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 * The rpcgen, librpcsvc and related headers will only be built and
15 installed when glibc is configured with --enable-obsolete-rpc.
16 This allows alternative RPC implementations, like TIRPC, to be used
17 by default. Applications needing features missing from TIRPC should
18 consider the rpcsvc-proto project developed by Thorsten Kukuk (SUSE).
20 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
21 libnss_compat, are deprecated, and will not be built or installed by
22 default. Replacement implementations based on TIRPC, which
23 additionally support IPv6, are available from
24 <https://github.com/thkukuk/libnss_{compat,nis,nisplus}>.
26 * The NIS(+) support library, libnsl, is deprecated. By default, a
27 compatibility shared library will be built and installed, but not
28 headers or development libraries.
30 Only a few NIS-related programs require this library.
31 A replacement implementation based on TIRPC is available from
32 <https://github.com/thkukuk/libnsl>. Like the replacement NIS(+)
33 name service modules, the replacement supports IPv6, and it can be
34 coinstalled with the compatibility shared library from glibc.
36 * New configure option --enable-obsolete-nsl will cause libnsl's
37 headers, and the NIS(+) name service modules, to be built and
38 installed. This option may be removed in a future release.
40 * Extensive new collation tests for Hungarian locales based on
41 "The Rules of Hungarian Orthography, 12th edition" and the work of
42 Egmont Koblinger (Bug 18934).
44 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
45 support is enabled, the configured recursive resolver must support EDNS.
46 (Responding to EDNS-enabled queries with responses which are not
47 EDNS-enabled is fine, but FORMERR responses are not.)
49 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
50 servers have not supported this opcode for a long time.
52 * The legacy cfree function has been removed. Applications should use the
53 free function instead.
55 Security related changes:
57 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
58 to avoid fragmentation-based spoofing attacks.
60 The following bugs are resolved with this release:
62 [The release manager will add the list generated by
63 scripts/list-fixed-bugs.py just before the release.]
68 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
69 24731-2:2010, is supported to enable declarations of functions from that
70 TR. Note that not all functions from that TR are supported by the GNU C
73 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
74 18661-1:2014, is supported to enable declarations of functions and macros
75 from that TS. Note that not all features from that TS are supported by
78 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
79 18661-4:2015, is supported to enable declarations of functions and macros
80 from that TS. Note that most features from that TS are not supported by
83 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
84 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
85 Since the GNU C Library defaults to a much newer revision of POSIX, this
86 will only affect programs that specifically request an old conformance
87 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
88 see a change in the visible declarations, but a program compiled with
89 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
92 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
93 defined by all multithreaded code, but glibc has not required this for
96 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
97 means that in a future release, the macros “major”, “minor”, and “makedev”
98 will only be available from <sys/sysmacros.h>.
100 These macros are not part of POSIX nor XSI, and their names frequently
101 collide with user code; see for instance glibc bug 19239 and Red Hat bug
102 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
103 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
106 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
107 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
108 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
110 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
111 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
112 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
113 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
114 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
115 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
116 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
117 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
118 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
119 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
120 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
123 * New <math.h> features are added from TS 18661-1:2014:
125 - Signaling NaN macros: SNANF, SNAN, SNANL.
127 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
128 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
129 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
131 - llogb functions: the llogb, llogbf and llogbl functions, and the
132 FP_LLOGB0 and FP_LLOGBNAN macros.
134 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
137 - Comparison macros: iseqsig.
139 - Classification macros: iscanonical, issubnormal, iszero.
141 - Total order functions: totalorder, totalorderf, totalorderl,
142 totalordermag, totalordermagf, totalordermagl.
144 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
146 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
147 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
149 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
150 are added to libc. They convert a floating-point number into string.
152 * Most of glibc can now be built with the stack smashing protector enabled.
153 It is recommended to build glibc with --enable-stack-protector=strong.
154 Implemented by Nick Alcock (Oracle).
156 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
157 intended to be used instead of memset() to erase sensitive data after use;
158 the compiler will not optimize out calls to explicit_bzero even if they
159 are "unnecessary" (in the sense that no _correct_ program can observe the
160 effects of the memory clear).
162 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
163 to float instead of double. This does not affect the ABI of any libraries
164 that are part of the GNU C Library, but may affect the ABI of other
165 libraries that use this type in their interfaces.
167 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
168 float_t and double_t types are now defined to long double instead of float
169 and double. These options are not the default, and this does not affect
170 the ABI of any libraries that are part of the GNU C Library, but it may
171 affect the ABI of other libraries that use this type in their interfaces,
172 if they are compiled or used with those options.
174 * The getentropy and getrandom functions, and the <sys/random.h> header file
177 * The buffer size for byte-oriented stdio streams is now limited to 8192
178 bytes by default. Previously, on Linux, the default buffer size on most
179 file systems was 4096 bytes (and thus remains unchanged), except on
180 network file systems, where the buffer size was unpredictable and could be
181 as large as several megabytes.
183 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
184 for the Linux quota interface which predates kernel version 2.4.22 has
187 * The malloc_get_state and malloc_set_state functions have been removed.
188 Already-existing binaries that dynamically link to these functions will
189 get a hidden implementation in which malloc_get_state is a stub. As far
190 as we know, these functions are used only by GNU Emacs and this change
191 will not adversely affect already-built Emacs executables. Any undumped
192 Emacs executables, which normally exist only during an Emacs build, should
193 be rebuilt by re-running “./configure; make” in the Emacs build tree.
195 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
196 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
197 “no-ip6-dotint” had already been the default, and support for the
198 “ip6-dotint” option was removed from the Internet in 2006.
200 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
201 flag from <resolv.h> have been removed. The option relied on a
202 backwards-incompatible DNS extension which was never deployed on the
205 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
206 RES_BLAST defined in the <resolv.h> header file have been deprecated.
207 They were already unimplemented.
209 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
210 _res.flags are deprecated. The flag was standardized in RFC 2133, but
211 removed again from the IETF name lookup interface specification in RFC
212 2553. Applications should use getaddrinfo instead.
214 * DNSSEC-related declarations and definitions have been removed from the
215 <arpa/nameser.h> header file, and libresolv will no longer attempt to
216 decode the data part of DNSSEC record types. Previous versions of glibc
217 only implemented minimal support for the previous version of DNSSEC, which
218 is incompatible with the currently deployed version.
220 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
221 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
222 <arpa/nameser.h> header file because the distinction between RR types and
223 meta-RR types is not officially standardized, subject to revision, and
224 thus not suitable for encoding in a macro.
226 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
227 and rhook members of the res_state type in <resolv.h> have been removed.
228 The glibc stub resolver did not support these hooks, but the header file
229 did not reflect that.
231 * For multi-arch support it is recommended to use a GCC which has
232 been built with support for GNU indirect functions. This ensures
233 that correct debugging information is generated for functions
234 selected by IFUNC resolvers. This support can either be enabled by
235 configuring GCC with '--enable-gnu-indirect-function', or by
236 enabling it by default by setting 'default_gnu_indirect_function'
237 variable for a particular architecture in the GCC source file
240 * GDB pretty printers have been added for mutex and condition variable
241 structures in POSIX Threads. When installed and loaded in gdb these pretty
242 printers show various pthread variables in human-readable form when read
243 using the 'print' or 'display' commands in gdb.
245 * Tunables feature added to allow tweaking of the runtime for an application
246 program. This feature can be enabled with the '--enable-tunables' configure
247 flag. The GNU C Library manual has details on usage and README.tunables has
248 instructions on adding new tunables to the library.
250 * A new version of condition variables functions have been implemented in
251 the NPTL implementation of POSIX Threads to provide stronger ordering
254 * A new version of pthread_rwlock functions have been implemented to use a more
255 scalable algorithm primarily through not using a critical section anymore to
258 Security related changes:
260 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
261 have been created with makecontext could fail to terminate due to a
262 missing .cantunwind annotation. This has been observed to lead to a hang
263 (denial of service) in some Go applications compiled with gccgo. Reported
264 by Andreas Schwab. (CVE-2016-6323)
266 * The DNS stub resolver functions would crash due to a NULL pointer
267 dereference when processing a query with a valid DNS question type which
268 was used internally in the implementation. The stub resolver now uses a
269 question type which is outside the range of valid question type values.
272 The following bugs are resolved with this release:
274 [4099] stdio: Overly agressive caching by stream i/o functions.
275 [7065] build: Support building glibc with -fstack-protector or -fstack-
277 [9842] localedata: en_CA: incorrect date format
278 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
279 before it started waiting
280 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
281 [15765] nptl: sem_open is wrongly a cancellation point
282 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
283 [16458] libc: endian.h and netinet/in.h byte order macros return values of
285 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
286 library linked with pthread
287 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
289 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
290 [17252] libc: getrandom and getentropy syscall
291 [17730] malloc: thread-local storage is sometimes improperly free()'d
292 after being __libc_memalign()'d
293 [18241] stdio: failed fseek on memstream does not set errno and can fail
295 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
297 [18463] nptl: pthread_cond_broadcast issue when surrounded by
298 PTHREAD_PRIO_INHERIT mutex on ARM
299 [18784] network: res_query and related function crash for special record
300 type queries (CVE-2015-5180)
301 [19380] math: strtod does not raise "inexact"
302 [19387] string: Integer overflow in memchr
303 [19390] string: Integer overflow in strncat
304 [19398] build: linknamespace tests fail with massively parallel build
305 [19402] nptl: Deadlock with robust shared mutex and asynchronous
307 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
308 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
309 [19514] libc: [PATCH] Fix spelling errors in spelling
310 "implement"/"implementation" in several places
311 [19582] network: Deprecate RES_USE_INET6
312 [19673] manual: clog10 docs appear to be erroneous
313 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
314 causes a segmentation fault
315 [19826] libc: invalid pointer returned from __tls_get_addr with static
317 [20016] network: resolv: Remove hooks support from the API
318 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
320 [20033] math: [x86_64] vectorized math function don't call the __finite
322 [20116] nptl: use after free in pthread_create
323 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
324 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
326 [20311] nptl: please install proc_service.h
327 [20366] libc: Compilation errors in installed headers in strict-compliance
329 [20370] malloc: malloc: Arena free list management is still racy
330 (incorrect fix in bug 19243)
331 [20386] libc: assert (X = 0) does not result in GCC warning
332 [20432] malloc: malloc: Minimize interface required for interposition
333 [20435] libc: Missing unwind info in __startcontext causes infinite loop
334 in _Unwind_Backtrace (CVE-2016-6323)
335 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
336 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
338 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
339 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
340 [20477] network: resolv: incorrect double-checked locking related to
342 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
344 [20495] math: x86_64 performance degradation due to AVX/SSE transition
346 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
347 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
348 cause transition penalty
349 [20517] math: sparcv9 missing fdiml compat symbol
350 [20524] manual: strverscmp is inconsistent
351 [20525] libc: <sys/quota.h> should be based on kernel headers
352 [20539] math: GCC 7 -static -lm fails to link at -O3
353 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
355 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
357 [20591] network: Remove obsolete DNSSEC support
358 [20592] network: DNS resource record type classification macros in
359 <arpa/nameser.h> are incorrect
360 [20593] network: Update DNS RR type definitions
361 [20611] network: getaddrinfo accepts invalid numeric scope IDs
362 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
364 [20629] network: libresolv: Remove support for bitstring labels
366 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
367 assertion: "maxidx >= 2"
368 [20660] math: [arm] Use VSQRT
369 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
370 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
371 [20689] libc: Test for FMA should also check for AVX.
372 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
373 [20715] math: iszero macro breaks existing code
374 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
375 [20728] libc: powerpc: Missing TOC stub in clone
376 [20729] build: glibc-2.24 fails to build for i486 with -Os
377 [20750] build: Build fails with default PIE enabled gcc-6.x
378 [20768] math: [alpha] sqrt fegetenv namespace
379 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
381 [20787] math: float_t is defined as float incorrectly on x86_64 even if
382 __FLT_EVAL_METHOD__ is 2
383 [20790] network: rpcgen buffer overrun in get_prog_declaration
384 [20822] nptl: race condition in __lll_unlock_elision on powerpc
385 [20829] libc: crypt snprintf namespace
386 [20847] libc: tst-vfork3 failure
387 [20855] math: Default bits/mathdef.h has inappropriate float_t
388 [20859] math: [sh4] FP_ILOGB0 invalid
389 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
390 U+20AC), not same as GBK
391 [20915] dynamic-link: global-dynamic TLS broken on aarch64
392 [20916] math: pow handling of sNaN arguments
393 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
394 [20919] math: Bad pow (qNaN, 0) result with -lieee
395 [20940] math: hypot sNaN handling
396 [20947] math: fmax, fmin sNaN handling
397 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
398 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
400 [20971] string: powerpc64/power7 memchr overflows internal pointer check
401 [20973] nptl: robust mutexes: Lost wake-ups
402 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
403 [20978] nis: strlen on null pointer in nss_nisplus
404 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
406 [21014] string: i686 memchr overflows internal pointer check
407 [21019] libc: [mips] n32 lseek incorrect on overflow
408 [21022] libc: [microblaze] __backtrace get_frame_size namespace
409 [21026] libc: [MIPS] readahead syscall is broken on n64
410 [21028] math: Fallback fesetexceptflag should always succeed
411 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
412 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
414 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
415 [21061] librt: [microblaze] librt lost clock_* exports
416 [21073] libc: tunables: insecure environment variables passed to
417 subprocesses with AT_SECURE
418 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
422 * The minimum Linux kernel version that this version of the GNU C Library
423 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
424 kernel version 2.6.32 or later suffices (on architectures that already
425 required kernel versions more recent than 3.2, those requirements remain
426 unchanged). Linux 3.2 or later kernel headers are required on all
429 * The pap_AN locale has been deleted. This has been deprecated for a long
430 time. It has been replaced by pap_AW & pap_CW, both of which have long
431 been included in previous releases.
433 * The readdir_r and readdir64_r functions have been deprecated. It is
434 recommended to use readdir and readdir64 instead.
436 * The type “union wait” has been removed. It was deprecated in the early
437 1990s and never part of POSIX. Application code should use the int type
438 instead of “union wait”.
440 * A new NSS action is added to facilitate large distributed system
441 administration. The action, MERGE, allows remote user stores like LDAP
442 to be merged into local user stores like /etc/groups in order to provide
443 easy to use, updated, and managed sets of merged credentials. The new
444 action can be used by configuring it in /etc/nsswitch.conf:
445 group: files [SUCCESS=merge] nis
446 Implemented by Stephen Gallagher (Red Hat).
448 * The deprecated __malloc_initialize_hook variable has been removed from the
451 * The long unused localedef --old-style option has been removed. It hasn't
452 done anything in over 16 years. Scripts using this option can safely
455 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
456 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
457 return the next representable value in the direction of positive infinity
458 and the nextdown functions return the next representable value in the
459 direction of negative infinity. These are currently enabled as GNU
462 Security related changes:
464 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
465 could result in a stack overflow when getnetbyname was called with an
466 overly long name. (CVE-2016-3075)
468 * Previously, getaddrinfo copied large amounts of address data to the stack,
469 even after the fix for CVE-2013-4458 has been applied, potentially
470 resulting in a stack overflow. getaddrinfo now uses a heap allocation
471 instead. Reported by Michael Petlan. (CVE-2016-3706)
473 * The glob function suffered from a stack-based buffer overflow when it was
474 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
475 Reported by Alexander Cherepanov. (CVE-2016-1234)
477 * The Sun RPC UDP client could exhaust all available stack space when
478 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
479 alloca plugin for GCC. (CVE-2016-4429)
481 * The IPv6 name server management code in libresolv could result in a memory
482 leak for each thread which is created, performs a failing naming lookup,
483 and exits. Over time, this could result in a denial of service due to
484 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
486 The following bugs are resolved with this release:
488 [1170] localedata: ne_NP: update Nepali locale definition file
489 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
491 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
492 [6796] math: fdim() does not set errno on overflow
493 [10354] libc: posix_spawn should use vfork() in more cases than presently
494 [11213] localedata: localedata: add copyright disclaimer to locale files
495 [12143] localedata: chr_US: new Cherokee locale
496 [12450] localedata: sgs_LT: new locale
497 [12676] localedata: ln_CD: new locale
498 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
500 [13304] math: fma, fmaf, fmal produce wrong results
501 [14259] build: --localedir arg to configure is ignored
502 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
503 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
504 [14934] localedata: es_CL: wrong first weekday chilean locale
505 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
507 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
509 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
510 [15368] nptl: raise() is not async-signal-safe
511 [15479] math: ceil, floor, round and trunc raise inexact exception
512 [15578] localedata: kk_KZ: various updates
513 [16003] localedata: pap_AN: punt old locale
514 [16137] localedata: iw_IL: punt old locale
515 [16190] localedata: eo: new esperanto locale
516 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
517 [16742] malloc: race condition: pthread_atfork() called before first
518 malloc() results in unexpected locking behaviour/deadlocks
519 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
521 [16983] localedata: postal_fmt does not allow %l and %n modifiers
522 [17565] localedata: pt_PT: wrong (work-)week start
523 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
524 [17950] build: Build fails with -msse
525 [18205] localedata: be_BY*: wrong first_weekday and first_workday
526 [18433] libc: posix_spawn does not return correctly upon failure to
528 [18453] localedata: charmaps/IBM875: incorrect codes
529 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
531 [18896] localedata: he_IL: improvements for currency
532 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
534 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
536 [19056] libc: Deprecate readdir_r
537 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
539 [19198] localedata: nl_NL: small improvements for Dutch locales
540 [19257] network: Per-thread memory leak in __res_vinit with IPv6
541 nameservers (CVE-2016-5417)
542 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
544 [19400] locale: Language missing in "iso-639.def", trivial fix in
546 [19431] malloc: Deadlock between fflush, getdelim, and fork
547 [19505] libc: Incorrect file descriptor validity checks in
548 posix_spawn_file_actions_add{open,close,dup2}
549 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
551 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
552 `intl/{gettextP,loadinfo}.h'
553 [19534] libc: execle, execlp may use malloc
554 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
555 [19573] network: res_nclose and __res_maybe_init disagree about name
556 server initialization, breaking Hesiod
557 [19575] localedata: Status of GB18030 tables
558 [19581] localedata: sr_* date_fmt string contains additional newline
559 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
561 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
562 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
563 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
564 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
566 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
568 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
570 [19610] dynamic-link: ldconfig -X removes stale symbolic links
571 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
572 [19633] locale: strfmon_l applies global locale to number formatting
573 [19642] network: Memory leak in getnameinfo
574 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
575 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
577 [19654] math: [x86_64] Need testcase for BZ #19590 fix
578 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
580 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
581 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
583 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
585 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
587 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
588 pointers and lengths in error-case.
589 [19727] locale: Converting from/to UTF-xx with iconv() does not always
590 report errors on UTF-16 surrogates values.
591 [19755] nscd: nscd assertion failure in gc
592 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
593 [19759] libc: mempcpy shouldn't be inlined
594 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
595 [19765] libc: s390 needs an optimized mempcpy
596 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
597 NAME_MAX limit assumption (CVE-2016-1234)
598 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
599 [19787] network: Missing and incorrect truncation checks in getnameinfo
600 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
602 [19791] network: Assertion failure in res_query.c with un-connectable name
604 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
605 [19822] math: libm.so install clobbers old version
606 [19825] network: resolv: send_vc can return uninitialized data in second
607 response to getaddrinfo
608 [19830] network: nss_dns: should check RDATA length against buffer length
609 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
610 confronted with A/AAAA records of invalid size
611 [19837] nss: nss_db: No retries for some long lines with a larger buffer
612 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
613 [19853] stdio: Printing IBM long double in decimal with high precision is
615 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
616 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
617 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
619 [19865] network: Assertion failure or memory leak in
620 _nss_dns_getcanonname_r
621 [19868] network: nss_dns: netent code does not skip over non-PTR records
622 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
624 [19881] string: Improve x86-64 memset
625 [19907] string: Incorrect memcpy tests
626 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
628 [19925] libc: termios.h XCASE namespace
629 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
630 [19929] libc: limits.h NL_NMAX namespace
631 [19931] stdio: Memory leak in vfprintf
632 [19957] libc: clone(CLONE_VM) access invalid parent memory
633 [19963] localedata: en_IL: New locale
634 [19989] stdio: stdio.h cuserid namespace
635 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
637 [19996] locale: langinfo.h nl_langinfo_l namespace
638 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
639 [20010] network: getaddrinfo: Stack overflow in hostent translation
641 [20012] stdio: libio: fmemopen append mode failure
642 [20014] stdio: stdio.h namespace for pre-threads POSIX
643 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
644 [20023] libc: fcntl.h timespec namespace
645 [20024] math: [x86_64] vectorized sincos trashes the stack
646 [20031] network: nss_hesiod: Heap overflow in get_txt_records
647 [20041] time: sys/time.h timespec namespace
648 [20043] libc: unistd.h missing cuserid for UNIX98 and before
649 [20044] libc: unistd.h missing pthread_atfork for UNIX98
650 [20051] libc: ttyslot in wrong header under wrong conditions
651 [20054] libc: gethostname not declared for XPG4
652 [20055] libc: termios.h missing tcgetsid for XPG4
653 [20072] dynamic-link: x86 init_cpu_features is called twice in static
655 [20073] libc: sys/stat.h fchmod namespace
656 [20074] libc: stdlib.h rand_r namespace
657 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
658 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
660 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
661 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
663 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
664 [20119] libc: Wrong mask for processors level type from CPUID
665 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
666 AS not supporting AVX512
667 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
669 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
670 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
672 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
673 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
674 [20178] libc: posix_spawn{p} should not call exit
675 [20191] stdio: libio: vtables hardening
676 [20195] string: FMA4 detection requires CPUID execution with register
678 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
679 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
681 [20212] math: acos (sNaN) returns sNaN
682 [20213] math: asin (sNaN) returns sNaN
683 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
684 [20218] math: [i386] asinhl (sNaN) returns sNaN
685 [20219] math: [i386] atanhl (sNaN) returns sNaN
686 [20222] stdio: fopencookie: Mangle function pointers
687 [20224] math: [i386] cbrtl (sNaN) returns sNaN
688 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
689 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
691 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
692 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
693 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
694 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
695 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
696 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
697 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
698 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
699 [20240] math: modf (sNaN) returns sNaN
700 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
701 [20250] math: frexp (sNaN) returns sNaN
702 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
703 [20255] math: [i386] fdim, fdimf return with excess range and precision /
705 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
706 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
707 subscript is below array bounds [-Werror=array-bounds]
708 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
709 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
711 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
712 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
714 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
716 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
717 [20316] localedata: id_ID: Februari instead of Pebruari
718 [20327] string: POWER8 strcasecmp returns incorrect result
719 [20347] math: Failure: Test: j0_downward (0xap+0)
720 [20348] libc: FAIL: misc/tst-preadvwritev64
721 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
722 [20350] libc: There is no test for p{read,write}64
723 [20357] math: Incorrect cos result for 1.5174239687223976
724 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
728 * Unicode 8.0.0 Support: Character encoding, character type info, and
729 transliteration tables are all updated to Unicode 8.0.0, using new
730 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
731 These updates cause user visible changes, such as the fixes for bugs
732 89, 16061, and 18568.
734 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
735 kernel-internal CPU set size. This means that requests that change the
736 CPU affinity which failed before (for example, an all-ones CPU mask) will
737 now succeed. Applications that need to determine the effective CPU
738 affinities need to call sched_getaffinity or pthread_getaffinity_np after
739 setting it because the kernel can adjust it (and the previous size check
740 would not detect this in the majority of cases).
742 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
743 the following new symbols are used: fts64_children, fts64_close,
744 fts64_open, fts64_read and fts64_set.
746 * getaddrinfo now detects certain invalid responses on an internal netlink
747 socket. If such responses are received, an affected process will
748 terminate with an error message of "Unexpected error <number> on netlink
749 descriptor <number>" or "Unexpected netlink response of size <number> on
750 descriptor <number>". The most likely cause for these errors is a
751 multi-threaded application which erroneously closes and reuses the netlink
752 file descriptor while it is used by getaddrinfo.
754 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
755 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
756 unnecessary serialization of memory allocation requests across threads.
757 The defect is now corrected. Users should see a substantial increase in
758 the concurent throughput of allocation requests for applications which
759 trigger this bug. Affected applications typically create and
760 destroy threads frequently. (Bug 19048 was reported and analyzed by
763 * There is now a --disable-timezone-tools configure option for disabling the
764 building and installing of the timezone related utilities (zic, zdump, and
765 tzselect). This is useful for people who build the timezone data and code
766 independent of the GNU C Library.
768 * The obsolete header <regexp.h> has been removed. Programs that require
769 this header must be updated to use <regex.h> instead.
771 * The obsolete functions bdflush, create_module, get_kernel_syms,
772 query_module and uselib are no longer available to newly linked binaries;
773 the header <sys/kdaemon.h> has been removed. These functions and header
774 were specific to systems using the Linux kernel and could not usefully be
775 used with the GNU C Library on systems with version 2.6 or later of the
778 * Optimized string, wcsmbs and memory functions for IBM z13.
779 Implemented by Stefan Liebler.
781 * Newly linked programs that define a variable called signgam will no longer
782 have it set by the lgamma, lgammaf and lgammal functions. Programs that
783 require signgam to be set by those functions must ensure that they use the
784 variable provided by the GNU C Library and declared in <math.h>, without
785 defining their own copy.
787 * The minimum GCC version that can be used to build this version of the GNU
788 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
789 still be used to compile programs using the GNU C Library.
791 Security related changes:
793 * An out-of-bounds value in a broken-out struct tm argument to strftime no
794 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
796 * The LD_POINTER_GUARD environment variable can no longer be used to disable
797 the pointer guard feature. It is always enabled. Previously,
798 LD_POINTER_GUARD could be used to disable security hardening in binaries
799 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
802 * An integer overflow in hcreate and hcreate_r could lead to an
803 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
805 * The catopen function no longer has unbounded stack usage. Reported by
808 * The nan, nanf and nanl functions no longer have unbounded stack usage
809 depending on the length of the string passed as an argument to the
810 functions. Reported by Joseph Myers. (CVE-2014-9761)
812 * A stack-based buffer overflow was found in libresolv when invoked from
813 libnss_dns, allowing specially crafted DNS responses to seize control
814 of execution flow in the DNS client. The buffer overflow occurs in
815 the functions send_dg (send datagram) and send_vc (send TCP) for the
816 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
817 family. The use of AF_UNSPEC triggers the low-level resolver code to
818 send out two parallel queries for A and AAAA. A mismanagement of the
819 buffers used for those queries could result in the response of a query
820 writing beyond the alloca allocated buffer created by
821 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
822 the overflow. Thanks to the Google Security Team and Red Hat for
823 reporting the security impact of this issue, and Robert Holiday of
824 Ciena for reporting the related bug 18665. (CVE-2015-7547)
826 The following bugs are resolved with this release:
828 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
829 [887] math: Math library function "logb" and "nextafter" inconsistent
830 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
831 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
832 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
833 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
835 [4404] localedata: German translation of "Alarm clock" is misleading
836 [6799] math: nextafter() and nexttoward() doen't set errno on
837 overflow/underflow errors
838 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
840 [10432] nis: _nss_nis_setnetgrent assertion failure
841 [11460] libc: fts has no LFS support
842 [12926] network: getaddrinfo()/make_request() may spin forever
843 [13065] nptl: Race condition in pthread barriers
844 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
845 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
847 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
848 [14912] libc: Rename non-installed bits/*.h headers
849 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
850 [15367] math: Let gcc use __builtin_isinf
851 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
852 [15421] math: lgamma wrongly sets signgam for ISO C
853 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
854 FE_INVALID with argument out of range
855 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
857 [15786] dynamic-link: ifunc resolver functions can smash function
859 [15918] math: Unnecessary check for equality in hypotf()
860 [16061] localedata: Review / update transliteration data
861 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
863 [16141] time: strptime %z offset restriction
864 [16171] math: drem should be alias of remainder
865 [16296] math: fegetround is pure?
866 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
867 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
869 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
870 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
872 [16422] math: [powerpc] math-float, math-double failing llrint tests with
873 "Exception "Inexact" set" on ppc32
874 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
875 [16517] math: Missing underflow exception from tanf/tan/tanl
876 [16519] math: Missing underflow exception from sinhf
877 [16520] math: Missing underflow exception from tanhf
878 [16521] math: Missing underflow exception from exp2
879 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
881 [16734] stdio: fopen calls mmap to allocate its buffer
882 [16961] math: nan function incorrect handling of bad sequences
883 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
884 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
885 [16985] locale: localedef: confusing error message when opening output
887 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
888 [17197] locale: Redundant shift character in iconv conversion output at
890 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
891 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
892 [17250] dynamic-link: static linking breaks nss loading
893 (getaddrinfo/getpwnam/etc...)
894 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
896 [17441] math: isnan() should use __builtin_isnan() in GCC
897 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
898 (related to lock elision)
899 [17787] manual: Exponent on page 324 of the PDF ends prematurely
900 [17886] time: strptime should be able to parse "Z" as a timezone with %z
901 [17887] time: strptime should be able to parse "+01:00" style timezones
902 [17905] libc: catopen() Multiple unbounded stack allocations
904 [18084] libc: backtrace (..., 0) dumps core on x86
905 [18086] libc: nice() sets errno to 0 on success
906 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
907 is too large (CVE-2015-8778)
908 [18251] dynamic-link: SONAME missing when audit modules provides path
909 [18265] libc: add attributes for wchar string and memory functions
910 [18370] math: csqrt missing underflows
911 [18421] libc: [hppa] read-only segment has dynamic relocations
912 [18472] libc: Obsolete syscall wrappers should be compat symbols
913 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
914 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
915 [18525] localedata: Remove locale timezone information
916 [18560] libc: [powerpc] spurious bits/ipc.h definitions
917 [18568] localedata: Update locale data to Unicode 8.0
918 [18589] locale: sort-test.sh fails at random
919 [18595] math: ctan, ctanh missing underflows
920 [18604] libc: assert macro-expands its argument
921 [18610] math: S390: fetestexcept() reports any exception if DXC-code
922 contains a vector instruction exception.
923 [18611] math: j1, jn missing errno setting on underflow
924 [18618] localedata: sync Chechen locale definitions with other *_RU
926 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
927 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
928 [18665] network: In send_dg, the recvfrom function is NOT always using the
929 buffer size of a newly created buffer (CVE-2015-7547)
930 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
932 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
934 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
935 [18699] math: tilegx cproj() for various complex infinities does not yield
937 [18724] libc: Harden put*ent functions against data injection
938 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
939 [18755] build: build errors with -DNDEBUG
940 [18757] stdio: fmemopen fails to set errno on failure
941 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
943 [18781] libc: openat64 lacks O_LARGEFILE
944 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
945 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
946 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
947 [18790] math: [ldbl-128ibm] tanhl inaccurate
948 [18795] libc: stpncpy fortification misses buffer lengths that are
950 [18796] build: build fails for --disable-mathvec
951 [18803] math: hypot missing underflows
952 [18820] stdio: fmemopen may leak memory on failure
953 [18823] math: csqrt spurious underflows
954 [18824] math: fma spurious underflows
955 [18825] math: pow missing underflows
956 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
957 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
958 [18870] build: sem_open.c fails to compile with missing symbol
960 [18872] stdio: Fix memory leak in printf_positional
961 [18873] libc: posix_fallocate overflow check ineffective
962 [18875] math: Excess precision leads incorrect libm
963 [18877] libc: arm: mmap offset regression
964 [18887] libc: memory corruption when using getmntent on blank lines
965 [18918] localedata: hu_HU: change time to HH:MM:SS format
966 [18921] libc: Regression: extraneous stat() and fstat() performed by
968 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
969 binaries (CVE-2015-8777)
970 [18951] math: tgamma missing underflows
971 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
973 [18953] localedata: lt_LT: change currency symbol to the euro
974 [18956] math: powf inaccuracy
975 [18961] math: [i386] exp missing underflows
976 [18966] math: [i386] exp10 missing underflows
977 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
978 [18969] build: multiple string test failures due to missing locale
980 [18970] libc: Reference of pthread_setcancelstate in libc.a
981 [18977] math: float / long double Bessel functions not in XSI POSIX
982 [18980] math: i386 libm functions return with excess range and precision
983 [18981] math: i386 scalb*, ldexp return with excess range and precision
984 [18982] stdio: va_list and vprintf
985 [18985] time: Passing out of range data to strftime() causes a segfault
987 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
988 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
989 [19012] locale: iconv_open leaks memory on error path
990 [19016] math: clog, clog10 inaccuracy
991 [19018] nptl: Mangle function pointers in tls_dtor_list
992 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
993 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
994 [19048] malloc: malloc: arena free list can become cyclic, increasing
996 [19049] math: [powerpc] erfc incorrect zero sign
997 [19050] math: [powerpc] log* incorrect zero sign
998 [19058] math: [x86_64] Link fail with -fopenmp and -flto
999 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1000 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1001 [19074] network: Data race in _res_hconf_reorder_addrs
1002 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1003 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1004 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1006 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1008 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1009 [19086] manual: posix_fallocate64 documented argument order is wrong.
1010 [19088] math: lround, llround missing exceptions close to overflow
1012 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1013 [19095] math: dbl-64 lrint incorrect for 64-bit long
1014 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1015 [19124] dynamic-link: ld.so failed to build with older assmebler
1016 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1017 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1018 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1019 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1020 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1021 pthread_setaffinity_np
1022 [19156] math: [ldbl-128] j0l spurious underflows
1023 [19164] nptl: tst-getcpu fails with many possible CPUs
1024 [19168] math: math/test-ildoubl and math/test-ldouble failure
1025 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1026 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1028 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1029 (FE_NOMASK_ENV) do not clear SSE exceptions
1030 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1031 _int_new_arena/reused_arena
1032 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1033 [19201] math: dbl-64 remainder incorrect sign of zero result
1034 [19205] math: bits/math-finite.h conditions do not match math.h and
1036 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1037 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1038 for C99-based standards
1039 [19212] libc: features.h not -Wundef clean
1040 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1042 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1043 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1044 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1046 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1047 "inexact" exceptions
1048 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1050 [19242] libc: strtol incorrect in Turkish locales
1051 [19243] malloc: reused_arena can pick an arena on the free list, leading
1052 to an assertion failure and reference count corruption
1053 [19253] time: tzset() ineffective when temporary TZ did not include DST
1055 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1056 [19270] math: [hppa] Shared libm missing __isnanl
1057 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1058 MAP_HUGETLB and MAP_STACK defines
1059 [19313] nptl: Wrong __cpu_mask for x32
1060 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1061 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1062 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1063 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1064 [19363] time: x32: times() return value wrongly truncates/sign extends
1066 [19367] dynamic-link: Improve branch prediction on Silvermont
1067 [19369] network: Default domain name not reset by res_ninit when "search"
1068 / "domain" entry is removed from resolv.conf
1069 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1071 [19385] localedata: bg_BG: time separator should be colon, not comma
1072 [19408] libc: linux personality syscall wrapper may erroneously return an
1073 error on 32-bit architectures
1074 [19415] libc: dladdr returns wrong names on hppa
1075 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1076 IBM905, IBM907, and IBM909
1077 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1078 [19443] build: build failures with -DDEBUG
1079 [19451] build: Make check fails on test-double-vlen2
1080 [19462] libc: Glibc failed to build with -Os
1081 [19465] math: Wrong code with -Os
1082 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1084 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1086 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1087 [19486] math: S390: Math tests fail with "Exception Inexact set".
1088 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1089 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1091 [19590] math: Fail to build shared objects that use libmvec.so functions.
1095 * The following bugs are resolved with this release:
1097 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1098 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1099 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1100 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1101 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1102 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1103 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1104 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1105 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1106 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1107 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1108 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1109 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1110 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1111 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1112 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1113 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1114 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1115 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1116 18657, 18676, 18694, 18696, 18887.
1118 * Cache information can be queried via sysconf() function on s390 e.g. with
1119 _SC_LEVEL1_ICACHE_SIZE as argument.
1121 * A buffer overflow in gethostbyname_r and related functions performing DNS
1122 requests has been fixed. If the NSS functions were called with a
1123 misaligned buffer, the buffer length change due to pointer alignment was
1124 not taken into account. This could result in application crashes or,
1125 potentially arbitrary code execution, using crafted, but syntactically
1126 valid DNS responses. (CVE-2015-1781)
1128 * The time zone file parser has been made more robust against crafted time
1129 zone files, avoiding heap buffer overflows related to the processing of
1130 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1131 large time zone data files. Overly long time zone specifiers in the TZ
1132 variable no longer result in stack overflows and crashes.
1134 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1135 for LD and GD on x86 and x86-64, has been implemented. You will need
1136 binutils-2.24 or later to enable this optimization.
1138 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1139 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1140 Hat). These updates cause user visible changes, such as the fix for bug
1143 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1144 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1145 condition in some applications.
1147 * Added vector math library named libmvec with the following vectorized x86_64
1148 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1150 The library can be disabled with --disable-mathvec. Use of the functions is
1151 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1152 Shared library libmvec.so is linked in as needed when using -lm (no need to
1153 specify -lmvec explicitly for not static builds).
1154 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1156 * A new fmemopen implementation has been added with the goal of POSIX
1157 compliance. The new implementation fixes the following long-standing
1158 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1159 old implementation is still present for use be by existing binaries.
1161 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1162 releases. It has been fixed to match 2.19 and older, but binaries built
1163 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1165 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1166 Contributed by Roland McGrath (Google).
1168 * The header <regexp.h> is deprecated, and will be removed in a future
1169 release. Use of this header will trigger a deprecation warning.
1170 Application developers should update their code to use <regex.h> instead.
1172 This header was formerly part of SUS, but was deprecated in 1994 and
1173 removed from the standard in 2001. Also, the glibc implementation
1174 leaks memory. See BZ#18681 for more details.
1178 * The following bugs are resolved with this release:
1180 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1181 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1182 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1183 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1184 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1185 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1186 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1187 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1188 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1189 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1192 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1193 too little memory for the to-be-scanned arguments and overflow the
1194 allocated buffer. The implementation now correctly computes the required
1195 buffer size when using malloc, and switches to malloc from alloca as
1198 * A new semaphore algorithm has been implemented in generic C code for all
1199 machines. Previous custom assembly implementations of semaphore were
1200 difficult to reason about or ensure that they were safe. The new version
1201 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1202 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1203 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1205 * Port to Altera Nios II has been contributed by Mentor Graphics.
1207 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1208 implementations for powerpc64/powerpc64le.
1209 Implemented by Adhemerval Zanella (IBM).
1211 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1212 and powerpc64le. This may improve lock scaling of existing programs on
1213 HTM capable systems. The lock elision code is only enabled with
1214 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1215 powerpc will issue a transaction abort on every syscall to avoid side
1216 effects being visible outside transactions.
1218 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1219 AArch64. Contributed by ARM Ltd.
1221 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1223 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1224 under certain input conditions resulting in the execution of a shell for
1225 command substitution when the application did not request it. The
1226 implementation now checks WRDE_NOCMD immediately before executing the
1227 shell and returns the error WRDE_CMDSUB as expected.
1229 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1230 processing format strings with a large number of format specifiers.
1232 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1233 infinite loop if the DNS response contained a PTR record of an unexpected
1236 * The minimum GCC version that can be used to build this version of the GNU
1237 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1238 still be used to compile programs using the GNU C Library.
1240 * The GNU C Library is now built with -Werror by default. This can be
1241 disabled by configuring with --disable-werror.
1243 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1245 * The obsolete sigvec function has been removed. This was the original
1246 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1247 programs have been using instead for about 25 years. Of course, ABI
1248 compatibility for old binaries using sigvec remains intact.
1250 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1251 with newer versions of bison.
1253 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1254 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1255 registers overlay two consecutive single-precision registers. MIPS32R2
1256 introduced a new FPU mode (FR=1) where double-precision registers extend the
1257 corresponding single-precision registers which is incompatible with the
1258 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1259 require the use of FR=1 making a transition necessary. New o32 ABI
1260 extensions enable users to migrate over time from the original o32 ABI
1261 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1262 tracks the ABI of any loaded object and verifies that new objects are
1263 compatible. Mode transitions will also be requested as required and
1264 unsupportable objects will be rejected. The ABI checks include both soft and
1265 hard float ABIs for o32, n32 and n64.
1267 GCC 5 with GNU binutils 2.25 onwards:
1268 It is strongly recommended that all o32 system libraries are built using the
1269 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1270 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1271 compiler using --with-fp-32=xx to set this by default.
1275 * The following bugs are resolved with this release:
1277 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1278 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1279 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1280 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1281 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1282 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1283 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1284 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1285 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1286 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1287 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1288 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1289 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1290 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1291 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1293 * Reverted change of ABI data structures for s390 and s390x:
1294 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1295 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1296 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1297 are preserved pointing straight to the same implementation as the old ones.
1298 Given that, new callers will simply provide a too-big buffer to these
1299 functions. Any applications/libraries out there that embed jmp_buf or
1300 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1301 against 2.19 headers will have to rebuilt again. This is necessary in any
1302 case to revert the breakage in their ABI caused by the glibc change.
1304 * Support for file description locks is added to systems running the
1305 Linux kernel. The standard file locking interfaces are extended to
1306 operate on file descriptions, not file descriptors, via the use of
1307 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1308 are associated with an open file instead of a process.
1310 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1312 * The minimum Linux kernel version that this version of the GNU C Library
1313 can be used with is 2.6.32.
1315 * Running the testsuite no longer terminates as soon as a test fails.
1316 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1317 with PASS or FAIL lines for individual tests. A summary of the results is
1318 printed, including a list of failing lists, and "make check" exits with
1319 error status if there were any unexpected failures. "make check
1320 stop-on-test-failure=y" may be used to keep the old behavior.
1322 * The am33 port, which had not worked for several years, has been removed
1325 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1326 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1327 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1328 conflicted with POSIX (support for which was removed in 2.19), the
1329 interfaces those macros enabled remain available when compiling with
1330 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1331 test macros defined.
1333 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1335 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1336 This may improve lock scaling of existing programs on TX capable systems.
1337 The lock elision code is only built with --enable-lock-elision=yes and
1338 then requires a GCC version supporting the TX builtins. With lock elision
1339 default mutexes are elided via __builtin_tbegin, if the cpu supports
1340 transactions. By default lock elision is not enabled and the elision code
1343 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1344 copy the path argument. This allowed programs to cause posix_spawn to
1345 deference a dangling pointer, or use an unexpected pathname argument if
1346 the string was modified after the posix_spawn_file_actions_addopen
1349 * All supported architectures now use the main glibc sysdeps directory
1350 instead of some being in a separate "ports" directory (which was
1351 distributed separately before glibc 2.17).
1353 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1354 On configurations that support it (all Linux configurations), it's now
1355 used regardless of the --enable-add-ons switch to configure. It is no
1356 longer possible to build such configurations without pthreads support.
1358 * Locale names, including those obtained from environment variables (LANG
1359 and the LC_* variables), are more tightly checked for proper syntax.
1360 setlocale will now fail (with EINVAL) for locale names that are overly
1361 long, contain slashes without starting with a slash, or contain ".." path
1362 components. (CVE-2014-0475) Previously, some valid locale names were
1363 silently replaced with the "C" locale when running in AT_SECURE mode
1364 (e.g., in a SUID program). This is no longer necessary because of the
1367 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1368 with application code using Intel MPX instructions. (With all previous
1369 versions, the MPX register state could be clobbered when making calls
1370 into or out of a shared library.) Note that while the new dynamic
1371 linker is compatible with all known x86 hardware whether or not it
1372 supports Intel MPX, some x86 instruction-set emulators might fail to
1373 handle the new instruction encodings. This is known to affect Valgrind
1374 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1375 release), and might affect other tools that do instruction emulation.
1377 * Support for loadable gconv transliteration modules has been removed.
1378 The support for transliteration modules has been non-functional for
1379 over a decade, and the removal is prompted by security defects. The
1380 normal gconv conversion modules are still supported. Transliteration
1381 with //TRANSLIT is still possible, and the //IGNORE specifier
1382 continues to be supported. (CVE-2014-5119)
1384 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1385 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1386 resulting a denial-of-service security vulnerability in applications which
1387 use functions related to iconv. (CVE-2014-6040)
1391 * The following bugs are resolved with this release:
1393 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1394 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1395 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1396 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1397 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1398 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1399 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1400 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1401 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1402 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1403 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1404 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1405 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1406 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1407 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1408 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1409 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1410 16431, 16453, 16474, 16506, 16510, 16529
1412 * Slovenian translations for glibc messages have been contributed by the
1413 Translation Project's Slovenian team of translators.
1415 * The public headers no longer use __unused nor __block. This change is to
1416 support compiling programs that are derived from BSD sources and use
1417 __unused internally, and to support compiling with Clang's -fblock
1418 extension which uses __block.
1420 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1421 large collation sequences to optimize multiple passes. This cache
1422 computation may overflow for large collation sequences and may cause a
1423 stack or buffer overflow. This is now fixed to use a slower algorithm
1424 which does not use a cache if there is an integer overflow.
1426 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1427 rules for large collation sequences to optimize multiple passes and falls
1428 back to alloca if malloc fails, resulting in a possible stack overflow.
1429 The implementation now falls back to an uncached collation sequence lookup
1432 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1433 initialized for static applications resulting in the security feature
1434 being disabled. The pointer guard is now correctly initialized to a
1435 random value for static applications. Existing static applications need
1436 to be recompiled to take advantage of the fix (bug 15754).
1438 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1439 to the d_name member of struct dirent, or omit the terminating NUL
1440 character. (Bugzilla #14699).
1442 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1443 aligned_alloc functions could allocate too few bytes or corrupt the
1444 heap when passed very large allocation size values (Bugzilla #15855,
1447 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1448 for AF_INET6 has been fixed (Bugzilla #16072).
1450 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1451 pap_CW, quz_PE, the_NP.
1453 * Substantially revised locales: gd_GB, ht_HT
1455 * The LC_ADDRESS field was updated to support country_car for almost all
1458 * ISO 1427 definitions were updated.
1460 * ISO 3166 definitions were updated.
1462 * The localedef utility now supports --big-endian and --little-endian
1463 command-line options to generate locales for a different system from that
1464 for which the C library was built.
1466 * Binary locale files now only depend on the endianness of the system for
1467 which they are generated and not on other properties of that system. As a
1468 consequence, binary files generated with new localedef may be incompatible
1469 with old versions of the GNU C Library, and binary files generated with
1470 old localedef may be incompatible with this version of the GNU C Library,
1471 in the following circumstances:
1473 + Locale files may be incompatible on m68k systems.
1475 + Locale archive files (but not separate files for individual locales) may
1476 be incompatible on systems where plain "char" is signed.
1478 * The configure option --disable-versioning has been removed. Builds with
1479 --disable-versioning had not worked for several years.
1481 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1482 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1484 * SystemTap probes for malloc have been introduced.
1486 * SystemTap probes for slow multiple precision fallback paths of
1487 transcendental functions have been introduced.
1489 * Support for powerpc64le has been added.
1491 * The soft-float powerpc port now supports e500 processors.
1493 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1495 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1496 set of header declarations that are enabled by default, even when other
1497 feature test macros or compiler options such as -std=c99 would otherwise
1498 disable some of those declarations.
1500 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1501 conflict with POSIX. The libbsd-compat library (which was a dummy library
1502 that did nothing) has also been removed.
1504 * Preliminary documentation about Multi-Thread, Async-Signal and
1505 Async-Cancel Safety has been added.
1507 * Change of ABI data structures for s390 and s390x:
1508 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1509 allow for future hardware extensions. All library functions that accept or
1510 return these structures were versioned in order to provide backwards
1511 compatibility with existing code. However, not all cases can be handled
1512 with interface versioning. If an application embeds either structure into
1513 another structure and that structure is passed to another compilation unit
1514 which expects the newer larger-sized object then any access to the new
1515 fields will result in undefined behaviour. Similarly any access to fields
1516 that were moved by the enlarging of the structures will cause undefined
1517 behaviour. To fix the undefined behaviour all objects that are part of the
1518 final in-memory image must agree on the size of structures, and this may
1519 require recompilation.
1523 * The following bugs are resolved with this release:
1525 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1526 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1527 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1528 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1529 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1530 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1531 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1532 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1533 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1534 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1535 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1536 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1537 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1540 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1541 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1542 Distributions can re-enable building and using pt_chown via the new configure
1543 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1544 considerable security risks and should only be used if the distribution
1545 understands and accepts the risks.
1547 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1550 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1551 fixed (Bugzilla #15330).
1553 * Add support for calling C++11 thread_local object destructors on thread
1554 and program exit. This needs compiler support for offloading C++11
1555 destructor calls to glibc.
1557 * Improved worst case performance of libm functions with double inputs and
1560 * Support for priority inherited mutexes in pthread condition variables on
1561 non-x86 architectures.
1563 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1565 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1567 * Optimized string functions for ARM. Implemented by Will Newton and
1570 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1572 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1575 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1576 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1578 * Added a benchmark framework to track performance of functions in glibc.
1580 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1581 It is based on draft TS 18661 and currently enabled as a GNU extension.
1583 * On Linux, the clock function now uses the clock_gettime system call
1584 for improved precision, rather than old times system call.
1586 * Added support for version-3 tz format files. This is needed when using
1587 the tz database release 2013e or later, and affects a few unusual cases --
1588 currently only TZ='America/Godthab' for time stamps after 2037.
1590 * Added new API functions pthread_getattr_default_np and
1591 pthread_setattr_default_np to get and set the default pthread
1592 attributes of a process.
1594 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1595 This may improve lock scaling of existing programs on TSX capable systems.
1596 When the --enable-lock-elision=yes parameter is specified at configure
1597 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1600 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1601 availability). Implemented by Ryan S. Arnold.
1603 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1607 * The following bugs are resolved with this release:
1609 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1610 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1611 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1612 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1613 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1614 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1615 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1616 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1617 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1618 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1619 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1620 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1621 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1622 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1623 14879, 14889, 14893, 14898, 14914.
1625 * Optimization of memcpy for MIPS.
1627 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1628 EMFILE has been fixed (Bugzilla #14889).
1630 * The add-on ports collection is now distributed in the "ports" subdirectory
1631 of the main GNU C Library distribution, rather than separately.
1633 * Port to ARM AArch64 contributed by Linaro.
1635 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1636 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1638 Implemented by Andreas Krebbel.
1640 * The new function secure_getenv allows secure access to the environment,
1641 returning NULL if running in a SUID/SGID process. This function replaces
1642 the internal function __secure_getenv.
1644 * SystemTap static probes have been added into the dynamic linker.
1645 Implemented by Gary Benson.
1647 * Optimizations of string functions strstr, strcasestr and memmem.
1648 Implemented by Maxim Kuvyrkov.
1650 * The minimum Linux kernel version that this version of the GNU C Library
1651 can be used with is 2.6.16.
1653 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1654 powerpc POWER7. Implemented by Will Schmidt.
1656 * New configure option --disable-nscd builds the C library such that it
1657 never attempts to contact the Name Service Caching Daemon (nscd).
1658 New configure option --disable-build-nscd avoids building nscd itself;
1659 this is the default if --disable-nscd is used.
1661 * Improved support for cross-compilation, including cross-testing and
1662 bootstrap builds without a previously built glibc.
1664 * Several testsuite tests are now able to test multiple IFUNC variants of an
1665 interface, rather than just testing the one that would be chooen by
1668 * New configure options --with-bugurl and --with-pkgversion, for
1669 distributors to use to embed their bug-reporting and package version
1670 information in --help and --version output.
1672 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1673 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1674 allows creation of chroots without the procfs mounted on /proc.
1676 * The `crypt' function now fails if passed salt bytes that violate the
1677 specification for those values. On Linux, the `crypt' function will
1678 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1679 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1680 when the mode is enabled.
1682 * The `clock_*' suite of functions (declared in <time.h>) is now available
1683 directly in the main C library. Previously it was necessary to link with
1684 -lrt to use these functions. This change has the effect that a
1685 single-threaded program that uses a function such as `clock_gettime' (and
1686 is not linked with -lrt) will no longer implicitly load the pthreads
1687 library at runtime and so will not suffer the overheads associated with
1688 multi-thread support in other code such as the C++ runtime library.
1690 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1695 * The following bugs are resolved with this release:
1697 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1698 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1699 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1700 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1701 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1702 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1703 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1704 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1705 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1706 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1707 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1708 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1709 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1710 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1711 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1712 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1713 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1714 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1715 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1716 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1717 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1718 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1721 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1722 configuring glibc with:
1723 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1724 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1725 Implemented by H.J. Lu.
1729 + define static_assert
1731 + do not declare gets
1733 + declare at_quick_exit and quick_exit also for ISO C11
1735 + aligned_alloc. NB: The code is deliberately allows the size parameter
1736 to not be a multiple of the alignment. This is a moronic requirement
1737 in the standard but it is only a requirement on the caller, not the
1740 + timespec_get added
1742 + uchar.h support added
1744 + CMPLX, CMPLXF, CMPLXL added
1746 Implemented by Ulrich Drepper.
1748 * Support for the IA-64 has been moved to ports.
1750 * Remove support for anything but ELF binary format
1752 * Checking versions of poll, ppoll added.
1753 Implemented by Ulrich Drepper.
1755 * More generic and 64-bit performance optimizations to math functions.
1756 Implemented by Ulrich Drepper.
1758 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1759 headers and functions available at compile time as they were before
1760 version 2.14. This option will be removed at some time in the future
1761 after the TI-RPC library becomes fully sufficient for the needs of
1762 existing applications.
1764 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1765 Note that glibc is not expected to work with any Linux kernel version
1768 * New header <sys/auxv.h> and function getauxval allowing easy access to
1769 the AT_* key-value pairs passed from the Linux kernel. The header also
1770 defines the HWCAP_* bits associated with the AT_HWCAP key.
1772 * New locales: mag_IN
1774 * New configure option --enable-systemtap builds SystemTap static probes
1775 into libc for setjmp and longjmp and into libpthread for various operations.
1776 So far the setjmp/longjmp probes and some of the libpthread probes are
1777 provided only for i*86 and x86_64.
1778 Implemented by Roland McGrath and Rayson Ho.
1780 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1782 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1785 * More optimized functions for SPARC. Implemented by David S. Miller.
1787 * Improved support for cross-compilation, especially bootstrap builds
1788 without a previously built glibc.
1790 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1791 Chris Metcalf from Tilera.
1793 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1794 now supported for ARM processors.
1796 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1797 as the name of the dynamic linker, to distinguish it from the
1798 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1800 * Support for CRIS has been removed from ports.
1802 * A new class of installed header has been documented for low-level
1803 platform-specific functionality. PowerPC added the first instance with a
1804 function to provide time base register access. Contributed by Tulio
1805 Magno Quites Machado Filho.
1807 * ix86 configurations now install header files that are consistent with
1808 what x86-64 configurations install. These same header files can be used
1809 for -m32, -m64, or -mx32 builds.
1810 Contributed by H.J. Lu.
1812 * Math library bug fixes. A thorough audit of all open math library bugs was
1813 conducted by Joseph Myers. Significant progress was made on many math
1814 library bugs resulting in more accurate exceptions and function results.
1815 Many thanks to all those that contributed including Andreas Jaeger for his
1816 patch review and work on the x87 trigonometric instruction issues.
1818 * Timezone data is no longer installed. Timezone-related binaries and scripts
1819 will continue to be installed. Users should obtain their timezone data from
1820 their distribution provider or from the tzdata package at
1821 <ftp://munnari.oz.au/pub/>.
1825 * The following bugs are resolved with this release:
1827 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1828 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1829 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1830 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1831 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1832 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1833 13484, 13506, 13515, 13523, 13524, 13538, 13540
1835 * New program pldd to list loaded object of a process
1836 Implemented by Ulrich Drepper.
1838 * Add nss_db support back to glibc. No more dependency on Berkeley db
1839 and support for initgroups lookups.
1840 Implemented by Ulrich Drepper.
1842 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1843 Contributed by HJ Lu.
1845 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1846 Contributed by HJ Lu.
1848 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1849 on x86-32 and x86-64.
1850 Contributed by Liubov Dmitrieva.
1852 * Optimized strchr and strrchr for SSE on x86-32.
1853 Contributed by Liubov Dmitrieva.
1855 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1856 for x86-64 and x86-32.
1857 Contributed by Liubov Dmitrieva.
1859 * New interfaces: scandirat, scandirat64
1860 Implemented by Ulrich Drepper.
1862 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1863 Implemented by Ulrich Drepper.
1865 * nscd now also caches the netgroup database.
1866 Implemented by Ulrich Drepper.
1868 * Integrate libm with gcc's -ffinite-math-only option.
1869 Implemented by Ulrich Drepper.
1871 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1872 to math functions. Implemented by Ulrich Drepper.
1874 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1875 Implemented by Ulrich Drepper.
1877 * New Linux interfaces: process_vm_readv, process_vm_writev
1879 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1880 Implemented by Ulrich Drepper.
1882 * Optimized nearbyint and strcasecmp for PPC.
1883 Implemented by Adhemerval Zanella.
1885 * New locales: bho_IN, unm_US, es_CU, ta_LK
1889 * The following bugs are resolved with this release:
1891 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1892 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1893 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1894 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1895 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1896 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1897 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1898 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1899 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1900 12795, 12811, 12813, 12814, 12841
1902 * The RPC implementation in libc is obsoleted. Old programs keep working
1903 but new programs cannot be linked with the routines in libc anymore.
1904 Programs in need of RPC functionality must be linked against TI-RPC.
1905 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1907 Visible changes of this change include (obviously) the inability to link
1908 programs using RPC functions without referencing the TI-RPC library and the
1909 removal of the RPC headers from the glibc headers.
1910 Implemented by Ulrich Drepper.
1912 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1913 syncfs, setns, sendmmsg
1915 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1916 yue_HK, lij_IT, mhr_RU
1918 * New encodings: CP770, CP771, CP772, CP773, CP774
1920 * New program sotruss to trace calls through PLTs
1921 Implemented by Ulrich Drepper.
1923 * The malloc hook implementation is marked deprecated and will be removed
1924 from the default implementation in the next version. The design never
1925 worked ever since the introduction of threads. Even programs which do
1926 not create threads themselves can use multiple threads created internally.
1930 * The following bugs are resolved with this release:
1932 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1933 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1934 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1935 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1938 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1940 * POWER7 optimizations: memset, memcmp, strncmp
1942 * New optimized string functions for x86-64: strnlen (SSE2),
1943 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1944 Implemented by Ulrich Drepper.
1948 * The following bugs are resolved with this release:
1950 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1951 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1952 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1953 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1954 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1955 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1956 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1957 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1959 * New interfaces: pthread_getname_np, pthread_setname_np
1961 * New Linux interface: recvmmsg
1963 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1965 * The dynamic linker now recognizes supported ABI versions from the
1966 EI_ABIVERSION field in the ELF header.
1967 Implemented by Ulrich Drepper.
1969 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1970 will not be used to fill in password fields in the passwd.byname replies.
1971 Instead it is used to synthesize the shadow.byname table, should it be
1972 missing. This is a useful mode in some installations involving Solaris.
1973 Implemented by Ulrich Drepper.
1975 * New locales: kok_IN, sq_MK, cv_RU
1979 * The following bugs are resolved with this release:
1981 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1982 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1983 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1984 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1985 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1986 10789, 10817, 10823, 10840, 10847
1988 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1990 Implemented by Ulrich Drepper.
1992 * Checking version of longjmp added that fails if an uninitialized stack
1993 frame would be created. Implemented by Ulrich Drepper.
1995 * STT_GNU_IFUNC is now supported in static executables.
1996 Implemented by H.J. Lu.
1998 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2000 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2001 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2002 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2003 Contributed by H.J. Lu.
2005 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2006 Implemented by Ulrich Drepper.
2008 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2010 Contributed by H.J. Lu.
2012 * Support for fma instruction in AVX on x86-64.
2013 Implemented by H.J. Lu and Ulrich Drepper.
2015 * AVX support in x86-64 auditing support in ld.so.
2016 Implemented by H.J. Lu.
2018 * STB_GNU_UNIQUE support added.
2019 Implemented by Ulrich Drepper.
2021 * Implement second fallback mode for optimized DNS lookup for even more
2022 broken environments. If two requests from the same port are not
2023 handled correctly close the socket and open a new one before sending
2024 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2025 can be used to select this mode right away, instead of rediscovering the
2026 necessity is every process again.
2027 Implemented by Ulrich Drepper.
2029 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2030 Implemented by Adam Tkac.
2032 * Optimized iconv conversions for S390x.
2033 Implemented by Andreas Krebbel.
2035 * Using condvars with PI mutexes is now more efficient due to kernel
2036 support for requeueing to PI futexes. NPTL support added for x86-64.
2037 Implemented by Ulrich Drepper.
2039 * New locales: ps_AF, my_MM
2043 * The following bugs are resolved with this release:
2045 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2046 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2047 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2048 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2049 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2051 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2052 Implemented by Ulrich Drepper.
2054 * New Linux interfaces: accept4, fallocate, fallocate64.
2055 Implemented by Ulrich Drepper.
2057 * Correct declarations of string function when used in C++ code. This
2058 could lead to compile errors for invalid C++ code.
2060 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2063 * New POSIX 2008 interface: psiginfo
2064 Implemented by Ulrich Drepper.
2066 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2067 Implemented by Ulrich Drepper.
2069 * Support for selecting between multiple function definitions at runtime
2070 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2072 * The libcrypt library can now use the hash function implementations in
2073 NSS. Implemented by Ulrich Drepper.
2075 * The malloc implementation can be compiled to be less memory efficient
2076 but higher performing in multi-threaded programs.
2077 Implemented by Ulrich Drepper.
2079 * New locales: nan_TW@latin, ks_IN
2081 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2082 Implemented by Ulrich Drepper.
2084 * Extended printf hook support. It is possible to use user-defined types
2085 and extend existing format specifiers.
2086 Implemented by Ulrich Drepper.
2088 * Handling for group shadow files has been added.
2089 Implemented by Ulrich Drepper.
2091 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2092 servers (the case, e.g., for some people using the built-in DNS
2093 server in ADSL modems/routers). There is a once-per-process timeout
2094 in case of a broken server. To avoid it, users can run nscd or put
2095 'options single-request' in /etc/resolv.conf.
2096 Implemented by Ulrich Drepper.
2100 * The following bugs are resolved with this release:
2102 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2103 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2104 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2105 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2106 6968, 6974, 6980, 6995, 7008, 7009, 7029
2108 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2109 up at the same time. Implemented by Ulrich Drepper.
2111 * TLS descriptors for LD and GD on x86 and x86-64.
2112 Implemented by Alexandre Oliva.
2114 * getaddrinfo now handles DCCP and UDPlite.
2115 Implemented by Ulrich Drepper.
2117 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2118 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2119 Implemented by Ulrich Drepper.
2121 * New implementation of memmem, strstr, and strcasestr which is O(n).
2122 Implemented by Eric Blake.
2124 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2126 * Implement "e" option for popen to open file descriptor with the
2127 close-on-exec flag set. Implemented by Ulrich Drepper.
2129 * Implement "b" mode for fmemopen. In this mode writes writes don't
2130 implicitly add a NUL byte and seeks from the end of the buffer really
2131 use the buffer end, not the string length as the basis.
2132 Implemented by Ulrich Drepper.
2134 * Many functions, exported and internal, now atomically set the close-on-exec
2135 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2137 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2139 Implemented by Pravin Satpute.
2141 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2145 * The following bugs are resolved with this release:
2147 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2148 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2149 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2150 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2151 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2152 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2153 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2155 * New locales: bo_CN, bo_IN, shs_CA.
2157 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2159 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2160 Implemented by Pravin Satpute.
2162 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2164 * nscd caches now all timeouts for DNS entries
2165 Implemented by Ulrich Drepper.
2167 * nscd is more efficient and wakes up less often.
2168 Implemented by Ulrich Drepper.
2170 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2171 vdprintf, and obstack_vprintf.
2172 Implemented by Jakub Jelinek.
2174 * Faster memset for x86-64.
2175 Implemented by Harsha Jagasia and H.J. Lu.
2177 * Faster memcpy on x86.
2178 Implemented by Ulrich Drepper.
2180 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2181 Implemented by Ulrich Drepper.
2183 * Faster sqrt and sqrtf implemention for some PPC variants.
2184 Implemented by Steven Munroe.
2188 * The following bugs are resolved with this release:
2190 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2191 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2192 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2193 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2194 5071, 5103, 5104, 5112, 5113, 5184, 5186
2196 * More checking functions: fread, fread_unlocked, open*, mq_open.
2197 Implemented by Jakub Jelinek and Ulrich Drepper.
2199 * Extend fortification to C++. Implemented by Jakub Jelinek.
2201 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2202 by not recognizing 'a' as a modifier when those specs are requested.
2203 Implemented by Jakub Jelinek.
2205 * PPC optimizations to math and string functions.
2206 Implemented by Steven Munroe.
2208 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2209 options to be passed. Implemented by Ulrich Drepper.
2211 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2213 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2214 Implemented by Ulrich Drepper.
2216 * Handle private futexes in the NPTL implementation.
2217 Implemented by Jakub Jelinek and Ulrich Drepper.
2219 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2220 Implemented by Roland McGrath and Ulrich Drepper.
2222 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2224 * SHA-256 and SHA-512 based password encryption.
2225 Implemented by Ulrich Drepper.
2227 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2228 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2231 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2232 Implemented by Ulrich Drepper.
2236 * The following bugs are resolved with this release:
2238 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2239 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2240 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2241 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2242 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2243 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2244 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2247 * New Linux interfaces: epoll_pwait, sched_getcpu.
2249 * New generic interfaces: strerror_l.
2251 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2255 * The following bugs are resolved with this release:
2257 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2258 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2259 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2260 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2261 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2262 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2263 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2264 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2265 3225, 3251, 3252, 3253, 3273
2267 * For Linux, the sorting of addresses returned by getaddrinfo now also
2268 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2269 Implemented by Ulrich Drepper.
2271 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2272 Implemented by Ulrich Drepper.
2274 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2276 * New iconv module for MIK. Contributed by Alexander Shopov.
2278 * For sites with broken group and/or passwd database, the auto-propagate
2279 option of nscd can prevent creating ID lookup entries from the results
2280 of a name lookup and vice versa. This usually is no problem but some
2281 site might have problems with the default behavior.
2282 Implemented by Ulrich Drepper.
2284 * Iterating over entire database in NIS can be slow. With the
2285 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2286 to trade time for memory. The entire database will be read at once.
2287 Implemented by Ulrich Drepper.
2289 * The interfaces introduced in RFC 3542 have been implemented by
2292 * Support for the new ELF hash table format was added by Ulrich Drepper.
2294 * Support for priority inheritance mutexes added by Jakub Jelinek and
2297 * Support for priority protected mutexes added by Jakub Jelinek.
2299 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2300 or_IN, csb_PL, fy_NL, sr_ME.
2304 * More overflow detection functions.
2306 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2307 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2309 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2310 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2311 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2312 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2313 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2314 by Masahide Washizawa.
2316 * It is now possible to install audit libraries for the dynamic linker using
2317 LD_AUDIT. Implemented by Ulrich Drepper.
2319 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2320 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2321 For a libc and libpthread that works well on Linux 2.4 kernels, we
2322 recommend using the stable 2.3 branch.
2324 * The new function open_wmemstream acts like open_memstream,
2325 but uses a wchar_t wide character buffer.
2327 * The new function ppoll is an improved version of poll, similar to pselect.
2329 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2330 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2331 renameat, unlinkat, linkat, symlinkat, readlinkat.
2333 * New Linux kernel system calls: unshare,
2334 inotify_init, inotify_add_watch, inotify_rm_watch.
2336 * The euidaccess function is now also known by the alias eaccess,
2337 for compatibility with some other systems.
2339 * Timezone data updated to 2006b version.
2343 * The following bugs are resolved with this release:
2345 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2346 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2347 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2348 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2349 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2350 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2352 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2354 * As of this release, GCC 4 can be used to compile the C Library.
2356 * Timezone data updated to 2005m version.
2360 * The following bugs are resolved with this release:
2362 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2363 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2364 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2365 777, 787, 821, 822, 823, 825
2367 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2371 * Support for RFC 3678. Real implementations exist only for Linux so far.
2372 Implemented by Ulrich Drepper.
2374 * nscd can now cache entries persistently. Expiring entries are reloaded.
2375 For speedups the cache can be shared in memory with client processes.
2376 Implemented by Ulrich Drepper.
2378 * nscd can now perform SELinux checks.
2379 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2381 * getaddrinfo queries are now cached. Canonical name lookup is performed
2383 Implemented by Ulrich Drepper.
2385 * The nothrow function attribute is used when headers are used by gcc when
2386 compiling C code. This can avoid significant amounts of exception
2389 * The malloc functions perform more error checking and are stricter when
2390 it comes to reacting on errors. The default action is to terminate
2391 the process after showing an error message. Implemented by Ulrich Drepper.
2393 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2394 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2396 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2397 namespaces using the new function dlmopen(). This feature is of course,
2398 like most other dynamic loading functionality, not available in statically
2399 linked applications. Implemented by Ulrich Drepper.
2401 * Low-overhead boundary checking variants of string and some stdio functions
2402 were added. These are to be used in conjunction with a gcc patch by
2403 Jakub Jelinek which adds calls to these functions if possible.
2404 Implemented by Jakub Jelinek and Ulrich Drepper.
2406 * Old code for several operating systems and machine architectures that
2407 have not been in working condition in a long time have been removed from
2408 the main source tree maintained by the GNU C Library's maintainers.
2409 These files are now reside in the separate `ports' source module
2410 that is usable as an add-on when building the library.
2414 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2415 interrogate the dynamic linker, compatible with the Solaris interface.
2417 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2418 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2420 * getifaddrs now uses the netlink interface on Linux to get its information.
2421 Implemented by Thorsten Kukuk.
2423 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2424 Implemented by Ulrich Drepper.
2426 * support for non-executable stacks on x86 has been added. Changes mostly
2429 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2432 * getaddrinfo now performs destination address selection according to
2437 * Thread-safe interfaces for many functions that access locale data
2438 were added in version 2.3, but these features were omitted from NEWS.
2439 Many functions have variants with an `_l' suffix that take a `locale_t'
2440 object as a parameter rather than consulting the current locale.
2441 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2442 create and maintain `locale_t' objects. Additionally, the new function
2443 `uselocale' sets "the current locale" (as used by functions not so
2444 parameterized) set for an individual thread. These features were added
2445 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2447 * The functions getresuid, getresgid, setresuid, and setresgid, which
2448 have long been available on Linux, are now declared in <unistd.h>
2449 and are now also available on the Hurd.
2451 * ELF thread-local storage support (TLS) now works on x86-64.
2453 * The new dynamic string token $LIB is expanded in shared library names.
2454 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2456 * Aldy Hernandez contributed complete software floating point support for
2457 PowerPC machines with no FPU.
2459 * fexecve is implemented on Linux.
2461 * The `btowc' function should work at least twice as fast due to
2462 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2464 * With approriate thread add-ons cancelable functions are now implemented
2465 in libc.so as well. No need to call the function in libpthread. This
2466 change allowed to finally disable the incorrect and expensive handling
2467 of weak definition in ld.so.
2469 * Yet more PLT entries in libc.so have been removed. We finally arrived
2470 at the bare minimum. Startup times improved appropriately.
2472 * Support for the new Linux/x86 system call interface was added. The
2473 AT_SYSINFO auxiliary vector entry is recognized and handled.
2477 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2480 * iconv (the program and the interface) now accepts empty names (excluding
2481 options like //TRANSLIT) to mean "use charset of current locale".
2483 * localedef can now transliterate characters in strings which are not in
2484 the provided charmap. The information from the input locale is used.
2486 * Prelinking support was added for ELF targets. This requires additional
2487 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2489 * Read-only stdio streams now use mmap to speed up operation by eliminating
2490 copying and buffer underflows. To use add 'm' to the mode string of
2491 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2493 * The malloc functions were completely rewritten by Wolfram Gloger based
2494 on Doug Lea's malloc-2.7.0.c.
2496 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2497 implementation of regex.
2499 * Bruno Haible upgraded the iconv and locale implementation to support
2502 * Contents of the LC_* and LANG environment variables in the CEN style are
2503 not recognized anymore. It never was used. Change by Ulrich Drepper.
2505 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2506 thread-local storage (TLS) ABI on some platforms.
2507 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2509 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2510 EUC-JISX0213, and TSCII.
2512 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2513 BSD-compatible interface for getting all network interface addresses.
2514 Implementation for IPv4 by Roland McGrath.
2516 * Loading of locale data is faster due to the introduction of a locale
2517 archive. Implemented by Roland McGrath and Ulrich Drepper.
2519 * Startup times are significantly reduced by not using exported functions
2520 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2523 * Steven Munroe contributed a port to PowerPC64/Linux.
2527 * The Hurd now uses the GNU libio implementation of stdio rather than the
2528 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2530 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2531 with the same behavior as the Linux system call of the same name.
2535 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2536 128-bit long double format.
2538 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2539 IBM1161, and IBM1162 charsets.
2541 * Andreas Jaeger contributed a port to x86-64/Linux.
2543 * Peter Bruin contributed a port to PowerPC/Hurd.
2545 * libc functions using I/O streams now can handle wide-oriented streams
2548 * optimizations in the dynamic linker. Binaries created by recent binutils
2549 versions start up quicker due to reduced time spent on relocations.
2551 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2555 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2556 asinh, atanh, j0 for the 128-bit long double format.
2558 * Bruno Haible updated all the code handling Unicode in some form to
2559 support Unicode 3.1.
2561 * Speed of regex for single-byte locales is back to previous levels.
2562 Patch by Isamu Hasegawa.
2564 * Alpha, SPARC, and IA-64 now also using floating stacks.
2566 * Startup time of internationalized applications greatly improved through
2567 iconv cache. Use iconvconfig to generate the cache file.
2568 Contributed by Ulrich Drepper.
2570 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2571 severe performance problems. Patch by David Mosberger.
2573 * The Hurd port got a lot more functionality like AIO, various stdio
2574 extensions, etc. Mainly done by Roland McGrath.
2576 * mtrace can now lookup symbols in shared libraries.
2580 * Intel's IA-64 math library is largely integrated. It provides fast and
2581 accurate implementatations for most basic and standard math functions
2582 in float, double, and long double format.
2584 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2585 and asin for the 96-bit long double format and asin, log, tan for the
2586 128-bit long double format.
2588 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2589 The basic double functions exist today. Contributed by Abraham Ziv
2590 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2591 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2593 * An asynchronous name lookup library was added. The interface is designed
2594 after POSIX AIO. The proposal was circulated beforehand to get comments.
2595 No negative ones came in. Implemented by Ulrich Drepper.
2597 * Port to S390/64bit contributed by Martin Schwidefsky
2598 <schwidefsky@de.ibm.com>.
2600 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2601 of functions for Linux/IA-64.
2603 * The RPC code is now thread safe. Threads can now use the same service
2604 of different services at the same time. Patch by Eric Norum
2605 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2607 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2608 family of functions for Linux/S390.
2610 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2611 of functions for Linux/x86.
2613 * Port to Linux/CRIS contributed by Axis Communications.
2617 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2618 we can now check for namespace violations and missing declarations. The
2619 result is that almost all headers are now Unix-compliant (as defined in
2620 the upcoming XPG6). The negative side is that some programs might need
2621 corrections, too, if they depend on the incorrect form of the headers in
2622 previous versions which defined too many symbols and included too many
2625 * regex now handles multibyte character sets correctly.
2626 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2628 * iconv (the program) does now conform to the upcoming XPG6 and handles
2629 charmaps. Instead of the charset names the path of charmaps can be
2630 provided and the conversion happens based on this data.
2631 Contributed by Ulrich Drepper.
2633 * The locale program now provides detailed information about the installed
2638 only lists the names of the supported locales
2642 provides details such as country, language, and codeset name.
2643 Contributed by Ulrich Drepper.
2647 * The gencat program now parses the input file according to the charset
2648 selected by the LC_CTYPE category. This is important for stateful
2649 character sets. To make generating catalogs easier there is a way
2650 to overwrite the charset selected by the locale: before the first
2651 message or $ quote line the catalog can contain a line like
2653 $ codeset=ISO-8859-2
2655 to select the charset (ISO-8859-2 in this case).
2657 Implemented by Shinya Hanataka and Ulrich Drepper.
2659 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2660 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2661 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2663 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2664 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2666 * The LANGUAGE environment variable is now ignored unless the locale is
2667 changed from the default "C" locale.
2669 * The usual bug fixes.
2673 * Greg McGary added runtime support for bounds checking using gcc's
2674 new -fbounded-pointers option. ix86 target is complete. PowerPC
2677 * Thorsten Kukuk added secure mode support to nscd.
2679 * The Berkeley db libraries have been removed.
2681 Related, the nss_db module is now in a separate package since it
2682 obviously requires a database library being available.
2684 * Wide character I/O streams implemented by Ulrich Drepper.
2686 * Functions from the extended socket API added by Ulrich Drepper.
2688 * Functions feenableexcept and fedisableexcept to control the
2689 behaviour of individual exceptions have been added by Andreas Jaeger.
2691 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2693 * The resolver code has been updated from bind 8.2.3-T5B which supports
2694 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2697 This change could in some situations effect backward compatibility. Since
2698 now `_res' is a thread-local instead of a global variable, modifying it
2699 in one thread does not have any effect in other threads.
2701 The resolver library was also extended to allow IPv6 as the transport
2702 protocol for the requests. This work was done by Stig Venaas.
2704 * Compatibility code for K&R C compilers has been removed from the
2705 header files. A ISO C compiler is needed to use the library
2706 (conforming to either C89 or C99 standard).
2708 * Complete rewrite of the localedef program to support multibyte character
2709 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2710 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2711 Implemented by Ulrich Drepper.
2713 Bruno Haible significantly improved the generation and use of the data
2714 structures for the wide character tables.
2716 * Plural handling in gettext implemented by Ulrich Drepper.
2718 * The utmp daemon has been removed.
2720 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2722 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2725 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2727 * POSIX barriers implemented by Kaz Kylheku.
2729 * POSIX spawn function family implemented by Ulrich Drepper.
2731 * POSIX spinlocks are now available.
2733 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2735 * the configure option --enable-kernel=X.Y.Z allows to strip out
2736 compatibility for kernel versions before X.Y.Z. This is currently only
2737 implemented for Linux.
2739 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2740 field sin6_scope_id. This means that all programs using IPv6 should be
2741 recompiled. Don't expect binary compatibility with previous glibc
2744 * various conversion modules for IBM character sets contributed by
2747 * IA-64 port by Jes Sorensen and HJ Lu.
2761 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2763 * New xdr functions are added; some rpc functions are now 64bit clean.
2765 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2767 * Fixed known incompatibilities with glibc 2.0.
2769 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2771 * Optimized a number of functions (especially the ELF dynamic loader).
2773 * Update timezone data files.
2775 * lots of charmaps corrections
2777 * some new locale definitions and charmaps
2782 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2783 conform to POSIX member type requirements. Symbol versions have been
2784 adjusted as needed within the library, and for direct use by applications,
2785 but there is potential for problems if third-party libraries use
2786 struct timeval as part of their interface. This does not present
2787 a problem for X and other "essential" system libraries.
2789 * An additional locale model to support C++ Standard Library locale
2790 model and probably more was implemented by Ulrich Drepper.
2792 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2795 * Miles Bader provided the `argp' function family to support hierachical
2796 command line argument parsing, layered on top of getopt.
2798 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2800 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2803 * scanf recognizes the %a and %A format for scanning floating point numbers.
2805 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2806 information and interfaces for the available integer types.
2808 * about 130 new math functions were added to implement the ISO C9x math
2811 * the new header <complex.h> contains definitions of the complex math
2812 functions from ISO C 9X.
2814 * the new header <tgmath.h> defines generic macros to use complex or
2815 real valued functions.
2817 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2819 * Andreas Jaeger provided a test suite for the math library.
2821 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2823 * Ulrich Drepper added character set conversion functions (iconv).
2825 * Optimized string functions have been added.
2827 * The localedata addon is now part of glibc.
2829 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2831 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2832 daemon for NSS (nscd).
2834 Missing a better place here are some numbers on improvements. Under
2835 Linux 2.1.125 un-tar-ing the kernel sources takes
2839 using local files 12.19s 6.88s 22.91s
2841 using NIS 13.92s 8.91s 26.34s
2843 using NIS & nscd 10.37s 7.34s 25.30s
2845 using NIS+ 27.57s 30.37s 640.46s
2847 using NIS+ & nscd 10.25s 7.83s 26.51s
2849 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2851 Keep in mind that non-namelookup related operations dominate above times.
2852 It was just a common complain that using NIS+ unpacking the kernel is
2855 [1] The old nscd implementation is not available anymore since it was
2856 distributed with glibc up to version 2.0.98 and thus is now replaced.
2858 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2860 * Mark Kettenis provided a Hesiod NSS module.
2862 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2863 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2865 * rcmd can now handle netgroups (Dick Streefland).
2867 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2870 * Support for the IPv6 protocol has been added to the socket API, as per the
2871 latest draft standards.
2873 * Support for Linux 2.2 has been added.
2875 * Interface changes relative to the latest 2.0.x release:
2876 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2877 addseverity NEW: Unix98
2878 alphasort64 NEW: LFS
2879 argp_err_exit_status NEW: argp, GNU ext
2880 argp_error NEW: argp, GNU ext
2881 argp_failure NEW: argp, GNU ext
2882 argp_help NEW: argp, GNU ext
2883 argp_parse NEW: argp, GNU ext
2884 argp_program_bug_address NEW: argp, GNU ext
2885 argp_program_version NEW: argp, GNU ext
2886 argp_program_version_hook NEW: argp, GNU ext
2887 argp_state_help NEW: argp, GNU ext
2888 argp_usage NEW: argp, GNU ext
2889 authdes_create NEW: Secure RPC
2890 authdes_getucred NEW: Secure RPC
2891 authdes_pk_create NEW: Secure RPC
2892 backtrace NEW: GNU ext.
2893 backtrace_symbols NEW: GNU ext.
2894 backtrace_symbols_fd NEW: GNU ext.
2896 cacosf NEW: ISO C 9x
2897 cacosh NEW: ISO C 9x
2898 cacoshf NEW: ISO C 9x
2899 cacoshl NEW: ISO C 9x
2900 cacosl NEW: ISO C 9x
2907 casinf NEW: ISO C 9x
2908 casinh NEW: ISO C 9x
2909 casinhf NEW: ISO C 9x
2910 casinhl NEW: ISO C 9x
2911 casinl NEW: ISO C 9x
2913 catanf NEW: ISO C 9x
2914 catanh NEW: ISO C 9x
2915 catanhf NEW: ISO C 9x
2916 catanhl NEW: ISO C 9x
2917 catanl NEW: ISO C 9x
2918 cbc_crypt NEW: Secure RPC
2922 ccoshf NEW: ISO C 9x
2923 ccoshl NEW: ISO C 9x
2929 cimagf NEW: ISO C 9x
2930 cimagl NEW: ISO C 9x
2931 clearerr_locked REMOVED
2932 clntunix_create NEW: sunrpc ext
2934 clog10 NEW: ISO C 9x
2935 clog10f NEW: ISO C 9x
2936 clog10l NEW: ISO C 9x
2946 cprojf NEW: ISO C 9x
2947 cprojl NEW: ISO C 9x
2949 crealf NEW: ISO C 9x
2950 creall NEW: ISO C 9x
2955 csinhf NEW: ISO C 9x
2956 csinhl NEW: ISO C 9x
2959 csqrtf NEW: ISO C 9x
2960 csqrtl NEW: ISO C 9x
2964 ctanhf NEW: ISO C 9x
2965 ctanhl NEW: ISO C 9x
2967 des_setparity NEW: Secure RPC
2968 ecb_crypt NEW: Secure RPC
2969 endutxent NEW: Unix98
2971 exp10f NEW: ISO C 9x
2972 exp10l NEW: ISO C 9x
2976 fattach NEW: STREAMS
2977 fdetach NEW: STREAMS
2981 feclearexcept NEW: ISO C 9x
2982 fegetenv NEW: ISO C 9x
2983 fegetexceptflag NEW: ISO C 9x
2984 fegetround NEW: ISO C 9x
2985 feholdexcept NEW: ISO C 9x
2987 feraiseexcept NEW: ISO C 9x
2988 ferror_locked REMOVED
2989 fesetenv NEW: ISO C 9x
2990 fesetexceptflag NEW: ISO C 9x
2991 fesetround NEW: ISO C 9x
2992 fetestexcept NEW: ISO C 9x
2993 feupdateenv NEW: ISO C 9x
2994 fflush_locked REMOVED
2998 fgets_unlocked NEW: GNU ext.
2999 fileno_locked REMOVED
3011 fputc_locked REMOVED
3012 fputs_unlocked NEW: GNU ext.
3013 fread_unlocked NEW: GNU ext.
3018 fstatvfs NEW: Unix98
3022 ftruncate64 NEW: LFS
3024 fwrite_unlocked NEW: GNU ext.
3025 gai_strerror NEW: IPv6
3029 getchar_locked REMOVED
3031 getdate_err NEW: Unix98
3032 getdate_r NEW: GNU ext.
3034 getnameinfo NEW: IPv6
3035 getnetname NEW: Secure RPC
3036 getpmsg NEW: STREAMS
3037 getpt NEW: Unix98 PTY
3038 getrlimit64 NEW: LFS
3039 getutxent NEW: Unix98
3040 getutxid NEW: Unix98
3041 getutxline NEW: Unix98
3042 glob64 NEW: GNU ext.
3043 globfree64 NEW: GNU ext.
3044 gnu_get_libc_release NEW: GNU ext.
3045 gnu_get_libc_version NEW: GNU ext.
3046 grantpt NEW: Unix98 PTY
3047 host2netname NEW: Secure RPC
3049 iconv_close NEW: iconv
3050 iconv_open NEW: iconv
3051 if_freenameindex NEW: IPv6
3052 if_indextoname NEW: IPv6
3053 if_nameindex NEW: IPv6
3054 if_nametoindex NEW: IPv6
3055 in6addr_any NEW: IPv6
3056 in6addr_loopback NEW: IPv6
3057 inet6_isipv4mapped NEW: IPv6
3058 isastream NEW: STREAMS
3059 iswblank NEW: GNU ext.
3060 key_decryptsession NEW: Secure RPC
3061 key_decryptsession_pk NEW: Secure RPC
3062 key_encryptsession NEW: Secure RPC
3063 key_encryptsession_pk NEW: Secure RPC
3064 key_gendes NEW: Secure RPC
3065 key_get_conv NEW: Secure RPC
3066 key_secretkey_is_set NEW: Secure RPC
3067 key_setnet NEW: Secure RPC
3068 key_setsecret NEW: Secure RPC
3069 llrint NEW: ISO C 9x
3070 llrintf NEW: ISO C 9x
3071 llrintl NEW: ISO C 9x
3072 llround NEW: ISO C 9x
3073 llroundf NEW: ISO C 9x
3074 llroundl NEW: ISO C 9x
3079 lrintf NEW: ISO C 9x
3080 lrintl NEW: ISO C 9x
3081 lround NEW: ISO C 9x
3082 lroundf NEW: ISO C 9x
3083 lroundl NEW: ISO C 9x
3085 makecontext NEW: Unix98
3086 mempcpy NEW: GNU ext.
3089 modify_ldt NEW: kernel
3093 nearbyint NEW: ISO C 9x
3094 nearbyintf NEW: ISO C 9x
3095 nearbyintl NEW: ISO C 9x
3096 netname2host NEW: Secure RPC
3097 netname2user NEW: Secure RPC
3098 nexttoward NEW: ISO C 9x
3099 nexttowardf NEW: ISO C 9x
3100 nexttowardl NEW: ISO C 9x
3104 passwd2des NEW: Secure RPC
3106 pow10f NEW: GNU ext.
3107 pow10l NEW: GNU ext.
3110 printf_size NEW: GNU ext.
3111 printf_size_info NEW: GNU ext.
3112 profil_counter REMOVED
3113 pthread_mutexattr_getkind_np REPLACED
3114 pthread_mutexattr_setkind_np REPLACED
3115 ptsname NEW: Unix98 PTY
3116 ptsname_r NEW: Unix98 PTY
3118 putchar_locked REMOVED
3119 putgrent NEW: GNU ext.
3121 putpmsg NEW: STREAMS
3122 pututxline NEW: Unix98
3126 readdir64_r NEW: LFS
3127 remquo NEW: ISO C 9x
3128 remquof NEW: ISO C 9x
3129 remquol NEW: ISO C 9x
3131 roundf NEW: ISO C 9x
3132 roundl NEW: ISO C 9x
3134 scalbln NEW: ISO C 9x
3135 scalblnf NEW: ISO C 9x
3136 scalblnl NEW: ISO C 9x
3138 sendfile NEW: kernel
3139 setcontext NEW: Unix98
3140 setrlimit64 NEW: LFS
3141 setutxent NEW: Unix98
3143 sigignore NEW: Unix98
3144 sigqueue NEW: POSIX.1b
3145 sigrelse NEW: Unix98
3146 sigset NEW: POSIX.1b
3147 sigtimedwait NEW: POSIX.1b
3148 sigwaitinfo NEW: POSIX.1b
3149 sincos NEW: GNU ext.
3150 sincosf NEW: GNU ext.
3151 sincosl NEW: GNU ext.
3155 strcasestr NEW: GNU ext.
3156 strtoimax NEW: ISO C 9x
3157 strtoumax NEW: ISO C 9x
3158 strverscmp NEW: GNU ext.
3159 svcauthdes_stats NEW: Secure RPC
3160 svcunix_create NEW: sunrpc etx
3161 svcunixfd_create NEW: sunrpc ext
3162 swapcontext NEW: Unix98
3163 tcgetsid NEW: Unix98 PTY
3164 tdestroy NEW: GNU ext.
3165 tgamma NEW: ISO C 9x
3166 tgammaf NEW: ISO C 9x
3167 tgammal NEW: ISO C 9x
3171 truncf NEW: ISO C 9x
3172 truncl NEW: ISO C 9x
3174 unlockpt NEW: Unix98 PTY
3175 updwtmpx NEW: Unix98
3176 user2netname NEW: Secure RPC
3177 utmpxname NEW: Unix98
3178 versionsort NEW: GNU ext.
3179 versionsort64 NEW: GNU ext.
3181 wcscasecmp NEW: GNU ext.
3182 wcsncasecmp NEW: GNU ext.
3183 wcsnlen NEW: GNU ext.
3184 wcstoimax NEW: ISO C 9x
3185 wcstoll NEW: ISO C 9x
3186 wcstoull NEW: ISO C 9x
3187 wcstoumax NEW: ISO C 9x
3189 wordexp NEW: POSIX.2
3190 wordfree NEW: POSIX.2
3191 write_profiling REMOVED
3192 xdecrypt NEW: Secure RPC
3193 xdr_authdes_cred NEW: Secure RPC
3194 xdr_authdes_verf NEW: Secure RPC
3195 xdr_cryptkeyarg NEW: Secure RPC
3196 xdr_cryptkeyarg2 NEW: Secure RPC
3197 xdr_cryptkeyres NEW: Secure RPC
3198 xdr_getcredres NEW: Secure RPC
3199 xdr_key_netstarg NEW: Secure RPC
3200 xdr_key_netstres NEW: Secure RPC
3201 xdr_keybuf NEW: Secure RPC
3202 xdr_keystatus NEW: Secure RPC
3203 xdr_netnamestr NEW: Secure RPC
3204 xdr_sizeof NEW: Secure RPC
3205 xdr_unixcred NEW: sunrpc ext
3206 xencrypt NEW: Secure RPC
3207 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3218 * inet_ntoa is thread-safe
3220 * updwtmp is moved from libutil to libc
3222 * rewrite of cbrt function
3224 * update of timezone data
3238 * add atoll function
3240 * fix complex problems in Berkeley DB code
3242 * fix math functions
3246 * fixed lots of header problems (especially Linux/GNU specific)
3248 * dynamic loader preserves all registers
3250 * Roland McGrath provided support for handling of auxiliary objects in
3251 the ELF dynamic loader.
3253 * support for parallel builds is improved
3257 * GNU extensions are no longer declared by default. To enable them you
3258 must define the macro `_GNU_SOURCE' in your program or compile with
3261 * The library has changed from using GNU ld symbol aliases to using weak
3262 symbols where available. The ELF object file format supports weak
3263 symbols; GNU ld also supports weak symbols in the a.out format. (There
3264 is also now support for other GNU ld extensions in ELF. Use the
3265 `--with-elf' option to configure to indicate you have ELF, and
3266 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3267 of many files which contained only symbol aliases, reducing the size of
3268 the source and the compiled library; many other files were renamed to
3269 less cryptic names previously occupied by the symbol alias files.
3270 There is a new header file <elf.h> for programs which operate on
3271 files in the ELF format.
3273 * Converted to Autoconf version 2, so `configure' has more options.
3274 Run `configure --help' to see the details.
3276 * The library can now be configured to build profiling, highly-optimized
3277 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3278 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3279 `configure' enable building these extra libraries. The shared library is
3280 built by default when using both ELF and GNU ld. When shared libraries
3281 are enabled, the new library `-ldl' is available for arbitrary run-time
3282 loading of shared objects; its interface is defined in <dlfcn.h>. The
3283 new header file <link.h> gives access to the internals of the run-time
3284 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3285 application of same name on other systems and it provides information
3286 about dynamically linked binaries.
3288 * The C library now provides the run-time support code for profiling
3289 executables compiled with `-pg'. Programs can control the profiling code
3290 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3291 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3292 the support for this file format was contributed by David Mosberger-Tang.
3294 * The math code has been replaced with a math library based on fdlibm from
3295 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3296 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3297 functions. The math functions now reside in a separate library, so
3298 programs using them will need to use `-lm' their linking commands.
3300 * John C. Bowman contributed optimized ix87 assembler inline functions.
3302 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3303 mechanism similar to that found in Solaris 2. This is now used for the
3304 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3305 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3306 file controls what services are used for each individual database. This
3307 works by loading shared libraries with names specified in `nsswitch.conf',
3308 so service modules can be changed or added at any time without even
3309 relinking any program. Currently there are the file, db, and NIS based
3310 NSS services available.
3312 * The new functions `strtoq' and `strtouq' parse integer values from
3313 strings, like `strtol' and `strtoul', but they return `long long int' and
3314 `unsigned long long int' values, respectively (64-bit quantities).
3316 * The new functions `strtof' and `strtold' parse floating-point values from
3317 strings, like `strtod', but they return `float' and `long double' values,
3318 respectively (on some machines `double' and `long double' are the same).
3320 * Ulrich Drepper has contributed new implementations of the floating-point
3321 printing and reading code used in the `printf' family of functions and
3322 `strtod', `strtof', and `strtold'. These new functions are perfectly
3323 accurate, and much faster than the old ones.
3325 * The implementation of the POSIX locale model was completely rewritten by
3326 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3327 to compile the POSIX locale definition.
3329 * The former dummy implementations of the strcoll and strxfrm function are
3330 now replaced by fully functional code contributed by Ulrich Drepper. The
3331 collation information comes from the POSIX locale definitions.
3333 * The new header <langinfo.h> defines an interface for accessing
3334 various locale-dependent data (using the locale chosen with `setlocale').
3336 * Ulrich Drepper has contributed a new suite of functions for operation on
3337 wide-character and multibyte-character strings, in <wchar.h>;
3338 and classification and case conversion of wide characters, in <wctype.h>.
3339 These new functions are conforming to the ISO C, Amendement 1 specification.
3341 * There is now a second implementation of the standard I/O library available.
3342 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3343 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3345 * You can now use positional parameter specifications in format strings
3346 for the `printf' and `scanf' families of functions. For example,
3347 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3348 ``Number 6, Mr Jones''. This is mainly useful when providing different
3349 format strings for different languages, whose grammars may dictate
3350 different orderings of the values being printed. To support this
3351 feature, the interface for `register_printf_handler' has changed; see
3352 the header file <printf.h> for details.
3354 * The `printf' and `scanf' families of functions now understand a new
3355 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3356 to group numbers as indicated by the locale; for `scanf' and friends, this
3357 says to accept as valid only a number with all the proper grouping
3358 separators in the right places. In the default "C" locale, numbers are
3359 not grouped; but locales for specific countries will define the usual
3360 conventions (i.e. separate thousands with `,' in the US locale).
3362 * The pgrp functions have been regularized, slightly incompatibly but much
3363 less confusingly. The core functions are now `getpgid' and `setpgid',
3364 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3365 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3366 provided for compatibility. There is no longer an incompatible `getpgrp'
3367 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3369 * The new header file <fts.h> and suite of functions simplify programs that
3370 operate on directory trees. This code comes from 4.4 BSD.
3372 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3373 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3374 NSS scheme used in glibc.
3376 * There is a new malloc debugging hook `__memalign_hook'.
3378 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3379 `unsigned int' in <sys/types.h>. These are for compatibility only and
3380 their use is discouraged.
3382 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3383 done differently, so that it works even without GNU ld.
3385 * New function `euidaccess' checks allowed access to a file like `access',
3386 but using the effective IDs instead of the real IDs.
3388 * The time zone data files have been updated for the latest and greatest
3389 local time conventions of the countries of the world.
3391 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3394 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3395 for formatting floating-point numbers. They are provided only for
3396 compatibility; new programs should use `sprintf' instead. There are
3397 also equivalent function for the `long double' floating-point type and
3398 all functions also exist in a reentrant form.
3400 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3401 functions for maintaining the login-record files (primarily of use to
3402 system programs such as `login'), and convenient functions for
3403 allocating and initializing a pseudo-terminal (pty) device.
3405 * Ulrich Drepper has contributed new support for System V style
3406 shared memory and IPC on systems that support it.
3408 * Ulrich Drepper has contributed several miscellaneous new functions found
3409 in System V: The `hsearch' family of functions provide an effective
3410 implementation of hash tables; `a64l' and `l64a' provide a very simple
3411 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3414 * Ulrich Drepper has contributed new reentrant counterparts for the
3415 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3417 * Ulrich Drepper has contributed new, highly-optimized versions of several
3418 string functions for the i486/Pentium family of processors.
3420 * Ulrich Drepper has updated the Linux-specific code, based largely
3421 on work done in Hongjiu Lu's version of GNU libc for Linux.
3422 The GNU library now supports Linux versions 2.0.10 and later,
3423 using the ELF object file format (i[3456]86-*-linux).
3425 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3427 * David Mosberger-Tang and Richard Henderson have ported the C library
3428 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3429 dynamic linking support for ELF/Alpha.
3431 * Richard Henderson contributed several Alpha optimized assembler function
3432 for arithmetic and string handling.
3434 * Ulrich Drepper has contributed a new set of message catalog functions to
3435 support multiple languages using the <libintl.h> interface, for use with
3436 his new package GNU gettext. Translation volunteers have contributed
3437 catalogs of the library's messages in Spanish, German, and Korean.
3439 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3440 program and the `catgets' function for reading the catalog files it
3441 creates. (The <libintl.h> interface is preferred; we include the
3442 <nl_types.h> interface using `catgets' only for source compatibility with
3443 programs already written to use it.)
3445 * New header file <values.h> gives SVID-compatible names for <limits.h>
3448 * Various new macros, declarations, and small header files for compatibility
3451 * New function `group_member' is a convenient way to check if a process has
3452 a given effective group ID.
3454 * When using GCC 2.7 and later, the socket functions are now declared in a
3455 special way so that passing an argument of type `struct sockaddr_in *',
3456 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3457 `struct sockaddr *' type, does not generate a type-clash warning.
3459 * New function `error' declared in header file <error.h> is a convenient
3460 function for printing error messages and optionally exiting; this is the
3461 canonical function used in GNU programs. The new functions `err', `warn',
3462 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3463 doing the same thing.
3465 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3466 POSIX.2 `glob' function to do ~ and {...} expansion.
3468 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3469 `clearenv' which is used in POSIX.9 is also available.
3471 * New function `getsid' returns session ID number on systems that support it.
3473 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3474 files <db.h> and <mpool.h> provide a rich set of functions for several
3475 types of simple databases stored in memory and in files, and <ndbm.h> is
3476 an old `ndbm'-compatible interface using the `db' functions. Link with
3477 `-ldb' to get these functions.
3479 * New macro `strdupa' copies a string like `strdup', but uses local stack
3480 space from `alloca' instead of dynamic heap space from `malloc'.
3482 * New function `strnlen' is like `strlen' but searches only a given maximum
3483 number of characters for the null terminator. `stpncpy', `strndup' and
3484 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3487 * New function `statfs' in header <sys/statfs.h>.
3489 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3490 provide convenient functions for operating on blocks of null-terminated
3493 * A new suite of functions in <utmp.h> handle all the details of reading
3494 and writing the utmp file.
3496 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3499 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3500 completely POSIX compliant and also implemented the extended functionality
3501 to handle alternate digit representation and alternate era date formats.
3503 * Ulrich Drepper provided an implementation of the `strptime' function
3504 defined in XPG4.2 which transforms a string into a `struct tm' value.
3506 * Paul Eggert provided the tzselect shell script as part of the timezone
3507 code. The shell script makes it easy to select the correct timezone
3510 * The implementation of the malloc family of functions is completely replaced
3511 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3512 The implementation uses the mmap function (if available) and it is
3513 optimized for the use in multi threaded programs.
3515 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3516 functions. This new functionality is usable by specifying a special
3517 salt string and it is compatible with implementation on *BSD systems.
3519 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3520 `getsubopt' to handle second level command line options, `bsd_signal'
3521 to access BSD style `signal' functionality, the obsolete `regexp' style
3524 * the `lchown' function is available on system which support this
3527 * The implementation of the shadow password handling function was contributed
3530 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3532 * POSIX.1g support was added. The <sys/select.h> header is available,
3533 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3534 implementation of `getaddrinfo'.
3538 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3540 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3541 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3543 * The new function `daemon' from 4.4 BSD is useful for server programs that
3544 want to put themselves in the background.
3546 * Joel Sherrill has contributed support for several standalone boards that
3547 run without an operating system.
3549 * `printf', `scanf' and friends now accept a `q' type modifier for long
3550 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3552 * All of the code taken from BSD (notably most of the math and networking
3553 routines) has been updated from the BSD 4.4-Lite release.
3555 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3557 * The new functions `getdomainname' and `setdomainname' fetch or change the
3558 YP/NIS domain name. These are system calls which exist on systems which
3561 * The time zone data files have been updated for the latest international
3564 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3565 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3569 * The C library now includes support for Sun RPC, from Sun's free
3570 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3571 are included. (There is still no support for YP.)
3573 * Tom Quinn has contributed a port of the C library to SGI machines running
3574 Irix 4 (mips-sgi-irix4).
3576 * The new `lockf' function is a simplified interface to the locking
3577 facilities of `fcntl', included for compatibility.
3579 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3581 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3583 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3586 * The error message printed by `assert' for a failed assertion now includes
3587 the name of the program (if using GNU ld) and the name of the calling
3588 function (with versions of GCC that support this).
3590 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3592 * The library now includes the <sys/mman.h> header file and memory
3593 management functions `mmap', `munmap', `mprotect', `msync', and
3594 `madvise', on systems that support those facilities.
3596 * The interface for `mcheck' has changed slightly: the function called to
3597 abort the program when an allocation inconsistency is detected now takes
3598 an argument that indicates the type of failure. The new function
3599 `mprobe' lets you request a consistency check for a particular block at
3600 any time (checks are normally done only when you call `free' or `realloc'
3603 * It is now possible to easily cross-compile the C library, building on one
3604 system a library to run on another machine and/or operating system. All
3605 you need to do is set the variable `HOST_CC' in `configparms' to the
3606 native compiler for programs to run on the machine you are building on (a
3607 few generator programs are used on Unix systems); set `CC' to the
3610 * The new function `fexecve' (only implemented on the GNU system) executes
3611 a program file given a file descriptor already open on the file.
3615 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3616 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3618 * You can set the variable `libprefix' in `configparms' to specify a prefix
3619 to be prepended to installed library files; this makes it easy to install
3620 the GNU C library to be linked as `-lgnuc' or whatever.
3622 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3623 copies a limited number of characters from a string, and returns the
3624 address of the last character written.
3626 * You no longer need to check for whether the installed `stddef.h' is
3627 compatible with the GNU C library. configure now checks for you.
3629 * You can now define a per-stream `fileno' function to convert the
3630 stream's cookie into an integral file descriptor.
3632 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3633 allocates zero bytes of storage, and returns a unique pointer which
3634 you can pass to `realloc' or `free'. The behavior is undefined if
3635 you dereference this pointer.
3637 * The C library now runs on Sony NEWS m68k machines running either
3638 NewsOS 3 or NewsOS 4.
3640 * The new `syscall' function is a system-dependent primitive function
3641 for invoking system calls. It has the canonical behavior on Unix
3642 systems, including unreliable return values for some calls (such as
3643 `pipe', `fork' and `getppid').
3645 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3646 to `EAGAIN', which is the preferred name. On systems whose kernels
3647 use two distinct codes, the C library now translates EWOULDBLOCK to
3648 EAGAIN in every system call function.
3652 * The GNU C Library Reference Manual is now distributed with the library.
3653 `make dvi' will produce a DVI file of the printed manual.
3654 `make info' will produce Info files that you can read on line using C-h i
3655 in Emacs or the `info' program.
3656 Please send comments on the manual to bug-glibc-manual@gnu.org.
3658 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3660 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3662 * Jason Merrill has contributed a port to the Sequent Symmetry running
3663 Dynix version 3 (i386-sequent-dynix).
3665 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3666 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3668 * New function `memory_warnings' lets you arrange to get warnings when
3669 malloc is running out of memory to allocate, like Emacs gives you.
3671 * The C library now contains the relocating allocator used in Emacs 19 for
3672 its editing buffers. This allocator (ralloc) minimizes allocation
3673 overhead and fragmentation by moving allocated regions around whenever it
3674 needs to. You always refer to a ralloc'd region with a "handle" (a
3675 pointer to a pointer--an object of type `void **').
3677 * There is a new `printf' format: `%m' gives you the string corresponding
3678 to the error code in `errno'.
3680 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3681 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3682 pass, the `a' modifier says to fill in a `char **' you pass with a
3685 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3686 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3687 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3689 * `mkstemp' is a traditional Unix function to atomically create and open a
3690 uniquely-named temporary file.
3694 * The standard location for the file that says what the local timezone is
3695 has changed again. It is now `/usr/local/etc/localtime' (or more
3696 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3698 * The distribution no longer contains any files with names longer than 14
3701 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3702 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3704 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3708 * The configuration process has changed quite a bit. The `configure'
3709 script is now used just like the configuration scripts for other GNU
3710 packages. The `sysdeps' directory hierarchy is much rearranged.
3711 The file `INSTALL' explains the new scheme in detail.
3713 * The header files no longer need to be processed into ANSI C and
3714 traditional C versions. There is just one set of files to install, and
3715 it will work with ANSI or old C compilers (including `gcc -traditional').
3717 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3718 MIPS DECStation running Ultrix 4.
3720 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3721 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3722 made itself into a shared library.
3724 * Yet further improved support for the i386, running 4.3 BSD-like systems
3725 (such as Mach 3 with the Unix single-server), or System V.
3727 * New function `strncasecmp' to do case-insensitive string comparison
3728 with limited length.
3730 * New function `strsep' is a reentrant alternative to `strtok'.
3732 * New functions `scandir' and `alphasort' for searching directories.
3734 * New function `setenv' is a better interface to `putenv'.
3736 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3737 function for traversing a directory tree.
3739 * The GNU obstack package is now also part of the C library.
3740 The new function `open_obstack_stream' creates a stdio stream that
3741 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3742 formatted output directly to an obstack.
3744 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3745 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3747 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3749 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3750 things to your strings.
3752 * There are some new test programs: test-fseek, testmb, and testrand.
3754 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3755 These ports are not finished, but are a good starting place for really
3756 supporting those systems.
3758 * `/etc/localtime' is now the standard location for the file that says what
3759 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3760 This follows the general principle that `/etc' is the place for all local
3761 configuration files.
3763 * The C library header files now use `extern "C"' when used by the C++
3764 compiler, so the C library should now work with C++ code.
3766 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3767 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3770 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3771 function which uses the `merge sort' algorithm, and is said to be
3772 significantly faster than the old GNU `qsort' function. Merge sort is
3773 now the standard `qsort' function. The new algorithm can require a lot
3774 of temporary storage; so, the old sorting function is called when the
3775 required storage is not available.
3777 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3778 provides the Unix `crypt' function, plus some other entry points.
3780 * The code and header files taken from 4.4 BSD have been updated with the
3781 latest files released from Berkeley.
3783 ----------------------------------------------------------------------
3784 Copyright information:
3786 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3788 Permission is granted to anyone to make or distribute verbatim copies
3789 of this document as received, in any medium, provided that the
3790 copyright notice and this permission notice are preserved,
3791 thus giving the recipient permission to redistribute in turn.
3793 Permission is granted to distribute modified versions
3794 of this document, or of portions of it,
3795 under the above conditions, provided also that they
3796 carry prominent notices stating who last changed them.
3799 version-control: never