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 nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
26 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
27 Since the GNU C Library defaults to a much newer revision of POSIX, this
28 will only affect programs that specifically request an old conformance
29 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
30 see a change in the visible declarations, but a program compiled with
31 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
34 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
35 defined by all multithreaded code, but glibc has not required this for
38 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
39 means that in a future release, the macros “major”, “minor”, and “makedev”
40 will only be available from <sys/sysmacros.h>.
42 These macros are not part of POSIX nor XSI, and their names frequently
43 collide with user code; see for instance glibc bug 19239 and Red Hat bug
44 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
45 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
48 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
49 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
50 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
52 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
53 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
54 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
55 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
56 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
57 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
58 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
59 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
60 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
61 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
62 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
65 * New <math.h> features are added from TS 18661-1:2014:
67 - Signaling NaN macros: SNANF, SNAN, SNANL.
69 - Nearest integer functions: roundeven, roundevenf, roundevenl.
71 - llogb functions: the llogb, llogbf and llogbl functions, and the
72 FP_LLOGB0 and FP_LLOGBNAN macros.
74 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
77 - Comparison macros: iseqsig.
79 - Classification macros: iscanonical, issubnormal, iszero.
81 - Total order functions: totalorder, totalorderf, totalorderl,
82 totalordermag, totalordermagf, totalordermagl.
84 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
86 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
87 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
89 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
90 are added to libc. They convert a floating-point number into string.
92 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
93 intended to be used instead of memset() to erase sensitive data after use;
94 the compiler will not optimize out calls to explicit_bzero even if they
95 are "unnecessary" (in the sense that no _correct_ program can observe the
96 effects of the memory clear).
98 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
99 to float instead of double. This does not affect the ABI of any libraries
100 that are part of the GNU C Library, but may affect the ABI of other
101 libraries that use this type in their interfaces.
103 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
104 float_t and double_t types are now defined to long double instead of float
105 and double. These options are not the default, and this does not affect
106 the ABI of any libraries that are part of the GNU C Library, but it may
107 affect the ABI of other libraries that use this type in their interfaces,
108 if they are compiled or used with those options.
110 * The getentropy and getrandom functions, and the <sys/random.h> header file
113 * The buffer size for byte-oriented stdio streams is now limited to 8192
114 bytes by default. Previously, on Linux, the default buffer size on most
115 file systems was 4096 bytes (and thus remains unchanged), except on
116 network file systems, where the buffer size was unpredictable and could be
117 as large as several megabytes.
119 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
120 for the Linux quota interface which predates kernel version 2.4.22 has
123 * The malloc_get_state and malloc_set_state functions have been removed.
124 Already-existing binaries that dynamically link to these functions will
125 get a hidden implementation in which malloc_get_state is a stub. As far
126 as we know, these functions are used only by GNU Emacs and this change
127 will not adversely affect already-built Emacs executables. Any undumped
128 Emacs executables, which normally exist only during an Emacs build, should
129 be rebuilt by re-running “./configure; make” in the Emacs build tree.
131 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
132 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
133 “no-ip6-dotint” had already been the default, and support for the
134 “ip6-dotint” option was removed from the Internet in 2006.
136 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
137 flag from <resolv.h> have been removed. The option relied on a
138 backwards-incompatible DNS extension which was never deployed on the
141 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
142 in the <resolv.h> header file have been deprecated. They were already
145 * DNSSEC-related declarations and definitions have been removed from the
146 <arpa/nameser.h> header file, and libresolv will no longer attempt to
147 decode the data part of DNSSEC record types. Previous versions of glibc
148 only implemented minimal support for the previous version of DNSSEC, which
149 is incompatible with the currently deployed version.
151 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
152 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
153 <arpa/nameser.h> header file because the distinction between RR types and
154 meta-RR types is not officially standardized, subject to revision, and
155 thus not suitable for encoding in a macro.
157 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
158 and rhook members of the res_state type in <resolv.h> have been removed.
159 The glibc stub resolver did not support these hooks, but the header file
160 did not reflect that.
162 * For multi-arch support it is recommended to use a GCC which has
163 been built with support for GNU indirect functions. This ensures
164 that correct debugging information is generated for functions
165 selected by IFUNC resolvers. This support can either be enabled by
166 configuring GCC with '--enable-gnu-indirect-function', or by
167 enabling it by default by setting 'default_gnu_indirect_function'
168 variable for a particular architecture in the GCC source file
171 * GDB pretty printers have been added for mutex and condition variable
172 structures in POSIX Threads. When installed and loaded in gdb these pretty
173 printers show various pthread variables in human-readable form when read
174 using the 'print' or 'display' commands in gdb.
176 Security related changes:
178 On ARM EABI (32-bit), generating a backtrace for execution contexts which
179 have been created with makecontext could fail to terminate due to a
180 missing .cantunwind annotation. This has been observed to lead to a hang
181 (denial of service) in some Go applications compiled with gccgo. Reported
182 by Andreas Schwab. (CVE-2016-6323)
184 The following bugs are resolved with this release:
186 [The release manager will add the list generated by
187 scripts/list-fixed-bugs.py just before the release.]
191 * The minimum Linux kernel version that this version of the GNU C Library
192 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
193 kernel version 2.6.32 or later suffices (on architectures that already
194 required kernel versions more recent than 3.2, those requirements remain
195 unchanged). Linux 3.2 or later kernel headers are required on all
198 * The pap_AN locale has been deleted. This has been deprecated for a long
199 time. It has been replaced by pap_AW & pap_CW, both of which have long
200 been included in previous releases.
202 * The readdir_r and readdir64_r functions have been deprecated. It is
203 recommended to use readdir and readdir64 instead.
205 * The type “union wait” has been removed. It was deprecated in the early
206 1990s and never part of POSIX. Application code should use the int type
207 instead of “union wait”.
209 * A new NSS action is added to facilitate large distributed system
210 administration. The action, MERGE, allows remote user stores like LDAP
211 to be merged into local user stores like /etc/groups in order to provide
212 easy to use, updated, and managed sets of merged credentials. The new
213 action can be used by configuring it in /etc/nsswitch.conf:
214 group: files [SUCCESS=merge] nis
215 Implemented by Stephen Gallagher (Red Hat).
217 * The deprecated __malloc_initialize_hook variable has been removed from the
220 * The long unused localedef --old-style option has been removed. It hasn't
221 done anything in over 16 years. Scripts using this option can safely
224 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
225 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
226 return the next representable value in the direction of positive infinity
227 and the nextdown functions return the next representable value in the
228 direction of negative infinity. These are currently enabled as GNU
231 Security related changes:
233 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
234 could result in a stack overflow when getnetbyname was called with an
235 overly long name. (CVE-2016-3075)
237 * Previously, getaddrinfo copied large amounts of address data to the stack,
238 even after the fix for CVE-2013-4458 has been applied, potentially
239 resulting in a stack overflow. getaddrinfo now uses a heap allocation
240 instead. Reported by Michael Petlan. (CVE-2016-3706)
242 * The glob function suffered from a stack-based buffer overflow when it was
243 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
244 Reported by Alexander Cherepanov. (CVE-2016-1234)
246 * The Sun RPC UDP client could exhaust all available stack space when
247 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
248 alloca plugin for GCC. (CVE-2016-4429)
250 * The IPv6 name server management code in libresolv could result in a memory
251 leak for each thread which is created, performs a failing naming lookup,
252 and exits. Over time, this could result in a denial of service due to
253 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
255 The following bugs are resolved with this release:
257 [1170] localedata: ne_NP: update Nepali locale definition file
258 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
260 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
261 [6796] math: fdim() does not set errno on overflow
262 [10354] libc: posix_spawn should use vfork() in more cases than presently
263 [11213] localedata: localedata: add copyright disclaimer to locale files
264 [12143] localedata: chr_US: new Cherokee locale
265 [12450] localedata: sgs_LT: new locale
266 [12676] localedata: ln_CD: new locale
267 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
269 [13304] math: fma, fmaf, fmal produce wrong results
270 [14259] build: --localedir arg to configure is ignored
271 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
272 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
273 [14934] localedata: es_CL: wrong first weekday chilean locale
274 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
276 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
278 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
279 [15368] nptl: raise() is not async-signal-safe
280 [15479] math: ceil, floor, round and trunc raise inexact exception
281 [15578] localedata: kk_KZ: various updates
282 [16003] localedata: pap_AN: punt old locale
283 [16137] localedata: iw_IL: punt old locale
284 [16190] localedata: eo: new esperanto locale
285 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
286 [16742] malloc: race condition: pthread_atfork() called before first
287 malloc() results in unexpected locking behaviour/deadlocks
288 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
290 [16983] localedata: postal_fmt does not allow %l and %n modifiers
291 [17565] localedata: pt_PT: wrong (work-)week start
292 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
293 [17950] build: Build fails with -msse
294 [18205] localedata: be_BY*: wrong first_weekday and first_workday
295 [18433] libc: posix_spawn does not return correctly upon failure to
297 [18453] localedata: charmaps/IBM875: incorrect codes
298 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
300 [18896] localedata: he_IL: improvements for currency
301 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
303 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
305 [19056] libc: Deprecate readdir_r
306 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
308 [19198] localedata: nl_NL: small improvements for Dutch locales
309 [19257] network: Per-thread memory leak in __res_vinit with IPv6
310 nameservers (CVE-2016-5417)
311 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
313 [19400] locale: Language missing in "iso-639.def", trivial fix in
315 [19431] malloc: Deadlock between fflush, getdelim, and fork
316 [19505] libc: Incorrect file descriptor validity checks in
317 posix_spawn_file_actions_add{open,close,dup2}
318 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
320 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
321 `intl/{gettextP,loadinfo}.h'
322 [19534] libc: execle, execlp may use malloc
323 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
324 [19573] network: res_nclose and __res_maybe_init disagree about name
325 server initialization, breaking Hesiod
326 [19575] localedata: Status of GB18030 tables
327 [19581] localedata: sr_* date_fmt string contains additional newline
328 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
330 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
331 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
332 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
333 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
335 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
337 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
339 [19610] dynamic-link: ldconfig -X removes stale symbolic links
340 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
341 [19633] locale: strfmon_l applies global locale to number formatting
342 [19642] network: Memory leak in getnameinfo
343 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
344 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
346 [19654] math: [x86_64] Need testcase for BZ #19590 fix
347 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
349 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
350 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
352 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
354 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
356 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
357 pointers and lengths in error-case.
358 [19727] locale: Converting from/to UTF-xx with iconv() does not always
359 report errors on UTF-16 surrogates values.
360 [19755] nscd: nscd assertion failure in gc
361 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
362 [19759] libc: mempcpy shouldn't be inlined
363 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
364 [19765] libc: s390 needs an optimized mempcpy
365 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
366 NAME_MAX limit assumption (CVE-2016-1234)
367 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
368 [19787] network: Missing and incorrect truncation checks in getnameinfo
369 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
371 [19791] network: Assertion failure in res_query.c with un-connectable name
373 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
374 [19822] math: libm.so install clobbers old version
375 [19825] network: resolv: send_vc can return uninitialized data in second
376 response to getaddrinfo
377 [19830] network: nss_dns: should check RDATA length against buffer length
378 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
379 confronted with A/AAAA records of invalid size
380 [19837] nss: nss_db: No retries for some long lines with a larger buffer
381 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
382 [19853] stdio: Printing IBM long double in decimal with high precision is
384 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
385 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
386 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
388 [19865] network: Assertion failure or memory leak in
389 _nss_dns_getcanonname_r
390 [19868] network: nss_dns: netent code does not skip over non-PTR records
391 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
393 [19881] string: Improve x86-64 memset
394 [19907] string: Incorrect memcpy tests
395 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
397 [19925] libc: termios.h XCASE namespace
398 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
399 [19929] libc: limits.h NL_NMAX namespace
400 [19931] stdio: Memory leak in vfprintf
401 [19957] libc: clone(CLONE_VM) access invalid parent memory
402 [19963] localedata: en_IL: New locale
403 [19989] stdio: stdio.h cuserid namespace
404 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
406 [19996] locale: langinfo.h nl_langinfo_l namespace
407 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
408 [20010] network: getaddrinfo: Stack overflow in hostent translation
410 [20012] stdio: libio: fmemopen append mode failure
411 [20014] stdio: stdio.h namespace for pre-threads POSIX
412 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
413 [20023] libc: fcntl.h timespec namespace
414 [20024] math: [x86_64] vectorized sincos trashes the stack
415 [20031] network: nss_hesiod: Heap overflow in get_txt_records
416 [20041] time: sys/time.h timespec namespace
417 [20043] libc: unistd.h missing cuserid for UNIX98 and before
418 [20044] libc: unistd.h missing pthread_atfork for UNIX98
419 [20051] libc: ttyslot in wrong header under wrong conditions
420 [20054] libc: gethostname not declared for XPG4
421 [20055] libc: termios.h missing tcgetsid for XPG4
422 [20072] dynamic-link: x86 init_cpu_features is called twice in static
424 [20073] libc: sys/stat.h fchmod namespace
425 [20074] libc: stdlib.h rand_r namespace
426 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
427 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
429 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
430 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
432 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
433 [20119] libc: Wrong mask for processors level type from CPUID
434 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
435 AS not supporting AVX512
436 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
438 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
439 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
441 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
442 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
443 [20178] libc: posix_spawn{p} should not call exit
444 [20191] stdio: libio: vtables hardening
445 [20195] string: FMA4 detection requires CPUID execution with register
447 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
448 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
450 [20212] math: acos (sNaN) returns sNaN
451 [20213] math: asin (sNaN) returns sNaN
452 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
453 [20218] math: [i386] asinhl (sNaN) returns sNaN
454 [20219] math: [i386] atanhl (sNaN) returns sNaN
455 [20222] stdio: fopencookie: Mangle function pointers
456 [20224] math: [i386] cbrtl (sNaN) returns sNaN
457 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
458 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
460 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
461 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
462 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
463 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
464 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
465 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
466 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
467 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
468 [20240] math: modf (sNaN) returns sNaN
469 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
470 [20250] math: frexp (sNaN) returns sNaN
471 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
472 [20255] math: [i386] fdim, fdimf return with excess range and precision /
474 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
475 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
476 subscript is below array bounds [-Werror=array-bounds]
477 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
478 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
480 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
481 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
483 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
485 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
486 [20316] localedata: id_ID: Februari instead of Pebruari
487 [20327] string: POWER8 strcasecmp returns incorrect result
488 [20347] math: Failure: Test: j0_downward (0xap+0)
489 [20348] libc: FAIL: misc/tst-preadvwritev64
490 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
491 [20350] libc: There is no test for p{read,write}64
492 [20357] math: Incorrect cos result for 1.5174239687223976
493 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
497 * Unicode 8.0.0 Support: Character encoding, character type info, and
498 transliteration tables are all updated to Unicode 8.0.0, using new
499 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
500 These updates cause user visible changes, such as the fixes for bugs
501 89, 16061, and 18568.
503 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
504 kernel-internal CPU set size. This means that requests that change the
505 CPU affinity which failed before (for example, an all-ones CPU mask) will
506 now succeed. Applications that need to determine the effective CPU
507 affinities need to call sched_getaffinity or pthread_getaffinity_np after
508 setting it because the kernel can adjust it (and the previous size check
509 would not detect this in the majority of cases).
511 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
512 the following new symbols are used: fts64_children, fts64_close,
513 fts64_open, fts64_read and fts64_set.
515 * getaddrinfo now detects certain invalid responses on an internal netlink
516 socket. If such responses are received, an affected process will
517 terminate with an error message of "Unexpected error <number> on netlink
518 descriptor <number>" or "Unexpected netlink response of size <number> on
519 descriptor <number>". The most likely cause for these errors is a
520 multi-threaded application which erroneously closes and reuses the netlink
521 file descriptor while it is used by getaddrinfo.
523 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
524 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
525 unnecessary serialization of memory allocation requests across threads.
526 The defect is now corrected. Users should see a substantial increase in
527 the concurent throughput of allocation requests for applications which
528 trigger this bug. Affected applications typically create and
529 destroy threads frequently. (Bug 19048 was reported and analyzed by
532 * There is now a --disable-timezone-tools configure option for disabling the
533 building and installing of the timezone related utilities (zic, zdump, and
534 tzselect). This is useful for people who build the timezone data and code
535 independent of the GNU C Library.
537 * The obsolete header <regexp.h> has been removed. Programs that require
538 this header must be updated to use <regex.h> instead.
540 * The obsolete functions bdflush, create_module, get_kernel_syms,
541 query_module and uselib are no longer available to newly linked binaries;
542 the header <sys/kdaemon.h> has been removed. These functions and header
543 were specific to systems using the Linux kernel and could not usefully be
544 used with the GNU C Library on systems with version 2.6 or later of the
547 * Optimized string, wcsmbs and memory functions for IBM z13.
548 Implemented by Stefan Liebler.
550 * Newly linked programs that define a variable called signgam will no longer
551 have it set by the lgamma, lgammaf and lgammal functions. Programs that
552 require signgam to be set by those functions must ensure that they use the
553 variable provided by the GNU C Library and declared in <math.h>, without
554 defining their own copy.
556 * The minimum GCC version that can be used to build this version of the GNU
557 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
558 still be used to compile programs using the GNU C Library.
560 Security related changes:
562 * An out-of-bounds value in a broken-out struct tm argument to strftime no
563 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
565 * The LD_POINTER_GUARD environment variable can no longer be used to disable
566 the pointer guard feature. It is always enabled. Previously,
567 LD_POINTER_GUARD could be used to disable security hardening in binaries
568 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
571 * An integer overflow in hcreate and hcreate_r could lead to an
572 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
574 * The catopen function no longer has unbounded stack usage. Reported by
577 * The nan, nanf and nanl functions no longer have unbounded stack usage
578 depending on the length of the string passed as an argument to the
579 functions. Reported by Joseph Myers. (CVE-2014-9761)
581 * A stack-based buffer overflow was found in libresolv when invoked from
582 libnss_dns, allowing specially crafted DNS responses to seize control
583 of execution flow in the DNS client. The buffer overflow occurs in
584 the functions send_dg (send datagram) and send_vc (send TCP) for the
585 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
586 family. The use of AF_UNSPEC triggers the low-level resolver code to
587 send out two parallel queries for A and AAAA. A mismanagement of the
588 buffers used for those queries could result in the response of a query
589 writing beyond the alloca allocated buffer created by
590 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
591 the overflow. Thanks to the Google Security Team and Red Hat for
592 reporting the security impact of this issue, and Robert Holiday of
593 Ciena for reporting the related bug 18665. (CVE-2015-7547)
595 The following bugs are resolved with this release:
597 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
598 [887] math: Math library function "logb" and "nextafter" inconsistent
599 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
600 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
601 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
602 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
604 [4404] localedata: German translation of "Alarm clock" is misleading
605 [6799] math: nextafter() and nexttoward() doen't set errno on
606 overflow/underflow errors
607 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
609 [10432] nis: _nss_nis_setnetgrent assertion failure
610 [11460] libc: fts has no LFS support
611 [12926] network: getaddrinfo()/make_request() may spin forever
612 [13065] nptl: Race condition in pthread barriers
613 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
614 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
616 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
617 [14912] libc: Rename non-installed bits/*.h headers
618 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
619 [15367] math: Let gcc use __builtin_isinf
620 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
621 [15421] math: lgamma wrongly sets signgam for ISO C
622 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
623 FE_INVALID with argument out of range
624 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
626 [15786] dynamic-link: ifunc resolver functions can smash function
628 [15918] math: Unnecessary check for equality in hypotf()
629 [16061] localedata: Review / update transliteration data
630 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
632 [16141] time: strptime %z offset restriction
633 [16171] math: drem should be alias of remainder
634 [16296] math: fegetround is pure?
635 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
636 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
638 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
639 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
641 [16422] math: [powerpc] math-float, math-double failing llrint tests with
642 "Exception "Inexact" set" on ppc32
643 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
644 [16517] math: Missing underflow exception from tanf/tan/tanl
645 [16519] math: Missing underflow exception from sinhf
646 [16520] math: Missing underflow exception from tanhf
647 [16521] math: Missing underflow exception from exp2
648 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
650 [16734] stdio: fopen calls mmap to allocate its buffer
651 [16961] math: nan function incorrect handling of bad sequences
652 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
653 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
654 [16985] locale: localedef: confusing error message when opening output
656 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
657 [17197] locale: Redundant shift character in iconv conversion output at
659 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
660 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
661 [17250] dynamic-link: static linking breaks nss loading
662 (getaddrinfo/getpwnam/etc...)
663 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
665 [17441] math: isnan() should use __builtin_isnan() in GCC
666 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
667 (related to lock elision)
668 [17787] manual: Exponent on page 324 of the PDF ends prematurely
669 [17886] time: strptime should be able to parse "Z" as a timezone with %z
670 [17887] time: strptime should be able to parse "+01:00" style timezones
671 [17905] libc: catopen() Multiple unbounded stack allocations
673 [18084] libc: backtrace (..., 0) dumps core on x86
674 [18086] libc: nice() sets errno to 0 on success
675 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
676 is too large (CVE-2015-8778)
677 [18251] dynamic-link: SONAME missing when audit modules provides path
678 [18265] libc: add attributes for wchar string and memory functions
679 [18370] math: csqrt missing underflows
680 [18421] libc: [hppa] read-only segment has dynamic relocations
681 [18472] libc: Obsolete syscall wrappers should be compat symbols
682 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
683 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
684 [18525] localedata: Remove locale timezone information
685 [18560] libc: [powerpc] spurious bits/ipc.h definitions
686 [18568] localedata: Update locale data to Unicode 8.0
687 [18589] locale: sort-test.sh fails at random
688 [18595] math: ctan, ctanh missing underflows
689 [18604] libc: assert macro-expands its argument
690 [18610] math: S390: fetestexcept() reports any exception if DXC-code
691 contains a vector instruction exception.
692 [18611] math: j1, jn missing errno setting on underflow
693 [18618] localedata: sync Chechen locale definitions with other *_RU
695 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
696 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
697 [18665] network: In send_dg, the recvfrom function is NOT always using the
698 buffer size of a newly created buffer (CVE-2015-7547)
699 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
701 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
703 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
704 [18699] math: tilegx cproj() for various complex infinities does not yield
706 [18724] libc: Harden put*ent functions against data injection
707 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
708 [18755] build: build errors with -DNDEBUG
709 [18757] stdio: fmemopen fails to set errno on failure
710 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
712 [18781] libc: openat64 lacks O_LARGEFILE
713 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
714 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
715 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
716 [18790] math: [ldbl-128ibm] tanhl inaccurate
717 [18795] libc: stpncpy fortification misses buffer lengths that are
719 [18796] build: build fails for --disable-mathvec
720 [18803] math: hypot missing underflows
721 [18820] stdio: fmemopen may leak memory on failure
722 [18823] math: csqrt spurious underflows
723 [18824] math: fma spurious underflows
724 [18825] math: pow missing underflows
725 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
726 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
727 [18870] build: sem_open.c fails to compile with missing symbol
729 [18872] stdio: Fix memory leak in printf_positional
730 [18873] libc: posix_fallocate overflow check ineffective
731 [18875] math: Excess precision leads incorrect libm
732 [18877] libc: arm: mmap offset regression
733 [18887] libc: memory corruption when using getmntent on blank lines
734 [18918] localedata: hu_HU: change time to HH:MM:SS format
735 [18921] libc: Regression: extraneous stat() and fstat() performed by
737 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
738 binaries (CVE-2015-8777)
739 [18951] math: tgamma missing underflows
740 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
742 [18953] localedata: lt_LT: change currency symbol to the euro
743 [18956] math: powf inaccuracy
744 [18961] math: [i386] exp missing underflows
745 [18966] math: [i386] exp10 missing underflows
746 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
747 [18969] build: multiple string test failures due to missing locale
749 [18970] libc: Reference of pthread_setcancelstate in libc.a
750 [18977] math: float / long double Bessel functions not in XSI POSIX
751 [18980] math: i386 libm functions return with excess range and precision
752 [18981] math: i386 scalb*, ldexp return with excess range and precision
753 [18982] stdio: va_list and vprintf
754 [18985] time: Passing out of range data to strftime() causes a segfault
756 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
757 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
758 [19012] locale: iconv_open leaks memory on error path
759 [19016] math: clog, clog10 inaccuracy
760 [19018] nptl: Mangle function pointers in tls_dtor_list
761 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
762 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
763 [19048] malloc: malloc: arena free list can become cyclic, increasing
765 [19049] math: [powerpc] erfc incorrect zero sign
766 [19050] math: [powerpc] log* incorrect zero sign
767 [19058] math: [x86_64] Link fail with -fopenmp and -flto
768 [19059] math: nexttoward overflow incorrect in non-default rounding modes
769 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
770 [19074] network: Data race in _res_hconf_reorder_addrs
771 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
772 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
773 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
775 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
777 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
778 [19086] manual: posix_fallocate64 documented argument order is wrong.
779 [19088] math: lround, llround missing exceptions close to overflow
781 [19094] math: lrint, llrint missing exceptions close to overflow threshold
782 [19095] math: dbl-64 lrint incorrect for 64-bit long
783 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
784 [19124] dynamic-link: ld.so failed to build with older assmebler
785 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
786 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
787 [19134] math: [powerpc32] lround, lroundf spurious exceptions
788 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
789 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
790 pthread_setaffinity_np
791 [19156] math: [ldbl-128] j0l spurious underflows
792 [19164] nptl: tst-getcpu fails with many possible CPUs
793 [19168] math: math/test-ildoubl and math/test-ldouble failure
794 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
795 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
797 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
798 (FE_NOMASK_ENV) do not clear SSE exceptions
799 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
800 _int_new_arena/reused_arena
801 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
802 [19201] math: dbl-64 remainder incorrect sign of zero result
803 [19205] math: bits/math-finite.h conditions do not match math.h and
805 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
806 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
807 for C99-based standards
808 [19212] libc: features.h not -Wundef clean
809 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
811 [19214] libc: Family and model identification for AMD CPU's are incorrect.
812 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
813 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
815 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
817 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
819 [19242] libc: strtol incorrect in Turkish locales
820 [19243] malloc: reused_arena can pick an arena on the free list, leading
821 to an assertion failure and reference count corruption
822 [19253] time: tzset() ineffective when temporary TZ did not include DST
824 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
825 [19270] math: [hppa] Shared libm missing __isnanl
826 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
827 MAP_HUGETLB and MAP_STACK defines
828 [19313] nptl: Wrong __cpu_mask for x32
829 [19347] libc: grantpt: try to force a specific gid even without pt_chown
830 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
831 [19350] math: [ldbl-128ibm] sinhl spurious overflows
832 [19351] math: [ldbl-128ibm] logl inaccurate near 1
833 [19363] time: x32: times() return value wrongly truncates/sign extends
835 [19367] dynamic-link: Improve branch prediction on Silvermont
836 [19369] network: Default domain name not reset by res_ninit when "search"
837 / "domain" entry is removed from resolv.conf
838 [19375] math: powerpc: incorrect results for POWER7 logb with negative
840 [19385] localedata: bg_BG: time separator should be colon, not comma
841 [19408] libc: linux personality syscall wrapper may erroneously return an
842 error on 32-bit architectures
843 [19415] libc: dladdr returns wrong names on hppa
844 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
845 IBM905, IBM907, and IBM909
846 [19439] math: Unix98 isinf and isnan functions conflict with C++11
847 [19443] build: build failures with -DDEBUG
848 [19451] build: Make check fails on test-double-vlen2
849 [19462] libc: Glibc failed to build with -Os
850 [19465] math: Wrong code with -Os
851 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
853 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
855 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
856 [19486] math: S390: Math tests fail with "Exception Inexact set".
857 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
858 [19550] libc: [mips] mmap negative offset handling inconsistent with other
860 [19590] math: Fail to build shared objects that use libmvec.so functions.
864 * The following bugs are resolved with this release:
866 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
867 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
868 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
869 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
870 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
871 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
872 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
873 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
874 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
875 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
876 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
877 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
878 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
879 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
880 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
881 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
882 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
883 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
884 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
885 18657, 18676, 18694, 18696, 18887.
887 * Cache information can be queried via sysconf() function on s390 e.g. with
888 _SC_LEVEL1_ICACHE_SIZE as argument.
890 * A buffer overflow in gethostbyname_r and related functions performing DNS
891 requests has been fixed. If the NSS functions were called with a
892 misaligned buffer, the buffer length change due to pointer alignment was
893 not taken into account. This could result in application crashes or,
894 potentially arbitrary code execution, using crafted, but syntactically
895 valid DNS responses. (CVE-2015-1781)
897 * The time zone file parser has been made more robust against crafted time
898 zone files, avoiding heap buffer overflows related to the processing of
899 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
900 large time zone data files. Overly long time zone specifiers in the TZ
901 variable no longer result in stack overflows and crashes.
903 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
904 for LD and GD on x86 and x86-64, has been implemented. You will need
905 binutils-2.24 or later to enable this optimization.
907 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
908 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
909 Hat). These updates cause user visible changes, such as the fix for bug
912 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
913 and getXXbyYY NSS calls for the same database, causing a denial-of-service
914 condition in some applications.
916 * Added vector math library named libmvec with the following vectorized x86_64
917 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
919 The library can be disabled with --disable-mathvec. Use of the functions is
920 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
921 Shared library libmvec.so is linked in as needed when using -lm (no need to
922 specify -lmvec explicitly for not static builds).
923 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
925 * A new fmemopen implementation has been added with the goal of POSIX
926 compliance. The new implementation fixes the following long-standing
927 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
928 old implementation is still present for use be by existing binaries.
930 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
931 releases. It has been fixed to match 2.19 and older, but binaries built
932 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
934 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
935 Contributed by Roland McGrath (Google).
937 * The header <regexp.h> is deprecated, and will be removed in a future
938 release. Use of this header will trigger a deprecation warning.
939 Application developers should update their code to use <regex.h> instead.
941 This header was formerly part of SUS, but was deprecated in 1994 and
942 removed from the standard in 2001. Also, the glibc implementation
943 leaks memory. See BZ#18681 for more details.
947 * The following bugs are resolved with this release:
949 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
950 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
951 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
952 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
953 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
954 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
955 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
956 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
957 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
958 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
961 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
962 too little memory for the to-be-scanned arguments and overflow the
963 allocated buffer. The implementation now correctly computes the required
964 buffer size when using malloc, and switches to malloc from alloca as
967 * A new semaphore algorithm has been implemented in generic C code for all
968 machines. Previous custom assembly implementations of semaphore were
969 difficult to reason about or ensure that they were safe. The new version
970 of semaphore supports machines with 64-bit or 32-bit atomic operations.
971 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
972 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
974 * Port to Altera Nios II has been contributed by Mentor Graphics.
976 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
977 implementations for powerpc64/powerpc64le.
978 Implemented by Adhemerval Zanella (IBM).
980 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
981 and powerpc64le. This may improve lock scaling of existing programs on
982 HTM capable systems. The lock elision code is only enabled with
983 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
984 powerpc will issue a transaction abort on every syscall to avoid side
985 effects being visible outside transactions.
987 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
988 AArch64. Contributed by ARM Ltd.
990 * i386 memcpy functions optimized with SSE2 unaligned load/store.
992 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
993 under certain input conditions resulting in the execution of a shell for
994 command substitution when the application did not request it. The
995 implementation now checks WRDE_NOCMD immediately before executing the
996 shell and returns the error WRDE_CMDSUB as expected.
998 * CVE-2012-3406 printf-style functions could run into a stack overflow when
999 processing format strings with a large number of format specifiers.
1001 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1002 infinite loop if the DNS response contained a PTR record of an unexpected
1005 * The minimum GCC version that can be used to build this version of the GNU
1006 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1007 still be used to compile programs using the GNU C Library.
1009 * The GNU C Library is now built with -Werror by default. This can be
1010 disabled by configuring with --disable-werror.
1012 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1014 * The obsolete sigvec function has been removed. This was the original
1015 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1016 programs have been using instead for about 25 years. Of course, ABI
1017 compatibility for old binaries using sigvec remains intact.
1019 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1020 with newer versions of bison.
1022 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1023 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1024 registers overlay two consecutive single-precision registers. MIPS32R2
1025 introduced a new FPU mode (FR=1) where double-precision registers extend the
1026 corresponding single-precision registers which is incompatible with the
1027 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1028 require the use of FR=1 making a transition necessary. New o32 ABI
1029 extensions enable users to migrate over time from the original o32 ABI
1030 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1031 tracks the ABI of any loaded object and verifies that new objects are
1032 compatible. Mode transitions will also be requested as required and
1033 unsupportable objects will be rejected. The ABI checks include both soft and
1034 hard float ABIs for o32, n32 and n64.
1036 GCC 5 with GNU binutils 2.25 onwards:
1037 It is strongly recommended that all o32 system libraries are built using the
1038 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1039 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1040 compiler using --with-fp-32=xx to set this by default.
1044 * The following bugs are resolved with this release:
1046 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1047 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1048 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1049 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1050 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1051 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1052 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1053 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1054 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1055 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1056 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1057 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1058 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1059 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1060 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1062 * Reverted change of ABI data structures for s390 and s390x:
1063 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1064 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1065 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1066 are preserved pointing straight to the same implementation as the old ones.
1067 Given that, new callers will simply provide a too-big buffer to these
1068 functions. Any applications/libraries out there that embed jmp_buf or
1069 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1070 against 2.19 headers will have to rebuilt again. This is necessary in any
1071 case to revert the breakage in their ABI caused by the glibc change.
1073 * Support for file description locks is added to systems running the
1074 Linux kernel. The standard file locking interfaces are extended to
1075 operate on file descriptions, not file descriptors, via the use of
1076 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1077 are associated with an open file instead of a process.
1079 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1081 * The minimum Linux kernel version that this version of the GNU C Library
1082 can be used with is 2.6.32.
1084 * Running the testsuite no longer terminates as soon as a test fails.
1085 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1086 with PASS or FAIL lines for individual tests. A summary of the results is
1087 printed, including a list of failing lists, and "make check" exits with
1088 error status if there were any unexpected failures. "make check
1089 stop-on-test-failure=y" may be used to keep the old behavior.
1091 * The am33 port, which had not worked for several years, has been removed
1094 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1095 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1096 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1097 conflicted with POSIX (support for which was removed in 2.19), the
1098 interfaces those macros enabled remain available when compiling with
1099 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1100 test macros defined.
1102 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1104 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1105 This may improve lock scaling of existing programs on TX capable systems.
1106 The lock elision code is only built with --enable-lock-elision=yes and
1107 then requires a GCC version supporting the TX builtins. With lock elision
1108 default mutexes are elided via __builtin_tbegin, if the cpu supports
1109 transactions. By default lock elision is not enabled and the elision code
1112 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1113 copy the path argument. This allowed programs to cause posix_spawn to
1114 deference a dangling pointer, or use an unexpected pathname argument if
1115 the string was modified after the posix_spawn_file_actions_addopen
1118 * All supported architectures now use the main glibc sysdeps directory
1119 instead of some being in a separate "ports" directory (which was
1120 distributed separately before glibc 2.17).
1122 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1123 On configurations that support it (all Linux configurations), it's now
1124 used regardless of the --enable-add-ons switch to configure. It is no
1125 longer possible to build such configurations without pthreads support.
1127 * Locale names, including those obtained from environment variables (LANG
1128 and the LC_* variables), are more tightly checked for proper syntax.
1129 setlocale will now fail (with EINVAL) for locale names that are overly
1130 long, contain slashes without starting with a slash, or contain ".." path
1131 components. (CVE-2014-0475) Previously, some valid locale names were
1132 silently replaced with the "C" locale when running in AT_SECURE mode
1133 (e.g., in a SUID program). This is no longer necessary because of the
1136 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1137 with application code using Intel MPX instructions. (With all previous
1138 versions, the MPX register state could be clobbered when making calls
1139 into or out of a shared library.) Note that while the new dynamic
1140 linker is compatible with all known x86 hardware whether or not it
1141 supports Intel MPX, some x86 instruction-set emulators might fail to
1142 handle the new instruction encodings. This is known to affect Valgrind
1143 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1144 release), and might affect other tools that do instruction emulation.
1146 * Support for loadable gconv transliteration modules has been removed.
1147 The support for transliteration modules has been non-functional for
1148 over a decade, and the removal is prompted by security defects. The
1149 normal gconv conversion modules are still supported. Transliteration
1150 with //TRANSLIT is still possible, and the //IGNORE specifier
1151 continues to be supported. (CVE-2014-5119)
1153 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1154 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1155 resulting a denial-of-service security vulnerability in applications which
1156 use functions related to iconv. (CVE-2014-6040)
1160 * The following bugs are resolved with this release:
1162 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1163 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1164 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1165 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1166 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1167 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1168 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1169 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1170 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1171 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1172 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1173 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1174 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1175 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1176 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1177 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1178 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1179 16431, 16453, 16474, 16506, 16510, 16529
1181 * Slovenian translations for glibc messages have been contributed by the
1182 Translation Project's Slovenian team of translators.
1184 * The public headers no longer use __unused nor __block. This change is to
1185 support compiling programs that are derived from BSD sources and use
1186 __unused internally, and to support compiling with Clang's -fblock
1187 extension which uses __block.
1189 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1190 large collation sequences to optimize multiple passes. This cache
1191 computation may overflow for large collation sequences and may cause a
1192 stack or buffer overflow. This is now fixed to use a slower algorithm
1193 which does not use a cache if there is an integer overflow.
1195 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1196 rules for large collation sequences to optimize multiple passes and falls
1197 back to alloca if malloc fails, resulting in a possible stack overflow.
1198 The implementation now falls back to an uncached collation sequence lookup
1201 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1202 initialized for static applications resulting in the security feature
1203 being disabled. The pointer guard is now correctly initialized to a
1204 random value for static applications. Existing static applications need
1205 to be recompiled to take advantage of the fix (bug 15754).
1207 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1208 to the d_name member of struct dirent, or omit the terminating NUL
1209 character. (Bugzilla #14699).
1211 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1212 aligned_alloc functions could allocate too few bytes or corrupt the
1213 heap when passed very large allocation size values (Bugzilla #15855,
1216 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1217 for AF_INET6 has been fixed (Bugzilla #16072).
1219 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1220 pap_CW, quz_PE, the_NP.
1222 * Substantially revised locales: gd_GB, ht_HT
1224 * The LC_ADDRESS field was updated to support country_car for almost all
1227 * ISO 1427 definitions were updated.
1229 * ISO 3166 definitions were updated.
1231 * The localedef utility now supports --big-endian and --little-endian
1232 command-line options to generate locales for a different system from that
1233 for which the C library was built.
1235 * Binary locale files now only depend on the endianness of the system for
1236 which they are generated and not on other properties of that system. As a
1237 consequence, binary files generated with new localedef may be incompatible
1238 with old versions of the GNU C Library, and binary files generated with
1239 old localedef may be incompatible with this version of the GNU C Library,
1240 in the following circumstances:
1242 + Locale files may be incompatible on m68k systems.
1244 + Locale archive files (but not separate files for individual locales) may
1245 be incompatible on systems where plain "char" is signed.
1247 * The configure option --disable-versioning has been removed. Builds with
1248 --disable-versioning had not worked for several years.
1250 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1251 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1253 * SystemTap probes for malloc have been introduced.
1255 * SystemTap probes for slow multiple precision fallback paths of
1256 transcendental functions have been introduced.
1258 * Support for powerpc64le has been added.
1260 * The soft-float powerpc port now supports e500 processors.
1262 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1264 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1265 set of header declarations that are enabled by default, even when other
1266 feature test macros or compiler options such as -std=c99 would otherwise
1267 disable some of those declarations.
1269 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1270 conflict with POSIX. The libbsd-compat library (which was a dummy library
1271 that did nothing) has also been removed.
1273 * Preliminary documentation about Multi-Thread, Async-Signal and
1274 Async-Cancel Safety has been added.
1276 * Change of ABI data structures for s390 and s390x:
1277 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1278 allow for future hardware extensions. All library functions that accept or
1279 return these structures were versioned in order to provide backwards
1280 compatibility with existing code. However, not all cases can be handled
1281 with interface versioning. If an application embeds either structure into
1282 another structure and that structure is passed to another compilation unit
1283 which expects the newer larger-sized object then any access to the new
1284 fields will result in undefined behaviour. Similarly any access to fields
1285 that were moved by the enlarging of the structures will cause undefined
1286 behaviour. To fix the undefined behaviour all objects that are part of the
1287 final in-memory image must agree on the size of structures, and this may
1288 require recompilation.
1292 * The following bugs are resolved with this release:
1294 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1295 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1296 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1297 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1298 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1299 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1300 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1301 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1302 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1303 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1304 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1305 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1306 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1309 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1310 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1311 Distributions can re-enable building and using pt_chown via the new configure
1312 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1313 considerable security risks and should only be used if the distribution
1314 understands and accepts the risks.
1316 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1319 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1320 fixed (Bugzilla #15330).
1322 * Add support for calling C++11 thread_local object destructors on thread
1323 and program exit. This needs compiler support for offloading C++11
1324 destructor calls to glibc.
1326 * Improved worst case performance of libm functions with double inputs and
1329 * Support for priority inherited mutexes in pthread condition variables on
1330 non-x86 architectures.
1332 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1334 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1336 * Optimized string functions for ARM. Implemented by Will Newton and
1339 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1341 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1344 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1345 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1347 * Added a benchmark framework to track performance of functions in glibc.
1349 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1350 It is based on draft TS 18661 and currently enabled as a GNU extension.
1352 * On Linux, the clock function now uses the clock_gettime system call
1353 for improved precision, rather than old times system call.
1355 * Added support for version-3 tz format files. This is needed when using
1356 the tz database release 2013e or later, and affects a few unusual cases --
1357 currently only TZ='America/Godthab' for time stamps after 2037.
1359 * Added new API functions pthread_getattr_default_np and
1360 pthread_setattr_default_np to get and set the default pthread
1361 attributes of a process.
1363 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1364 This may improve lock scaling of existing programs on TSX capable systems.
1365 When the --enable-lock-elision=yes parameter is specified at configure
1366 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1369 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1370 availability). Implemented by Ryan S. Arnold.
1372 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1376 * The following bugs are resolved with this release:
1378 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1379 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1380 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1381 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1382 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1383 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1384 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1385 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1386 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1387 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1388 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1389 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1390 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1391 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1392 14879, 14889, 14893, 14898, 14914.
1394 * Optimization of memcpy for MIPS.
1396 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1397 EMFILE has been fixed (Bugzilla #14889).
1399 * The add-on ports collection is now distributed in the "ports" subdirectory
1400 of the main GNU C Library distribution, rather than separately.
1402 * Port to ARM AArch64 contributed by Linaro.
1404 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1405 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1407 Implemented by Andreas Krebbel.
1409 * The new function secure_getenv allows secure access to the environment,
1410 returning NULL if running in a SUID/SGID process. This function replaces
1411 the internal function __secure_getenv.
1413 * SystemTap static probes have been added into the dynamic linker.
1414 Implemented by Gary Benson.
1416 * Optimizations of string functions strstr, strcasestr and memmem.
1417 Implemented by Maxim Kuvyrkov.
1419 * The minimum Linux kernel version that this version of the GNU C Library
1420 can be used with is 2.6.16.
1422 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1423 powerpc POWER7. Implemented by Will Schmidt.
1425 * New configure option --disable-nscd builds the C library such that it
1426 never attempts to contact the Name Service Caching Daemon (nscd).
1427 New configure option --disable-build-nscd avoids building nscd itself;
1428 this is the default if --disable-nscd is used.
1430 * Improved support for cross-compilation, including cross-testing and
1431 bootstrap builds without a previously built glibc.
1433 * Several testsuite tests are now able to test multiple IFUNC variants of an
1434 interface, rather than just testing the one that would be chooen by
1437 * New configure options --with-bugurl and --with-pkgversion, for
1438 distributors to use to embed their bug-reporting and package version
1439 information in --help and --version output.
1441 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1442 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1443 allows creation of chroots without the procfs mounted on /proc.
1445 * The `crypt' function now fails if passed salt bytes that violate the
1446 specification for those values. On Linux, the `crypt' function will
1447 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1448 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1449 when the mode is enabled.
1451 * The `clock_*' suite of functions (declared in <time.h>) is now available
1452 directly in the main C library. Previously it was necessary to link with
1453 -lrt to use these functions. This change has the effect that a
1454 single-threaded program that uses a function such as `clock_gettime' (and
1455 is not linked with -lrt) will no longer implicitly load the pthreads
1456 library at runtime and so will not suffer the overheads associated with
1457 multi-thread support in other code such as the C++ runtime library.
1459 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1464 * The following bugs are resolved with this release:
1466 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1467 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1468 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1469 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1470 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1471 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1472 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1473 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1474 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1475 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1476 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1477 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1478 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1479 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1480 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1481 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1482 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1483 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1484 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1485 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1486 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1487 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1490 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1491 configuring glibc with:
1492 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1493 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1494 Implemented by H.J. Lu.
1498 + define static_assert
1500 + do not declare gets
1502 + declare at_quick_exit and quick_exit also for ISO C11
1504 + aligned_alloc. NB: The code is deliberately allows the size parameter
1505 to not be a multiple of the alignment. This is a moronic requirement
1506 in the standard but it is only a requirement on the caller, not the
1509 + timespec_get added
1511 + uchar.h support added
1513 + CMPLX, CMPLXF, CMPLXL added
1515 Implemented by Ulrich Drepper.
1517 * Support for the IA-64 has been moved to ports.
1519 * Remove support for anything but ELF binary format
1521 * Checking versions of poll, ppoll added.
1522 Implemented by Ulrich Drepper.
1524 * More generic and 64-bit performance optimizations to math functions.
1525 Implemented by Ulrich Drepper.
1527 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1528 headers and functions available at compile time as they were before
1529 version 2.14. This option will be removed at some time in the future
1530 after the TI-RPC library becomes fully sufficient for the needs of
1531 existing applications.
1533 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1534 Note that glibc is not expected to work with any Linux kernel version
1537 * New header <sys/auxv.h> and function getauxval allowing easy access to
1538 the AT_* key-value pairs passed from the Linux kernel. The header also
1539 defines the HWCAP_* bits associated with the AT_HWCAP key.
1541 * New locales: mag_IN
1543 * New configure option --enable-systemtap builds SystemTap static probes
1544 into libc for setjmp and longjmp and into libpthread for various operations.
1545 So far the setjmp/longjmp probes and some of the libpthread probes are
1546 provided only for i*86 and x86_64.
1547 Implemented by Roland McGrath and Rayson Ho.
1549 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1551 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1554 * More optimized functions for SPARC. Implemented by David S. Miller.
1556 * Improved support for cross-compilation, especially bootstrap builds
1557 without a previously built glibc.
1559 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1560 Chris Metcalf from Tilera.
1562 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1563 now supported for ARM processors.
1565 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1566 as the name of the dynamic linker, to distinguish it from the
1567 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1569 * Support for CRIS has been removed from ports.
1571 * A new class of installed header has been documented for low-level
1572 platform-specific functionality. PowerPC added the first instance with a
1573 function to provide time base register access. Contributed by Tulio
1574 Magno Quites Machado Filho.
1576 * ix86 configurations now install header files that are consistent with
1577 what x86-64 configurations install. These same header files can be used
1578 for -m32, -m64, or -mx32 builds.
1579 Contributed by H.J. Lu.
1581 * Math library bug fixes. A thorough audit of all open math library bugs was
1582 conducted by Joseph Myers. Significant progress was made on many math
1583 library bugs resulting in more accurate exceptions and function results.
1584 Many thanks to all those that contributed including Andreas Jaeger for his
1585 patch review and work on the x87 trigonometric instruction issues.
1587 * Timezone data is no longer installed. Timezone-related binaries and scripts
1588 will continue to be installed. Users should obtain their timezone data from
1589 their distribution provider or from the tzdata package at
1590 <ftp://munnari.oz.au/pub/>.
1594 * The following bugs are resolved with this release:
1596 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1597 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1598 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1599 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1600 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1601 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1602 13484, 13506, 13515, 13523, 13524, 13538, 13540
1604 * New program pldd to list loaded object of a process
1605 Implemented by Ulrich Drepper.
1607 * Add nss_db support back to glibc. No more dependency on Berkeley db
1608 and support for initgroups lookups.
1609 Implemented by Ulrich Drepper.
1611 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1612 Contributed by HJ Lu.
1614 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1615 Contributed by HJ Lu.
1617 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1618 on x86-32 and x86-64.
1619 Contributed by Liubov Dmitrieva.
1621 * Optimized strchr and strrchr for SSE on x86-32.
1622 Contributed by Liubov Dmitrieva.
1624 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1625 for x86-64 and x86-32.
1626 Contributed by Liubov Dmitrieva.
1628 * New interfaces: scandirat, scandirat64
1629 Implemented by Ulrich Drepper.
1631 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1632 Implemented by Ulrich Drepper.
1634 * nscd now also caches the netgroup database.
1635 Implemented by Ulrich Drepper.
1637 * Integrate libm with gcc's -ffinite-math-only option.
1638 Implemented by Ulrich Drepper.
1640 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1641 to math functions. Implemented by Ulrich Drepper.
1643 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1644 Implemented by Ulrich Drepper.
1646 * New Linux interfaces: process_vm_readv, process_vm_writev
1648 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1649 Implemented by Ulrich Drepper.
1651 * Optimized nearbyint and strcasecmp for PPC.
1652 Implemented by Adhemerval Zanella.
1654 * New locales: bho_IN, unm_US, es_CU, ta_LK
1658 * The following bugs are resolved with this release:
1660 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1661 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1662 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1663 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1664 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1665 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1666 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1667 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1668 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1669 12795, 12811, 12813, 12814, 12841
1671 * The RPC implementation in libc is obsoleted. Old programs keep working
1672 but new programs cannot be linked with the routines in libc anymore.
1673 Programs in need of RPC functionality must be linked against TI-RPC.
1674 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1676 Visible changes of this change include (obviously) the inability to link
1677 programs using RPC functions without referencing the TI-RPC library and the
1678 removal of the RPC headers from the glibc headers.
1679 Implemented by Ulrich Drepper.
1681 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1682 syncfs, setns, sendmmsg
1684 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1685 yue_HK, lij_IT, mhr_RU
1687 * New encodings: CP770, CP771, CP772, CP773, CP774
1689 * New program sotruss to trace calls through PLTs
1690 Implemented by Ulrich Drepper.
1692 * The malloc hook implementation is marked deprecated and will be removed
1693 from the default implementation in the next version. The design never
1694 worked ever since the introduction of threads. Even programs which do
1695 not create threads themselves can use multiple threads created internally.
1699 * The following bugs are resolved with this release:
1701 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1702 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1703 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1704 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1707 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1709 * POWER7 optimizations: memset, memcmp, strncmp
1711 * New optimized string functions for x86-64: strnlen (SSE2),
1712 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1713 Implemented by Ulrich Drepper.
1717 * The following bugs are resolved with this release:
1719 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1720 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1721 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1722 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1723 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1724 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1725 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1726 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1728 * New interfaces: pthread_getname_np, pthread_setname_np
1730 * New Linux interface: recvmmsg
1732 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1734 * The dynamic linker now recognizes supported ABI versions from the
1735 EI_ABIVERSION field in the ELF header.
1736 Implemented by Ulrich Drepper.
1738 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1739 will not be used to fill in password fields in the passwd.byname replies.
1740 Instead it is used to synthesize the shadow.byname table, should it be
1741 missing. This is a useful mode in some installations involving Solaris.
1742 Implemented by Ulrich Drepper.
1744 * New locales: kok_IN, sq_MK, cv_RU
1748 * The following bugs are resolved with this release:
1750 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1751 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1752 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1753 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1754 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1755 10789, 10817, 10823, 10840, 10847
1757 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1759 Implemented by Ulrich Drepper.
1761 * Checking version of longjmp added that fails if an uninitialized stack
1762 frame would be created. Implemented by Ulrich Drepper.
1764 * STT_GNU_IFUNC is now supported in static executables.
1765 Implemented by H.J. Lu.
1767 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1769 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1770 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1771 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1772 Contributed by H.J. Lu.
1774 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1775 Implemented by Ulrich Drepper.
1777 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1779 Contributed by H.J. Lu.
1781 * Support for fma instruction in AVX on x86-64.
1782 Implemented by H.J. Lu and Ulrich Drepper.
1784 * AVX support in x86-64 auditing support in ld.so.
1785 Implemented by H.J. Lu.
1787 * STB_GNU_UNIQUE support added.
1788 Implemented by Ulrich Drepper.
1790 * Implement second fallback mode for optimized DNS lookup for even more
1791 broken environments. If two requests from the same port are not
1792 handled correctly close the socket and open a new one before sending
1793 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1794 can be used to select this mode right away, instead of rediscovering the
1795 necessity is every process again.
1796 Implemented by Ulrich Drepper.
1798 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1799 Implemented by Adam Tkac.
1801 * Optimized iconv conversions for S390x.
1802 Implemented by Andreas Krebbel.
1804 * Using condvars with PI mutexes is now more efficient due to kernel
1805 support for requeueing to PI futexes. NPTL support added for x86-64.
1806 Implemented by Ulrich Drepper.
1808 * New locales: ps_AF, my_MM
1812 * The following bugs are resolved with this release:
1814 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1815 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1816 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1817 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1818 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1820 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1821 Implemented by Ulrich Drepper.
1823 * New Linux interfaces: accept4, fallocate, fallocate64.
1824 Implemented by Ulrich Drepper.
1826 * Correct declarations of string function when used in C++ code. This
1827 could lead to compile errors for invalid C++ code.
1829 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1832 * New POSIX 2008 interface: psiginfo
1833 Implemented by Ulrich Drepper.
1835 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1836 Implemented by Ulrich Drepper.
1838 * Support for selecting between multiple function definitions at runtime
1839 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1841 * The libcrypt library can now use the hash function implementations in
1842 NSS. Implemented by Ulrich Drepper.
1844 * The malloc implementation can be compiled to be less memory efficient
1845 but higher performing in multi-threaded programs.
1846 Implemented by Ulrich Drepper.
1848 * New locales: nan_TW@latin, ks_IN
1850 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1851 Implemented by Ulrich Drepper.
1853 * Extended printf hook support. It is possible to use user-defined types
1854 and extend existing format specifiers.
1855 Implemented by Ulrich Drepper.
1857 * Handling for group shadow files has been added.
1858 Implemented by Ulrich Drepper.
1860 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1861 servers (the case, e.g., for some people using the built-in DNS
1862 server in ADSL modems/routers). There is a once-per-process timeout
1863 in case of a broken server. To avoid it, users can run nscd or put
1864 'options single-request' in /etc/resolv.conf.
1865 Implemented by Ulrich Drepper.
1869 * The following bugs are resolved with this release:
1871 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1872 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1873 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1874 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1875 6968, 6974, 6980, 6995, 7008, 7009, 7029
1877 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1878 up at the same time. Implemented by Ulrich Drepper.
1880 * TLS descriptors for LD and GD on x86 and x86-64.
1881 Implemented by Alexandre Oliva.
1883 * getaddrinfo now handles DCCP and UDPlite.
1884 Implemented by Ulrich Drepper.
1886 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1887 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1888 Implemented by Ulrich Drepper.
1890 * New implementation of memmem, strstr, and strcasestr which is O(n).
1891 Implemented by Eric Blake.
1893 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1895 * Implement "e" option for popen to open file descriptor with the
1896 close-on-exec flag set. Implemented by Ulrich Drepper.
1898 * Implement "b" mode for fmemopen. In this mode writes writes don't
1899 implicitly add a NUL byte and seeks from the end of the buffer really
1900 use the buffer end, not the string length as the basis.
1901 Implemented by Ulrich Drepper.
1903 * Many functions, exported and internal, now atomically set the close-on-exec
1904 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1906 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1908 Implemented by Pravin Satpute.
1910 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1914 * The following bugs are resolved with this release:
1916 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1917 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1918 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1919 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1920 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1921 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1922 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1924 * New locales: bo_CN, bo_IN, shs_CA.
1926 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1928 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1929 Implemented by Pravin Satpute.
1931 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1933 * nscd caches now all timeouts for DNS entries
1934 Implemented by Ulrich Drepper.
1936 * nscd is more efficient and wakes up less often.
1937 Implemented by Ulrich Drepper.
1939 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1940 vdprintf, and obstack_vprintf.
1941 Implemented by Jakub Jelinek.
1943 * Faster memset for x86-64.
1944 Implemented by Harsha Jagasia and H.J. Lu.
1946 * Faster memcpy on x86.
1947 Implemented by Ulrich Drepper.
1949 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1950 Implemented by Ulrich Drepper.
1952 * Faster sqrt and sqrtf implemention for some PPC variants.
1953 Implemented by Steven Munroe.
1957 * The following bugs are resolved with this release:
1959 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1960 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1961 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1962 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1963 5071, 5103, 5104, 5112, 5113, 5184, 5186
1965 * More checking functions: fread, fread_unlocked, open*, mq_open.
1966 Implemented by Jakub Jelinek and Ulrich Drepper.
1968 * Extend fortification to C++. Implemented by Jakub Jelinek.
1970 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1971 by not recognizing 'a' as a modifier when those specs are requested.
1972 Implemented by Jakub Jelinek.
1974 * PPC optimizations to math and string functions.
1975 Implemented by Steven Munroe.
1977 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1978 options to be passed. Implemented by Ulrich Drepper.
1980 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1982 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1983 Implemented by Ulrich Drepper.
1985 * Handle private futexes in the NPTL implementation.
1986 Implemented by Jakub Jelinek and Ulrich Drepper.
1988 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1989 Implemented by Roland McGrath and Ulrich Drepper.
1991 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1993 * SHA-256 and SHA-512 based password encryption.
1994 Implemented by Ulrich Drepper.
1996 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1997 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2000 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2001 Implemented by Ulrich Drepper.
2005 * The following bugs are resolved with this release:
2007 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2008 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2009 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2010 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2011 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2012 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2013 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2016 * New Linux interfaces: epoll_pwait, sched_getcpu.
2018 * New generic interfaces: strerror_l.
2020 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2024 * The following bugs are resolved with this release:
2026 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2027 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2028 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2029 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2030 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2031 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2032 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2033 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2034 3225, 3251, 3252, 3253, 3273
2036 * For Linux, the sorting of addresses returned by getaddrinfo now also
2037 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2038 Implemented by Ulrich Drepper.
2040 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2041 Implemented by Ulrich Drepper.
2043 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2045 * New iconv module for MIK. Contributed by Alexander Shopov.
2047 * For sites with broken group and/or passwd database, the auto-propagate
2048 option of nscd can prevent creating ID lookup entries from the results
2049 of a name lookup and vice versa. This usually is no problem but some
2050 site might have problems with the default behavior.
2051 Implemented by Ulrich Drepper.
2053 * Iterating over entire database in NIS can be slow. With the
2054 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2055 to trade time for memory. The entire database will be read at once.
2056 Implemented by Ulrich Drepper.
2058 * The interfaces introduced in RFC 3542 have been implemented by
2061 * Support for the new ELF hash table format was added by Ulrich Drepper.
2063 * Support for priority inheritance mutexes added by Jakub Jelinek and
2066 * Support for priority protected mutexes added by Jakub Jelinek.
2068 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2069 or_IN, csb_PL, fy_NL, sr_ME.
2073 * More overflow detection functions.
2075 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2076 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2078 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2079 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2080 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2081 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2082 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2083 by Masahide Washizawa.
2085 * It is now possible to install audit libraries for the dynamic linker using
2086 LD_AUDIT. Implemented by Ulrich Drepper.
2088 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2089 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2090 For a libc and libpthread that works well on Linux 2.4 kernels, we
2091 recommend using the stable 2.3 branch.
2093 * The new function open_wmemstream acts like open_memstream,
2094 but uses a wchar_t wide character buffer.
2096 * The new function ppoll is an improved version of poll, similar to pselect.
2098 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2099 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2100 renameat, unlinkat, linkat, symlinkat, readlinkat.
2102 * New Linux kernel system calls: unshare,
2103 inotify_init, inotify_add_watch, inotify_rm_watch.
2105 * The euidaccess function is now also known by the alias eaccess,
2106 for compatibility with some other systems.
2108 * Timezone data updated to 2006b version.
2112 * The following bugs are resolved with this release:
2114 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2115 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2116 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2117 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2118 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2119 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2121 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2123 * As of this release, GCC 4 can be used to compile the C Library.
2125 * Timezone data updated to 2005m version.
2129 * The following bugs are resolved with this release:
2131 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2132 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2133 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2134 777, 787, 821, 822, 823, 825
2136 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2140 * Support for RFC 3678. Real implementations exist only for Linux so far.
2141 Implemented by Ulrich Drepper.
2143 * nscd can now cache entries persistently. Expiring entries are reloaded.
2144 For speedups the cache can be shared in memory with client processes.
2145 Implemented by Ulrich Drepper.
2147 * nscd can now perform SELinux checks.
2148 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2150 * getaddrinfo queries are now cached. Canonical name lookup is performed
2152 Implemented by Ulrich Drepper.
2154 * The nothrow function attribute is used when headers are used by gcc when
2155 compiling C code. This can avoid significant amounts of exception
2158 * The malloc functions perform more error checking and are stricter when
2159 it comes to reacting on errors. The default action is to terminate
2160 the process after showing an error message. Implemented by Ulrich Drepper.
2162 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2163 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2165 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2166 namespaces using the new function dlmopen(). This feature is of course,
2167 like most other dynamic loading functionality, not available in statically
2168 linked applications. Implemented by Ulrich Drepper.
2170 * Low-overhead boundary checking variants of string and some stdio functions
2171 were added. These are to be used in conjunction with a gcc patch by
2172 Jakub Jelinek which adds calls to these functions if possible.
2173 Implemented by Jakub Jelinek and Ulrich Drepper.
2175 * Old code for several operating systems and machine architectures that
2176 have not been in working condition in a long time have been removed from
2177 the main source tree maintained by the GNU C Library's maintainers.
2178 These files are now reside in the separate `ports' source module
2179 that is usable as an add-on when building the library.
2183 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2184 interrogate the dynamic linker, compatible with the Solaris interface.
2186 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2187 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2189 * getifaddrs now uses the netlink interface on Linux to get its information.
2190 Implemented by Thorsten Kukuk.
2192 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2193 Implemented by Ulrich Drepper.
2195 * support for non-executable stacks on x86 has been added. Changes mostly
2198 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2201 * getaddrinfo now performs destination address selection according to
2206 * Thread-safe interfaces for many functions that access locale data
2207 were added in version 2.3, but these features were omitted from NEWS.
2208 Many functions have variants with an `_l' suffix that take a `locale_t'
2209 object as a parameter rather than consulting the current locale.
2210 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2211 create and maintain `locale_t' objects. Additionally, the new function
2212 `uselocale' sets "the current locale" (as used by functions not so
2213 parameterized) set for an individual thread. These features were added
2214 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2216 * The functions getresuid, getresgid, setresuid, and setresgid, which
2217 have long been available on Linux, are now declared in <unistd.h>
2218 and are now also available on the Hurd.
2220 * ELF thread-local storage support (TLS) now works on x86-64.
2222 * The new dynamic string token $LIB is expanded in shared library names.
2223 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2225 * Aldy Hernandez contributed complete software floating point support for
2226 PowerPC machines with no FPU.
2228 * fexecve is implemented on Linux.
2230 * The `btowc' function should work at least twice as fast due to
2231 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2233 * With approriate thread add-ons cancelable functions are now implemented
2234 in libc.so as well. No need to call the function in libpthread. This
2235 change allowed to finally disable the incorrect and expensive handling
2236 of weak definition in ld.so.
2238 * Yet more PLT entries in libc.so have been removed. We finally arrived
2239 at the bare minimum. Startup times improved appropriately.
2241 * Support for the new Linux/x86 system call interface was added. The
2242 AT_SYSINFO auxiliary vector entry is recognized and handled.
2246 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2249 * iconv (the program and the interface) now accepts empty names (excluding
2250 options like //TRANSLIT) to mean "use charset of current locale".
2252 * localedef can now transliterate characters in strings which are not in
2253 the provided charmap. The information from the input locale is used.
2255 * Prelinking support was added for ELF targets. This requires additional
2256 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2258 * Read-only stdio streams now use mmap to speed up operation by eliminating
2259 copying and buffer underflows. To use add 'm' to the mode string of
2260 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2262 * The malloc functions were completely rewritten by Wolfram Gloger based
2263 on Doug Lea's malloc-2.7.0.c.
2265 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2266 implementation of regex.
2268 * Bruno Haible upgraded the iconv and locale implementation to support
2271 * Contents of the LC_* and LANG environment variables in the CEN style are
2272 not recognized anymore. It never was used. Change by Ulrich Drepper.
2274 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2275 thread-local storage (TLS) ABI on some platforms.
2276 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2278 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2279 EUC-JISX0213, and TSCII.
2281 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2282 BSD-compatible interface for getting all network interface addresses.
2283 Implementation for IPv4 by Roland McGrath.
2285 * Loading of locale data is faster due to the introduction of a locale
2286 archive. Implemented by Roland McGrath and Ulrich Drepper.
2288 * Startup times are significantly reduced by not using exported functions
2289 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2292 * Steven Munroe contributed a port to PowerPC64/Linux.
2296 * The Hurd now uses the GNU libio implementation of stdio rather than the
2297 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2299 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2300 with the same behavior as the Linux system call of the same name.
2304 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2305 128-bit long double format.
2307 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2308 IBM1161, and IBM1162 charsets.
2310 * Andreas Jaeger contributed a port to x86-64/Linux.
2312 * Peter Bruin contributed a port to PowerPC/Hurd.
2314 * libc functions using I/O streams now can handle wide-oriented streams
2317 * optimizations in the dynamic linker. Binaries created by recent binutils
2318 versions start up quicker due to reduced time spent on relocations.
2320 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2324 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2325 asinh, atanh, j0 for the 128-bit long double format.
2327 * Bruno Haible updated all the code handling Unicode in some form to
2328 support Unicode 3.1.
2330 * Speed of regex for single-byte locales is back to previous levels.
2331 Patch by Isamu Hasegawa.
2333 * Alpha, SPARC, and IA-64 now also using floating stacks.
2335 * Startup time of internationalized applications greatly improved through
2336 iconv cache. Use iconvconfig to generate the cache file.
2337 Contributed by Ulrich Drepper.
2339 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2340 severe performance problems. Patch by David Mosberger.
2342 * The Hurd port got a lot more functionality like AIO, various stdio
2343 extensions, etc. Mainly done by Roland McGrath.
2345 * mtrace can now lookup symbols in shared libraries.
2349 * Intel's IA-64 math library is largely integrated. It provides fast and
2350 accurate implementatations for most basic and standard math functions
2351 in float, double, and long double format.
2353 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2354 and asin for the 96-bit long double format and asin, log, tan for the
2355 128-bit long double format.
2357 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2358 The basic double functions exist today. Contributed by Abraham Ziv
2359 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2360 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2362 * An asynchronous name lookup library was added. The interface is designed
2363 after POSIX AIO. The proposal was circulated beforehand to get comments.
2364 No negative ones came in. Implemented by Ulrich Drepper.
2366 * Port to S390/64bit contributed by Martin Schwidefsky
2367 <schwidefsky@de.ibm.com>.
2369 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2370 of functions for Linux/IA-64.
2372 * The RPC code is now thread safe. Threads can now use the same service
2373 of different services at the same time. Patch by Eric Norum
2374 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2376 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2377 family of functions for Linux/S390.
2379 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2380 of functions for Linux/x86.
2382 * Port to Linux/CRIS contributed by Axis Communications.
2386 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2387 we can now check for namespace violations and missing declarations. The
2388 result is that almost all headers are now Unix-compliant (as defined in
2389 the upcoming XPG6). The negative side is that some programs might need
2390 corrections, too, if they depend on the incorrect form of the headers in
2391 previous versions which defined too many symbols and included too many
2394 * regex now handles multibyte character sets correctly.
2395 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2397 * iconv (the program) does now conform to the upcoming XPG6 and handles
2398 charmaps. Instead of the charset names the path of charmaps can be
2399 provided and the conversion happens based on this data.
2400 Contributed by Ulrich Drepper.
2402 * The locale program now provides detailed information about the installed
2407 only lists the names of the supported locales
2411 provides details such as country, language, and codeset name.
2412 Contributed by Ulrich Drepper.
2416 * The gencat program now parses the input file according to the charset
2417 selected by the LC_CTYPE category. This is important for stateful
2418 character sets. To make generating catalogs easier there is a way
2419 to overwrite the charset selected by the locale: before the first
2420 message or $ quote line the catalog can contain a line like
2422 $ codeset=ISO-8859-2
2424 to select the charset (ISO-8859-2 in this case).
2426 Implemented by Shinya Hanataka and Ulrich Drepper.
2428 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2429 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2430 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2432 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2433 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2435 * The LANGUAGE environment variable is now ignored unless the locale is
2436 changed from the default "C" locale.
2438 * The usual bug fixes.
2442 * Greg McGary added runtime support for bounds checking using gcc's
2443 new -fbounded-pointers option. ix86 target is complete. PowerPC
2446 * Thorsten Kukuk added secure mode support to nscd.
2448 * The Berkeley db libraries have been removed.
2450 Related, the nss_db module is now in a separate package since it
2451 obviously requires a database library being available.
2453 * Wide character I/O streams implemented by Ulrich Drepper.
2455 * Functions from the extended socket API added by Ulrich Drepper.
2457 * Functions feenableexcept and fedisableexcept to control the
2458 behaviour of individual exceptions have been added by Andreas Jaeger.
2460 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2462 * The resolver code has been updated from bind 8.2.3-T5B which supports
2463 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2466 This change could in some situations effect backward compatibility. Since
2467 now `_res' is a thread-local instead of a global variable, modifying it
2468 in one thread does not have any effect in other threads.
2470 The resolver library was also extended to allow IPv6 as the transport
2471 protocol for the requests. This work was done by Stig Venaas.
2473 * Compatibility code for K&R C compilers has been removed from the
2474 header files. A ISO C compiler is needed to use the library
2475 (conforming to either C89 or C99 standard).
2477 * Complete rewrite of the localedef program to support multibyte character
2478 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2479 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2480 Implemented by Ulrich Drepper.
2482 Bruno Haible significantly improved the generation and use of the data
2483 structures for the wide character tables.
2485 * Plural handling in gettext implemented by Ulrich Drepper.
2487 * The utmp daemon has been removed.
2489 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2491 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2494 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2496 * POSIX barriers implemented by Kaz Kylheku.
2498 * POSIX spawn function family implemented by Ulrich Drepper.
2500 * POSIX spinlocks are now available.
2502 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2504 * the configure option --enable-kernel=X.Y.Z allows to strip out
2505 compatibility for kernel versions before X.Y.Z. This is currently only
2506 implemented for Linux.
2508 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2509 field sin6_scope_id. This means that all programs using IPv6 should be
2510 recompiled. Don't expect binary compatibility with previous glibc
2513 * various conversion modules for IBM character sets contributed by
2516 * IA-64 port by Jes Sorensen and HJ Lu.
2530 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2532 * New xdr functions are added; some rpc functions are now 64bit clean.
2534 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2536 * Fixed known incompatibilities with glibc 2.0.
2538 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2540 * Optimized a number of functions (especially the ELF dynamic loader).
2542 * Update timezone data files.
2544 * lots of charmaps corrections
2546 * some new locale definitions and charmaps
2551 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2552 conform to POSIX member type requirements. Symbol versions have been
2553 adjusted as needed within the library, and for direct use by applications,
2554 but there is potential for problems if third-party libraries use
2555 struct timeval as part of their interface. This does not present
2556 a problem for X and other "essential" system libraries.
2558 * An additional locale model to support C++ Standard Library locale
2559 model and probably more was implemented by Ulrich Drepper.
2561 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2564 * Miles Bader provided the `argp' function family to support hierachical
2565 command line argument parsing, layered on top of getopt.
2567 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2569 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2572 * scanf recognizes the %a and %A format for scanning floating point numbers.
2574 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2575 information and interfaces for the available integer types.
2577 * about 130 new math functions were added to implement the ISO C9x math
2580 * the new header <complex.h> contains definitions of the complex math
2581 functions from ISO C 9X.
2583 * the new header <tgmath.h> defines generic macros to use complex or
2584 real valued functions.
2586 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2588 * Andreas Jaeger provided a test suite for the math library.
2590 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2592 * Ulrich Drepper added character set conversion functions (iconv).
2594 * Optimized string functions have been added.
2596 * The localedata addon is now part of glibc.
2598 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2600 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2601 daemon for NSS (nscd).
2603 Missing a better place here are some numbers on improvements. Under
2604 Linux 2.1.125 un-tar-ing the kernel sources takes
2608 using local files 12.19s 6.88s 22.91s
2610 using NIS 13.92s 8.91s 26.34s
2612 using NIS & nscd 10.37s 7.34s 25.30s
2614 using NIS+ 27.57s 30.37s 640.46s
2616 using NIS+ & nscd 10.25s 7.83s 26.51s
2618 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2620 Keep in mind that non-namelookup related operations dominate above times.
2621 It was just a common complain that using NIS+ unpacking the kernel is
2624 [1] The old nscd implementation is not available anymore since it was
2625 distributed with glibc up to version 2.0.98 and thus is now replaced.
2627 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2629 * Mark Kettenis provided a Hesiod NSS module.
2631 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2632 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2634 * rcmd can now handle netgroups (Dick Streefland).
2636 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2639 * Support for the IPv6 protocol has been added to the socket API, as per the
2640 latest draft standards.
2642 * Support for Linux 2.2 has been added.
2644 * Interface changes relative to the latest 2.0.x release:
2645 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2646 addseverity NEW: Unix98
2647 alphasort64 NEW: LFS
2648 argp_err_exit_status NEW: argp, GNU ext
2649 argp_error NEW: argp, GNU ext
2650 argp_failure NEW: argp, GNU ext
2651 argp_help NEW: argp, GNU ext
2652 argp_parse NEW: argp, GNU ext
2653 argp_program_bug_address NEW: argp, GNU ext
2654 argp_program_version NEW: argp, GNU ext
2655 argp_program_version_hook NEW: argp, GNU ext
2656 argp_state_help NEW: argp, GNU ext
2657 argp_usage NEW: argp, GNU ext
2658 authdes_create NEW: Secure RPC
2659 authdes_getucred NEW: Secure RPC
2660 authdes_pk_create NEW: Secure RPC
2661 backtrace NEW: GNU ext.
2662 backtrace_symbols NEW: GNU ext.
2663 backtrace_symbols_fd NEW: GNU ext.
2665 cacosf NEW: ISO C 9x
2666 cacosh NEW: ISO C 9x
2667 cacoshf NEW: ISO C 9x
2668 cacoshl NEW: ISO C 9x
2669 cacosl NEW: ISO C 9x
2676 casinf NEW: ISO C 9x
2677 casinh NEW: ISO C 9x
2678 casinhf NEW: ISO C 9x
2679 casinhl NEW: ISO C 9x
2680 casinl NEW: ISO C 9x
2682 catanf NEW: ISO C 9x
2683 catanh NEW: ISO C 9x
2684 catanhf NEW: ISO C 9x
2685 catanhl NEW: ISO C 9x
2686 catanl NEW: ISO C 9x
2687 cbc_crypt NEW: Secure RPC
2691 ccoshf NEW: ISO C 9x
2692 ccoshl NEW: ISO C 9x
2698 cimagf NEW: ISO C 9x
2699 cimagl NEW: ISO C 9x
2700 clearerr_locked REMOVED
2701 clntunix_create NEW: sunrpc ext
2703 clog10 NEW: ISO C 9x
2704 clog10f NEW: ISO C 9x
2705 clog10l NEW: ISO C 9x
2715 cprojf NEW: ISO C 9x
2716 cprojl NEW: ISO C 9x
2718 crealf NEW: ISO C 9x
2719 creall NEW: ISO C 9x
2724 csinhf NEW: ISO C 9x
2725 csinhl NEW: ISO C 9x
2728 csqrtf NEW: ISO C 9x
2729 csqrtl NEW: ISO C 9x
2733 ctanhf NEW: ISO C 9x
2734 ctanhl NEW: ISO C 9x
2736 des_setparity NEW: Secure RPC
2737 ecb_crypt NEW: Secure RPC
2738 endutxent NEW: Unix98
2740 exp10f NEW: ISO C 9x
2741 exp10l NEW: ISO C 9x
2745 fattach NEW: STREAMS
2746 fdetach NEW: STREAMS
2750 feclearexcept NEW: ISO C 9x
2751 fegetenv NEW: ISO C 9x
2752 fegetexceptflag NEW: ISO C 9x
2753 fegetround NEW: ISO C 9x
2754 feholdexcept NEW: ISO C 9x
2756 feraiseexcept NEW: ISO C 9x
2757 ferror_locked REMOVED
2758 fesetenv NEW: ISO C 9x
2759 fesetexceptflag NEW: ISO C 9x
2760 fesetround NEW: ISO C 9x
2761 fetestexcept NEW: ISO C 9x
2762 feupdateenv NEW: ISO C 9x
2763 fflush_locked REMOVED
2767 fgets_unlocked NEW: GNU ext.
2768 fileno_locked REMOVED
2780 fputc_locked REMOVED
2781 fputs_unlocked NEW: GNU ext.
2782 fread_unlocked NEW: GNU ext.
2787 fstatvfs NEW: Unix98
2791 ftruncate64 NEW: LFS
2793 fwrite_unlocked NEW: GNU ext.
2794 gai_strerror NEW: IPv6
2798 getchar_locked REMOVED
2800 getdate_err NEW: Unix98
2801 getdate_r NEW: GNU ext.
2803 getnameinfo NEW: IPv6
2804 getnetname NEW: Secure RPC
2805 getpmsg NEW: STREAMS
2806 getpt NEW: Unix98 PTY
2807 getrlimit64 NEW: LFS
2808 getutxent NEW: Unix98
2809 getutxid NEW: Unix98
2810 getutxline NEW: Unix98
2811 glob64 NEW: GNU ext.
2812 globfree64 NEW: GNU ext.
2813 gnu_get_libc_release NEW: GNU ext.
2814 gnu_get_libc_version NEW: GNU ext.
2815 grantpt NEW: Unix98 PTY
2816 host2netname NEW: Secure RPC
2818 iconv_close NEW: iconv
2819 iconv_open NEW: iconv
2820 if_freenameindex NEW: IPv6
2821 if_indextoname NEW: IPv6
2822 if_nameindex NEW: IPv6
2823 if_nametoindex NEW: IPv6
2824 in6addr_any NEW: IPv6
2825 in6addr_loopback NEW: IPv6
2826 inet6_isipv4mapped NEW: IPv6
2827 isastream NEW: STREAMS
2828 iswblank NEW: GNU ext.
2829 key_decryptsession NEW: Secure RPC
2830 key_decryptsession_pk NEW: Secure RPC
2831 key_encryptsession NEW: Secure RPC
2832 key_encryptsession_pk NEW: Secure RPC
2833 key_gendes NEW: Secure RPC
2834 key_get_conv NEW: Secure RPC
2835 key_secretkey_is_set NEW: Secure RPC
2836 key_setnet NEW: Secure RPC
2837 key_setsecret NEW: Secure RPC
2838 llrint NEW: ISO C 9x
2839 llrintf NEW: ISO C 9x
2840 llrintl NEW: ISO C 9x
2841 llround NEW: ISO C 9x
2842 llroundf NEW: ISO C 9x
2843 llroundl NEW: ISO C 9x
2848 lrintf NEW: ISO C 9x
2849 lrintl NEW: ISO C 9x
2850 lround NEW: ISO C 9x
2851 lroundf NEW: ISO C 9x
2852 lroundl NEW: ISO C 9x
2854 makecontext NEW: Unix98
2855 mempcpy NEW: GNU ext.
2858 modify_ldt NEW: kernel
2862 nearbyint NEW: ISO C 9x
2863 nearbyintf NEW: ISO C 9x
2864 nearbyintl NEW: ISO C 9x
2865 netname2host NEW: Secure RPC
2866 netname2user NEW: Secure RPC
2867 nexttoward NEW: ISO C 9x
2868 nexttowardf NEW: ISO C 9x
2869 nexttowardl NEW: ISO C 9x
2873 passwd2des NEW: Secure RPC
2875 pow10f NEW: GNU ext.
2876 pow10l NEW: GNU ext.
2879 printf_size NEW: GNU ext.
2880 printf_size_info NEW: GNU ext.
2881 profil_counter REMOVED
2882 pthread_mutexattr_getkind_np REPLACED
2883 pthread_mutexattr_setkind_np REPLACED
2884 ptsname NEW: Unix98 PTY
2885 ptsname_r NEW: Unix98 PTY
2887 putchar_locked REMOVED
2888 putgrent NEW: GNU ext.
2890 putpmsg NEW: STREAMS
2891 pututxline NEW: Unix98
2895 readdir64_r NEW: LFS
2896 remquo NEW: ISO C 9x
2897 remquof NEW: ISO C 9x
2898 remquol NEW: ISO C 9x
2900 roundf NEW: ISO C 9x
2901 roundl NEW: ISO C 9x
2903 scalbln NEW: ISO C 9x
2904 scalblnf NEW: ISO C 9x
2905 scalblnl NEW: ISO C 9x
2907 sendfile NEW: kernel
2908 setcontext NEW: Unix98
2909 setrlimit64 NEW: LFS
2910 setutxent NEW: Unix98
2912 sigignore NEW: Unix98
2913 sigqueue NEW: POSIX.1b
2914 sigrelse NEW: Unix98
2915 sigset NEW: POSIX.1b
2916 sigtimedwait NEW: POSIX.1b
2917 sigwaitinfo NEW: POSIX.1b
2918 sincos NEW: GNU ext.
2919 sincosf NEW: GNU ext.
2920 sincosl NEW: GNU ext.
2924 strcasestr NEW: GNU ext.
2925 strtoimax NEW: ISO C 9x
2926 strtoumax NEW: ISO C 9x
2927 strverscmp NEW: GNU ext.
2928 svcauthdes_stats NEW: Secure RPC
2929 svcunix_create NEW: sunrpc etx
2930 svcunixfd_create NEW: sunrpc ext
2931 swapcontext NEW: Unix98
2932 tcgetsid NEW: Unix98 PTY
2933 tdestroy NEW: GNU ext.
2934 tgamma NEW: ISO C 9x
2935 tgammaf NEW: ISO C 9x
2936 tgammal NEW: ISO C 9x
2940 truncf NEW: ISO C 9x
2941 truncl NEW: ISO C 9x
2943 unlockpt NEW: Unix98 PTY
2944 updwtmpx NEW: Unix98
2945 user2netname NEW: Secure RPC
2946 utmpxname NEW: Unix98
2947 versionsort NEW: GNU ext.
2948 versionsort64 NEW: GNU ext.
2950 wcscasecmp NEW: GNU ext.
2951 wcsncasecmp NEW: GNU ext.
2952 wcsnlen NEW: GNU ext.
2953 wcstoimax NEW: ISO C 9x
2954 wcstoll NEW: ISO C 9x
2955 wcstoull NEW: ISO C 9x
2956 wcstoumax NEW: ISO C 9x
2958 wordexp NEW: POSIX.2
2959 wordfree NEW: POSIX.2
2960 write_profiling REMOVED
2961 xdecrypt NEW: Secure RPC
2962 xdr_authdes_cred NEW: Secure RPC
2963 xdr_authdes_verf NEW: Secure RPC
2964 xdr_cryptkeyarg NEW: Secure RPC
2965 xdr_cryptkeyarg2 NEW: Secure RPC
2966 xdr_cryptkeyres NEW: Secure RPC
2967 xdr_getcredres NEW: Secure RPC
2968 xdr_key_netstarg NEW: Secure RPC
2969 xdr_key_netstres NEW: Secure RPC
2970 xdr_keybuf NEW: Secure RPC
2971 xdr_keystatus NEW: Secure RPC
2972 xdr_netnamestr NEW: Secure RPC
2973 xdr_sizeof NEW: Secure RPC
2974 xdr_unixcred NEW: sunrpc ext
2975 xencrypt NEW: Secure RPC
2976 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2987 * inet_ntoa is thread-safe
2989 * updwtmp is moved from libutil to libc
2991 * rewrite of cbrt function
2993 * update of timezone data
3007 * add atoll function
3009 * fix complex problems in Berkeley DB code
3011 * fix math functions
3015 * fixed lots of header problems (especially Linux/GNU specific)
3017 * dynamic loader preserves all registers
3019 * Roland McGrath provided support for handling of auxiliary objects in
3020 the ELF dynamic loader.
3022 * support for parallel builds is improved
3026 * GNU extensions are no longer declared by default. To enable them you
3027 must define the macro `_GNU_SOURCE' in your program or compile with
3030 * The library has changed from using GNU ld symbol aliases to using weak
3031 symbols where available. The ELF object file format supports weak
3032 symbols; GNU ld also supports weak symbols in the a.out format. (There
3033 is also now support for other GNU ld extensions in ELF. Use the
3034 `--with-elf' option to configure to indicate you have ELF, and
3035 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3036 of many files which contained only symbol aliases, reducing the size of
3037 the source and the compiled library; many other files were renamed to
3038 less cryptic names previously occupied by the symbol alias files.
3039 There is a new header file <elf.h> for programs which operate on
3040 files in the ELF format.
3042 * Converted to Autoconf version 2, so `configure' has more options.
3043 Run `configure --help' to see the details.
3045 * The library can now be configured to build profiling, highly-optimized
3046 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3047 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3048 `configure' enable building these extra libraries. The shared library is
3049 built by default when using both ELF and GNU ld. When shared libraries
3050 are enabled, the new library `-ldl' is available for arbitrary run-time
3051 loading of shared objects; its interface is defined in <dlfcn.h>. The
3052 new header file <link.h> gives access to the internals of the run-time
3053 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3054 application of same name on other systems and it provides information
3055 about dynamically linked binaries.
3057 * The C library now provides the run-time support code for profiling
3058 executables compiled with `-pg'. Programs can control the profiling code
3059 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3060 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3061 the support for this file format was contributed by David Mosberger-Tang.
3063 * The math code has been replaced with a math library based on fdlibm from
3064 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3065 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3066 functions. The math functions now reside in a separate library, so
3067 programs using them will need to use `-lm' their linking commands.
3069 * John C. Bowman contributed optimized ix87 assembler inline functions.
3071 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3072 mechanism similar to that found in Solaris 2. This is now used for the
3073 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3074 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3075 file controls what services are used for each individual database. This
3076 works by loading shared libraries with names specified in `nsswitch.conf',
3077 so service modules can be changed or added at any time without even
3078 relinking any program. Currently there are the file, db, and NIS based
3079 NSS services available.
3081 * The new functions `strtoq' and `strtouq' parse integer values from
3082 strings, like `strtol' and `strtoul', but they return `long long int' and
3083 `unsigned long long int' values, respectively (64-bit quantities).
3085 * The new functions `strtof' and `strtold' parse floating-point values from
3086 strings, like `strtod', but they return `float' and `long double' values,
3087 respectively (on some machines `double' and `long double' are the same).
3089 * Ulrich Drepper has contributed new implementations of the floating-point
3090 printing and reading code used in the `printf' family of functions and
3091 `strtod', `strtof', and `strtold'. These new functions are perfectly
3092 accurate, and much faster than the old ones.
3094 * The implementation of the POSIX locale model was completely rewritten by
3095 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3096 to compile the POSIX locale definition.
3098 * The former dummy implementations of the strcoll and strxfrm function are
3099 now replaced by fully functional code contributed by Ulrich Drepper. The
3100 collation information comes from the POSIX locale definitions.
3102 * The new header <langinfo.h> defines an interface for accessing
3103 various locale-dependent data (using the locale chosen with `setlocale').
3105 * Ulrich Drepper has contributed a new suite of functions for operation on
3106 wide-character and multibyte-character strings, in <wchar.h>;
3107 and classification and case conversion of wide characters, in <wctype.h>.
3108 These new functions are conforming to the ISO C, Amendement 1 specification.
3110 * There is now a second implementation of the standard I/O library available.
3111 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3112 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3114 * You can now use positional parameter specifications in format strings
3115 for the `printf' and `scanf' families of functions. For example,
3116 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3117 ``Number 6, Mr Jones''. This is mainly useful when providing different
3118 format strings for different languages, whose grammars may dictate
3119 different orderings of the values being printed. To support this
3120 feature, the interface for `register_printf_handler' has changed; see
3121 the header file <printf.h> for details.
3123 * The `printf' and `scanf' families of functions now understand a new
3124 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3125 to group numbers as indicated by the locale; for `scanf' and friends, this
3126 says to accept as valid only a number with all the proper grouping
3127 separators in the right places. In the default "C" locale, numbers are
3128 not grouped; but locales for specific countries will define the usual
3129 conventions (i.e. separate thousands with `,' in the US locale).
3131 * The pgrp functions have been regularized, slightly incompatibly but much
3132 less confusingly. The core functions are now `getpgid' and `setpgid',
3133 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3134 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3135 provided for compatibility. There is no longer an incompatible `getpgrp'
3136 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3138 * The new header file <fts.h> and suite of functions simplify programs that
3139 operate on directory trees. This code comes from 4.4 BSD.
3141 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3142 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3143 NSS scheme used in glibc.
3145 * There is a new malloc debugging hook `__memalign_hook'.
3147 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3148 `unsigned int' in <sys/types.h>. These are for compatibility only and
3149 their use is discouraged.
3151 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3152 done differently, so that it works even without GNU ld.
3154 * New function `euidaccess' checks allowed access to a file like `access',
3155 but using the effective IDs instead of the real IDs.
3157 * The time zone data files have been updated for the latest and greatest
3158 local time conventions of the countries of the world.
3160 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3163 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3164 for formatting floating-point numbers. They are provided only for
3165 compatibility; new programs should use `sprintf' instead. There are
3166 also equivalent function for the `long double' floating-point type and
3167 all functions also exist in a reentrant form.
3169 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3170 functions for maintaining the login-record files (primarily of use to
3171 system programs such as `login'), and convenient functions for
3172 allocating and initializing a pseudo-terminal (pty) device.
3174 * Ulrich Drepper has contributed new support for System V style
3175 shared memory and IPC on systems that support it.
3177 * Ulrich Drepper has contributed several miscellaneous new functions found
3178 in System V: The `hsearch' family of functions provide an effective
3179 implementation of hash tables; `a64l' and `l64a' provide a very simple
3180 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3183 * Ulrich Drepper has contributed new reentrant counterparts for the
3184 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3186 * Ulrich Drepper has contributed new, highly-optimized versions of several
3187 string functions for the i486/Pentium family of processors.
3189 * Ulrich Drepper has updated the Linux-specific code, based largely
3190 on work done in Hongjiu Lu's version of GNU libc for Linux.
3191 The GNU library now supports Linux versions 2.0.10 and later,
3192 using the ELF object file format (i[3456]86-*-linux).
3194 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3196 * David Mosberger-Tang and Richard Henderson have ported the C library
3197 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3198 dynamic linking support for ELF/Alpha.
3200 * Richard Henderson contributed several Alpha optimized assembler function
3201 for arithmetic and string handling.
3203 * Ulrich Drepper has contributed a new set of message catalog functions to
3204 support multiple languages using the <libintl.h> interface, for use with
3205 his new package GNU gettext. Translation volunteers have contributed
3206 catalogs of the library's messages in Spanish, German, and Korean.
3208 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3209 program and the `catgets' function for reading the catalog files it
3210 creates. (The <libintl.h> interface is preferred; we include the
3211 <nl_types.h> interface using `catgets' only for source compatibility with
3212 programs already written to use it.)
3214 * New header file <values.h> gives SVID-compatible names for <limits.h>
3217 * Various new macros, declarations, and small header files for compatibility
3220 * New function `group_member' is a convenient way to check if a process has
3221 a given effective group ID.
3223 * When using GCC 2.7 and later, the socket functions are now declared in a
3224 special way so that passing an argument of type `struct sockaddr_in *',
3225 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3226 `struct sockaddr *' type, does not generate a type-clash warning.
3228 * New function `error' declared in header file <error.h> is a convenient
3229 function for printing error messages and optionally exiting; this is the
3230 canonical function used in GNU programs. The new functions `err', `warn',
3231 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3232 doing the same thing.
3234 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3235 POSIX.2 `glob' function to do ~ and {...} expansion.
3237 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3238 `clearenv' which is used in POSIX.9 is also available.
3240 * New function `getsid' returns session ID number on systems that support it.
3242 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3243 files <db.h> and <mpool.h> provide a rich set of functions for several
3244 types of simple databases stored in memory and in files, and <ndbm.h> is
3245 an old `ndbm'-compatible interface using the `db' functions. Link with
3246 `-ldb' to get these functions.
3248 * New macro `strdupa' copies a string like `strdup', but uses local stack
3249 space from `alloca' instead of dynamic heap space from `malloc'.
3251 * New function `strnlen' is like `strlen' but searches only a given maximum
3252 number of characters for the null terminator. `stpncpy', `strndup' and
3253 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3256 * New function `statfs' in header <sys/statfs.h>.
3258 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3259 provide convenient functions for operating on blocks of null-terminated
3262 * A new suite of functions in <utmp.h> handle all the details of reading
3263 and writing the utmp file.
3265 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3268 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3269 completely POSIX compliant and also implemented the extended functionality
3270 to handle alternate digit representation and alternate era date formats.
3272 * Ulrich Drepper provided an implementation of the `strptime' function
3273 defined in XPG4.2 which transforms a string into a `struct tm' value.
3275 * Paul Eggert provided the tzselect shell script as part of the timezone
3276 code. The shell script makes it easy to select the correct timezone
3279 * The implementation of the malloc family of functions is completely replaced
3280 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3281 The implementation uses the mmap function (if available) and it is
3282 optimized for the use in multi threaded programs.
3284 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3285 functions. This new functionality is usable by specifying a special
3286 salt string and it is compatible with implementation on *BSD systems.
3288 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3289 `getsubopt' to handle second level command line options, `bsd_signal'
3290 to access BSD style `signal' functionality, the obsolete `regexp' style
3293 * the `lchown' function is available on system which support this
3296 * The implementation of the shadow password handling function was contributed
3299 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3301 * POSIX.1g support was added. The <sys/select.h> header is available,
3302 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3303 implementation of `getaddrinfo'.
3307 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3309 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3310 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3312 * The new function `daemon' from 4.4 BSD is useful for server programs that
3313 want to put themselves in the background.
3315 * Joel Sherrill has contributed support for several standalone boards that
3316 run without an operating system.
3318 * `printf', `scanf' and friends now accept a `q' type modifier for long
3319 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3321 * All of the code taken from BSD (notably most of the math and networking
3322 routines) has been updated from the BSD 4.4-Lite release.
3324 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3326 * The new functions `getdomainname' and `setdomainname' fetch or change the
3327 YP/NIS domain name. These are system calls which exist on systems which
3330 * The time zone data files have been updated for the latest international
3333 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3334 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3338 * The C library now includes support for Sun RPC, from Sun's free
3339 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3340 are included. (There is still no support for YP.)
3342 * Tom Quinn has contributed a port of the C library to SGI machines running
3343 Irix 4 (mips-sgi-irix4).
3345 * The new `lockf' function is a simplified interface to the locking
3346 facilities of `fcntl', included for compatibility.
3348 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3350 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3352 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3355 * The error message printed by `assert' for a failed assertion now includes
3356 the name of the program (if using GNU ld) and the name of the calling
3357 function (with versions of GCC that support this).
3359 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3361 * The library now includes the <sys/mman.h> header file and memory
3362 management functions `mmap', `munmap', `mprotect', `msync', and
3363 `madvise', on systems that support those facilities.
3365 * The interface for `mcheck' has changed slightly: the function called to
3366 abort the program when an allocation inconsistency is detected now takes
3367 an argument that indicates the type of failure. The new function
3368 `mprobe' lets you request a consistency check for a particular block at
3369 any time (checks are normally done only when you call `free' or `realloc'
3372 * It is now possible to easily cross-compile the C library, building on one
3373 system a library to run on another machine and/or operating system. All
3374 you need to do is set the variable `HOST_CC' in `configparms' to the
3375 native compiler for programs to run on the machine you are building on (a
3376 few generator programs are used on Unix systems); set `CC' to the
3379 * The new function `fexecve' (only implemented on the GNU system) executes
3380 a program file given a file descriptor already open on the file.
3384 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3385 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3387 * You can set the variable `libprefix' in `configparms' to specify a prefix
3388 to be prepended to installed library files; this makes it easy to install
3389 the GNU C library to be linked as `-lgnuc' or whatever.
3391 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3392 copies a limited number of characters from a string, and returns the
3393 address of the last character written.
3395 * You no longer need to check for whether the installed `stddef.h' is
3396 compatible with the GNU C library. configure now checks for you.
3398 * You can now define a per-stream `fileno' function to convert the
3399 stream's cookie into an integral file descriptor.
3401 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3402 allocates zero bytes of storage, and returns a unique pointer which
3403 you can pass to `realloc' or `free'. The behavior is undefined if
3404 you dereference this pointer.
3406 * The C library now runs on Sony NEWS m68k machines running either
3407 NewsOS 3 or NewsOS 4.
3409 * The new `syscall' function is a system-dependent primitive function
3410 for invoking system calls. It has the canonical behavior on Unix
3411 systems, including unreliable return values for some calls (such as
3412 `pipe', `fork' and `getppid').
3414 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3415 to `EAGAIN', which is the preferred name. On systems whose kernels
3416 use two distinct codes, the C library now translates EWOULDBLOCK to
3417 EAGAIN in every system call function.
3421 * The GNU C Library Reference Manual is now distributed with the library.
3422 `make dvi' will produce a DVI file of the printed manual.
3423 `make info' will produce Info files that you can read on line using C-h i
3424 in Emacs or the `info' program.
3425 Please send comments on the manual to bug-glibc-manual@gnu.org.
3427 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3429 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3431 * Jason Merrill has contributed a port to the Sequent Symmetry running
3432 Dynix version 3 (i386-sequent-dynix).
3434 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3435 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3437 * New function `memory_warnings' lets you arrange to get warnings when
3438 malloc is running out of memory to allocate, like Emacs gives you.
3440 * The C library now contains the relocating allocator used in Emacs 19 for
3441 its editing buffers. This allocator (ralloc) minimizes allocation
3442 overhead and fragmentation by moving allocated regions around whenever it
3443 needs to. You always refer to a ralloc'd region with a "handle" (a
3444 pointer to a pointer--an object of type `void **').
3446 * There is a new `printf' format: `%m' gives you the string corresponding
3447 to the error code in `errno'.
3449 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3450 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3451 pass, the `a' modifier says to fill in a `char **' you pass with a
3454 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3455 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3456 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3458 * `mkstemp' is a traditional Unix function to atomically create and open a
3459 uniquely-named temporary file.
3463 * The standard location for the file that says what the local timezone is
3464 has changed again. It is now `/usr/local/etc/localtime' (or more
3465 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3467 * The distribution no longer contains any files with names longer than 14
3470 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3471 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3473 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3477 * The configuration process has changed quite a bit. The `configure'
3478 script is now used just like the configuration scripts for other GNU
3479 packages. The `sysdeps' directory hierarchy is much rearranged.
3480 The file `INSTALL' explains the new scheme in detail.
3482 * The header files no longer need to be processed into ANSI C and
3483 traditional C versions. There is just one set of files to install, and
3484 it will work with ANSI or old C compilers (including `gcc -traditional').
3486 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3487 MIPS DECStation running Ultrix 4.
3489 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3490 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3491 made itself into a shared library.
3493 * Yet further improved support for the i386, running 4.3 BSD-like systems
3494 (such as Mach 3 with the Unix single-server), or System V.
3496 * New function `strncasecmp' to do case-insensitive string comparison
3497 with limited length.
3499 * New function `strsep' is a reentrant alternative to `strtok'.
3501 * New functions `scandir' and `alphasort' for searching directories.
3503 * New function `setenv' is a better interface to `putenv'.
3505 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3506 function for traversing a directory tree.
3508 * The GNU obstack package is now also part of the C library.
3509 The new function `open_obstack_stream' creates a stdio stream that
3510 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3511 formatted output directly to an obstack.
3513 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3514 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3516 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3518 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3519 things to your strings.
3521 * There are some new test programs: test-fseek, testmb, and testrand.
3523 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3524 These ports are not finished, but are a good starting place for really
3525 supporting those systems.
3527 * `/etc/localtime' is now the standard location for the file that says what
3528 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3529 This follows the general principle that `/etc' is the place for all local
3530 configuration files.
3532 * The C library header files now use `extern "C"' when used by the C++
3533 compiler, so the C library should now work with C++ code.
3535 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3536 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3539 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3540 function which uses the `merge sort' algorithm, and is said to be
3541 significantly faster than the old GNU `qsort' function. Merge sort is
3542 now the standard `qsort' function. The new algorithm can require a lot
3543 of temporary storage; so, the old sorting function is called when the
3544 required storage is not available.
3546 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3547 provides the Unix `crypt' function, plus some other entry points.
3549 * The code and header files taken from 4.4 BSD have been updated with the
3550 latest files released from Berkeley.
3552 ----------------------------------------------------------------------
3553 Copyright information:
3555 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3557 Permission is granted to anyone to make or distribute verbatim copies
3558 of this document as received, in any medium, provided that the
3559 copyright notice and this permission notice are preserved,
3560 thus giving the recipient permission to redistribute in turn.
3562 Permission is granted to distribute modified versions
3563 of this document, or of portions of it,
3564 under the above conditions, provided also that they
3565 carry prominent notices stating who last changed them.
3568 version-control: never