1466 default mutexes are elided via __builtin_tbegin, if the cpu supports
1467 transactions. By default lock elision is not enabled and the elision code
1470 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1471 copy the path argument. This allowed programs to cause posix_spawn to
1472 deference a dangling pointer, or use an unexpected pathname argument if
1473 the string was modified after the posix_spawn_file_actions_addopen
1476 * All supported architectures now use the main glibc sysdeps directory
1477 instead of some being in a separate "ports" directory (which was
1478 distributed separately before glibc 2.17).
1480 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1481 On configurations that support it (all Linux configurations), it's now
1482 used regardless of the --enable-add-ons switch to configure. It is no
1483 longer possible to build such configurations without pthreads support.
1485 * Locale names, including those obtained from environment variables (LANG
1486 and the LC_* variables), are more tightly checked for proper syntax.
1487 setlocale will now fail (with EINVAL) for locale names that are overly
1488 long, contain slashes without starting with a slash, or contain ".." path
1489 components. (CVE-2014-0475) Previously, some valid locale names were
1490 silently replaced with the "C" locale when running in AT_SECURE mode
1491 (e.g., in a SUID program). This is no longer necessary because of the
1494 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1495 with application code using Intel MPX instructions. (With all previous
1496 versions, the MPX register state could be clobbered when making calls
1497 into or out of a shared library.) Note that while the new dynamic
1498 linker is compatible with all known x86 hardware whether or not it
1499 supports Intel MPX, some x86 instruction-set emulators might fail to
1500 handle the new instruction encodings. This is known to affect Valgrind
1501 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1502 release), and might affect other tools that do instruction emulation.
1504 * Support for loadable gconv transliteration modules has been removed.
1505 The support for transliteration modules has been non-functional for
1506 over a decade, and the removal is prompted by security defects. The
1507 normal gconv conversion modules are still supported. Transliteration
1508 with //TRANSLIT is still possible, and the //IGNORE specifier
1509 continues to be supported. (CVE-2014-5119)
1511 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1512 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1513 resulting a denial-of-service security vulnerability in applications which
1514 use functions related to iconv. (CVE-2014-6040)
1518 * The following bugs are resolved with this release:
1520 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1521 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1522 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1523 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1524 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1525 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1526 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1527 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1528 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1529 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1530 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1531 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1532 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1533 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1534 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1535 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1536 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1537 16431, 16453, 16474, 16506, 16510, 16529
1539 * Slovenian translations for glibc messages have been contributed by the
1540 Translation Project's Slovenian team of translators.
1542 * The public headers no longer use __unused nor __block. This change is to
1543 support compiling programs that are derived from BSD sources and use
1544 __unused internally, and to support compiling with Clang's -fblock
1545 extension which uses __block.
1547 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1548 large collation sequences to optimize multiple passes. This cache
1549 computation may overflow for large collation sequences and may cause a
1550 stack or buffer overflow. This is now fixed to use a slower algorithm
1551 which does not use a cache if there is an integer overflow.
1553 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1554 rules for large collation sequences to optimize multiple passes and falls
1555 back to alloca if malloc fails, resulting in a possible stack overflow.
1556 The implementation now falls back to an uncached collation sequence lookup
1559 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1560 initialized for static applications resulting in the security feature
1561 being disabled. The pointer guard is now correctly initialized to a
1562 random value for static applications. Existing static applications need
1563 to be recompiled to take advantage of the fix (bug 15754).
1565 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1566 to the d_name member of struct dirent, or omit the terminating NUL
1567 character. (Bugzilla #14699).
1569 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1570 aligned_alloc functions could allocate too few bytes or corrupt the
1571 heap when passed very large allocation size values (Bugzilla #15855,
1574 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1575 for AF_INET6 has been fixed (Bugzilla #16072).
1577 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1578 pap_CW, quz_PE, the_NP.
1580 * Substantially revised locales: gd_GB, ht_HT
1582 * The LC_ADDRESS field was updated to support country_car for almost all
1585 * ISO 1427 definitions were updated.
1587 * ISO 3166 definitions were updated.
1589 * The localedef utility now supports --big-endian and --little-endian
1590 command-line options to generate locales for a different system from that
1591 for which the C library was built.
1593 * Binary locale files now only depend on the endianness of the system for
1594 which they are generated and not on other properties of that system. As a
1595 consequence, binary files generated with new localedef may be incompatible
1596 with old versions of the GNU C Library, and binary files generated with
1597 old localedef may be incompatible with this version of the GNU C Library,
1598 in the following circumstances:
1600 + Locale files may be incompatible on m68k systems.
1602 + Locale archive files (but not separate files for individual locales) may
1603 be incompatible on systems where plain "char" is signed.
1605 * The configure option --disable-versioning has been removed. Builds with
1606 --disable-versioning had not worked for several years.
1608 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1609 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1611 * SystemTap probes for malloc have been introduced.
1613 * SystemTap probes for slow multiple precision fallback paths of
1614 transcendental functions have been introduced.
1616 * Support for powerpc64le has been added.
1618 * The soft-float powerpc port now supports e500 processors.
1620 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1622 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1623 set of header declarations that are enabled by default, even when other
1624 feature test macros or compiler options such as -std=c99 would otherwise
1625 disable some of those declarations.
1627 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1628 conflict with POSIX. The libbsd-compat library (which was a dummy library
1629 that did nothing) has also been removed.
1631 * Preliminary documentation about Multi-Thread, Async-Signal and
1632 Async-Cancel Safety has been added.
1634 * Change of ABI data structures for s390 and s390x:
1635 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1636 allow for future hardware extensions. All library functions that accept or
1637 return these structures were versioned in order to provide backwards
1638 compatibility with existing code. However, not all cases can be handled
1639 with interface versioning. If an application embeds either structure into
1640 another structure and that structure is passed to another compilation unit
1641 which expects the newer larger-sized object then any access to the new
1642 fields will result in undefined behaviour. Similarly any access to fields
1643 that were moved by the enlarging of the structures will cause undefined
1644 behaviour. To fix the undefined behaviour all objects that are part of the
1645 final in-memory image must agree on the size of structures, and this may
1646 require recompilation.
1650 * The following bugs are resolved with this release:
1652 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1653 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1654 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1655 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1656 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1657 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1658 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1659 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1660 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1661 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1662 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1663 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1664 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1667 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1668 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1669 Distributions can re-enable building and using pt_chown via the new configure
1670 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1671 considerable security risks and should only be used if the distribution
1672 understands and accepts the risks.
1674 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1677 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1678 fixed (Bugzilla #15330).
1680 * Add support for calling C++11 thread_local object destructors on thread
1681 and program exit. This needs compiler support for offloading C++11
1682 destructor calls to glibc.
1684 * Improved worst case performance of libm functions with double inputs and
1687 * Support for priority inherited mutexes in pthread condition variables on
1688 non-x86 architectures.
1690 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1692 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1694 * Optimized string functions for ARM. Implemented by Will Newton and
1697 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1699 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1702 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1703 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1705 * Added a benchmark framework to track performance of functions in glibc.
1707 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1708 It is based on draft TS 18661 and currently enabled as a GNU extension.
1710 * On Linux, the clock function now uses the clock_gettime system call
1711 for improved precision, rather than old times system call.
1713 * Added support for version-3 tz format files. This is needed when using
1714 the tz database release 2013e or later, and affects a few unusual cases --
1715 currently only TZ='America/Godthab' for time stamps after 2037.
1717 * Added new API functions pthread_getattr_default_np and
1718 pthread_setattr_default_np to get and set the default pthread
1719 attributes of a process.
1721 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1722 This may improve lock scaling of existing programs on TSX capable systems.
1723 When the --enable-lock-elision=yes parameter is specified at configure
1724 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1727 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1728 availability). Implemented by Ryan S. Arnold.
1730 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1734 * The following bugs are resolved with this release:
1736 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1737 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1738 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1739 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1740 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1741 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1742 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1743 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1744 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1745 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1746 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1747 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1748 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1749 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1750 14879, 14889, 14893, 14898, 14914.
1752 * Optimization of memcpy for MIPS.
1754 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1755 EMFILE has been fixed (Bugzilla #14889).
1757 * The add-on ports collection is now distributed in the "ports" subdirectory
1758 of the main GNU C Library distribution, rather than separately.
1760 * Port to ARM AArch64 contributed by Linaro.
1762 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1763 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1765 Implemented by Andreas Krebbel.
1767 * The new function secure_getenv allows secure access to the environment,
1768 returning NULL if running in a SUID/SGID process. This function replaces
1769 the internal function __secure_getenv.
1771 * SystemTap static probes have been added into the dynamic linker.
1772 Implemented by Gary Benson.
1774 * Optimizations of string functions strstr, strcasestr and memmem.
1775 Implemented by Maxim Kuvyrkov.
1777 * The minimum Linux kernel version that this version of the GNU C Library
1778 can be used with is 2.6.16.
1780 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1781 powerpc POWER7. Implemented by Will Schmidt.
1783 * New configure option --disable-nscd builds the C library such that it
1784 never attempts to contact the Name Service Caching Daemon (nscd).
1785 New configure option --disable-build-nscd avoids building nscd itself;
1786 this is the default if --disable-nscd is used.
1788 * Improved support for cross-compilation, including cross-testing and
1789 bootstrap builds without a previously built glibc.
1791 * Several testsuite tests are now able to test multiple IFUNC variants of an
1792 interface, rather than just testing the one that would be chooen by
1795 * New configure options --with-bugurl and --with-pkgversion, for
1796 distributors to use to embed their bug-reporting and package version
1797 information in --help and --version output.
1799 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1800 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1801 allows creation of chroots without the procfs mounted on /proc.
1803 * The `crypt' function now fails if passed salt bytes that violate the
1804 specification for those values. On Linux, the `crypt' function will
1805 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1806 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1807 when the mode is enabled.
1809 * The `clock_*' suite of functions (declared in <time.h>) is now available
1810 directly in the main C library. Previously it was necessary to link with
1811 -lrt to use these functions. This change has the effect that a
1812 single-threaded program that uses a function such as `clock_gettime' (and
1813 is not linked with -lrt) will no longer implicitly load the pthreads
1814 library at runtime and so will not suffer the overheads associated with
1815 multi-thread support in other code such as the C++ runtime library.
1817 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1822 * The following bugs are resolved with this release:
1824 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1825 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1826 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1827 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1828 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1829 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1830 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1831 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1832 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1833 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1834 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1835 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1836 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1837 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1838 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1839 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1840 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1841 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1842 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1843 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1844 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1845 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1848 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1849 configuring glibc with:
1850 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1851 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1852 Implemented by H.J. Lu.
1856 + define static_assert
1858 + do not declare gets
1860 + declare at_quick_exit and quick_exit also for ISO C11
1862 + aligned_alloc. NB: The code is deliberately allows the size parameter
1863 to not be a multiple of the alignment. This is a moronic requirement
1864 in the standard but it is only a requirement on the caller, not the
1867 + timespec_get added
1869 + uchar.h support added
1871 + CMPLX, CMPLXF, CMPLXL added
1873 Implemented by Ulrich Drepper.
1875 * Support for the IA-64 has been moved to ports.
1877 * Remove support for anything but ELF binary format
1879 * Checking versions of poll, ppoll added.
1880 Implemented by Ulrich Drepper.
1882 * More generic and 64-bit performance optimizations to math functions.
1883 Implemented by Ulrich Drepper.
1885 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1886 headers and functions available at compile time as they were before
1887 version 2.14. This option will be removed at some time in the future
1888 after the TI-RPC library becomes fully sufficient for the needs of
1889 existing applications.
1891 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1892 Note that glibc is not expected to work with any Linux kernel version
1895 * New header <sys/auxv.h> and function getauxval allowing easy access to
1896 the AT_* key-value pairs passed from the Linux kernel. The header also
1897 defines the HWCAP_* bits associated with the AT_HWCAP key.
1899 * New locales: mag_IN
1901 * New configure option --enable-systemtap builds SystemTap static probes
1902 into libc for setjmp and longjmp and into libpthread for various operations.
1903 So far the setjmp/longjmp probes and some of the libpthread probes are
1904 provided only for i*86 and x86_64.
1905 Implemented by Roland McGrath and Rayson Ho.
1907 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1909 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1912 * More optimized functions for SPARC. Implemented by David S. Miller.
1914 * Improved support for cross-compilation, especially bootstrap builds
1915 without a previously built glibc.
1917 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1918 Chris Metcalf from Tilera.
1920 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1921 now supported for ARM processors.
1923 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1924 as the name of the dynamic linker, to distinguish it from the
1925 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1927 * Support for CRIS has been removed from ports.
1929 * A new class of installed header has been documented for low-level
1930 platform-specific functionality. PowerPC added the first instance with a
1931 function to provide time base register access. Contributed by Tulio
1932 Magno Quites Machado Filho.
1934 * ix86 configurations now install header files that are consistent with
1935 what x86-64 configurations install. These same header files can be used
1936 for -m32, -m64, or -mx32 builds.
1937 Contributed by H.J. Lu.
1939 * Math library bug fixes. A thorough audit of all open math library bugs was
1940 conducted by Joseph Myers. Significant progress was made on many math
1941 library bugs resulting in more accurate exceptions and function results.
1942 Many thanks to all those that contributed including Andreas Jaeger for his
1943 patch review and work on the x87 trigonometric instruction issues.
1945 * Timezone data is no longer installed. Timezone-related binaries and scripts
1946 will continue to be installed. Users should obtain their timezone data from
1947 their distribution provider or from the tzdata package at
1948 <ftp://munnari.oz.au/pub/>.
1952 * The following bugs are resolved with this release:
1954 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1955 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1956 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1957 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1958 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1959 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1960 13484, 13506, 13515, 13523, 13524, 13538, 13540
1962 * New program pldd to list loaded object of a process
1963 Implemented by Ulrich Drepper.
1965 * Add nss_db support back to glibc. No more dependency on Berkeley db
1966 and support for initgroups lookups.
1967 Implemented by Ulrich Drepper.
1969 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1970 Contributed by HJ Lu.
1972 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1973 Contributed by HJ Lu.
1975 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1976 on x86-32 and x86-64.
1977 Contributed by Liubov Dmitrieva.
1979 * Optimized strchr and strrchr for SSE on x86-32.
1980 Contributed by Liubov Dmitrieva.
1982 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1983 for x86-64 and x86-32.
1984 Contributed by Liubov Dmitrieva.
1986 * New interfaces: scandirat, scandirat64
1987 Implemented by Ulrich Drepper.
1989 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1990 Implemented by Ulrich Drepper.
1992 * nscd now also caches the netgroup database.
1993 Implemented by Ulrich Drepper.
1995 * Integrate libm with gcc's -ffinite-math-only option.
1996 Implemented by Ulrich Drepper.
1998 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1999 to math functions. Implemented by Ulrich Drepper.
2001 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2002 Implemented by Ulrich Drepper.
2004 * New Linux interfaces: process_vm_readv, process_vm_writev
2006 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2007 Implemented by Ulrich Drepper.
2009 * Optimized nearbyint and strcasecmp for PPC.
2010 Implemented by Adhemerval Zanella.
2012 * New locales: bho_IN, unm_US, es_CU, ta_LK
2016 * The following bugs are resolved with this release:
2018 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2019 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2020 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2021 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2022 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2023 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2024 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2025 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2026 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2027 12795, 12811, 12813, 12814, 12841
2029 * The RPC implementation in libc is obsoleted. Old programs keep working
2030 but new programs cannot be linked with the routines in libc anymore.
2031 Programs in need of RPC functionality must be linked against TI-RPC.
2032 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2034 Visible changes of this change include (obviously) the inability to link
2035 programs using RPC functions without referencing the TI-RPC library and the
2036 removal of the RPC headers from the glibc headers.
2037 Implemented by Ulrich Drepper.
2039 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2040 syncfs, setns, sendmmsg
2042 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2043 yue_HK, lij_IT, mhr_RU
2045 * New encodings: CP770, CP771, CP772, CP773, CP774
2047 * New program sotruss to trace calls through PLTs
2048 Implemented by Ulrich Drepper.
2050 * The malloc hook implementation is marked deprecated and will be removed
2051 from the default implementation in the next version. The design never
2052 worked ever since the introduction of threads. Even programs which do
2053 not create threads themselves can use multiple threads created internally.
2057 * The following bugs are resolved with this release:
2059 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2060 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2061 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2062 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2065 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2067 * POWER7 optimizations: memset, memcmp, strncmp
2069 * New optimized string functions for x86-64: strnlen (SSE2),
2070 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2071 Implemented by Ulrich Drepper.
2075 * The following bugs are resolved with this release:
2077 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2078 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2079 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2080 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2081 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2082 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2083 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2084 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2086 * New interfaces: pthread_getname_np, pthread_setname_np
2088 * New Linux interface: recvmmsg
2090 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2092 * The dynamic linker now recognizes supported ABI versions from the
2093 EI_ABIVERSION field in the ELF header.
2094 Implemented by Ulrich Drepper.
2096 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2097 will not be used to fill in password fields in the passwd.byname replies.
2098 Instead it is used to synthesize the shadow.byname table, should it be
2099 missing. This is a useful mode in some installations involving Solaris.
2100 Implemented by Ulrich Drepper.
2102 * New locales: kok_IN, sq_MK, cv_RU
2106 * The following bugs are resolved with this release:
2108 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2109 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2110 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2111 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2112 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2113 10789, 10817, 10823, 10840, 10847
2115 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2117 Implemented by Ulrich Drepper.
2119 * Checking version of longjmp added that fails if an uninitialized stack
2120 frame would be created. Implemented by Ulrich Drepper.
2122 * STT_GNU_IFUNC is now supported in static executables.
2123 Implemented by H.J. Lu.
2125 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2127 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2128 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2129 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2130 Contributed by H.J. Lu.
2132 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2133 Implemented by Ulrich Drepper.
2135 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2137 Contributed by H.J. Lu.
2139 * Support for fma instruction in AVX on x86-64.
2140 Implemented by H.J. Lu and Ulrich Drepper.
2142 * AVX support in x86-64 auditing support in ld.so.
2143 Implemented by H.J. Lu.
2145 * STB_GNU_UNIQUE support added.
2146 Implemented by Ulrich Drepper.
2148 * Implement second fallback mode for optimized DNS lookup for even more
2149 broken environments. If two requests from the same port are not
2150 handled correctly close the socket and open a new one before sending
2151 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2152 can be used to select this mode right away, instead of rediscovering the
2153 necessity is every process again.
2154 Implemented by Ulrich Drepper.
2156 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2157 Implemented by Adam Tkac.
2159 * Optimized iconv conversions for S390x.
2160 Implemented by Andreas Krebbel.
2162 * Using condvars with PI mutexes is now more efficient due to kernel
2163 support for requeueing to PI futexes. NPTL support added for x86-64.
2164 Implemented by Ulrich Drepper.
2166 * New locales: ps_AF, my_MM
2170 * The following bugs are resolved with this release:
2172 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2173 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2174 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2175 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2176 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2178 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2179 Implemented by Ulrich Drepper.
2181 * New Linux interfaces: accept4, fallocate, fallocate64.
2182 Implemented by Ulrich Drepper.
2184 * Correct declarations of string function when used in C++ code. This
2185 could lead to compile errors for invalid C++ code.
2187 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2190 * New POSIX 2008 interface: psiginfo
2191 Implemented by Ulrich Drepper.
2193 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2194 Implemented by Ulrich Drepper.
2196 * Support for selecting between multiple function definitions at runtime
2197 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2199 * The libcrypt library can now use the hash function implementations in
2200 NSS. Implemented by Ulrich Drepper.
2202 * The malloc implementation can be compiled to be less memory efficient
2203 but higher performing in multi-threaded programs.
2204 Implemented by Ulrich Drepper.
2206 * New locales: nan_TW@latin, ks_IN
2208 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2209 Implemented by Ulrich Drepper.
2211 * Extended printf hook support. It is possible to use user-defined types
2212 and extend existing format specifiers.
2213 Implemented by Ulrich Drepper.
2215 * Handling for group shadow files has been added.
2216 Implemented by Ulrich Drepper.
2218 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2219 servers (the case, e.g., for some people using the built-in DNS
2220 server in ADSL modems/routers). There is a once-per-process timeout
2221 in case of a broken server. To avoid it, users can run nscd or put
2222 'options single-request' in /etc/resolv.conf.
2223 Implemented by Ulrich Drepper.
2227 * The following bugs are resolved with this release:
2229 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2230 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2231 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2232 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2233 6968, 6974, 6980, 6995, 7008, 7009, 7029
2235 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2236 up at the same time. Implemented by Ulrich Drepper.
2238 * TLS descriptors for LD and GD on x86 and x86-64.
2239 Implemented by Alexandre Oliva.
2241 * getaddrinfo now handles DCCP and UDPlite.
2242 Implemented by Ulrich Drepper.
2244 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2245 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2246 Implemented by Ulrich Drepper.
2248 * New implementation of memmem, strstr, and strcasestr which is O(n).
2249 Implemented by Eric Blake.
2251 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2253 * Implement "e" option for popen to open file descriptor with the
2254 close-on-exec flag set. Implemented by Ulrich Drepper.
2256 * Implement "b" mode for fmemopen. In this mode writes writes don't
2257 implicitly add a NUL byte and seeks from the end of the buffer really
2258 use the buffer end, not the string length as the basis.
2259 Implemented by Ulrich Drepper.
2261 * Many functions, exported and internal, now atomically set the close-on-exec
2262 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2264 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2266 Implemented by Pravin Satpute.
2268 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2272 * The following bugs are resolved with this release:
2274 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2275 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2276 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2277 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2278 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2279 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2280 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2282 * New locales: bo_CN, bo_IN, shs_CA.
2284 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2286 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2287 Implemented by Pravin Satpute.
2289 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2291 * nscd caches now all timeouts for DNS entries
2292 Implemented by Ulrich Drepper.
2294 * nscd is more efficient and wakes up less often.
2295 Implemented by Ulrich Drepper.
2297 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2298 vdprintf, and obstack_vprintf.
2299 Implemented by Jakub Jelinek.
2301 * Faster memset for x86-64.
2302 Implemented by Harsha Jagasia and H.J. Lu.
2304 * Faster memcpy on x86.
2305 Implemented by Ulrich Drepper.
2307 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2308 Implemented by Ulrich Drepper.
2310 * Faster sqrt and sqrtf implemention for some PPC variants.
2311 Implemented by Steven Munroe.
2315 * The following bugs are resolved with this release:
2317 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2318 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2319 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2320 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2321 5071, 5103, 5104, 5112, 5113, 5184, 5186
2323 * More checking functions: fread, fread_unlocked, open*, mq_open.
2324 Implemented by Jakub Jelinek and Ulrich Drepper.
2326 * Extend fortification to C++. Implemented by Jakub Jelinek.
2328 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2329 by not recognizing 'a' as a modifier when those specs are requested.
2330 Implemented by Jakub Jelinek.
2332 * PPC optimizations to math and string functions.
2333 Implemented by Steven Munroe.
2335 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2336 options to be passed. Implemented by Ulrich Drepper.
2338 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2340 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2341 Implemented by Ulrich Drepper.
2343 * Handle private futexes in the NPTL implementation.
2344 Implemented by Jakub Jelinek and Ulrich Drepper.
2346 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2347 Implemented by Roland McGrath and Ulrich Drepper.
2349 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2351 * SHA-256 and SHA-512 based password encryption.
2352 Implemented by Ulrich Drepper.
2354 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2355 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2358 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2359 Implemented by Ulrich Drepper.
2363 * The following bugs are resolved with this release:
2365 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2366 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2367 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2368 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2369 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2370 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2371 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2374 * New Linux interfaces: epoll_pwait, sched_getcpu.
2376 * New generic interfaces: strerror_l.
2378 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2382 * The following bugs are resolved with this release:
2384 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2385 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2386 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2387 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2388 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2389 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2390 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2391 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2392 3225, 3251, 3252, 3253, 3273
2394 * For Linux, the sorting of addresses returned by getaddrinfo now also
2395 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2396 Implemented by Ulrich Drepper.
2398 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2399 Implemented by Ulrich Drepper.
2401 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2403 * New iconv module for MIK. Contributed by Alexander Shopov.
2405 * For sites with broken group and/or passwd database, the auto-propagate
2406 option of nscd can prevent creating ID lookup entries from the results
2407 of a name lookup and vice versa. This usually is no problem but some
2408 site might have problems with the default behavior.
2409 Implemented by Ulrich Drepper.
2411 * Iterating over entire database in NIS can be slow. With the
2412 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2413 to trade time for memory. The entire database will be read at once.
2414 Implemented by Ulrich Drepper.
2416 * The interfaces introduced in RFC 3542 have been implemented by
2419 * Support for the new ELF hash table format was added by Ulrich Drepper.
2421 * Support for priority inheritance mutexes added by Jakub Jelinek and
2424 * Support for priority protected mutexes added by Jakub Jelinek.
2426 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2427 or_IN, csb_PL, fy_NL, sr_ME.
2431 * More overflow detection functions.
2433 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2434 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2436 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2437 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2438 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2439 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2440 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2441 by Masahide Washizawa.
2443 * It is now possible to install audit libraries for the dynamic linker using
2444 LD_AUDIT. Implemented by Ulrich Drepper.
2446 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2447 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2448 For a libc and libpthread that works well on Linux 2.4 kernels, we
2449 recommend using the stable 2.3 branch.
2451 * The new function open_wmemstream acts like open_memstream,
2452 but uses a wchar_t wide character buffer.
2454 * The new function ppoll is an improved version of poll, similar to pselect.
2456 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2457 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2458 renameat, unlinkat, linkat, symlinkat, readlinkat.
2460 * New Linux kernel system calls: unshare,
2461 inotify_init, inotify_add_watch, inotify_rm_watch.
2463 * The euidaccess function is now also known by the alias eaccess,
2464 for compatibility with some other systems.
2466 * Timezone data updated to 2006b version.
2470 * The following bugs are resolved with this release:
2472 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2473 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2474 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2475 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2476 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2477 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2479 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2481 * As of this release, GCC 4 can be used to compile the C Library.
2483 * Timezone data updated to 2005m version.
2487 * The following bugs are resolved with this release:
2489 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2490 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2491 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2492 777, 787, 821, 822, 823, 825
2494 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2498 * Support for RFC 3678. Real implementations exist only for Linux so far.
2499 Implemented by Ulrich Drepper.
2501 * nscd can now cache entries persistently. Expiring entries are reloaded.
2502 For speedups the cache can be shared in memory with client processes.
2503 Implemented by Ulrich Drepper.
2505 * nscd can now perform SELinux checks.
2506 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2508 * getaddrinfo queries are now cached. Canonical name lookup is performed
2510 Implemented by Ulrich Drepper.
2512 * The nothrow function attribute is used when headers are used by gcc when
2513 compiling C code. This can avoid significant amounts of exception
2516 * The malloc functions perform more error checking and are stricter when
2517 it comes to reacting on errors. The default action is to terminate
2518 the process after showing an error message. Implemented by Ulrich Drepper.
2520 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2521 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2523 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2524 namespaces using the new function dlmopen(). This feature is of course,
2525 like most other dynamic loading functionality, not available in statically
2526 linked applications. Implemented by Ulrich Drepper.
2528 * Low-overhead boundary checking variants of string and some stdio functions
2529 were added. These are to be used in conjunction with a gcc patch by
2530 Jakub Jelinek which adds calls to these functions if possible.
2531 Implemented by Jakub Jelinek and Ulrich Drepper.
2533 * Old code for several operating systems and machine architectures that
2534 have not been in working condition in a long time have been removed from
2535 the main source tree maintained by the GNU C Library's maintainers.
2536 These files are now reside in the separate `ports' source module
2537 that is usable as an add-on when building the library.
2541 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2542 interrogate the dynamic linker, compatible with the Solaris interface.
2544 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2545 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2547 * getifaddrs now uses the netlink interface on Linux to get its information.
2548 Implemented by Thorsten Kukuk.
2550 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2551 Implemented by Ulrich Drepper.
2553 * support for non-executable stacks on x86 has been added. Changes mostly
2556 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2559 * getaddrinfo now performs destination address selection according to
2564 * Thread-safe interfaces for many functions that access locale data
2565 were added in version 2.3, but these features were omitted from NEWS.
2566 Many functions have variants with an `_l' suffix that take a `locale_t'
2567 object as a parameter rather than consulting the current locale.
2568 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2569 create and maintain `locale_t' objects. Additionally, the new function
2570 `uselocale' sets "the current locale" (as used by functions not so
2571 parameterized) set for an individual thread. These features were added
2572 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2574 * The functions getresuid, getresgid, setresuid, and setresgid, which
2575 have long been available on Linux, are now declared in <unistd.h>
2576 and are now also available on the Hurd.
2578 * ELF thread-local storage support (TLS) now works on x86-64.
2580 * The new dynamic string token $LIB is expanded in shared library names.
2581 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2583 * Aldy Hernandez contributed complete software floating point support for
2584 PowerPC machines with no FPU.
2586 * fexecve is implemented on Linux.
2588 * The `btowc' function should work at least twice as fast due to
2589 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2591 * With approriate thread add-ons cancelable functions are now implemented
2592 in libc.so as well. No need to call the function in libpthread. This
2593 change allowed to finally disable the incorrect and expensive handling
2594 of weak definition in ld.so.
2596 * Yet more PLT entries in libc.so have been removed. We finally arrived
2597 at the bare minimum. Startup times improved appropriately.
2599 * Support for the new Linux/x86 system call interface was added. The
2600 AT_SYSINFO auxiliary vector entry is recognized and handled.
2604 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2607 * iconv (the program and the interface) now accepts empty names (excluding
2608 options like //TRANSLIT) to mean "use charset of current locale".
2610 * localedef can now transliterate characters in strings which are not in
2611 the provided charmap. The information from the input locale is used.
2613 * Prelinking support was added for ELF targets. This requires additional
2614 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2616 * Read-only stdio streams now use mmap to speed up operation by eliminating
2617 copying and buffer underflows. To use add 'm' to the mode string of
2618 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2620 * The malloc functions were completely rewritten by Wolfram Gloger based
2621 on Doug Lea's malloc-2.7.0.c.
2623 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2624 implementation of regex.
2626 * Bruno Haible upgraded the iconv and locale implementation to support
2629 * Contents of the LC_* and LANG environment variables in the CEN style are
2630 not recognized anymore. It never was used. Change by Ulrich Drepper.
2632 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2633 thread-local storage (TLS) ABI on some platforms.
2634 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2636 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2637 EUC-JISX0213, and TSCII.
2639 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2640 BSD-compatible interface for getting all network interface addresses.
2641 Implementation for IPv4 by Roland McGrath.
2643 * Loading of locale data is faster due to the introduction of a locale
2644 archive. Implemented by Roland McGrath and Ulrich Drepper.
2646 * Startup times are significantly reduced by not using exported functions
2647 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2650 * Steven Munroe contributed a port to PowerPC64/Linux.
2654 * The Hurd now uses the GNU libio implementation of stdio rather than the
2655 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2657 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2658 with the same behavior as the Linux system call of the same name.
2662 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2663 128-bit long double format.
2665 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2666 IBM1161, and IBM1162 charsets.
2668 * Andreas Jaeger contributed a port to x86-64/Linux.
2670 * Peter Bruin contributed a port to PowerPC/Hurd.
2672 * libc functions using I/O streams now can handle wide-oriented streams
2675 * optimizations in the dynamic linker. Binaries created by recent binutils
2676 versions start up quicker due to reduced time spent on relocations.
2678 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2682 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2683 asinh, atanh, j0 for the 128-bit long double format.
2685 * Bruno Haible updated all the code handling Unicode in some form to
2686 support Unicode 3.1.
2688 * Speed of regex for single-byte locales is back to previous levels.
2689 Patch by Isamu Hasegawa.
2691 * Alpha, SPARC, and IA-64 now also using floating stacks.
2693 * Startup time of internationalized applications greatly improved through
2694 iconv cache. Use iconvconfig to generate the cache file.
2695 Contributed by Ulrich Drepper.
2697 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2698 severe performance problems. Patch by David Mosberger.
2700 * The Hurd port got a lot more functionality like AIO, various stdio
2701 extensions, etc. Mainly done by Roland McGrath.
2703 * mtrace can now lookup symbols in shared libraries.
2707 * Intel's IA-64 math library is largely integrated. It provides fast and
2708 accurate implementatations for most basic and standard math functions
2709 in float, double, and long double format.
2711 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2712 and asin for the 96-bit long double format and asin, log, tan for the
2713 128-bit long double format.
2715 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2716 The basic double functions exist today. Contributed by Abraham Ziv
2717 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2718 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2720 * An asynchronous name lookup library was added. The interface is designed
2721 after POSIX AIO. The proposal was circulated beforehand to get comments.
2722 No negative ones came in. Implemented by Ulrich Drepper.
2724 * Port to S390/64bit contributed by Martin Schwidefsky
2725 <schwidefsky@de.ibm.com>.
2727 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2728 of functions for Linux/IA-64.
2730 * The RPC code is now thread safe. Threads can now use the same service
2731 of different services at the same time. Patch by Eric Norum
2732 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2734 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2735 family of functions for Linux/S390.
2737 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2738 of functions for Linux/x86.
2740 * Port to Linux/CRIS contributed by Axis Communications.
2744 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2745 we can now check for namespace violations and missing declarations. The
2746 result is that almost all headers are now Unix-compliant (as defined in
2747 the upcoming XPG6). The negative side is that some programs might need
2748 corrections, too, if they depend on the incorrect form of the headers in
2749 previous versions which defined too many symbols and included too many
2752 * regex now handles multibyte character sets correctly.
2753 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2755 * iconv (the program) does now conform to the upcoming XPG6 and handles
2756 charmaps. Instead of the charset names the path of charmaps can be
2757 provided and the conversion happens based on this data.
2758 Contributed by Ulrich Drepper.
2760 * The locale program now provides detailed information about the installed
2765 only lists the names of the supported locales
2769 provides details such as country, language, and codeset name.
2770 Contributed by Ulrich Drepper.
2774 * The gencat program now parses the input file according to the charset
2775 selected by the LC_CTYPE category. This is important for stateful
2776 character sets. To make generating catalogs easier there is a way
2777 to overwrite the charset selected by the locale: before the first
2778 message or $ quote line the catalog can contain a line like
2780 $ codeset=ISO-8859-2
2782 to select the charset (ISO-8859-2 in this case).
2784 Implemented by Shinya Hanataka and Ulrich Drepper.
2786 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2787 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2788 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2790 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2791 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2793 * The LANGUAGE environment variable is now ignored unless the locale is
2794 changed from the default "C" locale.
2796 * The usual bug fixes.
2800 * Greg McGary added runtime support for bounds checking using gcc's
2801 new -fbounded-pointers option. ix86 target is complete. PowerPC
2804 * Thorsten Kukuk added secure mode support to nscd.
2806 * The Berkeley db libraries have been removed.
2808 Related, the nss_db module is now in a separate package since it
2809 obviously requires a database library being available.
2811 * Wide character I/O streams implemented by Ulrich Drepper.
2813 * Functions from the extended socket API added by Ulrich Drepper.
2815 * Functions feenableexcept and fedisableexcept to control the
2816 behaviour of individual exceptions have been added by Andreas Jaeger.
2818 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2820 * The resolver code has been updated from bind 8.2.3-T5B which supports
2821 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2824 This change could in some situations effect backward compatibility. Since
2825 now `_res' is a thread-local instead of a global variable, modifying it
2826 in one thread does not have any effect in other threads.
2828 The resolver library was also extended to allow IPv6 as the transport
2829 protocol for the requests. This work was done by Stig Venaas.
2831 * Compatibility code for K&R C compilers has been removed from the
2832 header files. A ISO C compiler is needed to use the library
2833 (conforming to either C89 or C99 standard).
2835 * Complete rewrite of the localedef program to support multibyte character
2836 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2837 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2838 Implemented by Ulrich Drepper.
2840 Bruno Haible significantly improved the generation and use of the data
2841 structures for the wide character tables.
2843 * Plural handling in gettext implemented by Ulrich Drepper.
2845 * The utmp daemon has been removed.
2847 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2849 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2852 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2854 * POSIX barriers implemented by Kaz Kylheku.
2856 * POSIX spawn function family implemented by Ulrich Drepper.
2858 * POSIX spinlocks are now available.
2860 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2862 * the configure option --enable-kernel=X.Y.Z allows to strip out
2863 compatibility for kernel versions before X.Y.Z. This is currently only
2864 implemented for Linux.
2866 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2867 field sin6_scope_id. This means that all programs using IPv6 should be
2868 recompiled. Don't expect binary compatibility with previous glibc
2871 * various conversion modules for IBM character sets contributed by
2874 * IA-64 port by Jes Sorensen and HJ Lu.
2888 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2890 * New xdr functions are added; some rpc functions are now 64bit clean.
2892 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2894 * Fixed known incompatibilities with glibc 2.0.
2896 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2898 * Optimized a number of functions (especially the ELF dynamic loader).
2900 * Update timezone data files.
2902 * lots of charmaps corrections
2904 * some new locale definitions and charmaps
2909 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2910 conform to POSIX member type requirements. Symbol versions have been
2911 adjusted as needed within the library, and for direct use by applications,
2912 but there is potential for problems if third-party libraries use
2913 struct timeval as part of their interface. This does not present
2914 a problem for X and other "essential" system libraries.
2916 * An additional locale model to support C++ Standard Library locale
2917 model and probably more was implemented by Ulrich Drepper.
2919 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2922 * Miles Bader provided the `argp' function family to support hierachical
2923 command line argument parsing, layered on top of getopt.
2925 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2927 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2930 * scanf recognizes the %a and %A format for scanning floating point numbers.
2932 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2933 information and interfaces for the available integer types.
2935 * about 130 new math functions were added to implement the ISO C9x math
2938 * the new header <complex.h> contains definitions of the complex math
2939 functions from ISO C 9X.
2941 * the new header <tgmath.h> defines generic macros to use complex or
2942 real valued functions.
2944 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2946 * Andreas Jaeger provided a test suite for the math library.
2948 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2950 * Ulrich Drepper added character set conversion functions (iconv).
2952 * Optimized string functions have been added.
2954 * The localedata addon is now part of glibc.
2956 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2958 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2959 daemon for NSS (nscd).
2961 Missing a better place here are some numbers on improvements. Under
2962 Linux 2.1.125 un-tar-ing the kernel sources takes
2966 using local files 12.19s 6.88s 22.91s
2968 using NIS 13.92s 8.91s 26.34s
2970 using NIS & nscd 10.37s 7.34s 25.30s
2972 using NIS+ 27.57s 30.37s 640.46s
2974 using NIS+ & nscd 10.25s 7.83s 26.51s
2976 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2978 Keep in mind that non-namelookup related operations dominate above times.
2979 It was just a common complain that using NIS+ unpacking the kernel is
2982 [1] The old nscd implementation is not available anymore since it was
2983 distributed with glibc up to version 2.0.98 and thus is now replaced.
2985 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2987 * Mark Kettenis provided a Hesiod NSS module.
2989 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2990 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2992 * rcmd can now handle netgroups (Dick Streefland).
2994 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2997 * Support for the IPv6 protocol has been added to the socket API, as per the
2998 latest draft standards.
3000 * Support for Linux 2.2 has been added.
3002 * Interface changes relative to the latest 2.0.x release:
3003 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3004 addseverity NEW: Unix98
3005 alphasort64 NEW: LFS
3006 argp_err_exit_status NEW: argp, GNU ext
3007 argp_error NEW: argp, GNU ext
3008 argp_failure NEW: argp, GNU ext
3009 argp_help NEW: argp, GNU ext
3010 argp_parse NEW: argp, GNU ext
3011 argp_program_bug_address NEW: argp, GNU ext
3012 argp_program_version NEW: argp, GNU ext
3013 argp_program_version_hook NEW: argp, GNU ext
3014 argp_state_help NEW: argp, GNU ext
3015 argp_usage NEW: argp, GNU ext
3016 authdes_create NEW: Secure RPC
3017 authdes_getucred NEW: Secure RPC
3018 authdes_pk_create NEW: Secure RPC
3019 backtrace NEW: GNU ext.
3020 backtrace_symbols NEW: GNU ext.
3021 backtrace_symbols_fd NEW: GNU ext.
3023 cacosf NEW: ISO C 9x
3024 cacosh NEW: ISO C 9x
3025 cacoshf NEW: ISO C 9x
3026 cacoshl NEW: ISO C 9x
3027 cacosl NEW: ISO C 9x
3034 casinf NEW: ISO C 9x
3035 casinh NEW: ISO C 9x
3036 casinhf NEW: ISO C 9x
3037 casinhl NEW: ISO C 9x
3038 casinl NEW: ISO C 9x
3040 catanf NEW: ISO C 9x
3041 catanh NEW: ISO C 9x
3042 catanhf NEW: ISO C 9x
3043 catanhl NEW: ISO C 9x
3044 catanl NEW: ISO C 9x
3045 cbc_crypt NEW: Secure RPC
3049 ccoshf NEW: ISO C 9x
3050 ccoshl NEW: ISO C 9x
3056 cimagf NEW: ISO C 9x
3057 cimagl NEW: ISO C 9x
3058 clearerr_locked REMOVED
3059 clntunix_create NEW: sunrpc ext
3061 clog10 NEW: ISO C 9x
3062 clog10f NEW: ISO C 9x
3063 clog10l NEW: ISO C 9x
3073 cprojf NEW: ISO C 9x
3074 cprojl NEW: ISO C 9x
3076 crealf NEW: ISO C 9x
3077 creall NEW: ISO C 9x
3082 csinhf NEW: ISO C 9x
3083 csinhl NEW: ISO C 9x
3086 csqrtf NEW: ISO C 9x
3087 csqrtl NEW: ISO C 9x
3091 ctanhf NEW: ISO C 9x
3092 ctanhl NEW: ISO C 9x
3094 des_setparity NEW: Secure RPC
3095 ecb_crypt NEW: Secure RPC
3096 endutxent NEW: Unix98
3098 exp10f NEW: ISO C 9x
3099 exp10l NEW: ISO C 9x
3103 fattach NEW: STREAMS
3104 fdetach NEW: STREAMS
3108 feclearexcept NEW: ISO C 9x
3109 fegetenv NEW: ISO C 9x
3110 fegetexceptflag NEW: ISO C 9x
3111 fegetround NEW: ISO C 9x
3112 feholdexcept NEW: ISO C 9x
3114 feraiseexcept NEW: ISO C 9x
3115 ferror_locked REMOVED
3116 fesetenv NEW: ISO C 9x
3117 fesetexceptflag NEW: ISO C 9x
3118 fesetround NEW: ISO C 9x
3119 fetestexcept NEW: ISO C 9x
3120 feupdateenv NEW: ISO C 9x
3121 fflush_locked REMOVED
3125 fgets_unlocked NEW: GNU ext.
3126 fileno_locked REMOVED
3138 fputc_locked REMOVED
3139 fputs_unlocked NEW: GNU ext.
3140 fread_unlocked NEW: GNU ext.
3145 fstatvfs NEW: Unix98
3149 ftruncate64 NEW: LFS
3151 fwrite_unlocked NEW: GNU ext.
3152 gai_strerror NEW: IPv6
3156 getchar_locked REMOVED
3158 getdate_err NEW: Unix98
3159 getdate_r NEW: GNU ext.
3161 getnameinfo NEW: IPv6
3162 getnetname NEW: Secure RPC
3163 getpmsg NEW: STREAMS
3164 getpt NEW: Unix98 PTY
3165 getrlimit64 NEW: LFS
3166 getutxent NEW: Unix98
3167 getutxid NEW: Unix98
3168 getutxline NEW: Unix98
3169 glob64 NEW: GNU ext.
3170 globfree64 NEW: GNU ext.
3171 gnu_get_libc_release NEW: GNU ext.
3172 gnu_get_libc_version NEW: GNU ext.
3173 grantpt NEW: Unix98 PTY
3174 host2netname NEW: Secure RPC
3176 iconv_close NEW: iconv
3177 iconv_open NEW: iconv
3178 if_freenameindex NEW: IPv6
3179 if_indextoname NEW: IPv6
3180 if_nameindex NEW: IPv6
3181 if_nametoindex NEW: IPv6
3182 in6addr_any NEW: IPv6
3183 in6addr_loopback NEW: IPv6
3184 inet6_isipv4mapped NEW: IPv6
3185 isastream NEW: STREAMS
3186 iswblank NEW: GNU ext.
3187 key_decryptsession NEW: Secure RPC
3188 key_decryptsession_pk NEW: Secure RPC
3189 key_encryptsession NEW: Secure RPC
3190 key_encryptsession_pk NEW: Secure RPC
3191 key_gendes NEW: Secure RPC
3192 key_get_conv NEW: Secure RPC
3193 key_secretkey_is_set NEW: Secure RPC
3194 key_setnet NEW: Secure RPC
3195 key_setsecret NEW: Secure RPC
3196 llrint NEW: ISO C 9x
3197 llrintf NEW: ISO C 9x
3198 llrintl NEW: ISO C 9x
3199 llround NEW: ISO C 9x
3200 llroundf NEW: ISO C 9x
3201 llroundl NEW: ISO C 9x
3206 lrintf NEW: ISO C 9x
3207 lrintl NEW: ISO C 9x
3208 lround NEW: ISO C 9x
3209 lroundf NEW: ISO C 9x
3210 lroundl NEW: ISO C 9x
3212 makecontext NEW: Unix98
3213 mempcpy NEW: GNU ext.
3216 modify_ldt NEW: kernel
3220 nearbyint NEW: ISO C 9x
3221 nearbyintf NEW: ISO C 9x
3222 nearbyintl NEW: ISO C 9x
3223 netname2host NEW: Secure RPC
3224 netname2user NEW: Secure RPC
3225 nexttoward NEW: ISO C 9x
3226 nexttowardf NEW: ISO C 9x
3227 nexttowardl NEW: ISO C 9x
3231 passwd2des NEW: Secure RPC
3233 pow10f NEW: GNU ext.
3234 pow10l NEW: GNU ext.
3237 printf_size NEW: GNU ext.
3238 printf_size_info NEW: GNU ext.
3239 profil_counter REMOVED
3240 pthread_mutexattr_getkind_np REPLACED
3241 pthread_mutexattr_setkind_np REPLACED
3242 ptsname NEW: Unix98 PTY
3243 ptsname_r NEW: Unix98 PTY
3245 putchar_locked REMOVED
3246 putgrent NEW: GNU ext.
3248 putpmsg NEW: STREAMS
3249 pututxline NEW: Unix98
3253 readdir64_r NEW: LFS
3254 remquo NEW: ISO C 9x
3255 remquof NEW: ISO C 9x
3256 remquol NEW: ISO C 9x
3258 roundf NEW: ISO C 9x
3259 roundl NEW: ISO C 9x
3261 scalbln NEW: ISO C 9x
3262 scalblnf NEW: ISO C 9x
3263 scalblnl NEW: ISO C 9x
3265 sendfile NEW: kernel
3266 setcontext NEW: Unix98
3267 setrlimit64 NEW: LFS
3268 setutxent NEW: Unix98
3270 sigignore NEW: Unix98
3271 sigqueue NEW: POSIX.1b
3272 sigrelse NEW: Unix98
3273 sigset NEW: POSIX.1b
3274 sigtimedwait NEW: POSIX.1b
3275 sigwaitinfo NEW: POSIX.1b
3276 sincos NEW: GNU ext.
3277 sincosf NEW: GNU ext.
3278 sincosl NEW: GNU ext.
3282 strcasestr NEW: GNU ext.
3283 strtoimax NEW: ISO C 9x
3284 strtoumax NEW: ISO C 9x
3285 strverscmp NEW: GNU ext.
3286 svcauthdes_stats NEW: Secure RPC
3287 svcunix_create NEW: sunrpc etx
3288 svcunixfd_create NEW: sunrpc ext
3289 swapcontext NEW: Unix98
3290 tcgetsid NEW: Unix98 PTY
3291 tdestroy NEW: GNU ext.
3292 tgamma NEW: ISO C 9x
3293 tgammaf NEW: ISO C 9x
3294 tgammal NEW: ISO C 9x
3298 truncf NEW: ISO C 9x
3299 truncl NEW: ISO C 9x
3301 unlockpt NEW: Unix98 PTY
3302 updwtmpx NEW: Unix98
3303 user2netname NEW: Secure RPC
3304 utmpxname NEW: Unix98
3305 versionsort NEW: GNU ext.
3306 versionsort64 NEW: GNU ext.
3308 wcscasecmp NEW: GNU ext.
3309 wcsncasecmp NEW: GNU ext.
3310 wcsnlen NEW: GNU ext.
3311 wcstoimax NEW: ISO C 9x
3312 wcstoll NEW: ISO C 9x
3313 wcstoull NEW: ISO C 9x
3314 wcstoumax NEW: ISO C 9x
3316 wordexp NEW: POSIX.2
3317 wordfree NEW: POSIX.2
3318 write_profiling REMOVED
3319 xdecrypt NEW: Secure RPC
3320 xdr_authdes_cred NEW: Secure RPC
3321 xdr_authdes_verf NEW: Secure RPC
3322 xdr_cryptkeyarg NEW: Secure RPC
3323 xdr_cryptkeyarg2 NEW: Secure RPC
3324 xdr_cryptkeyres NEW: Secure RPC
3325 xdr_getcredres NEW: Secure RPC
3326 xdr_key_netstarg NEW: Secure RPC
3327 xdr_key_netstres NEW: Secure RPC
3328 xdr_keybuf NEW: Secure RPC
3329 xdr_keystatus NEW: Secure RPC
3330 xdr_netnamestr NEW: Secure RPC
3331 xdr_sizeof NEW: Secure RPC
3332 xdr_unixcred NEW: sunrpc ext
3333 xencrypt NEW: Secure RPC
3334 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3345 * inet_ntoa is thread-safe
3347 * updwtmp is moved from libutil to libc
3349 * rewrite of cbrt function
3351 * update of timezone data
3365 * add atoll function
3367 * fix complex problems in Berkeley DB code
3369 * fix math functions
3373 * fixed lots of header problems (especially Linux/GNU specific)
3375 * dynamic loader preserves all registers
3377 * Roland McGrath provided support for handling of auxiliary objects in
3378 the ELF dynamic loader.
3380 * support for parallel builds is improved
3384 * GNU extensions are no longer declared by default. To enable them you
3385 must define the macro `_GNU_SOURCE' in your program or compile with
3388 * The library has changed from using GNU ld symbol aliases to using weak
3389 symbols where available. The ELF object file format supports weak
3390 symbols; GNU ld also supports weak symbols in the a.out format. (There
3391 is also now support for other GNU ld extensions in ELF. Use the
3392 `--with-elf' option to configure to indicate you have ELF, and
3393 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3394 of many files which contained only symbol aliases, reducing the size of
3395 the source and the compiled library; many other files were renamed to
3396 less cryptic names previously occupied by the symbol alias files.
3397 There is a new header file <elf.h> for programs which operate on
3398 files in the ELF format.
3400 * Converted to Autoconf version 2, so `configure' has more options.
3401 Run `configure --help' to see the details.
3403 * The library can now be configured to build profiling, highly-optimized
3404 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3405 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3406 `configure' enable building these extra libraries. The shared library is
3407 built by default when using both ELF and GNU ld. When shared libraries
3408 are enabled, the new library `-ldl' is available for arbitrary run-time
3409 loading of shared objects; its interface is defined in <dlfcn.h>. The
3410 new header file <link.h> gives access to the internals of the run-time
3411 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3412 application of same name on other systems and it provides information
3413 about dynamically linked binaries.
3415 * The C library now provides the run-time support code for profiling
3416 executables compiled with `-pg'. Programs can control the profiling code
3417 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3418 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3419 the support for this file format was contributed by David Mosberger-Tang.
3421 * The math code has been replaced with a math library based on fdlibm from
3422 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3423 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3424 functions. The math functions now reside in a separate library, so
3425 programs using them will need to use `-lm' their linking commands.
3427 * John C. Bowman contributed optimized ix87 assembler inline functions.
3429 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3430 mechanism similar to that found in Solaris 2. This is now used for the
3431 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3432 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3433 file controls what services are used for each individual database. This
3434 works by loading shared libraries with names specified in `nsswitch.conf',
3435 so service modules can be changed or added at any time without even
3436 relinking any program. Currently there are the file, db, and NIS based
3437 NSS services available.
3439 * The new functions `strtoq' and `strtouq' parse integer values from
3440 strings, like `strtol' and `strtoul', but they return `long long int' and
3441 `unsigned long long int' values, respectively (64-bit quantities).
3443 * The new functions `strtof' and `strtold' parse floating-point values from
3444 strings, like `strtod', but they return `float' and `long double' values,
3445 respectively (on some machines `double' and `long double' are the same).
3447 * Ulrich Drepper has contributed new implementations of the floating-point
3448 printing and reading code used in the `printf' family of functions and
3449 `strtod', `strtof', and `strtold'. These new functions are perfectly
3450 accurate, and much faster than the old ones.
3452 * The implementation of the POSIX locale model was completely rewritten by
3453 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3454 to compile the POSIX locale definition.
3456 * The former dummy implementations of the strcoll and strxfrm function are
3457 now replaced by fully functional code contributed by Ulrich Drepper. The
3458 collation information comes from the POSIX locale definitions.
3460 * The new header <langinfo.h> defines an interface for accessing
3461 various locale-dependent data (using the locale chosen with `setlocale').
3463 * Ulrich Drepper has contributed a new suite of functions for operation on
3464 wide-character and multibyte-character strings, in <wchar.h>;
3465 and classification and case conversion of wide characters, in <wctype.h>.
3466 These new functions are conforming to the ISO C, Amendement 1 specification.
3468 * There is now a second implementation of the standard I/O library available.
3469 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3470 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3472 * You can now use positional parameter specifications in format strings
3473 for the `printf' and `scanf' families of functions. For example,
3474 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3475 ``Number 6, Mr Jones''. This is mainly useful when providing different
3476 format strings for different languages, whose grammars may dictate
3477 different orderings of the values being printed. To support this
3478 feature, the interface for `register_printf_handler' has changed; see
3479 the header file <printf.h> for details.
3481 * The `printf' and `scanf' families of functions now understand a new
3482 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3483 to group numbers as indicated by the locale; for `scanf' and friends, this
3484 says to accept as valid only a number with all the proper grouping
3485 separators in the right places. In the default "C" locale, numbers are
3486 not grouped; but locales for specific countries will define the usual
3487 conventions (i.e. separate thousands with `,' in the US locale).
3489 * The pgrp functions have been regularized, slightly incompatibly but much
3490 less confusingly. The core functions are now `getpgid' and `setpgid',
3491 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3492 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3493 provided for compatibility. There is no longer an incompatible `getpgrp'
3494 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3496 * The new header file <fts.h> and suite of functions simplify programs that
3497 operate on directory trees. This code comes from 4.4 BSD.
3499 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3500 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3501 NSS scheme used in glibc.
3503 * There is a new malloc debugging hook `__memalign_hook'.
3505 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3506 `unsigned int' in <sys/types.h>. These are for compatibility only and
3507 their use is discouraged.
3509 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3510 done differently, so that it works even without GNU ld.
3512 * New function `euidaccess' checks allowed access to a file like `access',
3513 but using the effective IDs instead of the real IDs.
3515 * The time zone data files have been updated for the latest and greatest
3516 local time conventions of the countries of the world.
3518 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3521 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3522 for formatting floating-point numbers. They are provided only for
3523 compatibility; new programs should use `sprintf' instead. There are
3524 also equivalent function for the `long double' floating-point type and
3525 all functions also exist in a reentrant form.
3527 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3528 functions for maintaining the login-record files (primarily of use to
3529 system programs such as `login'), and convenient functions for
3530 allocating and initializing a pseudo-terminal (pty) device.
3532 * Ulrich Drepper has contributed new support for System V style
3533 shared memory and IPC on systems that support it.
3535 * Ulrich Drepper has contributed several miscellaneous new functions found
3536 in System V: The `hsearch' family of functions provide an effective
3537 implementation of hash tables; `a64l' and `l64a' provide a very simple
3538 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3541 * Ulrich Drepper has contributed new reentrant counterparts for the
3542 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3544 * Ulrich Drepper has contributed new, highly-optimized versions of several
3545 string functions for the i486/Pentium family of processors.
3547 * Ulrich Drepper has updated the Linux-specific code, based largely
3548 on work done in Hongjiu Lu's version of GNU libc for Linux.
3549 The GNU library now supports Linux versions 2.0.10 and later,
3550 using the ELF object file format (i[3456]86-*-linux).
3552 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3554 * David Mosberger-Tang and Richard Henderson have ported the C library
3555 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3556 dynamic linking support for ELF/Alpha.
3558 * Richard Henderson contributed several Alpha optimized assembler function
3559 for arithmetic and string handling.
3561 * Ulrich Drepper has contributed a new set of message catalog functions to
3562 support multiple languages using the <libintl.h> interface, for use with
3563 his new package GNU gettext. Translation volunteers have contributed
3564 catalogs of the library's messages in Spanish, German, and Korean.
3566 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3567 program and the `catgets' function for reading the catalog files it
3568 creates. (The <libintl.h> interface is preferred; we include the
3569 <nl_types.h> interface using `catgets' only for source compatibility with
3570 programs already written to use it.)
3572 * New header file <values.h> gives SVID-compatible names for <limits.h>
3575 * Various new macros, declarations, and small header files for compatibility
3578 * New function `group_member' is a convenient way to check if a process has
3579 a given effective group ID.
3581 * When using GCC 2.7 and later, the socket functions are now declared in a
3582 special way so that passing an argument of type `struct sockaddr_in *',
3583 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3584 `struct sockaddr *' type, does not generate a type-clash warning.
3586 * New function `error' declared in header file <error.h> is a convenient
3587 function for printing error messages and optionally exiting; this is the
3588 canonical function used in GNU programs. The new functions `err', `warn',
3589 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3590 doing the same thing.
3592 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3593 POSIX.2 `glob' function to do ~ and {...} expansion.
3595 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3596 `clearenv' which is used in POSIX.9 is also available.
3598 * New function `getsid' returns session ID number on systems that support it.
3600 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3601 files <db.h> and <mpool.h> provide a rich set of functions for several
3602 types of simple databases stored in memory and in files, and <ndbm.h> is
3603 an old `ndbm'-compatible interface using the `db' functions. Link with
3604 `-ldb' to get these functions.
3606 * New macro `strdupa' copies a string like `strdup', but uses local stack
3607 space from `alloca' instead of dynamic heap space from `malloc'.
3609 * New function `strnlen' is like `strlen' but searches only a given maximum
3610 number of characters for the null terminator. `stpncpy', `strndup' and
3611 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3614 * New function `statfs' in header <sys/statfs.h>.
3616 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3617 provide convenient functions for operating on blocks of null-terminated
3620 * A new suite of functions in <utmp.h> handle all the details of reading
3621 and writing the utmp file.
3623 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3626 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3627 completely POSIX compliant and also implemented the extended functionality
3628 to handle alternate digit representation and alternate era date formats.
3630 * Ulrich Drepper provided an implementation of the `strptime' function
3631 defined in XPG4.2 which transforms a string into a `struct tm' value.
3633 * Paul Eggert provided the tzselect shell script as part of the timezone
3634 code. The shell script makes it easy to select the correct timezone
3637 * The implementation of the malloc family of functions is completely replaced
3638 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3639 The implementation uses the mmap function (if available) and it is
3640 optimized for the use in multi threaded programs.
3642 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3643 functions. This new functionality is usable by specifying a special
3644 salt string and it is compatible with implementation on *BSD systems.
3646 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3647 `getsubopt' to handle second level command line options, `bsd_signal'
3648 to access BSD style `signal' functionality, the obsolete `regexp' style
3651 * the `lchown' function is available on system which support this
3654 * The implementation of the shadow password handling function was contributed
3657 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3659 * POSIX.1g support was added. The <sys/select.h> header is available,
3660 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3661 implementation of `getaddrinfo'.
3665 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3667 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3668 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3670 * The new function `daemon' from 4.4 BSD is useful for server programs that
3671 want to put themselves in the background.
3673 * Joel Sherrill has contributed support for several standalone boards that
3674 run without an operating system.
3676 * `printf', `scanf' and friends now accept a `q' type modifier for long
3677 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3679 * All of the code taken from BSD (notably most of the math and networking
3680 routines) has been updated from the BSD 4.4-Lite release.
3682 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3684 * The new functions `getdomainname' and `setdomainname' fetch or change the
3685 YP/NIS domain name. These are system calls which exist on systems which
3688 * The time zone data files have been updated for the latest international
3691 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3692 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3696 * The C library now includes support for Sun RPC, from Sun's free
3697 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3698 are included. (There is still no support for YP.)
3700 * Tom Quinn has contributed a port of the C library to SGI machines running
3701 Irix 4 (mips-sgi-irix4).
3703 * The new `lockf' function is a simplified interface to the locking
3704 facilities of `fcntl', included for compatibility.
3706 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3708 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3710 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3713 * The error message printed by `assert' for a failed assertion now includes
3714 the name of the program (if using GNU ld) and the name of the calling
3715 function (with versions of GCC that support this).
3717 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3719 * The library now includes the <sys/mman.h> header file and memory
3720 management functions `mmap', `munmap', `mprotect', `msync', and
3721 `madvise', on systems that support those facilities.
3723 * The interface for `mcheck' has changed slightly: the function called to
3724 abort the program when an allocation inconsistency is detected now takes
3725 an argument that indicates the type of failure. The new function
3726 `mprobe' lets you request a consistency check for a particular block at
3727 any time (checks are normally done only when you call `free' or `realloc'
3730 * It is now possible to easily cross-compile the C library, building on one
3731 system a library to run on another machine and/or operating system. All
3732 you need to do is set the variable `HOST_CC' in `configparms' to the
3733 native compiler for programs to run on the machine you are building on (a
3734 few generator programs are used on Unix systems); set `CC' to the
3737 * The new function `fexecve' (only implemented on the GNU system) executes
3738 a program file given a file descriptor already open on the file.
3742 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3743 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3745 * You can set the variable `libprefix' in `configparms' to specify a prefix
3746 to be prepended to installed library files; this makes it easy to install
3747 the GNU C library to be linked as `-lgnuc' or whatever.
3749 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3750 copies a limited number of characters from a string, and returns the
3751 address of the last character written.
3753 * You no longer need to check for whether the installed `stddef.h' is
3754 compatible with the GNU C library. configure now checks for you.
3756 * You can now define a per-stream `fileno' function to convert the
3757 stream's cookie into an integral file descriptor.
3759 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3760 allocates zero bytes of storage, and returns a unique pointer which
3761 you can pass to `realloc' or `free'. The behavior is undefined if
3762 you dereference this pointer.
3764 * The C library now runs on Sony NEWS m68k machines running either
3765 NewsOS 3 or NewsOS 4.
3767 * The new `syscall' function is a system-dependent primitive function
3768 for invoking system calls. It has the canonical behavior on Unix
3769 systems, including unreliable return values for some calls (such as
3770 `pipe', `fork' and `getppid').
3772 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3773 to `EAGAIN', which is the preferred name. On systems whose kernels
3774 use two distinct codes, the C library now translates EWOULDBLOCK to
3775 EAGAIN in every system call function.
3779 * The GNU C Library Reference Manual is now distributed with the library.
3780 `make dvi' will produce a DVI file of the printed manual.
3781 `make info' will produce Info files that you can read on line using C-h i
3782 in Emacs or the `info' program.
3783 Please send comments on the manual to bug-glibc-manual@gnu.org.
3785 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3787 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3789 * Jason Merrill has contributed a port to the Sequent Symmetry running
3790 Dynix version 3 (i386-sequent-dynix).
3792 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3793 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3795 * New function `memory_warnings' lets you arrange to get warnings when
3796 malloc is running out of memory to allocate, like Emacs gives you.
3798 * The C library now contains the relocating allocator used in Emacs 19 for
3799 its editing buffers. This allocator (ralloc) minimizes allocation
3800 overhead and fragmentation by moving allocated regions around whenever it
3801 needs to. You always refer to a ralloc'd region with a "handle" (a
3802 pointer to a pointer--an object of type `void **').
3804 * There is a new `printf' format: `%m' gives you the string corresponding
3805 to the error code in `errno'.
3807 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3808 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3809 pass, the `a' modifier says to fill in a `char **' you pass with a
3812 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3813 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3814 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3816 * `mkstemp' is a traditional Unix function to atomically create and open a
3817 uniquely-named temporary file.
3821 * The standard location for the file that says what the local timezone is
3822 has changed again. It is now `/usr/local/etc/localtime' (or more
3823 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3825 * The distribution no longer contains any files with names longer than 14
3828 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3829 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3831 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3835 * The configuration process has changed quite a bit. The `configure'
3836 script is now used just like the configuration scripts for other GNU
3837 packages. The `sysdeps' directory hierarchy is much rearranged.
3838 The file `INSTALL' explains the new scheme in detail.
3840 * The header files no longer need to be processed into ANSI C and
3841 traditional C versions. There is just one set of files to install, and
3842 it will work with ANSI or old C compilers (including `gcc -traditional').
3844 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3845 MIPS DECStation running Ultrix 4.
3847 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3848 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3849 made itself into a shared library.
3851 * Yet further improved support for the i386, running 4.3 BSD-like systems
3852 (such as Mach 3 with the Unix single-server), or System V.
3854 * New function `strncasecmp' to do case-insensitive string comparison
3855 with limited length.
3857 * New function `strsep' is a reentrant alternative to `strtok'.
3859 * New functions `scandir' and `alphasort' for searching directories.
3861 * New function `setenv' is a better interface to `putenv'.
3863 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3864 function for traversing a directory tree.
3866 * The GNU obstack package is now also part of the C library.
3867 The new function `open_obstack_stream' creates a stdio stream that
3868 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3869 formatted output directly to an obstack.
3871 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3872 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3874 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3876 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3877 things to your strings.
3879 * There are some new test programs: test-fseek, testmb, and testrand.
3881 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3882 These ports are not finished, but are a good starting place for really
3883 supporting those systems.
3885 * `/etc/localtime' is now the standard location for the file that says what
3886 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3887 This follows the general principle that `/etc' is the place for all local
3888 configuration files.
3890 * The C library header files now use `extern "C"' when used by the C++
3891 compiler, so the C library should now work with C++ code.
3893 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3894 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3897 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3898 function which uses the `merge sort' algorithm, and is said to be
3899 significantly faster than the old GNU `qsort' function. Merge sort is
3900 now the standard `qsort' function. The new algorithm can require a lot
3901 of temporary storage; so, the old sorting function is called when the
3902 required storage is not available.
3904 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3905 provides the Unix `crypt' function, plus some other entry points.
3907 * The code and header files taken from 4.4 BSD have been updated with the
3908 latest files released from Berkeley.
3910 ----------------------------------------------------------------------
3911 Copyright information:
3913 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3915 Permission is granted to anyone to make or distribute verbatim copies
3916 of this document as received, in any medium, provided that the
3917 copyright notice and this permission notice are preserved,
3918 thus giving the recipient permission to redistribute in turn.
3920 Permission is granted to distribute modified versions
3921 of this document, or of portions of it,
3922 under the above conditions, provided also that they
3923 carry prominent notices stating who last changed them.
3926 version-control: never