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 - Total order functions: totalorder, totalorderf, totalorderl.
60 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
61 for the Linux quota interface which predates kernel version 2.4.22 has
64 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
65 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
66 “no-ip6-dotint” had already been the default, and support for the
67 “ip6-dotint” option was removed from the Internet in 2006.
69 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
70 flag from <resolv.h> have been removed. The option relied on a
71 backwards-incompatible DNS extension which was never deployed on the
74 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
75 in the <resolv.h> header file have been deprecated. They were already
78 * DNSSEC-related declarations and definitions have been removed from the
79 <arpa/nameser.h> header file, and libresolv will no longer attempt to
80 decode the data part of DNSSEC record types. Previous versions of glibc
81 only implemented minimal support for the previous version of DNSSEC, which
82 is incompatible with the currently deployed version.
84 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
85 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
86 <arpa/nameser.h> header file because the distinction between RR types and
87 meta-RR types is not officially standardized, subject to revision, and
88 thus not suitable for encoding in a macro.
90 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
91 and rhook members of the res_state type in <resolv.h> have been removed.
92 The glibc stub resolver did not support these hooks, but the header file
95 * For multi-arch support it is recommended to use a GCC which has
96 been built with support for GNU indirect functions. This ensures
97 that correct debugging information is generated for functions
98 selected by IFUNC resolvers. This support can either be enabled by
99 configuring GCC with '--enable-gnu-indirect-function', or by
100 enabling it by default by setting 'default_gnu_indirect_function'
101 variable for a particular architecture in the GCC source file
104 Security related changes:
106 On ARM EABI (32-bit), generating a backtrace for execution contexts which
107 have been created with makecontext could fail to terminate due to a
108 missing .cantunwind annotation. This has been observed to lead to a hang
109 (denial of service) in some Go applications compiled with gccgo. Reported
110 by Andreas Schwab. (CVE-2016-6323)
112 The following bugs are resolved with this release:
114 [The release manager will add the list generated by
115 scripts/list-fixed-bugs.py just before the release.]
119 * The minimum Linux kernel version that this version of the GNU C Library
120 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
121 kernel version 2.6.32 or later suffices (on architectures that already
122 required kernel versions more recent than 3.2, those requirements remain
123 unchanged). Linux 3.2 or later kernel headers are required on all
126 * The pap_AN locale has been deleted. This has been deprecated for a long
127 time. It has been replaced by pap_AW & pap_CW, both of which have long
128 been included in previous releases.
130 * The readdir_r and readdir64_r functions have been deprecated. It is
131 recommended to use readdir and readdir64 instead.
133 * The type “union wait” has been removed. It was deprecated in the early
134 1990s and never part of POSIX. Application code should use the int type
135 instead of “union wait”.
137 * A new NSS action is added to facilitate large distributed system
138 administration. The action, MERGE, allows remote user stores like LDAP
139 to be merged into local user stores like /etc/groups in order to provide
140 easy to use, updated, and managed sets of merged credentials. The new
141 action can be used by configuring it in /etc/nsswitch.conf:
142 group: files [SUCCESS=merge] nis
143 Implemented by Stephen Gallagher (Red Hat).
145 * The deprecated __malloc_initialize_hook variable has been removed from the
148 * The long unused localedef --old-style option has been removed. It hasn't
149 done anything in over 16 years. Scripts using this option can safely
152 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
153 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
154 return the next representable value in the direction of positive infinity
155 and the nextdown functions return the next representable value in the
156 direction of negative infinity. These are currently enabled as GNU
159 Security related changes:
161 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
162 could result in a stack overflow when getnetbyname was called with an
163 overly long name. (CVE-2016-3075)
165 * Previously, getaddrinfo copied large amounts of address data to the stack,
166 even after the fix for CVE-2013-4458 has been applied, potentially
167 resulting in a stack overflow. getaddrinfo now uses a heap allocation
168 instead. Reported by Michael Petlan. (CVE-2016-3706)
170 * The glob function suffered from a stack-based buffer overflow when it was
171 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
172 Reported by Alexander Cherepanov. (CVE-2016-1234)
174 * The Sun RPC UDP client could exhaust all available stack space when
175 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
176 alloca plugin for GCC. (CVE-2016-4429)
178 * The IPv6 name server management code in libresolv could result in a memory
179 leak for each thread which is created, performs a failing naming lookup,
180 and exits. Over time, this could result in a denial of service due to
181 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
183 The following bugs are resolved with this release:
185 [1170] localedata: ne_NP: update Nepali locale definition file
186 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
188 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
189 [6796] math: fdim() does not set errno on overflow
190 [10354] libc: posix_spawn should use vfork() in more cases than presently
191 [11213] localedata: localedata: add copyright disclaimer to locale files
192 [12143] localedata: chr_US: new Cherokee locale
193 [12450] localedata: sgs_LT: new locale
194 [12676] localedata: ln_CD: new locale
195 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
197 [13304] math: fma, fmaf, fmal produce wrong results
198 [14259] build: --localedir arg to configure is ignored
199 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
200 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
201 [14934] localedata: es_CL: wrong first weekday chilean locale
202 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
204 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
206 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
207 [15368] nptl: raise() is not async-signal-safe
208 [15479] math: ceil, floor, round and trunc raise inexact exception
209 [15578] localedata: kk_KZ: various updates
210 [16003] localedata: pap_AN: punt old locale
211 [16137] localedata: iw_IL: punt old locale
212 [16190] localedata: eo: new esperanto locale
213 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
214 [16742] malloc: race condition: pthread_atfork() called before first
215 malloc() results in unexpected locking behaviour/deadlocks
216 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
218 [16983] localedata: postal_fmt does not allow %l and %n modifiers
219 [17565] localedata: pt_PT: wrong (work-)week start
220 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
221 [17950] build: Build fails with -msse
222 [18205] localedata: be_BY*: wrong first_weekday and first_workday
223 [18433] libc: posix_spawn does not return correctly upon failure to
225 [18453] localedata: charmaps/IBM875: incorrect codes
226 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
228 [18896] localedata: he_IL: improvements for currency
229 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
231 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
233 [19056] libc: Deprecate readdir_r
234 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
236 [19198] localedata: nl_NL: small improvements for Dutch locales
237 [19257] network: Per-thread memory leak in __res_vinit with IPv6
238 nameservers (CVE-2016-5417)
239 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
241 [19400] locale: Language missing in "iso-639.def", trivial fix in
243 [19431] malloc: Deadlock between fflush, getdelim, and fork
244 [19505] libc: Incorrect file descriptor validity checks in
245 posix_spawn_file_actions_add{open,close,dup2}
246 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
248 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
249 `intl/{gettextP,loadinfo}.h'
250 [19534] libc: execle, execlp may use malloc
251 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
252 [19573] network: res_nclose and __res_maybe_init disagree about name
253 server initialization, breaking Hesiod
254 [19575] localedata: Status of GB18030 tables
255 [19581] localedata: sr_* date_fmt string contains additional newline
256 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
258 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
259 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
260 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
261 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
263 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
265 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
267 [19610] dynamic-link: ldconfig -X removes stale symbolic links
268 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
269 [19633] locale: strfmon_l applies global locale to number formatting
270 [19642] network: Memory leak in getnameinfo
271 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
272 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
274 [19654] math: [x86_64] Need testcase for BZ #19590 fix
275 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
277 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
278 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
280 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
282 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
284 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
285 pointers and lengths in error-case.
286 [19727] locale: Converting from/to UTF-xx with iconv() does not always
287 report errors on UTF-16 surrogates values.
288 [19755] nscd: nscd assertion failure in gc
289 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
290 [19759] libc: mempcpy shouldn't be inlined
291 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
292 [19765] libc: s390 needs an optimized mempcpy
293 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
294 NAME_MAX limit assumption (CVE-2016-1234)
295 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
296 [19787] network: Missing and incorrect truncation checks in getnameinfo
297 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
299 [19791] network: Assertion failure in res_query.c with un-connectable name
301 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
302 [19822] math: libm.so install clobbers old version
303 [19825] network: resolv: send_vc can return uninitialized data in second
304 response to getaddrinfo
305 [19830] network: nss_dns: should check RDATA length against buffer length
306 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
307 confronted with A/AAAA records of invalid size
308 [19837] nss: nss_db: No retries for some long lines with a larger buffer
309 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
310 [19853] stdio: Printing IBM long double in decimal with high precision is
312 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
313 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
314 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
316 [19865] network: Assertion failure or memory leak in
317 _nss_dns_getcanonname_r
318 [19868] network: nss_dns: netent code does not skip over non-PTR records
319 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
321 [19881] string: Improve x86-64 memset
322 [19907] string: Incorrect memcpy tests
323 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
325 [19925] libc: termios.h XCASE namespace
326 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
327 [19929] libc: limits.h NL_NMAX namespace
328 [19931] stdio: Memory leak in vfprintf
329 [19957] libc: clone(CLONE_VM) access invalid parent memory
330 [19963] localedata: en_IL: New locale
331 [19989] stdio: stdio.h cuserid namespace
332 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
334 [19996] locale: langinfo.h nl_langinfo_l namespace
335 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
336 [20010] network: getaddrinfo: Stack overflow in hostent translation
338 [20012] stdio: libio: fmemopen append mode failure
339 [20014] stdio: stdio.h namespace for pre-threads POSIX
340 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
341 [20023] libc: fcntl.h timespec namespace
342 [20024] math: [x86_64] vectorized sincos trashes the stack
343 [20031] network: nss_hesiod: Heap overflow in get_txt_records
344 [20041] time: sys/time.h timespec namespace
345 [20043] libc: unistd.h missing cuserid for UNIX98 and before
346 [20044] libc: unistd.h missing pthread_atfork for UNIX98
347 [20051] libc: ttyslot in wrong header under wrong conditions
348 [20054] libc: gethostname not declared for XPG4
349 [20055] libc: termios.h missing tcgetsid for XPG4
350 [20072] dynamic-link: x86 init_cpu_features is called twice in static
352 [20073] libc: sys/stat.h fchmod namespace
353 [20074] libc: stdlib.h rand_r namespace
354 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
355 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
357 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
358 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
360 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
361 [20119] libc: Wrong mask for processors level type from CPUID
362 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
363 AS not supporting AVX512
364 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
366 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
367 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
369 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
370 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
371 [20178] libc: posix_spawn{p} should not call exit
372 [20191] stdio: libio: vtables hardening
373 [20195] string: FMA4 detection requires CPUID execution with register
375 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
376 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
378 [20212] math: acos (sNaN) returns sNaN
379 [20213] math: asin (sNaN) returns sNaN
380 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
381 [20218] math: [i386] asinhl (sNaN) returns sNaN
382 [20219] math: [i386] atanhl (sNaN) returns sNaN
383 [20222] stdio: fopencookie: Mangle function pointers
384 [20224] math: [i386] cbrtl (sNaN) returns sNaN
385 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
386 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
388 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
389 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
390 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
391 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
392 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
393 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
394 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
395 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
396 [20240] math: modf (sNaN) returns sNaN
397 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
398 [20250] math: frexp (sNaN) returns sNaN
399 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
400 [20255] math: [i386] fdim, fdimf return with excess range and precision /
402 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
403 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
404 subscript is below array bounds [-Werror=array-bounds]
405 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
406 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
408 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
409 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
411 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
413 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
414 [20316] localedata: id_ID: Februari instead of Pebruari
415 [20327] string: POWER8 strcasecmp returns incorrect result
416 [20347] math: Failure: Test: j0_downward (0xap+0)
417 [20348] libc: FAIL: misc/tst-preadvwritev64
418 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
419 [20350] libc: There is no test for p{read,write}64
420 [20357] math: Incorrect cos result for 1.5174239687223976
421 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
425 * Unicode 8.0.0 Support: Character encoding, character type info, and
426 transliteration tables are all updated to Unicode 8.0.0, using new
427 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
428 These updates cause user visible changes, such as the fixes for bugs
429 89, 16061, and 18568.
431 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
432 kernel-internal CPU set size. This means that requests that change the
433 CPU affinity which failed before (for example, an all-ones CPU mask) will
434 now succeed. Applications that need to determine the effective CPU
435 affinities need to call sched_getaffinity or pthread_getaffinity_np after
436 setting it because the kernel can adjust it (and the previous size check
437 would not detect this in the majority of cases).
439 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
440 the following new symbols are used: fts64_children, fts64_close,
441 fts64_open, fts64_read and fts64_set.
443 * getaddrinfo now detects certain invalid responses on an internal netlink
444 socket. If such responses are received, an affected process will
445 terminate with an error message of "Unexpected error <number> on netlink
446 descriptor <number>" or "Unexpected netlink response of size <number> on
447 descriptor <number>". The most likely cause for these errors is a
448 multi-threaded application which erroneously closes and reuses the netlink
449 file descriptor while it is used by getaddrinfo.
451 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
452 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
453 unnecessary serialization of memory allocation requests across threads.
454 The defect is now corrected. Users should see a substantial increase in
455 the concurent throughput of allocation requests for applications which
456 trigger this bug. Affected applications typically create and
457 destroy threads frequently. (Bug 19048 was reported and analyzed by
460 * There is now a --disable-timezone-tools configure option for disabling the
461 building and installing of the timezone related utilities (zic, zdump, and
462 tzselect). This is useful for people who build the timezone data and code
463 independent of the GNU C Library.
465 * The obsolete header <regexp.h> has been removed. Programs that require
466 this header must be updated to use <regex.h> instead.
468 * The obsolete functions bdflush, create_module, get_kernel_syms,
469 query_module and uselib are no longer available to newly linked binaries;
470 the header <sys/kdaemon.h> has been removed. These functions and header
471 were specific to systems using the Linux kernel and could not usefully be
472 used with the GNU C Library on systems with version 2.6 or later of the
475 * Optimized string, wcsmbs and memory functions for IBM z13.
476 Implemented by Stefan Liebler.
478 * Newly linked programs that define a variable called signgam will no longer
479 have it set by the lgamma, lgammaf and lgammal functions. Programs that
480 require signgam to be set by those functions must ensure that they use the
481 variable provided by the GNU C Library and declared in <math.h>, without
482 defining their own copy.
484 * The minimum GCC version that can be used to build this version of the GNU
485 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
486 still be used to compile programs using the GNU C Library.
488 Security related changes:
490 * An out-of-bounds value in a broken-out struct tm argument to strftime no
491 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
493 * The LD_POINTER_GUARD environment variable can no longer be used to disable
494 the pointer guard feature. It is always enabled. Previously,
495 LD_POINTER_GUARD could be used to disable security hardening in binaries
496 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
499 * An integer overflow in hcreate and hcreate_r could lead to an
500 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
502 * The catopen function no longer has unbounded stack usage. Reported by
505 * The nan, nanf and nanl functions no longer have unbounded stack usage
506 depending on the length of the string passed as an argument to the
507 functions. Reported by Joseph Myers. (CVE-2014-9761)
509 * A stack-based buffer overflow was found in libresolv when invoked from
510 libnss_dns, allowing specially crafted DNS responses to seize control
511 of execution flow in the DNS client. The buffer overflow occurs in
512 the functions send_dg (send datagram) and send_vc (send TCP) for the
513 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
514 family. The use of AF_UNSPEC triggers the low-level resolver code to
515 send out two parallel queries for A and AAAA. A mismanagement of the
516 buffers used for those queries could result in the response of a query
517 writing beyond the alloca allocated buffer created by
518 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
519 the overflow. Thanks to the Google Security Team and Red Hat for
520 reporting the security impact of this issue, and Robert Holiday of
521 Ciena for reporting the related bug 18665. (CVE-2015-7547)
523 The following bugs are resolved with this release:
525 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
526 [887] math: Math library function "logb" and "nextafter" inconsistent
527 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
528 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
529 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
530 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
532 [4404] localedata: German translation of "Alarm clock" is misleading
533 [6799] math: nextafter() and nexttoward() doen't set errno on
534 overflow/underflow errors
535 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
537 [10432] nis: _nss_nis_setnetgrent assertion failure
538 [11460] libc: fts has no LFS support
539 [12926] network: getaddrinfo()/make_request() may spin forever
540 [13065] nptl: Race condition in pthread barriers
541 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
542 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
544 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
545 [14912] libc: Rename non-installed bits/*.h headers
546 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
547 [15367] math: Let gcc use __builtin_isinf
548 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
549 [15421] math: lgamma wrongly sets signgam for ISO C
550 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
551 FE_INVALID with argument out of range
552 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
554 [15786] dynamic-link: ifunc resolver functions can smash function
556 [15918] math: Unnecessary check for equality in hypotf()
557 [16061] localedata: Review / update transliteration data
558 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
560 [16141] time: strptime %z offset restriction
561 [16171] math: drem should be alias of remainder
562 [16296] math: fegetround is pure?
563 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
564 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
566 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
567 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
569 [16422] math: [powerpc] math-float, math-double failing llrint tests with
570 "Exception "Inexact" set" on ppc32
571 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
572 [16517] math: Missing underflow exception from tanf/tan/tanl
573 [16519] math: Missing underflow exception from sinhf
574 [16520] math: Missing underflow exception from tanhf
575 [16521] math: Missing underflow exception from exp2
576 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
578 [16734] stdio: fopen calls mmap to allocate its buffer
579 [16961] math: nan function incorrect handling of bad sequences
580 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
581 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
582 [16985] locale: localedef: confusing error message when opening output
584 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
585 [17197] locale: Redundant shift character in iconv conversion output at
587 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
588 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
589 [17250] dynamic-link: static linking breaks nss loading
590 (getaddrinfo/getpwnam/etc...)
591 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
593 [17441] math: isnan() should use __builtin_isnan() in GCC
594 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
595 (related to lock elision)
596 [17787] manual: Exponent on page 324 of the PDF ends prematurely
597 [17886] time: strptime should be able to parse "Z" as a timezone with %z
598 [17887] time: strptime should be able to parse "+01:00" style timezones
599 [17905] libc: catopen() Multiple unbounded stack allocations
601 [18084] libc: backtrace (..., 0) dumps core on x86
602 [18086] libc: nice() sets errno to 0 on success
603 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
604 is too large (CVE-2015-8778)
605 [18251] dynamic-link: SONAME missing when audit modules provides path
606 [18265] libc: add attributes for wchar string and memory functions
607 [18370] math: csqrt missing underflows
608 [18421] libc: [hppa] read-only segment has dynamic relocations
609 [18472] libc: Obsolete syscall wrappers should be compat symbols
610 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
611 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
612 [18525] localedata: Remove locale timezone information
613 [18560] libc: [powerpc] spurious bits/ipc.h definitions
614 [18568] localedata: Update locale data to Unicode 8.0
615 [18589] locale: sort-test.sh fails at random
616 [18595] math: ctan, ctanh missing underflows
617 [18604] libc: assert macro-expands its argument
618 [18610] math: S390: fetestexcept() reports any exception if DXC-code
619 contains a vector instruction exception.
620 [18611] math: j1, jn missing errno setting on underflow
621 [18618] localedata: sync Chechen locale definitions with other *_RU
623 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
624 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
625 [18665] network: In send_dg, the recvfrom function is NOT always using the
626 buffer size of a newly created buffer (CVE-2015-7547)
627 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
629 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
631 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
632 [18699] math: tilegx cproj() for various complex infinities does not yield
634 [18724] libc: Harden put*ent functions against data injection
635 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
636 [18755] build: build errors with -DNDEBUG
637 [18757] stdio: fmemopen fails to set errno on failure
638 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
640 [18781] libc: openat64 lacks O_LARGEFILE
641 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
642 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
643 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
644 [18790] math: [ldbl-128ibm] tanhl inaccurate
645 [18795] libc: stpncpy fortification misses buffer lengths that are
647 [18796] build: build fails for --disable-mathvec
648 [18803] math: hypot missing underflows
649 [18820] stdio: fmemopen may leak memory on failure
650 [18823] math: csqrt spurious underflows
651 [18824] math: fma spurious underflows
652 [18825] math: pow missing underflows
653 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
654 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
655 [18870] build: sem_open.c fails to compile with missing symbol
657 [18872] stdio: Fix memory leak in printf_positional
658 [18873] libc: posix_fallocate overflow check ineffective
659 [18875] math: Excess precision leads incorrect libm
660 [18877] libc: arm: mmap offset regression
661 [18887] libc: memory corruption when using getmntent on blank lines
662 [18918] localedata: hu_HU: change time to HH:MM:SS format
663 [18921] libc: Regression: extraneous stat() and fstat() performed by
665 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
666 binaries (CVE-2015-8777)
667 [18951] math: tgamma missing underflows
668 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
670 [18953] localedata: lt_LT: change currency symbol to the euro
671 [18956] math: powf inaccuracy
672 [18961] math: [i386] exp missing underflows
673 [18966] math: [i386] exp10 missing underflows
674 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
675 [18969] build: multiple string test failures due to missing locale
677 [18970] libc: Reference of pthread_setcancelstate in libc.a
678 [18977] math: float / long double Bessel functions not in XSI POSIX
679 [18980] math: i386 libm functions return with excess range and precision
680 [18981] math: i386 scalb*, ldexp return with excess range and precision
681 [18982] stdio: va_list and vprintf
682 [18985] time: Passing out of range data to strftime() causes a segfault
684 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
685 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
686 [19012] locale: iconv_open leaks memory on error path
687 [19016] math: clog, clog10 inaccuracy
688 [19018] nptl: Mangle function pointers in tls_dtor_list
689 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
690 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
691 [19048] malloc: malloc: arena free list can become cyclic, increasing
693 [19049] math: [powerpc] erfc incorrect zero sign
694 [19050] math: [powerpc] log* incorrect zero sign
695 [19058] math: [x86_64] Link fail with -fopenmp and -flto
696 [19059] math: nexttoward overflow incorrect in non-default rounding modes
697 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
698 [19074] network: Data race in _res_hconf_reorder_addrs
699 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
700 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
701 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
703 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
705 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
706 [19086] manual: posix_fallocate64 documented argument order is wrong.
707 [19088] math: lround, llround missing exceptions close to overflow
709 [19094] math: lrint, llrint missing exceptions close to overflow threshold
710 [19095] math: dbl-64 lrint incorrect for 64-bit long
711 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
712 [19124] dynamic-link: ld.so failed to build with older assmebler
713 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
714 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
715 [19134] math: [powerpc32] lround, lroundf spurious exceptions
716 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
717 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
718 pthread_setaffinity_np
719 [19156] math: [ldbl-128] j0l spurious underflows
720 [19164] nptl: tst-getcpu fails with many possible CPUs
721 [19168] math: math/test-ildoubl and math/test-ldouble failure
722 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
723 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
725 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
726 (FE_NOMASK_ENV) do not clear SSE exceptions
727 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
728 _int_new_arena/reused_arena
729 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
730 [19201] math: dbl-64 remainder incorrect sign of zero result
731 [19205] math: bits/math-finite.h conditions do not match math.h and
733 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
734 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
735 for C99-based standards
736 [19212] libc: features.h not -Wundef clean
737 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
739 [19214] libc: Family and model identification for AMD CPU's are incorrect.
740 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
741 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
743 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
745 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
747 [19242] libc: strtol incorrect in Turkish locales
748 [19243] malloc: reused_arena can pick an arena on the free list, leading
749 to an assertion failure and reference count corruption
750 [19253] time: tzset() ineffective when temporary TZ did not include DST
752 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
753 [19270] math: [hppa] Shared libm missing __isnanl
754 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
755 MAP_HUGETLB and MAP_STACK defines
756 [19313] nptl: Wrong __cpu_mask for x32
757 [19347] libc: grantpt: try to force a specific gid even without pt_chown
758 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
759 [19350] math: [ldbl-128ibm] sinhl spurious overflows
760 [19351] math: [ldbl-128ibm] logl inaccurate near 1
761 [19363] time: x32: times() return value wrongly truncates/sign extends
763 [19367] dynamic-link: Improve branch prediction on Silvermont
764 [19369] network: Default domain name not reset by res_ninit when "search"
765 / "domain" entry is removed from resolv.conf
766 [19375] math: powerpc: incorrect results for POWER7 logb with negative
768 [19385] localedata: bg_BG: time separator should be colon, not comma
769 [19408] libc: linux personality syscall wrapper may erroneously return an
770 error on 32-bit architectures
771 [19415] libc: dladdr returns wrong names on hppa
772 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
773 IBM905, IBM907, and IBM909
774 [19439] math: Unix98 isinf and isnan functions conflict with C++11
775 [19443] build: build failures with -DDEBUG
776 [19451] build: Make check fails on test-double-vlen2
777 [19462] libc: Glibc failed to build with -Os
778 [19465] math: Wrong code with -Os
779 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
781 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
783 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
784 [19486] math: S390: Math tests fail with "Exception Inexact set".
785 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
786 [19550] libc: [mips] mmap negative offset handling inconsistent with other
788 [19590] math: Fail to build shared objects that use libmvec.so functions.
792 * The following bugs are resolved with this release:
794 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
795 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
796 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
797 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
798 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
799 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
800 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
801 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
802 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
803 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
804 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
805 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
806 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
807 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
808 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
809 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
810 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
811 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
812 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
813 18657, 18676, 18694, 18696, 18887.
815 * Cache information can be queried via sysconf() function on s390 e.g. with
816 _SC_LEVEL1_ICACHE_SIZE as argument.
818 * A buffer overflow in gethostbyname_r and related functions performing DNS
819 requests has been fixed. If the NSS functions were called with a
820 misaligned buffer, the buffer length change due to pointer alignment was
821 not taken into account. This could result in application crashes or,
822 potentially arbitrary code execution, using crafted, but syntactically
823 valid DNS responses. (CVE-2015-1781)
825 * The time zone file parser has been made more robust against crafted time
826 zone files, avoiding heap buffer overflows related to the processing of
827 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
828 large time zone data files. Overly long time zone specifiers in the TZ
829 variable no longer result in stack overflows and crashes.
831 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
832 for LD and GD on x86 and x86-64, has been implemented. You will need
833 binutils-2.24 or later to enable this optimization.
835 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
836 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
837 Hat). These updates cause user visible changes, such as the fix for bug
840 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
841 and getXXbyYY NSS calls for the same database, causing a denial-of-service
842 condition in some applications.
844 * Added vector math library named libmvec with the following vectorized x86_64
845 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
847 The library can be disabled with --disable-mathvec. Use of the functions is
848 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
849 Shared library libmvec.so is linked in as needed when using -lm (no need to
850 specify -lmvec explicitly for not static builds).
851 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
853 * A new fmemopen implementation has been added with the goal of POSIX
854 compliance. The new implementation fixes the following long-standing
855 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
856 old implementation is still present for use be by existing binaries.
858 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
859 releases. It has been fixed to match 2.19 and older, but binaries built
860 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
862 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
863 Contributed by Roland McGrath (Google).
865 * The header <regexp.h> is deprecated, and will be removed in a future
866 release. Use of this header will trigger a deprecation warning.
867 Application developers should update their code to use <regex.h> instead.
869 This header was formerly part of SUS, but was deprecated in 1994 and
870 removed from the standard in 2001. Also, the glibc implementation
871 leaks memory. See BZ#18681 for more details.
875 * The following bugs are resolved with this release:
877 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
878 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
879 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
880 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
881 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
882 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
883 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
884 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
885 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
886 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
889 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
890 too little memory for the to-be-scanned arguments and overflow the
891 allocated buffer. The implementation now correctly computes the required
892 buffer size when using malloc, and switches to malloc from alloca as
895 * A new semaphore algorithm has been implemented in generic C code for all
896 machines. Previous custom assembly implementations of semaphore were
897 difficult to reason about or ensure that they were safe. The new version
898 of semaphore supports machines with 64-bit or 32-bit atomic operations.
899 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
900 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
902 * Port to Altera Nios II has been contributed by Mentor Graphics.
904 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
905 implementations for powerpc64/powerpc64le.
906 Implemented by Adhemerval Zanella (IBM).
908 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
909 and powerpc64le. This may improve lock scaling of existing programs on
910 HTM capable systems. The lock elision code is only enabled with
911 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
912 powerpc will issue a transaction abort on every syscall to avoid side
913 effects being visible outside transactions.
915 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
916 AArch64. Contributed by ARM Ltd.
918 * i386 memcpy functions optimized with SSE2 unaligned load/store.
920 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
921 under certain input conditions resulting in the execution of a shell for
922 command substitution when the application did not request it. The
923 implementation now checks WRDE_NOCMD immediately before executing the
924 shell and returns the error WRDE_CMDSUB as expected.
926 * CVE-2012-3406 printf-style functions could run into a stack overflow when
927 processing format strings with a large number of format specifiers.
929 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
930 infinite loop if the DNS response contained a PTR record of an unexpected
933 * The minimum GCC version that can be used to build this version of the GNU
934 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
935 still be used to compile programs using the GNU C Library.
937 * The GNU C Library is now built with -Werror by default. This can be
938 disabled by configuring with --disable-werror.
940 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
942 * The obsolete sigvec function has been removed. This was the original
943 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
944 programs have been using instead for about 25 years. Of course, ABI
945 compatibility for old binaries using sigvec remains intact.
947 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
948 with newer versions of bison.
950 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
951 The original MIPS o32 hard-float ABI requires an FPU where double-precision
952 registers overlay two consecutive single-precision registers. MIPS32R2
953 introduced a new FPU mode (FR=1) where double-precision registers extend the
954 corresponding single-precision registers which is incompatible with the
955 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
956 require the use of FR=1 making a transition necessary. New o32 ABI
957 extensions enable users to migrate over time from the original o32 ABI
958 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
959 tracks the ABI of any loaded object and verifies that new objects are
960 compatible. Mode transitions will also be requested as required and
961 unsupportable objects will be rejected. The ABI checks include both soft and
962 hard float ABIs for o32, n32 and n64.
964 GCC 5 with GNU binutils 2.25 onwards:
965 It is strongly recommended that all o32 system libraries are built using the
966 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
967 with the original and all new o32 ABI extensions. Configure a MIPS GCC
968 compiler using --with-fp-32=xx to set this by default.
972 * The following bugs are resolved with this release:
974 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
975 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
976 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
977 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
978 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
979 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
980 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
981 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
982 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
983 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
984 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
985 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
986 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
987 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
988 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
990 * Reverted change of ABI data structures for s390 and s390x:
991 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
992 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
993 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
994 are preserved pointing straight to the same implementation as the old ones.
995 Given that, new callers will simply provide a too-big buffer to these
996 functions. Any applications/libraries out there that embed jmp_buf or
997 ucontext_t in an ABI-relevant data structure that have already been rebuilt
998 against 2.19 headers will have to rebuilt again. This is necessary in any
999 case to revert the breakage in their ABI caused by the glibc change.
1001 * Support for file description locks is added to systems running the
1002 Linux kernel. The standard file locking interfaces are extended to
1003 operate on file descriptions, not file descriptors, via the use of
1004 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1005 are associated with an open file instead of a process.
1007 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1009 * The minimum Linux kernel version that this version of the GNU C Library
1010 can be used with is 2.6.32.
1012 * Running the testsuite no longer terminates as soon as a test fails.
1013 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1014 with PASS or FAIL lines for individual tests. A summary of the results is
1015 printed, including a list of failing lists, and "make check" exits with
1016 error status if there were any unexpected failures. "make check
1017 stop-on-test-failure=y" may be used to keep the old behavior.
1019 * The am33 port, which had not worked for several years, has been removed
1022 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1023 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1024 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1025 conflicted with POSIX (support for which was removed in 2.19), the
1026 interfaces those macros enabled remain available when compiling with
1027 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1028 test macros defined.
1030 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1032 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1033 This may improve lock scaling of existing programs on TX capable systems.
1034 The lock elision code is only built with --enable-lock-elision=yes and
1035 then requires a GCC version supporting the TX builtins. With lock elision
1036 default mutexes are elided via __builtin_tbegin, if the cpu supports
1037 transactions. By default lock elision is not enabled and the elision code
1040 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1041 copy the path argument. This allowed programs to cause posix_spawn to
1042 deference a dangling pointer, or use an unexpected pathname argument if
1043 the string was modified after the posix_spawn_file_actions_addopen
1046 * All supported architectures now use the main glibc sysdeps directory
1047 instead of some being in a separate "ports" directory (which was
1048 distributed separately before glibc 2.17).
1050 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1051 On configurations that support it (all Linux configurations), it's now
1052 used regardless of the --enable-add-ons switch to configure. It is no
1053 longer possible to build such configurations without pthreads support.
1055 * Locale names, including those obtained from environment variables (LANG
1056 and the LC_* variables), are more tightly checked for proper syntax.
1057 setlocale will now fail (with EINVAL) for locale names that are overly
1058 long, contain slashes without starting with a slash, or contain ".." path
1059 components. (CVE-2014-0475) Previously, some valid locale names were
1060 silently replaced with the "C" locale when running in AT_SECURE mode
1061 (e.g., in a SUID program). This is no longer necessary because of the
1064 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1065 with application code using Intel MPX instructions. (With all previous
1066 versions, the MPX register state could be clobbered when making calls
1067 into or out of a shared library.) Note that while the new dynamic
1068 linker is compatible with all known x86 hardware whether or not it
1069 supports Intel MPX, some x86 instruction-set emulators might fail to
1070 handle the new instruction encodings. This is known to affect Valgrind
1071 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1072 release), and might affect other tools that do instruction emulation.
1074 * Support for loadable gconv transliteration modules has been removed.
1075 The support for transliteration modules has been non-functional for
1076 over a decade, and the removal is prompted by security defects. The
1077 normal gconv conversion modules are still supported. Transliteration
1078 with //TRANSLIT is still possible, and the //IGNORE specifier
1079 continues to be supported. (CVE-2014-5119)
1081 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1082 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1083 resulting a denial-of-service security vulnerability in applications which
1084 use functions related to iconv. (CVE-2014-6040)
1088 * The following bugs are resolved with this release:
1090 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1091 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1092 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1093 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1094 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1095 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1096 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1097 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1098 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1099 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1100 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1101 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1102 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1103 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1104 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1105 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1106 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1107 16431, 16453, 16474, 16506, 16510, 16529
1109 * Slovenian translations for glibc messages have been contributed by the
1110 Translation Project's Slovenian team of translators.
1112 * The public headers no longer use __unused nor __block. This change is to
1113 support compiling programs that are derived from BSD sources and use
1114 __unused internally, and to support compiling with Clang's -fblock
1115 extension which uses __block.
1117 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1118 large collation sequences to optimize multiple passes. This cache
1119 computation may overflow for large collation sequences and may cause a
1120 stack or buffer overflow. This is now fixed to use a slower algorithm
1121 which does not use a cache if there is an integer overflow.
1123 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1124 rules for large collation sequences to optimize multiple passes and falls
1125 back to alloca if malloc fails, resulting in a possible stack overflow.
1126 The implementation now falls back to an uncached collation sequence lookup
1129 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1130 initialized for static applications resulting in the security feature
1131 being disabled. The pointer guard is now correctly initialized to a
1132 random value for static applications. Existing static applications need
1133 to be recompiled to take advantage of the fix (bug 15754).
1135 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1136 to the d_name member of struct dirent, or omit the terminating NUL
1137 character. (Bugzilla #14699).
1139 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1140 aligned_alloc functions could allocate too few bytes or corrupt the
1141 heap when passed very large allocation size values (Bugzilla #15855,
1144 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1145 for AF_INET6 has been fixed (Bugzilla #16072).
1147 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1148 pap_CW, quz_PE, the_NP.
1150 * Substantially revised locales: gd_GB, ht_HT
1152 * The LC_ADDRESS field was updated to support country_car for almost all
1155 * ISO 1427 definitions were updated.
1157 * ISO 3166 definitions were updated.
1159 * The localedef utility now supports --big-endian and --little-endian
1160 command-line options to generate locales for a different system from that
1161 for which the C library was built.
1163 * Binary locale files now only depend on the endianness of the system for
1164 which they are generated and not on other properties of that system. As a
1165 consequence, binary files generated with new localedef may be incompatible
1166 with old versions of the GNU C Library, and binary files generated with
1167 old localedef may be incompatible with this version of the GNU C Library,
1168 in the following circumstances:
1170 + Locale files may be incompatible on m68k systems.
1172 + Locale archive files (but not separate files for individual locales) may
1173 be incompatible on systems where plain "char" is signed.
1175 * The configure option --disable-versioning has been removed. Builds with
1176 --disable-versioning had not worked for several years.
1178 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1179 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1181 * SystemTap probes for malloc have been introduced.
1183 * SystemTap probes for slow multiple precision fallback paths of
1184 transcendental functions have been introduced.
1186 * Support for powerpc64le has been added.
1188 * The soft-float powerpc port now supports e500 processors.
1190 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1192 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1193 set of header declarations that are enabled by default, even when other
1194 feature test macros or compiler options such as -std=c99 would otherwise
1195 disable some of those declarations.
1197 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1198 conflict with POSIX. The libbsd-compat library (which was a dummy library
1199 that did nothing) has also been removed.
1201 * Preliminary documentation about Multi-Thread, Async-Signal and
1202 Async-Cancel Safety has been added.
1204 * Change of ABI data structures for s390 and s390x:
1205 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1206 allow for future hardware extensions. All library functions that accept or
1207 return these structures were versioned in order to provide backwards
1208 compatibility with existing code. However, not all cases can be handled
1209 with interface versioning. If an application embeds either structure into
1210 another structure and that structure is passed to another compilation unit
1211 which expects the newer larger-sized object then any access to the new
1212 fields will result in undefined behaviour. Similarly any access to fields
1213 that were moved by the enlarging of the structures will cause undefined
1214 behaviour. To fix the undefined behaviour all objects that are part of the
1215 final in-memory image must agree on the size of structures, and this may
1216 require recompilation.
1220 * The following bugs are resolved with this release:
1222 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1223 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1224 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1225 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1226 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1227 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1228 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1229 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1230 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1231 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1232 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1233 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1234 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1237 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1238 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1239 Distributions can re-enable building and using pt_chown via the new configure
1240 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1241 considerable security risks and should only be used if the distribution
1242 understands and accepts the risks.
1244 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1247 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1248 fixed (Bugzilla #15330).
1250 * Add support for calling C++11 thread_local object destructors on thread
1251 and program exit. This needs compiler support for offloading C++11
1252 destructor calls to glibc.
1254 * Improved worst case performance of libm functions with double inputs and
1257 * Support for priority inherited mutexes in pthread condition variables on
1258 non-x86 architectures.
1260 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1262 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1264 * Optimized string functions for ARM. Implemented by Will Newton and
1267 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1269 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1272 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1273 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1275 * Added a benchmark framework to track performance of functions in glibc.
1277 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1278 It is based on draft TS 18661 and currently enabled as a GNU extension.
1280 * On Linux, the clock function now uses the clock_gettime system call
1281 for improved precision, rather than old times system call.
1283 * Added support for version-3 tz format files. This is needed when using
1284 the tz database release 2013e or later, and affects a few unusual cases --
1285 currently only TZ='America/Godthab' for time stamps after 2037.
1287 * Added new API functions pthread_getattr_default_np and
1288 pthread_setattr_default_np to get and set the default pthread
1289 attributes of a process.
1291 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1292 This may improve lock scaling of existing programs on TSX capable systems.
1293 When the --enable-lock-elision=yes parameter is specified at configure
1294 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1297 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1298 availability). Implemented by Ryan S. Arnold.
1300 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1304 * The following bugs are resolved with this release:
1306 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1307 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1308 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1309 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1310 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1311 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1312 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1313 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1314 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1315 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1316 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1317 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1318 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1319 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1320 14879, 14889, 14893, 14898, 14914.
1322 * Optimization of memcpy for MIPS.
1324 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1325 EMFILE has been fixed (Bugzilla #14889).
1327 * The add-on ports collection is now distributed in the "ports" subdirectory
1328 of the main GNU C Library distribution, rather than separately.
1330 * Port to ARM AArch64 contributed by Linaro.
1332 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1333 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1335 Implemented by Andreas Krebbel.
1337 * The new function secure_getenv allows secure access to the environment,
1338 returning NULL if running in a SUID/SGID process. This function replaces
1339 the internal function __secure_getenv.
1341 * SystemTap static probes have been added into the dynamic linker.
1342 Implemented by Gary Benson.
1344 * Optimizations of string functions strstr, strcasestr and memmem.
1345 Implemented by Maxim Kuvyrkov.
1347 * The minimum Linux kernel version that this version of the GNU C Library
1348 can be used with is 2.6.16.
1350 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1351 powerpc POWER7. Implemented by Will Schmidt.
1353 * New configure option --disable-nscd builds the C library such that it
1354 never attempts to contact the Name Service Caching Daemon (nscd).
1355 New configure option --disable-build-nscd avoids building nscd itself;
1356 this is the default if --disable-nscd is used.
1358 * Improved support for cross-compilation, including cross-testing and
1359 bootstrap builds without a previously built glibc.
1361 * Several testsuite tests are now able to test multiple IFUNC variants of an
1362 interface, rather than just testing the one that would be chooen by
1365 * New configure options --with-bugurl and --with-pkgversion, for
1366 distributors to use to embed their bug-reporting and package version
1367 information in --help and --version output.
1369 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1370 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1371 allows creation of chroots without the procfs mounted on /proc.
1373 * The `crypt' function now fails if passed salt bytes that violate the
1374 specification for those values. On Linux, the `crypt' function will
1375 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1376 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1377 when the mode is enabled.
1379 * The `clock_*' suite of functions (declared in <time.h>) is now available
1380 directly in the main C library. Previously it was necessary to link with
1381 -lrt to use these functions. This change has the effect that a
1382 single-threaded program that uses a function such as `clock_gettime' (and
1383 is not linked with -lrt) will no longer implicitly load the pthreads
1384 library at runtime and so will not suffer the overheads associated with
1385 multi-thread support in other code such as the C++ runtime library.
1387 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1392 * The following bugs are resolved with this release:
1394 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1395 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1396 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1397 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1398 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1399 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1400 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1401 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1402 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1403 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1404 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1405 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1406 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1407 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1408 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1409 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1410 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1411 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1412 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1413 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1414 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1415 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1418 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1419 configuring glibc with:
1420 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1421 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1422 Implemented by H.J. Lu.
1426 + define static_assert
1428 + do not declare gets
1430 + declare at_quick_exit and quick_exit also for ISO C11
1432 + aligned_alloc. NB: The code is deliberately allows the size parameter
1433 to not be a multiple of the alignment. This is a moronic requirement
1434 in the standard but it is only a requirement on the caller, not the
1437 + timespec_get added
1439 + uchar.h support added
1441 + CMPLX, CMPLXF, CMPLXL added
1443 Implemented by Ulrich Drepper.
1445 * Support for the IA-64 has been moved to ports.
1447 * Remove support for anything but ELF binary format
1449 * Checking versions of poll, ppoll added.
1450 Implemented by Ulrich Drepper.
1452 * More generic and 64-bit performance optimizations to math functions.
1453 Implemented by Ulrich Drepper.
1455 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1456 headers and functions available at compile time as they were before
1457 version 2.14. This option will be removed at some time in the future
1458 after the TI-RPC library becomes fully sufficient for the needs of
1459 existing applications.
1461 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1462 Note that glibc is not expected to work with any Linux kernel version
1465 * New header <sys/auxv.h> and function getauxval allowing easy access to
1466 the AT_* key-value pairs passed from the Linux kernel. The header also
1467 defines the HWCAP_* bits associated with the AT_HWCAP key.
1469 * New locales: mag_IN
1471 * New configure option --enable-systemtap builds SystemTap static probes
1472 into libc for setjmp and longjmp and into libpthread for various operations.
1473 So far the setjmp/longjmp probes and some of the libpthread probes are
1474 provided only for i*86 and x86_64.
1475 Implemented by Roland McGrath and Rayson Ho.
1477 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1479 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1482 * More optimized functions for SPARC. Implemented by David S. Miller.
1484 * Improved support for cross-compilation, especially bootstrap builds
1485 without a previously built glibc.
1487 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1488 Chris Metcalf from Tilera.
1490 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1491 now supported for ARM processors.
1493 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1494 as the name of the dynamic linker, to distinguish it from the
1495 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1497 * Support for CRIS has been removed from ports.
1499 * A new class of installed header has been documented for low-level
1500 platform-specific functionality. PowerPC added the first instance with a
1501 function to provide time base register access. Contributed by Tulio
1502 Magno Quites Machado Filho.
1504 * ix86 configurations now install header files that are consistent with
1505 what x86-64 configurations install. These same header files can be used
1506 for -m32, -m64, or -mx32 builds.
1507 Contributed by H.J. Lu.
1509 * Math library bug fixes. A thorough audit of all open math library bugs was
1510 conducted by Joseph Myers. Significant progress was made on many math
1511 library bugs resulting in more accurate exceptions and function results.
1512 Many thanks to all those that contributed including Andreas Jaeger for his
1513 patch review and work on the x87 trigonometric instruction issues.
1515 * Timezone data is no longer installed. Timezone-related binaries and scripts
1516 will continue to be installed. Users should obtain their timezone data from
1517 their distribution provider or from the tzdata package at
1518 <ftp://munnari.oz.au/pub/>.
1522 * The following bugs are resolved with this release:
1524 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1525 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1526 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1527 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1528 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1529 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1530 13484, 13506, 13515, 13523, 13524, 13538, 13540
1532 * New program pldd to list loaded object of a process
1533 Implemented by Ulrich Drepper.
1535 * Add nss_db support back to glibc. No more dependency on Berkeley db
1536 and support for initgroups lookups.
1537 Implemented by Ulrich Drepper.
1539 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1540 Contributed by HJ Lu.
1542 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1543 Contributed by HJ Lu.
1545 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1546 on x86-32 and x86-64.
1547 Contributed by Liubov Dmitrieva.
1549 * Optimized strchr and strrchr for SSE on x86-32.
1550 Contributed by Liubov Dmitrieva.
1552 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1553 for x86-64 and x86-32.
1554 Contributed by Liubov Dmitrieva.
1556 * New interfaces: scandirat, scandirat64
1557 Implemented by Ulrich Drepper.
1559 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1560 Implemented by Ulrich Drepper.
1562 * nscd now also caches the netgroup database.
1563 Implemented by Ulrich Drepper.
1565 * Integrate libm with gcc's -ffinite-math-only option.
1566 Implemented by Ulrich Drepper.
1568 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1569 to math functions. Implemented by Ulrich Drepper.
1571 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1572 Implemented by Ulrich Drepper.
1574 * New Linux interfaces: process_vm_readv, process_vm_writev
1576 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1577 Implemented by Ulrich Drepper.
1579 * Optimized nearbyint and strcasecmp for PPC.
1580 Implemented by Adhemerval Zanella.
1582 * New locales: bho_IN, unm_US, es_CU, ta_LK
1586 * The following bugs are resolved with this release:
1588 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1589 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1590 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1591 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1592 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1593 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1594 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1595 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1596 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1597 12795, 12811, 12813, 12814, 12841
1599 * The RPC implementation in libc is obsoleted. Old programs keep working
1600 but new programs cannot be linked with the routines in libc anymore.
1601 Programs in need of RPC functionality must be linked against TI-RPC.
1602 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1604 Visible changes of this change include (obviously) the inability to link
1605 programs using RPC functions without referencing the TI-RPC library and the
1606 removal of the RPC headers from the glibc headers.
1607 Implemented by Ulrich Drepper.
1609 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1610 syncfs, setns, sendmmsg
1612 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1613 yue_HK, lij_IT, mhr_RU
1615 * New encodings: CP770, CP771, CP772, CP773, CP774
1617 * New program sotruss to trace calls through PLTs
1618 Implemented by Ulrich Drepper.
1620 * The malloc hook implementation is marked deprecated and will be removed
1621 from the default implementation in the next version. The design never
1622 worked ever since the introduction of threads. Even programs which do
1623 not create threads themselves can use multiple threads created internally.
1627 * The following bugs are resolved with this release:
1629 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1630 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1631 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1632 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1635 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1637 * POWER7 optimizations: memset, memcmp, strncmp
1639 * New optimized string functions for x86-64: strnlen (SSE2),
1640 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1641 Implemented by Ulrich Drepper.
1645 * The following bugs are resolved with this release:
1647 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1648 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1649 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1650 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1651 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1652 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1653 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1654 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1656 * New interfaces: pthread_getname_np, pthread_setname_np
1658 * New Linux interface: recvmmsg
1660 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1662 * The dynamic linker now recognizes supported ABI versions from the
1663 EI_ABIVERSION field in the ELF header.
1664 Implemented by Ulrich Drepper.
1666 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1667 will not be used to fill in password fields in the passwd.byname replies.
1668 Instead it is used to synthesize the shadow.byname table, should it be
1669 missing. This is a useful mode in some installations involving Solaris.
1670 Implemented by Ulrich Drepper.
1672 * New locales: kok_IN, sq_MK, cv_RU
1676 * The following bugs are resolved with this release:
1678 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1679 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1680 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1681 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1682 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1683 10789, 10817, 10823, 10840, 10847
1685 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1687 Implemented by Ulrich Drepper.
1689 * Checking version of longjmp added that fails if an uninitialized stack
1690 frame would be created. Implemented by Ulrich Drepper.
1692 * STT_GNU_IFUNC is now supported in static executables.
1693 Implemented by H.J. Lu.
1695 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1697 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1698 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1699 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1700 Contributed by H.J. Lu.
1702 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1703 Implemented by Ulrich Drepper.
1705 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1707 Contributed by H.J. Lu.
1709 * Support for fma instruction in AVX on x86-64.
1710 Implemented by H.J. Lu and Ulrich Drepper.
1712 * AVX support in x86-64 auditing support in ld.so.
1713 Implemented by H.J. Lu.
1715 * STB_GNU_UNIQUE support added.
1716 Implemented by Ulrich Drepper.
1718 * Implement second fallback mode for optimized DNS lookup for even more
1719 broken environments. If two requests from the same port are not
1720 handled correctly close the socket and open a new one before sending
1721 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1722 can be used to select this mode right away, instead of rediscovering the
1723 necessity is every process again.
1724 Implemented by Ulrich Drepper.
1726 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1727 Implemented by Adam Tkac.
1729 * Optimized iconv conversions for S390x.
1730 Implemented by Andreas Krebbel.
1732 * Using condvars with PI mutexes is now more efficient due to kernel
1733 support for requeueing to PI futexes. NPTL support added for x86-64.
1734 Implemented by Ulrich Drepper.
1736 * New locales: ps_AF, my_MM
1740 * The following bugs are resolved with this release:
1742 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1743 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1744 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1745 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1746 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1748 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1749 Implemented by Ulrich Drepper.
1751 * New Linux interfaces: accept4, fallocate, fallocate64.
1752 Implemented by Ulrich Drepper.
1754 * Correct declarations of string function when used in C++ code. This
1755 could lead to compile errors for invalid C++ code.
1757 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1760 * New POSIX 2008 interface: psiginfo
1761 Implemented by Ulrich Drepper.
1763 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1764 Implemented by Ulrich Drepper.
1766 * Support for selecting between multiple function definitions at runtime
1767 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1769 * The libcrypt library can now use the hash function implementations in
1770 NSS. Implemented by Ulrich Drepper.
1772 * The malloc implementation can be compiled to be less memory efficient
1773 but higher performing in multi-threaded programs.
1774 Implemented by Ulrich Drepper.
1776 * New locales: nan_TW@latin, ks_IN
1778 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1779 Implemented by Ulrich Drepper.
1781 * Extended printf hook support. It is possible to use user-defined types
1782 and extend existing format specifiers.
1783 Implemented by Ulrich Drepper.
1785 * Handling for group shadow files has been added.
1786 Implemented by Ulrich Drepper.
1788 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1789 servers (the case, e.g., for some people using the built-in DNS
1790 server in ADSL modems/routers). There is a once-per-process timeout
1791 in case of a broken server. To avoid it, users can run nscd or put
1792 'options single-request' in /etc/resolv.conf.
1793 Implemented by Ulrich Drepper.
1797 * The following bugs are resolved with this release:
1799 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1800 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1801 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1802 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1803 6968, 6974, 6980, 6995, 7008, 7009, 7029
1805 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1806 up at the same time. Implemented by Ulrich Drepper.
1808 * TLS descriptors for LD and GD on x86 and x86-64.
1809 Implemented by Alexandre Oliva.
1811 * getaddrinfo now handles DCCP and UDPlite.
1812 Implemented by Ulrich Drepper.
1814 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1815 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1816 Implemented by Ulrich Drepper.
1818 * New implementation of memmem, strstr, and strcasestr which is O(n).
1819 Implemented by Eric Blake.
1821 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1823 * Implement "e" option for popen to open file descriptor with the
1824 close-on-exec flag set. Implemented by Ulrich Drepper.
1826 * Implement "b" mode for fmemopen. In this mode writes writes don't
1827 implicitly add a NUL byte and seeks from the end of the buffer really
1828 use the buffer end, not the string length as the basis.
1829 Implemented by Ulrich Drepper.
1831 * Many functions, exported and internal, now atomically set the close-on-exec
1832 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1834 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1836 Implemented by Pravin Satpute.
1838 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1842 * The following bugs are resolved with this release:
1844 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1845 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1846 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1847 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1848 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1849 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1850 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1852 * New locales: bo_CN, bo_IN, shs_CA.
1854 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1856 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1857 Implemented by Pravin Satpute.
1859 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1861 * nscd caches now all timeouts for DNS entries
1862 Implemented by Ulrich Drepper.
1864 * nscd is more efficient and wakes up less often.
1865 Implemented by Ulrich Drepper.
1867 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1868 vdprintf, and obstack_vprintf.
1869 Implemented by Jakub Jelinek.
1871 * Faster memset for x86-64.
1872 Implemented by Harsha Jagasia and H.J. Lu.
1874 * Faster memcpy on x86.
1875 Implemented by Ulrich Drepper.
1877 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1878 Implemented by Ulrich Drepper.
1880 * Faster sqrt and sqrtf implemention for some PPC variants.
1881 Implemented by Steven Munroe.
1885 * The following bugs are resolved with this release:
1887 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1888 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1889 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1890 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1891 5071, 5103, 5104, 5112, 5113, 5184, 5186
1893 * More checking functions: fread, fread_unlocked, open*, mq_open.
1894 Implemented by Jakub Jelinek and Ulrich Drepper.
1896 * Extend fortification to C++. Implemented by Jakub Jelinek.
1898 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1899 by not recognizing 'a' as a modifier when those specs are requested.
1900 Implemented by Jakub Jelinek.
1902 * PPC optimizations to math and string functions.
1903 Implemented by Steven Munroe.
1905 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1906 options to be passed. Implemented by Ulrich Drepper.
1908 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1910 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1911 Implemented by Ulrich Drepper.
1913 * Handle private futexes in the NPTL implementation.
1914 Implemented by Jakub Jelinek and Ulrich Drepper.
1916 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1917 Implemented by Roland McGrath and Ulrich Drepper.
1919 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1921 * SHA-256 and SHA-512 based password encryption.
1922 Implemented by Ulrich Drepper.
1924 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1925 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1928 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1929 Implemented by Ulrich Drepper.
1933 * The following bugs are resolved with this release:
1935 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
1936 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
1937 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
1938 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
1939 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
1940 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
1941 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
1944 * New Linux interfaces: epoll_pwait, sched_getcpu.
1946 * New generic interfaces: strerror_l.
1948 * nscd can now cache the services database. Implemented by Ulrich Drepper.
1952 * The following bugs are resolved with this release:
1954 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
1955 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
1956 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
1957 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
1958 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
1959 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
1960 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
1961 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
1962 3225, 3251, 3252, 3253, 3273
1964 * For Linux, the sorting of addresses returned by getaddrinfo now also
1965 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
1966 Implemented by Ulrich Drepper.
1968 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
1969 Implemented by Ulrich Drepper.
1971 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
1973 * New iconv module for MIK. Contributed by Alexander Shopov.
1975 * For sites with broken group and/or passwd database, the auto-propagate
1976 option of nscd can prevent creating ID lookup entries from the results
1977 of a name lookup and vice versa. This usually is no problem but some
1978 site might have problems with the default behavior.
1979 Implemented by Ulrich Drepper.
1981 * Iterating over entire database in NIS can be slow. With the
1982 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
1983 to trade time for memory. The entire database will be read at once.
1984 Implemented by Ulrich Drepper.
1986 * The interfaces introduced in RFC 3542 have been implemented by
1989 * Support for the new ELF hash table format was added by Ulrich Drepper.
1991 * Support for priority inheritance mutexes added by Jakub Jelinek and
1994 * Support for priority protected mutexes added by Jakub Jelinek.
1996 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
1997 or_IN, csb_PL, fy_NL, sr_ME.
2001 * More overflow detection functions.
2003 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2004 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2006 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2007 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2008 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2009 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2010 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2011 by Masahide Washizawa.
2013 * It is now possible to install audit libraries for the dynamic linker using
2014 LD_AUDIT. Implemented by Ulrich Drepper.
2016 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2017 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2018 For a libc and libpthread that works well on Linux 2.4 kernels, we
2019 recommend using the stable 2.3 branch.
2021 * The new function open_wmemstream acts like open_memstream,
2022 but uses a wchar_t wide character buffer.
2024 * The new function ppoll is an improved version of poll, similar to pselect.
2026 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2027 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2028 renameat, unlinkat, linkat, symlinkat, readlinkat.
2030 * New Linux kernel system calls: unshare,
2031 inotify_init, inotify_add_watch, inotify_rm_watch.
2033 * The euidaccess function is now also known by the alias eaccess,
2034 for compatibility with some other systems.
2036 * Timezone data updated to 2006b version.
2040 * The following bugs are resolved with this release:
2042 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2043 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2044 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2045 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2046 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2047 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2049 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2051 * As of this release, GCC 4 can be used to compile the C Library.
2053 * Timezone data updated to 2005m version.
2057 * The following bugs are resolved with this release:
2059 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2060 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2061 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2062 777, 787, 821, 822, 823, 825
2064 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2068 * Support for RFC 3678. Real implementations exist only for Linux so far.
2069 Implemented by Ulrich Drepper.
2071 * nscd can now cache entries persistently. Expiring entries are reloaded.
2072 For speedups the cache can be shared in memory with client processes.
2073 Implemented by Ulrich Drepper.
2075 * nscd can now perform SELinux checks.
2076 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2078 * getaddrinfo queries are now cached. Canonical name lookup is performed
2080 Implemented by Ulrich Drepper.
2082 * The nothrow function attribute is used when headers are used by gcc when
2083 compiling C code. This can avoid significant amounts of exception
2086 * The malloc functions perform more error checking and are stricter when
2087 it comes to reacting on errors. The default action is to terminate
2088 the process after showing an error message. Implemented by Ulrich Drepper.
2090 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2091 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2093 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2094 namespaces using the new function dlmopen(). This feature is of course,
2095 like most other dynamic loading functionality, not available in statically
2096 linked applications. Implemented by Ulrich Drepper.
2098 * Low-overhead boundary checking variants of string and some stdio functions
2099 were added. These are to be used in conjunction with a gcc patch by
2100 Jakub Jelinek which adds calls to these functions if possible.
2101 Implemented by Jakub Jelinek and Ulrich Drepper.
2103 * Old code for several operating systems and machine architectures that
2104 have not been in working condition in a long time have been removed from
2105 the main source tree maintained by the GNU C Library's maintainers.
2106 These files are now reside in the separate `ports' source module
2107 that is usable as an add-on when building the library.
2111 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2112 interrogate the dynamic linker, compatible with the Solaris interface.
2114 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2115 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2117 * getifaddrs now uses the netlink interface on Linux to get its information.
2118 Implemented by Thorsten Kukuk.
2120 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2121 Implemented by Ulrich Drepper.
2123 * support for non-executable stacks on x86 has been added. Changes mostly
2126 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2129 * getaddrinfo now performs destination address selection according to
2134 * Thread-safe interfaces for many functions that access locale data
2135 were added in version 2.3, but these features were omitted from NEWS.
2136 Many functions have variants with an `_l' suffix that take a `locale_t'
2137 object as a parameter rather than consulting the current locale.
2138 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2139 create and maintain `locale_t' objects. Additionally, the new function
2140 `uselocale' sets "the current locale" (as used by functions not so
2141 parameterized) set for an individual thread. These features were added
2142 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2144 * The functions getresuid, getresgid, setresuid, and setresgid, which
2145 have long been available on Linux, are now declared in <unistd.h>
2146 and are now also available on the Hurd.
2148 * ELF thread-local storage support (TLS) now works on x86-64.
2150 * The new dynamic string token $LIB is expanded in shared library names.
2151 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2153 * Aldy Hernandez contributed complete software floating point support for
2154 PowerPC machines with no FPU.
2156 * fexecve is implemented on Linux.
2158 * The `btowc' function should work at least twice as fast due to
2159 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2161 * With approriate thread add-ons cancelable functions are now implemented
2162 in libc.so as well. No need to call the function in libpthread. This
2163 change allowed to finally disable the incorrect and expensive handling
2164 of weak definition in ld.so.
2166 * Yet more PLT entries in libc.so have been removed. We finally arrived
2167 at the bare minimum. Startup times improved appropriately.
2169 * Support for the new Linux/x86 system call interface was added. The
2170 AT_SYSINFO auxiliary vector entry is recognized and handled.
2174 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2177 * iconv (the program and the interface) now accepts empty names (excluding
2178 options like //TRANSLIT) to mean "use charset of current locale".
2180 * localedef can now transliterate characters in strings which are not in
2181 the provided charmap. The information from the input locale is used.
2183 * Prelinking support was added for ELF targets. This requires additional
2184 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2186 * Read-only stdio streams now use mmap to speed up operation by eliminating
2187 copying and buffer underflows. To use add 'm' to the mode string of
2188 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2190 * The malloc functions were completely rewritten by Wolfram Gloger based
2191 on Doug Lea's malloc-2.7.0.c.
2193 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2194 implementation of regex.
2196 * Bruno Haible upgraded the iconv and locale implementation to support
2199 * Contents of the LC_* and LANG environment variables in the CEN style are
2200 not recognized anymore. It never was used. Change by Ulrich Drepper.
2202 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2203 thread-local storage (TLS) ABI on some platforms.
2204 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2206 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2207 EUC-JISX0213, and TSCII.
2209 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2210 BSD-compatible interface for getting all network interface addresses.
2211 Implementation for IPv4 by Roland McGrath.
2213 * Loading of locale data is faster due to the introduction of a locale
2214 archive. Implemented by Roland McGrath and Ulrich Drepper.
2216 * Startup times are significantly reduced by not using exported functions
2217 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2220 * Steven Munroe contributed a port to PowerPC64/Linux.
2224 * The Hurd now uses the GNU libio implementation of stdio rather than the
2225 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2227 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2228 with the same behavior as the Linux system call of the same name.
2232 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2233 128-bit long double format.
2235 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2236 IBM1161, and IBM1162 charsets.
2238 * Andreas Jaeger contributed a port to x86-64/Linux.
2240 * Peter Bruin contributed a port to PowerPC/Hurd.
2242 * libc functions using I/O streams now can handle wide-oriented streams
2245 * optimizations in the dynamic linker. Binaries created by recent binutils
2246 versions start up quicker due to reduced time spent on relocations.
2248 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2252 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2253 asinh, atanh, j0 for the 128-bit long double format.
2255 * Bruno Haible updated all the code handling Unicode in some form to
2256 support Unicode 3.1.
2258 * Speed of regex for single-byte locales is back to previous levels.
2259 Patch by Isamu Hasegawa.
2261 * Alpha, SPARC, and IA-64 now also using floating stacks.
2263 * Startup time of internationalized applications greatly improved through
2264 iconv cache. Use iconvconfig to generate the cache file.
2265 Contributed by Ulrich Drepper.
2267 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2268 severe performance problems. Patch by David Mosberger.
2270 * The Hurd port got a lot more functionality like AIO, various stdio
2271 extensions, etc. Mainly done by Roland McGrath.
2273 * mtrace can now lookup symbols in shared libraries.
2277 * Intel's IA-64 math library is largely integrated. It provides fast and
2278 accurate implementatations for most basic and standard math functions
2279 in float, double, and long double format.
2281 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2282 and asin for the 96-bit long double format and asin, log, tan for the
2283 128-bit long double format.
2285 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2286 The basic double functions exist today. Contributed by Abraham Ziv
2287 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2288 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2290 * An asynchronous name lookup library was added. The interface is designed
2291 after POSIX AIO. The proposal was circulated beforehand to get comments.
2292 No negative ones came in. Implemented by Ulrich Drepper.
2294 * Port to S390/64bit contributed by Martin Schwidefsky
2295 <schwidefsky@de.ibm.com>.
2297 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2298 of functions for Linux/IA-64.
2300 * The RPC code is now thread safe. Threads can now use the same service
2301 of different services at the same time. Patch by Eric Norum
2302 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2304 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2305 family of functions for Linux/S390.
2307 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2308 of functions for Linux/x86.
2310 * Port to Linux/CRIS contributed by Axis Communications.
2314 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2315 we can now check for namespace violations and missing declarations. The
2316 result is that almost all headers are now Unix-compliant (as defined in
2317 the upcoming XPG6). The negative side is that some programs might need
2318 corrections, too, if they depend on the incorrect form of the headers in
2319 previous versions which defined too many symbols and included too many
2322 * regex now handles multibyte character sets correctly.
2323 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2325 * iconv (the program) does now conform to the upcoming XPG6 and handles
2326 charmaps. Instead of the charset names the path of charmaps can be
2327 provided and the conversion happens based on this data.
2328 Contributed by Ulrich Drepper.
2330 * The locale program now provides detailed information about the installed
2335 only lists the names of the supported locales
2339 provides details such as country, language, and codeset name.
2340 Contributed by Ulrich Drepper.
2344 * The gencat program now parses the input file according to the charset
2345 selected by the LC_CTYPE category. This is important for stateful
2346 character sets. To make generating catalogs easier there is a way
2347 to overwrite the charset selected by the locale: before the first
2348 message or $ quote line the catalog can contain a line like
2350 $ codeset=ISO-8859-2
2352 to select the charset (ISO-8859-2 in this case).
2354 Implemented by Shinya Hanataka and Ulrich Drepper.
2356 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2357 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2358 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2360 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2361 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2363 * The LANGUAGE environment variable is now ignored unless the locale is
2364 changed from the default "C" locale.
2366 * The usual bug fixes.
2370 * Greg McGary added runtime support for bounds checking using gcc's
2371 new -fbounded-pointers option. ix86 target is complete. PowerPC
2374 * Thorsten Kukuk added secure mode support to nscd.
2376 * The Berkeley db libraries have been removed.
2378 Related, the nss_db module is now in a separate package since it
2379 obviously requires a database library being available.
2381 * Wide character I/O streams implemented by Ulrich Drepper.
2383 * Functions from the extended socket API added by Ulrich Drepper.
2385 * Functions feenableexcept and fedisableexcept to control the
2386 behaviour of individual exceptions have been added by Andreas Jaeger.
2388 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2390 * The resolver code has been updated from bind 8.2.3-T5B which supports
2391 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2394 This change could in some situations effect backward compatibility. Since
2395 now `_res' is a thread-local instead of a global variable, modifying it
2396 in one thread does not have any effect in other threads.
2398 The resolver library was also extended to allow IPv6 as the transport
2399 protocol for the requests. This work was done by Stig Venaas.
2401 * Compatibility code for K&R C compilers has been removed from the
2402 header files. A ISO C compiler is needed to use the library
2403 (conforming to either C89 or C99 standard).
2405 * Complete rewrite of the localedef program to support multibyte character
2406 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2407 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2408 Implemented by Ulrich Drepper.
2410 Bruno Haible significantly improved the generation and use of the data
2411 structures for the wide character tables.
2413 * Plural handling in gettext implemented by Ulrich Drepper.
2415 * The utmp daemon has been removed.
2417 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2419 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2422 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2424 * POSIX barriers implemented by Kaz Kylheku.
2426 * POSIX spawn function family implemented by Ulrich Drepper.
2428 * POSIX spinlocks are now available.
2430 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2432 * the configure option --enable-kernel=X.Y.Z allows to strip out
2433 compatibility for kernel versions before X.Y.Z. This is currently only
2434 implemented for Linux.
2436 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2437 field sin6_scope_id. This means that all programs using IPv6 should be
2438 recompiled. Don't expect binary compatibility with previous glibc
2441 * various conversion modules for IBM character sets contributed by
2444 * IA-64 port by Jes Sorensen and HJ Lu.
2458 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2460 * New xdr functions are added; some rpc functions are now 64bit clean.
2462 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2464 * Fixed known incompatibilities with glibc 2.0.
2466 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2468 * Optimized a number of functions (especially the ELF dynamic loader).
2470 * Update timezone data files.
2472 * lots of charmaps corrections
2474 * some new locale definitions and charmaps
2479 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2480 conform to POSIX member type requirements. Symbol versions have been
2481 adjusted as needed within the library, and for direct use by applications,
2482 but there is potential for problems if third-party libraries use
2483 struct timeval as part of their interface. This does not present
2484 a problem for X and other "essential" system libraries.
2486 * An additional locale model to support C++ Standard Library locale
2487 model and probably more was implemented by Ulrich Drepper.
2489 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2492 * Miles Bader provided the `argp' function family to support hierachical
2493 command line argument parsing, layered on top of getopt.
2495 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2497 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2500 * scanf recognizes the %a and %A format for scanning floating point numbers.
2502 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2503 information and interfaces for the available integer types.
2505 * about 130 new math functions were added to implement the ISO C9x math
2508 * the new header <complex.h> contains definitions of the complex math
2509 functions from ISO C 9X.
2511 * the new header <tgmath.h> defines generic macros to use complex or
2512 real valued functions.
2514 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2516 * Andreas Jaeger provided a test suite for the math library.
2518 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2520 * Ulrich Drepper added character set conversion functions (iconv).
2522 * Optimized string functions have been added.
2524 * The localedata addon is now part of glibc.
2526 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2528 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2529 daemon for NSS (nscd).
2531 Missing a better place here are some numbers on improvements. Under
2532 Linux 2.1.125 un-tar-ing the kernel sources takes
2536 using local files 12.19s 6.88s 22.91s
2538 using NIS 13.92s 8.91s 26.34s
2540 using NIS & nscd 10.37s 7.34s 25.30s
2542 using NIS+ 27.57s 30.37s 640.46s
2544 using NIS+ & nscd 10.25s 7.83s 26.51s
2546 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2548 Keep in mind that non-namelookup related operations dominate above times.
2549 It was just a common complain that using NIS+ unpacking the kernel is
2552 [1] The old nscd implementation is not available anymore since it was
2553 distributed with glibc up to version 2.0.98 and thus is now replaced.
2555 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2557 * Mark Kettenis provided a Hesiod NSS module.
2559 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2560 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2562 * rcmd can now handle netgroups (Dick Streefland).
2564 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2567 * Support for the IPv6 protocol has been added to the socket API, as per the
2568 latest draft standards.
2570 * Support for Linux 2.2 has been added.
2572 * Interface changes relative to the latest 2.0.x release:
2573 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2574 addseverity NEW: Unix98
2575 alphasort64 NEW: LFS
2576 argp_err_exit_status NEW: argp, GNU ext
2577 argp_error NEW: argp, GNU ext
2578 argp_failure NEW: argp, GNU ext
2579 argp_help NEW: argp, GNU ext
2580 argp_parse NEW: argp, GNU ext
2581 argp_program_bug_address NEW: argp, GNU ext
2582 argp_program_version NEW: argp, GNU ext
2583 argp_program_version_hook NEW: argp, GNU ext
2584 argp_state_help NEW: argp, GNU ext
2585 argp_usage NEW: argp, GNU ext
2586 authdes_create NEW: Secure RPC
2587 authdes_getucred NEW: Secure RPC
2588 authdes_pk_create NEW: Secure RPC
2589 backtrace NEW: GNU ext.
2590 backtrace_symbols NEW: GNU ext.
2591 backtrace_symbols_fd NEW: GNU ext.
2593 cacosf NEW: ISO C 9x
2594 cacosh NEW: ISO C 9x
2595 cacoshf NEW: ISO C 9x
2596 cacoshl NEW: ISO C 9x
2597 cacosl NEW: ISO C 9x
2604 casinf NEW: ISO C 9x
2605 casinh NEW: ISO C 9x
2606 casinhf NEW: ISO C 9x
2607 casinhl NEW: ISO C 9x
2608 casinl NEW: ISO C 9x
2610 catanf NEW: ISO C 9x
2611 catanh NEW: ISO C 9x
2612 catanhf NEW: ISO C 9x
2613 catanhl NEW: ISO C 9x
2614 catanl NEW: ISO C 9x
2615 cbc_crypt NEW: Secure RPC
2619 ccoshf NEW: ISO C 9x
2620 ccoshl NEW: ISO C 9x
2626 cimagf NEW: ISO C 9x
2627 cimagl NEW: ISO C 9x
2628 clearerr_locked REMOVED
2629 clntunix_create NEW: sunrpc ext
2631 clog10 NEW: ISO C 9x
2632 clog10f NEW: ISO C 9x
2633 clog10l NEW: ISO C 9x
2643 cprojf NEW: ISO C 9x
2644 cprojl NEW: ISO C 9x
2646 crealf NEW: ISO C 9x
2647 creall NEW: ISO C 9x
2652 csinhf NEW: ISO C 9x
2653 csinhl NEW: ISO C 9x
2656 csqrtf NEW: ISO C 9x
2657 csqrtl NEW: ISO C 9x
2661 ctanhf NEW: ISO C 9x
2662 ctanhl NEW: ISO C 9x
2664 des_setparity NEW: Secure RPC
2665 ecb_crypt NEW: Secure RPC
2666 endutxent NEW: Unix98
2668 exp10f NEW: ISO C 9x
2669 exp10l NEW: ISO C 9x
2673 fattach NEW: STREAMS
2674 fdetach NEW: STREAMS
2678 feclearexcept NEW: ISO C 9x
2679 fegetenv NEW: ISO C 9x
2680 fegetexceptflag NEW: ISO C 9x
2681 fegetround NEW: ISO C 9x
2682 feholdexcept NEW: ISO C 9x
2684 feraiseexcept NEW: ISO C 9x
2685 ferror_locked REMOVED
2686 fesetenv NEW: ISO C 9x
2687 fesetexceptflag NEW: ISO C 9x
2688 fesetround NEW: ISO C 9x
2689 fetestexcept NEW: ISO C 9x
2690 feupdateenv NEW: ISO C 9x
2691 fflush_locked REMOVED
2695 fgets_unlocked NEW: GNU ext.
2696 fileno_locked REMOVED
2708 fputc_locked REMOVED
2709 fputs_unlocked NEW: GNU ext.
2710 fread_unlocked NEW: GNU ext.
2715 fstatvfs NEW: Unix98
2719 ftruncate64 NEW: LFS
2721 fwrite_unlocked NEW: GNU ext.
2722 gai_strerror NEW: IPv6
2726 getchar_locked REMOVED
2728 getdate_err NEW: Unix98
2729 getdate_r NEW: GNU ext.
2731 getnameinfo NEW: IPv6
2732 getnetname NEW: Secure RPC
2733 getpmsg NEW: STREAMS
2734 getpt NEW: Unix98 PTY
2735 getrlimit64 NEW: LFS
2736 getutxent NEW: Unix98
2737 getutxid NEW: Unix98
2738 getutxline NEW: Unix98
2739 glob64 NEW: GNU ext.
2740 globfree64 NEW: GNU ext.
2741 gnu_get_libc_release NEW: GNU ext.
2742 gnu_get_libc_version NEW: GNU ext.
2743 grantpt NEW: Unix98 PTY
2744 host2netname NEW: Secure RPC
2746 iconv_close NEW: iconv
2747 iconv_open NEW: iconv
2748 if_freenameindex NEW: IPv6
2749 if_indextoname NEW: IPv6
2750 if_nameindex NEW: IPv6
2751 if_nametoindex NEW: IPv6
2752 in6addr_any NEW: IPv6
2753 in6addr_loopback NEW: IPv6
2754 inet6_isipv4mapped NEW: IPv6
2755 isastream NEW: STREAMS
2756 iswblank NEW: GNU ext.
2757 key_decryptsession NEW: Secure RPC
2758 key_decryptsession_pk NEW: Secure RPC
2759 key_encryptsession NEW: Secure RPC
2760 key_encryptsession_pk NEW: Secure RPC
2761 key_gendes NEW: Secure RPC
2762 key_get_conv NEW: Secure RPC
2763 key_secretkey_is_set NEW: Secure RPC
2764 key_setnet NEW: Secure RPC
2765 key_setsecret NEW: Secure RPC
2766 llrint NEW: ISO C 9x
2767 llrintf NEW: ISO C 9x
2768 llrintl NEW: ISO C 9x
2769 llround NEW: ISO C 9x
2770 llroundf NEW: ISO C 9x
2771 llroundl NEW: ISO C 9x
2776 lrintf NEW: ISO C 9x
2777 lrintl NEW: ISO C 9x
2778 lround NEW: ISO C 9x
2779 lroundf NEW: ISO C 9x
2780 lroundl NEW: ISO C 9x
2782 makecontext NEW: Unix98
2783 mempcpy NEW: GNU ext.
2786 modify_ldt NEW: kernel
2790 nearbyint NEW: ISO C 9x
2791 nearbyintf NEW: ISO C 9x
2792 nearbyintl NEW: ISO C 9x
2793 netname2host NEW: Secure RPC
2794 netname2user NEW: Secure RPC
2795 nexttoward NEW: ISO C 9x
2796 nexttowardf NEW: ISO C 9x
2797 nexttowardl NEW: ISO C 9x
2801 passwd2des NEW: Secure RPC
2803 pow10f NEW: GNU ext.
2804 pow10l NEW: GNU ext.
2807 printf_size NEW: GNU ext.
2808 printf_size_info NEW: GNU ext.
2809 profil_counter REMOVED
2810 pthread_mutexattr_getkind_np REPLACED
2811 pthread_mutexattr_setkind_np REPLACED
2812 ptsname NEW: Unix98 PTY
2813 ptsname_r NEW: Unix98 PTY
2815 putchar_locked REMOVED
2816 putgrent NEW: GNU ext.
2818 putpmsg NEW: STREAMS
2819 pututxline NEW: Unix98
2823 readdir64_r NEW: LFS
2824 remquo NEW: ISO C 9x
2825 remquof NEW: ISO C 9x
2826 remquol NEW: ISO C 9x
2828 roundf NEW: ISO C 9x
2829 roundl NEW: ISO C 9x
2831 scalbln NEW: ISO C 9x
2832 scalblnf NEW: ISO C 9x
2833 scalblnl NEW: ISO C 9x
2835 sendfile NEW: kernel
2836 setcontext NEW: Unix98
2837 setrlimit64 NEW: LFS
2838 setutxent NEW: Unix98
2840 sigignore NEW: Unix98
2841 sigqueue NEW: POSIX.1b
2842 sigrelse NEW: Unix98
2843 sigset NEW: POSIX.1b
2844 sigtimedwait NEW: POSIX.1b
2845 sigwaitinfo NEW: POSIX.1b
2846 sincos NEW: GNU ext.
2847 sincosf NEW: GNU ext.
2848 sincosl NEW: GNU ext.
2852 strcasestr NEW: GNU ext.
2853 strtoimax NEW: ISO C 9x
2854 strtoumax NEW: ISO C 9x
2855 strverscmp NEW: GNU ext.
2856 svcauthdes_stats NEW: Secure RPC
2857 svcunix_create NEW: sunrpc etx
2858 svcunixfd_create NEW: sunrpc ext
2859 swapcontext NEW: Unix98
2860 tcgetsid NEW: Unix98 PTY
2861 tdestroy NEW: GNU ext.
2862 tgamma NEW: ISO C 9x
2863 tgammaf NEW: ISO C 9x
2864 tgammal NEW: ISO C 9x
2868 truncf NEW: ISO C 9x
2869 truncl NEW: ISO C 9x
2871 unlockpt NEW: Unix98 PTY
2872 updwtmpx NEW: Unix98
2873 user2netname NEW: Secure RPC
2874 utmpxname NEW: Unix98
2875 versionsort NEW: GNU ext.
2876 versionsort64 NEW: GNU ext.
2878 wcscasecmp NEW: GNU ext.
2879 wcsncasecmp NEW: GNU ext.
2880 wcsnlen NEW: GNU ext.
2881 wcstoimax NEW: ISO C 9x
2882 wcstoll NEW: ISO C 9x
2883 wcstoull NEW: ISO C 9x
2884 wcstoumax NEW: ISO C 9x
2886 wordexp NEW: POSIX.2
2887 wordfree NEW: POSIX.2
2888 write_profiling REMOVED
2889 xdecrypt NEW: Secure RPC
2890 xdr_authdes_cred NEW: Secure RPC
2891 xdr_authdes_verf NEW: Secure RPC
2892 xdr_cryptkeyarg NEW: Secure RPC
2893 xdr_cryptkeyarg2 NEW: Secure RPC
2894 xdr_cryptkeyres NEW: Secure RPC
2895 xdr_getcredres NEW: Secure RPC
2896 xdr_key_netstarg NEW: Secure RPC
2897 xdr_key_netstres NEW: Secure RPC
2898 xdr_keybuf NEW: Secure RPC
2899 xdr_keystatus NEW: Secure RPC
2900 xdr_netnamestr NEW: Secure RPC
2901 xdr_sizeof NEW: Secure RPC
2902 xdr_unixcred NEW: sunrpc ext
2903 xencrypt NEW: Secure RPC
2904 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2915 * inet_ntoa is thread-safe
2917 * updwtmp is moved from libutil to libc
2919 * rewrite of cbrt function
2921 * update of timezone data
2935 * add atoll function
2937 * fix complex problems in Berkeley DB code
2939 * fix math functions
2943 * fixed lots of header problems (especially Linux/GNU specific)
2945 * dynamic loader preserves all registers
2947 * Roland McGrath provided support for handling of auxiliary objects in
2948 the ELF dynamic loader.
2950 * support for parallel builds is improved
2954 * GNU extensions are no longer declared by default. To enable them you
2955 must define the macro `_GNU_SOURCE' in your program or compile with
2958 * The library has changed from using GNU ld symbol aliases to using weak
2959 symbols where available. The ELF object file format supports weak
2960 symbols; GNU ld also supports weak symbols in the a.out format. (There
2961 is also now support for other GNU ld extensions in ELF. Use the
2962 `--with-elf' option to configure to indicate you have ELF, and
2963 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
2964 of many files which contained only symbol aliases, reducing the size of
2965 the source and the compiled library; many other files were renamed to
2966 less cryptic names previously occupied by the symbol alias files.
2967 There is a new header file <elf.h> for programs which operate on
2968 files in the ELF format.
2970 * Converted to Autoconf version 2, so `configure' has more options.
2971 Run `configure --help' to see the details.
2973 * The library can now be configured to build profiling, highly-optimized
2974 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
2975 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
2976 `configure' enable building these extra libraries. The shared library is
2977 built by default when using both ELF and GNU ld. When shared libraries
2978 are enabled, the new library `-ldl' is available for arbitrary run-time
2979 loading of shared objects; its interface is defined in <dlfcn.h>. The
2980 new header file <link.h> gives access to the internals of the run-time
2981 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
2982 application of same name on other systems and it provides information
2983 about dynamically linked binaries.
2985 * The C library now provides the run-time support code for profiling
2986 executables compiled with `-pg'. Programs can control the profiling code
2987 through the interface in <sys/gmon.h>. The `gmon.out' files written by
2988 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
2989 the support for this file format was contributed by David Mosberger-Tang.
2991 * The math code has been replaced with a math library based on fdlibm from
2992 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
2993 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
2994 functions. The math functions now reside in a separate library, so
2995 programs using them will need to use `-lm' their linking commands.
2997 * John C. Bowman contributed optimized ix87 assembler inline functions.
2999 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3000 mechanism similar to that found in Solaris 2. This is now used for the
3001 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3002 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3003 file controls what services are used for each individual database. This
3004 works by loading shared libraries with names specified in `nsswitch.conf',
3005 so service modules can be changed or added at any time without even
3006 relinking any program. Currently there are the file, db, and NIS based
3007 NSS services available.
3009 * The new functions `strtoq' and `strtouq' parse integer values from
3010 strings, like `strtol' and `strtoul', but they return `long long int' and
3011 `unsigned long long int' values, respectively (64-bit quantities).
3013 * The new functions `strtof' and `strtold' parse floating-point values from
3014 strings, like `strtod', but they return `float' and `long double' values,
3015 respectively (on some machines `double' and `long double' are the same).
3017 * Ulrich Drepper has contributed new implementations of the floating-point
3018 printing and reading code used in the `printf' family of functions and
3019 `strtod', `strtof', and `strtold'. These new functions are perfectly
3020 accurate, and much faster than the old ones.
3022 * The implementation of the POSIX locale model was completely rewritten by
3023 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3024 to compile the POSIX locale definition.
3026 * The former dummy implementations of the strcoll and strxfrm function are
3027 now replaced by fully functional code contributed by Ulrich Drepper. The
3028 collation information comes from the POSIX locale definitions.
3030 * The new header <langinfo.h> defines an interface for accessing
3031 various locale-dependent data (using the locale chosen with `setlocale').
3033 * Ulrich Drepper has contributed a new suite of functions for operation on
3034 wide-character and multibyte-character strings, in <wchar.h>;
3035 and classification and case conversion of wide characters, in <wctype.h>.
3036 These new functions are conforming to the ISO C, Amendement 1 specification.
3038 * There is now a second implementation of the standard I/O library available.
3039 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3040 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3042 * You can now use positional parameter specifications in format strings
3043 for the `printf' and `scanf' families of functions. For example,
3044 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3045 ``Number 6, Mr Jones''. This is mainly useful when providing different
3046 format strings for different languages, whose grammars may dictate
3047 different orderings of the values being printed. To support this
3048 feature, the interface for `register_printf_handler' has changed; see
3049 the header file <printf.h> for details.
3051 * The `printf' and `scanf' families of functions now understand a new
3052 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3053 to group numbers as indicated by the locale; for `scanf' and friends, this
3054 says to accept as valid only a number with all the proper grouping
3055 separators in the right places. In the default "C" locale, numbers are
3056 not grouped; but locales for specific countries will define the usual
3057 conventions (i.e. separate thousands with `,' in the US locale).
3059 * The pgrp functions have been regularized, slightly incompatibly but much
3060 less confusingly. The core functions are now `getpgid' and `setpgid',
3061 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3062 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3063 provided for compatibility. There is no longer an incompatible `getpgrp'
3064 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3066 * The new header file <fts.h> and suite of functions simplify programs that
3067 operate on directory trees. This code comes from 4.4 BSD.
3069 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3070 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3071 NSS scheme used in glibc.
3073 * There is a new malloc debugging hook `__memalign_hook'.
3075 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3076 `unsigned int' in <sys/types.h>. These are for compatibility only and
3077 their use is discouraged.
3079 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3080 done differently, so that it works even without GNU ld.
3082 * New function `euidaccess' checks allowed access to a file like `access',
3083 but using the effective IDs instead of the real IDs.
3085 * The time zone data files have been updated for the latest and greatest
3086 local time conventions of the countries of the world.
3088 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3091 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3092 for formatting floating-point numbers. They are provided only for
3093 compatibility; new programs should use `sprintf' instead. There are
3094 also equivalent function for the `long double' floating-point type and
3095 all functions also exist in a reentrant form.
3097 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3098 functions for maintaining the login-record files (primarily of use to
3099 system programs such as `login'), and convenient functions for
3100 allocating and initializing a pseudo-terminal (pty) device.
3102 * Ulrich Drepper has contributed new support for System V style
3103 shared memory and IPC on systems that support it.
3105 * Ulrich Drepper has contributed several miscellaneous new functions found
3106 in System V: The `hsearch' family of functions provide an effective
3107 implementation of hash tables; `a64l' and `l64a' provide a very simple
3108 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3111 * Ulrich Drepper has contributed new reentrant counterparts for the
3112 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3114 * Ulrich Drepper has contributed new, highly-optimized versions of several
3115 string functions for the i486/Pentium family of processors.
3117 * Ulrich Drepper has updated the Linux-specific code, based largely
3118 on work done in Hongjiu Lu's version of GNU libc for Linux.
3119 The GNU library now supports Linux versions 2.0.10 and later,
3120 using the ELF object file format (i[3456]86-*-linux).
3122 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3124 * David Mosberger-Tang and Richard Henderson have ported the C library
3125 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3126 dynamic linking support for ELF/Alpha.
3128 * Richard Henderson contributed several Alpha optimized assembler function
3129 for arithmetic and string handling.
3131 * Ulrich Drepper has contributed a new set of message catalog functions to
3132 support multiple languages using the <libintl.h> interface, for use with
3133 his new package GNU gettext. Translation volunteers have contributed
3134 catalogs of the library's messages in Spanish, German, and Korean.
3136 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3137 program and the `catgets' function for reading the catalog files it
3138 creates. (The <libintl.h> interface is preferred; we include the
3139 <nl_types.h> interface using `catgets' only for source compatibility with
3140 programs already written to use it.)
3142 * New header file <values.h> gives SVID-compatible names for <limits.h>
3145 * Various new macros, declarations, and small header files for compatibility
3148 * New function `group_member' is a convenient way to check if a process has
3149 a given effective group ID.
3151 * When using GCC 2.7 and later, the socket functions are now declared in a
3152 special way so that passing an argument of type `struct sockaddr_in *',
3153 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3154 `struct sockaddr *' type, does not generate a type-clash warning.
3156 * New function `error' declared in header file <error.h> is a convenient
3157 function for printing error messages and optionally exiting; this is the
3158 canonical function used in GNU programs. The new functions `err', `warn',
3159 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3160 doing the same thing.
3162 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3163 POSIX.2 `glob' function to do ~ and {...} expansion.
3165 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3166 `clearenv' which is used in POSIX.9 is also available.
3168 * New function `getsid' returns session ID number on systems that support it.
3170 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3171 files <db.h> and <mpool.h> provide a rich set of functions for several
3172 types of simple databases stored in memory and in files, and <ndbm.h> is
3173 an old `ndbm'-compatible interface using the `db' functions. Link with
3174 `-ldb' to get these functions.
3176 * New macro `strdupa' copies a string like `strdup', but uses local stack
3177 space from `alloca' instead of dynamic heap space from `malloc'.
3179 * New function `strnlen' is like `strlen' but searches only a given maximum
3180 number of characters for the null terminator. `stpncpy', `strndup' and
3181 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3184 * New function `statfs' in header <sys/statfs.h>.
3186 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3187 provide convenient functions for operating on blocks of null-terminated
3190 * A new suite of functions in <utmp.h> handle all the details of reading
3191 and writing the utmp file.
3193 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3196 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3197 completely POSIX compliant and also implemented the extended functionality
3198 to handle alternate digit representation and alternate era date formats.
3200 * Ulrich Drepper provided an implementation of the `strptime' function
3201 defined in XPG4.2 which transforms a string into a `struct tm' value.
3203 * Paul Eggert provided the tzselect shell script as part of the timezone
3204 code. The shell script makes it easy to select the correct timezone
3207 * The implementation of the malloc family of functions is completely replaced
3208 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3209 The implementation uses the mmap function (if available) and it is
3210 optimized for the use in multi threaded programs.
3212 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3213 functions. This new functionality is usable by specifying a special
3214 salt string and it is compatible with implementation on *BSD systems.
3216 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3217 `getsubopt' to handle second level command line options, `bsd_signal'
3218 to access BSD style `signal' functionality, the obsolete `regexp' style
3221 * the `lchown' function is available on system which support this
3224 * The implementation of the shadow password handling function was contributed
3227 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3229 * POSIX.1g support was added. The <sys/select.h> header is available,
3230 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3231 implementation of `getaddrinfo'.
3235 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3237 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3238 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3240 * The new function `daemon' from 4.4 BSD is useful for server programs that
3241 want to put themselves in the background.
3243 * Joel Sherrill has contributed support for several standalone boards that
3244 run without an operating system.
3246 * `printf', `scanf' and friends now accept a `q' type modifier for long
3247 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3249 * All of the code taken from BSD (notably most of the math and networking
3250 routines) has been updated from the BSD 4.4-Lite release.
3252 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3254 * The new functions `getdomainname' and `setdomainname' fetch or change the
3255 YP/NIS domain name. These are system calls which exist on systems which
3258 * The time zone data files have been updated for the latest international
3261 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3262 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3266 * The C library now includes support for Sun RPC, from Sun's free
3267 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3268 are included. (There is still no support for YP.)
3270 * Tom Quinn has contributed a port of the C library to SGI machines running
3271 Irix 4 (mips-sgi-irix4).
3273 * The new `lockf' function is a simplified interface to the locking
3274 facilities of `fcntl', included for compatibility.
3276 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3278 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3280 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3283 * The error message printed by `assert' for a failed assertion now includes
3284 the name of the program (if using GNU ld) and the name of the calling
3285 function (with versions of GCC that support this).
3287 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3289 * The library now includes the <sys/mman.h> header file and memory
3290 management functions `mmap', `munmap', `mprotect', `msync', and
3291 `madvise', on systems that support those facilities.
3293 * The interface for `mcheck' has changed slightly: the function called to
3294 abort the program when an allocation inconsistency is detected now takes
3295 an argument that indicates the type of failure. The new function
3296 `mprobe' lets you request a consistency check for a particular block at
3297 any time (checks are normally done only when you call `free' or `realloc'
3300 * It is now possible to easily cross-compile the C library, building on one
3301 system a library to run on another machine and/or operating system. All
3302 you need to do is set the variable `HOST_CC' in `configparms' to the
3303 native compiler for programs to run on the machine you are building on (a
3304 few generator programs are used on Unix systems); set `CC' to the
3307 * The new function `fexecve' (only implemented on the GNU system) executes
3308 a program file given a file descriptor already open on the file.
3312 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3313 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3315 * You can set the variable `libprefix' in `configparms' to specify a prefix
3316 to be prepended to installed library files; this makes it easy to install
3317 the GNU C library to be linked as `-lgnuc' or whatever.
3319 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3320 copies a limited number of characters from a string, and returns the
3321 address of the last character written.
3323 * You no longer need to check for whether the installed `stddef.h' is
3324 compatible with the GNU C library. configure now checks for you.
3326 * You can now define a per-stream `fileno' function to convert the
3327 stream's cookie into an integral file descriptor.
3329 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3330 allocates zero bytes of storage, and returns a unique pointer which
3331 you can pass to `realloc' or `free'. The behavior is undefined if
3332 you dereference this pointer.
3334 * The C library now runs on Sony NEWS m68k machines running either
3335 NewsOS 3 or NewsOS 4.
3337 * The new `syscall' function is a system-dependent primitive function
3338 for invoking system calls. It has the canonical behavior on Unix
3339 systems, including unreliable return values for some calls (such as
3340 `pipe', `fork' and `getppid').
3342 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3343 to `EAGAIN', which is the preferred name. On systems whose kernels
3344 use two distinct codes, the C library now translates EWOULDBLOCK to
3345 EAGAIN in every system call function.
3349 * The GNU C Library Reference Manual is now distributed with the library.
3350 `make dvi' will produce a DVI file of the printed manual.
3351 `make info' will produce Info files that you can read on line using C-h i
3352 in Emacs or the `info' program.
3353 Please send comments on the manual to bug-glibc-manual@gnu.org.
3355 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3357 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3359 * Jason Merrill has contributed a port to the Sequent Symmetry running
3360 Dynix version 3 (i386-sequent-dynix).
3362 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3363 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3365 * New function `memory_warnings' lets you arrange to get warnings when
3366 malloc is running out of memory to allocate, like Emacs gives you.
3368 * The C library now contains the relocating allocator used in Emacs 19 for
3369 its editing buffers. This allocator (ralloc) minimizes allocation
3370 overhead and fragmentation by moving allocated regions around whenever it
3371 needs to. You always refer to a ralloc'd region with a "handle" (a
3372 pointer to a pointer--an object of type `void **').
3374 * There is a new `printf' format: `%m' gives you the string corresponding
3375 to the error code in `errno'.
3377 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3378 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3379 pass, the `a' modifier says to fill in a `char **' you pass with a
3382 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3383 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3384 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3386 * `mkstemp' is a traditional Unix function to atomically create and open a
3387 uniquely-named temporary file.
3391 * The standard location for the file that says what the local timezone is
3392 has changed again. It is now `/usr/local/etc/localtime' (or more
3393 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3395 * The distribution no longer contains any files with names longer than 14
3398 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3399 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3401 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3405 * The configuration process has changed quite a bit. The `configure'
3406 script is now used just like the configuration scripts for other GNU
3407 packages. The `sysdeps' directory hierarchy is much rearranged.
3408 The file `INSTALL' explains the new scheme in detail.
3410 * The header files no longer need to be processed into ANSI C and
3411 traditional C versions. There is just one set of files to install, and
3412 it will work with ANSI or old C compilers (including `gcc -traditional').
3414 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3415 MIPS DECStation running Ultrix 4.
3417 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3418 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3419 made itself into a shared library.
3421 * Yet further improved support for the i386, running 4.3 BSD-like systems
3422 (such as Mach 3 with the Unix single-server), or System V.
3424 * New function `strncasecmp' to do case-insensitive string comparison
3425 with limited length.
3427 * New function `strsep' is a reentrant alternative to `strtok'.
3429 * New functions `scandir' and `alphasort' for searching directories.
3431 * New function `setenv' is a better interface to `putenv'.
3433 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3434 function for traversing a directory tree.
3436 * The GNU obstack package is now also part of the C library.
3437 The new function `open_obstack_stream' creates a stdio stream that
3438 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3439 formatted output directly to an obstack.
3441 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3442 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3444 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3446 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3447 things to your strings.
3449 * There are some new test programs: test-fseek, testmb, and testrand.
3451 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3452 These ports are not finished, but are a good starting place for really
3453 supporting those systems.
3455 * `/etc/localtime' is now the standard location for the file that says what
3456 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3457 This follows the general principle that `/etc' is the place for all local
3458 configuration files.
3460 * The C library header files now use `extern "C"' when used by the C++
3461 compiler, so the C library should now work with C++ code.
3463 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3464 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3467 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3468 function which uses the `merge sort' algorithm, and is said to be
3469 significantly faster than the old GNU `qsort' function. Merge sort is
3470 now the standard `qsort' function. The new algorithm can require a lot
3471 of temporary storage; so, the old sorting function is called when the
3472 required storage is not available.
3474 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3475 provides the Unix `crypt' function, plus some other entry points.
3477 * The code and header files taken from 4.4 BSD have been updated with the
3478 latest files released from Berkeley.
3480 ----------------------------------------------------------------------
3481 Copyright information:
3483 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3485 Permission is granted to anyone to make or distribute verbatim copies
3486 of this document as received, in any medium, provided that the
3487 copyright notice and this permission notice are preserved,
3488 thus giving the recipient permission to redistribute in turn.
3490 Permission is granted to distribute modified versions
3491 of this document, or of portions of it,
3492 under the above conditions, provided also that they
3493 carry prominent notices stating who last changed them.
3496 version-control: never