1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2016 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 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
11 24731-2:2010, is supported to enable declarations of functions from that
12 TR. Note that not all functions from that TR are supported by the GNU C
15 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
16 18661-1:2014, is supported to enable declarations of functions and macros
17 from that TS. Note that most features from that TS are not supported by
20 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
21 18661-4:2015, is supported to enable declarations of functions and macros
22 from that TS. Note that most features from that TS are not supported by
25 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
26 means that in a future release, the macros “major”, “minor”, and “makedev”
27 will only be available from <sys/sysmacros.h>.
29 These macros are not part of POSIX nor XSI, and their names frequently
30 collide with user code; see for instance glibc bug 19239 and Red Hat bug
31 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
32 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
35 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
36 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions,
37 the femode_t type and the FE_DFL_MODE macro.
39 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
40 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
41 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
42 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
43 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
44 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
45 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
46 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
47 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
48 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
49 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
52 * New <math.h> features are added from TS 18661-1:2014:
54 - Comparison macros: iseqsig.
56 - Classification macros: iscanonical, issubnormal, iszero.
58 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
59 for the Linux quota interface which predates kernel version 2.4.22 has
62 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
63 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
64 “no-ip6-dotint” had already been the default, and support for the
65 “ip6-dotint” option was removed from the Internet in 2006.
67 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
68 flag from <resolv.h> have been removed. The option relied on a
69 backwards-incompatible DNS extension which was never deployed on the
72 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
73 in the <resolv.h> header file have been deprecated. They were already
76 * DNSSEC-related declarations and definitions have been removed from the
77 <arpa/nameser.h> header file, and libresolv will no longer attempt to
78 decode the data part of DNSSEC record types. Previous versions of glibc
79 only implemented minimal support for the previous version of DNSSEC, which
80 is incompatible with the currently deployed version.
82 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
83 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
84 <arpa/nameser.h> header file because the distinction between RR types and
85 meta-RR types is not officially standardized, subject to revision, and
86 thus not suitable for encoding in a macro.
88 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
89 and rhook members of the res_state type in <resolv.h> have been removed.
90 The glibc stub resolver did not support these hooks, but the header file
93 * For multi-arch support it is recommended to use a GCC which has
94 been built with support for GNU indirect functions. This ensures
95 that correct debugging information is generated for functions
96 selected by IFUNC resolvers. This support can either be enabled by
97 configuring GCC with '--enable-gnu-indirect-function', or by
98 enabling it by default by setting 'default_gnu_indirect_function'
99 variable for a particular architecture in the GCC source file
102 Security related changes:
104 On ARM EABI (32-bit), generating a backtrace for execution contexts which
105 have been created with makecontext could fail to terminate due to a
106 missing .cantunwind annotation. This has been observed to lead to a hang
107 (denial of service) in some Go applications compiled with gccgo. Reported
108 by Andreas Schwab. (CVE-2016-6323)
110 The following bugs are resolved with this release:
112 [The release manager will add the list generated by
113 scripts/list-fixed-bugs.py just before the release.]
117 * The minimum Linux kernel version that this version of the GNU C Library
118 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
119 kernel version 2.6.32 or later suffices (on architectures that already
120 required kernel versions more recent than 3.2, those requirements remain
121 unchanged). Linux 3.2 or later kernel headers are required on all
124 * The pap_AN locale has been deleted. This has been deprecated for a long
125 time. It has been replaced by pap_AW & pap_CW, both of which have long
126 been included in previous releases.
128 * The readdir_r and readdir64_r functions have been deprecated. It is
129 recommended to use readdir and readdir64 instead.
131 * The type “union wait” has been removed. It was deprecated in the early
132 1990s and never part of POSIX. Application code should use the int type
133 instead of “union wait”.
135 * A new NSS action is added to facilitate large distributed system
136 administration. The action, MERGE, allows remote user stores like LDAP
137 to be merged into local user stores like /etc/groups in order to provide
138 easy to use, updated, and managed sets of merged credentials. The new
139 action can be used by configuring it in /etc/nsswitch.conf:
140 group: files [SUCCESS=merge] nis
141 Implemented by Stephen Gallagher (Red Hat).
143 * The deprecated __malloc_initialize_hook variable has been removed from the
146 * The long unused localedef --old-style option has been removed. It hasn't
147 done anything in over 16 years. Scripts using this option can safely
150 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
151 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
152 return the next representable value in the direction of positive infinity
153 and the nextdown functions return the next representable value in the
154 direction of negative infinity. These are currently enabled as GNU
157 Security related changes:
159 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
160 could result in a stack overflow when getnetbyname was called with an
161 overly long name. (CVE-2016-3075)
163 * Previously, getaddrinfo copied large amounts of address data to the stack,
164 even after the fix for CVE-2013-4458 has been applied, potentially
165 resulting in a stack overflow. getaddrinfo now uses a heap allocation
166 instead. Reported by Michael Petlan. (CVE-2016-3706)
168 * The glob function suffered from a stack-based buffer overflow when it was
169 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
170 Reported by Alexander Cherepanov. (CVE-2016-1234)
172 * The Sun RPC UDP client could exhaust all available stack space when
173 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
174 alloca plugin for GCC. (CVE-2016-4429)
176 * The IPv6 name server management code in libresolv could result in a memory
177 leak for each thread which is created, performs a failing naming lookup,
178 and exits. Over time, this could result in a denial of service due to
179 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
181 The following bugs are resolved with this release:
183 [1170] localedata: ne_NP: update Nepali locale definition file
184 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
186 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
187 [6796] math: fdim() does not set errno on overflow
188 [10354] libc: posix_spawn should use vfork() in more cases than presently
189 [11213] localedata: localedata: add copyright disclaimer to locale files
190 [12143] localedata: chr_US: new Cherokee locale
191 [12450] localedata: sgs_LT: new locale
192 [12676] localedata: ln_CD: new locale
193 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
195 [13304] math: fma, fmaf, fmal produce wrong results
196 [14259] build: --localedir arg to configure is ignored
197 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
198 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
199 [14934] localedata: es_CL: wrong first weekday chilean locale
200 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
202 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
204 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
205 [15368] nptl: raise() is not async-signal-safe
206 [15479] math: ceil, floor, round and trunc raise inexact exception
207 [15578] localedata: kk_KZ: various updates
208 [16003] localedata: pap_AN: punt old locale
209 [16137] localedata: iw_IL: punt old locale
210 [16190] localedata: eo: new esperanto locale
211 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
212 [16742] malloc: race condition: pthread_atfork() called before first
213 malloc() results in unexpected locking behaviour/deadlocks
214 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
216 [16983] localedata: postal_fmt does not allow %l and %n modifiers
217 [17565] localedata: pt_PT: wrong (work-)week start
218 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
219 [17950] build: Build fails with -msse
220 [18205] localedata: be_BY*: wrong first_weekday and first_workday
221 [18433] libc: posix_spawn does not return correctly upon failure to
223 [18453] localedata: charmaps/IBM875: incorrect codes
224 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
226 [18896] localedata: he_IL: improvements for currency
227 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
229 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
231 [19056] libc: Deprecate readdir_r
232 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
234 [19198] localedata: nl_NL: small improvements for Dutch locales
235 [19257] network: Per-thread memory leak in __res_vinit with IPv6
236 nameservers (CVE-2016-5417)
237 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
239 [19400] locale: Language missing in "iso-639.def", trivial fix in
241 [19431] malloc: Deadlock between fflush, getdelim, and fork
242 [19505] libc: Incorrect file descriptor validity checks in
243 posix_spawn_file_actions_add{open,close,dup2}
244 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
246 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
247 `intl/{gettextP,loadinfo}.h'
248 [19534] libc: execle, execlp may use malloc
249 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
250 [19573] network: res_nclose and __res_maybe_init disagree about name
251 server initialization, breaking Hesiod
252 [19575] localedata: Status of GB18030 tables
253 [19581] localedata: sr_* date_fmt string contains additional newline
254 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
256 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
257 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
258 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
259 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
261 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
263 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
265 [19610] dynamic-link: ldconfig -X removes stale symbolic links
266 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
267 [19633] locale: strfmon_l applies global locale to number formatting
268 [19642] network: Memory leak in getnameinfo
269 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
270 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
272 [19654] math: [x86_64] Need testcase for BZ #19590 fix
273 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
275 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
276 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
278 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
280 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
282 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
283 pointers and lengths in error-case.
284 [19727] locale: Converting from/to UTF-xx with iconv() does not always
285 report errors on UTF-16 surrogates values.
286 [19755] nscd: nscd assertion failure in gc
287 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
288 [19759] libc: mempcpy shouldn't be inlined
289 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
290 [19765] libc: s390 needs an optimized mempcpy
291 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
292 NAME_MAX limit assumption (CVE-2016-1234)
293 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
294 [19787] network: Missing and incorrect truncation checks in getnameinfo
295 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
297 [19791] network: Assertion failure in res_query.c with un-connectable name
299 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
300 [19822] math: libm.so install clobbers old version
301 [19825] network: resolv: send_vc can return uninitialized data in second
302 response to getaddrinfo
303 [19830] network: nss_dns: should check RDATA length against buffer length
304 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
305 confronted with A/AAAA records of invalid size
306 [19837] nss: nss_db: No retries for some long lines with a larger buffer
307 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
308 [19853] stdio: Printing IBM long double in decimal with high precision is
310 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
311 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
312 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
314 [19865] network: Assertion failure or memory leak in
315 _nss_dns_getcanonname_r
316 [19868] network: nss_dns: netent code does not skip over non-PTR records
317 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
319 [19881] string: Improve x86-64 memset
320 [19907] string: Incorrect memcpy tests
321 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
323 [19925] libc: termios.h XCASE namespace
324 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
325 [19929] libc: limits.h NL_NMAX namespace
326 [19931] stdio: Memory leak in vfprintf
327 [19957] libc: clone(CLONE_VM) access invalid parent memory
328 [19963] localedata: en_IL: New locale
329 [19989] stdio: stdio.h cuserid namespace
330 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
332 [19996] locale: langinfo.h nl_langinfo_l namespace
333 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
334 [20010] network: getaddrinfo: Stack overflow in hostent translation
336 [20012] stdio: libio: fmemopen append mode failure
337 [20014] stdio: stdio.h namespace for pre-threads POSIX
338 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
339 [20023] libc: fcntl.h timespec namespace
340 [20024] math: [x86_64] vectorized sincos trashes the stack
341 [20031] network: nss_hesiod: Heap overflow in get_txt_records
342 [20041] time: sys/time.h timespec namespace
343 [20043] libc: unistd.h missing cuserid for UNIX98 and before
344 [20044] libc: unistd.h missing pthread_atfork for UNIX98
345 [20051] libc: ttyslot in wrong header under wrong conditions
346 [20054] libc: gethostname not declared for XPG4
347 [20055] libc: termios.h missing tcgetsid for XPG4
348 [20072] dynamic-link: x86 init_cpu_features is called twice in static
350 [20073] libc: sys/stat.h fchmod namespace
351 [20074] libc: stdlib.h rand_r namespace
352 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
353 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
355 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
356 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
358 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
359 [20119] libc: Wrong mask for processors level type from CPUID
360 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
361 AS not supporting AVX512
362 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
364 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
365 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
367 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
368 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
369 [20178] libc: posix_spawn{p} should not call exit
370 [20191] stdio: libio: vtables hardening
371 [20195] string: FMA4 detection requires CPUID execution with register
373 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
374 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
376 [20212] math: acos (sNaN) returns sNaN
377 [20213] math: asin (sNaN) returns sNaN
378 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
379 [20218] math: [i386] asinhl (sNaN) returns sNaN
380 [20219] math: [i386] atanhl (sNaN) returns sNaN
381 [20222] stdio: fopencookie: Mangle function pointers
382 [20224] math: [i386] cbrtl (sNaN) returns sNaN
383 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
384 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
386 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
387 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
388 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
389 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
390 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
391 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
392 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
393 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
394 [20240] math: modf (sNaN) returns sNaN
395 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
396 [20250] math: frexp (sNaN) returns sNaN
397 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
398 [20255] math: [i386] fdim, fdimf return with excess range and precision /
400 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
401 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
402 subscript is below array bounds [-Werror=array-bounds]
403 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
404 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
406 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
407 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
409 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
411 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
412 [20316] localedata: id_ID: Februari instead of Pebruari
413 [20327] string: POWER8 strcasecmp returns incorrect result
414 [20347] math: Failure: Test: j0_downward (0xap+0)
415 [20348] libc: FAIL: misc/tst-preadvwritev64
416 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
417 [20350] libc: There is no test for p{read,write}64
418 [20357] math: Incorrect cos result for 1.5174239687223976
419 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
423 * Unicode 8.0.0 Support: Character encoding, character type info, and
424 transliteration tables are all updated to Unicode 8.0.0, using new
425 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
426 These updates cause user visible changes, such as the fixes for bugs
427 89, 16061, and 18568.
429 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
430 kernel-internal CPU set size. This means that requests that change the
431 CPU affinity which failed before (for example, an all-ones CPU mask) will
432 now succeed. Applications that need to determine the effective CPU
433 affinities need to call sched_getaffinity or pthread_getaffinity_np after
434 setting it because the kernel can adjust it (and the previous size check
435 would not detect this in the majority of cases).
437 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
438 the following new symbols are used: fts64_children, fts64_close,
439 fts64_open, fts64_read and fts64_set.
441 * getaddrinfo now detects certain invalid responses on an internal netlink
442 socket. If such responses are received, an affected process will
443 terminate with an error message of "Unexpected error <number> on netlink
444 descriptor <number>" or "Unexpected netlink response of size <number> on
445 descriptor <number>". The most likely cause for these errors is a
446 multi-threaded application which erroneously closes and reuses the netlink
447 file descriptor while it is used by getaddrinfo.
449 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
450 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
451 unnecessary serialization of memory allocation requests across threads.
452 The defect is now corrected. Users should see a substantial increase in
453 the concurent throughput of allocation requests for applications which
454 trigger this bug. Affected applications typically create and
455 destroy threads frequently. (Bug 19048 was reported and analyzed by
458 * There is now a --disable-timezone-tools configure option for disabling the
459 building and installing of the timezone related utilities (zic, zdump, and
460 tzselect). This is useful for people who build the timezone data and code
461 independent of the GNU C Library.
463 * The obsolete header <regexp.h> has been removed. Programs that require
464 this header must be updated to use <regex.h> instead.
466 * The obsolete functions bdflush, create_module, get_kernel_syms,
467 query_module and uselib are no longer available to newly linked binaries;
468 the header <sys/kdaemon.h> has been removed. These functions and header
469 were specific to systems using the Linux kernel and could not usefully be
470 used with the GNU C Library on systems with version 2.6 or later of the
473 * Optimized string, wcsmbs and memory functions for IBM z13.
474 Implemented by Stefan Liebler.
476 * Newly linked programs that define a variable called signgam will no longer
477 have it set by the lgamma, lgammaf and lgammal functions. Programs that
478 require signgam to be set by those functions must ensure that they use the
479 variable provided by the GNU C Library and declared in <math.h>, without
480 defining their own copy.
482 * The minimum GCC version that can be used to build this version of the GNU
483 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
484 still be used to compile programs using the GNU C Library.
486 Security related changes:
488 * An out-of-bounds value in a broken-out struct tm argument to strftime no
489 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
491 * The LD_POINTER_GUARD environment variable can no longer be used to disable
492 the pointer guard feature. It is always enabled. Previously,
493 LD_POINTER_GUARD could be used to disable security hardening in binaries
494 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
497 * An integer overflow in hcreate and hcreate_r could lead to an
498 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
500 * The catopen function no longer has unbounded stack usage. Reported by
503 * The nan, nanf and nanl functions no longer have unbounded stack usage
504 depending on the length of the string passed as an argument to the
505 functions. Reported by Joseph Myers. (CVE-2014-9761)
507 * A stack-based buffer overflow was found in libresolv when invoked from
508 libnss_dns, allowing specially crafted DNS responses to seize control
509 of execution flow in the DNS client. The buffer overflow occurs in
510 the functions send_dg (send datagram) and send_vc (send TCP) for the
511 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
512 family. The use of AF_UNSPEC triggers the low-level resolver code to
513 send out two parallel queries for A and AAAA. A mismanagement of the
514 buffers used for those queries could result in the response of a query
515 writing beyond the alloca allocated buffer created by
516 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
517 the overflow. Thanks to the Google Security Team and Red Hat for
518 reporting the security impact of this issue, and Robert Holiday of
519 Ciena for reporting the related bug 18665. (CVE-2015-7547)
521 The following bugs are resolved with this release:
523 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
524 [887] math: Math library function "logb" and "nextafter" inconsistent
525 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
526 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
527 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
528 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
530 [4404] localedata: German translation of "Alarm clock" is misleading
531 [6799] math: nextafter() and nexttoward() doen't set errno on
532 overflow/underflow errors
533 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
535 [10432] nis: _nss_nis_setnetgrent assertion failure
536 [11460] libc: fts has no LFS support
537 [12926] network: getaddrinfo()/make_request() may spin forever
538 [13065] nptl: Race condition in pthread barriers
539 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
540 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
542 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
543 [14912] libc: Rename non-installed bits/*.h headers
544 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
545 [15367] math: Let gcc use __builtin_isinf
546 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
547 [15421] math: lgamma wrongly sets signgam for ISO C
548 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
549 FE_INVALID with argument out of range
550 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
552 [15786] dynamic-link: ifunc resolver functions can smash function
554 [15918] math: Unnecessary check for equality in hypotf()
555 [16061] localedata: Review / update transliteration data
556 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
558 [16141] time: strptime %z offset restriction
559 [16171] math: drem should be alias of remainder
560 [16296] math: fegetround is pure?
561 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
562 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
564 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
565 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
567 [16422] math: [powerpc] math-float, math-double failing llrint tests with
568 "Exception "Inexact" set" on ppc32
569 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
570 [16517] math: Missing underflow exception from tanf/tan/tanl
571 [16519] math: Missing underflow exception from sinhf
572 [16520] math: Missing underflow exception from tanhf
573 [16521] math: Missing underflow exception from exp2
574 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
576 [16734] stdio: fopen calls mmap to allocate its buffer
577 [16961] math: nan function incorrect handling of bad sequences
578 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
579 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
580 [16985] locale: localedef: confusing error message when opening output
582 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
583 [17197] locale: Redundant shift character in iconv conversion output at
585 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
586 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
587 [17250] dynamic-link: static linking breaks nss loading
588 (getaddrinfo/getpwnam/etc...)
589 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
591 [17441] math: isnan() should use __builtin_isnan() in GCC
592 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
593 (related to lock elision)
594 [17787] manual: Exponent on page 324 of the PDF ends prematurely
595 [17886] time: strptime should be able to parse "Z" as a timezone with %z
596 [17887] time: strptime should be able to parse "+01:00" style timezones
597 [17905] libc: catopen() Multiple unbounded stack allocations
599 [18084] libc: backtrace (..., 0) dumps core on x86
600 [18086] libc: nice() sets errno to 0 on success
601 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
602 is too large (CVE-2015-8778)
603 [18251] dynamic-link: SONAME missing when audit modules provides path
604 [18265] libc: add attributes for wchar string and memory functions
605 [18370] math: csqrt missing underflows
606 [18421] libc: [hppa] read-only segment has dynamic relocations
607 [18472] libc: Obsolete syscall wrappers should be compat symbols
608 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
609 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
610 [18525] localedata: Remove locale timezone information
611 [18560] libc: [powerpc] spurious bits/ipc.h definitions
612 [18568] localedata: Update locale data to Unicode 8.0
613 [18589] locale: sort-test.sh fails at random
614 [18595] math: ctan, ctanh missing underflows
615 [18604] libc: assert macro-expands its argument
616 [18610] math: S390: fetestexcept() reports any exception if DXC-code
617 contains a vector instruction exception.
618 [18611] math: j1, jn missing errno setting on underflow
619 [18618] localedata: sync Chechen locale definitions with other *_RU
621 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
622 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
623 [18665] network: In send_dg, the recvfrom function is NOT always using the
624 buffer size of a newly created buffer (CVE-2015-7547)
625 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
627 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
629 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
630 [18699] math: tilegx cproj() for various complex infinities does not yield
632 [18724] libc: Harden put*ent functions against data injection
633 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
634 [18755] build: build errors with -DNDEBUG
635 [18757] stdio: fmemopen fails to set errno on failure
636 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
638 [18781] libc: openat64 lacks O_LARGEFILE
639 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
640 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
641 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
642 [18790] math: [ldbl-128ibm] tanhl inaccurate
643 [18795] libc: stpncpy fortification misses buffer lengths that are
645 [18796] build: build fails for --disable-mathvec
646 [18803] math: hypot missing underflows
647 [18820] stdio: fmemopen may leak memory on failure
648 [18823] math: csqrt spurious underflows
649 [18824] math: fma spurious underflows
650 [18825] math: pow missing underflows
651 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
652 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
653 [18870] build: sem_open.c fails to compile with missing symbol
655 [18872] stdio: Fix memory leak in printf_positional
656 [18873] libc: posix_fallocate overflow check ineffective
657 [18875] math: Excess precision leads incorrect libm
658 [18877] libc: arm: mmap offset regression
659 [18887] libc: memory corruption when using getmntent on blank lines
660 [18918] localedata: hu_HU: change time to HH:MM:SS format
661 [18921] libc: Regression: extraneous stat() and fstat() performed by
663 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
664 binaries (CVE-2015-8777)
665 [18951] math: tgamma missing underflows
666 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
668 [18953] localedata: lt_LT: change currency symbol to the euro
669 [18956] math: powf inaccuracy
670 [18961] math: [i386] exp missing underflows
671 [18966] math: [i386] exp10 missing underflows
672 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
673 [18969] build: multiple string test failures due to missing locale
675 [18970] libc: Reference of pthread_setcancelstate in libc.a
676 [18977] math: float / long double Bessel functions not in XSI POSIX
677 [18980] math: i386 libm functions return with excess range and precision
678 [18981] math: i386 scalb*, ldexp return with excess range and precision
679 [18982] stdio: va_list and vprintf
680 [18985] time: Passing out of range data to strftime() causes a segfault
682 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
683 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
684 [19012] locale: iconv_open leaks memory on error path
685 [19016] math: clog, clog10 inaccuracy
686 [19018] nptl: Mangle function pointers in tls_dtor_list
687 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
688 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
689 [19048] malloc: malloc: arena free list can become cyclic, increasing
691 [19049] math: [powerpc] erfc incorrect zero sign
692 [19050] math: [powerpc] log* incorrect zero sign
693 [19058] math: [x86_64] Link fail with -fopenmp and -flto
694 [19059] math: nexttoward overflow incorrect in non-default rounding modes
695 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
696 [19074] network: Data race in _res_hconf_reorder_addrs
697 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
698 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
699 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
701 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
703 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
704 [19086] manual: posix_fallocate64 documented argument order is wrong.
705 [19088] math: lround, llround missing exceptions close to overflow
707 [19094] math: lrint, llrint missing exceptions close to overflow threshold
708 [19095] math: dbl-64 lrint incorrect for 64-bit long
709 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
710 [19124] dynamic-link: ld.so failed to build with older assmebler
711 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
712 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
713 [19134] math: [powerpc32] lround, lroundf spurious exceptions
714 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
715 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
716 pthread_setaffinity_np
717 [19156] math: [ldbl-128] j0l spurious underflows
718 [19164] nptl: tst-getcpu fails with many possible CPUs
719 [19168] math: math/test-ildoubl and math/test-ldouble failure
720 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
721 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
723 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
724 (FE_NOMASK_ENV) do not clear SSE exceptions
725 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
726 _int_new_arena/reused_arena
727 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
728 [19201] math: dbl-64 remainder incorrect sign of zero result
729 [19205] math: bits/math-finite.h conditions do not match math.h and
731 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
732 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
733 for C99-based standards
734 [19212] libc: features.h not -Wundef clean
735 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
737 [19214] libc: Family and model identification for AMD CPU's are incorrect.
738 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
739 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
741 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
743 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
745 [19242] libc: strtol incorrect in Turkish locales
746 [19243] malloc: reused_arena can pick an arena on the free list, leading
747 to an assertion failure and reference count corruption
748 [19253] time: tzset() ineffective when temporary TZ did not include DST
750 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
751 [19270] math: [hppa] Shared libm missing __isnanl
752 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
753 MAP_HUGETLB and MAP_STACK defines
754 [19313] nptl: Wrong __cpu_mask for x32
755 [19347] libc: grantpt: try to force a specific gid even without pt_chown
756 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
757 [19350] math: [ldbl-128ibm] sinhl spurious overflows
758 [19351] math: [ldbl-128ibm] logl inaccurate near 1
759 [19363] time: x32: times() return value wrongly truncates/sign extends
761 [19367] dynamic-link: Improve branch prediction on Silvermont
762 [19369] network: Default domain name not reset by res_ninit when "search"
763 / "domain" entry is removed from resolv.conf
764 [19375] math: powerpc: incorrect results for POWER7 logb with negative
766 [19385] localedata: bg_BG: time separator should be colon, not comma
767 [19408] libc: linux personality syscall wrapper may erroneously return an
768 error on 32-bit architectures
769 [19415] libc: dladdr returns wrong names on hppa
770 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
771 IBM905, IBM907, and IBM909
772 [19439] math: Unix98 isinf and isnan functions conflict with C++11
773 [19443] build: build failures with -DDEBUG
774 [19451] build: Make check fails on test-double-vlen2
775 [19462] libc: Glibc failed to build with -Os
776 [19465] math: Wrong code with -Os
777 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
779 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
781 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
782 [19486] math: S390: Math tests fail with "Exception Inexact set".
783 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
784 [19550] libc: [mips] mmap negative offset handling inconsistent with other
786 [19590] math: Fail to build shared objects that use libmvec.so functions.
790 * The following bugs are resolved with this release:
792 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
793 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
794 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
795 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
796 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
797 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
798 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
799 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
800 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
801 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
802 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
803 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
804 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
805 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
806 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
807 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
808 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
809 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
810 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
811 18657, 18676, 18694, 18696, 18887.
813 * Cache information can be queried via sysconf() function on s390 e.g. with
814 _SC_LEVEL1_ICACHE_SIZE as argument.
816 * A buffer overflow in gethostbyname_r and related functions performing DNS
817 requests has been fixed. If the NSS functions were called with a
818 misaligned buffer, the buffer length change due to pointer alignment was
819 not taken into account. This could result in application crashes or,
820 potentially arbitrary code execution, using crafted, but syntactically
821 valid DNS responses. (CVE-2015-1781)
823 * The time zone file parser has been made more robust against crafted time
824 zone files, avoiding heap buffer overflows related to the processing of
825 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
826 large time zone data files. Overly long time zone specifiers in the TZ
827 variable no longer result in stack overflows and crashes.
829 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
830 for LD and GD on x86 and x86-64, has been implemented. You will need
831 binutils-2.24 or later to enable this optimization.
833 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
834 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
835 Hat). These updates cause user visible changes, such as the fix for bug
838 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
839 and getXXbyYY NSS calls for the same database, causing a denial-of-service
840 condition in some applications.
842 * Added vector math library named libmvec with the following vectorized x86_64
843 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
845 The library can be disabled with --disable-mathvec. Use of the functions is
846 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
847 Shared library libmvec.so is linked in as needed when using -lm (no need to
848 specify -lmvec explicitly for not static builds).
849 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
851 * A new fmemopen implementation has been added with the goal of POSIX
852 compliance. The new implementation fixes the following long-standing
853 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
854 old implementation is still present for use be by existing binaries.
856 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
857 releases. It has been fixed to match 2.19 and older, but binaries built
858 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
860 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
861 Contributed by Roland McGrath (Google).
863 * The header <regexp.h> is deprecated, and will be removed in a future
864 release. Use of this header will trigger a deprecation warning.
865 Application developers should update their code to use <regex.h> instead.
867 This header was formerly part of SUS, but was deprecated in 1994 and
868 removed from the standard in 2001. Also, the glibc implementation
869 leaks memory. See BZ#18681 for more details.
873 * The following bugs are resolved with this release:
875 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
876 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
877 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
878 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
879 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
880 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
881 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
882 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
883 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
884 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
887 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
888 too little memory for the to-be-scanned arguments and overflow the
889 allocated buffer. The implementation now correctly computes the required
890 buffer size when using malloc, and switches to malloc from alloca as
893 * A new semaphore algorithm has been implemented in generic C code for all
894 machines. Previous custom assembly implementations of semaphore were
895 difficult to reason about or ensure that they were safe. The new version
896 of semaphore supports machines with 64-bit or 32-bit atomic operations.
897 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
898 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
900 * Port to Altera Nios II has been contributed by Mentor Graphics.
902 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
903 implementations for powerpc64/powerpc64le.
904 Implemented by Adhemerval Zanella (IBM).
906 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
907 and powerpc64le. This may improve lock scaling of existing programs on
908 HTM capable systems. The lock elision code is only enabled with
909 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
910 powerpc will issue a transaction abort on every syscall to avoid side
911 effects being visible outside transactions.
913 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
914 AArch64. Contributed by ARM Ltd.
916 * i386 memcpy functions optimized with SSE2 unaligned load/store.
918 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
919 under certain input conditions resulting in the execution of a shell for
920 command substitution when the application did not request it. The
921 implementation now checks WRDE_NOCMD immediately before executing the
922 shell and returns the error WRDE_CMDSUB as expected.
924 * CVE-2012-3406 printf-style functions could run into a stack overflow when
925 processing format strings with a large number of format specifiers.
927 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
928 infinite loop if the DNS response contained a PTR record of an unexpected
931 * The minimum GCC version that can be used to build this version of the GNU
932 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
933 still be used to compile programs using the GNU C Library.
935 * The GNU C Library is now built with -Werror by default. This can be
936 disabled by configuring with --disable-werror.
938 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
940 * The obsolete sigvec function has been removed. This was the original
941 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
942 programs have been using instead for about 25 years. Of course, ABI
943 compatibility for old binaries using sigvec remains intact.
945 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
946 with newer versions of bison.
948 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
949 The original MIPS o32 hard-float ABI requires an FPU where double-precision
950 registers overlay two consecutive single-precision registers. MIPS32R2
951 introduced a new FPU mode (FR=1) where double-precision registers extend the
952 corresponding single-precision registers which is incompatible with the
953 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
954 require the use of FR=1 making a transition necessary. New o32 ABI
955 extensions enable users to migrate over time from the original o32 ABI
956 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
957 tracks the ABI of any loaded object and verifies that new objects are
958 compatible. Mode transitions will also be requested as required and
959 unsupportable objects will be rejected. The ABI checks include both soft and
960 hard float ABIs for o32, n32 and n64.
962 GCC 5 with GNU binutils 2.25 onwards:
963 It is strongly recommended that all o32 system libraries are built using the
964 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
965 with the original and all new o32 ABI extensions. Configure a MIPS GCC
966 compiler using --with-fp-32=xx to set this by default.
970 * The following bugs are resolved with this release:
972 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
973 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
974 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
975 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
976 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
977 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
978 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
979 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
980 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
981 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
982 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
983 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
984 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
985 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
986 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
988 * Reverted change of ABI data structures for s390 and s390x:
989 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
990 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
991 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
992 are preserved pointing straight to the same implementation as the old ones.
993 Given that, new callers will simply provide a too-big buffer to these
994 functions. Any applications/libraries out there that embed jmp_buf or
995 ucontext_t in an ABI-relevant data structure that have already been rebuilt
996 against 2.19 headers will have to rebuilt again. This is necessary in any
997 case to revert the breakage in their ABI caused by the glibc change.
999 * Support for file description locks is added to systems running the
1000 Linux kernel. The standard file locking interfaces are extended to
1001 operate on file descriptions, not file descriptors, via the use of
1002 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1003 are associated with an open file instead of a process.
1005 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1007 * The minimum Linux kernel version that this version of the GNU C Library
1008 can be used with is 2.6.32.
1010 * Running the testsuite no longer terminates as soon as a test fails.
1011 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1012 with PASS or FAIL lines for individual tests. A summary of the results is
1013 printed, including a list of failing lists, and "make check" exits with
1014 error status if there were any unexpected failures. "make check
1015 stop-on-test-failure=y" may be used to keep the old behavior.
1017 * The am33 port, which had not worked for several years, has been removed
1020 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1021 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1022 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1023 conflicted with POSIX (support for which was removed in 2.19), the
1024 interfaces those macros enabled remain available when compiling with
1025 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1026 test macros defined.
1028 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1030 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1031 This may improve lock scaling of existing programs on TX capable systems.
1032 The lock elision code is only built with --enable-lock-elision=yes and
1033 then requires a GCC version supporting the TX builtins. With lock elision
1034 default mutexes are elided via __builtin_tbegin, if the cpu supports
1035 transactions. By default lock elision is not enabled and the elision code
1038 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1039 copy the path argument. This allowed programs to cause posix_spawn to
1040 deference a dangling pointer, or use an unexpected pathname argument if
1041 the string was modified after the posix_spawn_file_actions_addopen
1044 * All supported architectures now use the main glibc sysdeps directory
1045 instead of some being in a separate "ports" directory (which was
1046 distributed separately before glibc 2.17).
1048 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1049 On configurations that support it (all Linux configurations), it's now
1050 used regardless of the --enable-add-ons switch to configure. It is no
1051 longer possible to build such configurations without pthreads support.
1053 * Locale names, including those obtained from environment variables (LANG
1054 and the LC_* variables), are more tightly checked for proper syntax.
1055 setlocale will now fail (with EINVAL) for locale names that are overly
1056 long, contain slashes without starting with a slash, or contain ".." path
1057 components. (CVE-2014-0475) Previously, some valid locale names were
1058 silently replaced with the "C" locale when running in AT_SECURE mode
1059 (e.g., in a SUID program). This is no longer necessary because of the
1062 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1063 with application code using Intel MPX instructions. (With all previous
1064 versions, the MPX register state could be clobbered when making calls
1065 into or out of a shared library.) Note that while the new dynamic
1066 linker is compatible with all known x86 hardware whether or not it
1067 supports Intel MPX, some x86 instruction-set emulators might fail to
1068 handle the new instruction encodings. This is known to affect Valgrind
1069 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1070 release), and might affect other tools that do instruction emulation.
1072 * Support for loadable gconv transliteration modules has been removed.
1073 The support for transliteration modules has been non-functional for
1074 over a decade, and the removal is prompted by security defects. The
1075 normal gconv conversion modules are still supported. Transliteration
1076 with //TRANSLIT is still possible, and the //IGNORE specifier
1077 continues to be supported. (CVE-2014-5119)
1079 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1080 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1081 resulting a denial-of-service security vulnerability in applications which
1082 use functions related to iconv. (CVE-2014-6040)
1086 * The following bugs are resolved with this release:
1088 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1089 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1090 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1091 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1092 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1093 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1094 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1095 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1096 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1097 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1098 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1099 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1100 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1101 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1102 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1103 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1104 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1105 16431, 16453, 16474, 16506, 16510, 16529
1107 * Slovenian translations for glibc messages have been contributed by the
1108 Translation Project's Slovenian team of translators.
1110 * The public headers no longer use __unused nor __block. This change is to
1111 support compiling programs that are derived from BSD sources and use
1112 __unused internally, and to support compiling with Clang's -fblock
1113 extension which uses __block.
1115 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1116 large collation sequences to optimize multiple passes. This cache
1117 computation may overflow for large collation sequences and may cause a
1118 stack or buffer overflow. This is now fixed to use a slower algorithm
1119 which does not use a cache if there is an integer overflow.
1121 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1122 rules for large collation sequences to optimize multiple passes and falls
1123 back to alloca if malloc fails, resulting in a possible stack overflow.
1124 The implementation now falls back to an uncached collation sequence lookup
1127 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1128 initialized for static applications resulting in the security feature
1129 being disabled. The pointer guard is now correctly initialized to a
1130 random value for static applications. Existing static applications need
1131 to be recompiled to take advantage of the fix (bug 15754).
1133 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1134 to the d_name member of struct dirent, or omit the terminating NUL
1135 character. (Bugzilla #14699).
1137 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1138 aligned_alloc functions could allocate too few bytes or corrupt the
1139 heap when passed very large allocation size values (Bugzilla #15855,
1142 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1143 for AF_INET6 has been fixed (Bugzilla #16072).
1145 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1146 pap_CW, quz_PE, the_NP.
1148 * Substantially revised locales: gd_GB, ht_HT
1150 * The LC_ADDRESS field was updated to support country_car for almost all
1153 * ISO 1427 definitions were updated.
1155 * ISO 3166 definitions were updated.
1157 * The localedef utility now supports --big-endian and --little-endian
1158 command-line options to generate locales for a different system from that
1159 for which the C library was built.
1161 * Binary locale files now only depend on the endianness of the system for
1162 which they are generated and not on other properties of that system. As a
1163 consequence, binary files generated with new localedef may be incompatible
1164 with old versions of the GNU C Library, and binary files generated with
1165 old localedef may be incompatible with this version of the GNU C Library,
1166 in the following circumstances:
1168 + Locale files may be incompatible on m68k systems.
1170 + Locale archive files (but not separate files for individual locales) may
1171 be incompatible on systems where plain "char" is signed.
1173 * The configure option --disable-versioning has been removed. Builds with
1174 --disable-versioning had not worked for several years.
1176 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1177 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1179 * SystemTap probes for malloc have been introduced.
1181 * SystemTap probes for slow multiple precision fallback paths of
1182 transcendental functions have been introduced.
1184 * Support for powerpc64le has been added.
1186 * The soft-float powerpc port now supports e500 processors.
1188 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1190 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1191 set of header declarations that are enabled by default, even when other
1192 feature test macros or compiler options such as -std=c99 would otherwise
1193 disable some of those declarations.
1195 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1196 conflict with POSIX. The libbsd-compat library (which was a dummy library
1197 that did nothing) has also been removed.
1199 * Preliminary documentation about Multi-Thread, Async-Signal and
1200 Async-Cancel Safety has been added.
1202 * Change of ABI data structures for s390 and s390x:
1203 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1204 allow for future hardware extensions. All library functions that accept or
1205 return these structures were versioned in order to provide backwards
1206 compatibility with existing code. However, not all cases can be handled
1207 with interface versioning. If an application embeds either structure into
1208 another structure and that structure is passed to another compilation unit
1209 which expects the newer larger-sized object then any access to the new
1210 fields will result in undefined behaviour. Similarly any access to fields
1211 that were moved by the enlarging of the structures will cause undefined
1212 behaviour. To fix the undefined behaviour all objects that are part of the
1213 final in-memory image must agree on the size of structures, and this may
1214 require recompilation.
1218 * The following bugs are resolved with this release:
1220 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1221 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1222 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1223 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1224 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1225 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1226 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1227 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1228 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1229 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1230 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1231 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1232 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1235 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1236 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1237 Distributions can re-enable building and using pt_chown via the new configure
1238 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1239 considerable security risks and should only be used if the distribution
1240 understands and accepts the risks.
1242 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1245 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1246 fixed (Bugzilla #15330).
1248 * Add support for calling C++11 thread_local object destructors on thread
1249 and program exit. This needs compiler support for offloading C++11
1250 destructor calls to glibc.
1252 * Improved worst case performance of libm functions with double inputs and
1255 * Support for priority inherited mutexes in pthread condition variables on
1256 non-x86 architectures.
1258 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1260 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1262 * Optimized string functions for ARM. Implemented by Will Newton and
1265 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1267 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1270 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1271 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1273 * Added a benchmark framework to track performance of functions in glibc.
1275 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1276 It is based on draft TS 18661 and currently enabled as a GNU extension.
1278 * On Linux, the clock function now uses the clock_gettime system call
1279 for improved precision, rather than old times system call.
1281 * Added support for version-3 tz format files. This is needed when using
1282 the tz database release 2013e or later, and affects a few unusual cases --
1283 currently only TZ='America/Godthab' for time stamps after 2037.
1285 * Added new API functions pthread_getattr_default_np and
1286 pthread_setattr_default_np to get and set the default pthread
1287 attributes of a process.
1289 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1290 This may improve lock scaling of existing programs on TSX capable systems.
1291 When the --enable-lock-elision=yes parameter is specified at configure
1292 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1295 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1296 availability). Implemented by Ryan S. Arnold.
1298 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1302 * The following bugs are resolved with this release:
1304 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1305 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1306 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1307 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1308 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1309 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1310 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1311 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1312 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1313 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1314 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1315 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1316 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1317 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1318 14879, 14889, 14893, 14898, 14914.
1320 * Optimization of memcpy for MIPS.
1322 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1323 EMFILE has been fixed (Bugzilla #14889).
1325 * The add-on ports collection is now distributed in the "ports" subdirectory
1326 of the main GNU C Library distribution, rather than separately.
1328 * Port to ARM AArch64 contributed by Linaro.
1330 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1331 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1333 Implemented by Andreas Krebbel.
1335 * The new function secure_getenv allows secure access to the environment,
1336 returning NULL if running in a SUID/SGID process. This function replaces
1337 the internal function __secure_getenv.
1339 * SystemTap static probes have been added into the dynamic linker.
1340 Implemented by Gary Benson.
1342 * Optimizations of string functions strstr, strcasestr and memmem.
1343 Implemented by Maxim Kuvyrkov.
1345 * The minimum Linux kernel version that this version of the GNU C Library
1346 can be used with is 2.6.16.
1348 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1349 powerpc POWER7. Implemented by Will Schmidt.
1351 * New configure option --disable-nscd builds the C library such that it
1352 never attempts to contact the Name Service Caching Daemon (nscd).
1353 New configure option --disable-build-nscd avoids building nscd itself;
1354 this is the default if --disable-nscd is used.
1356 * Improved support for cross-compilation, including cross-testing and
1357 bootstrap builds without a previously built glibc.
1359 * Several testsuite tests are now able to test multiple IFUNC variants of an
1360 interface, rather than just testing the one that would be chooen by
1363 * New configure options --with-bugurl and --with-pkgversion, for
1364 distributors to use to embed their bug-reporting and package version
1365 information in --help and --version output.
1367 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1368 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1369 allows creation of chroots without the procfs mounted on /proc.
1371 * The `crypt' function now fails if passed salt bytes that violate the
1372 specification for those values. On Linux, the `crypt' function will
1373 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1374 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1375 when the mode is enabled.
1377 * The `clock_*' suite of functions (declared in <time.h>) is now available
1378 directly in the main C library. Previously it was necessary to link with
1379 -lrt to use these functions. This change has the effect that a
1380 single-threaded program that uses a function such as `clock_gettime' (and
1381 is not linked with -lrt) will no longer implicitly load the pthreads
1382 library at runtime and so will not suffer the overheads associated with
1383 multi-thread support in other code such as the C++ runtime library.
1385 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1390 * The following bugs are resolved with this release:
1392 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1393 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1394 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1395 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1396 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1397 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1398 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1399 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1400 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1401 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1402 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1403 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1404 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1405 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1406 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1407 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1408 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1409 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1410 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1411 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1412 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1413 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1416 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1417 configuring glibc with:
1418 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1419 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1420 Implemented by H.J. Lu.
1424 + define static_assert
1426 + do not declare gets
1428 + declare at_quick_exit and quick_exit also for ISO C11
1430 + aligned_alloc. NB: The code is deliberately allows the size parameter
1431 to not be a multiple of the alignment. This is a moronic requirement
1432 in the standard but it is only a requirement on the caller, not the
1435 + timespec_get added
1437 + uchar.h support added
1439 + CMPLX, CMPLXF, CMPLXL added
1441 Implemented by Ulrich Drepper.
1443 * Support for the IA-64 has been moved to ports.
1445 * Remove support for anything but ELF binary format
1447 * Checking versions of poll, ppoll added.
1448 Implemented by Ulrich Drepper.
1450 * More generic and 64-bit performance optimizations to math functions.
1451 Implemented by Ulrich Drepper.
1453 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1454 headers and functions available at compile time as they were before
1455 version 2.14. This option will be removed at some time in the future
1456 after the TI-RPC library becomes fully sufficient for the needs of
1457 existing applications.
1459 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1460 Note that glibc is not expected to work with any Linux kernel version
1463 * New header <sys/auxv.h> and function getauxval allowing easy access to
1464 the AT_* key-value pairs passed from the Linux kernel. The header also
1465 defines the HWCAP_* bits associated with the AT_HWCAP key.
1467 * New locales: mag_IN
1469 * New configure option --enable-systemtap builds SystemTap static probes
1470 into libc for setjmp and longjmp and into libpthread for various operations.
1471 So far the setjmp/longjmp probes and some of the libpthread probes are
1472 provided only for i*86 and x86_64.
1473 Implemented by Roland McGrath and Rayson Ho.
1475 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1477 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1480 * More optimized functions for SPARC. Implemented by David S. Miller.
1482 * Improved support for cross-compilation, especially bootstrap builds
1483 without a previously built glibc.
1485 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1486 Chris Metcalf from Tilera.
1488 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1489 now supported for ARM processors.
1491 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1492 as the name of the dynamic linker, to distinguish it from the
1493 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1495 * Support for CRIS has been removed from ports.
1497 * A new class of installed header has been documented for low-level
1498 platform-specific functionality. PowerPC added the first instance with a
1499 function to provide time base register access. Contributed by Tulio
1500 Magno Quites Machado Filho.
1502 * ix86 configurations now install header files that are consistent with
1503 what x86-64 configurations install. These same header files can be used
1504 for -m32, -m64, or -mx32 builds.
1505 Contributed by H.J. Lu.
1507 * Math library bug fixes. A thorough audit of all open math library bugs was
1508 conducted by Joseph Myers. Significant progress was made on many math
1509 library bugs resulting in more accurate exceptions and function results.
1510 Many thanks to all those that contributed including Andreas Jaeger for his
1511 patch review and work on the x87 trigonometric instruction issues.
1513 * Timezone data is no longer installed. Timezone-related binaries and scripts
1514 will continue to be installed. Users should obtain their timezone data from
1515 their distribution provider or from the tzdata package at
1516 <ftp://munnari.oz.au/pub/>.
1520 * The following bugs are resolved with this release:
1522 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1523 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1524 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1525 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1526 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1527 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1528 13484, 13506, 13515, 13523, 13524, 13538, 13540
1530 * New program pldd to list loaded object of a process
1531 Implemented by Ulrich Drepper.
1533 * Add nss_db support back to glibc. No more dependency on Berkeley db
1534 and support for initgroups lookups.
1535 Implemented by Ulrich Drepper.
1537 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1538 Contributed by HJ Lu.
1540 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1541 Contributed by HJ Lu.
1543 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1544 on x86-32 and x86-64.
1545 Contributed by Liubov Dmitrieva.
1547 * Optimized strchr and strrchr for SSE on x86-32.
1548 Contributed by Liubov Dmitrieva.
1550 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1551 for x86-64 and x86-32.
1552 Contributed by Liubov Dmitrieva.
1554 * New interfaces: scandirat, scandirat64
1555 Implemented by Ulrich Drepper.
1557 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1558 Implemented by Ulrich Drepper.
1560 * nscd now also caches the netgroup database.
1561 Implemented by Ulrich Drepper.
1563 * Integrate libm with gcc's -ffinite-math-only option.
1564 Implemented by Ulrich Drepper.
1566 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1567 to math functions. Implemented by Ulrich Drepper.
1569 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1570 Implemented by Ulrich Drepper.
1572 * New Linux interfaces: process_vm_readv, process_vm_writev
1574 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1575 Implemented by Ulrich Drepper.
1577 * Optimized nearbyint and strcasecmp for PPC.
1578 Implemented by Adhemerval Zanella.
1580 * New locales: bho_IN, unm_US, es_CU, ta_LK
1584 * The following bugs are resolved with this release:
1586 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1587 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1588 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1589 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1590 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1591 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1592 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1593 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1594 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1595 12795, 12811, 12813, 12814, 12841
1597 * The RPC implementation in libc is obsoleted. Old programs keep working
1598 but new programs cannot be linked with the routines in libc anymore.
1599 Programs in need of RPC functionality must be linked against TI-RPC.
1600 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1602 Visible changes of this change include (obviously) the inability to link
1603 programs using RPC functions without referencing the TI-RPC library and the
1604 removal of the RPC headers from the glibc headers.
1605 Implemented by Ulrich Drepper.
1607 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1608 syncfs, setns, sendmmsg
1610 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1611 yue_HK, lij_IT, mhr_RU
1613 * New encodings: CP770, CP771, CP772, CP773, CP774
1615 * New program sotruss to trace calls through PLTs
1616 Implemented by Ulrich Drepper.
1618 * The malloc hook implementation is marked deprecated and will be removed
1619 from the default implementation in the next version. The design never
1620 worked ever since the introduction of threads. Even programs which do
1621 not create threads themselves can use multiple threads created internally.
1625 * The following bugs are resolved with this release:
1627 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1628 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1629 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1630 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1633 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1635 * POWER7 optimizations: memset, memcmp, strncmp
1637 * New optimized string functions for x86-64: strnlen (SSE2),
1638 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1639 Implemented by Ulrich Drepper.
1643 * The following bugs are resolved with this release:
1645 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1646 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1647 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1648 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1649 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1650 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1651 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1652 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1654 * New interfaces: pthread_getname_np, pthread_setname_np
1656 * New Linux interface: recvmmsg
1658 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1660 * The dynamic linker now recognizes supported ABI versions from the
1661 EI_ABIVERSION field in the ELF header.
1662 Implemented by Ulrich Drepper.
1664 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1665 will not be used to fill in password fields in the passwd.byname replies.
1666 Instead it is used to synthesize the shadow.byname table, should it be
1667 missing. This is a useful mode in some installations involving Solaris.
1668 Implemented by Ulrich Drepper.
1670 * New locales: kok_IN, sq_MK, cv_RU
1674 * The following bugs are resolved with this release:
1676 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1677 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1678 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1679 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1680 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1681 10789, 10817, 10823, 10840, 10847
1683 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1685 Implemented by Ulrich Drepper.
1687 * Checking version of longjmp added that fails if an uninitialized stack
1688 frame would be created. Implemented by Ulrich Drepper.
1690 * STT_GNU_IFUNC is now supported in static executables.
1691 Implemented by H.J. Lu.
1693 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1695 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1696 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1697 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1698 Contributed by H.J. Lu.
1700 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1701 Implemented by Ulrich Drepper.
1703 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1705 Contributed by H.J. Lu.
1707 * Support for fma instruction in AVX on x86-64.
1708 Implemented by H.J. Lu and Ulrich Drepper.
1710 * AVX support in x86-64 auditing support in ld.so.
1711 Implemented by H.J. Lu.
1713 * STB_GNU_UNIQUE support added.
1714 Implemented by Ulrich Drepper.
1716 * Implement second fallback mode for optimized DNS lookup for even more
1717 broken environments. If two requests from the same port are not
1718 handled correctly close the socket and open a new one before sending
1719 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1720 can be used to select this mode right away, instead of rediscovering the
1721 necessity is every process again.
1722 Implemented by Ulrich Drepper.
1724 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1725 Implemented by Adam Tkac.
1727 * Optimized iconv conversions for S390x.
1728 Implemented by Andreas Krebbel.
1730 * Using condvars with PI mutexes is now more efficient due to kernel
1731 support for requeueing to PI futexes. NPTL support added for x86-64.
1732 Implemented by Ulrich Drepper.
1734 * New locales: ps_AF, my_MM
1738 * The following bugs are resolved with this release:
1740 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1741 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1742 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1743 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1744 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1746 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1747 Implemented by Ulrich Drepper.
1749 * New Linux interfaces: accept4, fallocate, fallocate64.
1750 Implemented by Ulrich Drepper.
1752 * Correct declarations of string function when used in C++ code. This
1753 could lead to compile errors for invalid C++ code.
1755 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1758 * New POSIX 2008 interface: psiginfo
1759 Implemented by Ulrich Drepper.
1761 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1762 Implemented by Ulrich Drepper.
1764 * Support for selecting between multiple function definitions at runtime
1765 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1767 * The libcrypt library can now use the hash function implementations in
1768 NSS. Implemented by Ulrich Drepper.
1770 * The malloc implementation can be compiled to be less memory efficient
1771 but higher performing in multi-threaded programs.
1772 Implemented by Ulrich Drepper.
1774 * New locales: nan_TW@latin, ks_IN
1776 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1777 Implemented by Ulrich Drepper.
1779 * Extended printf hook support. It is possible to use user-defined types
1780 and extend existing format specifiers.
1781 Implemented by Ulrich Drepper.
1783 * Handling for group shadow files has been added.
1784 Implemented by Ulrich Drepper.
1786 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1787 servers (the case, e.g., for some people using the built-in DNS
1788 server in ADSL modems/routers). There is a once-per-process timeout
1789 in case of a broken server. To avoid it, users can run nscd or put
1790 'options single-request' in /etc/resolv.conf.
1791 Implemented by Ulrich Drepper.
1795 * The following bugs are resolved with this release:
1797 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1798 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1799 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1800 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1801 6968, 6974, 6980, 6995, 7008, 7009, 7029
1803 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1804 up at the same time. Implemented by Ulrich Drepper.
1806 * TLS descriptors for LD and GD on x86 and x86-64.
1807 Implemented by Alexandre Oliva.
1809 * getaddrinfo now handles DCCP and UDPlite.
1810 Implemented by Ulrich Drepper.
1812 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1813 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1814 Implemented by Ulrich Drepper.
1816 * New implementation of memmem, strstr, and strcasestr which is O(n).
1817 Implemented by Eric Blake.
1819 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1821 * Implement "e" option for popen to open file descriptor with the
1822 close-on-exec flag set. Implemented by Ulrich Drepper.
1824 * Implement "b" mode for fmemopen. In this mode writes writes don't
1825 implicitly add a NUL byte and seeks from the end of the buffer really
1826 use the buffer end, not the string length as the basis.
1827 Implemented by Ulrich Drepper.
1829 * Many functions, exported and internal, now atomically set the close-on-exec
1830 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1832 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1834 Implemented by Pravin Satpute.
1836 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1840 * The following bugs are resolved with this release:
1842 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1843 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1844 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1845 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1846 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1847 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1848 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1850 * New locales: bo_CN, bo_IN, shs_CA.
1852 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1854 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1855 Implemented by Pravin Satpute.
1857 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1859 * nscd caches now all timeouts for DNS entries
1860 Implemented by Ulrich Drepper.
1862 * nscd is more efficient and wakes up less often.
1863 Implemented by Ulrich Drepper.
1865 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1866 vdprintf, and obstack_vprintf.
1867 Implemented by Jakub Jelinek.
1869 * Faster memset for x86-64.
1870 Implemented by Harsha Jagasia and H.J. Lu.
1872 * Faster memcpy on x86.
1873 Implemented by Ulrich Drepper.
1875 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1876 Implemented by Ulrich Drepper.
1878 * Faster sqrt and sqrtf implemention for some PPC variants.
1879 Implemented by Steven Munroe.
1883 * The following bugs are resolved with this release:
1885 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1886 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1887 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1888 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1889 5071, 5103, 5104, 5112, 5113, 5184, 5186
1891 * More checking functions: fread, fread_unlocked, open*, mq_open.
1892 Implemented by Jakub Jelinek and Ulrich Drepper.
1894 * Extend fortification to C++. Implemented by Jakub Jelinek.
1896 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1897 by not recognizing 'a' as a modifier when those specs are requested.
1898 Implemented by Jakub Jelinek.
1900 * PPC optimizations to math and string functions.
1901 Implemented by Steven Munroe.
1903 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1904 options to be passed. Implemented by Ulrich Drepper.
1906 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1908 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1909 Implemented by Ulrich Drepper.
1911 * Handle private futexes in the NPTL implementation.
1912 Implemented by Jakub Jelinek and Ulrich Drepper.
1914 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1915 Implemented by Roland McGrath and Ulrich Drepper.
1917 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1919 * SHA-256 and SHA-512 based password encryption.
1920 Implemented by Ulrich Drepper.
1922 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1923 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1926 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1927 Implemented by Ulrich Drepper.
1931 * The following bugs are resolved with this release:
1933 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
1934 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
1935 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
1936 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
1937 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
1938 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
1939 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
1942 * New Linux interfaces: epoll_pwait, sched_getcpu.
1944 * New generic interfaces: strerror_l.
1946 * nscd can now cache the services database. Implemented by Ulrich Drepper.
1950 * The following bugs are resolved with this release:
1952 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
1953 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
1954 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
1955 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
1956 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
1957 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
1958 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
1959 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
1960 3225, 3251, 3252, 3253, 3273
1962 * For Linux, the sorting of addresses returned by getaddrinfo now also
1963 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
1964 Implemented by Ulrich Drepper.
1966 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
1967 Implemented by Ulrich Drepper.
1969 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
1971 * New iconv module for MIK. Contributed by Alexander Shopov.
1973 * For sites with broken group and/or passwd database, the auto-propagate
1974 option of nscd can prevent creating ID lookup entries from the results
1975 of a name lookup and vice versa. This usually is no problem but some
1976 site might have problems with the default behavior.
1977 Implemented by Ulrich Drepper.
1979 * Iterating over entire database in NIS can be slow. With the
1980 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
1981 to trade time for memory. The entire database will be read at once.
1982 Implemented by Ulrich Drepper.
1984 * The interfaces introduced in RFC 3542 have been implemented by
1987 * Support for the new ELF hash table format was added by Ulrich Drepper.
1989 * Support for priority inheritance mutexes added by Jakub Jelinek and
1992 * Support for priority protected mutexes added by Jakub Jelinek.
1994 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
1995 or_IN, csb_PL, fy_NL, sr_ME.
1999 * More overflow detection functions.
2001 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2002 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2004 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2005 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2006 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2007 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2008 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2009 by Masahide Washizawa.
2011 * It is now possible to install audit libraries for the dynamic linker using
2012 LD_AUDIT. Implemented by Ulrich Drepper.
2014 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2015 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2016 For a libc and libpthread that works well on Linux 2.4 kernels, we
2017 recommend using the stable 2.3 branch.
2019 * The new function open_wmemstream acts like open_memstream,
2020 but uses a wchar_t wide character buffer.
2022 * The new function ppoll is an improved version of poll, similar to pselect.
2024 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2025 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2026 renameat, unlinkat, linkat, symlinkat, readlinkat.
2028 * New Linux kernel system calls: unshare,
2029 inotify_init, inotify_add_watch, inotify_rm_watch.
2031 * The euidaccess function is now also known by the alias eaccess,
2032 for compatibility with some other systems.
2034 * Timezone data updated to 2006b version.
2038 * The following bugs are resolved with this release:
2040 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2041 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2042 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2043 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2044 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2045 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2047 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2049 * As of this release, GCC 4 can be used to compile the C Library.
2051 * Timezone data updated to 2005m version.
2055 * The following bugs are resolved with this release:
2057 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2058 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2059 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2060 777, 787, 821, 822, 823, 825
2062 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2066 * Support for RFC 3678. Real implementations exist only for Linux so far.
2067 Implemented by Ulrich Drepper.
2069 * nscd can now cache entries persistently. Expiring entries are reloaded.
2070 For speedups the cache can be shared in memory with client processes.
2071 Implemented by Ulrich Drepper.
2073 * nscd can now perform SELinux checks.
2074 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2076 * getaddrinfo queries are now cached. Canonical name lookup is performed
2078 Implemented by Ulrich Drepper.
2080 * The nothrow function attribute is used when headers are used by gcc when
2081 compiling C code. This can avoid significant amounts of exception
2084 * The malloc functions perform more error checking and are stricter when
2085 it comes to reacting on errors. The default action is to terminate
2086 the process after showing an error message. Implemented by Ulrich Drepper.
2088 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2089 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2091 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2092 namespaces using the new function dlmopen(). This feature is of course,
2093 like most other dynamic loading functionality, not available in statically
2094 linked applications. Implemented by Ulrich Drepper.
2096 * Low-overhead boundary checking variants of string and some stdio functions
2097 were added. These are to be used in conjunction with a gcc patch by
2098 Jakub Jelinek which adds calls to these functions if possible.
2099 Implemented by Jakub Jelinek and Ulrich Drepper.
2101 * Old code for several operating systems and machine architectures that
2102 have not been in working condition in a long time have been removed from
2103 the main source tree maintained by the GNU C Library's maintainers.
2104 These files are now reside in the separate `ports' source module
2105 that is usable as an add-on when building the library.
2109 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2110 interrogate the dynamic linker, compatible with the Solaris interface.
2112 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2113 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2115 * getifaddrs now uses the netlink interface on Linux to get its information.
2116 Implemented by Thorsten Kukuk.
2118 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2119 Implemented by Ulrich Drepper.
2121 * support for non-executable stacks on x86 has been added. Changes mostly
2124 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2127 * getaddrinfo now performs destination address selection according to
2132 * Thread-safe interfaces for many functions that access locale data
2133 were added in version 2.3, but these features were omitted from NEWS.
2134 Many functions have variants with an `_l' suffix that take a `locale_t'
2135 object as a parameter rather than consulting the current locale.
2136 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2137 create and maintain `locale_t' objects. Additionally, the new function
2138 `uselocale' sets "the current locale" (as used by functions not so
2139 parameterized) set for an individual thread. These features were added
2140 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2142 * The functions getresuid, getresgid, setresuid, and setresgid, which
2143 have long been available on Linux, are now declared in <unistd.h>
2144 and are now also available on the Hurd.
2146 * ELF thread-local storage support (TLS) now works on x86-64.
2148 * The new dynamic string token $LIB is expanded in shared library names.
2149 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2151 * Aldy Hernandez contributed complete software floating point support for
2152 PowerPC machines with no FPU.
2154 * fexecve is implemented on Linux.
2156 * The `btowc' function should work at least twice as fast due to
2157 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2159 * With approriate thread add-ons cancelable functions are now implemented
2160 in libc.so as well. No need to call the function in libpthread. This
2161 change allowed to finally disable the incorrect and expensive handling
2162 of weak definition in ld.so.
2164 * Yet more PLT entries in libc.so have been removed. We finally arrived
2165 at the bare minimum. Startup times improved appropriately.
2167 * Support for the new Linux/x86 system call interface was added. The
2168 AT_SYSINFO auxiliary vector entry is recognized and handled.
2172 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2175 * iconv (the program and the interface) now accepts empty names (excluding
2176 options like //TRANSLIT) to mean "use charset of current locale".
2178 * localedef can now transliterate characters in strings which are not in
2179 the provided charmap. The information from the input locale is used.
2181 * Prelinking support was added for ELF targets. This requires additional
2182 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2184 * Read-only stdio streams now use mmap to speed up operation by eliminating
2185 copying and buffer underflows. To use add 'm' to the mode string of
2186 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2188 * The malloc functions were completely rewritten by Wolfram Gloger based
2189 on Doug Lea's malloc-2.7.0.c.
2191 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2192 implementation of regex.
2194 * Bruno Haible upgraded the iconv and locale implementation to support
2197 * Contents of the LC_* and LANG environment variables in the CEN style are
2198 not recognized anymore. It never was used. Change by Ulrich Drepper.
2200 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2201 thread-local storage (TLS) ABI on some platforms.
2202 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2204 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2205 EUC-JISX0213, and TSCII.
2207 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2208 BSD-compatible interface for getting all network interface addresses.
2209 Implementation for IPv4 by Roland McGrath.
2211 * Loading of locale data is faster due to the introduction of a locale
2212 archive. Implemented by Roland McGrath and Ulrich Drepper.
2214 * Startup times are significantly reduced by not using exported functions
2215 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2218 * Steven Munroe contributed a port to PowerPC64/Linux.
2222 * The Hurd now uses the GNU libio implementation of stdio rather than the
2223 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2225 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2226 with the same behavior as the Linux system call of the same name.
2230 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2231 128-bit long double format.
2233 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2234 IBM1161, and IBM1162 charsets.
2236 * Andreas Jaeger contributed a port to x86-64/Linux.
2238 * Peter Bruin contributed a port to PowerPC/Hurd.
2240 * libc functions using I/O streams now can handle wide-oriented streams
2243 * optimizations in the dynamic linker. Binaries created by recent binutils
2244 versions start up quicker due to reduced time spent on relocations.
2246 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2250 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2251 asinh, atanh, j0 for the 128-bit long double format.
2253 * Bruno Haible updated all the code handling Unicode in some form to
2254 support Unicode 3.1.
2256 * Speed of regex for single-byte locales is back to previous levels.
2257 Patch by Isamu Hasegawa.
2259 * Alpha, SPARC, and IA-64 now also using floating stacks.
2261 * Startup time of internationalized applications greatly improved through
2262 iconv cache. Use iconvconfig to generate the cache file.
2263 Contributed by Ulrich Drepper.
2265 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2266 severe performance problems. Patch by David Mosberger.
2268 * The Hurd port got a lot more functionality like AIO, various stdio
2269 extensions, etc. Mainly done by Roland McGrath.
2271 * mtrace can now lookup symbols in shared libraries.
2275 * Intel's IA-64 math library is largely integrated. It provides fast and
2276 accurate implementatations for most basic and standard math functions
2277 in float, double, and long double format.
2279 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2280 and asin for the 96-bit long double format and asin, log, tan for the
2281 128-bit long double format.
2283 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2284 The basic double functions exist today. Contributed by Abraham Ziv
2285 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2286 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2288 * An asynchronous name lookup library was added. The interface is designed
2289 after POSIX AIO. The proposal was circulated beforehand to get comments.
2290 No negative ones came in. Implemented by Ulrich Drepper.
2292 * Port to S390/64bit contributed by Martin Schwidefsky
2293 <schwidefsky@de.ibm.com>.
2295 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2296 of functions for Linux/IA-64.
2298 * The RPC code is now thread safe. Threads can now use the same service
2299 of different services at the same time. Patch by Eric Norum
2300 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2302 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2303 family of functions for Linux/S390.
2305 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2306 of functions for Linux/x86.
2308 * Port to Linux/CRIS contributed by Axis Communications.
2312 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2313 we can now check for namespace violations and missing declarations. The
2314 result is that almost all headers are now Unix-compliant (as defined in
2315 the upcoming XPG6). The negative side is that some programs might need
2316 corrections, too, if they depend on the incorrect form of the headers in
2317 previous versions which defined too many symbols and included too many
2320 * regex now handles multibyte character sets correctly.
2321 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2323 * iconv (the program) does now conform to the upcoming XPG6 and handles
2324 charmaps. Instead of the charset names the path of charmaps can be
2325 provided and the conversion happens based on this data.
2326 Contributed by Ulrich Drepper.
2328 * The locale program now provides detailed information about the installed
2333 only lists the names of the supported locales
2337 provides details such as country, language, and codeset name.
2338 Contributed by Ulrich Drepper.
2342 * The gencat program now parses the input file according to the charset
2343 selected by the LC_CTYPE category. This is important for stateful
2344 character sets. To make generating catalogs easier there is a way
2345 to overwrite the charset selected by the locale: before the first
2346 message or $ quote line the catalog can contain a line like
2348 $ codeset=ISO-8859-2
2350 to select the charset (ISO-8859-2 in this case).
2352 Implemented by Shinya Hanataka and Ulrich Drepper.
2354 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2355 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2356 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2358 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2359 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2361 * The LANGUAGE environment variable is now ignored unless the locale is
2362 changed from the default "C" locale.
2364 * The usual bug fixes.
2368 * Greg McGary added runtime support for bounds checking using gcc's
2369 new -fbounded-pointers option. ix86 target is complete. PowerPC
2372 * Thorsten Kukuk added secure mode support to nscd.
2374 * The Berkeley db libraries have been removed.
2376 Related, the nss_db module is now in a separate package since it
2377 obviously requires a database library being available.
2379 * Wide character I/O streams implemented by Ulrich Drepper.
2381 * Functions from the extended socket API added by Ulrich Drepper.
2383 * Functions feenableexcept and fedisableexcept to control the
2384 behaviour of individual exceptions have been added by Andreas Jaeger.
2386 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2388 * The resolver code has been updated from bind 8.2.3-T5B which supports
2389 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2392 This change could in some situations effect backward compatibility. Since
2393 now `_res' is a thread-local instead of a global variable, modifying it
2394 in one thread does not have any effect in other threads.
2396 The resolver library was also extended to allow IPv6 as the transport
2397 protocol for the requests. This work was done by Stig Venaas.
2399 * Compatibility code for K&R C compilers has been removed from the
2400 header files. A ISO C compiler is needed to use the library
2401 (conforming to either C89 or C99 standard).
2403 * Complete rewrite of the localedef program to support multibyte character
2404 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2405 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2406 Implemented by Ulrich Drepper.
2408 Bruno Haible significantly improved the generation and use of the data
2409 structures for the wide character tables.
2411 * Plural handling in gettext implemented by Ulrich Drepper.
2413 * The utmp daemon has been removed.
2415 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2417 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2420 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2422 * POSIX barriers implemented by Kaz Kylheku.
2424 * POSIX spawn function family implemented by Ulrich Drepper.
2426 * POSIX spinlocks are now available.
2428 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2430 * the configure option --enable-kernel=X.Y.Z allows to strip out
2431 compatibility for kernel versions before X.Y.Z. This is currently only
2432 implemented for Linux.
2434 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2435 field sin6_scope_id. This means that all programs using IPv6 should be
2436 recompiled. Don't expect binary compatibility with previous glibc
2439 * various conversion modules for IBM character sets contributed by
2442 * IA-64 port by Jes Sorensen and HJ Lu.
2456 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2458 * New xdr functions are added; some rpc functions are now 64bit clean.
2460 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2462 * Fixed known incompatibilities with glibc 2.0.
2464 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2466 * Optimized a number of functions (especially the ELF dynamic loader).
2468 * Update timezone data files.
2470 * lots of charmaps corrections
2472 * some new locale definitions and charmaps
2477 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2478 conform to POSIX member type requirements. Symbol versions have been
2479 adjusted as needed within the library, and for direct use by applications,
2480 but there is potential for problems if third-party libraries use
2481 struct timeval as part of their interface. This does not present
2482 a problem for X and other "essential" system libraries.
2484 * An additional locale model to support C++ Standard Library locale
2485 model and probably more was implemented by Ulrich Drepper.
2487 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2490 * Miles Bader provided the `argp' function family to support hierachical
2491 command line argument parsing, layered on top of getopt.
2493 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2495 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2498 * scanf recognizes the %a and %A format for scanning floating point numbers.
2500 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2501 information and interfaces for the available integer types.
2503 * about 130 new math functions were added to implement the ISO C9x math
2506 * the new header <complex.h> contains definitions of the complex math
2507 functions from ISO C 9X.
2509 * the new header <tgmath.h> defines generic macros to use complex or
2510 real valued functions.
2512 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2514 * Andreas Jaeger provided a test suite for the math library.
2516 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2518 * Ulrich Drepper added character set conversion functions (iconv).
2520 * Optimized string functions have been added.
2522 * The localedata addon is now part of glibc.
2524 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2526 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2527 daemon for NSS (nscd).
2529 Missing a better place here are some numbers on improvements. Under
2530 Linux 2.1.125 un-tar-ing the kernel sources takes
2534 using local files 12.19s 6.88s 22.91s
2536 using NIS 13.92s 8.91s 26.34s
2538 using NIS & nscd 10.37s 7.34s 25.30s
2540 using NIS+ 27.57s 30.37s 640.46s
2542 using NIS+ & nscd 10.25s 7.83s 26.51s
2544 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2546 Keep in mind that non-namelookup related operations dominate above times.
2547 It was just a common complain that using NIS+ unpacking the kernel is
2550 [1] The old nscd implementation is not available anymore since it was
2551 distributed with glibc up to version 2.0.98 and thus is now replaced.
2553 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2555 * Mark Kettenis provided a Hesiod NSS module.
2557 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2558 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2560 * rcmd can now handle netgroups (Dick Streefland).
2562 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2565 * Support for the IPv6 protocol has been added to the socket API, as per the
2566 latest draft standards.
2568 * Support for Linux 2.2 has been added.
2570 * Interface changes relative to the latest 2.0.x release:
2571 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2572 addseverity NEW: Unix98
2573 alphasort64 NEW: LFS
2574 argp_err_exit_status NEW: argp, GNU ext
2575 argp_error NEW: argp, GNU ext
2576 argp_failure NEW: argp, GNU ext
2577 argp_help NEW: argp, GNU ext
2578 argp_parse NEW: argp, GNU ext
2579 argp_program_bug_address NEW: argp, GNU ext
2580 argp_program_version NEW: argp, GNU ext
2581 argp_program_version_hook NEW: argp, GNU ext
2582 argp_state_help NEW: argp, GNU ext
2583 argp_usage NEW: argp, GNU ext
2584 authdes_create NEW: Secure RPC
2585 authdes_getucred NEW: Secure RPC
2586 authdes_pk_create NEW: Secure RPC
2587 backtrace NEW: GNU ext.
2588 backtrace_symbols NEW: GNU ext.
2589 backtrace_symbols_fd NEW: GNU ext.
2591 cacosf NEW: ISO C 9x
2592 cacosh NEW: ISO C 9x
2593 cacoshf NEW: ISO C 9x
2594 cacoshl NEW: ISO C 9x
2595 cacosl NEW: ISO C 9x
2602 casinf NEW: ISO C 9x
2603 casinh NEW: ISO C 9x
2604 casinhf NEW: ISO C 9x
2605 casinhl NEW: ISO C 9x
2606 casinl NEW: ISO C 9x
2608 catanf NEW: ISO C 9x
2609 catanh NEW: ISO C 9x
2610 catanhf NEW: ISO C 9x
2611 catanhl NEW: ISO C 9x
2612 catanl NEW: ISO C 9x
2613 cbc_crypt NEW: Secure RPC
2617 ccoshf NEW: ISO C 9x
2618 ccoshl NEW: ISO C 9x
2624 cimagf NEW: ISO C 9x
2625 cimagl NEW: ISO C 9x
2626 clearerr_locked REMOVED
2627 clntunix_create NEW: sunrpc ext
2629 clog10 NEW: ISO C 9x
2630 clog10f NEW: ISO C 9x
2631 clog10l NEW: ISO C 9x
2641 cprojf NEW: ISO C 9x
2642 cprojl NEW: ISO C 9x
2644 crealf NEW: ISO C 9x
2645 creall NEW: ISO C 9x
2650 csinhf NEW: ISO C 9x
2651 csinhl NEW: ISO C 9x
2654 csqrtf NEW: ISO C 9x
2655 csqrtl NEW: ISO C 9x
2659 ctanhf NEW: ISO C 9x
2660 ctanhl NEW: ISO C 9x
2662 des_setparity NEW: Secure RPC
2663 ecb_crypt NEW: Secure RPC
2664 endutxent NEW: Unix98
2666 exp10f NEW: ISO C 9x
2667 exp10l NEW: ISO C 9x
2671 fattach NEW: STREAMS
2672 fdetach NEW: STREAMS
2676 feclearexcept NEW: ISO C 9x
2677 fegetenv NEW: ISO C 9x
2678 fegetexceptflag NEW: ISO C 9x
2679 fegetround NEW: ISO C 9x
2680 feholdexcept NEW: ISO C 9x
2682 feraiseexcept NEW: ISO C 9x
2683 ferror_locked REMOVED
2684 fesetenv NEW: ISO C 9x
2685 fesetexceptflag NEW: ISO C 9x
2686 fesetround NEW: ISO C 9x
2687 fetestexcept NEW: ISO C 9x
2688 feupdateenv NEW: ISO C 9x
2689 fflush_locked REMOVED
2693 fgets_unlocked NEW: GNU ext.
2694 fileno_locked REMOVED
2706 fputc_locked REMOVED
2707 fputs_unlocked NEW: GNU ext.
2708 fread_unlocked NEW: GNU ext.
2713 fstatvfs NEW: Unix98
2717 ftruncate64 NEW: LFS
2719 fwrite_unlocked NEW: GNU ext.
2720 gai_strerror NEW: IPv6
2724 getchar_locked REMOVED
2726 getdate_err NEW: Unix98
2727 getdate_r NEW: GNU ext.
2729 getnameinfo NEW: IPv6
2730 getnetname NEW: Secure RPC
2731 getpmsg NEW: STREAMS
2732 getpt NEW: Unix98 PTY
2733 getrlimit64 NEW: LFS
2734 getutxent NEW: Unix98
2735 getutxid NEW: Unix98
2736 getutxline NEW: Unix98
2737 glob64 NEW: GNU ext.
2738 globfree64 NEW: GNU ext.
2739 gnu_get_libc_release NEW: GNU ext.
2740 gnu_get_libc_version NEW: GNU ext.
2741 grantpt NEW: Unix98 PTY
2742 host2netname NEW: Secure RPC
2744 iconv_close NEW: iconv
2745 iconv_open NEW: iconv
2746 if_freenameindex NEW: IPv6
2747 if_indextoname NEW: IPv6
2748 if_nameindex NEW: IPv6
2749 if_nametoindex NEW: IPv6
2750 in6addr_any NEW: IPv6
2751 in6addr_loopback NEW: IPv6
2752 inet6_isipv4mapped NEW: IPv6
2753 isastream NEW: STREAMS
2754 iswblank NEW: GNU ext.
2755 key_decryptsession NEW: Secure RPC
2756 key_decryptsession_pk NEW: Secure RPC
2757 key_encryptsession NEW: Secure RPC
2758 key_encryptsession_pk NEW: Secure RPC
2759 key_gendes NEW: Secure RPC
2760 key_get_conv NEW: Secure RPC
2761 key_secretkey_is_set NEW: Secure RPC
2762 key_setnet NEW: Secure RPC
2763 key_setsecret NEW: Secure RPC
2764 llrint NEW: ISO C 9x
2765 llrintf NEW: ISO C 9x
2766 llrintl NEW: ISO C 9x
2767 llround NEW: ISO C 9x
2768 llroundf NEW: ISO C 9x
2769 llroundl NEW: ISO C 9x
2774 lrintf NEW: ISO C 9x
2775 lrintl NEW: ISO C 9x
2776 lround NEW: ISO C 9x
2777 lroundf NEW: ISO C 9x
2778 lroundl NEW: ISO C 9x
2780 makecontext NEW: Unix98
2781 mempcpy NEW: GNU ext.
2784 modify_ldt NEW: kernel
2788 nearbyint NEW: ISO C 9x
2789 nearbyintf NEW: ISO C 9x
2790 nearbyintl NEW: ISO C 9x
2791 netname2host NEW: Secure RPC
2792 netname2user NEW: Secure RPC
2793 nexttoward NEW: ISO C 9x
2794 nexttowardf NEW: ISO C 9x
2795 nexttowardl NEW: ISO C 9x
2799 passwd2des NEW: Secure RPC
2801 pow10f NEW: GNU ext.
2802 pow10l NEW: GNU ext.
2805 printf_size NEW: GNU ext.
2806 printf_size_info NEW: GNU ext.
2807 profil_counter REMOVED
2808 pthread_mutexattr_getkind_np REPLACED
2809 pthread_mutexattr_setkind_np REPLACED
2810 ptsname NEW: Unix98 PTY
2811 ptsname_r NEW: Unix98 PTY
2813 putchar_locked REMOVED
2814 putgrent NEW: GNU ext.
2816 putpmsg NEW: STREAMS
2817 pututxline NEW: Unix98
2821 readdir64_r NEW: LFS
2822 remquo NEW: ISO C 9x
2823 remquof NEW: ISO C 9x
2824 remquol NEW: ISO C 9x
2826 roundf NEW: ISO C 9x
2827 roundl NEW: ISO C 9x
2829 scalbln NEW: ISO C 9x
2830 scalblnf NEW: ISO C 9x
2831 scalblnl NEW: ISO C 9x
2833 sendfile NEW: kernel
2834 setcontext NEW: Unix98
2835 setrlimit64 NEW: LFS
2836 setutxent NEW: Unix98
2838 sigignore NEW: Unix98
2839 sigqueue NEW: POSIX.1b
2840 sigrelse NEW: Unix98
2841 sigset NEW: POSIX.1b
2842 sigtimedwait NEW: POSIX.1b
2843 sigwaitinfo NEW: POSIX.1b
2844 sincos NEW: GNU ext.
2845 sincosf NEW: GNU ext.
2846 sincosl NEW: GNU ext.
2850 strcasestr NEW: GNU ext.
2851 strtoimax NEW: ISO C 9x
2852 strtoumax NEW: ISO C 9x
2853 strverscmp NEW: GNU ext.
2854 svcauthdes_stats NEW: Secure RPC
2855 svcunix_create NEW: sunrpc etx
2856 svcunixfd_create NEW: sunrpc ext
2857 swapcontext NEW: Unix98
2858 tcgetsid NEW: Unix98 PTY
2859 tdestroy NEW: GNU ext.
2860 tgamma NEW: ISO C 9x
2861 tgammaf NEW: ISO C 9x
2862 tgammal NEW: ISO C 9x
2866 truncf NEW: ISO C 9x
2867 truncl NEW: ISO C 9x
2869 unlockpt NEW: Unix98 PTY
2870 updwtmpx NEW: Unix98
2871 user2netname NEW: Secure RPC
2872 utmpxname NEW: Unix98
2873 versionsort NEW: GNU ext.
2874 versionsort64 NEW: GNU ext.
2876 wcscasecmp NEW: GNU ext.
2877 wcsncasecmp NEW: GNU ext.
2878 wcsnlen NEW: GNU ext.
2879 wcstoimax NEW: ISO C 9x
2880 wcstoll NEW: ISO C 9x
2881 wcstoull NEW: ISO C 9x
2882 wcstoumax NEW: ISO C 9x
2884 wordexp NEW: POSIX.2
2885 wordfree NEW: POSIX.2
2886 write_profiling REMOVED
2887 xdecrypt NEW: Secure RPC
2888 xdr_authdes_cred NEW: Secure RPC
2889 xdr_authdes_verf NEW: Secure RPC
2890 xdr_cryptkeyarg NEW: Secure RPC
2891 xdr_cryptkeyarg2 NEW: Secure RPC
2892 xdr_cryptkeyres NEW: Secure RPC
2893 xdr_getcredres NEW: Secure RPC
2894 xdr_key_netstarg NEW: Secure RPC
2895 xdr_key_netstres NEW: Secure RPC
2896 xdr_keybuf NEW: Secure RPC
2897 xdr_keystatus NEW: Secure RPC
2898 xdr_netnamestr NEW: Secure RPC
2899 xdr_sizeof NEW: Secure RPC
2900 xdr_unixcred NEW: sunrpc ext
2901 xencrypt NEW: Secure RPC
2902 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2913 * inet_ntoa is thread-safe
2915 * updwtmp is moved from libutil to libc
2917 * rewrite of cbrt function
2919 * update of timezone data
2933 * add atoll function
2935 * fix complex problems in Berkeley DB code
2937 * fix math functions
2941 * fixed lots of header problems (especially Linux/GNU specific)
2943 * dynamic loader preserves all registers
2945 * Roland McGrath provided support for handling of auxiliary objects in
2946 the ELF dynamic loader.
2948 * support for parallel builds is improved
2952 * GNU extensions are no longer declared by default. To enable them you
2953 must define the macro `_GNU_SOURCE' in your program or compile with
2956 * The library has changed from using GNU ld symbol aliases to using weak
2957 symbols where available. The ELF object file format supports weak
2958 symbols; GNU ld also supports weak symbols in the a.out format. (There
2959 is also now support for other GNU ld extensions in ELF. Use the
2960 `--with-elf' option to configure to indicate you have ELF, and
2961 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
2962 of many files which contained only symbol aliases, reducing the size of
2963 the source and the compiled library; many other files were renamed to
2964 less cryptic names previously occupied by the symbol alias files.
2965 There is a new header file <elf.h> for programs which operate on
2966 files in the ELF format.
2968 * Converted to Autoconf version 2, so `configure' has more options.
2969 Run `configure --help' to see the details.
2971 * The library can now be configured to build profiling, highly-optimized
2972 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
2973 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
2974 `configure' enable building these extra libraries. The shared library is
2975 built by default when using both ELF and GNU ld. When shared libraries
2976 are enabled, the new library `-ldl' is available for arbitrary run-time
2977 loading of shared objects; its interface is defined in <dlfcn.h>. The
2978 new header file <link.h> gives access to the internals of the run-time
2979 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
2980 application of same name on other systems and it provides information
2981 about dynamically linked binaries.
2983 * The C library now provides the run-time support code for profiling
2984 executables compiled with `-pg'. Programs can control the profiling code
2985 through the interface in <sys/gmon.h>. The `gmon.out' files written by
2986 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
2987 the support for this file format was contributed by David Mosberger-Tang.
2989 * The math code has been replaced with a math library based on fdlibm from
2990 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
2991 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
2992 functions. The math functions now reside in a separate library, so
2993 programs using them will need to use `-lm' their linking commands.
2995 * John C. Bowman contributed optimized ix87 assembler inline functions.
2997 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
2998 mechanism similar to that found in Solaris 2. This is now used for the
2999 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3000 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3001 file controls what services are used for each individual database. This
3002 works by loading shared libraries with names specified in `nsswitch.conf',
3003 so service modules can be changed or added at any time without even
3004 relinking any program. Currently there are the file, db, and NIS based
3005 NSS services available.
3007 * The new functions `strtoq' and `strtouq' parse integer values from
3008 strings, like `strtol' and `strtoul', but they return `long long int' and
3009 `unsigned long long int' values, respectively (64-bit quantities).
3011 * The new functions `strtof' and `strtold' parse floating-point values from
3012 strings, like `strtod', but they return `float' and `long double' values,
3013 respectively (on some machines `double' and `long double' are the same).
3015 * Ulrich Drepper has contributed new implementations of the floating-point
3016 printing and reading code used in the `printf' family of functions and
3017 `strtod', `strtof', and `strtold'. These new functions are perfectly
3018 accurate, and much faster than the old ones.
3020 * The implementation of the POSIX locale model was completely rewritten by
3021 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3022 to compile the POSIX locale definition.
3024 * The former dummy implementations of the strcoll and strxfrm function are
3025 now replaced by fully functional code contributed by Ulrich Drepper. The
3026 collation information comes from the POSIX locale definitions.
3028 * The new header <langinfo.h> defines an interface for accessing
3029 various locale-dependent data (using the locale chosen with `setlocale').
3031 * Ulrich Drepper has contributed a new suite of functions for operation on
3032 wide-character and multibyte-character strings, in <wchar.h>;
3033 and classification and case conversion of wide characters, in <wctype.h>.
3034 These new functions are conforming to the ISO C, Amendement 1 specification.
3036 * There is now a second implementation of the standard I/O library available.
3037 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3038 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3040 * You can now use positional parameter specifications in format strings
3041 for the `printf' and `scanf' families of functions. For example,
3042 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3043 ``Number 6, Mr Jones''. This is mainly useful when providing different
3044 format strings for different languages, whose grammars may dictate
3045 different orderings of the values being printed. To support this
3046 feature, the interface for `register_printf_handler' has changed; see
3047 the header file <printf.h> for details.
3049 * The `printf' and `scanf' families of functions now understand a new
3050 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3051 to group numbers as indicated by the locale; for `scanf' and friends, this
3052 says to accept as valid only a number with all the proper grouping
3053 separators in the right places. In the default "C" locale, numbers are
3054 not grouped; but locales for specific countries will define the usual
3055 conventions (i.e. separate thousands with `,' in the US locale).
3057 * The pgrp functions have been regularized, slightly incompatibly but much
3058 less confusingly. The core functions are now `getpgid' and `setpgid',
3059 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3060 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3061 provided for compatibility. There is no longer an incompatible `getpgrp'
3062 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3064 * The new header file <fts.h> and suite of functions simplify programs that
3065 operate on directory trees. This code comes from 4.4 BSD.
3067 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3068 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3069 NSS scheme used in glibc.
3071 * There is a new malloc debugging hook `__memalign_hook'.
3073 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3074 `unsigned int' in <sys/types.h>. These are for compatibility only and
3075 their use is discouraged.
3077 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3078 done differently, so that it works even without GNU ld.
3080 * New function `euidaccess' checks allowed access to a file like `access',
3081 but using the effective IDs instead of the real IDs.
3083 * The time zone data files have been updated for the latest and greatest
3084 local time conventions of the countries of the world.
3086 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3089 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3090 for formatting floating-point numbers. They are provided only for
3091 compatibility; new programs should use `sprintf' instead. There are
3092 also equivalent function for the `long double' floating-point type and
3093 all functions also exist in a reentrant form.
3095 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3096 functions for maintaining the login-record files (primarily of use to
3097 system programs such as `login'), and convenient functions for
3098 allocating and initializing a pseudo-terminal (pty) device.
3100 * Ulrich Drepper has contributed new support for System V style
3101 shared memory and IPC on systems that support it.
3103 * Ulrich Drepper has contributed several miscellaneous new functions found
3104 in System V: The `hsearch' family of functions provide an effective
3105 implementation of hash tables; `a64l' and `l64a' provide a very simple
3106 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3109 * Ulrich Drepper has contributed new reentrant counterparts for the
3110 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3112 * Ulrich Drepper has contributed new, highly-optimized versions of several
3113 string functions for the i486/Pentium family of processors.
3115 * Ulrich Drepper has updated the Linux-specific code, based largely
3116 on work done in Hongjiu Lu's version of GNU libc for Linux.
3117 The GNU library now supports Linux versions 2.0.10 and later,
3118 using the ELF object file format (i[3456]86-*-linux).
3120 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3122 * David Mosberger-Tang and Richard Henderson have ported the C library
3123 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3124 dynamic linking support for ELF/Alpha.
3126 * Richard Henderson contributed several Alpha optimized assembler function
3127 for arithmetic and string handling.
3129 * Ulrich Drepper has contributed a new set of message catalog functions to
3130 support multiple languages using the <libintl.h> interface, for use with
3131 his new package GNU gettext. Translation volunteers have contributed
3132 catalogs of the library's messages in Spanish, German, and Korean.
3134 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3135 program and the `catgets' function for reading the catalog files it
3136 creates. (The <libintl.h> interface is preferred; we include the
3137 <nl_types.h> interface using `catgets' only for source compatibility with
3138 programs already written to use it.)
3140 * New header file <values.h> gives SVID-compatible names for <limits.h>
3143 * Various new macros, declarations, and small header files for compatibility
3146 * New function `group_member' is a convenient way to check if a process has
3147 a given effective group ID.
3149 * When using GCC 2.7 and later, the socket functions are now declared in a
3150 special way so that passing an argument of type `struct sockaddr_in *',
3151 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3152 `struct sockaddr *' type, does not generate a type-clash warning.
3154 * New function `error' declared in header file <error.h> is a convenient
3155 function for printing error messages and optionally exiting; this is the
3156 canonical function used in GNU programs. The new functions `err', `warn',
3157 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3158 doing the same thing.
3160 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3161 POSIX.2 `glob' function to do ~ and {...} expansion.
3163 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3164 `clearenv' which is used in POSIX.9 is also available.
3166 * New function `getsid' returns session ID number on systems that support it.
3168 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3169 files <db.h> and <mpool.h> provide a rich set of functions for several
3170 types of simple databases stored in memory and in files, and <ndbm.h> is
3171 an old `ndbm'-compatible interface using the `db' functions. Link with
3172 `-ldb' to get these functions.
3174 * New macro `strdupa' copies a string like `strdup', but uses local stack
3175 space from `alloca' instead of dynamic heap space from `malloc'.
3177 * New function `strnlen' is like `strlen' but searches only a given maximum
3178 number of characters for the null terminator. `stpncpy', `strndup' and
3179 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3182 * New function `statfs' in header <sys/statfs.h>.
3184 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3185 provide convenient functions for operating on blocks of null-terminated
3188 * A new suite of functions in <utmp.h> handle all the details of reading
3189 and writing the utmp file.
3191 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3194 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3195 completely POSIX compliant and also implemented the extended functionality
3196 to handle alternate digit representation and alternate era date formats.
3198 * Ulrich Drepper provided an implementation of the `strptime' function
3199 defined in XPG4.2 which transforms a string into a `struct tm' value.
3201 * Paul Eggert provided the tzselect shell script as part of the timezone
3202 code. The shell script makes it easy to select the correct timezone
3205 * The implementation of the malloc family of functions is completely replaced
3206 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3207 The implementation uses the mmap function (if available) and it is
3208 optimized for the use in multi threaded programs.
3210 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3211 functions. This new functionality is usable by specifying a special
3212 salt string and it is compatible with implementation on *BSD systems.
3214 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3215 `getsubopt' to handle second level command line options, `bsd_signal'
3216 to access BSD style `signal' functionality, the obsolete `regexp' style
3219 * the `lchown' function is available on system which support this
3222 * The implementation of the shadow password handling function was contributed
3225 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3227 * POSIX.1g support was added. The <sys/select.h> header is available,
3228 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3229 implementation of `getaddrinfo'.
3233 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3235 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3236 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3238 * The new function `daemon' from 4.4 BSD is useful for server programs that
3239 want to put themselves in the background.
3241 * Joel Sherrill has contributed support for several standalone boards that
3242 run without an operating system.
3244 * `printf', `scanf' and friends now accept a `q' type modifier for long
3245 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3247 * All of the code taken from BSD (notably most of the math and networking
3248 routines) has been updated from the BSD 4.4-Lite release.
3250 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3252 * The new functions `getdomainname' and `setdomainname' fetch or change the
3253 YP/NIS domain name. These are system calls which exist on systems which
3256 * The time zone data files have been updated for the latest international
3259 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3260 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3264 * The C library now includes support for Sun RPC, from Sun's free
3265 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3266 are included. (There is still no support for YP.)
3268 * Tom Quinn has contributed a port of the C library to SGI machines running
3269 Irix 4 (mips-sgi-irix4).
3271 * The new `lockf' function is a simplified interface to the locking
3272 facilities of `fcntl', included for compatibility.
3274 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3276 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3278 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3281 * The error message printed by `assert' for a failed assertion now includes
3282 the name of the program (if using GNU ld) and the name of the calling
3283 function (with versions of GCC that support this).
3285 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3287 * The library now includes the <sys/mman.h> header file and memory
3288 management functions `mmap', `munmap', `mprotect', `msync', and
3289 `madvise', on systems that support those facilities.
3291 * The interface for `mcheck' has changed slightly: the function called to
3292 abort the program when an allocation inconsistency is detected now takes
3293 an argument that indicates the type of failure. The new function
3294 `mprobe' lets you request a consistency check for a particular block at
3295 any time (checks are normally done only when you call `free' or `realloc'
3298 * It is now possible to easily cross-compile the C library, building on one
3299 system a library to run on another machine and/or operating system. All
3300 you need to do is set the variable `HOST_CC' in `configparms' to the
3301 native compiler for programs to run on the machine you are building on (a
3302 few generator programs are used on Unix systems); set `CC' to the
3305 * The new function `fexecve' (only implemented on the GNU system) executes
3306 a program file given a file descriptor already open on the file.
3310 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3311 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3313 * You can set the variable `libprefix' in `configparms' to specify a prefix
3314 to be prepended to installed library files; this makes it easy to install
3315 the GNU C library to be linked as `-lgnuc' or whatever.
3317 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3318 copies a limited number of characters from a string, and returns the
3319 address of the last character written.
3321 * You no longer need to check for whether the installed `stddef.h' is
3322 compatible with the GNU C library. configure now checks for you.
3324 * You can now define a per-stream `fileno' function to convert the
3325 stream's cookie into an integral file descriptor.
3327 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3328 allocates zero bytes of storage, and returns a unique pointer which
3329 you can pass to `realloc' or `free'. The behavior is undefined if
3330 you dereference this pointer.
3332 * The C library now runs on Sony NEWS m68k machines running either
3333 NewsOS 3 or NewsOS 4.
3335 * The new `syscall' function is a system-dependent primitive function
3336 for invoking system calls. It has the canonical behavior on Unix
3337 systems, including unreliable return values for some calls (such as
3338 `pipe', `fork' and `getppid').
3340 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3341 to `EAGAIN', which is the preferred name. On systems whose kernels
3342 use two distinct codes, the C library now translates EWOULDBLOCK to
3343 EAGAIN in every system call function.
3347 * The GNU C Library Reference Manual is now distributed with the library.
3348 `make dvi' will produce a DVI file of the printed manual.
3349 `make info' will produce Info files that you can read on line using C-h i
3350 in Emacs or the `info' program.
3351 Please send comments on the manual to bug-glibc-manual@gnu.org.
3353 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3355 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3357 * Jason Merrill has contributed a port to the Sequent Symmetry running
3358 Dynix version 3 (i386-sequent-dynix).
3360 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3361 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3363 * New function `memory_warnings' lets you arrange to get warnings when
3364 malloc is running out of memory to allocate, like Emacs gives you.
3366 * The C library now contains the relocating allocator used in Emacs 19 for
3367 its editing buffers. This allocator (ralloc) minimizes allocation
3368 overhead and fragmentation by moving allocated regions around whenever it
3369 needs to. You always refer to a ralloc'd region with a "handle" (a
3370 pointer to a pointer--an object of type `void **').
3372 * There is a new `printf' format: `%m' gives you the string corresponding
3373 to the error code in `errno'.
3375 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3376 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3377 pass, the `a' modifier says to fill in a `char **' you pass with a
3380 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3381 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3382 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3384 * `mkstemp' is a traditional Unix function to atomically create and open a
3385 uniquely-named temporary file.
3389 * The standard location for the file that says what the local timezone is
3390 has changed again. It is now `/usr/local/etc/localtime' (or more
3391 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3393 * The distribution no longer contains any files with names longer than 14
3396 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3397 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3399 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3403 * The configuration process has changed quite a bit. The `configure'
3404 script is now used just like the configuration scripts for other GNU
3405 packages. The `sysdeps' directory hierarchy is much rearranged.
3406 The file `INSTALL' explains the new scheme in detail.
3408 * The header files no longer need to be processed into ANSI C and
3409 traditional C versions. There is just one set of files to install, and
3410 it will work with ANSI or old C compilers (including `gcc -traditional').
3412 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3413 MIPS DECStation running Ultrix 4.
3415 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3416 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3417 made itself into a shared library.
3419 * Yet further improved support for the i386, running 4.3 BSD-like systems
3420 (such as Mach 3 with the Unix single-server), or System V.
3422 * New function `strncasecmp' to do case-insensitive string comparison
3423 with limited length.
3425 * New function `strsep' is a reentrant alternative to `strtok'.
3427 * New functions `scandir' and `alphasort' for searching directories.
3429 * New function `setenv' is a better interface to `putenv'.
3431 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3432 function for traversing a directory tree.
3434 * The GNU obstack package is now also part of the C library.
3435 The new function `open_obstack_stream' creates a stdio stream that
3436 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3437 formatted output directly to an obstack.
3439 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3440 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3442 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3444 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3445 things to your strings.
3447 * There are some new test programs: test-fseek, testmb, and testrand.
3449 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3450 These ports are not finished, but are a good starting place for really
3451 supporting those systems.
3453 * `/etc/localtime' is now the standard location for the file that says what
3454 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3455 This follows the general principle that `/etc' is the place for all local
3456 configuration files.
3458 * The C library header files now use `extern "C"' when used by the C++
3459 compiler, so the C library should now work with C++ code.
3461 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3462 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3465 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3466 function which uses the `merge sort' algorithm, and is said to be
3467 significantly faster than the old GNU `qsort' function. Merge sort is
3468 now the standard `qsort' function. The new algorithm can require a lot
3469 of temporary storage; so, the old sorting function is called when the
3470 required storage is not available.
3472 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3473 provides the Unix `crypt' function, plus some other entry points.
3475 * The code and header files taken from 4.4 BSD have been updated with the
3476 latest files released from Berkeley.
3478 ----------------------------------------------------------------------
3479 Copyright information:
3481 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3483 Permission is granted to anyone to make or distribute verbatim copies
3484 of this document as received, in any medium, provided that the
3485 copyright notice and this permission notice are preserved,
3486 thus giving the recipient permission to redistribute in turn.
3488 Permission is granted to distribute modified versions
3489 of this document, or of portions of it,
3490 under the above conditions, provided also that they
3491 carry prominent notices stating who last changed them.
3494 version-control: never