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.
10 * Unicode 9.0.0 Support: Character encoding, character type info, and
11 transliteration tables are all updated to Unicode 9.0.0, using
12 generator scripts contributed by Mike FABIAN (Red Hat).
14 * errno.h is now safe to use from C-preprocessed assembly language on all
15 supported operating systems. In this context, it will only define the
16 Exxxx constants, as preprocessor macros expanding to integer literals.
18 * The rpcgen, librpcsvc and related headers will only be built and
19 installed when glibc is configured with --enable-obsolete-rpc.
20 This allows alternative RPC implementations, like TIRPC, to be used
21 by default. Applications needing features missing from TIRPC should
22 consider the rpcsvc-proto project developed by Thorsten Kukuk (SUSE).
24 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
25 libnss_compat, are deprecated, and will not be built or installed by
26 default. Replacement implementations based on TIRPC, which
27 additionally support IPv6, are available from
28 <https://github.com/thkukuk/libnss_{compat,nis,nisplus}>.
30 * The NIS(+) support library, libnsl, is deprecated. By default, a
31 compatibility shared library will be built and installed, but not
32 headers or development libraries.
34 Only a few NIS-related programs require this library.
35 A replacement implementation based on TIRPC is available from
36 <https://github.com/thkukuk/libnsl>. Like the replacement NIS(+)
37 name service modules, the replacement supports IPv6, and it can be
38 coinstalled with the compatibility shared library from glibc.
40 * New configure option --enable-obsolete-nsl will cause libnsl's
41 headers, and the NIS(+) name service modules, to be built and
42 installed. This option may be removed in a future release.
44 * Extensive new collation tests for Hungarian locales based on
45 "The Rules of Hungarian Orthography, 12th edition" and the work of
46 Egmont Koblinger (Bug 18934).
48 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
49 support is enabled, the configured recursive resolver must support EDNS.
50 (Responding to EDNS-enabled queries with responses which are not
51 EDNS-enabled is fine, but FORMERR responses are not.)
53 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
54 servers have not supported this opcode for a long time.
56 * The legacy cfree function has been removed. Applications should use the
57 free function instead.
59 * posix_spawnattr_setflags now supports POSIX_SPAWN_SETSID flag to create a
60 new session ID for the posix_spawn and posix_spawnp. It is scheduled to
61 be added on next major revision of POSIX, so current support is enabled
64 * The minimum Linux kernel version that this version of the GNU C Library
65 can be used with on i[4567]86 and x86_64 is 3.2. A Linux 3.2 or later
66 kernel was already required on all other architectures.
68 * The obsolete <sys/ultrasound.h> header file has been removed.
70 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
73 * The reallocarray function has been added to libc. It is a realloc
74 replacement with a check for integer overflow when calculating total
77 * New preadv2 and pwritev2 has been added. They are Linux extensions to
78 preadv and pwritev with an additional flag argument where it is possible
79 to set high priority or use O_DSYNC or O_SYNC for a specific IO operation.
80 For complete support it requires Linux kernel version 4.6, otherwise a
81 compat implementation will be used (which refuses all flags and routes it
82 to preadv or pwritev).
84 * The stack_t type no longer has the name struct sigaltstack. This changes
85 the C++ name mangling for interfaces involving this type.
87 * The synchronization that pthread_spin_unlock performs has been changed
88 to now be equivalent to a C11 atomic store with release memory order to
89 the spin lock's memory location. This ensures correct synchronization
90 for the spin lock's operations and critical sections protected by a spin
91 lock. Previously, several (but not all) architectures used stronger
92 synchronization (e.g., containing what is often called a full barrier).
93 This change can improve performance, but may affect odd fringe uses of
94 spin locks that depend on the previous behavior (e.g., using spin locks
95 as atomic variables to try to implement Dekker's mutual exclusion
98 * The tunables feature is now enabled by default. This allows users to tweak
99 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
101 * The s390 specific ptrace requests are adjusted to the kernel ones. Request 12
102 is now used for PTRACE_SINGLEBLOCK instead of PTRACE_GETREGS. The requests
103 PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS were
104 removed as those are not supported by the s390 kernel. The requests
105 PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
106 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
107 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND were added as those are supported
110 * The minimum GCC version that can be used to build this version of the GNU
111 C Library is GCC 4.9. Older GCC versions, and non-GNU compilers, can
112 still be used to compile programs using the GNU C Library.
114 Security related changes:
116 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
117 to avoid fragmentation-based spoofing attacks.
119 The following bugs are resolved with this release:
121 [The release manager will add the list generated by
122 scripts/list-fixed-bugs.py just before the release.]
127 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
128 24731-2:2010, is supported to enable declarations of functions from that
129 TR. Note that not all functions from that TR are supported by the GNU C
132 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
133 18661-1:2014, is supported to enable declarations of functions and macros
134 from that TS. Note that not all features from that TS are supported by
137 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
138 18661-4:2015, is supported to enable declarations of functions and macros
139 from that TS. Note that most features from that TS are not supported by
142 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
143 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
144 Since the GNU C Library defaults to a much newer revision of POSIX, this
145 will only affect programs that specifically request an old conformance
146 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
147 see a change in the visible declarations, but a program compiled with
148 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
151 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
152 defined by all multithreaded code, but glibc has not required this for
155 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
156 means that in a future release, the macros “major”, “minor”, and “makedev”
157 will only be available from <sys/sysmacros.h>.
159 These macros are not part of POSIX nor XSI, and their names frequently
160 collide with user code; see for instance glibc bug 19239 and Red Hat bug
161 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
162 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
165 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
166 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
167 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
169 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
170 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
171 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
172 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
173 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
174 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
175 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
176 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
177 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
178 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
179 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
182 * New <math.h> features are added from TS 18661-1:2014:
184 - Signaling NaN macros: SNANF, SNAN, SNANL.
186 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
187 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
188 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
190 - llogb functions: the llogb, llogbf and llogbl functions, and the
191 FP_LLOGB0 and FP_LLOGBNAN macros.
193 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
196 - Comparison macros: iseqsig.
198 - Classification macros: iscanonical, issubnormal, iszero.
200 - Total order functions: totalorder, totalorderf, totalorderl,
201 totalordermag, totalordermagf, totalordermagl.
203 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
205 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
206 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
208 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
209 are added to libc. They convert a floating-point number into string.
211 * Most of glibc can now be built with the stack smashing protector enabled.
212 It is recommended to build glibc with --enable-stack-protector=strong.
213 Implemented by Nick Alcock (Oracle).
215 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
216 intended to be used instead of memset() to erase sensitive data after use;
217 the compiler will not optimize out calls to explicit_bzero even if they
218 are "unnecessary" (in the sense that no _correct_ program can observe the
219 effects of the memory clear).
221 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
222 to float instead of double. This does not affect the ABI of any libraries
223 that are part of the GNU C Library, but may affect the ABI of other
224 libraries that use this type in their interfaces.
226 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
227 float_t and double_t types are now defined to long double instead of float
228 and double. These options are not the default, and this does not affect
229 the ABI of any libraries that are part of the GNU C Library, but it may
230 affect the ABI of other libraries that use this type in their interfaces,
231 if they are compiled or used with those options.
233 * The getentropy and getrandom functions, and the <sys/random.h> header file
236 * The buffer size for byte-oriented stdio streams is now limited to 8192
237 bytes by default. Previously, on Linux, the default buffer size on most
238 file systems was 4096 bytes (and thus remains unchanged), except on
239 network file systems, where the buffer size was unpredictable and could be
240 as large as several megabytes.
242 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
243 for the Linux quota interface which predates kernel version 2.4.22 has
246 * The malloc_get_state and malloc_set_state functions have been removed.
247 Already-existing binaries that dynamically link to these functions will
248 get a hidden implementation in which malloc_get_state is a stub. As far
249 as we know, these functions are used only by GNU Emacs and this change
250 will not adversely affect already-built Emacs executables. Any undumped
251 Emacs executables, which normally exist only during an Emacs build, should
252 be rebuilt by re-running “./configure; make” in the Emacs build tree.
254 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
255 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
256 “no-ip6-dotint” had already been the default, and support for the
257 “ip6-dotint” option was removed from the Internet in 2006.
259 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
260 flag from <resolv.h> have been removed. The option relied on a
261 backwards-incompatible DNS extension which was never deployed on the
264 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
265 RES_BLAST defined in the <resolv.h> header file have been deprecated.
266 They were already unimplemented.
268 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
269 _res.flags are deprecated. The flag was standardized in RFC 2133, but
270 removed again from the IETF name lookup interface specification in RFC
271 2553. Applications should use getaddrinfo instead.
273 * DNSSEC-related declarations and definitions have been removed from the
274 <arpa/nameser.h> header file, and libresolv will no longer attempt to
275 decode the data part of DNSSEC record types. Previous versions of glibc
276 only implemented minimal support for the previous version of DNSSEC, which
277 is incompatible with the currently deployed version.
279 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
280 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
281 <arpa/nameser.h> header file because the distinction between RR types and
282 meta-RR types is not officially standardized, subject to revision, and
283 thus not suitable for encoding in a macro.
285 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
286 and rhook members of the res_state type in <resolv.h> have been removed.
287 The glibc stub resolver did not support these hooks, but the header file
288 did not reflect that.
290 * For multi-arch support it is recommended to use a GCC which has
291 been built with support for GNU indirect functions. This ensures
292 that correct debugging information is generated for functions
293 selected by IFUNC resolvers. This support can either be enabled by
294 configuring GCC with '--enable-gnu-indirect-function', or by
295 enabling it by default by setting 'default_gnu_indirect_function'
296 variable for a particular architecture in the GCC source file
299 * GDB pretty printers have been added for mutex and condition variable
300 structures in POSIX Threads. When installed and loaded in gdb these pretty
301 printers show various pthread variables in human-readable form when read
302 using the 'print' or 'display' commands in gdb.
304 * Tunables feature added to allow tweaking of the runtime for an application
305 program. This feature can be enabled with the '--enable-tunables' configure
306 flag. The GNU C Library manual has details on usage and README.tunables has
307 instructions on adding new tunables to the library.
309 * A new version of condition variables functions have been implemented in
310 the NPTL implementation of POSIX Threads to provide stronger ordering
313 * A new version of pthread_rwlock functions have been implemented to use a more
314 scalable algorithm primarily through not using a critical section anymore to
317 Security related changes:
319 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
320 have been created with makecontext could fail to terminate due to a
321 missing .cantunwind annotation. This has been observed to lead to a hang
322 (denial of service) in some Go applications compiled with gccgo. Reported
323 by Andreas Schwab. (CVE-2016-6323)
325 * The DNS stub resolver functions would crash due to a NULL pointer
326 dereference when processing a query with a valid DNS question type which
327 was used internally in the implementation. The stub resolver now uses a
328 question type which is outside the range of valid question type values.
331 The following bugs are resolved with this release:
333 [4099] stdio: Overly agressive caching by stream i/o functions.
334 [7065] build: Support building glibc with -fstack-protector or -fstack-
336 [9842] localedata: en_CA: incorrect date format
337 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
338 before it started waiting
339 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
340 [15765] nptl: sem_open is wrongly a cancellation point
341 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
342 [16458] libc: endian.h and netinet/in.h byte order macros return values of
344 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
345 library linked with pthread
346 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
348 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
349 [17252] libc: getrandom and getentropy syscall
350 [17730] malloc: thread-local storage is sometimes improperly free()'d
351 after being __libc_memalign()'d
352 [18241] stdio: failed fseek on memstream does not set errno and can fail
354 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
356 [18463] nptl: pthread_cond_broadcast issue when surrounded by
357 PTHREAD_PRIO_INHERIT mutex on ARM
358 [18784] network: res_query and related function crash for special record
359 type queries (CVE-2015-5180)
360 [19380] math: strtod does not raise "inexact"
361 [19387] string: Integer overflow in memchr
362 [19390] string: Integer overflow in strncat
363 [19398] build: linknamespace tests fail with massively parallel build
364 [19402] nptl: Deadlock with robust shared mutex and asynchronous
366 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
367 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
368 [19514] libc: [PATCH] Fix spelling errors in spelling
369 "implement"/"implementation" in several places
370 [19582] network: Deprecate RES_USE_INET6
371 [19673] manual: clog10 docs appear to be erroneous
372 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
373 causes a segmentation fault
374 [19826] libc: invalid pointer returned from __tls_get_addr with static
376 [20016] network: resolv: Remove hooks support from the API
377 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
379 [20033] math: [x86_64] vectorized math function don't call the __finite
381 [20116] nptl: use after free in pthread_create
382 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
383 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
385 [20311] nptl: please install proc_service.h
386 [20366] libc: Compilation errors in installed headers in strict-compliance
388 [20370] malloc: malloc: Arena free list management is still racy
389 (incorrect fix in bug 19243)
390 [20386] libc: assert (X = 0) does not result in GCC warning
391 [20432] malloc: malloc: Minimize interface required for interposition
392 [20435] libc: Missing unwind info in __startcontext causes infinite loop
393 in _Unwind_Backtrace (CVE-2016-6323)
394 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
395 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
397 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
398 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
399 [20477] network: resolv: incorrect double-checked locking related to
401 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
403 [20495] math: x86_64 performance degradation due to AVX/SSE transition
405 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
406 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
407 cause transition penalty
408 [20517] math: sparcv9 missing fdiml compat symbol
409 [20524] manual: strverscmp is inconsistent
410 [20525] libc: <sys/quota.h> should be based on kernel headers
411 [20539] math: GCC 7 -static -lm fails to link at -O3
412 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
414 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
416 [20591] network: Remove obsolete DNSSEC support
417 [20592] network: DNS resource record type classification macros in
418 <arpa/nameser.h> are incorrect
419 [20593] network: Update DNS RR type definitions
420 [20611] network: getaddrinfo accepts invalid numeric scope IDs
421 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
423 [20629] network: libresolv: Remove support for bitstring labels
425 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
426 assertion: "maxidx >= 2"
427 [20660] math: [arm] Use VSQRT
428 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
429 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
430 [20689] libc: Test for FMA should also check for AVX.
431 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
432 [20715] math: iszero macro breaks existing code
433 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
434 [20728] libc: powerpc: Missing TOC stub in clone
435 [20729] build: glibc-2.24 fails to build for i486 with -Os
436 [20750] build: Build fails with default PIE enabled gcc-6.x
437 [20768] math: [alpha] sqrt fegetenv namespace
438 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
440 [20787] math: float_t is defined as float incorrectly on x86_64 even if
441 __FLT_EVAL_METHOD__ is 2
442 [20790] network: rpcgen buffer overrun in get_prog_declaration
443 [20822] nptl: race condition in __lll_unlock_elision on powerpc
444 [20829] libc: crypt snprintf namespace
445 [20847] libc: tst-vfork3 failure
446 [20855] math: Default bits/mathdef.h has inappropriate float_t
447 [20859] math: [sh4] FP_ILOGB0 invalid
448 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
449 U+20AC), not same as GBK
450 [20915] dynamic-link: global-dynamic TLS broken on aarch64
451 [20916] math: pow handling of sNaN arguments
452 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
453 [20919] math: Bad pow (qNaN, 0) result with -lieee
454 [20940] math: hypot sNaN handling
455 [20947] math: fmax, fmin sNaN handling
456 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
457 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
459 [20971] string: powerpc64/power7 memchr overflows internal pointer check
460 [20973] nptl: robust mutexes: Lost wake-ups
461 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
462 [20978] nis: strlen on null pointer in nss_nisplus
463 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
465 [21014] string: i686 memchr overflows internal pointer check
466 [21019] libc: [mips] n32 lseek incorrect on overflow
467 [21022] libc: [microblaze] __backtrace get_frame_size namespace
468 [21026] libc: [MIPS] readahead syscall is broken on n64
469 [21028] math: Fallback fesetexceptflag should always succeed
470 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
471 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
473 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
474 [21061] librt: [microblaze] librt lost clock_* exports
475 [21073] libc: tunables: insecure environment variables passed to
476 subprocesses with AT_SECURE
477 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
481 * The minimum Linux kernel version that this version of the GNU C Library
482 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
483 kernel version 2.6.32 or later suffices (on architectures that already
484 required kernel versions more recent than 3.2, those requirements remain
485 unchanged). Linux 3.2 or later kernel headers are required on all
488 * The pap_AN locale has been deleted. This has been deprecated for a long
489 time. It has been replaced by pap_AW & pap_CW, both of which have long
490 been included in previous releases.
492 * The readdir_r and readdir64_r functions have been deprecated. It is
493 recommended to use readdir and readdir64 instead.
495 * The type “union wait” has been removed. It was deprecated in the early
496 1990s and never part of POSIX. Application code should use the int type
497 instead of “union wait”.
499 * A new NSS action is added to facilitate large distributed system
500 administration. The action, MERGE, allows remote user stores like LDAP
501 to be merged into local user stores like /etc/groups in order to provide
502 easy to use, updated, and managed sets of merged credentials. The new
503 action can be used by configuring it in /etc/nsswitch.conf:
504 group: files [SUCCESS=merge] nis
505 Implemented by Stephen Gallagher (Red Hat).
507 * The deprecated __malloc_initialize_hook variable has been removed from the
510 * The long unused localedef --old-style option has been removed. It hasn't
511 done anything in over 16 years. Scripts using this option can safely
514 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
515 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
516 return the next representable value in the direction of positive infinity
517 and the nextdown functions return the next representable value in the
518 direction of negative infinity. These are currently enabled as GNU
521 Security related changes:
523 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
524 could result in a stack overflow when getnetbyname was called with an
525 overly long name. (CVE-2016-3075)
527 * Previously, getaddrinfo copied large amounts of address data to the stack,
528 even after the fix for CVE-2013-4458 has been applied, potentially
529 resulting in a stack overflow. getaddrinfo now uses a heap allocation
530 instead. Reported by Michael Petlan. (CVE-2016-3706)
532 * The glob function suffered from a stack-based buffer overflow when it was
533 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
534 Reported by Alexander Cherepanov. (CVE-2016-1234)
536 * The Sun RPC UDP client could exhaust all available stack space when
537 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
538 alloca plugin for GCC. (CVE-2016-4429)
540 * The IPv6 name server management code in libresolv could result in a memory
541 leak for each thread which is created, performs a failing naming lookup,
542 and exits. Over time, this could result in a denial of service due to
543 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
545 The following bugs are resolved with this release:
547 [1170] localedata: ne_NP: update Nepali locale definition file
548 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
550 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
551 [6796] math: fdim() does not set errno on overflow
552 [10354] libc: posix_spawn should use vfork() in more cases than presently
553 [11213] localedata: localedata: add copyright disclaimer to locale files
554 [12143] localedata: chr_US: new Cherokee locale
555 [12450] localedata: sgs_LT: new locale
556 [12676] localedata: ln_CD: new locale
557 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
559 [13304] math: fma, fmaf, fmal produce wrong results
560 [14259] build: --localedir arg to configure is ignored
561 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
562 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
563 [14934] localedata: es_CL: wrong first weekday chilean locale
564 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
566 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
568 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
569 [15368] nptl: raise() is not async-signal-safe
570 [15479] math: ceil, floor, round and trunc raise inexact exception
571 [15578] localedata: kk_KZ: various updates
572 [16003] localedata: pap_AN: punt old locale
573 [16137] localedata: iw_IL: punt old locale
574 [16190] localedata: eo: new esperanto locale
575 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
576 [16742] malloc: race condition: pthread_atfork() called before first
577 malloc() results in unexpected locking behaviour/deadlocks
578 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
580 [16983] localedata: postal_fmt does not allow %l and %n modifiers
581 [17565] localedata: pt_PT: wrong (work-)week start
582 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
583 [17950] build: Build fails with -msse
584 [18205] localedata: be_BY*: wrong first_weekday and first_workday
585 [18433] libc: posix_spawn does not return correctly upon failure to
587 [18453] localedata: charmaps/IBM875: incorrect codes
588 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
590 [18896] localedata: he_IL: improvements for currency
591 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
593 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
595 [19056] libc: Deprecate readdir_r
596 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
598 [19198] localedata: nl_NL: small improvements for Dutch locales
599 [19257] network: Per-thread memory leak in __res_vinit with IPv6
600 nameservers (CVE-2016-5417)
601 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
603 [19400] locale: Language missing in "iso-639.def", trivial fix in
605 [19431] malloc: Deadlock between fflush, getdelim, and fork
606 [19505] libc: Incorrect file descriptor validity checks in
607 posix_spawn_file_actions_add{open,close,dup2}
608 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
610 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
611 `intl/{gettextP,loadinfo}.h'
612 [19534] libc: execle, execlp may use malloc
613 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
614 [19573] network: res_nclose and __res_maybe_init disagree about name
615 server initialization, breaking Hesiod
616 [19575] localedata: Status of GB18030 tables
617 [19581] localedata: sr_* date_fmt string contains additional newline
618 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
620 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
621 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
622 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
623 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
625 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
627 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
629 [19610] dynamic-link: ldconfig -X removes stale symbolic links
630 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
631 [19633] locale: strfmon_l applies global locale to number formatting
632 [19642] network: Memory leak in getnameinfo
633 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
634 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
636 [19654] math: [x86_64] Need testcase for BZ #19590 fix
637 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
639 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
640 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
642 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
644 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
646 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
647 pointers and lengths in error-case.
648 [19727] locale: Converting from/to UTF-xx with iconv() does not always
649 report errors on UTF-16 surrogates values.
650 [19755] nscd: nscd assertion failure in gc
651 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
652 [19759] libc: mempcpy shouldn't be inlined
653 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
654 [19765] libc: s390 needs an optimized mempcpy
655 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
656 NAME_MAX limit assumption (CVE-2016-1234)
657 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
658 [19787] network: Missing and incorrect truncation checks in getnameinfo
659 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
661 [19791] network: Assertion failure in res_query.c with un-connectable name
663 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
664 [19822] math: libm.so install clobbers old version
665 [19825] network: resolv: send_vc can return uninitialized data in second
666 response to getaddrinfo
667 [19830] network: nss_dns: should check RDATA length against buffer length
668 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
669 confronted with A/AAAA records of invalid size
670 [19837] nss: nss_db: No retries for some long lines with a larger buffer
671 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
672 [19853] stdio: Printing IBM long double in decimal with high precision is
674 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
675 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
676 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
678 [19865] network: Assertion failure or memory leak in
679 _nss_dns_getcanonname_r
680 [19868] network: nss_dns: netent code does not skip over non-PTR records
681 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
683 [19881] string: Improve x86-64 memset
684 [19907] string: Incorrect memcpy tests
685 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
687 [19925] libc: termios.h XCASE namespace
688 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
689 [19929] libc: limits.h NL_NMAX namespace
690 [19931] stdio: Memory leak in vfprintf
691 [19957] libc: clone(CLONE_VM) access invalid parent memory
692 [19963] localedata: en_IL: New locale
693 [19989] stdio: stdio.h cuserid namespace
694 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
696 [19996] locale: langinfo.h nl_langinfo_l namespace
697 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
698 [20010] network: getaddrinfo: Stack overflow in hostent translation
700 [20012] stdio: libio: fmemopen append mode failure
701 [20014] stdio: stdio.h namespace for pre-threads POSIX
702 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
703 [20023] libc: fcntl.h timespec namespace
704 [20024] math: [x86_64] vectorized sincos trashes the stack
705 [20031] network: nss_hesiod: Heap overflow in get_txt_records
706 [20041] time: sys/time.h timespec namespace
707 [20043] libc: unistd.h missing cuserid for UNIX98 and before
708 [20044] libc: unistd.h missing pthread_atfork for UNIX98
709 [20051] libc: ttyslot in wrong header under wrong conditions
710 [20054] libc: gethostname not declared for XPG4
711 [20055] libc: termios.h missing tcgetsid for XPG4
712 [20072] dynamic-link: x86 init_cpu_features is called twice in static
714 [20073] libc: sys/stat.h fchmod namespace
715 [20074] libc: stdlib.h rand_r namespace
716 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
717 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
719 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
720 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
722 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
723 [20119] libc: Wrong mask for processors level type from CPUID
724 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
725 AS not supporting AVX512
726 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
728 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
729 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
731 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
732 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
733 [20178] libc: posix_spawn{p} should not call exit
734 [20191] stdio: libio: vtables hardening
735 [20195] string: FMA4 detection requires CPUID execution with register
737 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
738 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
740 [20212] math: acos (sNaN) returns sNaN
741 [20213] math: asin (sNaN) returns sNaN
742 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
743 [20218] math: [i386] asinhl (sNaN) returns sNaN
744 [20219] math: [i386] atanhl (sNaN) returns sNaN
745 [20222] stdio: fopencookie: Mangle function pointers
746 [20224] math: [i386] cbrtl (sNaN) returns sNaN
747 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
748 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
750 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
751 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
752 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
753 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
754 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
755 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
756 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
757 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
758 [20240] math: modf (sNaN) returns sNaN
759 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
760 [20250] math: frexp (sNaN) returns sNaN
761 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
762 [20255] math: [i386] fdim, fdimf return with excess range and precision /
764 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
765 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
766 subscript is below array bounds [-Werror=array-bounds]
767 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
768 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
770 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
771 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
773 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
775 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
776 [20316] localedata: id_ID: Februari instead of Pebruari
777 [20327] string: POWER8 strcasecmp returns incorrect result
778 [20347] math: Failure: Test: j0_downward (0xap+0)
779 [20348] libc: FAIL: misc/tst-preadvwritev64
780 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
781 [20350] libc: There is no test for p{read,write}64
782 [20357] math: Incorrect cos result for 1.5174239687223976
783 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
787 * Unicode 8.0.0 Support: Character encoding, character type info, and
788 transliteration tables are all updated to Unicode 8.0.0, using new
789 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
790 These updates cause user visible changes, such as the fixes for bugs
791 89, 16061, and 18568.
793 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
794 kernel-internal CPU set size. This means that requests that change the
795 CPU affinity which failed before (for example, an all-ones CPU mask) will
796 now succeed. Applications that need to determine the effective CPU
797 affinities need to call sched_getaffinity or pthread_getaffinity_np after
798 setting it because the kernel can adjust it (and the previous size check
799 would not detect this in the majority of cases).
801 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
802 the following new symbols are used: fts64_children, fts64_close,
803 fts64_open, fts64_read and fts64_set.
805 * getaddrinfo now detects certain invalid responses on an internal netlink
806 socket. If such responses are received, an affected process will
807 terminate with an error message of "Unexpected error <number> on netlink
808 descriptor <number>" or "Unexpected netlink response of size <number> on
809 descriptor <number>". The most likely cause for these errors is a
810 multi-threaded application which erroneously closes and reuses the netlink
811 file descriptor while it is used by getaddrinfo.
813 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
814 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
815 unnecessary serialization of memory allocation requests across threads.
816 The defect is now corrected. Users should see a substantial increase in
817 the concurent throughput of allocation requests for applications which
818 trigger this bug. Affected applications typically create and
819 destroy threads frequently. (Bug 19048 was reported and analyzed by
822 * There is now a --disable-timezone-tools configure option for disabling the
823 building and installing of the timezone related utilities (zic, zdump, and
824 tzselect). This is useful for people who build the timezone data and code
825 independent of the GNU C Library.
827 * The obsolete header <regexp.h> has been removed. Programs that require
828 this header must be updated to use <regex.h> instead.
830 * The obsolete functions bdflush, create_module, get_kernel_syms,
831 query_module and uselib are no longer available to newly linked binaries;
832 the header <sys/kdaemon.h> has been removed. These functions and header
833 were specific to systems using the Linux kernel and could not usefully be
834 used with the GNU C Library on systems with version 2.6 or later of the
837 * Optimized string, wcsmbs and memory functions for IBM z13.
838 Implemented by Stefan Liebler.
840 * Newly linked programs that define a variable called signgam will no longer
841 have it set by the lgamma, lgammaf and lgammal functions. Programs that
842 require signgam to be set by those functions must ensure that they use the
843 variable provided by the GNU C Library and declared in <math.h>, without
844 defining their own copy.
846 * The minimum GCC version that can be used to build this version of the GNU
847 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
848 still be used to compile programs using the GNU C Library.
850 Security related changes:
852 * An out-of-bounds value in a broken-out struct tm argument to strftime no
853 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
855 * The LD_POINTER_GUARD environment variable can no longer be used to disable
856 the pointer guard feature. It is always enabled. Previously,
857 LD_POINTER_GUARD could be used to disable security hardening in binaries
858 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
861 * An integer overflow in hcreate and hcreate_r could lead to an
862 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
864 * The catopen function no longer has unbounded stack usage. Reported by
867 * The nan, nanf and nanl functions no longer have unbounded stack usage
868 depending on the length of the string passed as an argument to the
869 functions. Reported by Joseph Myers. (CVE-2014-9761)
871 * A stack-based buffer overflow was found in libresolv when invoked from
872 libnss_dns, allowing specially crafted DNS responses to seize control
873 of execution flow in the DNS client. The buffer overflow occurs in
874 the functions send_dg (send datagram) and send_vc (send TCP) for the
875 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
876 family. The use of AF_UNSPEC triggers the low-level resolver code to
877 send out two parallel queries for A and AAAA. A mismanagement of the
878 buffers used for those queries could result in the response of a query
879 writing beyond the alloca allocated buffer created by
880 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
881 the overflow. Thanks to the Google Security Team and Red Hat for
882 reporting the security impact of this issue, and Robert Holiday of
883 Ciena for reporting the related bug 18665. (CVE-2015-7547)
885 The following bugs are resolved with this release:
887 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
888 [887] math: Math library function "logb" and "nextafter" inconsistent
889 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
890 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
891 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
892 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
894 [4404] localedata: German translation of "Alarm clock" is misleading
895 [6799] math: nextafter() and nexttoward() doen't set errno on
896 overflow/underflow errors
897 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
899 [10432] nis: _nss_nis_setnetgrent assertion failure
900 [11460] libc: fts has no LFS support
901 [12926] network: getaddrinfo()/make_request() may spin forever
902 [13065] nptl: Race condition in pthread barriers
903 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
904 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
906 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
907 [14912] libc: Rename non-installed bits/*.h headers
908 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
909 [15367] math: Let gcc use __builtin_isinf
910 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
911 [15421] math: lgamma wrongly sets signgam for ISO C
912 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
913 FE_INVALID with argument out of range
914 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
916 [15786] dynamic-link: ifunc resolver functions can smash function
918 [15918] math: Unnecessary check for equality in hypotf()
919 [16061] localedata: Review / update transliteration data
920 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
922 [16141] time: strptime %z offset restriction
923 [16171] math: drem should be alias of remainder
924 [16296] math: fegetround is pure?
925 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
926 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
928 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
929 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
931 [16422] math: [powerpc] math-float, math-double failing llrint tests with
932 "Exception "Inexact" set" on ppc32
933 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
934 [16517] math: Missing underflow exception from tanf/tan/tanl
935 [16519] math: Missing underflow exception from sinhf
936 [16520] math: Missing underflow exception from tanhf
937 [16521] math: Missing underflow exception from exp2
938 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
940 [16734] stdio: fopen calls mmap to allocate its buffer
941 [16961] math: nan function incorrect handling of bad sequences
942 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
943 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
944 [16985] locale: localedef: confusing error message when opening output
946 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
947 [17197] locale: Redundant shift character in iconv conversion output at
949 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
950 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
951 [17250] dynamic-link: static linking breaks nss loading
952 (getaddrinfo/getpwnam/etc...)
953 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
955 [17441] math: isnan() should use __builtin_isnan() in GCC
956 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
957 (related to lock elision)
958 [17787] manual: Exponent on page 324 of the PDF ends prematurely
959 [17886] time: strptime should be able to parse "Z" as a timezone with %z
960 [17887] time: strptime should be able to parse "+01:00" style timezones
961 [17905] libc: catopen() Multiple unbounded stack allocations
963 [18084] libc: backtrace (..., 0) dumps core on x86
964 [18086] libc: nice() sets errno to 0 on success
965 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
966 is too large (CVE-2015-8778)
967 [18251] dynamic-link: SONAME missing when audit modules provides path
968 [18265] libc: add attributes for wchar string and memory functions
969 [18370] math: csqrt missing underflows
970 [18421] libc: [hppa] read-only segment has dynamic relocations
971 [18472] libc: Obsolete syscall wrappers should be compat symbols
972 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
973 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
974 [18525] localedata: Remove locale timezone information
975 [18560] libc: [powerpc] spurious bits/ipc.h definitions
976 [18568] localedata: Update locale data to Unicode 8.0
977 [18589] locale: sort-test.sh fails at random
978 [18595] math: ctan, ctanh missing underflows
979 [18604] libc: assert macro-expands its argument
980 [18610] math: S390: fetestexcept() reports any exception if DXC-code
981 contains a vector instruction exception.
982 [18611] math: j1, jn missing errno setting on underflow
983 [18618] localedata: sync Chechen locale definitions with other *_RU
985 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
986 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
987 [18665] network: In send_dg, the recvfrom function is NOT always using the
988 buffer size of a newly created buffer (CVE-2015-7547)
989 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
991 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
993 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
994 [18699] math: tilegx cproj() for various complex infinities does not yield
996 [18724] libc: Harden put*ent functions against data injection
997 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
998 [18755] build: build errors with -DNDEBUG
999 [18757] stdio: fmemopen fails to set errno on failure
1000 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1002 [18781] libc: openat64 lacks O_LARGEFILE
1003 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1004 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1005 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1006 [18790] math: [ldbl-128ibm] tanhl inaccurate
1007 [18795] libc: stpncpy fortification misses buffer lengths that are
1008 statically too large
1009 [18796] build: build fails for --disable-mathvec
1010 [18803] math: hypot missing underflows
1011 [18820] stdio: fmemopen may leak memory on failure
1012 [18823] math: csqrt spurious underflows
1013 [18824] math: fma spurious underflows
1014 [18825] math: pow missing underflows
1015 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1016 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1017 [18870] build: sem_open.c fails to compile with missing symbol
1019 [18872] stdio: Fix memory leak in printf_positional
1020 [18873] libc: posix_fallocate overflow check ineffective
1021 [18875] math: Excess precision leads incorrect libm
1022 [18877] libc: arm: mmap offset regression
1023 [18887] libc: memory corruption when using getmntent on blank lines
1024 [18918] localedata: hu_HU: change time to HH:MM:SS format
1025 [18921] libc: Regression: extraneous stat() and fstat() performed by
1027 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1028 binaries (CVE-2015-8777)
1029 [18951] math: tgamma missing underflows
1030 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1032 [18953] localedata: lt_LT: change currency symbol to the euro
1033 [18956] math: powf inaccuracy
1034 [18961] math: [i386] exp missing underflows
1035 [18966] math: [i386] exp10 missing underflows
1036 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1037 [18969] build: multiple string test failures due to missing locale
1039 [18970] libc: Reference of pthread_setcancelstate in libc.a
1040 [18977] math: float / long double Bessel functions not in XSI POSIX
1041 [18980] math: i386 libm functions return with excess range and precision
1042 [18981] math: i386 scalb*, ldexp return with excess range and precision
1043 [18982] stdio: va_list and vprintf
1044 [18985] time: Passing out of range data to strftime() causes a segfault
1046 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1047 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1048 [19012] locale: iconv_open leaks memory on error path
1049 [19016] math: clog, clog10 inaccuracy
1050 [19018] nptl: Mangle function pointers in tls_dtor_list
1051 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1052 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1053 [19048] malloc: malloc: arena free list can become cyclic, increasing
1055 [19049] math: [powerpc] erfc incorrect zero sign
1056 [19050] math: [powerpc] log* incorrect zero sign
1057 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1058 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1059 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1060 [19074] network: Data race in _res_hconf_reorder_addrs
1061 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1062 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1063 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1065 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1067 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1068 [19086] manual: posix_fallocate64 documented argument order is wrong.
1069 [19088] math: lround, llround missing exceptions close to overflow
1071 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1072 [19095] math: dbl-64 lrint incorrect for 64-bit long
1073 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1074 [19124] dynamic-link: ld.so failed to build with older assmebler
1075 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1076 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1077 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1078 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1079 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1080 pthread_setaffinity_np
1081 [19156] math: [ldbl-128] j0l spurious underflows
1082 [19164] nptl: tst-getcpu fails with many possible CPUs
1083 [19168] math: math/test-ildoubl and math/test-ldouble failure
1084 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1085 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1087 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1088 (FE_NOMASK_ENV) do not clear SSE exceptions
1089 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1090 _int_new_arena/reused_arena
1091 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1092 [19201] math: dbl-64 remainder incorrect sign of zero result
1093 [19205] math: bits/math-finite.h conditions do not match math.h and
1095 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1096 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1097 for C99-based standards
1098 [19212] libc: features.h not -Wundef clean
1099 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1101 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1102 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1103 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1105 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1106 "inexact" exceptions
1107 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1109 [19242] libc: strtol incorrect in Turkish locales
1110 [19243] malloc: reused_arena can pick an arena on the free list, leading
1111 to an assertion failure and reference count corruption
1112 [19253] time: tzset() ineffective when temporary TZ did not include DST
1114 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1115 [19270] math: [hppa] Shared libm missing __isnanl
1116 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1117 MAP_HUGETLB and MAP_STACK defines
1118 [19313] nptl: Wrong __cpu_mask for x32
1119 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1120 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1121 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1122 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1123 [19363] time: x32: times() return value wrongly truncates/sign extends
1125 [19367] dynamic-link: Improve branch prediction on Silvermont
1126 [19369] network: Default domain name not reset by res_ninit when "search"
1127 / "domain" entry is removed from resolv.conf
1128 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1130 [19385] localedata: bg_BG: time separator should be colon, not comma
1131 [19408] libc: linux personality syscall wrapper may erroneously return an
1132 error on 32-bit architectures
1133 [19415] libc: dladdr returns wrong names on hppa
1134 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1135 IBM905, IBM907, and IBM909
1136 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1137 [19443] build: build failures with -DDEBUG
1138 [19451] build: Make check fails on test-double-vlen2
1139 [19462] libc: Glibc failed to build with -Os
1140 [19465] math: Wrong code with -Os
1141 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1143 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1145 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1146 [19486] math: S390: Math tests fail with "Exception Inexact set".
1147 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1148 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1150 [19590] math: Fail to build shared objects that use libmvec.so functions.
1154 * The following bugs are resolved with this release:
1156 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1157 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1158 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1159 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1160 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1161 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1162 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1163 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1164 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1165 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1166 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1167 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1168 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1169 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1170 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1171 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1172 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1173 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1174 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1175 18657, 18676, 18694, 18696, 18887.
1177 * Cache information can be queried via sysconf() function on s390 e.g. with
1178 _SC_LEVEL1_ICACHE_SIZE as argument.
1180 * A buffer overflow in gethostbyname_r and related functions performing DNS
1181 requests has been fixed. If the NSS functions were called with a
1182 misaligned buffer, the buffer length change due to pointer alignment was
1183 not taken into account. This could result in application crashes or,
1184 potentially arbitrary code execution, using crafted, but syntactically
1185 valid DNS responses. (CVE-2015-1781)
1187 * The time zone file parser has been made more robust against crafted time
1188 zone files, avoiding heap buffer overflows related to the processing of
1189 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1190 large time zone data files. Overly long time zone specifiers in the TZ
1191 variable no longer result in stack overflows and crashes.
1193 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1194 for LD and GD on x86 and x86-64, has been implemented. You will need
1195 binutils-2.24 or later to enable this optimization.
1197 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1198 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1199 Hat). These updates cause user visible changes, such as the fix for bug
1202 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1203 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1204 condition in some applications.
1206 * Added vector math library named libmvec with the following vectorized x86_64
1207 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1209 The library can be disabled with --disable-mathvec. Use of the functions is
1210 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1211 Shared library libmvec.so is linked in as needed when using -lm (no need to
1212 specify -lmvec explicitly for not static builds).
1213 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1215 * A new fmemopen implementation has been added with the goal of POSIX
1216 compliance. The new implementation fixes the following long-standing
1217 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1218 old implementation is still present for use be by existing binaries.
1220 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1221 releases. It has been fixed to match 2.19 and older, but binaries built
1222 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1224 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1225 Contributed by Roland McGrath (Google).
1227 * The header <regexp.h> is deprecated, and will be removed in a future
1228 release. Use of this header will trigger a deprecation warning.
1229 Application developers should update their code to use <regex.h> instead.
1231 This header was formerly part of SUS, but was deprecated in 1994 and
1232 removed from the standard in 2001. Also, the glibc implementation
1233 leaks memory. See BZ#18681 for more details.
1237 * The following bugs are resolved with this release:
1239 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1240 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1241 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1242 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1243 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1244 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1245 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1246 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1247 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1248 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1251 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1252 too little memory for the to-be-scanned arguments and overflow the
1253 allocated buffer. The implementation now correctly computes the required
1254 buffer size when using malloc, and switches to malloc from alloca as
1257 * A new semaphore algorithm has been implemented in generic C code for all
1258 machines. Previous custom assembly implementations of semaphore were
1259 difficult to reason about or ensure that they were safe. The new version
1260 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1261 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1262 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1264 * Port to Altera Nios II has been contributed by Mentor Graphics.
1266 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1267 implementations for powerpc64/powerpc64le.
1268 Implemented by Adhemerval Zanella (IBM).
1270 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1271 and powerpc64le. This may improve lock scaling of existing programs on
1272 HTM capable systems. The lock elision code is only enabled with
1273 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1274 powerpc will issue a transaction abort on every syscall to avoid side
1275 effects being visible outside transactions.
1277 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1278 AArch64. Contributed by ARM Ltd.
1280 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1282 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1283 under certain input conditions resulting in the execution of a shell for
1284 command substitution when the application did not request it. The
1285 implementation now checks WRDE_NOCMD immediately before executing the
1286 shell and returns the error WRDE_CMDSUB as expected.
1288 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1289 processing format strings with a large number of format specifiers.
1291 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1292 infinite loop if the DNS response contained a PTR record of an unexpected
1295 * The minimum GCC version that can be used to build this version of the GNU
1296 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1297 still be used to compile programs using the GNU C Library.
1299 * The GNU C Library is now built with -Werror by default. This can be
1300 disabled by configuring with --disable-werror.
1302 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1304 * The obsolete sigvec function has been removed. This was the original
1305 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1306 programs have been using instead for about 25 years. Of course, ABI
1307 compatibility for old binaries using sigvec remains intact.
1309 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1310 with newer versions of bison.
1312 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1313 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1314 registers overlay two consecutive single-precision registers. MIPS32R2
1315 introduced a new FPU mode (FR=1) where double-precision registers extend the
1316 corresponding single-precision registers which is incompatible with the
1317 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1318 require the use of FR=1 making a transition necessary. New o32 ABI
1319 extensions enable users to migrate over time from the original o32 ABI
1320 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1321 tracks the ABI of any loaded object and verifies that new objects are
1322 compatible. Mode transitions will also be requested as required and
1323 unsupportable objects will be rejected. The ABI checks include both soft and
1324 hard float ABIs for o32, n32 and n64.
1326 GCC 5 with GNU binutils 2.25 onwards:
1327 It is strongly recommended that all o32 system libraries are built using the
1328 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1329 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1330 compiler using --with-fp-32=xx to set this by default.
1334 * The following bugs are resolved with this release:
1336 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1337 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1338 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1339 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1340 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1341 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1342 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1343 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1344 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1345 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1346 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1347 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1348 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1349 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1350 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1352 * Reverted change of ABI data structures for s390 and s390x:
1353 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1354 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1355 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1356 are preserved pointing straight to the same implementation as the old ones.
1357 Given that, new callers will simply provide a too-big buffer to these
1358 functions. Any applications/libraries out there that embed jmp_buf or
1359 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1360 against 2.19 headers will have to rebuilt again. This is necessary in any
1361 case to revert the breakage in their ABI caused by the glibc change.
1363 * Support for file description locks is added to systems running the
1364 Linux kernel. The standard file locking interfaces are extended to
1365 operate on file descriptions, not file descriptors, via the use of
1366 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1367 are associated with an open file instead of a process.
1369 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1371 * The minimum Linux kernel version that this version of the GNU C Library
1372 can be used with is 2.6.32.
1374 * Running the testsuite no longer terminates as soon as a test fails.
1375 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1376 with PASS or FAIL lines for individual tests. A summary of the results is
1377 printed, including a list of failing lists, and "make check" exits with
1378 error status if there were any unexpected failures. "make check
1379 stop-on-test-failure=y" may be used to keep the old behavior.
1381 * The am33 port, which had not worked for several years, has been removed
1384 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1385 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1386 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1387 conflicted with POSIX (support for which was removed in 2.19), the
1388 interfaces those macros enabled remain available when compiling with
1389 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1390 test macros defined.
1392 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1394 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1395 This may improve lock scaling of existing programs on TX capable systems.
1396 The lock elision code is only built with --enable-lock-elision=yes and
1397 then requires a GCC version supporting the TX builtins. With lock elision
1398 default mutexes are elided via __builtin_tbegin, if the cpu supports
1399 transactions. By default lock elision is not enabled and the elision code
1402 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1403 copy the path argument. This allowed programs to cause posix_spawn to
1404 deference a dangling pointer, or use an unexpected pathname argument if
1405 the string was modified after the posix_spawn_file_actions_addopen
1408 * All supported architectures now use the main glibc sysdeps directory
1409 instead of some being in a separate "ports" directory (which was
1410 distributed separately before glibc 2.17).
1412 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1413 On configurations that support it (all Linux configurations), it's now
1414 used regardless of the --enable-add-ons switch to configure. It is no
1415 longer possible to build such configurations without pthreads support.
1417 * Locale names, including those obtained from environment variables (LANG
1418 and the LC_* variables), are more tightly checked for proper syntax.
1419 setlocale will now fail (with EINVAL) for locale names that are overly
1420 long, contain slashes without starting with a slash, or contain ".." path
1421 components. (CVE-2014-0475) Previously, some valid locale names were
1422 silently replaced with the "C" locale when running in AT_SECURE mode
1423 (e.g., in a SUID program). This is no longer necessary because of the
1426 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1427 with application code using Intel MPX instructions. (With all previous
1428 versions, the MPX register state could be clobbered when making calls
1429 into or out of a shared library.) Note that while the new dynamic
1430 linker is compatible with all known x86 hardware whether or not it
1431 supports Intel MPX, some x86 instruction-set emulators might fail to
1432 handle the new instruction encodings. This is known to affect Valgrind
1433 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1434 release), and might affect other tools that do instruction emulation.
1436 * Support for loadable gconv transliteration modules has been removed.
1437 The support for transliteration modules has been non-functional for
1438 over a decade, and the removal is prompted by security defects. The
1439 normal gconv conversion modules are still supported. Transliteration
1440 with //TRANSLIT is still possible, and the //IGNORE specifier
1441 continues to be supported. (CVE-2014-5119)
1443 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1444 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1445 resulting a denial-of-service security vulnerability in applications which
1446 use functions related to iconv. (CVE-2014-6040)
1450 * The following bugs are resolved with this release:
1452 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1453 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1454 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1455 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1456 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1457 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1458 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1459 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1460 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1461 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1462 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1463 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1464 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1465 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1466 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1467 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1468 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1469 16431, 16453, 16474, 16506, 16510, 16529
1471 * Slovenian translations for glibc messages have been contributed by the
1472 Translation Project's Slovenian team of translators.
1474 * The public headers no longer use __unused nor __block. This change is to
1475 support compiling programs that are derived from BSD sources and use
1476 __unused internally, and to support compiling with Clang's -fblock
1477 extension which uses __block.
1479 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1480 large collation sequences to optimize multiple passes. This cache
1481 computation may overflow for large collation sequences and may cause a
1482 stack or buffer overflow. This is now fixed to use a slower algorithm
1483 which does not use a cache if there is an integer overflow.
1485 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1486 rules for large collation sequences to optimize multiple passes and falls
1487 back to alloca if malloc fails, resulting in a possible stack overflow.
1488 The implementation now falls back to an uncached collation sequence lookup
1491 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1492 initialized for static applications resulting in the security feature
1493 being disabled. The pointer guard is now correctly initialized to a
1494 random value for static applications. Existing static applications need
1495 to be recompiled to take advantage of the fix (bug 15754).
1497 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1498 to the d_name member of struct dirent, or omit the terminating NUL
1499 character. (Bugzilla #14699).
1501 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1502 aligned_alloc functions could allocate too few bytes or corrupt the
1503 heap when passed very large allocation size values (Bugzilla #15855,
1506 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1507 for AF_INET6 has been fixed (Bugzilla #16072).
1509 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1510 pap_CW, quz_PE, the_NP.
1512 * Substantially revised locales: gd_GB, ht_HT
1514 * The LC_ADDRESS field was updated to support country_car for almost all
1517 * ISO 1427 definitions were updated.
1519 * ISO 3166 definitions were updated.
1521 * The localedef utility now supports --big-endian and --little-endian
1522 command-line options to generate locales for a different system from that
1523 for which the C library was built.
1525 * Binary locale files now only depend on the endianness of the system for
1526 which they are generated and not on other properties of that system. As a
1527 consequence, binary files generated with new localedef may be incompatible
1528 with old versions of the GNU C Library, and binary files generated with
1529 old localedef may be incompatible with this version of the GNU C Library,
1530 in the following circumstances:
1532 + Locale files may be incompatible on m68k systems.
1534 + Locale archive files (but not separate files for individual locales) may
1535 be incompatible on systems where plain "char" is signed.
1537 * The configure option --disable-versioning has been removed. Builds with
1538 --disable-versioning had not worked for several years.
1540 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1541 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1543 * SystemTap probes for malloc have been introduced.
1545 * SystemTap probes for slow multiple precision fallback paths of
1546 transcendental functions have been introduced.
1548 * Support for powerpc64le has been added.
1550 * The soft-float powerpc port now supports e500 processors.
1552 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1554 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1555 set of header declarations that are enabled by default, even when other
1556 feature test macros or compiler options such as -std=c99 would otherwise
1557 disable some of those declarations.
1559 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1560 conflict with POSIX. The libbsd-compat library (which was a dummy library
1561 that did nothing) has also been removed.
1563 * Preliminary documentation about Multi-Thread, Async-Signal and
1564 Async-Cancel Safety has been added.
1566 * Change of ABI data structures for s390 and s390x:
1567 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1568 allow for future hardware extensions. All library functions that accept or
1569 return these structures were versioned in order to provide backwards
1570 compatibility with existing code. However, not all cases can be handled
1571 with interface versioning. If an application embeds either structure into
1572 another structure and that structure is passed to another compilation unit
1573 which expects the newer larger-sized object then any access to the new
1574 fields will result in undefined behaviour. Similarly any access to fields
1575 that were moved by the enlarging of the structures will cause undefined
1576 behaviour. To fix the undefined behaviour all objects that are part of the
1577 final in-memory image must agree on the size of structures, and this may
1578 require recompilation.
1582 * The following bugs are resolved with this release:
1584 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1585 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1586 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1587 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1588 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1589 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1590 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1591 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1592 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1593 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1594 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1595 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1596 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1599 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1600 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1601 Distributions can re-enable building and using pt_chown via the new configure
1602 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1603 considerable security risks and should only be used if the distribution
1604 understands and accepts the risks.
1606 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1609 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1610 fixed (Bugzilla #15330).
1612 * Add support for calling C++11 thread_local object destructors on thread
1613 and program exit. This needs compiler support for offloading C++11
1614 destructor calls to glibc.
1616 * Improved worst case performance of libm functions with double inputs and
1619 * Support for priority inherited mutexes in pthread condition variables on
1620 non-x86 architectures.
1622 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1624 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1626 * Optimized string functions for ARM. Implemented by Will Newton and
1629 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1631 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1634 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1635 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1637 * Added a benchmark framework to track performance of functions in glibc.
1639 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1640 It is based on draft TS 18661 and currently enabled as a GNU extension.
1642 * On Linux, the clock function now uses the clock_gettime system call
1643 for improved precision, rather than old times system call.
1645 * Added support for version-3 tz format files. This is needed when using
1646 the tz database release 2013e or later, and affects a few unusual cases --
1647 currently only TZ='America/Godthab' for time stamps after 2037.
1649 * Added new API functions pthread_getattr_default_np and
1650 pthread_setattr_default_np to get and set the default pthread
1651 attributes of a process.
1653 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1654 This may improve lock scaling of existing programs on TSX capable systems.
1655 When the --enable-lock-elision=yes parameter is specified at configure
1656 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1659 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1660 availability). Implemented by Ryan S. Arnold.
1662 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1666 * The following bugs are resolved with this release:
1668 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1669 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1670 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1671 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1672 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1673 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1674 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1675 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1676 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1677 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1678 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1679 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1680 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1681 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1682 14879, 14889, 14893, 14898, 14914.
1684 * Optimization of memcpy for MIPS.
1686 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1687 EMFILE has been fixed (Bugzilla #14889).
1689 * The add-on ports collection is now distributed in the "ports" subdirectory
1690 of the main GNU C Library distribution, rather than separately.
1692 * Port to ARM AArch64 contributed by Linaro.
1694 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1695 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1697 Implemented by Andreas Krebbel.
1699 * The new function secure_getenv allows secure access to the environment,
1700 returning NULL if running in a SUID/SGID process. This function replaces
1701 the internal function __secure_getenv.
1703 * SystemTap static probes have been added into the dynamic linker.
1704 Implemented by Gary Benson.
1706 * Optimizations of string functions strstr, strcasestr and memmem.
1707 Implemented by Maxim Kuvyrkov.
1709 * The minimum Linux kernel version that this version of the GNU C Library
1710 can be used with is 2.6.16.
1712 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1713 powerpc POWER7. Implemented by Will Schmidt.
1715 * New configure option --disable-nscd builds the C library such that it
1716 never attempts to contact the Name Service Caching Daemon (nscd).
1717 New configure option --disable-build-nscd avoids building nscd itself;
1718 this is the default if --disable-nscd is used.
1720 * Improved support for cross-compilation, including cross-testing and
1721 bootstrap builds without a previously built glibc.
1723 * Several testsuite tests are now able to test multiple IFUNC variants of an
1724 interface, rather than just testing the one that would be chooen by
1727 * New configure options --with-bugurl and --with-pkgversion, for
1728 distributors to use to embed their bug-reporting and package version
1729 information in --help and --version output.
1731 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1732 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1733 allows creation of chroots without the procfs mounted on /proc.
1735 * The `crypt' function now fails if passed salt bytes that violate the
1736 specification for those values. On Linux, the `crypt' function will
1737 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1738 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1739 when the mode is enabled.
1741 * The `clock_*' suite of functions (declared in <time.h>) is now available
1742 directly in the main C library. Previously it was necessary to link with
1743 -lrt to use these functions. This change has the effect that a
1744 single-threaded program that uses a function such as `clock_gettime' (and
1745 is not linked with -lrt) will no longer implicitly load the pthreads
1746 library at runtime and so will not suffer the overheads associated with
1747 multi-thread support in other code such as the C++ runtime library.
1749 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1754 * The following bugs are resolved with this release:
1756 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1757 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1758 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1759 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1760 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1761 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1762 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1763 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1764 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1765 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1766 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1767 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1768 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1769 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1770 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1771 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1772 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1773 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1774 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1775 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1776 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1777 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1780 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1781 configuring glibc with:
1782 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1783 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1784 Implemented by H.J. Lu.
1788 + define static_assert
1790 + do not declare gets
1792 + declare at_quick_exit and quick_exit also for ISO C11
1794 + aligned_alloc. NB: The code is deliberately allows the size parameter
1795 to not be a multiple of the alignment. This is a moronic requirement
1796 in the standard but it is only a requirement on the caller, not the
1799 + timespec_get added
1801 + uchar.h support added
1803 + CMPLX, CMPLXF, CMPLXL added
1805 Implemented by Ulrich Drepper.
1807 * Support for the IA-64 has been moved to ports.
1809 * Remove support for anything but ELF binary format
1811 * Checking versions of poll, ppoll added.
1812 Implemented by Ulrich Drepper.
1814 * More generic and 64-bit performance optimizations to math functions.
1815 Implemented by Ulrich Drepper.
1817 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1818 headers and functions available at compile time as they were before
1819 version 2.14. This option will be removed at some time in the future
1820 after the TI-RPC library becomes fully sufficient for the needs of
1821 existing applications.
1823 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1824 Note that glibc is not expected to work with any Linux kernel version
1827 * New header <sys/auxv.h> and function getauxval allowing easy access to
1828 the AT_* key-value pairs passed from the Linux kernel. The header also
1829 defines the HWCAP_* bits associated with the AT_HWCAP key.
1831 * New locales: mag_IN
1833 * New configure option --enable-systemtap builds SystemTap static probes
1834 into libc for setjmp and longjmp and into libpthread for various operations.
1835 So far the setjmp/longjmp probes and some of the libpthread probes are
1836 provided only for i*86 and x86_64.
1837 Implemented by Roland McGrath and Rayson Ho.
1839 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1841 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1844 * More optimized functions for SPARC. Implemented by David S. Miller.
1846 * Improved support for cross-compilation, especially bootstrap builds
1847 without a previously built glibc.
1849 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1850 Chris Metcalf from Tilera.
1852 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1853 now supported for ARM processors.
1855 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1856 as the name of the dynamic linker, to distinguish it from the
1857 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1859 * Support for CRIS has been removed from ports.
1861 * A new class of installed header has been documented for low-level
1862 platform-specific functionality. PowerPC added the first instance with a
1863 function to provide time base register access. Contributed by Tulio
1864 Magno Quites Machado Filho.
1866 * ix86 configurations now install header files that are consistent with
1867 what x86-64 configurations install. These same header files can be used
1868 for -m32, -m64, or -mx32 builds.
1869 Contributed by H.J. Lu.
1871 * Math library bug fixes. A thorough audit of all open math library bugs was
1872 conducted by Joseph Myers. Significant progress was made on many math
1873 library bugs resulting in more accurate exceptions and function results.
1874 Many thanks to all those that contributed including Andreas Jaeger for his
1875 patch review and work on the x87 trigonometric instruction issues.
1877 * Timezone data is no longer installed. Timezone-related binaries and scripts
1878 will continue to be installed. Users should obtain their timezone data from
1879 their distribution provider or from the tzdata package at
1880 <ftp://munnari.oz.au/pub/>.
1884 * The following bugs are resolved with this release:
1886 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1887 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1888 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1889 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1890 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1891 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1892 13484, 13506, 13515, 13523, 13524, 13538, 13540
1894 * New program pldd to list loaded object of a process
1895 Implemented by Ulrich Drepper.
1897 * Add nss_db support back to glibc. No more dependency on Berkeley db
1898 and support for initgroups lookups.
1899 Implemented by Ulrich Drepper.
1901 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1902 Contributed by HJ Lu.
1904 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1905 Contributed by HJ Lu.
1907 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1908 on x86-32 and x86-64.
1909 Contributed by Liubov Dmitrieva.
1911 * Optimized strchr and strrchr for SSE on x86-32.
1912 Contributed by Liubov Dmitrieva.
1914 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1915 for x86-64 and x86-32.
1916 Contributed by Liubov Dmitrieva.
1918 * New interfaces: scandirat, scandirat64
1919 Implemented by Ulrich Drepper.
1921 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1922 Implemented by Ulrich Drepper.
1924 * nscd now also caches the netgroup database.
1925 Implemented by Ulrich Drepper.
1927 * Integrate libm with gcc's -ffinite-math-only option.
1928 Implemented by Ulrich Drepper.
1930 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1931 to math functions. Implemented by Ulrich Drepper.
1933 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1934 Implemented by Ulrich Drepper.
1936 * New Linux interfaces: process_vm_readv, process_vm_writev
1938 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1939 Implemented by Ulrich Drepper.
1941 * Optimized nearbyint and strcasecmp for PPC.
1942 Implemented by Adhemerval Zanella.
1944 * New locales: bho_IN, unm_US, es_CU, ta_LK
1948 * The following bugs are resolved with this release:
1950 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1951 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1952 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1953 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1954 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1955 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1956 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1957 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1958 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1959 12795, 12811, 12813, 12814, 12841
1961 * The RPC implementation in libc is obsoleted. Old programs keep working
1962 but new programs cannot be linked with the routines in libc anymore.
1963 Programs in need of RPC functionality must be linked against TI-RPC.
1964 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1966 Visible changes of this change include (obviously) the inability to link
1967 programs using RPC functions without referencing the TI-RPC library and the
1968 removal of the RPC headers from the glibc headers.
1969 Implemented by Ulrich Drepper.
1971 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1972 syncfs, setns, sendmmsg
1974 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1975 yue_HK, lij_IT, mhr_RU
1977 * New encodings: CP770, CP771, CP772, CP773, CP774
1979 * New program sotruss to trace calls through PLTs
1980 Implemented by Ulrich Drepper.
1982 * The malloc hook implementation is marked deprecated and will be removed
1983 from the default implementation in the next version. The design never
1984 worked ever since the introduction of threads. Even programs which do
1985 not create threads themselves can use multiple threads created internally.
1989 * The following bugs are resolved with this release:
1991 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1992 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1993 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1994 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1997 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1999 * POWER7 optimizations: memset, memcmp, strncmp
2001 * New optimized string functions for x86-64: strnlen (SSE2),
2002 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2003 Implemented by Ulrich Drepper.
2007 * The following bugs are resolved with this release:
2009 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2010 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2011 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2012 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2013 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2014 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2015 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2016 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2018 * New interfaces: pthread_getname_np, pthread_setname_np
2020 * New Linux interface: recvmmsg
2022 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2024 * The dynamic linker now recognizes supported ABI versions from the
2025 EI_ABIVERSION field in the ELF header.
2026 Implemented by Ulrich Drepper.
2028 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2029 will not be used to fill in password fields in the passwd.byname replies.
2030 Instead it is used to synthesize the shadow.byname table, should it be
2031 missing. This is a useful mode in some installations involving Solaris.
2032 Implemented by Ulrich Drepper.
2034 * New locales: kok_IN, sq_MK, cv_RU
2038 * The following bugs are resolved with this release:
2040 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2041 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2042 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2043 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2044 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2045 10789, 10817, 10823, 10840, 10847
2047 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2049 Implemented by Ulrich Drepper.
2051 * Checking version of longjmp added that fails if an uninitialized stack
2052 frame would be created. Implemented by Ulrich Drepper.
2054 * STT_GNU_IFUNC is now supported in static executables.
2055 Implemented by H.J. Lu.
2057 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2059 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2060 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2061 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2062 Contributed by H.J. Lu.
2064 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2065 Implemented by Ulrich Drepper.
2067 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2069 Contributed by H.J. Lu.
2071 * Support for fma instruction in AVX on x86-64.
2072 Implemented by H.J. Lu and Ulrich Drepper.
2074 * AVX support in x86-64 auditing support in ld.so.
2075 Implemented by H.J. Lu.
2077 * STB_GNU_UNIQUE support added.
2078 Implemented by Ulrich Drepper.
2080 * Implement second fallback mode for optimized DNS lookup for even more
2081 broken environments. If two requests from the same port are not
2082 handled correctly close the socket and open a new one before sending
2083 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2084 can be used to select this mode right away, instead of rediscovering the
2085 necessity is every process again.
2086 Implemented by Ulrich Drepper.
2088 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2089 Implemented by Adam Tkac.
2091 * Optimized iconv conversions for S390x.
2092 Implemented by Andreas Krebbel.
2094 * Using condvars with PI mutexes is now more efficient due to kernel
2095 support for requeueing to PI futexes. NPTL support added for x86-64.
2096 Implemented by Ulrich Drepper.
2098 * New locales: ps_AF, my_MM
2102 * The following bugs are resolved with this release:
2104 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2105 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2106 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2107 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2108 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2110 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2111 Implemented by Ulrich Drepper.
2113 * New Linux interfaces: accept4, fallocate, fallocate64.
2114 Implemented by Ulrich Drepper.
2116 * Correct declarations of string function when used in C++ code. This
2117 could lead to compile errors for invalid C++ code.
2119 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2122 * New POSIX 2008 interface: psiginfo
2123 Implemented by Ulrich Drepper.
2125 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2126 Implemented by Ulrich Drepper.
2128 * Support for selecting between multiple function definitions at runtime
2129 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2131 * The libcrypt library can now use the hash function implementations in
2132 NSS. Implemented by Ulrich Drepper.
2134 * The malloc implementation can be compiled to be less memory efficient
2135 but higher performing in multi-threaded programs.
2136 Implemented by Ulrich Drepper.
2138 * New locales: nan_TW@latin, ks_IN
2140 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2141 Implemented by Ulrich Drepper.
2143 * Extended printf hook support. It is possible to use user-defined types
2144 and extend existing format specifiers.
2145 Implemented by Ulrich Drepper.
2147 * Handling for group shadow files has been added.
2148 Implemented by Ulrich Drepper.
2150 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2151 servers (the case, e.g., for some people using the built-in DNS
2152 server in ADSL modems/routers). There is a once-per-process timeout
2153 in case of a broken server. To avoid it, users can run nscd or put
2154 'options single-request' in /etc/resolv.conf.
2155 Implemented by Ulrich Drepper.
2159 * The following bugs are resolved with this release:
2161 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2162 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2163 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2164 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2165 6968, 6974, 6980, 6995, 7008, 7009, 7029
2167 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2168 up at the same time. Implemented by Ulrich Drepper.
2170 * TLS descriptors for LD and GD on x86 and x86-64.
2171 Implemented by Alexandre Oliva.
2173 * getaddrinfo now handles DCCP and UDPlite.
2174 Implemented by Ulrich Drepper.
2176 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2177 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2178 Implemented by Ulrich Drepper.
2180 * New implementation of memmem, strstr, and strcasestr which is O(n).
2181 Implemented by Eric Blake.
2183 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2185 * Implement "e" option for popen to open file descriptor with the
2186 close-on-exec flag set. Implemented by Ulrich Drepper.
2188 * Implement "b" mode for fmemopen. In this mode writes writes don't
2189 implicitly add a NUL byte and seeks from the end of the buffer really
2190 use the buffer end, not the string length as the basis.
2191 Implemented by Ulrich Drepper.
2193 * Many functions, exported and internal, now atomically set the close-on-exec
2194 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2196 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2198 Implemented by Pravin Satpute.
2200 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2204 * The following bugs are resolved with this release:
2206 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2207 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2208 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2209 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2210 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2211 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2212 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2214 * New locales: bo_CN, bo_IN, shs_CA.
2216 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2218 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2219 Implemented by Pravin Satpute.
2221 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2223 * nscd caches now all timeouts for DNS entries
2224 Implemented by Ulrich Drepper.
2226 * nscd is more efficient and wakes up less often.
2227 Implemented by Ulrich Drepper.
2229 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2230 vdprintf, and obstack_vprintf.
2231 Implemented by Jakub Jelinek.
2233 * Faster memset for x86-64.
2234 Implemented by Harsha Jagasia and H.J. Lu.
2236 * Faster memcpy on x86.
2237 Implemented by Ulrich Drepper.
2239 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2240 Implemented by Ulrich Drepper.
2242 * Faster sqrt and sqrtf implemention for some PPC variants.
2243 Implemented by Steven Munroe.
2247 * The following bugs are resolved with this release:
2249 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2250 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2251 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2252 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2253 5071, 5103, 5104, 5112, 5113, 5184, 5186
2255 * More checking functions: fread, fread_unlocked, open*, mq_open.
2256 Implemented by Jakub Jelinek and Ulrich Drepper.
2258 * Extend fortification to C++. Implemented by Jakub Jelinek.
2260 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2261 by not recognizing 'a' as a modifier when those specs are requested.
2262 Implemented by Jakub Jelinek.
2264 * PPC optimizations to math and string functions.
2265 Implemented by Steven Munroe.
2267 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2268 options to be passed. Implemented by Ulrich Drepper.
2270 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2272 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2273 Implemented by Ulrich Drepper.
2275 * Handle private futexes in the NPTL implementation.
2276 Implemented by Jakub Jelinek and Ulrich Drepper.
2278 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2279 Implemented by Roland McGrath and Ulrich Drepper.
2281 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2283 * SHA-256 and SHA-512 based password encryption.
2284 Implemented by Ulrich Drepper.
2286 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2287 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2290 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2291 Implemented by Ulrich Drepper.
2295 * The following bugs are resolved with this release:
2297 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2298 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2299 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2300 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2301 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2302 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2303 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2306 * New Linux interfaces: epoll_pwait, sched_getcpu.
2308 * New generic interfaces: strerror_l.
2310 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2314 * The following bugs are resolved with this release:
2316 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2317 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2318 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2319 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2320 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2321 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2322 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2323 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2324 3225, 3251, 3252, 3253, 3273
2326 * For Linux, the sorting of addresses returned by getaddrinfo now also
2327 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2328 Implemented by Ulrich Drepper.
2330 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2331 Implemented by Ulrich Drepper.
2333 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2335 * New iconv module for MIK. Contributed by Alexander Shopov.
2337 * For sites with broken group and/or passwd database, the auto-propagate
2338 option of nscd can prevent creating ID lookup entries from the results
2339 of a name lookup and vice versa. This usually is no problem but some
2340 site might have problems with the default behavior.
2341 Implemented by Ulrich Drepper.
2343 * Iterating over entire database in NIS can be slow. With the
2344 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2345 to trade time for memory. The entire database will be read at once.
2346 Implemented by Ulrich Drepper.
2348 * The interfaces introduced in RFC 3542 have been implemented by
2351 * Support for the new ELF hash table format was added by Ulrich Drepper.
2353 * Support for priority inheritance mutexes added by Jakub Jelinek and
2356 * Support for priority protected mutexes added by Jakub Jelinek.
2358 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2359 or_IN, csb_PL, fy_NL, sr_ME.
2363 * More overflow detection functions.
2365 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2366 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2368 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2369 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2370 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2371 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2372 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2373 by Masahide Washizawa.
2375 * It is now possible to install audit libraries for the dynamic linker using
2376 LD_AUDIT. Implemented by Ulrich Drepper.
2378 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2379 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2380 For a libc and libpthread that works well on Linux 2.4 kernels, we
2381 recommend using the stable 2.3 branch.
2383 * The new function open_wmemstream acts like open_memstream,
2384 but uses a wchar_t wide character buffer.
2386 * The new function ppoll is an improved version of poll, similar to pselect.
2388 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2389 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2390 renameat, unlinkat, linkat, symlinkat, readlinkat.
2392 * New Linux kernel system calls: unshare,
2393 inotify_init, inotify_add_watch, inotify_rm_watch.
2395 * The euidaccess function is now also known by the alias eaccess,
2396 for compatibility with some other systems.
2398 * Timezone data updated to 2006b version.
2402 * The following bugs are resolved with this release:
2404 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2405 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2406 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2407 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2408 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2409 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2411 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2413 * As of this release, GCC 4 can be used to compile the C Library.
2415 * Timezone data updated to 2005m version.
2419 * The following bugs are resolved with this release:
2421 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2422 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2423 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2424 777, 787, 821, 822, 823, 825
2426 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2430 * Support for RFC 3678. Real implementations exist only for Linux so far.
2431 Implemented by Ulrich Drepper.
2433 * nscd can now cache entries persistently. Expiring entries are reloaded.
2434 For speedups the cache can be shared in memory with client processes.
2435 Implemented by Ulrich Drepper.
2437 * nscd can now perform SELinux checks.
2438 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2440 * getaddrinfo queries are now cached. Canonical name lookup is performed
2442 Implemented by Ulrich Drepper.
2444 * The nothrow function attribute is used when headers are used by gcc when
2445 compiling C code. This can avoid significant amounts of exception
2448 * The malloc functions perform more error checking and are stricter when
2449 it comes to reacting on errors. The default action is to terminate
2450 the process after showing an error message. Implemented by Ulrich Drepper.
2452 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2453 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2455 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2456 namespaces using the new function dlmopen(). This feature is of course,
2457 like most other dynamic loading functionality, not available in statically
2458 linked applications. Implemented by Ulrich Drepper.
2460 * Low-overhead boundary checking variants of string and some stdio functions
2461 were added. These are to be used in conjunction with a gcc patch by
2462 Jakub Jelinek which adds calls to these functions if possible.
2463 Implemented by Jakub Jelinek and Ulrich Drepper.
2465 * Old code for several operating systems and machine architectures that
2466 have not been in working condition in a long time have been removed from
2467 the main source tree maintained by the GNU C Library's maintainers.
2468 These files are now reside in the separate `ports' source module
2469 that is usable as an add-on when building the library.
2473 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2474 interrogate the dynamic linker, compatible with the Solaris interface.
2476 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2477 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2479 * getifaddrs now uses the netlink interface on Linux to get its information.
2480 Implemented by Thorsten Kukuk.
2482 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2483 Implemented by Ulrich Drepper.
2485 * support for non-executable stacks on x86 has been added. Changes mostly
2488 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2491 * getaddrinfo now performs destination address selection according to
2496 * Thread-safe interfaces for many functions that access locale data
2497 were added in version 2.3, but these features were omitted from NEWS.
2498 Many functions have variants with an `_l' suffix that take a `locale_t'
2499 object as a parameter rather than consulting the current locale.
2500 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2501 create and maintain `locale_t' objects. Additionally, the new function
2502 `uselocale' sets "the current locale" (as used by functions not so
2503 parameterized) set for an individual thread. These features were added
2504 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2506 * The functions getresuid, getresgid, setresuid, and setresgid, which
2507 have long been available on Linux, are now declared in <unistd.h>
2508 and are now also available on the Hurd.
2510 * ELF thread-local storage support (TLS) now works on x86-64.
2512 * The new dynamic string token $LIB is expanded in shared library names.
2513 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2515 * Aldy Hernandez contributed complete software floating point support for
2516 PowerPC machines with no FPU.
2518 * fexecve is implemented on Linux.
2520 * The `btowc' function should work at least twice as fast due to
2521 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2523 * With approriate thread add-ons cancelable functions are now implemented
2524 in libc.so as well. No need to call the function in libpthread. This
2525 change allowed to finally disable the incorrect and expensive handling
2526 of weak definition in ld.so.
2528 * Yet more PLT entries in libc.so have been removed. We finally arrived
2529 at the bare minimum. Startup times improved appropriately.
2531 * Support for the new Linux/x86 system call interface was added. The
2532 AT_SYSINFO auxiliary vector entry is recognized and handled.
2536 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2539 * iconv (the program and the interface) now accepts empty names (excluding
2540 options like //TRANSLIT) to mean "use charset of current locale".
2542 * localedef can now transliterate characters in strings which are not in
2543 the provided charmap. The information from the input locale is used.
2545 * Prelinking support was added for ELF targets. This requires additional
2546 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2548 * Read-only stdio streams now use mmap to speed up operation by eliminating
2549 copying and buffer underflows. To use add 'm' to the mode string of
2550 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2552 * The malloc functions were completely rewritten by Wolfram Gloger based
2553 on Doug Lea's malloc-2.7.0.c.
2555 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2556 implementation of regex.
2558 * Bruno Haible upgraded the iconv and locale implementation to support
2561 * Contents of the LC_* and LANG environment variables in the CEN style are
2562 not recognized anymore. It never was used. Change by Ulrich Drepper.
2564 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2565 thread-local storage (TLS) ABI on some platforms.
2566 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2568 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2569 EUC-JISX0213, and TSCII.
2571 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2572 BSD-compatible interface for getting all network interface addresses.
2573 Implementation for IPv4 by Roland McGrath.
2575 * Loading of locale data is faster due to the introduction of a locale
2576 archive. Implemented by Roland McGrath and Ulrich Drepper.
2578 * Startup times are significantly reduced by not using exported functions
2579 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2582 * Steven Munroe contributed a port to PowerPC64/Linux.
2586 * The Hurd now uses the GNU libio implementation of stdio rather than the
2587 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2589 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2590 with the same behavior as the Linux system call of the same name.
2594 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2595 128-bit long double format.
2597 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2598 IBM1161, and IBM1162 charsets.
2600 * Andreas Jaeger contributed a port to x86-64/Linux.
2602 * Peter Bruin contributed a port to PowerPC/Hurd.
2604 * libc functions using I/O streams now can handle wide-oriented streams
2607 * optimizations in the dynamic linker. Binaries created by recent binutils
2608 versions start up quicker due to reduced time spent on relocations.
2610 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2614 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2615 asinh, atanh, j0 for the 128-bit long double format.
2617 * Bruno Haible updated all the code handling Unicode in some form to
2618 support Unicode 3.1.
2620 * Speed of regex for single-byte locales is back to previous levels.
2621 Patch by Isamu Hasegawa.
2623 * Alpha, SPARC, and IA-64 now also using floating stacks.
2625 * Startup time of internationalized applications greatly improved through
2626 iconv cache. Use iconvconfig to generate the cache file.
2627 Contributed by Ulrich Drepper.
2629 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2630 severe performance problems. Patch by David Mosberger.
2632 * The Hurd port got a lot more functionality like AIO, various stdio
2633 extensions, etc. Mainly done by Roland McGrath.
2635 * mtrace can now lookup symbols in shared libraries.
2639 * Intel's IA-64 math library is largely integrated. It provides fast and
2640 accurate implementatations for most basic and standard math functions
2641 in float, double, and long double format.
2643 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2644 and asin for the 96-bit long double format and asin, log, tan for the
2645 128-bit long double format.
2647 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2648 The basic double functions exist today. Contributed by Abraham Ziv
2649 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2650 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2652 * An asynchronous name lookup library was added. The interface is designed
2653 after POSIX AIO. The proposal was circulated beforehand to get comments.
2654 No negative ones came in. Implemented by Ulrich Drepper.
2656 * Port to S390/64bit contributed by Martin Schwidefsky
2657 <schwidefsky@de.ibm.com>.
2659 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2660 of functions for Linux/IA-64.
2662 * The RPC code is now thread safe. Threads can now use the same service
2663 of different services at the same time. Patch by Eric Norum
2664 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2666 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2667 family of functions for Linux/S390.
2669 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2670 of functions for Linux/x86.
2672 * Port to Linux/CRIS contributed by Axis Communications.
2676 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2677 we can now check for namespace violations and missing declarations. The
2678 result is that almost all headers are now Unix-compliant (as defined in
2679 the upcoming XPG6). The negative side is that some programs might need
2680 corrections, too, if they depend on the incorrect form of the headers in
2681 previous versions which defined too many symbols and included too many
2684 * regex now handles multibyte character sets correctly.
2685 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2687 * iconv (the program) does now conform to the upcoming XPG6 and handles
2688 charmaps. Instead of the charset names the path of charmaps can be
2689 provided and the conversion happens based on this data.
2690 Contributed by Ulrich Drepper.
2692 * The locale program now provides detailed information about the installed
2697 only lists the names of the supported locales
2701 provides details such as country, language, and codeset name.
2702 Contributed by Ulrich Drepper.
2706 * The gencat program now parses the input file according to the charset
2707 selected by the LC_CTYPE category. This is important for stateful
2708 character sets. To make generating catalogs easier there is a way
2709 to overwrite the charset selected by the locale: before the first
2710 message or $ quote line the catalog can contain a line like
2712 $ codeset=ISO-8859-2
2714 to select the charset (ISO-8859-2 in this case).
2716 Implemented by Shinya Hanataka and Ulrich Drepper.
2718 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2719 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2720 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2722 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2723 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2725 * The LANGUAGE environment variable is now ignored unless the locale is
2726 changed from the default "C" locale.
2728 * The usual bug fixes.
2732 * Greg McGary added runtime support for bounds checking using gcc's
2733 new -fbounded-pointers option. ix86 target is complete. PowerPC
2736 * Thorsten Kukuk added secure mode support to nscd.
2738 * The Berkeley db libraries have been removed.
2740 Related, the nss_db module is now in a separate package since it
2741 obviously requires a database library being available.
2743 * Wide character I/O streams implemented by Ulrich Drepper.
2745 * Functions from the extended socket API added by Ulrich Drepper.
2747 * Functions feenableexcept and fedisableexcept to control the
2748 behaviour of individual exceptions have been added by Andreas Jaeger.
2750 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2752 * The resolver code has been updated from bind 8.2.3-T5B which supports
2753 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2756 This change could in some situations effect backward compatibility. Since
2757 now `_res' is a thread-local instead of a global variable, modifying it
2758 in one thread does not have any effect in other threads.
2760 The resolver library was also extended to allow IPv6 as the transport
2761 protocol for the requests. This work was done by Stig Venaas.
2763 * Compatibility code for K&R C compilers has been removed from the
2764 header files. A ISO C compiler is needed to use the library
2765 (conforming to either C89 or C99 standard).
2767 * Complete rewrite of the localedef program to support multibyte character
2768 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2769 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2770 Implemented by Ulrich Drepper.
2772 Bruno Haible significantly improved the generation and use of the data
2773 structures for the wide character tables.
2775 * Plural handling in gettext implemented by Ulrich Drepper.
2777 * The utmp daemon has been removed.
2779 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2781 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2784 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2786 * POSIX barriers implemented by Kaz Kylheku.
2788 * POSIX spawn function family implemented by Ulrich Drepper.
2790 * POSIX spinlocks are now available.
2792 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2794 * the configure option --enable-kernel=X.Y.Z allows to strip out
2795 compatibility for kernel versions before X.Y.Z. This is currently only
2796 implemented for Linux.
2798 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2799 field sin6_scope_id. This means that all programs using IPv6 should be
2800 recompiled. Don't expect binary compatibility with previous glibc
2803 * various conversion modules for IBM character sets contributed by
2806 * IA-64 port by Jes Sorensen and HJ Lu.
2820 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2822 * New xdr functions are added; some rpc functions are now 64bit clean.
2824 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2826 * Fixed known incompatibilities with glibc 2.0.
2828 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2830 * Optimized a number of functions (especially the ELF dynamic loader).
2832 * Update timezone data files.
2834 * lots of charmaps corrections
2836 * some new locale definitions and charmaps
2841 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2842 conform to POSIX member type requirements. Symbol versions have been
2843 adjusted as needed within the library, and for direct use by applications,
2844 but there is potential for problems if third-party libraries use
2845 struct timeval as part of their interface. This does not present
2846 a problem for X and other "essential" system libraries.
2848 * An additional locale model to support C++ Standard Library locale
2849 model and probably more was implemented by Ulrich Drepper.
2851 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2854 * Miles Bader provided the `argp' function family to support hierachical
2855 command line argument parsing, layered on top of getopt.
2857 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2859 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2862 * scanf recognizes the %a and %A format for scanning floating point numbers.
2864 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2865 information and interfaces for the available integer types.
2867 * about 130 new math functions were added to implement the ISO C9x math
2870 * the new header <complex.h> contains definitions of the complex math
2871 functions from ISO C 9X.
2873 * the new header <tgmath.h> defines generic macros to use complex or
2874 real valued functions.
2876 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2878 * Andreas Jaeger provided a test suite for the math library.
2880 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2882 * Ulrich Drepper added character set conversion functions (iconv).
2884 * Optimized string functions have been added.
2886 * The localedata addon is now part of glibc.
2888 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2890 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2891 daemon for NSS (nscd).
2893 Missing a better place here are some numbers on improvements. Under
2894 Linux 2.1.125 un-tar-ing the kernel sources takes
2898 using local files 12.19s 6.88s 22.91s
2900 using NIS 13.92s 8.91s 26.34s
2902 using NIS & nscd 10.37s 7.34s 25.30s
2904 using NIS+ 27.57s 30.37s 640.46s
2906 using NIS+ & nscd 10.25s 7.83s 26.51s
2908 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2910 Keep in mind that non-namelookup related operations dominate above times.
2911 It was just a common complain that using NIS+ unpacking the kernel is
2914 [1] The old nscd implementation is not available anymore since it was
2915 distributed with glibc up to version 2.0.98 and thus is now replaced.
2917 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2919 * Mark Kettenis provided a Hesiod NSS module.
2921 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2922 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2924 * rcmd can now handle netgroups (Dick Streefland).
2926 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2929 * Support for the IPv6 protocol has been added to the socket API, as per the
2930 latest draft standards.
2932 * Support for Linux 2.2 has been added.
2934 * Interface changes relative to the latest 2.0.x release:
2935 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2936 addseverity NEW: Unix98
2937 alphasort64 NEW: LFS
2938 argp_err_exit_status NEW: argp, GNU ext
2939 argp_error NEW: argp, GNU ext
2940 argp_failure NEW: argp, GNU ext
2941 argp_help NEW: argp, GNU ext
2942 argp_parse NEW: argp, GNU ext
2943 argp_program_bug_address NEW: argp, GNU ext
2944 argp_program_version NEW: argp, GNU ext
2945 argp_program_version_hook NEW: argp, GNU ext
2946 argp_state_help NEW: argp, GNU ext
2947 argp_usage NEW: argp, GNU ext
2948 authdes_create NEW: Secure RPC
2949 authdes_getucred NEW: Secure RPC
2950 authdes_pk_create NEW: Secure RPC
2951 backtrace NEW: GNU ext.
2952 backtrace_symbols NEW: GNU ext.
2953 backtrace_symbols_fd NEW: GNU ext.
2955 cacosf NEW: ISO C 9x
2956 cacosh NEW: ISO C 9x
2957 cacoshf NEW: ISO C 9x
2958 cacoshl NEW: ISO C 9x
2959 cacosl NEW: ISO C 9x
2966 casinf NEW: ISO C 9x
2967 casinh NEW: ISO C 9x
2968 casinhf NEW: ISO C 9x
2969 casinhl NEW: ISO C 9x
2970 casinl NEW: ISO C 9x
2972 catanf NEW: ISO C 9x
2973 catanh NEW: ISO C 9x
2974 catanhf NEW: ISO C 9x
2975 catanhl NEW: ISO C 9x
2976 catanl NEW: ISO C 9x
2977 cbc_crypt NEW: Secure RPC
2981 ccoshf NEW: ISO C 9x
2982 ccoshl NEW: ISO C 9x
2988 cimagf NEW: ISO C 9x
2989 cimagl NEW: ISO C 9x
2990 clearerr_locked REMOVED
2991 clntunix_create NEW: sunrpc ext
2993 clog10 NEW: ISO C 9x
2994 clog10f NEW: ISO C 9x
2995 clog10l NEW: ISO C 9x
3005 cprojf NEW: ISO C 9x
3006 cprojl NEW: ISO C 9x
3008 crealf NEW: ISO C 9x
3009 creall NEW: ISO C 9x
3014 csinhf NEW: ISO C 9x
3015 csinhl NEW: ISO C 9x
3018 csqrtf NEW: ISO C 9x
3019 csqrtl NEW: ISO C 9x
3023 ctanhf NEW: ISO C 9x
3024 ctanhl NEW: ISO C 9x
3026 des_setparity NEW: Secure RPC
3027 ecb_crypt NEW: Secure RPC
3028 endutxent NEW: Unix98
3030 exp10f NEW: ISO C 9x
3031 exp10l NEW: ISO C 9x
3035 fattach NEW: STREAMS
3036 fdetach NEW: STREAMS
3040 feclearexcept NEW: ISO C 9x
3041 fegetenv NEW: ISO C 9x
3042 fegetexceptflag NEW: ISO C 9x
3043 fegetround NEW: ISO C 9x
3044 feholdexcept NEW: ISO C 9x
3046 feraiseexcept NEW: ISO C 9x
3047 ferror_locked REMOVED
3048 fesetenv NEW: ISO C 9x
3049 fesetexceptflag NEW: ISO C 9x
3050 fesetround NEW: ISO C 9x
3051 fetestexcept NEW: ISO C 9x
3052 feupdateenv NEW: ISO C 9x
3053 fflush_locked REMOVED
3057 fgets_unlocked NEW: GNU ext.
3058 fileno_locked REMOVED
3070 fputc_locked REMOVED
3071 fputs_unlocked NEW: GNU ext.
3072 fread_unlocked NEW: GNU ext.
3077 fstatvfs NEW: Unix98
3081 ftruncate64 NEW: LFS
3083 fwrite_unlocked NEW: GNU ext.
3084 gai_strerror NEW: IPv6
3088 getchar_locked REMOVED
3090 getdate_err NEW: Unix98
3091 getdate_r NEW: GNU ext.
3093 getnameinfo NEW: IPv6
3094 getnetname NEW: Secure RPC
3095 getpmsg NEW: STREAMS
3096 getpt NEW: Unix98 PTY
3097 getrlimit64 NEW: LFS
3098 getutxent NEW: Unix98
3099 getutxid NEW: Unix98
3100 getutxline NEW: Unix98
3101 glob64 NEW: GNU ext.
3102 globfree64 NEW: GNU ext.
3103 gnu_get_libc_release NEW: GNU ext.
3104 gnu_get_libc_version NEW: GNU ext.
3105 grantpt NEW: Unix98 PTY
3106 host2netname NEW: Secure RPC
3108 iconv_close NEW: iconv
3109 iconv_open NEW: iconv
3110 if_freenameindex NEW: IPv6
3111 if_indextoname NEW: IPv6
3112 if_nameindex NEW: IPv6
3113 if_nametoindex NEW: IPv6
3114 in6addr_any NEW: IPv6
3115 in6addr_loopback NEW: IPv6
3116 inet6_isipv4mapped NEW: IPv6
3117 isastream NEW: STREAMS
3118 iswblank NEW: GNU ext.
3119 key_decryptsession NEW: Secure RPC
3120 key_decryptsession_pk NEW: Secure RPC
3121 key_encryptsession NEW: Secure RPC
3122 key_encryptsession_pk NEW: Secure RPC
3123 key_gendes NEW: Secure RPC
3124 key_get_conv NEW: Secure RPC
3125 key_secretkey_is_set NEW: Secure RPC
3126 key_setnet NEW: Secure RPC
3127 key_setsecret NEW: Secure RPC
3128 llrint NEW: ISO C 9x
3129 llrintf NEW: ISO C 9x
3130 llrintl NEW: ISO C 9x
3131 llround NEW: ISO C 9x
3132 llroundf NEW: ISO C 9x
3133 llroundl NEW: ISO C 9x
3138 lrintf NEW: ISO C 9x
3139 lrintl NEW: ISO C 9x
3140 lround NEW: ISO C 9x
3141 lroundf NEW: ISO C 9x
3142 lroundl NEW: ISO C 9x
3144 makecontext NEW: Unix98
3145 mempcpy NEW: GNU ext.
3148 modify_ldt NEW: kernel
3152 nearbyint NEW: ISO C 9x
3153 nearbyintf NEW: ISO C 9x
3154 nearbyintl NEW: ISO C 9x
3155 netname2host NEW: Secure RPC
3156 netname2user NEW: Secure RPC
3157 nexttoward NEW: ISO C 9x
3158 nexttowardf NEW: ISO C 9x
3159 nexttowardl NEW: ISO C 9x
3163 passwd2des NEW: Secure RPC
3165 pow10f NEW: GNU ext.
3166 pow10l NEW: GNU ext.
3169 printf_size NEW: GNU ext.
3170 printf_size_info NEW: GNU ext.
3171 profil_counter REMOVED
3172 pthread_mutexattr_getkind_np REPLACED
3173 pthread_mutexattr_setkind_np REPLACED
3174 ptsname NEW: Unix98 PTY
3175 ptsname_r NEW: Unix98 PTY
3177 putchar_locked REMOVED
3178 putgrent NEW: GNU ext.
3180 putpmsg NEW: STREAMS
3181 pututxline NEW: Unix98
3185 readdir64_r NEW: LFS
3186 remquo NEW: ISO C 9x
3187 remquof NEW: ISO C 9x
3188 remquol NEW: ISO C 9x
3190 roundf NEW: ISO C 9x
3191 roundl NEW: ISO C 9x
3193 scalbln NEW: ISO C 9x
3194 scalblnf NEW: ISO C 9x
3195 scalblnl NEW: ISO C 9x
3197 sendfile NEW: kernel
3198 setcontext NEW: Unix98
3199 setrlimit64 NEW: LFS
3200 setutxent NEW: Unix98
3202 sigignore NEW: Unix98
3203 sigqueue NEW: POSIX.1b
3204 sigrelse NEW: Unix98
3205 sigset NEW: POSIX.1b
3206 sigtimedwait NEW: POSIX.1b
3207 sigwaitinfo NEW: POSIX.1b
3208 sincos NEW: GNU ext.
3209 sincosf NEW: GNU ext.
3210 sincosl NEW: GNU ext.
3214 strcasestr NEW: GNU ext.
3215 strtoimax NEW: ISO C 9x
3216 strtoumax NEW: ISO C 9x
3217 strverscmp NEW: GNU ext.
3218 svcauthdes_stats NEW: Secure RPC
3219 svcunix_create NEW: sunrpc etx
3220 svcunixfd_create NEW: sunrpc ext
3221 swapcontext NEW: Unix98
3222 tcgetsid NEW: Unix98 PTY
3223 tdestroy NEW: GNU ext.
3224 tgamma NEW: ISO C 9x
3225 tgammaf NEW: ISO C 9x
3226 tgammal NEW: ISO C 9x
3230 truncf NEW: ISO C 9x
3231 truncl NEW: ISO C 9x
3233 unlockpt NEW: Unix98 PTY
3234 updwtmpx NEW: Unix98
3235 user2netname NEW: Secure RPC
3236 utmpxname NEW: Unix98
3237 versionsort NEW: GNU ext.
3238 versionsort64 NEW: GNU ext.
3240 wcscasecmp NEW: GNU ext.
3241 wcsncasecmp NEW: GNU ext.
3242 wcsnlen NEW: GNU ext.
3243 wcstoimax NEW: ISO C 9x
3244 wcstoll NEW: ISO C 9x
3245 wcstoull NEW: ISO C 9x
3246 wcstoumax NEW: ISO C 9x
3248 wordexp NEW: POSIX.2
3249 wordfree NEW: POSIX.2
3250 write_profiling REMOVED
3251 xdecrypt NEW: Secure RPC
3252 xdr_authdes_cred NEW: Secure RPC
3253 xdr_authdes_verf NEW: Secure RPC
3254 xdr_cryptkeyarg NEW: Secure RPC
3255 xdr_cryptkeyarg2 NEW: Secure RPC
3256 xdr_cryptkeyres NEW: Secure RPC
3257 xdr_getcredres NEW: Secure RPC
3258 xdr_key_netstarg NEW: Secure RPC
3259 xdr_key_netstres NEW: Secure RPC
3260 xdr_keybuf NEW: Secure RPC
3261 xdr_keystatus NEW: Secure RPC
3262 xdr_netnamestr NEW: Secure RPC
3263 xdr_sizeof NEW: Secure RPC
3264 xdr_unixcred NEW: sunrpc ext
3265 xencrypt NEW: Secure RPC
3266 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3277 * inet_ntoa is thread-safe
3279 * updwtmp is moved from libutil to libc
3281 * rewrite of cbrt function
3283 * update of timezone data
3297 * add atoll function
3299 * fix complex problems in Berkeley DB code
3301 * fix math functions
3305 * fixed lots of header problems (especially Linux/GNU specific)
3307 * dynamic loader preserves all registers
3309 * Roland McGrath provided support for handling of auxiliary objects in
3310 the ELF dynamic loader.
3312 * support for parallel builds is improved
3316 * GNU extensions are no longer declared by default. To enable them you
3317 must define the macro `_GNU_SOURCE' in your program or compile with
3320 * The library has changed from using GNU ld symbol aliases to using weak
3321 symbols where available. The ELF object file format supports weak
3322 symbols; GNU ld also supports weak symbols in the a.out format. (There
3323 is also now support for other GNU ld extensions in ELF. Use the
3324 `--with-elf' option to configure to indicate you have ELF, and
3325 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3326 of many files which contained only symbol aliases, reducing the size of
3327 the source and the compiled library; many other files were renamed to
3328 less cryptic names previously occupied by the symbol alias files.
3329 There is a new header file <elf.h> for programs which operate on
3330 files in the ELF format.
3332 * Converted to Autoconf version 2, so `configure' has more options.
3333 Run `configure --help' to see the details.
3335 * The library can now be configured to build profiling, highly-optimized
3336 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3337 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3338 `configure' enable building these extra libraries. The shared library is
3339 built by default when using both ELF and GNU ld. When shared libraries
3340 are enabled, the new library `-ldl' is available for arbitrary run-time
3341 loading of shared objects; its interface is defined in <dlfcn.h>. The
3342 new header file <link.h> gives access to the internals of the run-time
3343 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3344 application of same name on other systems and it provides information
3345 about dynamically linked binaries.
3347 * The C library now provides the run-time support code for profiling
3348 executables compiled with `-pg'. Programs can control the profiling code
3349 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3350 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3351 the support for this file format was contributed by David Mosberger-Tang.
3353 * The math code has been replaced with a math library based on fdlibm from
3354 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3355 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3356 functions. The math functions now reside in a separate library, so
3357 programs using them will need to use `-lm' their linking commands.
3359 * John C. Bowman contributed optimized ix87 assembler inline functions.
3361 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3362 mechanism similar to that found in Solaris 2. This is now used for the
3363 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3364 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3365 file controls what services are used for each individual database. This
3366 works by loading shared libraries with names specified in `nsswitch.conf',
3367 so service modules can be changed or added at any time without even
3368 relinking any program. Currently there are the file, db, and NIS based
3369 NSS services available.
3371 * The new functions `strtoq' and `strtouq' parse integer values from
3372 strings, like `strtol' and `strtoul', but they return `long long int' and
3373 `unsigned long long int' values, respectively (64-bit quantities).
3375 * The new functions `strtof' and `strtold' parse floating-point values from
3376 strings, like `strtod', but they return `float' and `long double' values,
3377 respectively (on some machines `double' and `long double' are the same).
3379 * Ulrich Drepper has contributed new implementations of the floating-point
3380 printing and reading code used in the `printf' family of functions and
3381 `strtod', `strtof', and `strtold'. These new functions are perfectly
3382 accurate, and much faster than the old ones.
3384 * The implementation of the POSIX locale model was completely rewritten by
3385 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3386 to compile the POSIX locale definition.
3388 * The former dummy implementations of the strcoll and strxfrm function are
3389 now replaced by fully functional code contributed by Ulrich Drepper. The
3390 collation information comes from the POSIX locale definitions.
3392 * The new header <langinfo.h> defines an interface for accessing
3393 various locale-dependent data (using the locale chosen with `setlocale').
3395 * Ulrich Drepper has contributed a new suite of functions for operation on
3396 wide-character and multibyte-character strings, in <wchar.h>;
3397 and classification and case conversion of wide characters, in <wctype.h>.
3398 These new functions are conforming to the ISO C, Amendement 1 specification.
3400 * There is now a second implementation of the standard I/O library available.
3401 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3402 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3404 * You can now use positional parameter specifications in format strings
3405 for the `printf' and `scanf' families of functions. For example,
3406 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3407 ``Number 6, Mr Jones''. This is mainly useful when providing different
3408 format strings for different languages, whose grammars may dictate
3409 different orderings of the values being printed. To support this
3410 feature, the interface for `register_printf_handler' has changed; see
3411 the header file <printf.h> for details.
3413 * The `printf' and `scanf' families of functions now understand a new
3414 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3415 to group numbers as indicated by the locale; for `scanf' and friends, this
3416 says to accept as valid only a number with all the proper grouping
3417 separators in the right places. In the default "C" locale, numbers are
3418 not grouped; but locales for specific countries will define the usual
3419 conventions (i.e. separate thousands with `,' in the US locale).
3421 * The pgrp functions have been regularized, slightly incompatibly but much
3422 less confusingly. The core functions are now `getpgid' and `setpgid',
3423 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3424 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3425 provided for compatibility. There is no longer an incompatible `getpgrp'
3426 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3428 * The new header file <fts.h> and suite of functions simplify programs that
3429 operate on directory trees. This code comes from 4.4 BSD.
3431 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3432 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3433 NSS scheme used in glibc.
3435 * There is a new malloc debugging hook `__memalign_hook'.
3437 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3438 `unsigned int' in <sys/types.h>. These are for compatibility only and
3439 their use is discouraged.
3441 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3442 done differently, so that it works even without GNU ld.
3444 * New function `euidaccess' checks allowed access to a file like `access',
3445 but using the effective IDs instead of the real IDs.
3447 * The time zone data files have been updated for the latest and greatest
3448 local time conventions of the countries of the world.
3450 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3453 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3454 for formatting floating-point numbers. They are provided only for
3455 compatibility; new programs should use `sprintf' instead. There are
3456 also equivalent function for the `long double' floating-point type and
3457 all functions also exist in a reentrant form.
3459 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3460 functions for maintaining the login-record files (primarily of use to
3461 system programs such as `login'), and convenient functions for
3462 allocating and initializing a pseudo-terminal (pty) device.
3464 * Ulrich Drepper has contributed new support for System V style
3465 shared memory and IPC on systems that support it.
3467 * Ulrich Drepper has contributed several miscellaneous new functions found
3468 in System V: The `hsearch' family of functions provide an effective
3469 implementation of hash tables; `a64l' and `l64a' provide a very simple
3470 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3473 * Ulrich Drepper has contributed new reentrant counterparts for the
3474 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3476 * Ulrich Drepper has contributed new, highly-optimized versions of several
3477 string functions for the i486/Pentium family of processors.
3479 * Ulrich Drepper has updated the Linux-specific code, based largely
3480 on work done in Hongjiu Lu's version of GNU libc for Linux.
3481 The GNU library now supports Linux versions 2.0.10 and later,
3482 using the ELF object file format (i[3456]86-*-linux).
3484 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3486 * David Mosberger-Tang and Richard Henderson have ported the C library
3487 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3488 dynamic linking support for ELF/Alpha.
3490 * Richard Henderson contributed several Alpha optimized assembler function
3491 for arithmetic and string handling.
3493 * Ulrich Drepper has contributed a new set of message catalog functions to
3494 support multiple languages using the <libintl.h> interface, for use with
3495 his new package GNU gettext. Translation volunteers have contributed
3496 catalogs of the library's messages in Spanish, German, and Korean.
3498 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3499 program and the `catgets' function for reading the catalog files it
3500 creates. (The <libintl.h> interface is preferred; we include the
3501 <nl_types.h> interface using `catgets' only for source compatibility with
3502 programs already written to use it.)
3504 * New header file <values.h> gives SVID-compatible names for <limits.h>
3507 * Various new macros, declarations, and small header files for compatibility
3510 * New function `group_member' is a convenient way to check if a process has
3511 a given effective group ID.
3513 * When using GCC 2.7 and later, the socket functions are now declared in a
3514 special way so that passing an argument of type `struct sockaddr_in *',
3515 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3516 `struct sockaddr *' type, does not generate a type-clash warning.
3518 * New function `error' declared in header file <error.h> is a convenient
3519 function for printing error messages and optionally exiting; this is the
3520 canonical function used in GNU programs. The new functions `err', `warn',
3521 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3522 doing the same thing.
3524 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3525 POSIX.2 `glob' function to do ~ and {...} expansion.
3527 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3528 `clearenv' which is used in POSIX.9 is also available.
3530 * New function `getsid' returns session ID number on systems that support it.
3532 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3533 files <db.h> and <mpool.h> provide a rich set of functions for several
3534 types of simple databases stored in memory and in files, and <ndbm.h> is
3535 an old `ndbm'-compatible interface using the `db' functions. Link with
3536 `-ldb' to get these functions.
3538 * New macro `strdupa' copies a string like `strdup', but uses local stack
3539 space from `alloca' instead of dynamic heap space from `malloc'.
3541 * New function `strnlen' is like `strlen' but searches only a given maximum
3542 number of characters for the null terminator. `stpncpy', `strndup' and
3543 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3546 * New function `statfs' in header <sys/statfs.h>.
3548 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3549 provide convenient functions for operating on blocks of null-terminated
3552 * A new suite of functions in <utmp.h> handle all the details of reading
3553 and writing the utmp file.
3555 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3558 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3559 completely POSIX compliant and also implemented the extended functionality
3560 to handle alternate digit representation and alternate era date formats.
3562 * Ulrich Drepper provided an implementation of the `strptime' function
3563 defined in XPG4.2 which transforms a string into a `struct tm' value.
3565 * Paul Eggert provided the tzselect shell script as part of the timezone
3566 code. The shell script makes it easy to select the correct timezone
3569 * The implementation of the malloc family of functions is completely replaced
3570 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3571 The implementation uses the mmap function (if available) and it is
3572 optimized for the use in multi threaded programs.
3574 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3575 functions. This new functionality is usable by specifying a special
3576 salt string and it is compatible with implementation on *BSD systems.
3578 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3579 `getsubopt' to handle second level command line options, `bsd_signal'
3580 to access BSD style `signal' functionality, the obsolete `regexp' style
3583 * the `lchown' function is available on system which support this
3586 * The implementation of the shadow password handling function was contributed
3589 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3591 * POSIX.1g support was added. The <sys/select.h> header is available,
3592 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3593 implementation of `getaddrinfo'.
3597 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3599 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3600 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3602 * The new function `daemon' from 4.4 BSD is useful for server programs that
3603 want to put themselves in the background.
3605 * Joel Sherrill has contributed support for several standalone boards that
3606 run without an operating system.
3608 * `printf', `scanf' and friends now accept a `q' type modifier for long
3609 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3611 * All of the code taken from BSD (notably most of the math and networking
3612 routines) has been updated from the BSD 4.4-Lite release.
3614 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3616 * The new functions `getdomainname' and `setdomainname' fetch or change the
3617 YP/NIS domain name. These are system calls which exist on systems which
3620 * The time zone data files have been updated for the latest international
3623 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3624 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3628 * The C library now includes support for Sun RPC, from Sun's free
3629 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3630 are included. (There is still no support for YP.)
3632 * Tom Quinn has contributed a port of the C library to SGI machines running
3633 Irix 4 (mips-sgi-irix4).
3635 * The new `lockf' function is a simplified interface to the locking
3636 facilities of `fcntl', included for compatibility.
3638 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3640 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3642 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3645 * The error message printed by `assert' for a failed assertion now includes
3646 the name of the program (if using GNU ld) and the name of the calling
3647 function (with versions of GCC that support this).
3649 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3651 * The library now includes the <sys/mman.h> header file and memory
3652 management functions `mmap', `munmap', `mprotect', `msync', and
3653 `madvise', on systems that support those facilities.
3655 * The interface for `mcheck' has changed slightly: the function called to
3656 abort the program when an allocation inconsistency is detected now takes
3657 an argument that indicates the type of failure. The new function
3658 `mprobe' lets you request a consistency check for a particular block at
3659 any time (checks are normally done only when you call `free' or `realloc'
3662 * It is now possible to easily cross-compile the C library, building on one
3663 system a library to run on another machine and/or operating system. All
3664 you need to do is set the variable `HOST_CC' in `configparms' to the
3665 native compiler for programs to run on the machine you are building on (a
3666 few generator programs are used on Unix systems); set `CC' to the
3669 * The new function `fexecve' (only implemented on the GNU system) executes
3670 a program file given a file descriptor already open on the file.
3674 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3675 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3677 * You can set the variable `libprefix' in `configparms' to specify a prefix
3678 to be prepended to installed library files; this makes it easy to install
3679 the GNU C library to be linked as `-lgnuc' or whatever.
3681 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3682 copies a limited number of characters from a string, and returns the
3683 address of the last character written.
3685 * You no longer need to check for whether the installed `stddef.h' is
3686 compatible with the GNU C library. configure now checks for you.
3688 * You can now define a per-stream `fileno' function to convert the
3689 stream's cookie into an integral file descriptor.
3691 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3692 allocates zero bytes of storage, and returns a unique pointer which
3693 you can pass to `realloc' or `free'. The behavior is undefined if
3694 you dereference this pointer.
3696 * The C library now runs on Sony NEWS m68k machines running either
3697 NewsOS 3 or NewsOS 4.
3699 * The new `syscall' function is a system-dependent primitive function
3700 for invoking system calls. It has the canonical behavior on Unix
3701 systems, including unreliable return values for some calls (such as
3702 `pipe', `fork' and `getppid').
3704 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3705 to `EAGAIN', which is the preferred name. On systems whose kernels
3706 use two distinct codes, the C library now translates EWOULDBLOCK to
3707 EAGAIN in every system call function.
3711 * The GNU C Library Reference Manual is now distributed with the library.
3712 `make dvi' will produce a DVI file of the printed manual.
3713 `make info' will produce Info files that you can read on line using C-h i
3714 in Emacs or the `info' program.
3715 Please send comments on the manual to bug-glibc-manual@gnu.org.
3717 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3719 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3721 * Jason Merrill has contributed a port to the Sequent Symmetry running
3722 Dynix version 3 (i386-sequent-dynix).
3724 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3725 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3727 * New function `memory_warnings' lets you arrange to get warnings when
3728 malloc is running out of memory to allocate, like Emacs gives you.
3730 * The C library now contains the relocating allocator used in Emacs 19 for
3731 its editing buffers. This allocator (ralloc) minimizes allocation
3732 overhead and fragmentation by moving allocated regions around whenever it
3733 needs to. You always refer to a ralloc'd region with a "handle" (a
3734 pointer to a pointer--an object of type `void **').
3736 * There is a new `printf' format: `%m' gives you the string corresponding
3737 to the error code in `errno'.
3739 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3740 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3741 pass, the `a' modifier says to fill in a `char **' you pass with a
3744 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3745 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3746 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3748 * `mkstemp' is a traditional Unix function to atomically create and open a
3749 uniquely-named temporary file.
3753 * The standard location for the file that says what the local timezone is
3754 has changed again. It is now `/usr/local/etc/localtime' (or more
3755 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3757 * The distribution no longer contains any files with names longer than 14
3760 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3761 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3763 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3767 * The configuration process has changed quite a bit. The `configure'
3768 script is now used just like the configuration scripts for other GNU
3769 packages. The `sysdeps' directory hierarchy is much rearranged.
3770 The file `INSTALL' explains the new scheme in detail.
3772 * The header files no longer need to be processed into ANSI C and
3773 traditional C versions. There is just one set of files to install, and
3774 it will work with ANSI or old C compilers (including `gcc -traditional').
3776 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3777 MIPS DECStation running Ultrix 4.
3779 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3780 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3781 made itself into a shared library.
3783 * Yet further improved support for the i386, running 4.3 BSD-like systems
3784 (such as Mach 3 with the Unix single-server), or System V.
3786 * New function `strncasecmp' to do case-insensitive string comparison
3787 with limited length.
3789 * New function `strsep' is a reentrant alternative to `strtok'.
3791 * New functions `scandir' and `alphasort' for searching directories.
3793 * New function `setenv' is a better interface to `putenv'.
3795 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3796 function for traversing a directory tree.
3798 * The GNU obstack package is now also part of the C library.
3799 The new function `open_obstack_stream' creates a stdio stream that
3800 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3801 formatted output directly to an obstack.
3803 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3804 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3806 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3808 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3809 things to your strings.
3811 * There are some new test programs: test-fseek, testmb, and testrand.
3813 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3814 These ports are not finished, but are a good starting place for really
3815 supporting those systems.
3817 * `/etc/localtime' is now the standard location for the file that says what
3818 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3819 This follows the general principle that `/etc' is the place for all local
3820 configuration files.
3822 * The C library header files now use `extern "C"' when used by the C++
3823 compiler, so the C library should now work with C++ code.
3825 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3826 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3829 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3830 function which uses the `merge sort' algorithm, and is said to be
3831 significantly faster than the old GNU `qsort' function. Merge sort is
3832 now the standard `qsort' function. The new algorithm can require a lot
3833 of temporary storage; so, the old sorting function is called when the
3834 required storage is not available.
3836 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3837 provides the Unix `crypt' function, plus some other entry points.
3839 * The code and header files taken from 4.4 BSD have been updated with the
3840 latest files released from Berkeley.
3842 ----------------------------------------------------------------------
3843 Copyright information:
3845 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3847 Permission is granted to anyone to make or distribute verbatim copies
3848 of this document as received, in any medium, provided that the
3849 copyright notice and this permission notice are preserved,
3850 thus giving the recipient permission to redistribute in turn.
3852 Permission is granted to distribute modified versions
3853 of this document, or of portions of it,
3854 under the above conditions, provided also that they
3855 carry prominent notices stating who last changed them.
3858 version-control: never