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 - llogb functions: the llogb, llogbf and llogbl functions, and the
70 FP_LLOGB0 and FP_LLOGBNAN macros.
72 - Comparison macros: iseqsig.
74 - Classification macros: iscanonical, issubnormal, iszero.
76 - Total order functions: totalorder, totalorderf, totalorderl,
77 totalordermag, totalordermagf, totalordermagl.
79 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
81 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
82 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
84 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
85 are added to libc. They convert a floating-point number into string.
87 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
88 intended to be used instead of memset() to erase sensitive data after use;
89 the compiler will not optimize out calls to explicit_bzero even if they
90 are "unnecessary" (in the sense that no _correct_ program can observe the
91 effects of the memory clear).
93 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
94 to float instead of double. This does not affect the ABI of any libraries
95 that are part of the GNU C Library, but may affect the ABI of other
96 libraries that use this type in their interfaces.
98 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
99 float_t and double_t types are now defined to long double instead of float
100 and double. These options are not the default, and this does not affect
101 the ABI of any libraries that are part of the GNU C Library, but it may
102 affect the ABI of other libraries that use this type in their interfaces,
103 if they are compiled or used with those options.
105 * The getentropy and getrandom functions, and the <sys/random.h> header file
108 * The buffer size for byte-oriented stdio streams is now limited to 8192
109 bytes by default. Previously, on Linux, the default buffer size on most
110 file systems was 4096 bytes (and thus remains unchanged), except on
111 network file systems, where the buffer size was unpredictable and could be
112 as large as several megabytes.
114 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
115 for the Linux quota interface which predates kernel version 2.4.22 has
118 * The malloc_get_state and malloc_set_state functions have been removed.
119 Already-existing binaries that dynamically link to these functions will
120 get a hidden implementation in which malloc_get_state is a stub. As far
121 as we know, these functions are used only by GNU Emacs and this change
122 will not adversely affect already-built Emacs executables. Any undumped
123 Emacs executables, which normally exist only during an Emacs build, should
124 be rebuilt by re-running “./configure; make” in the Emacs build tree.
126 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
127 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
128 “no-ip6-dotint” had already been the default, and support for the
129 “ip6-dotint” option was removed from the Internet in 2006.
131 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
132 flag from <resolv.h> have been removed. The option relied on a
133 backwards-incompatible DNS extension which was never deployed on the
136 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
137 in the <resolv.h> header file have been deprecated. They were already
140 * DNSSEC-related declarations and definitions have been removed from the
141 <arpa/nameser.h> header file, and libresolv will no longer attempt to
142 decode the data part of DNSSEC record types. Previous versions of glibc
143 only implemented minimal support for the previous version of DNSSEC, which
144 is incompatible with the currently deployed version.
146 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
147 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
148 <arpa/nameser.h> header file because the distinction between RR types and
149 meta-RR types is not officially standardized, subject to revision, and
150 thus not suitable for encoding in a macro.
152 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
153 and rhook members of the res_state type in <resolv.h> have been removed.
154 The glibc stub resolver did not support these hooks, but the header file
155 did not reflect that.
157 * For multi-arch support it is recommended to use a GCC which has
158 been built with support for GNU indirect functions. This ensures
159 that correct debugging information is generated for functions
160 selected by IFUNC resolvers. This support can either be enabled by
161 configuring GCC with '--enable-gnu-indirect-function', or by
162 enabling it by default by setting 'default_gnu_indirect_function'
163 variable for a particular architecture in the GCC source file
166 * GDB pretty printers have been added for mutex and condition variable
167 structures in POSIX Threads. When installed and loaded in gdb these pretty
168 printers show various pthread variables in human-readable form when read
169 using the 'print' or 'display' commands in gdb.
171 Security related changes:
173 On ARM EABI (32-bit), generating a backtrace for execution contexts which
174 have been created with makecontext could fail to terminate due to a
175 missing .cantunwind annotation. This has been observed to lead to a hang
176 (denial of service) in some Go applications compiled with gccgo. Reported
177 by Andreas Schwab. (CVE-2016-6323)
179 The following bugs are resolved with this release:
181 [The release manager will add the list generated by
182 scripts/list-fixed-bugs.py just before the release.]
186 * The minimum Linux kernel version that this version of the GNU C Library
187 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
188 kernel version 2.6.32 or later suffices (on architectures that already
189 required kernel versions more recent than 3.2, those requirements remain
190 unchanged). Linux 3.2 or later kernel headers are required on all
193 * The pap_AN locale has been deleted. This has been deprecated for a long
194 time. It has been replaced by pap_AW & pap_CW, both of which have long
195 been included in previous releases.
197 * The readdir_r and readdir64_r functions have been deprecated. It is
198 recommended to use readdir and readdir64 instead.
200 * The type “union wait” has been removed. It was deprecated in the early
201 1990s and never part of POSIX. Application code should use the int type
202 instead of “union wait”.
204 * A new NSS action is added to facilitate large distributed system
205 administration. The action, MERGE, allows remote user stores like LDAP
206 to be merged into local user stores like /etc/groups in order to provide
207 easy to use, updated, and managed sets of merged credentials. The new
208 action can be used by configuring it in /etc/nsswitch.conf:
209 group: files [SUCCESS=merge] nis
210 Implemented by Stephen Gallagher (Red Hat).
212 * The deprecated __malloc_initialize_hook variable has been removed from the
215 * The long unused localedef --old-style option has been removed. It hasn't
216 done anything in over 16 years. Scripts using this option can safely
219 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
220 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
221 return the next representable value in the direction of positive infinity
222 and the nextdown functions return the next representable value in the
223 direction of negative infinity. These are currently enabled as GNU
226 Security related changes:
228 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
229 could result in a stack overflow when getnetbyname was called with an
230 overly long name. (CVE-2016-3075)
232 * Previously, getaddrinfo copied large amounts of address data to the stack,
233 even after the fix for CVE-2013-4458 has been applied, potentially
234 resulting in a stack overflow. getaddrinfo now uses a heap allocation
235 instead. Reported by Michael Petlan. (CVE-2016-3706)
237 * The glob function suffered from a stack-based buffer overflow when it was
238 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
239 Reported by Alexander Cherepanov. (CVE-2016-1234)
241 * The Sun RPC UDP client could exhaust all available stack space when
242 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
243 alloca plugin for GCC. (CVE-2016-4429)
245 * The IPv6 name server management code in libresolv could result in a memory
246 leak for each thread which is created, performs a failing naming lookup,
247 and exits. Over time, this could result in a denial of service due to
248 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
250 The following bugs are resolved with this release:
252 [1170] localedata: ne_NP: update Nepali locale definition file
253 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
255 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
256 [6796] math: fdim() does not set errno on overflow
257 [10354] libc: posix_spawn should use vfork() in more cases than presently
258 [11213] localedata: localedata: add copyright disclaimer to locale files
259 [12143] localedata: chr_US: new Cherokee locale
260 [12450] localedata: sgs_LT: new locale
261 [12676] localedata: ln_CD: new locale
262 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
264 [13304] math: fma, fmaf, fmal produce wrong results
265 [14259] build: --localedir arg to configure is ignored
266 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
267 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
268 [14934] localedata: es_CL: wrong first weekday chilean locale
269 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
271 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
273 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
274 [15368] nptl: raise() is not async-signal-safe
275 [15479] math: ceil, floor, round and trunc raise inexact exception
276 [15578] localedata: kk_KZ: various updates
277 [16003] localedata: pap_AN: punt old locale
278 [16137] localedata: iw_IL: punt old locale
279 [16190] localedata: eo: new esperanto locale
280 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
281 [16742] malloc: race condition: pthread_atfork() called before first
282 malloc() results in unexpected locking behaviour/deadlocks
283 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
285 [16983] localedata: postal_fmt does not allow %l and %n modifiers
286 [17565] localedata: pt_PT: wrong (work-)week start
287 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
288 [17950] build: Build fails with -msse
289 [18205] localedata: be_BY*: wrong first_weekday and first_workday
290 [18433] libc: posix_spawn does not return correctly upon failure to
292 [18453] localedata: charmaps/IBM875: incorrect codes
293 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
295 [18896] localedata: he_IL: improvements for currency
296 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
298 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
300 [19056] libc: Deprecate readdir_r
301 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
303 [19198] localedata: nl_NL: small improvements for Dutch locales
304 [19257] network: Per-thread memory leak in __res_vinit with IPv6
305 nameservers (CVE-2016-5417)
306 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
308 [19400] locale: Language missing in "iso-639.def", trivial fix in
310 [19431] malloc: Deadlock between fflush, getdelim, and fork
311 [19505] libc: Incorrect file descriptor validity checks in
312 posix_spawn_file_actions_add{open,close,dup2}
313 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
315 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
316 `intl/{gettextP,loadinfo}.h'
317 [19534] libc: execle, execlp may use malloc
318 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
319 [19573] network: res_nclose and __res_maybe_init disagree about name
320 server initialization, breaking Hesiod
321 [19575] localedata: Status of GB18030 tables
322 [19581] localedata: sr_* date_fmt string contains additional newline
323 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
325 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
326 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
327 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
328 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
330 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
332 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
334 [19610] dynamic-link: ldconfig -X removes stale symbolic links
335 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
336 [19633] locale: strfmon_l applies global locale to number formatting
337 [19642] network: Memory leak in getnameinfo
338 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
339 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
341 [19654] math: [x86_64] Need testcase for BZ #19590 fix
342 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
344 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
345 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
347 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
349 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
351 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
352 pointers and lengths in error-case.
353 [19727] locale: Converting from/to UTF-xx with iconv() does not always
354 report errors on UTF-16 surrogates values.
355 [19755] nscd: nscd assertion failure in gc
356 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
357 [19759] libc: mempcpy shouldn't be inlined
358 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
359 [19765] libc: s390 needs an optimized mempcpy
360 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
361 NAME_MAX limit assumption (CVE-2016-1234)
362 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
363 [19787] network: Missing and incorrect truncation checks in getnameinfo
364 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
366 [19791] network: Assertion failure in res_query.c with un-connectable name
368 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
369 [19822] math: libm.so install clobbers old version
370 [19825] network: resolv: send_vc can return uninitialized data in second
371 response to getaddrinfo
372 [19830] network: nss_dns: should check RDATA length against buffer length
373 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
374 confronted with A/AAAA records of invalid size
375 [19837] nss: nss_db: No retries for some long lines with a larger buffer
376 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
377 [19853] stdio: Printing IBM long double in decimal with high precision is
379 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
380 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
381 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
383 [19865] network: Assertion failure or memory leak in
384 _nss_dns_getcanonname_r
385 [19868] network: nss_dns: netent code does not skip over non-PTR records
386 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
388 [19881] string: Improve x86-64 memset
389 [19907] string: Incorrect memcpy tests
390 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
392 [19925] libc: termios.h XCASE namespace
393 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
394 [19929] libc: limits.h NL_NMAX namespace
395 [19931] stdio: Memory leak in vfprintf
396 [19957] libc: clone(CLONE_VM) access invalid parent memory
397 [19963] localedata: en_IL: New locale
398 [19989] stdio: stdio.h cuserid namespace
399 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
401 [19996] locale: langinfo.h nl_langinfo_l namespace
402 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
403 [20010] network: getaddrinfo: Stack overflow in hostent translation
405 [20012] stdio: libio: fmemopen append mode failure
406 [20014] stdio: stdio.h namespace for pre-threads POSIX
407 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
408 [20023] libc: fcntl.h timespec namespace
409 [20024] math: [x86_64] vectorized sincos trashes the stack
410 [20031] network: nss_hesiod: Heap overflow in get_txt_records
411 [20041] time: sys/time.h timespec namespace
412 [20043] libc: unistd.h missing cuserid for UNIX98 and before
413 [20044] libc: unistd.h missing pthread_atfork for UNIX98
414 [20051] libc: ttyslot in wrong header under wrong conditions
415 [20054] libc: gethostname not declared for XPG4
416 [20055] libc: termios.h missing tcgetsid for XPG4
417 [20072] dynamic-link: x86 init_cpu_features is called twice in static
419 [20073] libc: sys/stat.h fchmod namespace
420 [20074] libc: stdlib.h rand_r namespace
421 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
422 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
424 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
425 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
427 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
428 [20119] libc: Wrong mask for processors level type from CPUID
429 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
430 AS not supporting AVX512
431 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
433 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
434 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
436 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
437 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
438 [20178] libc: posix_spawn{p} should not call exit
439 [20191] stdio: libio: vtables hardening
440 [20195] string: FMA4 detection requires CPUID execution with register
442 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
443 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
445 [20212] math: acos (sNaN) returns sNaN
446 [20213] math: asin (sNaN) returns sNaN
447 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
448 [20218] math: [i386] asinhl (sNaN) returns sNaN
449 [20219] math: [i386] atanhl (sNaN) returns sNaN
450 [20222] stdio: fopencookie: Mangle function pointers
451 [20224] math: [i386] cbrtl (sNaN) returns sNaN
452 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
453 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
455 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
456 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
457 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
458 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
459 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
460 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
461 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
462 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
463 [20240] math: modf (sNaN) returns sNaN
464 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
465 [20250] math: frexp (sNaN) returns sNaN
466 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
467 [20255] math: [i386] fdim, fdimf return with excess range and precision /
469 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
470 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
471 subscript is below array bounds [-Werror=array-bounds]
472 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
473 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
475 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
476 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
478 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
480 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
481 [20316] localedata: id_ID: Februari instead of Pebruari
482 [20327] string: POWER8 strcasecmp returns incorrect result
483 [20347] math: Failure: Test: j0_downward (0xap+0)
484 [20348] libc: FAIL: misc/tst-preadvwritev64
485 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
486 [20350] libc: There is no test for p{read,write}64
487 [20357] math: Incorrect cos result for 1.5174239687223976
488 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
492 * Unicode 8.0.0 Support: Character encoding, character type info, and
493 transliteration tables are all updated to Unicode 8.0.0, using new
494 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
495 These updates cause user visible changes, such as the fixes for bugs
496 89, 16061, and 18568.
498 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
499 kernel-internal CPU set size. This means that requests that change the
500 CPU affinity which failed before (for example, an all-ones CPU mask) will
501 now succeed. Applications that need to determine the effective CPU
502 affinities need to call sched_getaffinity or pthread_getaffinity_np after
503 setting it because the kernel can adjust it (and the previous size check
504 would not detect this in the majority of cases).
506 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
507 the following new symbols are used: fts64_children, fts64_close,
508 fts64_open, fts64_read and fts64_set.
510 * getaddrinfo now detects certain invalid responses on an internal netlink
511 socket. If such responses are received, an affected process will
512 terminate with an error message of "Unexpected error <number> on netlink
513 descriptor <number>" or "Unexpected netlink response of size <number> on
514 descriptor <number>". The most likely cause for these errors is a
515 multi-threaded application which erroneously closes and reuses the netlink
516 file descriptor while it is used by getaddrinfo.
518 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
519 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
520 unnecessary serialization of memory allocation requests across threads.
521 The defect is now corrected. Users should see a substantial increase in
522 the concurent throughput of allocation requests for applications which
523 trigger this bug. Affected applications typically create and
524 destroy threads frequently. (Bug 19048 was reported and analyzed by
527 * There is now a --disable-timezone-tools configure option for disabling the
528 building and installing of the timezone related utilities (zic, zdump, and
529 tzselect). This is useful for people who build the timezone data and code
530 independent of the GNU C Library.
532 * The obsolete header <regexp.h> has been removed. Programs that require
533 this header must be updated to use <regex.h> instead.
535 * The obsolete functions bdflush, create_module, get_kernel_syms,
536 query_module and uselib are no longer available to newly linked binaries;
537 the header <sys/kdaemon.h> has been removed. These functions and header
538 were specific to systems using the Linux kernel and could not usefully be
539 used with the GNU C Library on systems with version 2.6 or later of the
542 * Optimized string, wcsmbs and memory functions for IBM z13.
543 Implemented by Stefan Liebler.
545 * Newly linked programs that define a variable called signgam will no longer
546 have it set by the lgamma, lgammaf and lgammal functions. Programs that
547 require signgam to be set by those functions must ensure that they use the
548 variable provided by the GNU C Library and declared in <math.h>, without
549 defining their own copy.
551 * The minimum GCC version that can be used to build this version of the GNU
552 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
553 still be used to compile programs using the GNU C Library.
555 Security related changes:
557 * An out-of-bounds value in a broken-out struct tm argument to strftime no
558 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
560 * The LD_POINTER_GUARD environment variable can no longer be used to disable
561 the pointer guard feature. It is always enabled. Previously,
562 LD_POINTER_GUARD could be used to disable security hardening in binaries
563 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
566 * An integer overflow in hcreate and hcreate_r could lead to an
567 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
569 * The catopen function no longer has unbounded stack usage. Reported by
572 * The nan, nanf and nanl functions no longer have unbounded stack usage
573 depending on the length of the string passed as an argument to the
574 functions. Reported by Joseph Myers. (CVE-2014-9761)
576 * A stack-based buffer overflow was found in libresolv when invoked from
577 libnss_dns, allowing specially crafted DNS responses to seize control
578 of execution flow in the DNS client. The buffer overflow occurs in
579 the functions send_dg (send datagram) and send_vc (send TCP) for the
580 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
581 family. The use of AF_UNSPEC triggers the low-level resolver code to
582 send out two parallel queries for A and AAAA. A mismanagement of the
583 buffers used for those queries could result in the response of a query
584 writing beyond the alloca allocated buffer created by
585 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
586 the overflow. Thanks to the Google Security Team and Red Hat for
587 reporting the security impact of this issue, and Robert Holiday of
588 Ciena for reporting the related bug 18665. (CVE-2015-7547)
590 The following bugs are resolved with this release:
592 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
593 [887] math: Math library function "logb" and "nextafter" inconsistent
594 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
595 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
596 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
597 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
599 [4404] localedata: German translation of "Alarm clock" is misleading
600 [6799] math: nextafter() and nexttoward() doen't set errno on
601 overflow/underflow errors
602 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
604 [10432] nis: _nss_nis_setnetgrent assertion failure
605 [11460] libc: fts has no LFS support
606 [12926] network: getaddrinfo()/make_request() may spin forever
607 [13065] nptl: Race condition in pthread barriers
608 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
609 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
611 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
612 [14912] libc: Rename non-installed bits/*.h headers
613 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
614 [15367] math: Let gcc use __builtin_isinf
615 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
616 [15421] math: lgamma wrongly sets signgam for ISO C
617 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
618 FE_INVALID with argument out of range
619 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
621 [15786] dynamic-link: ifunc resolver functions can smash function
623 [15918] math: Unnecessary check for equality in hypotf()
624 [16061] localedata: Review / update transliteration data
625 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
627 [16141] time: strptime %z offset restriction
628 [16171] math: drem should be alias of remainder
629 [16296] math: fegetround is pure?
630 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
631 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
633 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
634 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
636 [16422] math: [powerpc] math-float, math-double failing llrint tests with
637 "Exception "Inexact" set" on ppc32
638 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
639 [16517] math: Missing underflow exception from tanf/tan/tanl
640 [16519] math: Missing underflow exception from sinhf
641 [16520] math: Missing underflow exception from tanhf
642 [16521] math: Missing underflow exception from exp2
643 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
645 [16734] stdio: fopen calls mmap to allocate its buffer
646 [16961] math: nan function incorrect handling of bad sequences
647 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
648 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
649 [16985] locale: localedef: confusing error message when opening output
651 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
652 [17197] locale: Redundant shift character in iconv conversion output at
654 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
655 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
656 [17250] dynamic-link: static linking breaks nss loading
657 (getaddrinfo/getpwnam/etc...)
658 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
660 [17441] math: isnan() should use __builtin_isnan() in GCC
661 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
662 (related to lock elision)
663 [17787] manual: Exponent on page 324 of the PDF ends prematurely
664 [17886] time: strptime should be able to parse "Z" as a timezone with %z
665 [17887] time: strptime should be able to parse "+01:00" style timezones
666 [17905] libc: catopen() Multiple unbounded stack allocations
668 [18084] libc: backtrace (..., 0) dumps core on x86
669 [18086] libc: nice() sets errno to 0 on success
670 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
671 is too large (CVE-2015-8778)
672 [18251] dynamic-link: SONAME missing when audit modules provides path
673 [18265] libc: add attributes for wchar string and memory functions
674 [18370] math: csqrt missing underflows
675 [18421] libc: [hppa] read-only segment has dynamic relocations
676 [18472] libc: Obsolete syscall wrappers should be compat symbols
677 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
678 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
679 [18525] localedata: Remove locale timezone information
680 [18560] libc: [powerpc] spurious bits/ipc.h definitions
681 [18568] localedata: Update locale data to Unicode 8.0
682 [18589] locale: sort-test.sh fails at random
683 [18595] math: ctan, ctanh missing underflows
684 [18604] libc: assert macro-expands its argument
685 [18610] math: S390: fetestexcept() reports any exception if DXC-code
686 contains a vector instruction exception.
687 [18611] math: j1, jn missing errno setting on underflow
688 [18618] localedata: sync Chechen locale definitions with other *_RU
690 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
691 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
692 [18665] network: In send_dg, the recvfrom function is NOT always using the
693 buffer size of a newly created buffer (CVE-2015-7547)
694 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
696 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
698 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
699 [18699] math: tilegx cproj() for various complex infinities does not yield
701 [18724] libc: Harden put*ent functions against data injection
702 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
703 [18755] build: build errors with -DNDEBUG
704 [18757] stdio: fmemopen fails to set errno on failure
705 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
707 [18781] libc: openat64 lacks O_LARGEFILE
708 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
709 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
710 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
711 [18790] math: [ldbl-128ibm] tanhl inaccurate
712 [18795] libc: stpncpy fortification misses buffer lengths that are
714 [18796] build: build fails for --disable-mathvec
715 [18803] math: hypot missing underflows
716 [18820] stdio: fmemopen may leak memory on failure
717 [18823] math: csqrt spurious underflows
718 [18824] math: fma spurious underflows
719 [18825] math: pow missing underflows
720 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
721 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
722 [18870] build: sem_open.c fails to compile with missing symbol
724 [18872] stdio: Fix memory leak in printf_positional
725 [18873] libc: posix_fallocate overflow check ineffective
726 [18875] math: Excess precision leads incorrect libm
727 [18877] libc: arm: mmap offset regression
728 [18887] libc: memory corruption when using getmntent on blank lines
729 [18918] localedata: hu_HU: change time to HH:MM:SS format
730 [18921] libc: Regression: extraneous stat() and fstat() performed by
732 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
733 binaries (CVE-2015-8777)
734 [18951] math: tgamma missing underflows
735 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
737 [18953] localedata: lt_LT: change currency symbol to the euro
738 [18956] math: powf inaccuracy
739 [18961] math: [i386] exp missing underflows
740 [18966] math: [i386] exp10 missing underflows
741 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
742 [18969] build: multiple string test failures due to missing locale
744 [18970] libc: Reference of pthread_setcancelstate in libc.a
745 [18977] math: float / long double Bessel functions not in XSI POSIX
746 [18980] math: i386 libm functions return with excess range and precision
747 [18981] math: i386 scalb*, ldexp return with excess range and precision
748 [18982] stdio: va_list and vprintf
749 [18985] time: Passing out of range data to strftime() causes a segfault
751 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
752 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
753 [19012] locale: iconv_open leaks memory on error path
754 [19016] math: clog, clog10 inaccuracy
755 [19018] nptl: Mangle function pointers in tls_dtor_list
756 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
757 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
758 [19048] malloc: malloc: arena free list can become cyclic, increasing
760 [19049] math: [powerpc] erfc incorrect zero sign
761 [19050] math: [powerpc] log* incorrect zero sign
762 [19058] math: [x86_64] Link fail with -fopenmp and -flto
763 [19059] math: nexttoward overflow incorrect in non-default rounding modes
764 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
765 [19074] network: Data race in _res_hconf_reorder_addrs
766 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
767 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
768 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
770 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
772 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
773 [19086] manual: posix_fallocate64 documented argument order is wrong.
774 [19088] math: lround, llround missing exceptions close to overflow
776 [19094] math: lrint, llrint missing exceptions close to overflow threshold
777 [19095] math: dbl-64 lrint incorrect for 64-bit long
778 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
779 [19124] dynamic-link: ld.so failed to build with older assmebler
780 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
781 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
782 [19134] math: [powerpc32] lround, lroundf spurious exceptions
783 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
784 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
785 pthread_setaffinity_np
786 [19156] math: [ldbl-128] j0l spurious underflows
787 [19164] nptl: tst-getcpu fails with many possible CPUs
788 [19168] math: math/test-ildoubl and math/test-ldouble failure
789 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
790 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
792 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
793 (FE_NOMASK_ENV) do not clear SSE exceptions
794 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
795 _int_new_arena/reused_arena
796 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
797 [19201] math: dbl-64 remainder incorrect sign of zero result
798 [19205] math: bits/math-finite.h conditions do not match math.h and
800 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
801 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
802 for C99-based standards
803 [19212] libc: features.h not -Wundef clean
804 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
806 [19214] libc: Family and model identification for AMD CPU's are incorrect.
807 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
808 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
810 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
812 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
814 [19242] libc: strtol incorrect in Turkish locales
815 [19243] malloc: reused_arena can pick an arena on the free list, leading
816 to an assertion failure and reference count corruption
817 [19253] time: tzset() ineffective when temporary TZ did not include DST
819 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
820 [19270] math: [hppa] Shared libm missing __isnanl
821 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
822 MAP_HUGETLB and MAP_STACK defines
823 [19313] nptl: Wrong __cpu_mask for x32
824 [19347] libc: grantpt: try to force a specific gid even without pt_chown
825 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
826 [19350] math: [ldbl-128ibm] sinhl spurious overflows
827 [19351] math: [ldbl-128ibm] logl inaccurate near 1
828 [19363] time: x32: times() return value wrongly truncates/sign extends
830 [19367] dynamic-link: Improve branch prediction on Silvermont
831 [19369] network: Default domain name not reset by res_ninit when "search"
832 / "domain" entry is removed from resolv.conf
833 [19375] math: powerpc: incorrect results for POWER7 logb with negative
835 [19385] localedata: bg_BG: time separator should be colon, not comma
836 [19408] libc: linux personality syscall wrapper may erroneously return an
837 error on 32-bit architectures
838 [19415] libc: dladdr returns wrong names on hppa
839 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
840 IBM905, IBM907, and IBM909
841 [19439] math: Unix98 isinf and isnan functions conflict with C++11
842 [19443] build: build failures with -DDEBUG
843 [19451] build: Make check fails on test-double-vlen2
844 [19462] libc: Glibc failed to build with -Os
845 [19465] math: Wrong code with -Os
846 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
848 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
850 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
851 [19486] math: S390: Math tests fail with "Exception Inexact set".
852 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
853 [19550] libc: [mips] mmap negative offset handling inconsistent with other
855 [19590] math: Fail to build shared objects that use libmvec.so functions.
859 * The following bugs are resolved with this release:
861 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
862 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
863 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
864 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
865 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
866 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
867 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
868 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
869 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
870 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
871 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
872 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
873 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
874 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
875 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
876 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
877 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
878 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
879 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
880 18657, 18676, 18694, 18696, 18887.
882 * Cache information can be queried via sysconf() function on s390 e.g. with
883 _SC_LEVEL1_ICACHE_SIZE as argument.
885 * A buffer overflow in gethostbyname_r and related functions performing DNS
886 requests has been fixed. If the NSS functions were called with a
887 misaligned buffer, the buffer length change due to pointer alignment was
888 not taken into account. This could result in application crashes or,
889 potentially arbitrary code execution, using crafted, but syntactically
890 valid DNS responses. (CVE-2015-1781)
892 * The time zone file parser has been made more robust against crafted time
893 zone files, avoiding heap buffer overflows related to the processing of
894 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
895 large time zone data files. Overly long time zone specifiers in the TZ
896 variable no longer result in stack overflows and crashes.
898 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
899 for LD and GD on x86 and x86-64, has been implemented. You will need
900 binutils-2.24 or later to enable this optimization.
902 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
903 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
904 Hat). These updates cause user visible changes, such as the fix for bug
907 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
908 and getXXbyYY NSS calls for the same database, causing a denial-of-service
909 condition in some applications.
911 * Added vector math library named libmvec with the following vectorized x86_64
912 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
914 The library can be disabled with --disable-mathvec. Use of the functions is
915 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
916 Shared library libmvec.so is linked in as needed when using -lm (no need to
917 specify -lmvec explicitly for not static builds).
918 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
920 * A new fmemopen implementation has been added with the goal of POSIX
921 compliance. The new implementation fixes the following long-standing
922 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
923 old implementation is still present for use be by existing binaries.
925 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
926 releases. It has been fixed to match 2.19 and older, but binaries built
927 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
929 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
930 Contributed by Roland McGrath (Google).
932 * The header <regexp.h> is deprecated, and will be removed in a future
933 release. Use of this header will trigger a deprecation warning.
934 Application developers should update their code to use <regex.h> instead.
936 This header was formerly part of SUS, but was deprecated in 1994 and
937 removed from the standard in 2001. Also, the glibc implementation
938 leaks memory. See BZ#18681 for more details.
942 * The following bugs are resolved with this release:
944 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
945 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
946 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
947 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
948 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
949 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
950 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
951 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
952 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
953 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
956 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
957 too little memory for the to-be-scanned arguments and overflow the
958 allocated buffer. The implementation now correctly computes the required
959 buffer size when using malloc, and switches to malloc from alloca as
962 * A new semaphore algorithm has been implemented in generic C code for all
963 machines. Previous custom assembly implementations of semaphore were
964 difficult to reason about or ensure that they were safe. The new version
965 of semaphore supports machines with 64-bit or 32-bit atomic operations.
966 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
967 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
969 * Port to Altera Nios II has been contributed by Mentor Graphics.
971 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
972 implementations for powerpc64/powerpc64le.
973 Implemented by Adhemerval Zanella (IBM).
975 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
976 and powerpc64le. This may improve lock scaling of existing programs on
977 HTM capable systems. The lock elision code is only enabled with
978 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
979 powerpc will issue a transaction abort on every syscall to avoid side
980 effects being visible outside transactions.
982 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
983 AArch64. Contributed by ARM Ltd.
985 * i386 memcpy functions optimized with SSE2 unaligned load/store.
987 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
988 under certain input conditions resulting in the execution of a shell for
989 command substitution when the application did not request it. The
990 implementation now checks WRDE_NOCMD immediately before executing the
991 shell and returns the error WRDE_CMDSUB as expected.
993 * CVE-2012-3406 printf-style functions could run into a stack overflow when
994 processing format strings with a large number of format specifiers.
996 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
997 infinite loop if the DNS response contained a PTR record of an unexpected
1000 * The minimum GCC version that can be used to build this version of the GNU
1001 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1002 still be used to compile programs using the GNU C Library.
1004 * The GNU C Library is now built with -Werror by default. This can be
1005 disabled by configuring with --disable-werror.
1007 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1009 * The obsolete sigvec function has been removed. This was the original
1010 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1011 programs have been using instead for about 25 years. Of course, ABI
1012 compatibility for old binaries using sigvec remains intact.
1014 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1015 with newer versions of bison.
1017 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1018 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1019 registers overlay two consecutive single-precision registers. MIPS32R2
1020 introduced a new FPU mode (FR=1) where double-precision registers extend the
1021 corresponding single-precision registers which is incompatible with the
1022 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1023 require the use of FR=1 making a transition necessary. New o32 ABI
1024 extensions enable users to migrate over time from the original o32 ABI
1025 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1026 tracks the ABI of any loaded object and verifies that new objects are
1027 compatible. Mode transitions will also be requested as required and
1028 unsupportable objects will be rejected. The ABI checks include both soft and
1029 hard float ABIs for o32, n32 and n64.
1031 GCC 5 with GNU binutils 2.25 onwards:
1032 It is strongly recommended that all o32 system libraries are built using the
1033 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1034 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1035 compiler using --with-fp-32=xx to set this by default.
1039 * The following bugs are resolved with this release:
1041 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1042 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1043 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1044 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1045 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1046 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1047 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1048 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1049 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1050 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1051 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1052 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1053 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1054 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1055 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1057 * Reverted change of ABI data structures for s390 and s390x:
1058 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1059 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1060 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1061 are preserved pointing straight to the same implementation as the old ones.
1062 Given that, new callers will simply provide a too-big buffer to these
1063 functions. Any applications/libraries out there that embed jmp_buf or
1064 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1065 against 2.19 headers will have to rebuilt again. This is necessary in any
1066 case to revert the breakage in their ABI caused by the glibc change.
1068 * Support for file description locks is added to systems running the
1069 Linux kernel. The standard file locking interfaces are extended to
1070 operate on file descriptions, not file descriptors, via the use of
1071 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1072 are associated with an open file instead of a process.
1074 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1076 * The minimum Linux kernel version that this version of the GNU C Library
1077 can be used with is 2.6.32.
1079 * Running the testsuite no longer terminates as soon as a test fails.
1080 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1081 with PASS or FAIL lines for individual tests. A summary of the results is
1082 printed, including a list of failing lists, and "make check" exits with
1083 error status if there were any unexpected failures. "make check
1084 stop-on-test-failure=y" may be used to keep the old behavior.
1086 * The am33 port, which had not worked for several years, has been removed
1089 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1090 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1091 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1092 conflicted with POSIX (support for which was removed in 2.19), the
1093 interfaces those macros enabled remain available when compiling with
1094 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1095 test macros defined.
1097 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1099 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1100 This may improve lock scaling of existing programs on TX capable systems.
1101 The lock elision code is only built with --enable-lock-elision=yes and
1102 then requires a GCC version supporting the TX builtins. With lock elision
1103 default mutexes are elided via __builtin_tbegin, if the cpu supports
1104 transactions. By default lock elision is not enabled and the elision code
1107 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1108 copy the path argument. This allowed programs to cause posix_spawn to
1109 deference a dangling pointer, or use an unexpected pathname argument if
1110 the string was modified after the posix_spawn_file_actions_addopen
1113 * All supported architectures now use the main glibc sysdeps directory
1114 instead of some being in a separate "ports" directory (which was
1115 distributed separately before glibc 2.17).
1117 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1118 On configurations that support it (all Linux configurations), it's now
1119 used regardless of the --enable-add-ons switch to configure. It is no
1120 longer possible to build such configurations without pthreads support.
1122 * Locale names, including those obtained from environment variables (LANG
1123 and the LC_* variables), are more tightly checked for proper syntax.
1124 setlocale will now fail (with EINVAL) for locale names that are overly
1125 long, contain slashes without starting with a slash, or contain ".." path
1126 components. (CVE-2014-0475) Previously, some valid locale names were
1127 silently replaced with the "C" locale when running in AT_SECURE mode
1128 (e.g., in a SUID program). This is no longer necessary because of the
1131 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1132 with application code using Intel MPX instructions. (With all previous
1133 versions, the MPX register state could be clobbered when making calls
1134 into or out of a shared library.) Note that while the new dynamic
1135 linker is compatible with all known x86 hardware whether or not it
1136 supports Intel MPX, some x86 instruction-set emulators might fail to
1137 handle the new instruction encodings. This is known to affect Valgrind
1138 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1139 release), and might affect other tools that do instruction emulation.
1141 * Support for loadable gconv transliteration modules has been removed.
1142 The support for transliteration modules has been non-functional for
1143 over a decade, and the removal is prompted by security defects. The
1144 normal gconv conversion modules are still supported. Transliteration
1145 with //TRANSLIT is still possible, and the //IGNORE specifier
1146 continues to be supported. (CVE-2014-5119)
1148 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1149 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1150 resulting a denial-of-service security vulnerability in applications which
1151 use functions related to iconv. (CVE-2014-6040)
1155 * The following bugs are resolved with this release:
1157 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1158 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1159 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1160 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1161 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1162 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1163 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1164 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1165 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1166 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1167 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1168 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1169 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1170 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1171 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1172 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1173 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1174 16431, 16453, 16474, 16506, 16510, 16529
1176 * Slovenian translations for glibc messages have been contributed by the
1177 Translation Project's Slovenian team of translators.
1179 * The public headers no longer use __unused nor __block. This change is to
1180 support compiling programs that are derived from BSD sources and use
1181 __unused internally, and to support compiling with Clang's -fblock
1182 extension which uses __block.
1184 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1185 large collation sequences to optimize multiple passes. This cache
1186 computation may overflow for large collation sequences and may cause a
1187 stack or buffer overflow. This is now fixed to use a slower algorithm
1188 which does not use a cache if there is an integer overflow.
1190 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1191 rules for large collation sequences to optimize multiple passes and falls
1192 back to alloca if malloc fails, resulting in a possible stack overflow.
1193 The implementation now falls back to an uncached collation sequence lookup
1196 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1197 initialized for static applications resulting in the security feature
1198 being disabled. The pointer guard is now correctly initialized to a
1199 random value for static applications. Existing static applications need
1200 to be recompiled to take advantage of the fix (bug 15754).
1202 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1203 to the d_name member of struct dirent, or omit the terminating NUL
1204 character. (Bugzilla #14699).
1206 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1207 aligned_alloc functions could allocate too few bytes or corrupt the
1208 heap when passed very large allocation size values (Bugzilla #15855,
1211 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1212 for AF_INET6 has been fixed (Bugzilla #16072).
1214 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1215 pap_CW, quz_PE, the_NP.
1217 * Substantially revised locales: gd_GB, ht_HT
1219 * The LC_ADDRESS field was updated to support country_car for almost all
1222 * ISO 1427 definitions were updated.
1224 * ISO 3166 definitions were updated.
1226 * The localedef utility now supports --big-endian and --little-endian
1227 command-line options to generate locales for a different system from that
1228 for which the C library was built.
1230 * Binary locale files now only depend on the endianness of the system for
1231 which they are generated and not on other properties of that system. As a
1232 consequence, binary files generated with new localedef may be incompatible
1233 with old versions of the GNU C Library, and binary files generated with
1234 old localedef may be incompatible with this version of the GNU C Library,
1235 in the following circumstances:
1237 + Locale files may be incompatible on m68k systems.
1239 + Locale archive files (but not separate files for individual locales) may
1240 be incompatible on systems where plain "char" is signed.
1242 * The configure option --disable-versioning has been removed. Builds with
1243 --disable-versioning had not worked for several years.
1245 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1246 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1248 * SystemTap probes for malloc have been introduced.
1250 * SystemTap probes for slow multiple precision fallback paths of
1251 transcendental functions have been introduced.
1253 * Support for powerpc64le has been added.
1255 * The soft-float powerpc port now supports e500 processors.
1257 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1259 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1260 set of header declarations that are enabled by default, even when other
1261 feature test macros or compiler options such as -std=c99 would otherwise
1262 disable some of those declarations.
1264 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1265 conflict with POSIX. The libbsd-compat library (which was a dummy library
1266 that did nothing) has also been removed.
1268 * Preliminary documentation about Multi-Thread, Async-Signal and
1269 Async-Cancel Safety has been added.
1271 * Change of ABI data structures for s390 and s390x:
1272 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1273 allow for future hardware extensions. All library functions that accept or
1274 return these structures were versioned in order to provide backwards
1275 compatibility with existing code. However, not all cases can be handled
1276 with interface versioning. If an application embeds either structure into
1277 another structure and that structure is passed to another compilation unit
1278 which expects the newer larger-sized object then any access to the new
1279 fields will result in undefined behaviour. Similarly any access to fields
1280 that were moved by the enlarging of the structures will cause undefined
1281 behaviour. To fix the undefined behaviour all objects that are part of the
1282 final in-memory image must agree on the size of structures, and this may
1283 require recompilation.
1287 * The following bugs are resolved with this release:
1289 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1290 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1291 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1292 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1293 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1294 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1295 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1296 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1297 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1298 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1299 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1300 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1301 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1304 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1305 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1306 Distributions can re-enable building and using pt_chown via the new configure
1307 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1308 considerable security risks and should only be used if the distribution
1309 understands and accepts the risks.
1311 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1314 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1315 fixed (Bugzilla #15330).
1317 * Add support for calling C++11 thread_local object destructors on thread
1318 and program exit. This needs compiler support for offloading C++11
1319 destructor calls to glibc.
1321 * Improved worst case performance of libm functions with double inputs and
1324 * Support for priority inherited mutexes in pthread condition variables on
1325 non-x86 architectures.
1327 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1329 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1331 * Optimized string functions for ARM. Implemented by Will Newton and
1334 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1336 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1339 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1340 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1342 * Added a benchmark framework to track performance of functions in glibc.
1344 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1345 It is based on draft TS 18661 and currently enabled as a GNU extension.
1347 * On Linux, the clock function now uses the clock_gettime system call
1348 for improved precision, rather than old times system call.
1350 * Added support for version-3 tz format files. This is needed when using
1351 the tz database release 2013e or later, and affects a few unusual cases --
1352 currently only TZ='America/Godthab' for time stamps after 2037.
1354 * Added new API functions pthread_getattr_default_np and
1355 pthread_setattr_default_np to get and set the default pthread
1356 attributes of a process.
1358 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1359 This may improve lock scaling of existing programs on TSX capable systems.
1360 When the --enable-lock-elision=yes parameter is specified at configure
1361 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1364 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1365 availability). Implemented by Ryan S. Arnold.
1367 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1371 * The following bugs are resolved with this release:
1373 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1374 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1375 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1376 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1377 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1378 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1379 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1380 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1381 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1382 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1383 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1384 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1385 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1386 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1387 14879, 14889, 14893, 14898, 14914.
1389 * Optimization of memcpy for MIPS.
1391 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1392 EMFILE has been fixed (Bugzilla #14889).
1394 * The add-on ports collection is now distributed in the "ports" subdirectory
1395 of the main GNU C Library distribution, rather than separately.
1397 * Port to ARM AArch64 contributed by Linaro.
1399 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1400 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1402 Implemented by Andreas Krebbel.
1404 * The new function secure_getenv allows secure access to the environment,
1405 returning NULL if running in a SUID/SGID process. This function replaces
1406 the internal function __secure_getenv.
1408 * SystemTap static probes have been added into the dynamic linker.
1409 Implemented by Gary Benson.
1411 * Optimizations of string functions strstr, strcasestr and memmem.
1412 Implemented by Maxim Kuvyrkov.
1414 * The minimum Linux kernel version that this version of the GNU C Library
1415 can be used with is 2.6.16.
1417 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1418 powerpc POWER7. Implemented by Will Schmidt.
1420 * New configure option --disable-nscd builds the C library such that it
1421 never attempts to contact the Name Service Caching Daemon (nscd).
1422 New configure option --disable-build-nscd avoids building nscd itself;
1423 this is the default if --disable-nscd is used.
1425 * Improved support for cross-compilation, including cross-testing and
1426 bootstrap builds without a previously built glibc.
1428 * Several testsuite tests are now able to test multiple IFUNC variants of an
1429 interface, rather than just testing the one that would be chooen by
1432 * New configure options --with-bugurl and --with-pkgversion, for
1433 distributors to use to embed their bug-reporting and package version
1434 information in --help and --version output.
1436 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1437 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1438 allows creation of chroots without the procfs mounted on /proc.
1440 * The `crypt' function now fails if passed salt bytes that violate the
1441 specification for those values. On Linux, the `crypt' function will
1442 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1443 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1444 when the mode is enabled.
1446 * The `clock_*' suite of functions (declared in <time.h>) is now available
1447 directly in the main C library. Previously it was necessary to link with
1448 -lrt to use these functions. This change has the effect that a
1449 single-threaded program that uses a function such as `clock_gettime' (and
1450 is not linked with -lrt) will no longer implicitly load the pthreads
1451 library at runtime and so will not suffer the overheads associated with
1452 multi-thread support in other code such as the C++ runtime library.
1454 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1459 * The following bugs are resolved with this release:
1461 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1462 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1463 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1464 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1465 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1466 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1467 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1468 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1469 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1470 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1471 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1472 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1473 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1474 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1475 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1476 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1477 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1478 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1479 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1480 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1481 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1482 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1485 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1486 configuring glibc with:
1487 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1488 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1489 Implemented by H.J. Lu.
1493 + define static_assert
1495 + do not declare gets
1497 + declare at_quick_exit and quick_exit also for ISO C11
1499 + aligned_alloc. NB: The code is deliberately allows the size parameter
1500 to not be a multiple of the alignment. This is a moronic requirement
1501 in the standard but it is only a requirement on the caller, not the
1504 + timespec_get added
1506 + uchar.h support added
1508 + CMPLX, CMPLXF, CMPLXL added
1510 Implemented by Ulrich Drepper.
1512 * Support for the IA-64 has been moved to ports.
1514 * Remove support for anything but ELF binary format
1516 * Checking versions of poll, ppoll added.
1517 Implemented by Ulrich Drepper.
1519 * More generic and 64-bit performance optimizations to math functions.
1520 Implemented by Ulrich Drepper.
1522 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1523 headers and functions available at compile time as they were before
1524 version 2.14. This option will be removed at some time in the future
1525 after the TI-RPC library becomes fully sufficient for the needs of
1526 existing applications.
1528 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1529 Note that glibc is not expected to work with any Linux kernel version
1532 * New header <sys/auxv.h> and function getauxval allowing easy access to
1533 the AT_* key-value pairs passed from the Linux kernel. The header also
1534 defines the HWCAP_* bits associated with the AT_HWCAP key.
1536 * New locales: mag_IN
1538 * New configure option --enable-systemtap builds SystemTap static probes
1539 into libc for setjmp and longjmp and into libpthread for various operations.
1540 So far the setjmp/longjmp probes and some of the libpthread probes are
1541 provided only for i*86 and x86_64.
1542 Implemented by Roland McGrath and Rayson Ho.
1544 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1546 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1549 * More optimized functions for SPARC. Implemented by David S. Miller.
1551 * Improved support for cross-compilation, especially bootstrap builds
1552 without a previously built glibc.
1554 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1555 Chris Metcalf from Tilera.
1557 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1558 now supported for ARM processors.
1560 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1561 as the name of the dynamic linker, to distinguish it from the
1562 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1564 * Support for CRIS has been removed from ports.
1566 * A new class of installed header has been documented for low-level
1567 platform-specific functionality. PowerPC added the first instance with a
1568 function to provide time base register access. Contributed by Tulio
1569 Magno Quites Machado Filho.
1571 * ix86 configurations now install header files that are consistent with
1572 what x86-64 configurations install. These same header files can be used
1573 for -m32, -m64, or -mx32 builds.
1574 Contributed by H.J. Lu.
1576 * Math library bug fixes. A thorough audit of all open math library bugs was
1577 conducted by Joseph Myers. Significant progress was made on many math
1578 library bugs resulting in more accurate exceptions and function results.
1579 Many thanks to all those that contributed including Andreas Jaeger for his
1580 patch review and work on the x87 trigonometric instruction issues.
1582 * Timezone data is no longer installed. Timezone-related binaries and scripts
1583 will continue to be installed. Users should obtain their timezone data from
1584 their distribution provider or from the tzdata package at
1585 <ftp://munnari.oz.au/pub/>.
1589 * The following bugs are resolved with this release:
1591 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1592 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1593 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1594 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1595 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1596 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1597 13484, 13506, 13515, 13523, 13524, 13538, 13540
1599 * New program pldd to list loaded object of a process
1600 Implemented by Ulrich Drepper.
1602 * Add nss_db support back to glibc. No more dependency on Berkeley db
1603 and support for initgroups lookups.
1604 Implemented by Ulrich Drepper.
1606 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1607 Contributed by HJ Lu.
1609 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1610 Contributed by HJ Lu.
1612 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1613 on x86-32 and x86-64.
1614 Contributed by Liubov Dmitrieva.
1616 * Optimized strchr and strrchr for SSE on x86-32.
1617 Contributed by Liubov Dmitrieva.
1619 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1620 for x86-64 and x86-32.
1621 Contributed by Liubov Dmitrieva.
1623 * New interfaces: scandirat, scandirat64
1624 Implemented by Ulrich Drepper.
1626 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1627 Implemented by Ulrich Drepper.
1629 * nscd now also caches the netgroup database.
1630 Implemented by Ulrich Drepper.
1632 * Integrate libm with gcc's -ffinite-math-only option.
1633 Implemented by Ulrich Drepper.
1635 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1636 to math functions. Implemented by Ulrich Drepper.
1638 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1639 Implemented by Ulrich Drepper.
1641 * New Linux interfaces: process_vm_readv, process_vm_writev
1643 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1644 Implemented by Ulrich Drepper.
1646 * Optimized nearbyint and strcasecmp for PPC.
1647 Implemented by Adhemerval Zanella.
1649 * New locales: bho_IN, unm_US, es_CU, ta_LK
1653 * The following bugs are resolved with this release:
1655 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1656 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1657 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1658 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1659 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1660 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1661 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1662 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1663 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1664 12795, 12811, 12813, 12814, 12841
1666 * The RPC implementation in libc is obsoleted. Old programs keep working
1667 but new programs cannot be linked with the routines in libc anymore.
1668 Programs in need of RPC functionality must be linked against TI-RPC.
1669 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1671 Visible changes of this change include (obviously) the inability to link
1672 programs using RPC functions without referencing the TI-RPC library and the
1673 removal of the RPC headers from the glibc headers.
1674 Implemented by Ulrich Drepper.
1676 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1677 syncfs, setns, sendmmsg
1679 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1680 yue_HK, lij_IT, mhr_RU
1682 * New encodings: CP770, CP771, CP772, CP773, CP774
1684 * New program sotruss to trace calls through PLTs
1685 Implemented by Ulrich Drepper.
1687 * The malloc hook implementation is marked deprecated and will be removed
1688 from the default implementation in the next version. The design never
1689 worked ever since the introduction of threads. Even programs which do
1690 not create threads themselves can use multiple threads created internally.
1694 * The following bugs are resolved with this release:
1696 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1697 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1698 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1699 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1702 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1704 * POWER7 optimizations: memset, memcmp, strncmp
1706 * New optimized string functions for x86-64: strnlen (SSE2),
1707 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1708 Implemented by Ulrich Drepper.
1712 * The following bugs are resolved with this release:
1714 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1715 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1716 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1717 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1718 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1719 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1720 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1721 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1723 * New interfaces: pthread_getname_np, pthread_setname_np
1725 * New Linux interface: recvmmsg
1727 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1729 * The dynamic linker now recognizes supported ABI versions from the
1730 EI_ABIVERSION field in the ELF header.
1731 Implemented by Ulrich Drepper.
1733 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1734 will not be used to fill in password fields in the passwd.byname replies.
1735 Instead it is used to synthesize the shadow.byname table, should it be
1736 missing. This is a useful mode in some installations involving Solaris.
1737 Implemented by Ulrich Drepper.
1739 * New locales: kok_IN, sq_MK, cv_RU
1743 * The following bugs are resolved with this release:
1745 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1746 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1747 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1748 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1749 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1750 10789, 10817, 10823, 10840, 10847
1752 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1754 Implemented by Ulrich Drepper.
1756 * Checking version of longjmp added that fails if an uninitialized stack
1757 frame would be created. Implemented by Ulrich Drepper.
1759 * STT_GNU_IFUNC is now supported in static executables.
1760 Implemented by H.J. Lu.
1762 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1764 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1765 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1766 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1767 Contributed by H.J. Lu.
1769 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1770 Implemented by Ulrich Drepper.
1772 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1774 Contributed by H.J. Lu.
1776 * Support for fma instruction in AVX on x86-64.
1777 Implemented by H.J. Lu and Ulrich Drepper.
1779 * AVX support in x86-64 auditing support in ld.so.
1780 Implemented by H.J. Lu.
1782 * STB_GNU_UNIQUE support added.
1783 Implemented by Ulrich Drepper.
1785 * Implement second fallback mode for optimized DNS lookup for even more
1786 broken environments. If two requests from the same port are not
1787 handled correctly close the socket and open a new one before sending
1788 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1789 can be used to select this mode right away, instead of rediscovering the
1790 necessity is every process again.
1791 Implemented by Ulrich Drepper.
1793 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1794 Implemented by Adam Tkac.
1796 * Optimized iconv conversions for S390x.
1797 Implemented by Andreas Krebbel.
1799 * Using condvars with PI mutexes is now more efficient due to kernel
1800 support for requeueing to PI futexes. NPTL support added for x86-64.
1801 Implemented by Ulrich Drepper.
1803 * New locales: ps_AF, my_MM
1807 * The following bugs are resolved with this release:
1809 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1810 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1811 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1812 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1813 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1815 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1816 Implemented by Ulrich Drepper.
1818 * New Linux interfaces: accept4, fallocate, fallocate64.
1819 Implemented by Ulrich Drepper.
1821 * Correct declarations of string function when used in C++ code. This
1822 could lead to compile errors for invalid C++ code.
1824 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1827 * New POSIX 2008 interface: psiginfo
1828 Implemented by Ulrich Drepper.
1830 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1831 Implemented by Ulrich Drepper.
1833 * Support for selecting between multiple function definitions at runtime
1834 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1836 * The libcrypt library can now use the hash function implementations in
1837 NSS. Implemented by Ulrich Drepper.
1839 * The malloc implementation can be compiled to be less memory efficient
1840 but higher performing in multi-threaded programs.
1841 Implemented by Ulrich Drepper.
1843 * New locales: nan_TW@latin, ks_IN
1845 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1846 Implemented by Ulrich Drepper.
1848 * Extended printf hook support. It is possible to use user-defined types
1849 and extend existing format specifiers.
1850 Implemented by Ulrich Drepper.
1852 * Handling for group shadow files has been added.
1853 Implemented by Ulrich Drepper.
1855 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1856 servers (the case, e.g., for some people using the built-in DNS
1857 server in ADSL modems/routers). There is a once-per-process timeout
1858 in case of a broken server. To avoid it, users can run nscd or put
1859 'options single-request' in /etc/resolv.conf.
1860 Implemented by Ulrich Drepper.
1864 * The following bugs are resolved with this release:
1866 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1867 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1868 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1869 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1870 6968, 6974, 6980, 6995, 7008, 7009, 7029
1872 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1873 up at the same time. Implemented by Ulrich Drepper.
1875 * TLS descriptors for LD and GD on x86 and x86-64.
1876 Implemented by Alexandre Oliva.
1878 * getaddrinfo now handles DCCP and UDPlite.
1879 Implemented by Ulrich Drepper.
1881 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1882 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1883 Implemented by Ulrich Drepper.
1885 * New implementation of memmem, strstr, and strcasestr which is O(n).
1886 Implemented by Eric Blake.
1888 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1890 * Implement "e" option for popen to open file descriptor with the
1891 close-on-exec flag set. Implemented by Ulrich Drepper.
1893 * Implement "b" mode for fmemopen. In this mode writes writes don't
1894 implicitly add a NUL byte and seeks from the end of the buffer really
1895 use the buffer end, not the string length as the basis.
1896 Implemented by Ulrich Drepper.
1898 * Many functions, exported and internal, now atomically set the close-on-exec
1899 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1901 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1903 Implemented by Pravin Satpute.
1905 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1909 * The following bugs are resolved with this release:
1911 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1912 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1913 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1914 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1915 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1916 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1917 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1919 * New locales: bo_CN, bo_IN, shs_CA.
1921 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1923 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1924 Implemented by Pravin Satpute.
1926 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1928 * nscd caches now all timeouts for DNS entries
1929 Implemented by Ulrich Drepper.
1931 * nscd is more efficient and wakes up less often.
1932 Implemented by Ulrich Drepper.
1934 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1935 vdprintf, and obstack_vprintf.
1936 Implemented by Jakub Jelinek.
1938 * Faster memset for x86-64.
1939 Implemented by Harsha Jagasia and H.J. Lu.
1941 * Faster memcpy on x86.
1942 Implemented by Ulrich Drepper.
1944 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1945 Implemented by Ulrich Drepper.
1947 * Faster sqrt and sqrtf implemention for some PPC variants.
1948 Implemented by Steven Munroe.
1952 * The following bugs are resolved with this release:
1954 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1955 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1956 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1957 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1958 5071, 5103, 5104, 5112, 5113, 5184, 5186
1960 * More checking functions: fread, fread_unlocked, open*, mq_open.
1961 Implemented by Jakub Jelinek and Ulrich Drepper.
1963 * Extend fortification to C++. Implemented by Jakub Jelinek.
1965 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1966 by not recognizing 'a' as a modifier when those specs are requested.
1967 Implemented by Jakub Jelinek.
1969 * PPC optimizations to math and string functions.
1970 Implemented by Steven Munroe.
1972 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1973 options to be passed. Implemented by Ulrich Drepper.
1975 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1977 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1978 Implemented by Ulrich Drepper.
1980 * Handle private futexes in the NPTL implementation.
1981 Implemented by Jakub Jelinek and Ulrich Drepper.
1983 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1984 Implemented by Roland McGrath and Ulrich Drepper.
1986 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1988 * SHA-256 and SHA-512 based password encryption.
1989 Implemented by Ulrich Drepper.
1991 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1992 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1995 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1996 Implemented by Ulrich Drepper.
2000 * The following bugs are resolved with this release:
2002 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2003 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2004 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2005 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2006 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2007 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2008 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2011 * New Linux interfaces: epoll_pwait, sched_getcpu.
2013 * New generic interfaces: strerror_l.
2015 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2019 * The following bugs are resolved with this release:
2021 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2022 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2023 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2024 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2025 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2026 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2027 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2028 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2029 3225, 3251, 3252, 3253, 3273
2031 * For Linux, the sorting of addresses returned by getaddrinfo now also
2032 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2033 Implemented by Ulrich Drepper.
2035 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2036 Implemented by Ulrich Drepper.
2038 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2040 * New iconv module for MIK. Contributed by Alexander Shopov.
2042 * For sites with broken group and/or passwd database, the auto-propagate
2043 option of nscd can prevent creating ID lookup entries from the results
2044 of a name lookup and vice versa. This usually is no problem but some
2045 site might have problems with the default behavior.
2046 Implemented by Ulrich Drepper.
2048 * Iterating over entire database in NIS can be slow. With the
2049 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2050 to trade time for memory. The entire database will be read at once.
2051 Implemented by Ulrich Drepper.
2053 * The interfaces introduced in RFC 3542 have been implemented by
2056 * Support for the new ELF hash table format was added by Ulrich Drepper.
2058 * Support for priority inheritance mutexes added by Jakub Jelinek and
2061 * Support for priority protected mutexes added by Jakub Jelinek.
2063 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2064 or_IN, csb_PL, fy_NL, sr_ME.
2068 * More overflow detection functions.
2070 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2071 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2073 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2074 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2075 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2076 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2077 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2078 by Masahide Washizawa.
2080 * It is now possible to install audit libraries for the dynamic linker using
2081 LD_AUDIT. Implemented by Ulrich Drepper.
2083 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2084 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2085 For a libc and libpthread that works well on Linux 2.4 kernels, we
2086 recommend using the stable 2.3 branch.
2088 * The new function open_wmemstream acts like open_memstream,
2089 but uses a wchar_t wide character buffer.
2091 * The new function ppoll is an improved version of poll, similar to pselect.
2093 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2094 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2095 renameat, unlinkat, linkat, symlinkat, readlinkat.
2097 * New Linux kernel system calls: unshare,
2098 inotify_init, inotify_add_watch, inotify_rm_watch.
2100 * The euidaccess function is now also known by the alias eaccess,
2101 for compatibility with some other systems.
2103 * Timezone data updated to 2006b version.
2107 * The following bugs are resolved with this release:
2109 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2110 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2111 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2112 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2113 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2114 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2116 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2118 * As of this release, GCC 4 can be used to compile the C Library.
2120 * Timezone data updated to 2005m version.
2124 * The following bugs are resolved with this release:
2126 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2127 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2128 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2129 777, 787, 821, 822, 823, 825
2131 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2135 * Support for RFC 3678. Real implementations exist only for Linux so far.
2136 Implemented by Ulrich Drepper.
2138 * nscd can now cache entries persistently. Expiring entries are reloaded.
2139 For speedups the cache can be shared in memory with client processes.
2140 Implemented by Ulrich Drepper.
2142 * nscd can now perform SELinux checks.
2143 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2145 * getaddrinfo queries are now cached. Canonical name lookup is performed
2147 Implemented by Ulrich Drepper.
2149 * The nothrow function attribute is used when headers are used by gcc when
2150 compiling C code. This can avoid significant amounts of exception
2153 * The malloc functions perform more error checking and are stricter when
2154 it comes to reacting on errors. The default action is to terminate
2155 the process after showing an error message. Implemented by Ulrich Drepper.
2157 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2158 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2160 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2161 namespaces using the new function dlmopen(). This feature is of course,
2162 like most other dynamic loading functionality, not available in statically
2163 linked applications. Implemented by Ulrich Drepper.
2165 * Low-overhead boundary checking variants of string and some stdio functions
2166 were added. These are to be used in conjunction with a gcc patch by
2167 Jakub Jelinek which adds calls to these functions if possible.
2168 Implemented by Jakub Jelinek and Ulrich Drepper.
2170 * Old code for several operating systems and machine architectures that
2171 have not been in working condition in a long time have been removed from
2172 the main source tree maintained by the GNU C Library's maintainers.
2173 These files are now reside in the separate `ports' source module
2174 that is usable as an add-on when building the library.
2178 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2179 interrogate the dynamic linker, compatible with the Solaris interface.
2181 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2182 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2184 * getifaddrs now uses the netlink interface on Linux to get its information.
2185 Implemented by Thorsten Kukuk.
2187 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2188 Implemented by Ulrich Drepper.
2190 * support for non-executable stacks on x86 has been added. Changes mostly
2193 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2196 * getaddrinfo now performs destination address selection according to
2201 * Thread-safe interfaces for many functions that access locale data
2202 were added in version 2.3, but these features were omitted from NEWS.
2203 Many functions have variants with an `_l' suffix that take a `locale_t'
2204 object as a parameter rather than consulting the current locale.
2205 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2206 create and maintain `locale_t' objects. Additionally, the new function
2207 `uselocale' sets "the current locale" (as used by functions not so
2208 parameterized) set for an individual thread. These features were added
2209 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2211 * The functions getresuid, getresgid, setresuid, and setresgid, which
2212 have long been available on Linux, are now declared in <unistd.h>
2213 and are now also available on the Hurd.
2215 * ELF thread-local storage support (TLS) now works on x86-64.
2217 * The new dynamic string token $LIB is expanded in shared library names.
2218 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2220 * Aldy Hernandez contributed complete software floating point support for
2221 PowerPC machines with no FPU.
2223 * fexecve is implemented on Linux.
2225 * The `btowc' function should work at least twice as fast due to
2226 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2228 * With approriate thread add-ons cancelable functions are now implemented
2229 in libc.so as well. No need to call the function in libpthread. This
2230 change allowed to finally disable the incorrect and expensive handling
2231 of weak definition in ld.so.
2233 * Yet more PLT entries in libc.so have been removed. We finally arrived
2234 at the bare minimum. Startup times improved appropriately.
2236 * Support for the new Linux/x86 system call interface was added. The
2237 AT_SYSINFO auxiliary vector entry is recognized and handled.
2241 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2244 * iconv (the program and the interface) now accepts empty names (excluding
2245 options like //TRANSLIT) to mean "use charset of current locale".
2247 * localedef can now transliterate characters in strings which are not in
2248 the provided charmap. The information from the input locale is used.
2250 * Prelinking support was added for ELF targets. This requires additional
2251 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2253 * Read-only stdio streams now use mmap to speed up operation by eliminating
2254 copying and buffer underflows. To use add 'm' to the mode string of
2255 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2257 * The malloc functions were completely rewritten by Wolfram Gloger based
2258 on Doug Lea's malloc-2.7.0.c.
2260 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2261 implementation of regex.
2263 * Bruno Haible upgraded the iconv and locale implementation to support
2266 * Contents of the LC_* and LANG environment variables in the CEN style are
2267 not recognized anymore. It never was used. Change by Ulrich Drepper.
2269 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2270 thread-local storage (TLS) ABI on some platforms.
2271 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2273 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2274 EUC-JISX0213, and TSCII.
2276 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2277 BSD-compatible interface for getting all network interface addresses.
2278 Implementation for IPv4 by Roland McGrath.
2280 * Loading of locale data is faster due to the introduction of a locale
2281 archive. Implemented by Roland McGrath and Ulrich Drepper.
2283 * Startup times are significantly reduced by not using exported functions
2284 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2287 * Steven Munroe contributed a port to PowerPC64/Linux.
2291 * The Hurd now uses the GNU libio implementation of stdio rather than the
2292 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2294 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2295 with the same behavior as the Linux system call of the same name.
2299 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2300 128-bit long double format.
2302 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2303 IBM1161, and IBM1162 charsets.
2305 * Andreas Jaeger contributed a port to x86-64/Linux.
2307 * Peter Bruin contributed a port to PowerPC/Hurd.
2309 * libc functions using I/O streams now can handle wide-oriented streams
2312 * optimizations in the dynamic linker. Binaries created by recent binutils
2313 versions start up quicker due to reduced time spent on relocations.
2315 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2319 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2320 asinh, atanh, j0 for the 128-bit long double format.
2322 * Bruno Haible updated all the code handling Unicode in some form to
2323 support Unicode 3.1.
2325 * Speed of regex for single-byte locales is back to previous levels.
2326 Patch by Isamu Hasegawa.
2328 * Alpha, SPARC, and IA-64 now also using floating stacks.
2330 * Startup time of internationalized applications greatly improved through
2331 iconv cache. Use iconvconfig to generate the cache file.
2332 Contributed by Ulrich Drepper.
2334 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2335 severe performance problems. Patch by David Mosberger.
2337 * The Hurd port got a lot more functionality like AIO, various stdio
2338 extensions, etc. Mainly done by Roland McGrath.
2340 * mtrace can now lookup symbols in shared libraries.
2344 * Intel's IA-64 math library is largely integrated. It provides fast and
2345 accurate implementatations for most basic and standard math functions
2346 in float, double, and long double format.
2348 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2349 and asin for the 96-bit long double format and asin, log, tan for the
2350 128-bit long double format.
2352 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2353 The basic double functions exist today. Contributed by Abraham Ziv
2354 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2355 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2357 * An asynchronous name lookup library was added. The interface is designed
2358 after POSIX AIO. The proposal was circulated beforehand to get comments.
2359 No negative ones came in. Implemented by Ulrich Drepper.
2361 * Port to S390/64bit contributed by Martin Schwidefsky
2362 <schwidefsky@de.ibm.com>.
2364 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2365 of functions for Linux/IA-64.
2367 * The RPC code is now thread safe. Threads can now use the same service
2368 of different services at the same time. Patch by Eric Norum
2369 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2371 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2372 family of functions for Linux/S390.
2374 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2375 of functions for Linux/x86.
2377 * Port to Linux/CRIS contributed by Axis Communications.
2381 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2382 we can now check for namespace violations and missing declarations. The
2383 result is that almost all headers are now Unix-compliant (as defined in
2384 the upcoming XPG6). The negative side is that some programs might need
2385 corrections, too, if they depend on the incorrect form of the headers in
2386 previous versions which defined too many symbols and included too many
2389 * regex now handles multibyte character sets correctly.
2390 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2392 * iconv (the program) does now conform to the upcoming XPG6 and handles
2393 charmaps. Instead of the charset names the path of charmaps can be
2394 provided and the conversion happens based on this data.
2395 Contributed by Ulrich Drepper.
2397 * The locale program now provides detailed information about the installed
2402 only lists the names of the supported locales
2406 provides details such as country, language, and codeset name.
2407 Contributed by Ulrich Drepper.
2411 * The gencat program now parses the input file according to the charset
2412 selected by the LC_CTYPE category. This is important for stateful
2413 character sets. To make generating catalogs easier there is a way
2414 to overwrite the charset selected by the locale: before the first
2415 message or $ quote line the catalog can contain a line like
2417 $ codeset=ISO-8859-2
2419 to select the charset (ISO-8859-2 in this case).
2421 Implemented by Shinya Hanataka and Ulrich Drepper.
2423 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2424 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2425 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2427 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2428 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2430 * The LANGUAGE environment variable is now ignored unless the locale is
2431 changed from the default "C" locale.
2433 * The usual bug fixes.
2437 * Greg McGary added runtime support for bounds checking using gcc's
2438 new -fbounded-pointers option. ix86 target is complete. PowerPC
2441 * Thorsten Kukuk added secure mode support to nscd.
2443 * The Berkeley db libraries have been removed.
2445 Related, the nss_db module is now in a separate package since it
2446 obviously requires a database library being available.
2448 * Wide character I/O streams implemented by Ulrich Drepper.
2450 * Functions from the extended socket API added by Ulrich Drepper.
2452 * Functions feenableexcept and fedisableexcept to control the
2453 behaviour of individual exceptions have been added by Andreas Jaeger.
2455 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2457 * The resolver code has been updated from bind 8.2.3-T5B which supports
2458 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2461 This change could in some situations effect backward compatibility. Since
2462 now `_res' is a thread-local instead of a global variable, modifying it
2463 in one thread does not have any effect in other threads.
2465 The resolver library was also extended to allow IPv6 as the transport
2466 protocol for the requests. This work was done by Stig Venaas.
2468 * Compatibility code for K&R C compilers has been removed from the
2469 header files. A ISO C compiler is needed to use the library
2470 (conforming to either C89 or C99 standard).
2472 * Complete rewrite of the localedef program to support multibyte character
2473 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2474 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2475 Implemented by Ulrich Drepper.
2477 Bruno Haible significantly improved the generation and use of the data
2478 structures for the wide character tables.
2480 * Plural handling in gettext implemented by Ulrich Drepper.
2482 * The utmp daemon has been removed.
2484 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2486 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2489 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2491 * POSIX barriers implemented by Kaz Kylheku.
2493 * POSIX spawn function family implemented by Ulrich Drepper.
2495 * POSIX spinlocks are now available.
2497 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2499 * the configure option --enable-kernel=X.Y.Z allows to strip out
2500 compatibility for kernel versions before X.Y.Z. This is currently only
2501 implemented for Linux.
2503 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2504 field sin6_scope_id. This means that all programs using IPv6 should be
2505 recompiled. Don't expect binary compatibility with previous glibc
2508 * various conversion modules for IBM character sets contributed by
2511 * IA-64 port by Jes Sorensen and HJ Lu.
2525 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2527 * New xdr functions are added; some rpc functions are now 64bit clean.
2529 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2531 * Fixed known incompatibilities with glibc 2.0.
2533 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2535 * Optimized a number of functions (especially the ELF dynamic loader).
2537 * Update timezone data files.
2539 * lots of charmaps corrections
2541 * some new locale definitions and charmaps
2546 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2547 conform to POSIX member type requirements. Symbol versions have been
2548 adjusted as needed within the library, and for direct use by applications,
2549 but there is potential for problems if third-party libraries use
2550 struct timeval as part of their interface. This does not present
2551 a problem for X and other "essential" system libraries.
2553 * An additional locale model to support C++ Standard Library locale
2554 model and probably more was implemented by Ulrich Drepper.
2556 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2559 * Miles Bader provided the `argp' function family to support hierachical
2560 command line argument parsing, layered on top of getopt.
2562 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2564 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2567 * scanf recognizes the %a and %A format for scanning floating point numbers.
2569 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2570 information and interfaces for the available integer types.
2572 * about 130 new math functions were added to implement the ISO C9x math
2575 * the new header <complex.h> contains definitions of the complex math
2576 functions from ISO C 9X.
2578 * the new header <tgmath.h> defines generic macros to use complex or
2579 real valued functions.
2581 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2583 * Andreas Jaeger provided a test suite for the math library.
2585 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2587 * Ulrich Drepper added character set conversion functions (iconv).
2589 * Optimized string functions have been added.
2591 * The localedata addon is now part of glibc.
2593 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2595 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2596 daemon for NSS (nscd).
2598 Missing a better place here are some numbers on improvements. Under
2599 Linux 2.1.125 un-tar-ing the kernel sources takes
2603 using local files 12.19s 6.88s 22.91s
2605 using NIS 13.92s 8.91s 26.34s
2607 using NIS & nscd 10.37s 7.34s 25.30s
2609 using NIS+ 27.57s 30.37s 640.46s
2611 using NIS+ & nscd 10.25s 7.83s 26.51s
2613 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2615 Keep in mind that non-namelookup related operations dominate above times.
2616 It was just a common complain that using NIS+ unpacking the kernel is
2619 [1] The old nscd implementation is not available anymore since it was
2620 distributed with glibc up to version 2.0.98 and thus is now replaced.
2622 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2624 * Mark Kettenis provided a Hesiod NSS module.
2626 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2627 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2629 * rcmd can now handle netgroups (Dick Streefland).
2631 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2634 * Support for the IPv6 protocol has been added to the socket API, as per the
2635 latest draft standards.
2637 * Support for Linux 2.2 has been added.
2639 * Interface changes relative to the latest 2.0.x release:
2640 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2641 addseverity NEW: Unix98
2642 alphasort64 NEW: LFS
2643 argp_err_exit_status NEW: argp, GNU ext
2644 argp_error NEW: argp, GNU ext
2645 argp_failure NEW: argp, GNU ext
2646 argp_help NEW: argp, GNU ext
2647 argp_parse NEW: argp, GNU ext
2648 argp_program_bug_address NEW: argp, GNU ext
2649 argp_program_version NEW: argp, GNU ext
2650 argp_program_version_hook NEW: argp, GNU ext
2651 argp_state_help NEW: argp, GNU ext
2652 argp_usage NEW: argp, GNU ext
2653 authdes_create NEW: Secure RPC
2654 authdes_getucred NEW: Secure RPC
2655 authdes_pk_create NEW: Secure RPC
2656 backtrace NEW: GNU ext.
2657 backtrace_symbols NEW: GNU ext.
2658 backtrace_symbols_fd NEW: GNU ext.
2660 cacosf NEW: ISO C 9x
2661 cacosh NEW: ISO C 9x
2662 cacoshf NEW: ISO C 9x
2663 cacoshl NEW: ISO C 9x
2664 cacosl NEW: ISO C 9x
2671 casinf NEW: ISO C 9x
2672 casinh NEW: ISO C 9x
2673 casinhf NEW: ISO C 9x
2674 casinhl NEW: ISO C 9x
2675 casinl NEW: ISO C 9x
2677 catanf NEW: ISO C 9x
2678 catanh NEW: ISO C 9x
2679 catanhf NEW: ISO C 9x
2680 catanhl NEW: ISO C 9x
2681 catanl NEW: ISO C 9x
2682 cbc_crypt NEW: Secure RPC
2686 ccoshf NEW: ISO C 9x
2687 ccoshl NEW: ISO C 9x
2693 cimagf NEW: ISO C 9x
2694 cimagl NEW: ISO C 9x
2695 clearerr_locked REMOVED
2696 clntunix_create NEW: sunrpc ext
2698 clog10 NEW: ISO C 9x
2699 clog10f NEW: ISO C 9x
2700 clog10l NEW: ISO C 9x
2710 cprojf NEW: ISO C 9x
2711 cprojl NEW: ISO C 9x
2713 crealf NEW: ISO C 9x
2714 creall NEW: ISO C 9x
2719 csinhf NEW: ISO C 9x
2720 csinhl NEW: ISO C 9x
2723 csqrtf NEW: ISO C 9x
2724 csqrtl NEW: ISO C 9x
2728 ctanhf NEW: ISO C 9x
2729 ctanhl NEW: ISO C 9x
2731 des_setparity NEW: Secure RPC
2732 ecb_crypt NEW: Secure RPC
2733 endutxent NEW: Unix98
2735 exp10f NEW: ISO C 9x
2736 exp10l NEW: ISO C 9x
2740 fattach NEW: STREAMS
2741 fdetach NEW: STREAMS
2745 feclearexcept NEW: ISO C 9x
2746 fegetenv NEW: ISO C 9x
2747 fegetexceptflag NEW: ISO C 9x
2748 fegetround NEW: ISO C 9x
2749 feholdexcept NEW: ISO C 9x
2751 feraiseexcept NEW: ISO C 9x
2752 ferror_locked REMOVED
2753 fesetenv NEW: ISO C 9x
2754 fesetexceptflag NEW: ISO C 9x
2755 fesetround NEW: ISO C 9x
2756 fetestexcept NEW: ISO C 9x
2757 feupdateenv NEW: ISO C 9x
2758 fflush_locked REMOVED
2762 fgets_unlocked NEW: GNU ext.
2763 fileno_locked REMOVED
2775 fputc_locked REMOVED
2776 fputs_unlocked NEW: GNU ext.
2777 fread_unlocked NEW: GNU ext.
2782 fstatvfs NEW: Unix98
2786 ftruncate64 NEW: LFS
2788 fwrite_unlocked NEW: GNU ext.
2789 gai_strerror NEW: IPv6
2793 getchar_locked REMOVED
2795 getdate_err NEW: Unix98
2796 getdate_r NEW: GNU ext.
2798 getnameinfo NEW: IPv6
2799 getnetname NEW: Secure RPC
2800 getpmsg NEW: STREAMS
2801 getpt NEW: Unix98 PTY
2802 getrlimit64 NEW: LFS
2803 getutxent NEW: Unix98
2804 getutxid NEW: Unix98
2805 getutxline NEW: Unix98
2806 glob64 NEW: GNU ext.
2807 globfree64 NEW: GNU ext.
2808 gnu_get_libc_release NEW: GNU ext.
2809 gnu_get_libc_version NEW: GNU ext.
2810 grantpt NEW: Unix98 PTY
2811 host2netname NEW: Secure RPC
2813 iconv_close NEW: iconv
2814 iconv_open NEW: iconv
2815 if_freenameindex NEW: IPv6
2816 if_indextoname NEW: IPv6
2817 if_nameindex NEW: IPv6
2818 if_nametoindex NEW: IPv6
2819 in6addr_any NEW: IPv6
2820 in6addr_loopback NEW: IPv6
2821 inet6_isipv4mapped NEW: IPv6
2822 isastream NEW: STREAMS
2823 iswblank NEW: GNU ext.
2824 key_decryptsession NEW: Secure RPC
2825 key_decryptsession_pk NEW: Secure RPC
2826 key_encryptsession NEW: Secure RPC
2827 key_encryptsession_pk NEW: Secure RPC
2828 key_gendes NEW: Secure RPC
2829 key_get_conv NEW: Secure RPC
2830 key_secretkey_is_set NEW: Secure RPC
2831 key_setnet NEW: Secure RPC
2832 key_setsecret NEW: Secure RPC
2833 llrint NEW: ISO C 9x
2834 llrintf NEW: ISO C 9x
2835 llrintl NEW: ISO C 9x
2836 llround NEW: ISO C 9x
2837 llroundf NEW: ISO C 9x
2838 llroundl NEW: ISO C 9x
2843 lrintf NEW: ISO C 9x
2844 lrintl NEW: ISO C 9x
2845 lround NEW: ISO C 9x
2846 lroundf NEW: ISO C 9x
2847 lroundl NEW: ISO C 9x
2849 makecontext NEW: Unix98
2850 mempcpy NEW: GNU ext.
2853 modify_ldt NEW: kernel
2857 nearbyint NEW: ISO C 9x
2858 nearbyintf NEW: ISO C 9x
2859 nearbyintl NEW: ISO C 9x
2860 netname2host NEW: Secure RPC
2861 netname2user NEW: Secure RPC
2862 nexttoward NEW: ISO C 9x
2863 nexttowardf NEW: ISO C 9x
2864 nexttowardl NEW: ISO C 9x
2868 passwd2des NEW: Secure RPC
2870 pow10f NEW: GNU ext.
2871 pow10l NEW: GNU ext.
2874 printf_size NEW: GNU ext.
2875 printf_size_info NEW: GNU ext.
2876 profil_counter REMOVED
2877 pthread_mutexattr_getkind_np REPLACED
2878 pthread_mutexattr_setkind_np REPLACED
2879 ptsname NEW: Unix98 PTY
2880 ptsname_r NEW: Unix98 PTY
2882 putchar_locked REMOVED
2883 putgrent NEW: GNU ext.
2885 putpmsg NEW: STREAMS
2886 pututxline NEW: Unix98
2890 readdir64_r NEW: LFS
2891 remquo NEW: ISO C 9x
2892 remquof NEW: ISO C 9x
2893 remquol NEW: ISO C 9x
2895 roundf NEW: ISO C 9x
2896 roundl NEW: ISO C 9x
2898 scalbln NEW: ISO C 9x
2899 scalblnf NEW: ISO C 9x
2900 scalblnl NEW: ISO C 9x
2902 sendfile NEW: kernel
2903 setcontext NEW: Unix98
2904 setrlimit64 NEW: LFS
2905 setutxent NEW: Unix98
2907 sigignore NEW: Unix98
2908 sigqueue NEW: POSIX.1b
2909 sigrelse NEW: Unix98
2910 sigset NEW: POSIX.1b
2911 sigtimedwait NEW: POSIX.1b
2912 sigwaitinfo NEW: POSIX.1b
2913 sincos NEW: GNU ext.
2914 sincosf NEW: GNU ext.
2915 sincosl NEW: GNU ext.
2919 strcasestr NEW: GNU ext.
2920 strtoimax NEW: ISO C 9x
2921 strtoumax NEW: ISO C 9x
2922 strverscmp NEW: GNU ext.
2923 svcauthdes_stats NEW: Secure RPC
2924 svcunix_create NEW: sunrpc etx
2925 svcunixfd_create NEW: sunrpc ext
2926 swapcontext NEW: Unix98
2927 tcgetsid NEW: Unix98 PTY
2928 tdestroy NEW: GNU ext.
2929 tgamma NEW: ISO C 9x
2930 tgammaf NEW: ISO C 9x
2931 tgammal NEW: ISO C 9x
2935 truncf NEW: ISO C 9x
2936 truncl NEW: ISO C 9x
2938 unlockpt NEW: Unix98 PTY
2939 updwtmpx NEW: Unix98
2940 user2netname NEW: Secure RPC
2941 utmpxname NEW: Unix98
2942 versionsort NEW: GNU ext.
2943 versionsort64 NEW: GNU ext.
2945 wcscasecmp NEW: GNU ext.
2946 wcsncasecmp NEW: GNU ext.
2947 wcsnlen NEW: GNU ext.
2948 wcstoimax NEW: ISO C 9x
2949 wcstoll NEW: ISO C 9x
2950 wcstoull NEW: ISO C 9x
2951 wcstoumax NEW: ISO C 9x
2953 wordexp NEW: POSIX.2
2954 wordfree NEW: POSIX.2
2955 write_profiling REMOVED
2956 xdecrypt NEW: Secure RPC
2957 xdr_authdes_cred NEW: Secure RPC
2958 xdr_authdes_verf NEW: Secure RPC
2959 xdr_cryptkeyarg NEW: Secure RPC
2960 xdr_cryptkeyarg2 NEW: Secure RPC
2961 xdr_cryptkeyres NEW: Secure RPC
2962 xdr_getcredres NEW: Secure RPC
2963 xdr_key_netstarg NEW: Secure RPC
2964 xdr_key_netstres NEW: Secure RPC
2965 xdr_keybuf NEW: Secure RPC
2966 xdr_keystatus NEW: Secure RPC
2967 xdr_netnamestr NEW: Secure RPC
2968 xdr_sizeof NEW: Secure RPC
2969 xdr_unixcred NEW: sunrpc ext
2970 xencrypt NEW: Secure RPC
2971 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2982 * inet_ntoa is thread-safe
2984 * updwtmp is moved from libutil to libc
2986 * rewrite of cbrt function
2988 * update of timezone data
3002 * add atoll function
3004 * fix complex problems in Berkeley DB code
3006 * fix math functions
3010 * fixed lots of header problems (especially Linux/GNU specific)
3012 * dynamic loader preserves all registers
3014 * Roland McGrath provided support for handling of auxiliary objects in
3015 the ELF dynamic loader.
3017 * support for parallel builds is improved
3021 * GNU extensions are no longer declared by default. To enable them you
3022 must define the macro `_GNU_SOURCE' in your program or compile with
3025 * The library has changed from using GNU ld symbol aliases to using weak
3026 symbols where available. The ELF object file format supports weak
3027 symbols; GNU ld also supports weak symbols in the a.out format. (There
3028 is also now support for other GNU ld extensions in ELF. Use the
3029 `--with-elf' option to configure to indicate you have ELF, and
3030 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3031 of many files which contained only symbol aliases, reducing the size of
3032 the source and the compiled library; many other files were renamed to
3033 less cryptic names previously occupied by the symbol alias files.
3034 There is a new header file <elf.h> for programs which operate on
3035 files in the ELF format.
3037 * Converted to Autoconf version 2, so `configure' has more options.
3038 Run `configure --help' to see the details.
3040 * The library can now be configured to build profiling, highly-optimized
3041 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3042 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3043 `configure' enable building these extra libraries. The shared library is
3044 built by default when using both ELF and GNU ld. When shared libraries
3045 are enabled, the new library `-ldl' is available for arbitrary run-time
3046 loading of shared objects; its interface is defined in <dlfcn.h>. The
3047 new header file <link.h> gives access to the internals of the run-time
3048 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3049 application of same name on other systems and it provides information
3050 about dynamically linked binaries.
3052 * The C library now provides the run-time support code for profiling
3053 executables compiled with `-pg'. Programs can control the profiling code
3054 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3055 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3056 the support for this file format was contributed by David Mosberger-Tang.
3058 * The math code has been replaced with a math library based on fdlibm from
3059 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3060 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3061 functions. The math functions now reside in a separate library, so
3062 programs using them will need to use `-lm' their linking commands.
3064 * John C. Bowman contributed optimized ix87 assembler inline functions.
3066 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3067 mechanism similar to that found in Solaris 2. This is now used for the
3068 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3069 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3070 file controls what services are used for each individual database. This
3071 works by loading shared libraries with names specified in `nsswitch.conf',
3072 so service modules can be changed or added at any time without even
3073 relinking any program. Currently there are the file, db, and NIS based
3074 NSS services available.
3076 * The new functions `strtoq' and `strtouq' parse integer values from
3077 strings, like `strtol' and `strtoul', but they return `long long int' and
3078 `unsigned long long int' values, respectively (64-bit quantities).
3080 * The new functions `strtof' and `strtold' parse floating-point values from
3081 strings, like `strtod', but they return `float' and `long double' values,
3082 respectively (on some machines `double' and `long double' are the same).
3084 * Ulrich Drepper has contributed new implementations of the floating-point
3085 printing and reading code used in the `printf' family of functions and
3086 `strtod', `strtof', and `strtold'. These new functions are perfectly
3087 accurate, and much faster than the old ones.
3089 * The implementation of the POSIX locale model was completely rewritten by
3090 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3091 to compile the POSIX locale definition.
3093 * The former dummy implementations of the strcoll and strxfrm function are
3094 now replaced by fully functional code contributed by Ulrich Drepper. The
3095 collation information comes from the POSIX locale definitions.
3097 * The new header <langinfo.h> defines an interface for accessing
3098 various locale-dependent data (using the locale chosen with `setlocale').
3100 * Ulrich Drepper has contributed a new suite of functions for operation on
3101 wide-character and multibyte-character strings, in <wchar.h>;
3102 and classification and case conversion of wide characters, in <wctype.h>.
3103 These new functions are conforming to the ISO C, Amendement 1 specification.
3105 * There is now a second implementation of the standard I/O library available.
3106 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3107 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3109 * You can now use positional parameter specifications in format strings
3110 for the `printf' and `scanf' families of functions. For example,
3111 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3112 ``Number 6, Mr Jones''. This is mainly useful when providing different
3113 format strings for different languages, whose grammars may dictate
3114 different orderings of the values being printed. To support this
3115 feature, the interface for `register_printf_handler' has changed; see
3116 the header file <printf.h> for details.
3118 * The `printf' and `scanf' families of functions now understand a new
3119 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3120 to group numbers as indicated by the locale; for `scanf' and friends, this
3121 says to accept as valid only a number with all the proper grouping
3122 separators in the right places. In the default "C" locale, numbers are
3123 not grouped; but locales for specific countries will define the usual
3124 conventions (i.e. separate thousands with `,' in the US locale).
3126 * The pgrp functions have been regularized, slightly incompatibly but much
3127 less confusingly. The core functions are now `getpgid' and `setpgid',
3128 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3129 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3130 provided for compatibility. There is no longer an incompatible `getpgrp'
3131 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3133 * The new header file <fts.h> and suite of functions simplify programs that
3134 operate on directory trees. This code comes from 4.4 BSD.
3136 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3137 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3138 NSS scheme used in glibc.
3140 * There is a new malloc debugging hook `__memalign_hook'.
3142 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3143 `unsigned int' in <sys/types.h>. These are for compatibility only and
3144 their use is discouraged.
3146 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3147 done differently, so that it works even without GNU ld.
3149 * New function `euidaccess' checks allowed access to a file like `access',
3150 but using the effective IDs instead of the real IDs.
3152 * The time zone data files have been updated for the latest and greatest
3153 local time conventions of the countries of the world.
3155 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3158 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3159 for formatting floating-point numbers. They are provided only for
3160 compatibility; new programs should use `sprintf' instead. There are
3161 also equivalent function for the `long double' floating-point type and
3162 all functions also exist in a reentrant form.
3164 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3165 functions for maintaining the login-record files (primarily of use to
3166 system programs such as `login'), and convenient functions for
3167 allocating and initializing a pseudo-terminal (pty) device.
3169 * Ulrich Drepper has contributed new support for System V style
3170 shared memory and IPC on systems that support it.
3172 * Ulrich Drepper has contributed several miscellaneous new functions found
3173 in System V: The `hsearch' family of functions provide an effective
3174 implementation of hash tables; `a64l' and `l64a' provide a very simple
3175 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3178 * Ulrich Drepper has contributed new reentrant counterparts for the
3179 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3181 * Ulrich Drepper has contributed new, highly-optimized versions of several
3182 string functions for the i486/Pentium family of processors.
3184 * Ulrich Drepper has updated the Linux-specific code, based largely
3185 on work done in Hongjiu Lu's version of GNU libc for Linux.
3186 The GNU library now supports Linux versions 2.0.10 and later,
3187 using the ELF object file format (i[3456]86-*-linux).
3189 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3191 * David Mosberger-Tang and Richard Henderson have ported the C library
3192 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3193 dynamic linking support for ELF/Alpha.
3195 * Richard Henderson contributed several Alpha optimized assembler function
3196 for arithmetic and string handling.
3198 * Ulrich Drepper has contributed a new set of message catalog functions to
3199 support multiple languages using the <libintl.h> interface, for use with
3200 his new package GNU gettext. Translation volunteers have contributed
3201 catalogs of the library's messages in Spanish, German, and Korean.
3203 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3204 program and the `catgets' function for reading the catalog files it
3205 creates. (The <libintl.h> interface is preferred; we include the
3206 <nl_types.h> interface using `catgets' only for source compatibility with
3207 programs already written to use it.)
3209 * New header file <values.h> gives SVID-compatible names for <limits.h>
3212 * Various new macros, declarations, and small header files for compatibility
3215 * New function `group_member' is a convenient way to check if a process has
3216 a given effective group ID.
3218 * When using GCC 2.7 and later, the socket functions are now declared in a
3219 special way so that passing an argument of type `struct sockaddr_in *',
3220 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3221 `struct sockaddr *' type, does not generate a type-clash warning.
3223 * New function `error' declared in header file <error.h> is a convenient
3224 function for printing error messages and optionally exiting; this is the
3225 canonical function used in GNU programs. The new functions `err', `warn',
3226 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3227 doing the same thing.
3229 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3230 POSIX.2 `glob' function to do ~ and {...} expansion.
3232 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3233 `clearenv' which is used in POSIX.9 is also available.
3235 * New function `getsid' returns session ID number on systems that support it.
3237 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3238 files <db.h> and <mpool.h> provide a rich set of functions for several
3239 types of simple databases stored in memory and in files, and <ndbm.h> is
3240 an old `ndbm'-compatible interface using the `db' functions. Link with
3241 `-ldb' to get these functions.
3243 * New macro `strdupa' copies a string like `strdup', but uses local stack
3244 space from `alloca' instead of dynamic heap space from `malloc'.
3246 * New function `strnlen' is like `strlen' but searches only a given maximum
3247 number of characters for the null terminator. `stpncpy', `strndup' and
3248 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3251 * New function `statfs' in header <sys/statfs.h>.
3253 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3254 provide convenient functions for operating on blocks of null-terminated
3257 * A new suite of functions in <utmp.h> handle all the details of reading
3258 and writing the utmp file.
3260 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3263 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3264 completely POSIX compliant and also implemented the extended functionality
3265 to handle alternate digit representation and alternate era date formats.
3267 * Ulrich Drepper provided an implementation of the `strptime' function
3268 defined in XPG4.2 which transforms a string into a `struct tm' value.
3270 * Paul Eggert provided the tzselect shell script as part of the timezone
3271 code. The shell script makes it easy to select the correct timezone
3274 * The implementation of the malloc family of functions is completely replaced
3275 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3276 The implementation uses the mmap function (if available) and it is
3277 optimized for the use in multi threaded programs.
3279 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3280 functions. This new functionality is usable by specifying a special
3281 salt string and it is compatible with implementation on *BSD systems.
3283 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3284 `getsubopt' to handle second level command line options, `bsd_signal'
3285 to access BSD style `signal' functionality, the obsolete `regexp' style
3288 * the `lchown' function is available on system which support this
3291 * The implementation of the shadow password handling function was contributed
3294 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3296 * POSIX.1g support was added. The <sys/select.h> header is available,
3297 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3298 implementation of `getaddrinfo'.
3302 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3304 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3305 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3307 * The new function `daemon' from 4.4 BSD is useful for server programs that
3308 want to put themselves in the background.
3310 * Joel Sherrill has contributed support for several standalone boards that
3311 run without an operating system.
3313 * `printf', `scanf' and friends now accept a `q' type modifier for long
3314 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3316 * All of the code taken from BSD (notably most of the math and networking
3317 routines) has been updated from the BSD 4.4-Lite release.
3319 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3321 * The new functions `getdomainname' and `setdomainname' fetch or change the
3322 YP/NIS domain name. These are system calls which exist on systems which
3325 * The time zone data files have been updated for the latest international
3328 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3329 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3333 * The C library now includes support for Sun RPC, from Sun's free
3334 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3335 are included. (There is still no support for YP.)
3337 * Tom Quinn has contributed a port of the C library to SGI machines running
3338 Irix 4 (mips-sgi-irix4).
3340 * The new `lockf' function is a simplified interface to the locking
3341 facilities of `fcntl', included for compatibility.
3343 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3345 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3347 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3350 * The error message printed by `assert' for a failed assertion now includes
3351 the name of the program (if using GNU ld) and the name of the calling
3352 function (with versions of GCC that support this).
3354 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3356 * The library now includes the <sys/mman.h> header file and memory
3357 management functions `mmap', `munmap', `mprotect', `msync', and
3358 `madvise', on systems that support those facilities.
3360 * The interface for `mcheck' has changed slightly: the function called to
3361 abort the program when an allocation inconsistency is detected now takes
3362 an argument that indicates the type of failure. The new function
3363 `mprobe' lets you request a consistency check for a particular block at
3364 any time (checks are normally done only when you call `free' or `realloc'
3367 * It is now possible to easily cross-compile the C library, building on one
3368 system a library to run on another machine and/or operating system. All
3369 you need to do is set the variable `HOST_CC' in `configparms' to the
3370 native compiler for programs to run on the machine you are building on (a
3371 few generator programs are used on Unix systems); set `CC' to the
3374 * The new function `fexecve' (only implemented on the GNU system) executes
3375 a program file given a file descriptor already open on the file.
3379 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3380 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3382 * You can set the variable `libprefix' in `configparms' to specify a prefix
3383 to be prepended to installed library files; this makes it easy to install
3384 the GNU C library to be linked as `-lgnuc' or whatever.
3386 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3387 copies a limited number of characters from a string, and returns the
3388 address of the last character written.
3390 * You no longer need to check for whether the installed `stddef.h' is
3391 compatible with the GNU C library. configure now checks for you.
3393 * You can now define a per-stream `fileno' function to convert the
3394 stream's cookie into an integral file descriptor.
3396 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3397 allocates zero bytes of storage, and returns a unique pointer which
3398 you can pass to `realloc' or `free'. The behavior is undefined if
3399 you dereference this pointer.
3401 * The C library now runs on Sony NEWS m68k machines running either
3402 NewsOS 3 or NewsOS 4.
3404 * The new `syscall' function is a system-dependent primitive function
3405 for invoking system calls. It has the canonical behavior on Unix
3406 systems, including unreliable return values for some calls (such as
3407 `pipe', `fork' and `getppid').
3409 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3410 to `EAGAIN', which is the preferred name. On systems whose kernels
3411 use two distinct codes, the C library now translates EWOULDBLOCK to
3412 EAGAIN in every system call function.
3416 * The GNU C Library Reference Manual is now distributed with the library.
3417 `make dvi' will produce a DVI file of the printed manual.
3418 `make info' will produce Info files that you can read on line using C-h i
3419 in Emacs or the `info' program.
3420 Please send comments on the manual to bug-glibc-manual@gnu.org.
3422 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3424 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3426 * Jason Merrill has contributed a port to the Sequent Symmetry running
3427 Dynix version 3 (i386-sequent-dynix).
3429 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3430 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3432 * New function `memory_warnings' lets you arrange to get warnings when
3433 malloc is running out of memory to allocate, like Emacs gives you.
3435 * The C library now contains the relocating allocator used in Emacs 19 for
3436 its editing buffers. This allocator (ralloc) minimizes allocation
3437 overhead and fragmentation by moving allocated regions around whenever it
3438 needs to. You always refer to a ralloc'd region with a "handle" (a
3439 pointer to a pointer--an object of type `void **').
3441 * There is a new `printf' format: `%m' gives you the string corresponding
3442 to the error code in `errno'.
3444 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3445 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3446 pass, the `a' modifier says to fill in a `char **' you pass with a
3449 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3450 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3451 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3453 * `mkstemp' is a traditional Unix function to atomically create and open a
3454 uniquely-named temporary file.
3458 * The standard location for the file that says what the local timezone is
3459 has changed again. It is now `/usr/local/etc/localtime' (or more
3460 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3462 * The distribution no longer contains any files with names longer than 14
3465 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3466 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3468 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3472 * The configuration process has changed quite a bit. The `configure'
3473 script is now used just like the configuration scripts for other GNU
3474 packages. The `sysdeps' directory hierarchy is much rearranged.
3475 The file `INSTALL' explains the new scheme in detail.
3477 * The header files no longer need to be processed into ANSI C and
3478 traditional C versions. There is just one set of files to install, and
3479 it will work with ANSI or old C compilers (including `gcc -traditional').
3481 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3482 MIPS DECStation running Ultrix 4.
3484 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3485 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3486 made itself into a shared library.
3488 * Yet further improved support for the i386, running 4.3 BSD-like systems
3489 (such as Mach 3 with the Unix single-server), or System V.
3491 * New function `strncasecmp' to do case-insensitive string comparison
3492 with limited length.
3494 * New function `strsep' is a reentrant alternative to `strtok'.
3496 * New functions `scandir' and `alphasort' for searching directories.
3498 * New function `setenv' is a better interface to `putenv'.
3500 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3501 function for traversing a directory tree.
3503 * The GNU obstack package is now also part of the C library.
3504 The new function `open_obstack_stream' creates a stdio stream that
3505 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3506 formatted output directly to an obstack.
3508 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3509 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3511 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3513 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3514 things to your strings.
3516 * There are some new test programs: test-fseek, testmb, and testrand.
3518 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3519 These ports are not finished, but are a good starting place for really
3520 supporting those systems.
3522 * `/etc/localtime' is now the standard location for the file that says what
3523 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3524 This follows the general principle that `/etc' is the place for all local
3525 configuration files.
3527 * The C library header files now use `extern "C"' when used by the C++
3528 compiler, so the C library should now work with C++ code.
3530 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3531 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3534 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3535 function which uses the `merge sort' algorithm, and is said to be
3536 significantly faster than the old GNU `qsort' function. Merge sort is
3537 now the standard `qsort' function. The new algorithm can require a lot
3538 of temporary storage; so, the old sorting function is called when the
3539 required storage is not available.
3541 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3542 provides the Unix `crypt' function, plus some other entry points.
3544 * The code and header files taken from 4.4 BSD have been updated with the
3545 latest files released from Berkeley.
3547 ----------------------------------------------------------------------
3548 Copyright information:
3550 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3552 Permission is granted to anyone to make or distribute verbatim copies
3553 of this document as received, in any medium, provided that the
3554 copyright notice and this permission notice are preserved,
3555 thus giving the recipient permission to redistribute in turn.
3557 Permission is granted to distribute modified versions
3558 of this document, or of portions of it,
3559 under the above conditions, provided also that they
3560 carry prominent notices stating who last changed them.
3563 version-control: never