1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
10 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
11 24731-2:2010, is supported to enable declarations of functions from that
12 TR. Note that not all functions from that TR are supported by the GNU C
15 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
16 18661-1:2014, is supported to enable declarations of functions and macros
17 from that TS. Note that most features from that TS are not supported by
20 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
21 18661-4:2015, is supported to enable declarations of functions and macros
22 from that TS. Note that most features from that TS are not supported by
25 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
26 means that in a future release, the macros “major”, “minor”, and “makedev”
27 will only be available from <sys/sysmacros.h>.
29 These macros are not part of POSIX nor XSI, and their names frequently
30 collide with user code; see for instance glibc bug 19239 and Red Hat bug
31 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
32 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
35 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
36 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions,
37 the femode_t type and the FE_DFL_MODE macro.
39 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
40 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
41 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
42 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
43 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
44 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
45 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
46 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
47 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
48 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
49 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
52 * New <math.h> features are added from TS 18661-1:2014:
54 - Signaling NaN macros: SNANF, SNAN, SNANL.
56 - llogb functions: the llogb, llogbf and llogbl functions, and the
57 FP_LLOGB0 and FP_LLOGBNAN macros.
59 - Comparison macros: iseqsig.
61 - Classification macros: iscanonical, issubnormal, iszero.
63 - Total order functions: totalorder, totalorderf, totalorderl,
64 totalordermag, totalordermagf, totalordermagl.
66 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
68 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
69 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
71 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
72 are added to libc. They convert a floating-point number into string.
74 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
75 to float instead of double. This does not affect the ABI of any libraries
76 that are part of the GNU C Library, but may affect the ABI of other
77 libraries that use this type in their interfaces.
79 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
80 float_t and double_t types are now defined to long double instead of float
81 and double. These options are not the default, and this does not affect
82 the ABI of any libraries that are part of the GNU C Library, but it may
83 affect the ABI of other libraries that use this type in their interfaces,
84 if they are compiled or used with those options.
86 * The buffer size for byte-oriented stdio streams is now limited to 8192
87 bytes by default. Previously, on Linux, the default buffer size on most
88 file systems was 4096 bytes (and thus remains unchanged), except on
89 network file systems, where the buffer size was unpredictable and could be
90 as large as several megabytes.
92 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
93 for the Linux quota interface which predates kernel version 2.4.22 has
96 * The malloc_get_state and malloc_set_state functions have been removed.
97 Already-existing binaries that dynamically link to these functions will
98 get a hidden implementation in which malloc_get_state is a stub. As far
99 as we know, these functions are used only by GNU Emacs and this change
100 will not adversely affect already-built Emacs executables. Any undumped
101 Emacs executables, which normally exist only during an Emacs build, should
102 be rebuilt by re-running “./configure; make” in the Emacs build tree.
104 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
105 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
106 “no-ip6-dotint” had already been the default, and support for the
107 “ip6-dotint” option was removed from the Internet in 2006.
109 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
110 flag from <resolv.h> have been removed. The option relied on a
111 backwards-incompatible DNS extension which was never deployed on the
114 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
115 in the <resolv.h> header file have been deprecated. They were already
118 * DNSSEC-related declarations and definitions have been removed from the
119 <arpa/nameser.h> header file, and libresolv will no longer attempt to
120 decode the data part of DNSSEC record types. Previous versions of glibc
121 only implemented minimal support for the previous version of DNSSEC, which
122 is incompatible with the currently deployed version.
124 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
125 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
126 <arpa/nameser.h> header file because the distinction between RR types and
127 meta-RR types is not officially standardized, subject to revision, and
128 thus not suitable for encoding in a macro.
130 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
131 and rhook members of the res_state type in <resolv.h> have been removed.
132 The glibc stub resolver did not support these hooks, but the header file
133 did not reflect that.
135 * For multi-arch support it is recommended to use a GCC which has
136 been built with support for GNU indirect functions. This ensures
137 that correct debugging information is generated for functions
138 selected by IFUNC resolvers. This support can either be enabled by
139 configuring GCC with '--enable-gnu-indirect-function', or by
140 enabling it by default by setting 'default_gnu_indirect_function'
141 variable for a particular architecture in the GCC source file
144 * GDB pretty printers have been added for mutex and condition variable
145 structures in POSIX Threads. When installed and loaded in gdb these pretty
146 printers show various pthread variables in human-readable form when read
147 using the 'print' or 'display' commands in gdb.
149 Security related changes:
151 On ARM EABI (32-bit), generating a backtrace for execution contexts which
152 have been created with makecontext could fail to terminate due to a
153 missing .cantunwind annotation. This has been observed to lead to a hang
154 (denial of service) in some Go applications compiled with gccgo. Reported
155 by Andreas Schwab. (CVE-2016-6323)
157 The following bugs are resolved with this release:
159 [The release manager will add the list generated by
160 scripts/list-fixed-bugs.py just before the release.]
164 * The minimum Linux kernel version that this version of the GNU C Library
165 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
166 kernel version 2.6.32 or later suffices (on architectures that already
167 required kernel versions more recent than 3.2, those requirements remain
168 unchanged). Linux 3.2 or later kernel headers are required on all
171 * The pap_AN locale has been deleted. This has been deprecated for a long
172 time. It has been replaced by pap_AW & pap_CW, both of which have long
173 been included in previous releases.
175 * The readdir_r and readdir64_r functions have been deprecated. It is
176 recommended to use readdir and readdir64 instead.
178 * The type “union wait” has been removed. It was deprecated in the early
179 1990s and never part of POSIX. Application code should use the int type
180 instead of “union wait”.
182 * A new NSS action is added to facilitate large distributed system
183 administration. The action, MERGE, allows remote user stores like LDAP
184 to be merged into local user stores like /etc/groups in order to provide
185 easy to use, updated, and managed sets of merged credentials. The new
186 action can be used by configuring it in /etc/nsswitch.conf:
187 group: files [SUCCESS=merge] nis
188 Implemented by Stephen Gallagher (Red Hat).
190 * The deprecated __malloc_initialize_hook variable has been removed from the
193 * The long unused localedef --old-style option has been removed. It hasn't
194 done anything in over 16 years. Scripts using this option can safely
197 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
198 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
199 return the next representable value in the direction of positive infinity
200 and the nextdown functions return the next representable value in the
201 direction of negative infinity. These are currently enabled as GNU
204 Security related changes:
206 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
207 could result in a stack overflow when getnetbyname was called with an
208 overly long name. (CVE-2016-3075)
210 * Previously, getaddrinfo copied large amounts of address data to the stack,
211 even after the fix for CVE-2013-4458 has been applied, potentially
212 resulting in a stack overflow. getaddrinfo now uses a heap allocation
213 instead. Reported by Michael Petlan. (CVE-2016-3706)
215 * The glob function suffered from a stack-based buffer overflow when it was
216 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
217 Reported by Alexander Cherepanov. (CVE-2016-1234)
219 * The Sun RPC UDP client could exhaust all available stack space when
220 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
221 alloca plugin for GCC. (CVE-2016-4429)
223 * The IPv6 name server management code in libresolv could result in a memory
224 leak for each thread which is created, performs a failing naming lookup,
225 and exits. Over time, this could result in a denial of service due to
226 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
228 The following bugs are resolved with this release:
230 [1170] localedata: ne_NP: update Nepali locale definition file
231 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
233 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
234 [6796] math: fdim() does not set errno on overflow
235 [10354] libc: posix_spawn should use vfork() in more cases than presently
236 [11213] localedata: localedata: add copyright disclaimer to locale files
237 [12143] localedata: chr_US: new Cherokee locale
238 [12450] localedata: sgs_LT: new locale
239 [12676] localedata: ln_CD: new locale
240 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
242 [13304] math: fma, fmaf, fmal produce wrong results
243 [14259] build: --localedir arg to configure is ignored
244 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
245 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
246 [14934] localedata: es_CL: wrong first weekday chilean locale
247 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
249 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
251 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
252 [15368] nptl: raise() is not async-signal-safe
253 [15479] math: ceil, floor, round and trunc raise inexact exception
254 [15578] localedata: kk_KZ: various updates
255 [16003] localedata: pap_AN: punt old locale
256 [16137] localedata: iw_IL: punt old locale
257 [16190] localedata: eo: new esperanto locale
258 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
259 [16742] malloc: race condition: pthread_atfork() called before first
260 malloc() results in unexpected locking behaviour/deadlocks
261 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
263 [16983] localedata: postal_fmt does not allow %l and %n modifiers
264 [17565] localedata: pt_PT: wrong (work-)week start
265 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
266 [17950] build: Build fails with -msse
267 [18205] localedata: be_BY*: wrong first_weekday and first_workday
268 [18433] libc: posix_spawn does not return correctly upon failure to
270 [18453] localedata: charmaps/IBM875: incorrect codes
271 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
273 [18896] localedata: he_IL: improvements for currency
274 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
276 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
278 [19056] libc: Deprecate readdir_r
279 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
281 [19198] localedata: nl_NL: small improvements for Dutch locales
282 [19257] network: Per-thread memory leak in __res_vinit with IPv6
283 nameservers (CVE-2016-5417)
284 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
286 [19400] locale: Language missing in "iso-639.def", trivial fix in
288 [19431] malloc: Deadlock between fflush, getdelim, and fork
289 [19505] libc: Incorrect file descriptor validity checks in
290 posix_spawn_file_actions_add{open,close,dup2}
291 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
293 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
294 `intl/{gettextP,loadinfo}.h'
295 [19534] libc: execle, execlp may use malloc
296 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
297 [19573] network: res_nclose and __res_maybe_init disagree about name
298 server initialization, breaking Hesiod
299 [19575] localedata: Status of GB18030 tables
300 [19581] localedata: sr_* date_fmt string contains additional newline
301 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
303 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
304 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
305 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
306 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
308 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
310 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
312 [19610] dynamic-link: ldconfig -X removes stale symbolic links
313 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
314 [19633] locale: strfmon_l applies global locale to number formatting
315 [19642] network: Memory leak in getnameinfo
316 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
317 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
319 [19654] math: [x86_64] Need testcase for BZ #19590 fix
320 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
322 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
323 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
325 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
327 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
329 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
330 pointers and lengths in error-case.
331 [19727] locale: Converting from/to UTF-xx with iconv() does not always
332 report errors on UTF-16 surrogates values.
333 [19755] nscd: nscd assertion failure in gc
334 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
335 [19759] libc: mempcpy shouldn't be inlined
336 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
337 [19765] libc: s390 needs an optimized mempcpy
338 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
339 NAME_MAX limit assumption (CVE-2016-1234)
340 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
341 [19787] network: Missing and incorrect truncation checks in getnameinfo
342 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
344 [19791] network: Assertion failure in res_query.c with un-connectable name
346 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
347 [19822] math: libm.so install clobbers old version
348 [19825] network: resolv: send_vc can return uninitialized data in second
349 response to getaddrinfo
350 [19830] network: nss_dns: should check RDATA length against buffer length
351 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
352 confronted with A/AAAA records of invalid size
353 [19837] nss: nss_db: No retries for some long lines with a larger buffer
354 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
355 [19853] stdio: Printing IBM long double in decimal with high precision is
357 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
358 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
359 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
361 [19865] network: Assertion failure or memory leak in
362 _nss_dns_getcanonname_r
363 [19868] network: nss_dns: netent code does not skip over non-PTR records
364 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
366 [19881] string: Improve x86-64 memset
367 [19907] string: Incorrect memcpy tests
368 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
370 [19925] libc: termios.h XCASE namespace
371 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
372 [19929] libc: limits.h NL_NMAX namespace
373 [19931] stdio: Memory leak in vfprintf
374 [19957] libc: clone(CLONE_VM) access invalid parent memory
375 [19963] localedata: en_IL: New locale
376 [19989] stdio: stdio.h cuserid namespace
377 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
379 [19996] locale: langinfo.h nl_langinfo_l namespace
380 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
381 [20010] network: getaddrinfo: Stack overflow in hostent translation
383 [20012] stdio: libio: fmemopen append mode failure
384 [20014] stdio: stdio.h namespace for pre-threads POSIX
385 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
386 [20023] libc: fcntl.h timespec namespace
387 [20024] math: [x86_64] vectorized sincos trashes the stack
388 [20031] network: nss_hesiod: Heap overflow in get_txt_records
389 [20041] time: sys/time.h timespec namespace
390 [20043] libc: unistd.h missing cuserid for UNIX98 and before
391 [20044] libc: unistd.h missing pthread_atfork for UNIX98
392 [20051] libc: ttyslot in wrong header under wrong conditions
393 [20054] libc: gethostname not declared for XPG4
394 [20055] libc: termios.h missing tcgetsid for XPG4
395 [20072] dynamic-link: x86 init_cpu_features is called twice in static
397 [20073] libc: sys/stat.h fchmod namespace
398 [20074] libc: stdlib.h rand_r namespace
399 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
400 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
402 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
403 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
405 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
406 [20119] libc: Wrong mask for processors level type from CPUID
407 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
408 AS not supporting AVX512
409 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
411 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
412 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
414 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
415 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
416 [20178] libc: posix_spawn{p} should not call exit
417 [20191] stdio: libio: vtables hardening
418 [20195] string: FMA4 detection requires CPUID execution with register
420 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
421 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
423 [20212] math: acos (sNaN) returns sNaN
424 [20213] math: asin (sNaN) returns sNaN
425 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
426 [20218] math: [i386] asinhl (sNaN) returns sNaN
427 [20219] math: [i386] atanhl (sNaN) returns sNaN
428 [20222] stdio: fopencookie: Mangle function pointers
429 [20224] math: [i386] cbrtl (sNaN) returns sNaN
430 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
431 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
433 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
434 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
435 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
436 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
437 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
438 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
439 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
440 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
441 [20240] math: modf (sNaN) returns sNaN
442 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
443 [20250] math: frexp (sNaN) returns sNaN
444 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
445 [20255] math: [i386] fdim, fdimf return with excess range and precision /
447 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
448 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
449 subscript is below array bounds [-Werror=array-bounds]
450 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
451 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
453 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
454 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
456 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
458 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
459 [20316] localedata: id_ID: Februari instead of Pebruari
460 [20327] string: POWER8 strcasecmp returns incorrect result
461 [20347] math: Failure: Test: j0_downward (0xap+0)
462 [20348] libc: FAIL: misc/tst-preadvwritev64
463 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
464 [20350] libc: There is no test for p{read,write}64
465 [20357] math: Incorrect cos result for 1.5174239687223976
466 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
470 * Unicode 8.0.0 Support: Character encoding, character type info, and
471 transliteration tables are all updated to Unicode 8.0.0, using new
472 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
473 These updates cause user visible changes, such as the fixes for bugs
474 89, 16061, and 18568.
476 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
477 kernel-internal CPU set size. This means that requests that change the
478 CPU affinity which failed before (for example, an all-ones CPU mask) will
479 now succeed. Applications that need to determine the effective CPU
480 affinities need to call sched_getaffinity or pthread_getaffinity_np after
481 setting it because the kernel can adjust it (and the previous size check
482 would not detect this in the majority of cases).
484 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
485 the following new symbols are used: fts64_children, fts64_close,
486 fts64_open, fts64_read and fts64_set.
488 * getaddrinfo now detects certain invalid responses on an internal netlink
489 socket. If such responses are received, an affected process will
490 terminate with an error message of "Unexpected error <number> on netlink
491 descriptor <number>" or "Unexpected netlink response of size <number> on
492 descriptor <number>". The most likely cause for these errors is a
493 multi-threaded application which erroneously closes and reuses the netlink
494 file descriptor while it is used by getaddrinfo.
496 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
497 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
498 unnecessary serialization of memory allocation requests across threads.
499 The defect is now corrected. Users should see a substantial increase in
500 the concurent throughput of allocation requests for applications which
501 trigger this bug. Affected applications typically create and
502 destroy threads frequently. (Bug 19048 was reported and analyzed by
505 * There is now a --disable-timezone-tools configure option for disabling the
506 building and installing of the timezone related utilities (zic, zdump, and
507 tzselect). This is useful for people who build the timezone data and code
508 independent of the GNU C Library.
510 * The obsolete header <regexp.h> has been removed. Programs that require
511 this header must be updated to use <regex.h> instead.
513 * The obsolete functions bdflush, create_module, get_kernel_syms,
514 query_module and uselib are no longer available to newly linked binaries;
515 the header <sys/kdaemon.h> has been removed. These functions and header
516 were specific to systems using the Linux kernel and could not usefully be
517 used with the GNU C Library on systems with version 2.6 or later of the
520 * Optimized string, wcsmbs and memory functions for IBM z13.
521 Implemented by Stefan Liebler.
523 * Newly linked programs that define a variable called signgam will no longer
524 have it set by the lgamma, lgammaf and lgammal functions. Programs that
525 require signgam to be set by those functions must ensure that they use the
526 variable provided by the GNU C Library and declared in <math.h>, without
527 defining their own copy.
529 * The minimum GCC version that can be used to build this version of the GNU
530 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
531 still be used to compile programs using the GNU C Library.
533 Security related changes:
535 * An out-of-bounds value in a broken-out struct tm argument to strftime no
536 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
538 * The LD_POINTER_GUARD environment variable can no longer be used to disable
539 the pointer guard feature. It is always enabled. Previously,
540 LD_POINTER_GUARD could be used to disable security hardening in binaries
541 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
544 * An integer overflow in hcreate and hcreate_r could lead to an
545 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
547 * The catopen function no longer has unbounded stack usage. Reported by
550 * The nan, nanf and nanl functions no longer have unbounded stack usage
551 depending on the length of the string passed as an argument to the
552 functions. Reported by Joseph Myers. (CVE-2014-9761)
554 * A stack-based buffer overflow was found in libresolv when invoked from
555 libnss_dns, allowing specially crafted DNS responses to seize control
556 of execution flow in the DNS client. The buffer overflow occurs in
557 the functions send_dg (send datagram) and send_vc (send TCP) for the
558 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
559 family. The use of AF_UNSPEC triggers the low-level resolver code to
560 send out two parallel queries for A and AAAA. A mismanagement of the
561 buffers used for those queries could result in the response of a query
562 writing beyond the alloca allocated buffer created by
563 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
564 the overflow. Thanks to the Google Security Team and Red Hat for
565 reporting the security impact of this issue, and Robert Holiday of
566 Ciena for reporting the related bug 18665. (CVE-2015-7547)
568 The following bugs are resolved with this release:
570 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
571 [887] math: Math library function "logb" and "nextafter" inconsistent
572 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
573 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
574 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
575 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
577 [4404] localedata: German translation of "Alarm clock" is misleading
578 [6799] math: nextafter() and nexttoward() doen't set errno on
579 overflow/underflow errors
580 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
582 [10432] nis: _nss_nis_setnetgrent assertion failure
583 [11460] libc: fts has no LFS support
584 [12926] network: getaddrinfo()/make_request() may spin forever
585 [13065] nptl: Race condition in pthread barriers
586 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
587 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
589 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
590 [14912] libc: Rename non-installed bits/*.h headers
591 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
592 [15367] math: Let gcc use __builtin_isinf
593 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
594 [15421] math: lgamma wrongly sets signgam for ISO C
595 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
596 FE_INVALID with argument out of range
597 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
599 [15786] dynamic-link: ifunc resolver functions can smash function
601 [15918] math: Unnecessary check for equality in hypotf()
602 [16061] localedata: Review / update transliteration data
603 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
605 [16141] time: strptime %z offset restriction
606 [16171] math: drem should be alias of remainder
607 [16296] math: fegetround is pure?
608 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
609 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
611 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
612 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
614 [16422] math: [powerpc] math-float, math-double failing llrint tests with
615 "Exception "Inexact" set" on ppc32
616 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
617 [16517] math: Missing underflow exception from tanf/tan/tanl
618 [16519] math: Missing underflow exception from sinhf
619 [16520] math: Missing underflow exception from tanhf
620 [16521] math: Missing underflow exception from exp2
621 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
623 [16734] stdio: fopen calls mmap to allocate its buffer
624 [16961] math: nan function incorrect handling of bad sequences
625 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
626 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
627 [16985] locale: localedef: confusing error message when opening output
629 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
630 [17197] locale: Redundant shift character in iconv conversion output at
632 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
633 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
634 [17250] dynamic-link: static linking breaks nss loading
635 (getaddrinfo/getpwnam/etc...)
636 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
638 [17441] math: isnan() should use __builtin_isnan() in GCC
639 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
640 (related to lock elision)
641 [17787] manual: Exponent on page 324 of the PDF ends prematurely
642 [17886] time: strptime should be able to parse "Z" as a timezone with %z
643 [17887] time: strptime should be able to parse "+01:00" style timezones
644 [17905] libc: catopen() Multiple unbounded stack allocations
646 [18084] libc: backtrace (..., 0) dumps core on x86
647 [18086] libc: nice() sets errno to 0 on success
648 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
649 is too large (CVE-2015-8778)
650 [18251] dynamic-link: SONAME missing when audit modules provides path
651 [18265] libc: add attributes for wchar string and memory functions
652 [18370] math: csqrt missing underflows
653 [18421] libc: [hppa] read-only segment has dynamic relocations
654 [18472] libc: Obsolete syscall wrappers should be compat symbols
655 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
656 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
657 [18525] localedata: Remove locale timezone information
658 [18560] libc: [powerpc] spurious bits/ipc.h definitions
659 [18568] localedata: Update locale data to Unicode 8.0
660 [18589] locale: sort-test.sh fails at random
661 [18595] math: ctan, ctanh missing underflows
662 [18604] libc: assert macro-expands its argument
663 [18610] math: S390: fetestexcept() reports any exception if DXC-code
664 contains a vector instruction exception.
665 [18611] math: j1, jn missing errno setting on underflow
666 [18618] localedata: sync Chechen locale definitions with other *_RU
668 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
669 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
670 [18665] network: In send_dg, the recvfrom function is NOT always using the
671 buffer size of a newly created buffer (CVE-2015-7547)
672 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
674 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
676 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
677 [18699] math: tilegx cproj() for various complex infinities does not yield
679 [18724] libc: Harden put*ent functions against data injection
680 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
681 [18755] build: build errors with -DNDEBUG
682 [18757] stdio: fmemopen fails to set errno on failure
683 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
685 [18781] libc: openat64 lacks O_LARGEFILE
686 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
687 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
688 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
689 [18790] math: [ldbl-128ibm] tanhl inaccurate
690 [18795] libc: stpncpy fortification misses buffer lengths that are
692 [18796] build: build fails for --disable-mathvec
693 [18803] math: hypot missing underflows
694 [18820] stdio: fmemopen may leak memory on failure
695 [18823] math: csqrt spurious underflows
696 [18824] math: fma spurious underflows
697 [18825] math: pow missing underflows
698 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
699 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
700 [18870] build: sem_open.c fails to compile with missing symbol
702 [18872] stdio: Fix memory leak in printf_positional
703 [18873] libc: posix_fallocate overflow check ineffective
704 [18875] math: Excess precision leads incorrect libm
705 [18877] libc: arm: mmap offset regression
706 [18887] libc: memory corruption when using getmntent on blank lines
707 [18918] localedata: hu_HU: change time to HH:MM:SS format
708 [18921] libc: Regression: extraneous stat() and fstat() performed by
710 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
711 binaries (CVE-2015-8777)
712 [18951] math: tgamma missing underflows
713 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
715 [18953] localedata: lt_LT: change currency symbol to the euro
716 [18956] math: powf inaccuracy
717 [18961] math: [i386] exp missing underflows
718 [18966] math: [i386] exp10 missing underflows
719 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
720 [18969] build: multiple string test failures due to missing locale
722 [18970] libc: Reference of pthread_setcancelstate in libc.a
723 [18977] math: float / long double Bessel functions not in XSI POSIX
724 [18980] math: i386 libm functions return with excess range and precision
725 [18981] math: i386 scalb*, ldexp return with excess range and precision
726 [18982] stdio: va_list and vprintf
727 [18985] time: Passing out of range data to strftime() causes a segfault
729 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
730 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
731 [19012] locale: iconv_open leaks memory on error path
732 [19016] math: clog, clog10 inaccuracy
733 [19018] nptl: Mangle function pointers in tls_dtor_list
734 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
735 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
736 [19048] malloc: malloc: arena free list can become cyclic, increasing
738 [19049] math: [powerpc] erfc incorrect zero sign
739 [19050] math: [powerpc] log* incorrect zero sign
740 [19058] math: [x86_64] Link fail with -fopenmp and -flto
741 [19059] math: nexttoward overflow incorrect in non-default rounding modes
742 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
743 [19074] network: Data race in _res_hconf_reorder_addrs
744 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
745 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
746 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
748 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
750 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
751 [19086] manual: posix_fallocate64 documented argument order is wrong.
752 [19088] math: lround, llround missing exceptions close to overflow
754 [19094] math: lrint, llrint missing exceptions close to overflow threshold
755 [19095] math: dbl-64 lrint incorrect for 64-bit long
756 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
757 [19124] dynamic-link: ld.so failed to build with older assmebler
758 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
759 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
760 [19134] math: [powerpc32] lround, lroundf spurious exceptions
761 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
762 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
763 pthread_setaffinity_np
764 [19156] math: [ldbl-128] j0l spurious underflows
765 [19164] nptl: tst-getcpu fails with many possible CPUs
766 [19168] math: math/test-ildoubl and math/test-ldouble failure
767 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
768 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
770 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
771 (FE_NOMASK_ENV) do not clear SSE exceptions
772 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
773 _int_new_arena/reused_arena
774 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
775 [19201] math: dbl-64 remainder incorrect sign of zero result
776 [19205] math: bits/math-finite.h conditions do not match math.h and
778 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
779 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
780 for C99-based standards
781 [19212] libc: features.h not -Wundef clean
782 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
784 [19214] libc: Family and model identification for AMD CPU's are incorrect.
785 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
786 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
788 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
790 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
792 [19242] libc: strtol incorrect in Turkish locales
793 [19243] malloc: reused_arena can pick an arena on the free list, leading
794 to an assertion failure and reference count corruption
795 [19253] time: tzset() ineffective when temporary TZ did not include DST
797 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
798 [19270] math: [hppa] Shared libm missing __isnanl
799 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
800 MAP_HUGETLB and MAP_STACK defines
801 [19313] nptl: Wrong __cpu_mask for x32
802 [19347] libc: grantpt: try to force a specific gid even without pt_chown
803 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
804 [19350] math: [ldbl-128ibm] sinhl spurious overflows
805 [19351] math: [ldbl-128ibm] logl inaccurate near 1
806 [19363] time: x32: times() return value wrongly truncates/sign extends
808 [19367] dynamic-link: Improve branch prediction on Silvermont
809 [19369] network: Default domain name not reset by res_ninit when "search"
810 / "domain" entry is removed from resolv.conf
811 [19375] math: powerpc: incorrect results for POWER7 logb with negative
813 [19385] localedata: bg_BG: time separator should be colon, not comma
814 [19408] libc: linux personality syscall wrapper may erroneously return an
815 error on 32-bit architectures
816 [19415] libc: dladdr returns wrong names on hppa
817 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
818 IBM905, IBM907, and IBM909
819 [19439] math: Unix98 isinf and isnan functions conflict with C++11
820 [19443] build: build failures with -DDEBUG
821 [19451] build: Make check fails on test-double-vlen2
822 [19462] libc: Glibc failed to build with -Os
823 [19465] math: Wrong code with -Os
824 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
826 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
828 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
829 [19486] math: S390: Math tests fail with "Exception Inexact set".
830 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
831 [19550] libc: [mips] mmap negative offset handling inconsistent with other
833 [19590] math: Fail to build shared objects that use libmvec.so functions.
837 * The following bugs are resolved with this release:
839 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
840 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
841 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
842 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
843 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
844 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
845 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
846 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
847 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
848 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
849 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
850 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
851 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
852 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
853 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
854 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
855 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
856 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
857 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
858 18657, 18676, 18694, 18696, 18887.
860 * Cache information can be queried via sysconf() function on s390 e.g. with
861 _SC_LEVEL1_ICACHE_SIZE as argument.
863 * A buffer overflow in gethostbyname_r and related functions performing DNS
864 requests has been fixed. If the NSS functions were called with a
865 misaligned buffer, the buffer length change due to pointer alignment was
866 not taken into account. This could result in application crashes or,
867 potentially arbitrary code execution, using crafted, but syntactically
868 valid DNS responses. (CVE-2015-1781)
870 * The time zone file parser has been made more robust against crafted time
871 zone files, avoiding heap buffer overflows related to the processing of
872 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
873 large time zone data files. Overly long time zone specifiers in the TZ
874 variable no longer result in stack overflows and crashes.
876 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
877 for LD and GD on x86 and x86-64, has been implemented. You will need
878 binutils-2.24 or later to enable this optimization.
880 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
881 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
882 Hat). These updates cause user visible changes, such as the fix for bug
885 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
886 and getXXbyYY NSS calls for the same database, causing a denial-of-service
887 condition in some applications.
889 * Added vector math library named libmvec with the following vectorized x86_64
890 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
892 The library can be disabled with --disable-mathvec. Use of the functions is
893 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
894 Shared library libmvec.so is linked in as needed when using -lm (no need to
895 specify -lmvec explicitly for not static builds).
896 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
898 * A new fmemopen implementation has been added with the goal of POSIX
899 compliance. The new implementation fixes the following long-standing
900 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
901 old implementation is still present for use be by existing binaries.
903 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
904 releases. It has been fixed to match 2.19 and older, but binaries built
905 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
907 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
908 Contributed by Roland McGrath (Google).
910 * The header <regexp.h> is deprecated, and will be removed in a future
911 release. Use of this header will trigger a deprecation warning.
912 Application developers should update their code to use <regex.h> instead.
914 This header was formerly part of SUS, but was deprecated in 1994 and
915 removed from the standard in 2001. Also, the glibc implementation
916 leaks memory. See BZ#18681 for more details.
920 * The following bugs are resolved with this release:
922 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
923 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
924 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
925 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
926 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
927 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
928 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
929 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
930 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
931 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
934 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
935 too little memory for the to-be-scanned arguments and overflow the
936 allocated buffer. The implementation now correctly computes the required
937 buffer size when using malloc, and switches to malloc from alloca as
940 * A new semaphore algorithm has been implemented in generic C code for all
941 machines. Previous custom assembly implementations of semaphore were
942 difficult to reason about or ensure that they were safe. The new version
943 of semaphore supports machines with 64-bit or 32-bit atomic operations.
944 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
945 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
947 * Port to Altera Nios II has been contributed by Mentor Graphics.
949 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
950 implementations for powerpc64/powerpc64le.
951 Implemented by Adhemerval Zanella (IBM).
953 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
954 and powerpc64le. This may improve lock scaling of existing programs on
955 HTM capable systems. The lock elision code is only enabled with
956 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
957 powerpc will issue a transaction abort on every syscall to avoid side
958 effects being visible outside transactions.
960 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
961 AArch64. Contributed by ARM Ltd.
963 * i386 memcpy functions optimized with SSE2 unaligned load/store.
965 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
966 under certain input conditions resulting in the execution of a shell for
967 command substitution when the application did not request it. The
968 implementation now checks WRDE_NOCMD immediately before executing the
969 shell and returns the error WRDE_CMDSUB as expected.
971 * CVE-2012-3406 printf-style functions could run into a stack overflow when
972 processing format strings with a large number of format specifiers.
974 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
975 infinite loop if the DNS response contained a PTR record of an unexpected
978 * The minimum GCC version that can be used to build this version of the GNU
979 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
980 still be used to compile programs using the GNU C Library.
982 * The GNU C Library is now built with -Werror by default. This can be
983 disabled by configuring with --disable-werror.
985 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
987 * The obsolete sigvec function has been removed. This was the original
988 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
989 programs have been using instead for about 25 years. Of course, ABI
990 compatibility for old binaries using sigvec remains intact.
992 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
993 with newer versions of bison.
995 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
996 The original MIPS o32 hard-float ABI requires an FPU where double-precision
997 registers overlay two consecutive single-precision registers. MIPS32R2
998 introduced a new FPU mode (FR=1) where double-precision registers extend the
999 corresponding single-precision registers which is incompatible with the
1000 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1001 require the use of FR=1 making a transition necessary. New o32 ABI
1002 extensions enable users to migrate over time from the original o32 ABI
1003 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1004 tracks the ABI of any loaded object and verifies that new objects are
1005 compatible. Mode transitions will also be requested as required and
1006 unsupportable objects will be rejected. The ABI checks include both soft and
1007 hard float ABIs for o32, n32 and n64.
1009 GCC 5 with GNU binutils 2.25 onwards:
1010 It is strongly recommended that all o32 system libraries are built using the
1011 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1012 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1013 compiler using --with-fp-32=xx to set this by default.
1017 * The following bugs are resolved with this release:
1019 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1020 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1021 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1022 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1023 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1024 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1025 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1026 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1027 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1028 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1029 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1030 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1031 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1032 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1033 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1035 * Reverted change of ABI data structures for s390 and s390x:
1036 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1037 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1038 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1039 are preserved pointing straight to the same implementation as the old ones.
1040 Given that, new callers will simply provide a too-big buffer to these
1041 functions. Any applications/libraries out there that embed jmp_buf or
1042 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1043 against 2.19 headers will have to rebuilt again. This is necessary in any
1044 case to revert the breakage in their ABI caused by the glibc change.
1046 * Support for file description locks is added to systems running the
1047 Linux kernel. The standard file locking interfaces are extended to
1048 operate on file descriptions, not file descriptors, via the use of
1049 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1050 are associated with an open file instead of a process.
1052 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1054 * The minimum Linux kernel version that this version of the GNU C Library
1055 can be used with is 2.6.32.
1057 * Running the testsuite no longer terminates as soon as a test fails.
1058 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1059 with PASS or FAIL lines for individual tests. A summary of the results is
1060 printed, including a list of failing lists, and "make check" exits with
1061 error status if there were any unexpected failures. "make check
1062 stop-on-test-failure=y" may be used to keep the old behavior.
1064 * The am33 port, which had not worked for several years, has been removed
1067 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1068 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1069 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1070 conflicted with POSIX (support for which was removed in 2.19), the
1071 interfaces those macros enabled remain available when compiling with
1072 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1073 test macros defined.
1075 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1077 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1078 This may improve lock scaling of existing programs on TX capable systems.
1079 The lock elision code is only built with --enable-lock-elision=yes and
1080 then requires a GCC version supporting the TX builtins. With lock elision
1081 default mutexes are elided via __builtin_tbegin, if the cpu supports
1082 transactions. By default lock elision is not enabled and the elision code
1085 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1086 copy the path argument. This allowed programs to cause posix_spawn to
1087 deference a dangling pointer, or use an unexpected pathname argument if
1088 the string was modified after the posix_spawn_file_actions_addopen
1091 * All supported architectures now use the main glibc sysdeps directory
1092 instead of some being in a separate "ports" directory (which was
1093 distributed separately before glibc 2.17).
1095 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1096 On configurations that support it (all Linux configurations), it's now
1097 used regardless of the --enable-add-ons switch to configure. It is no
1098 longer possible to build such configurations without pthreads support.
1100 * Locale names, including those obtained from environment variables (LANG
1101 and the LC_* variables), are more tightly checked for proper syntax.
1102 setlocale will now fail (with EINVAL) for locale names that are overly
1103 long, contain slashes without starting with a slash, or contain ".." path
1104 components. (CVE-2014-0475) Previously, some valid locale names were
1105 silently replaced with the "C" locale when running in AT_SECURE mode
1106 (e.g., in a SUID program). This is no longer necessary because of the
1109 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1110 with application code using Intel MPX instructions. (With all previous
1111 versions, the MPX register state could be clobbered when making calls
1112 into or out of a shared library.) Note that while the new dynamic
1113 linker is compatible with all known x86 hardware whether or not it
1114 supports Intel MPX, some x86 instruction-set emulators might fail to
1115 handle the new instruction encodings. This is known to affect Valgrind
1116 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1117 release), and might affect other tools that do instruction emulation.
1119 * Support for loadable gconv transliteration modules has been removed.
1120 The support for transliteration modules has been non-functional for
1121 over a decade, and the removal is prompted by security defects. The
1122 normal gconv conversion modules are still supported. Transliteration
1123 with //TRANSLIT is still possible, and the //IGNORE specifier
1124 continues to be supported. (CVE-2014-5119)
1126 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1127 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1128 resulting a denial-of-service security vulnerability in applications which
1129 use functions related to iconv. (CVE-2014-6040)
1133 * The following bugs are resolved with this release:
1135 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1136 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1137 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1138 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1139 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1140 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1141 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1142 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1143 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1144 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1145 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1146 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1147 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1148 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1149 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1150 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1151 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1152 16431, 16453, 16474, 16506, 16510, 16529
1154 * Slovenian translations for glibc messages have been contributed by the
1155 Translation Project's Slovenian team of translators.
1157 * The public headers no longer use __unused nor __block. This change is to
1158 support compiling programs that are derived from BSD sources and use
1159 __unused internally, and to support compiling with Clang's -fblock
1160 extension which uses __block.
1162 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1163 large collation sequences to optimize multiple passes. This cache
1164 computation may overflow for large collation sequences and may cause a
1165 stack or buffer overflow. This is now fixed to use a slower algorithm
1166 which does not use a cache if there is an integer overflow.
1168 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1169 rules for large collation sequences to optimize multiple passes and falls
1170 back to alloca if malloc fails, resulting in a possible stack overflow.
1171 The implementation now falls back to an uncached collation sequence lookup
1174 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1175 initialized for static applications resulting in the security feature
1176 being disabled. The pointer guard is now correctly initialized to a
1177 random value for static applications. Existing static applications need
1178 to be recompiled to take advantage of the fix (bug 15754).
1180 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1181 to the d_name member of struct dirent, or omit the terminating NUL
1182 character. (Bugzilla #14699).
1184 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1185 aligned_alloc functions could allocate too few bytes or corrupt the
1186 heap when passed very large allocation size values (Bugzilla #15855,
1189 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1190 for AF_INET6 has been fixed (Bugzilla #16072).
1192 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1193 pap_CW, quz_PE, the_NP.
1195 * Substantially revised locales: gd_GB, ht_HT
1197 * The LC_ADDRESS field was updated to support country_car for almost all
1200 * ISO 1427 definitions were updated.
1202 * ISO 3166 definitions were updated.
1204 * The localedef utility now supports --big-endian and --little-endian
1205 command-line options to generate locales for a different system from that
1206 for which the C library was built.
1208 * Binary locale files now only depend on the endianness of the system for
1209 which they are generated and not on other properties of that system. As a
1210 consequence, binary files generated with new localedef may be incompatible
1211 with old versions of the GNU C Library, and binary files generated with
1212 old localedef may be incompatible with this version of the GNU C Library,
1213 in the following circumstances:
1215 + Locale files may be incompatible on m68k systems.
1217 + Locale archive files (but not separate files for individual locales) may
1218 be incompatible on systems where plain "char" is signed.
1220 * The configure option --disable-versioning has been removed. Builds with
1221 --disable-versioning had not worked for several years.
1223 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1224 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1226 * SystemTap probes for malloc have been introduced.
1228 * SystemTap probes for slow multiple precision fallback paths of
1229 transcendental functions have been introduced.
1231 * Support for powerpc64le has been added.
1233 * The soft-float powerpc port now supports e500 processors.
1235 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1237 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1238 set of header declarations that are enabled by default, even when other
1239 feature test macros or compiler options such as -std=c99 would otherwise
1240 disable some of those declarations.
1242 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1243 conflict with POSIX. The libbsd-compat library (which was a dummy library
1244 that did nothing) has also been removed.
1246 * Preliminary documentation about Multi-Thread, Async-Signal and
1247 Async-Cancel Safety has been added.
1249 * Change of ABI data structures for s390 and s390x:
1250 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1251 allow for future hardware extensions. All library functions that accept or
1252 return these structures were versioned in order to provide backwards
1253 compatibility with existing code. However, not all cases can be handled
1254 with interface versioning. If an application embeds either structure into
1255 another structure and that structure is passed to another compilation unit
1256 which expects the newer larger-sized object then any access to the new
1257 fields will result in undefined behaviour. Similarly any access to fields
1258 that were moved by the enlarging of the structures will cause undefined
1259 behaviour. To fix the undefined behaviour all objects that are part of the
1260 final in-memory image must agree on the size of structures, and this may
1261 require recompilation.
1265 * The following bugs are resolved with this release:
1267 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1268 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1269 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1270 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1271 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1272 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1273 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1274 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1275 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1276 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1277 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1278 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1279 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1282 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1283 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1284 Distributions can re-enable building and using pt_chown via the new configure
1285 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1286 considerable security risks and should only be used if the distribution
1287 understands and accepts the risks.
1289 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1292 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1293 fixed (Bugzilla #15330).
1295 * Add support for calling C++11 thread_local object destructors on thread
1296 and program exit. This needs compiler support for offloading C++11
1297 destructor calls to glibc.
1299 * Improved worst case performance of libm functions with double inputs and
1302 * Support for priority inherited mutexes in pthread condition variables on
1303 non-x86 architectures.
1305 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1307 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1309 * Optimized string functions for ARM. Implemented by Will Newton and
1312 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1314 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1317 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1318 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1320 * Added a benchmark framework to track performance of functions in glibc.
1322 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1323 It is based on draft TS 18661 and currently enabled as a GNU extension.
1325 * On Linux, the clock function now uses the clock_gettime system call
1326 for improved precision, rather than old times system call.
1328 * Added support for version-3 tz format files. This is needed when using
1329 the tz database release 2013e or later, and affects a few unusual cases --
1330 currently only TZ='America/Godthab' for time stamps after 2037.
1332 * Added new API functions pthread_getattr_default_np and
1333 pthread_setattr_default_np to get and set the default pthread
1334 attributes of a process.
1336 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1337 This may improve lock scaling of existing programs on TSX capable systems.
1338 When the --enable-lock-elision=yes parameter is specified at configure
1339 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1342 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1343 availability). Implemented by Ryan S. Arnold.
1345 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1349 * The following bugs are resolved with this release:
1351 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1352 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1353 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1354 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1355 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1356 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1357 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1358 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1359 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1360 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1361 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1362 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1363 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1364 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1365 14879, 14889, 14893, 14898, 14914.
1367 * Optimization of memcpy for MIPS.
1369 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1370 EMFILE has been fixed (Bugzilla #14889).
1372 * The add-on ports collection is now distributed in the "ports" subdirectory
1373 of the main GNU C Library distribution, rather than separately.
1375 * Port to ARM AArch64 contributed by Linaro.
1377 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1378 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1380 Implemented by Andreas Krebbel.
1382 * The new function secure_getenv allows secure access to the environment,
1383 returning NULL if running in a SUID/SGID process. This function replaces
1384 the internal function __secure_getenv.
1386 * SystemTap static probes have been added into the dynamic linker.
1387 Implemented by Gary Benson.
1389 * Optimizations of string functions strstr, strcasestr and memmem.
1390 Implemented by Maxim Kuvyrkov.
1392 * The minimum Linux kernel version that this version of the GNU C Library
1393 can be used with is 2.6.16.
1395 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1396 powerpc POWER7. Implemented by Will Schmidt.
1398 * New configure option --disable-nscd builds the C library such that it
1399 never attempts to contact the Name Service Caching Daemon (nscd).
1400 New configure option --disable-build-nscd avoids building nscd itself;
1401 this is the default if --disable-nscd is used.
1403 * Improved support for cross-compilation, including cross-testing and
1404 bootstrap builds without a previously built glibc.
1406 * Several testsuite tests are now able to test multiple IFUNC variants of an
1407 interface, rather than just testing the one that would be chooen by
1410 * New configure options --with-bugurl and --with-pkgversion, for
1411 distributors to use to embed their bug-reporting and package version
1412 information in --help and --version output.
1414 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1415 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1416 allows creation of chroots without the procfs mounted on /proc.
1418 * The `crypt' function now fails if passed salt bytes that violate the
1419 specification for those values. On Linux, the `crypt' function will
1420 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1421 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1422 when the mode is enabled.
1424 * The `clock_*' suite of functions (declared in <time.h>) is now available
1425 directly in the main C library. Previously it was necessary to link with
1426 -lrt to use these functions. This change has the effect that a
1427 single-threaded program that uses a function such as `clock_gettime' (and
1428 is not linked with -lrt) will no longer implicitly load the pthreads
1429 library at runtime and so will not suffer the overheads associated with
1430 multi-thread support in other code such as the C++ runtime library.
1432 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1437 * The following bugs are resolved with this release:
1439 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1440 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1441 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1442 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1443 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1444 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1445 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1446 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1447 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1448 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1449 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1450 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1451 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1452 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1453 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1454 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1455 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1456 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1457 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1458 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1459 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1460 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1463 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1464 configuring glibc with:
1465 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1466 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1467 Implemented by H.J. Lu.
1471 + define static_assert
1473 + do not declare gets
1475 + declare at_quick_exit and quick_exit also for ISO C11
1477 + aligned_alloc. NB: The code is deliberately allows the size parameter
1478 to not be a multiple of the alignment. This is a moronic requirement
1479 in the standard but it is only a requirement on the caller, not the
1482 + timespec_get added
1484 + uchar.h support added
1486 + CMPLX, CMPLXF, CMPLXL added
1488 Implemented by Ulrich Drepper.
1490 * Support for the IA-64 has been moved to ports.
1492 * Remove support for anything but ELF binary format
1494 * Checking versions of poll, ppoll added.
1495 Implemented by Ulrich Drepper.
1497 * More generic and 64-bit performance optimizations to math functions.
1498 Implemented by Ulrich Drepper.
1500 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1501 headers and functions available at compile time as they were before
1502 version 2.14. This option will be removed at some time in the future
1503 after the TI-RPC library becomes fully sufficient for the needs of
1504 existing applications.
1506 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1507 Note that glibc is not expected to work with any Linux kernel version
1510 * New header <sys/auxv.h> and function getauxval allowing easy access to
1511 the AT_* key-value pairs passed from the Linux kernel. The header also
1512 defines the HWCAP_* bits associated with the AT_HWCAP key.
1514 * New locales: mag_IN
1516 * New configure option --enable-systemtap builds SystemTap static probes
1517 into libc for setjmp and longjmp and into libpthread for various operations.
1518 So far the setjmp/longjmp probes and some of the libpthread probes are
1519 provided only for i*86 and x86_64.
1520 Implemented by Roland McGrath and Rayson Ho.
1522 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1524 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1527 * More optimized functions for SPARC. Implemented by David S. Miller.
1529 * Improved support for cross-compilation, especially bootstrap builds
1530 without a previously built glibc.
1532 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1533 Chris Metcalf from Tilera.
1535 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1536 now supported for ARM processors.
1538 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1539 as the name of the dynamic linker, to distinguish it from the
1540 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1542 * Support for CRIS has been removed from ports.
1544 * A new class of installed header has been documented for low-level
1545 platform-specific functionality. PowerPC added the first instance with a
1546 function to provide time base register access. Contributed by Tulio
1547 Magno Quites Machado Filho.
1549 * ix86 configurations now install header files that are consistent with
1550 what x86-64 configurations install. These same header files can be used
1551 for -m32, -m64, or -mx32 builds.
1552 Contributed by H.J. Lu.
1554 * Math library bug fixes. A thorough audit of all open math library bugs was
1555 conducted by Joseph Myers. Significant progress was made on many math
1556 library bugs resulting in more accurate exceptions and function results.
1557 Many thanks to all those that contributed including Andreas Jaeger for his
1558 patch review and work on the x87 trigonometric instruction issues.
1560 * Timezone data is no longer installed. Timezone-related binaries and scripts
1561 will continue to be installed. Users should obtain their timezone data from
1562 their distribution provider or from the tzdata package at
1563 <ftp://munnari.oz.au/pub/>.
1567 * The following bugs are resolved with this release:
1569 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1570 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1571 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1572 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1573 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1574 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1575 13484, 13506, 13515, 13523, 13524, 13538, 13540
1577 * New program pldd to list loaded object of a process
1578 Implemented by Ulrich Drepper.
1580 * Add nss_db support back to glibc. No more dependency on Berkeley db
1581 and support for initgroups lookups.
1582 Implemented by Ulrich Drepper.
1584 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1585 Contributed by HJ Lu.
1587 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1588 Contributed by HJ Lu.
1590 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1591 on x86-32 and x86-64.
1592 Contributed by Liubov Dmitrieva.
1594 * Optimized strchr and strrchr for SSE on x86-32.
1595 Contributed by Liubov Dmitrieva.
1597 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1598 for x86-64 and x86-32.
1599 Contributed by Liubov Dmitrieva.
1601 * New interfaces: scandirat, scandirat64
1602 Implemented by Ulrich Drepper.
1604 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1605 Implemented by Ulrich Drepper.
1607 * nscd now also caches the netgroup database.
1608 Implemented by Ulrich Drepper.
1610 * Integrate libm with gcc's -ffinite-math-only option.
1611 Implemented by Ulrich Drepper.
1613 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1614 to math functions. Implemented by Ulrich Drepper.
1616 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1617 Implemented by Ulrich Drepper.
1619 * New Linux interfaces: process_vm_readv, process_vm_writev
1621 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1622 Implemented by Ulrich Drepper.
1624 * Optimized nearbyint and strcasecmp for PPC.
1625 Implemented by Adhemerval Zanella.
1627 * New locales: bho_IN, unm_US, es_CU, ta_LK
1631 * The following bugs are resolved with this release:
1633 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1634 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1635 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1636 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1637 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1638 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1639 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1640 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1641 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1642 12795, 12811, 12813, 12814, 12841
1644 * The RPC implementation in libc is obsoleted. Old programs keep working
1645 but new programs cannot be linked with the routines in libc anymore.
1646 Programs in need of RPC functionality must be linked against TI-RPC.
1647 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1649 Visible changes of this change include (obviously) the inability to link
1650 programs using RPC functions without referencing the TI-RPC library and the
1651 removal of the RPC headers from the glibc headers.
1652 Implemented by Ulrich Drepper.
1654 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1655 syncfs, setns, sendmmsg
1657 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1658 yue_HK, lij_IT, mhr_RU
1660 * New encodings: CP770, CP771, CP772, CP773, CP774
1662 * New program sotruss to trace calls through PLTs
1663 Implemented by Ulrich Drepper.
1665 * The malloc hook implementation is marked deprecated and will be removed
1666 from the default implementation in the next version. The design never
1667 worked ever since the introduction of threads. Even programs which do
1668 not create threads themselves can use multiple threads created internally.
1672 * The following bugs are resolved with this release:
1674 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1675 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1676 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1677 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1680 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1682 * POWER7 optimizations: memset, memcmp, strncmp
1684 * New optimized string functions for x86-64: strnlen (SSE2),
1685 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1686 Implemented by Ulrich Drepper.
1690 * The following bugs are resolved with this release:
1692 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1693 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1694 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1695 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1696 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1697 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1698 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1699 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1701 * New interfaces: pthread_getname_np, pthread_setname_np
1703 * New Linux interface: recvmmsg
1705 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1707 * The dynamic linker now recognizes supported ABI versions from the
1708 EI_ABIVERSION field in the ELF header.
1709 Implemented by Ulrich Drepper.
1711 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1712 will not be used to fill in password fields in the passwd.byname replies.
1713 Instead it is used to synthesize the shadow.byname table, should it be
1714 missing. This is a useful mode in some installations involving Solaris.
1715 Implemented by Ulrich Drepper.
1717 * New locales: kok_IN, sq_MK, cv_RU
1721 * The following bugs are resolved with this release:
1723 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1724 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1725 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1726 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1727 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1728 10789, 10817, 10823, 10840, 10847
1730 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1732 Implemented by Ulrich Drepper.
1734 * Checking version of longjmp added that fails if an uninitialized stack
1735 frame would be created. Implemented by Ulrich Drepper.
1737 * STT_GNU_IFUNC is now supported in static executables.
1738 Implemented by H.J. Lu.
1740 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1742 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1743 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1744 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1745 Contributed by H.J. Lu.
1747 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1748 Implemented by Ulrich Drepper.
1750 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1752 Contributed by H.J. Lu.
1754 * Support for fma instruction in AVX on x86-64.
1755 Implemented by H.J. Lu and Ulrich Drepper.
1757 * AVX support in x86-64 auditing support in ld.so.
1758 Implemented by H.J. Lu.
1760 * STB_GNU_UNIQUE support added.
1761 Implemented by Ulrich Drepper.
1763 * Implement second fallback mode for optimized DNS lookup for even more
1764 broken environments. If two requests from the same port are not
1765 handled correctly close the socket and open a new one before sending
1766 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1767 can be used to select this mode right away, instead of rediscovering the
1768 necessity is every process again.
1769 Implemented by Ulrich Drepper.
1771 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1772 Implemented by Adam Tkac.
1774 * Optimized iconv conversions for S390x.
1775 Implemented by Andreas Krebbel.
1777 * Using condvars with PI mutexes is now more efficient due to kernel
1778 support for requeueing to PI futexes. NPTL support added for x86-64.
1779 Implemented by Ulrich Drepper.
1781 * New locales: ps_AF, my_MM
1785 * The following bugs are resolved with this release:
1787 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1788 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1789 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1790 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1791 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1793 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1794 Implemented by Ulrich Drepper.
1796 * New Linux interfaces: accept4, fallocate, fallocate64.
1797 Implemented by Ulrich Drepper.
1799 * Correct declarations of string function when used in C++ code. This
1800 could lead to compile errors for invalid C++ code.
1802 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1805 * New POSIX 2008 interface: psiginfo
1806 Implemented by Ulrich Drepper.
1808 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1809 Implemented by Ulrich Drepper.
1811 * Support for selecting between multiple function definitions at runtime
1812 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1814 * The libcrypt library can now use the hash function implementations in
1815 NSS. Implemented by Ulrich Drepper.
1817 * The malloc implementation can be compiled to be less memory efficient
1818 but higher performing in multi-threaded programs.
1819 Implemented by Ulrich Drepper.
1821 * New locales: nan_TW@latin, ks_IN
1823 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1824 Implemented by Ulrich Drepper.
1826 * Extended printf hook support. It is possible to use user-defined types
1827 and extend existing format specifiers.
1828 Implemented by Ulrich Drepper.
1830 * Handling for group shadow files has been added.
1831 Implemented by Ulrich Drepper.
1833 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1834 servers (the case, e.g., for some people using the built-in DNS
1835 server in ADSL modems/routers). There is a once-per-process timeout
1836 in case of a broken server. To avoid it, users can run nscd or put
1837 'options single-request' in /etc/resolv.conf.
1838 Implemented by Ulrich Drepper.
1842 * The following bugs are resolved with this release:
1844 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1845 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1846 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1847 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1848 6968, 6974, 6980, 6995, 7008, 7009, 7029
1850 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1851 up at the same time. Implemented by Ulrich Drepper.
1853 * TLS descriptors for LD and GD on x86 and x86-64.
1854 Implemented by Alexandre Oliva.
1856 * getaddrinfo now handles DCCP and UDPlite.
1857 Implemented by Ulrich Drepper.
1859 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1860 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1861 Implemented by Ulrich Drepper.
1863 * New implementation of memmem, strstr, and strcasestr which is O(n).
1864 Implemented by Eric Blake.
1866 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1868 * Implement "e" option for popen to open file descriptor with the
1869 close-on-exec flag set. Implemented by Ulrich Drepper.
1871 * Implement "b" mode for fmemopen. In this mode writes writes don't
1872 implicitly add a NUL byte and seeks from the end of the buffer really
1873 use the buffer end, not the string length as the basis.
1874 Implemented by Ulrich Drepper.
1876 * Many functions, exported and internal, now atomically set the close-on-exec
1877 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1879 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1881 Implemented by Pravin Satpute.
1883 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1887 * The following bugs are resolved with this release:
1889 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1890 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1891 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1892 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1893 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1894 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1895 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1897 * New locales: bo_CN, bo_IN, shs_CA.
1899 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1901 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1902 Implemented by Pravin Satpute.
1904 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1906 * nscd caches now all timeouts for DNS entries
1907 Implemented by Ulrich Drepper.
1909 * nscd is more efficient and wakes up less often.
1910 Implemented by Ulrich Drepper.
1912 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1913 vdprintf, and obstack_vprintf.
1914 Implemented by Jakub Jelinek.
1916 * Faster memset for x86-64.
1917 Implemented by Harsha Jagasia and H.J. Lu.
1919 * Faster memcpy on x86.
1920 Implemented by Ulrich Drepper.
1922 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1923 Implemented by Ulrich Drepper.
1925 * Faster sqrt and sqrtf implemention for some PPC variants.
1926 Implemented by Steven Munroe.
1930 * The following bugs are resolved with this release:
1932 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1933 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1934 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1935 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1936 5071, 5103, 5104, 5112, 5113, 5184, 5186
1938 * More checking functions: fread, fread_unlocked, open*, mq_open.
1939 Implemented by Jakub Jelinek and Ulrich Drepper.
1941 * Extend fortification to C++. Implemented by Jakub Jelinek.
1943 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1944 by not recognizing 'a' as a modifier when those specs are requested.
1945 Implemented by Jakub Jelinek.
1947 * PPC optimizations to math and string functions.
1948 Implemented by Steven Munroe.
1950 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1951 options to be passed. Implemented by Ulrich Drepper.
1953 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1955 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1956 Implemented by Ulrich Drepper.
1958 * Handle private futexes in the NPTL implementation.
1959 Implemented by Jakub Jelinek and Ulrich Drepper.
1961 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1962 Implemented by Roland McGrath and Ulrich Drepper.
1964 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1966 * SHA-256 and SHA-512 based password encryption.
1967 Implemented by Ulrich Drepper.
1969 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1970 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1973 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1974 Implemented by Ulrich Drepper.
1978 * The following bugs are resolved with this release:
1980 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
1981 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
1982 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
1983 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
1984 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
1985 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
1986 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
1989 * New Linux interfaces: epoll_pwait, sched_getcpu.
1991 * New generic interfaces: strerror_l.
1993 * nscd can now cache the services database. Implemented by Ulrich Drepper.
1997 * The following bugs are resolved with this release:
1999 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2000 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2001 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2002 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2003 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2004 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2005 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2006 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2007 3225, 3251, 3252, 3253, 3273
2009 * For Linux, the sorting of addresses returned by getaddrinfo now also
2010 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2011 Implemented by Ulrich Drepper.
2013 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2014 Implemented by Ulrich Drepper.
2016 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2018 * New iconv module for MIK. Contributed by Alexander Shopov.
2020 * For sites with broken group and/or passwd database, the auto-propagate
2021 option of nscd can prevent creating ID lookup entries from the results
2022 of a name lookup and vice versa. This usually is no problem but some
2023 site might have problems with the default behavior.
2024 Implemented by Ulrich Drepper.
2026 * Iterating over entire database in NIS can be slow. With the
2027 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2028 to trade time for memory. The entire database will be read at once.
2029 Implemented by Ulrich Drepper.
2031 * The interfaces introduced in RFC 3542 have been implemented by
2034 * Support for the new ELF hash table format was added by Ulrich Drepper.
2036 * Support for priority inheritance mutexes added by Jakub Jelinek and
2039 * Support for priority protected mutexes added by Jakub Jelinek.
2041 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2042 or_IN, csb_PL, fy_NL, sr_ME.
2046 * More overflow detection functions.
2048 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2049 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2051 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2052 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2053 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2054 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2055 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2056 by Masahide Washizawa.
2058 * It is now possible to install audit libraries for the dynamic linker using
2059 LD_AUDIT. Implemented by Ulrich Drepper.
2061 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2062 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2063 For a libc and libpthread that works well on Linux 2.4 kernels, we
2064 recommend using the stable 2.3 branch.
2066 * The new function open_wmemstream acts like open_memstream,
2067 but uses a wchar_t wide character buffer.
2069 * The new function ppoll is an improved version of poll, similar to pselect.
2071 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2072 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2073 renameat, unlinkat, linkat, symlinkat, readlinkat.
2075 * New Linux kernel system calls: unshare,
2076 inotify_init, inotify_add_watch, inotify_rm_watch.
2078 * The euidaccess function is now also known by the alias eaccess,
2079 for compatibility with some other systems.
2081 * Timezone data updated to 2006b version.
2085 * The following bugs are resolved with this release:
2087 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2088 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2089 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2090 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2091 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2092 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2094 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2096 * As of this release, GCC 4 can be used to compile the C Library.
2098 * Timezone data updated to 2005m version.
2102 * The following bugs are resolved with this release:
2104 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2105 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2106 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2107 777, 787, 821, 822, 823, 825
2109 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2113 * Support for RFC 3678. Real implementations exist only for Linux so far.
2114 Implemented by Ulrich Drepper.
2116 * nscd can now cache entries persistently. Expiring entries are reloaded.
2117 For speedups the cache can be shared in memory with client processes.
2118 Implemented by Ulrich Drepper.
2120 * nscd can now perform SELinux checks.
2121 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2123 * getaddrinfo queries are now cached. Canonical name lookup is performed
2125 Implemented by Ulrich Drepper.
2127 * The nothrow function attribute is used when headers are used by gcc when
2128 compiling C code. This can avoid significant amounts of exception
2131 * The malloc functions perform more error checking and are stricter when
2132 it comes to reacting on errors. The default action is to terminate
2133 the process after showing an error message. Implemented by Ulrich Drepper.
2135 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2136 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2138 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2139 namespaces using the new function dlmopen(). This feature is of course,
2140 like most other dynamic loading functionality, not available in statically
2141 linked applications. Implemented by Ulrich Drepper.
2143 * Low-overhead boundary checking variants of string and some stdio functions
2144 were added. These are to be used in conjunction with a gcc patch by
2145 Jakub Jelinek which adds calls to these functions if possible.
2146 Implemented by Jakub Jelinek and Ulrich Drepper.
2148 * Old code for several operating systems and machine architectures that
2149 have not been in working condition in a long time have been removed from
2150 the main source tree maintained by the GNU C Library's maintainers.
2151 These files are now reside in the separate `ports' source module
2152 that is usable as an add-on when building the library.
2156 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2157 interrogate the dynamic linker, compatible with the Solaris interface.
2159 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2160 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2162 * getifaddrs now uses the netlink interface on Linux to get its information.
2163 Implemented by Thorsten Kukuk.
2165 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2166 Implemented by Ulrich Drepper.
2168 * support for non-executable stacks on x86 has been added. Changes mostly
2171 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2174 * getaddrinfo now performs destination address selection according to
2179 * Thread-safe interfaces for many functions that access locale data
2180 were added in version 2.3, but these features were omitted from NEWS.
2181 Many functions have variants with an `_l' suffix that take a `locale_t'
2182 object as a parameter rather than consulting the current locale.
2183 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2184 create and maintain `locale_t' objects. Additionally, the new function
2185 `uselocale' sets "the current locale" (as used by functions not so
2186 parameterized) set for an individual thread. These features were added
2187 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2189 * The functions getresuid, getresgid, setresuid, and setresgid, which
2190 have long been available on Linux, are now declared in <unistd.h>
2191 and are now also available on the Hurd.
2193 * ELF thread-local storage support (TLS) now works on x86-64.
2195 * The new dynamic string token $LIB is expanded in shared library names.
2196 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2198 * Aldy Hernandez contributed complete software floating point support for
2199 PowerPC machines with no FPU.
2201 * fexecve is implemented on Linux.
2203 * The `btowc' function should work at least twice as fast due to
2204 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2206 * With approriate thread add-ons cancelable functions are now implemented
2207 in libc.so as well. No need to call the function in libpthread. This
2208 change allowed to finally disable the incorrect and expensive handling
2209 of weak definition in ld.so.
2211 * Yet more PLT entries in libc.so have been removed. We finally arrived
2212 at the bare minimum. Startup times improved appropriately.
2214 * Support for the new Linux/x86 system call interface was added. The
2215 AT_SYSINFO auxiliary vector entry is recognized and handled.
2219 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2222 * iconv (the program and the interface) now accepts empty names (excluding
2223 options like //TRANSLIT) to mean "use charset of current locale".
2225 * localedef can now transliterate characters in strings which are not in
2226 the provided charmap. The information from the input locale is used.
2228 * Prelinking support was added for ELF targets. This requires additional
2229 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2231 * Read-only stdio streams now use mmap to speed up operation by eliminating
2232 copying and buffer underflows. To use add 'm' to the mode string of
2233 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2235 * The malloc functions were completely rewritten by Wolfram Gloger based
2236 on Doug Lea's malloc-2.7.0.c.
2238 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2239 implementation of regex.
2241 * Bruno Haible upgraded the iconv and locale implementation to support
2244 * Contents of the LC_* and LANG environment variables in the CEN style are
2245 not recognized anymore. It never was used. Change by Ulrich Drepper.
2247 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2248 thread-local storage (TLS) ABI on some platforms.
2249 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2251 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2252 EUC-JISX0213, and TSCII.
2254 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2255 BSD-compatible interface for getting all network interface addresses.
2256 Implementation for IPv4 by Roland McGrath.
2258 * Loading of locale data is faster due to the introduction of a locale
2259 archive. Implemented by Roland McGrath and Ulrich Drepper.
2261 * Startup times are significantly reduced by not using exported functions
2262 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2265 * Steven Munroe contributed a port to PowerPC64/Linux.
2269 * The Hurd now uses the GNU libio implementation of stdio rather than the
2270 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2272 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2273 with the same behavior as the Linux system call of the same name.
2277 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2278 128-bit long double format.
2280 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2281 IBM1161, and IBM1162 charsets.
2283 * Andreas Jaeger contributed a port to x86-64/Linux.
2285 * Peter Bruin contributed a port to PowerPC/Hurd.
2287 * libc functions using I/O streams now can handle wide-oriented streams
2290 * optimizations in the dynamic linker. Binaries created by recent binutils
2291 versions start up quicker due to reduced time spent on relocations.
2293 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2297 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2298 asinh, atanh, j0 for the 128-bit long double format.
2300 * Bruno Haible updated all the code handling Unicode in some form to
2301 support Unicode 3.1.
2303 * Speed of regex for single-byte locales is back to previous levels.
2304 Patch by Isamu Hasegawa.
2306 * Alpha, SPARC, and IA-64 now also using floating stacks.
2308 * Startup time of internationalized applications greatly improved through
2309 iconv cache. Use iconvconfig to generate the cache file.
2310 Contributed by Ulrich Drepper.
2312 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2313 severe performance problems. Patch by David Mosberger.
2315 * The Hurd port got a lot more functionality like AIO, various stdio
2316 extensions, etc. Mainly done by Roland McGrath.
2318 * mtrace can now lookup symbols in shared libraries.
2322 * Intel's IA-64 math library is largely integrated. It provides fast and
2323 accurate implementatations for most basic and standard math functions
2324 in float, double, and long double format.
2326 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2327 and asin for the 96-bit long double format and asin, log, tan for the
2328 128-bit long double format.
2330 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2331 The basic double functions exist today. Contributed by Abraham Ziv
2332 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2333 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2335 * An asynchronous name lookup library was added. The interface is designed
2336 after POSIX AIO. The proposal was circulated beforehand to get comments.
2337 No negative ones came in. Implemented by Ulrich Drepper.
2339 * Port to S390/64bit contributed by Martin Schwidefsky
2340 <schwidefsky@de.ibm.com>.
2342 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2343 of functions for Linux/IA-64.
2345 * The RPC code is now thread safe. Threads can now use the same service
2346 of different services at the same time. Patch by Eric Norum
2347 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2349 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2350 family of functions for Linux/S390.
2352 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2353 of functions for Linux/x86.
2355 * Port to Linux/CRIS contributed by Axis Communications.
2359 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2360 we can now check for namespace violations and missing declarations. The
2361 result is that almost all headers are now Unix-compliant (as defined in
2362 the upcoming XPG6). The negative side is that some programs might need
2363 corrections, too, if they depend on the incorrect form of the headers in
2364 previous versions which defined too many symbols and included too many
2367 * regex now handles multibyte character sets correctly.
2368 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2370 * iconv (the program) does now conform to the upcoming XPG6 and handles
2371 charmaps. Instead of the charset names the path of charmaps can be
2372 provided and the conversion happens based on this data.
2373 Contributed by Ulrich Drepper.
2375 * The locale program now provides detailed information about the installed
2380 only lists the names of the supported locales
2384 provides details such as country, language, and codeset name.
2385 Contributed by Ulrich Drepper.
2389 * The gencat program now parses the input file according to the charset
2390 selected by the LC_CTYPE category. This is important for stateful
2391 character sets. To make generating catalogs easier there is a way
2392 to overwrite the charset selected by the locale: before the first
2393 message or $ quote line the catalog can contain a line like
2395 $ codeset=ISO-8859-2
2397 to select the charset (ISO-8859-2 in this case).
2399 Implemented by Shinya Hanataka and Ulrich Drepper.
2401 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2402 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2403 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2405 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2406 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2408 * The LANGUAGE environment variable is now ignored unless the locale is
2409 changed from the default "C" locale.
2411 * The usual bug fixes.
2415 * Greg McGary added runtime support for bounds checking using gcc's
2416 new -fbounded-pointers option. ix86 target is complete. PowerPC
2419 * Thorsten Kukuk added secure mode support to nscd.
2421 * The Berkeley db libraries have been removed.
2423 Related, the nss_db module is now in a separate package since it
2424 obviously requires a database library being available.
2426 * Wide character I/O streams implemented by Ulrich Drepper.
2428 * Functions from the extended socket API added by Ulrich Drepper.
2430 * Functions feenableexcept and fedisableexcept to control the
2431 behaviour of individual exceptions have been added by Andreas Jaeger.
2433 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2435 * The resolver code has been updated from bind 8.2.3-T5B which supports
2436 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2439 This change could in some situations effect backward compatibility. Since
2440 now `_res' is a thread-local instead of a global variable, modifying it
2441 in one thread does not have any effect in other threads.
2443 The resolver library was also extended to allow IPv6 as the transport
2444 protocol for the requests. This work was done by Stig Venaas.
2446 * Compatibility code for K&R C compilers has been removed from the
2447 header files. A ISO C compiler is needed to use the library
2448 (conforming to either C89 or C99 standard).
2450 * Complete rewrite of the localedef program to support multibyte character
2451 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2452 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2453 Implemented by Ulrich Drepper.
2455 Bruno Haible significantly improved the generation and use of the data
2456 structures for the wide character tables.
2458 * Plural handling in gettext implemented by Ulrich Drepper.
2460 * The utmp daemon has been removed.
2462 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2464 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2467 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2469 * POSIX barriers implemented by Kaz Kylheku.
2471 * POSIX spawn function family implemented by Ulrich Drepper.
2473 * POSIX spinlocks are now available.
2475 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2477 * the configure option --enable-kernel=X.Y.Z allows to strip out
2478 compatibility for kernel versions before X.Y.Z. This is currently only
2479 implemented for Linux.
2481 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2482 field sin6_scope_id. This means that all programs using IPv6 should be
2483 recompiled. Don't expect binary compatibility with previous glibc
2486 * various conversion modules for IBM character sets contributed by
2489 * IA-64 port by Jes Sorensen and HJ Lu.
2503 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2505 * New xdr functions are added; some rpc functions are now 64bit clean.
2507 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2509 * Fixed known incompatibilities with glibc 2.0.
2511 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2513 * Optimized a number of functions (especially the ELF dynamic loader).
2515 * Update timezone data files.
2517 * lots of charmaps corrections
2519 * some new locale definitions and charmaps
2524 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2525 conform to POSIX member type requirements. Symbol versions have been
2526 adjusted as needed within the library, and for direct use by applications,
2527 but there is potential for problems if third-party libraries use
2528 struct timeval as part of their interface. This does not present
2529 a problem for X and other "essential" system libraries.
2531 * An additional locale model to support C++ Standard Library locale
2532 model and probably more was implemented by Ulrich Drepper.
2534 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2537 * Miles Bader provided the `argp' function family to support hierachical
2538 command line argument parsing, layered on top of getopt.
2540 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2542 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2545 * scanf recognizes the %a and %A format for scanning floating point numbers.
2547 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2548 information and interfaces for the available integer types.
2550 * about 130 new math functions were added to implement the ISO C9x math
2553 * the new header <complex.h> contains definitions of the complex math
2554 functions from ISO C 9X.
2556 * the new header <tgmath.h> defines generic macros to use complex or
2557 real valued functions.
2559 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2561 * Andreas Jaeger provided a test suite for the math library.
2563 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2565 * Ulrich Drepper added character set conversion functions (iconv).
2567 * Optimized string functions have been added.
2569 * The localedata addon is now part of glibc.
2571 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2573 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2574 daemon for NSS (nscd).
2576 Missing a better place here are some numbers on improvements. Under
2577 Linux 2.1.125 un-tar-ing the kernel sources takes
2581 using local files 12.19s 6.88s 22.91s
2583 using NIS 13.92s 8.91s 26.34s
2585 using NIS & nscd 10.37s 7.34s 25.30s
2587 using NIS+ 27.57s 30.37s 640.46s
2589 using NIS+ & nscd 10.25s 7.83s 26.51s
2591 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2593 Keep in mind that non-namelookup related operations dominate above times.
2594 It was just a common complain that using NIS+ unpacking the kernel is
2597 [1] The old nscd implementation is not available anymore since it was
2598 distributed with glibc up to version 2.0.98 and thus is now replaced.
2600 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2602 * Mark Kettenis provided a Hesiod NSS module.
2604 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2605 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2607 * rcmd can now handle netgroups (Dick Streefland).
2609 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2612 * Support for the IPv6 protocol has been added to the socket API, as per the
2613 latest draft standards.
2615 * Support for Linux 2.2 has been added.
2617 * Interface changes relative to the latest 2.0.x release:
2618 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2619 addseverity NEW: Unix98
2620 alphasort64 NEW: LFS
2621 argp_err_exit_status NEW: argp, GNU ext
2622 argp_error NEW: argp, GNU ext
2623 argp_failure NEW: argp, GNU ext
2624 argp_help NEW: argp, GNU ext
2625 argp_parse NEW: argp, GNU ext
2626 argp_program_bug_address NEW: argp, GNU ext
2627 argp_program_version NEW: argp, GNU ext
2628 argp_program_version_hook NEW: argp, GNU ext
2629 argp_state_help NEW: argp, GNU ext
2630 argp_usage NEW: argp, GNU ext
2631 authdes_create NEW: Secure RPC
2632 authdes_getucred NEW: Secure RPC
2633 authdes_pk_create NEW: Secure RPC
2634 backtrace NEW: GNU ext.
2635 backtrace_symbols NEW: GNU ext.
2636 backtrace_symbols_fd NEW: GNU ext.
2638 cacosf NEW: ISO C 9x
2639 cacosh NEW: ISO C 9x
2640 cacoshf NEW: ISO C 9x
2641 cacoshl NEW: ISO C 9x
2642 cacosl NEW: ISO C 9x
2649 casinf NEW: ISO C 9x
2650 casinh NEW: ISO C 9x
2651 casinhf NEW: ISO C 9x
2652 casinhl NEW: ISO C 9x
2653 casinl NEW: ISO C 9x
2655 catanf NEW: ISO C 9x
2656 catanh NEW: ISO C 9x
2657 catanhf NEW: ISO C 9x
2658 catanhl NEW: ISO C 9x
2659 catanl NEW: ISO C 9x
2660 cbc_crypt NEW: Secure RPC
2664 ccoshf NEW: ISO C 9x
2665 ccoshl NEW: ISO C 9x
2671 cimagf NEW: ISO C 9x
2672 cimagl NEW: ISO C 9x
2673 clearerr_locked REMOVED
2674 clntunix_create NEW: sunrpc ext
2676 clog10 NEW: ISO C 9x
2677 clog10f NEW: ISO C 9x
2678 clog10l NEW: ISO C 9x
2688 cprojf NEW: ISO C 9x
2689 cprojl NEW: ISO C 9x
2691 crealf NEW: ISO C 9x
2692 creall NEW: ISO C 9x
2697 csinhf NEW: ISO C 9x
2698 csinhl NEW: ISO C 9x
2701 csqrtf NEW: ISO C 9x
2702 csqrtl NEW: ISO C 9x
2706 ctanhf NEW: ISO C 9x
2707 ctanhl NEW: ISO C 9x
2709 des_setparity NEW: Secure RPC
2710 ecb_crypt NEW: Secure RPC
2711 endutxent NEW: Unix98
2713 exp10f NEW: ISO C 9x
2714 exp10l NEW: ISO C 9x
2718 fattach NEW: STREAMS
2719 fdetach NEW: STREAMS
2723 feclearexcept NEW: ISO C 9x
2724 fegetenv NEW: ISO C 9x
2725 fegetexceptflag NEW: ISO C 9x
2726 fegetround NEW: ISO C 9x
2727 feholdexcept NEW: ISO C 9x
2729 feraiseexcept NEW: ISO C 9x
2730 ferror_locked REMOVED
2731 fesetenv NEW: ISO C 9x
2732 fesetexceptflag NEW: ISO C 9x
2733 fesetround NEW: ISO C 9x
2734 fetestexcept NEW: ISO C 9x
2735 feupdateenv NEW: ISO C 9x
2736 fflush_locked REMOVED
2740 fgets_unlocked NEW: GNU ext.
2741 fileno_locked REMOVED
2753 fputc_locked REMOVED
2754 fputs_unlocked NEW: GNU ext.
2755 fread_unlocked NEW: GNU ext.
2760 fstatvfs NEW: Unix98
2764 ftruncate64 NEW: LFS
2766 fwrite_unlocked NEW: GNU ext.
2767 gai_strerror NEW: IPv6
2771 getchar_locked REMOVED
2773 getdate_err NEW: Unix98
2774 getdate_r NEW: GNU ext.
2776 getnameinfo NEW: IPv6
2777 getnetname NEW: Secure RPC
2778 getpmsg NEW: STREAMS
2779 getpt NEW: Unix98 PTY
2780 getrlimit64 NEW: LFS
2781 getutxent NEW: Unix98
2782 getutxid NEW: Unix98
2783 getutxline NEW: Unix98
2784 glob64 NEW: GNU ext.
2785 globfree64 NEW: GNU ext.
2786 gnu_get_libc_release NEW: GNU ext.
2787 gnu_get_libc_version NEW: GNU ext.
2788 grantpt NEW: Unix98 PTY
2789 host2netname NEW: Secure RPC
2791 iconv_close NEW: iconv
2792 iconv_open NEW: iconv
2793 if_freenameindex NEW: IPv6
2794 if_indextoname NEW: IPv6
2795 if_nameindex NEW: IPv6
2796 if_nametoindex NEW: IPv6
2797 in6addr_any NEW: IPv6
2798 in6addr_loopback NEW: IPv6
2799 inet6_isipv4mapped NEW: IPv6
2800 isastream NEW: STREAMS
2801 iswblank NEW: GNU ext.
2802 key_decryptsession NEW: Secure RPC
2803 key_decryptsession_pk NEW: Secure RPC
2804 key_encryptsession NEW: Secure RPC
2805 key_encryptsession_pk NEW: Secure RPC
2806 key_gendes NEW: Secure RPC
2807 key_get_conv NEW: Secure RPC
2808 key_secretkey_is_set NEW: Secure RPC
2809 key_setnet NEW: Secure RPC
2810 key_setsecret NEW: Secure RPC
2811 llrint NEW: ISO C 9x
2812 llrintf NEW: ISO C 9x
2813 llrintl NEW: ISO C 9x
2814 llround NEW: ISO C 9x
2815 llroundf NEW: ISO C 9x
2816 llroundl NEW: ISO C 9x
2821 lrintf NEW: ISO C 9x
2822 lrintl NEW: ISO C 9x
2823 lround NEW: ISO C 9x
2824 lroundf NEW: ISO C 9x
2825 lroundl NEW: ISO C 9x
2827 makecontext NEW: Unix98
2828 mempcpy NEW: GNU ext.
2831 modify_ldt NEW: kernel
2835 nearbyint NEW: ISO C 9x
2836 nearbyintf NEW: ISO C 9x
2837 nearbyintl NEW: ISO C 9x
2838 netname2host NEW: Secure RPC
2839 netname2user NEW: Secure RPC
2840 nexttoward NEW: ISO C 9x
2841 nexttowardf NEW: ISO C 9x
2842 nexttowardl NEW: ISO C 9x
2846 passwd2des NEW: Secure RPC
2848 pow10f NEW: GNU ext.
2849 pow10l NEW: GNU ext.
2852 printf_size NEW: GNU ext.
2853 printf_size_info NEW: GNU ext.
2854 profil_counter REMOVED
2855 pthread_mutexattr_getkind_np REPLACED
2856 pthread_mutexattr_setkind_np REPLACED
2857 ptsname NEW: Unix98 PTY
2858 ptsname_r NEW: Unix98 PTY
2860 putchar_locked REMOVED
2861 putgrent NEW: GNU ext.
2863 putpmsg NEW: STREAMS
2864 pututxline NEW: Unix98
2868 readdir64_r NEW: LFS
2869 remquo NEW: ISO C 9x
2870 remquof NEW: ISO C 9x
2871 remquol NEW: ISO C 9x
2873 roundf NEW: ISO C 9x
2874 roundl NEW: ISO C 9x
2876 scalbln NEW: ISO C 9x
2877 scalblnf NEW: ISO C 9x
2878 scalblnl NEW: ISO C 9x
2880 sendfile NEW: kernel
2881 setcontext NEW: Unix98
2882 setrlimit64 NEW: LFS
2883 setutxent NEW: Unix98
2885 sigignore NEW: Unix98
2886 sigqueue NEW: POSIX.1b
2887 sigrelse NEW: Unix98
2888 sigset NEW: POSIX.1b
2889 sigtimedwait NEW: POSIX.1b
2890 sigwaitinfo NEW: POSIX.1b
2891 sincos NEW: GNU ext.
2892 sincosf NEW: GNU ext.
2893 sincosl NEW: GNU ext.
2897 strcasestr NEW: GNU ext.
2898 strtoimax NEW: ISO C 9x
2899 strtoumax NEW: ISO C 9x
2900 strverscmp NEW: GNU ext.
2901 svcauthdes_stats NEW: Secure RPC
2902 svcunix_create NEW: sunrpc etx
2903 svcunixfd_create NEW: sunrpc ext
2904 swapcontext NEW: Unix98
2905 tcgetsid NEW: Unix98 PTY
2906 tdestroy NEW: GNU ext.
2907 tgamma NEW: ISO C 9x
2908 tgammaf NEW: ISO C 9x
2909 tgammal NEW: ISO C 9x
2913 truncf NEW: ISO C 9x
2914 truncl NEW: ISO C 9x
2916 unlockpt NEW: Unix98 PTY
2917 updwtmpx NEW: Unix98
2918 user2netname NEW: Secure RPC
2919 utmpxname NEW: Unix98
2920 versionsort NEW: GNU ext.
2921 versionsort64 NEW: GNU ext.
2923 wcscasecmp NEW: GNU ext.
2924 wcsncasecmp NEW: GNU ext.
2925 wcsnlen NEW: GNU ext.
2926 wcstoimax NEW: ISO C 9x
2927 wcstoll NEW: ISO C 9x
2928 wcstoull NEW: ISO C 9x
2929 wcstoumax NEW: ISO C 9x
2931 wordexp NEW: POSIX.2
2932 wordfree NEW: POSIX.2
2933 write_profiling REMOVED
2934 xdecrypt NEW: Secure RPC
2935 xdr_authdes_cred NEW: Secure RPC
2936 xdr_authdes_verf NEW: Secure RPC
2937 xdr_cryptkeyarg NEW: Secure RPC
2938 xdr_cryptkeyarg2 NEW: Secure RPC
2939 xdr_cryptkeyres NEW: Secure RPC
2940 xdr_getcredres NEW: Secure RPC
2941 xdr_key_netstarg NEW: Secure RPC
2942 xdr_key_netstres NEW: Secure RPC
2943 xdr_keybuf NEW: Secure RPC
2944 xdr_keystatus NEW: Secure RPC
2945 xdr_netnamestr NEW: Secure RPC
2946 xdr_sizeof NEW: Secure RPC
2947 xdr_unixcred NEW: sunrpc ext
2948 xencrypt NEW: Secure RPC
2949 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2960 * inet_ntoa is thread-safe
2962 * updwtmp is moved from libutil to libc
2964 * rewrite of cbrt function
2966 * update of timezone data
2980 * add atoll function
2982 * fix complex problems in Berkeley DB code
2984 * fix math functions
2988 * fixed lots of header problems (especially Linux/GNU specific)
2990 * dynamic loader preserves all registers
2992 * Roland McGrath provided support for handling of auxiliary objects in
2993 the ELF dynamic loader.
2995 * support for parallel builds is improved
2999 * GNU extensions are no longer declared by default. To enable them you
3000 must define the macro `_GNU_SOURCE' in your program or compile with
3003 * The library has changed from using GNU ld symbol aliases to using weak
3004 symbols where available. The ELF object file format supports weak
3005 symbols; GNU ld also supports weak symbols in the a.out format. (There
3006 is also now support for other GNU ld extensions in ELF. Use the
3007 `--with-elf' option to configure to indicate you have ELF, and
3008 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3009 of many files which contained only symbol aliases, reducing the size of
3010 the source and the compiled library; many other files were renamed to
3011 less cryptic names previously occupied by the symbol alias files.
3012 There is a new header file <elf.h> for programs which operate on
3013 files in the ELF format.
3015 * Converted to Autoconf version 2, so `configure' has more options.
3016 Run `configure --help' to see the details.
3018 * The library can now be configured to build profiling, highly-optimized
3019 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3020 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3021 `configure' enable building these extra libraries. The shared library is
3022 built by default when using both ELF and GNU ld. When shared libraries
3023 are enabled, the new library `-ldl' is available for arbitrary run-time
3024 loading of shared objects; its interface is defined in <dlfcn.h>. The
3025 new header file <link.h> gives access to the internals of the run-time
3026 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3027 application of same name on other systems and it provides information
3028 about dynamically linked binaries.
3030 * The C library now provides the run-time support code for profiling
3031 executables compiled with `-pg'. Programs can control the profiling code
3032 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3033 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3034 the support for this file format was contributed by David Mosberger-Tang.
3036 * The math code has been replaced with a math library based on fdlibm from
3037 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3038 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3039 functions. The math functions now reside in a separate library, so
3040 programs using them will need to use `-lm' their linking commands.
3042 * John C. Bowman contributed optimized ix87 assembler inline functions.
3044 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3045 mechanism similar to that found in Solaris 2. This is now used for the
3046 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3047 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3048 file controls what services are used for each individual database. This
3049 works by loading shared libraries with names specified in `nsswitch.conf',
3050 so service modules can be changed or added at any time without even
3051 relinking any program. Currently there are the file, db, and NIS based
3052 NSS services available.
3054 * The new functions `strtoq' and `strtouq' parse integer values from
3055 strings, like `strtol' and `strtoul', but they return `long long int' and
3056 `unsigned long long int' values, respectively (64-bit quantities).
3058 * The new functions `strtof' and `strtold' parse floating-point values from
3059 strings, like `strtod', but they return `float' and `long double' values,
3060 respectively (on some machines `double' and `long double' are the same).
3062 * Ulrich Drepper has contributed new implementations of the floating-point
3063 printing and reading code used in the `printf' family of functions and
3064 `strtod', `strtof', and `strtold'. These new functions are perfectly
3065 accurate, and much faster than the old ones.
3067 * The implementation of the POSIX locale model was completely rewritten by
3068 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3069 to compile the POSIX locale definition.
3071 * The former dummy implementations of the strcoll and strxfrm function are
3072 now replaced by fully functional code contributed by Ulrich Drepper. The
3073 collation information comes from the POSIX locale definitions.
3075 * The new header <langinfo.h> defines an interface for accessing
3076 various locale-dependent data (using the locale chosen with `setlocale').
3078 * Ulrich Drepper has contributed a new suite of functions for operation on
3079 wide-character and multibyte-character strings, in <wchar.h>;
3080 and classification and case conversion of wide characters, in <wctype.h>.
3081 These new functions are conforming to the ISO C, Amendement 1 specification.
3083 * There is now a second implementation of the standard I/O library available.
3084 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3085 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3087 * You can now use positional parameter specifications in format strings
3088 for the `printf' and `scanf' families of functions. For example,
3089 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3090 ``Number 6, Mr Jones''. This is mainly useful when providing different
3091 format strings for different languages, whose grammars may dictate
3092 different orderings of the values being printed. To support this
3093 feature, the interface for `register_printf_handler' has changed; see
3094 the header file <printf.h> for details.
3096 * The `printf' and `scanf' families of functions now understand a new
3097 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3098 to group numbers as indicated by the locale; for `scanf' and friends, this
3099 says to accept as valid only a number with all the proper grouping
3100 separators in the right places. In the default "C" locale, numbers are
3101 not grouped; but locales for specific countries will define the usual
3102 conventions (i.e. separate thousands with `,' in the US locale).
3104 * The pgrp functions have been regularized, slightly incompatibly but much
3105 less confusingly. The core functions are now `getpgid' and `setpgid',
3106 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3107 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3108 provided for compatibility. There is no longer an incompatible `getpgrp'
3109 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3111 * The new header file <fts.h> and suite of functions simplify programs that
3112 operate on directory trees. This code comes from 4.4 BSD.
3114 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3115 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3116 NSS scheme used in glibc.
3118 * There is a new malloc debugging hook `__memalign_hook'.
3120 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3121 `unsigned int' in <sys/types.h>. These are for compatibility only and
3122 their use is discouraged.
3124 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3125 done differently, so that it works even without GNU ld.
3127 * New function `euidaccess' checks allowed access to a file like `access',
3128 but using the effective IDs instead of the real IDs.
3130 * The time zone data files have been updated for the latest and greatest
3131 local time conventions of the countries of the world.
3133 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3136 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3137 for formatting floating-point numbers. They are provided only for
3138 compatibility; new programs should use `sprintf' instead. There are
3139 also equivalent function for the `long double' floating-point type and
3140 all functions also exist in a reentrant form.
3142 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3143 functions for maintaining the login-record files (primarily of use to
3144 system programs such as `login'), and convenient functions for
3145 allocating and initializing a pseudo-terminal (pty) device.
3147 * Ulrich Drepper has contributed new support for System V style
3148 shared memory and IPC on systems that support it.
3150 * Ulrich Drepper has contributed several miscellaneous new functions found
3151 in System V: The `hsearch' family of functions provide an effective
3152 implementation of hash tables; `a64l' and `l64a' provide a very simple
3153 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3156 * Ulrich Drepper has contributed new reentrant counterparts for the
3157 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3159 * Ulrich Drepper has contributed new, highly-optimized versions of several
3160 string functions for the i486/Pentium family of processors.
3162 * Ulrich Drepper has updated the Linux-specific code, based largely
3163 on work done in Hongjiu Lu's version of GNU libc for Linux.
3164 The GNU library now supports Linux versions 2.0.10 and later,
3165 using the ELF object file format (i[3456]86-*-linux).
3167 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3169 * David Mosberger-Tang and Richard Henderson have ported the C library
3170 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3171 dynamic linking support for ELF/Alpha.
3173 * Richard Henderson contributed several Alpha optimized assembler function
3174 for arithmetic and string handling.
3176 * Ulrich Drepper has contributed a new set of message catalog functions to
3177 support multiple languages using the <libintl.h> interface, for use with
3178 his new package GNU gettext. Translation volunteers have contributed
3179 catalogs of the library's messages in Spanish, German, and Korean.
3181 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3182 program and the `catgets' function for reading the catalog files it
3183 creates. (The <libintl.h> interface is preferred; we include the
3184 <nl_types.h> interface using `catgets' only for source compatibility with
3185 programs already written to use it.)
3187 * New header file <values.h> gives SVID-compatible names for <limits.h>
3190 * Various new macros, declarations, and small header files for compatibility
3193 * New function `group_member' is a convenient way to check if a process has
3194 a given effective group ID.
3196 * When using GCC 2.7 and later, the socket functions are now declared in a
3197 special way so that passing an argument of type `struct sockaddr_in *',
3198 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3199 `struct sockaddr *' type, does not generate a type-clash warning.
3201 * New function `error' declared in header file <error.h> is a convenient
3202 function for printing error messages and optionally exiting; this is the
3203 canonical function used in GNU programs. The new functions `err', `warn',
3204 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3205 doing the same thing.
3207 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3208 POSIX.2 `glob' function to do ~ and {...} expansion.
3210 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3211 `clearenv' which is used in POSIX.9 is also available.
3213 * New function `getsid' returns session ID number on systems that support it.
3215 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3216 files <db.h> and <mpool.h> provide a rich set of functions for several
3217 types of simple databases stored in memory and in files, and <ndbm.h> is
3218 an old `ndbm'-compatible interface using the `db' functions. Link with
3219 `-ldb' to get these functions.
3221 * New macro `strdupa' copies a string like `strdup', but uses local stack
3222 space from `alloca' instead of dynamic heap space from `malloc'.
3224 * New function `strnlen' is like `strlen' but searches only a given maximum
3225 number of characters for the null terminator. `stpncpy', `strndup' and
3226 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3229 * New function `statfs' in header <sys/statfs.h>.
3231 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3232 provide convenient functions for operating on blocks of null-terminated
3235 * A new suite of functions in <utmp.h> handle all the details of reading
3236 and writing the utmp file.
3238 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3241 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3242 completely POSIX compliant and also implemented the extended functionality
3243 to handle alternate digit representation and alternate era date formats.
3245 * Ulrich Drepper provided an implementation of the `strptime' function
3246 defined in XPG4.2 which transforms a string into a `struct tm' value.
3248 * Paul Eggert provided the tzselect shell script as part of the timezone
3249 code. The shell script makes it easy to select the correct timezone
3252 * The implementation of the malloc family of functions is completely replaced
3253 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3254 The implementation uses the mmap function (if available) and it is
3255 optimized for the use in multi threaded programs.
3257 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3258 functions. This new functionality is usable by specifying a special
3259 salt string and it is compatible with implementation on *BSD systems.
3261 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3262 `getsubopt' to handle second level command line options, `bsd_signal'
3263 to access BSD style `signal' functionality, the obsolete `regexp' style
3266 * the `lchown' function is available on system which support this
3269 * The implementation of the shadow password handling function was contributed
3272 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3274 * POSIX.1g support was added. The <sys/select.h> header is available,
3275 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3276 implementation of `getaddrinfo'.
3280 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3282 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3283 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3285 * The new function `daemon' from 4.4 BSD is useful for server programs that
3286 want to put themselves in the background.
3288 * Joel Sherrill has contributed support for several standalone boards that
3289 run without an operating system.
3291 * `printf', `scanf' and friends now accept a `q' type modifier for long
3292 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3294 * All of the code taken from BSD (notably most of the math and networking
3295 routines) has been updated from the BSD 4.4-Lite release.
3297 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3299 * The new functions `getdomainname' and `setdomainname' fetch or change the
3300 YP/NIS domain name. These are system calls which exist on systems which
3303 * The time zone data files have been updated for the latest international
3306 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3307 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3311 * The C library now includes support for Sun RPC, from Sun's free
3312 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3313 are included. (There is still no support for YP.)
3315 * Tom Quinn has contributed a port of the C library to SGI machines running
3316 Irix 4 (mips-sgi-irix4).
3318 * The new `lockf' function is a simplified interface to the locking
3319 facilities of `fcntl', included for compatibility.
3321 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3323 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3325 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3328 * The error message printed by `assert' for a failed assertion now includes
3329 the name of the program (if using GNU ld) and the name of the calling
3330 function (with versions of GCC that support this).
3332 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3334 * The library now includes the <sys/mman.h> header file and memory
3335 management functions `mmap', `munmap', `mprotect', `msync', and
3336 `madvise', on systems that support those facilities.
3338 * The interface for `mcheck' has changed slightly: the function called to
3339 abort the program when an allocation inconsistency is detected now takes
3340 an argument that indicates the type of failure. The new function
3341 `mprobe' lets you request a consistency check for a particular block at
3342 any time (checks are normally done only when you call `free' or `realloc'
3345 * It is now possible to easily cross-compile the C library, building on one
3346 system a library to run on another machine and/or operating system. All
3347 you need to do is set the variable `HOST_CC' in `configparms' to the
3348 native compiler for programs to run on the machine you are building on (a
3349 few generator programs are used on Unix systems); set `CC' to the
3352 * The new function `fexecve' (only implemented on the GNU system) executes
3353 a program file given a file descriptor already open on the file.
3357 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3358 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3360 * You can set the variable `libprefix' in `configparms' to specify a prefix
3361 to be prepended to installed library files; this makes it easy to install
3362 the GNU C library to be linked as `-lgnuc' or whatever.
3364 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3365 copies a limited number of characters from a string, and returns the
3366 address of the last character written.
3368 * You no longer need to check for whether the installed `stddef.h' is
3369 compatible with the GNU C library. configure now checks for you.
3371 * You can now define a per-stream `fileno' function to convert the
3372 stream's cookie into an integral file descriptor.
3374 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3375 allocates zero bytes of storage, and returns a unique pointer which
3376 you can pass to `realloc' or `free'. The behavior is undefined if
3377 you dereference this pointer.
3379 * The C library now runs on Sony NEWS m68k machines running either
3380 NewsOS 3 or NewsOS 4.
3382 * The new `syscall' function is a system-dependent primitive function
3383 for invoking system calls. It has the canonical behavior on Unix
3384 systems, including unreliable return values for some calls (such as
3385 `pipe', `fork' and `getppid').
3387 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3388 to `EAGAIN', which is the preferred name. On systems whose kernels
3389 use two distinct codes, the C library now translates EWOULDBLOCK to
3390 EAGAIN in every system call function.
3394 * The GNU C Library Reference Manual is now distributed with the library.
3395 `make dvi' will produce a DVI file of the printed manual.
3396 `make info' will produce Info files that you can read on line using C-h i
3397 in Emacs or the `info' program.
3398 Please send comments on the manual to bug-glibc-manual@gnu.org.
3400 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3402 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3404 * Jason Merrill has contributed a port to the Sequent Symmetry running
3405 Dynix version 3 (i386-sequent-dynix).
3407 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3408 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3410 * New function `memory_warnings' lets you arrange to get warnings when
3411 malloc is running out of memory to allocate, like Emacs gives you.
3413 * The C library now contains the relocating allocator used in Emacs 19 for
3414 its editing buffers. This allocator (ralloc) minimizes allocation
3415 overhead and fragmentation by moving allocated regions around whenever it
3416 needs to. You always refer to a ralloc'd region with a "handle" (a
3417 pointer to a pointer--an object of type `void **').
3419 * There is a new `printf' format: `%m' gives you the string corresponding
3420 to the error code in `errno'.
3422 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3423 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3424 pass, the `a' modifier says to fill in a `char **' you pass with a
3427 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3428 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3429 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3431 * `mkstemp' is a traditional Unix function to atomically create and open a
3432 uniquely-named temporary file.
3436 * The standard location for the file that says what the local timezone is
3437 has changed again. It is now `/usr/local/etc/localtime' (or more
3438 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3440 * The distribution no longer contains any files with names longer than 14
3443 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3444 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3446 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3450 * The configuration process has changed quite a bit. The `configure'
3451 script is now used just like the configuration scripts for other GNU
3452 packages. The `sysdeps' directory hierarchy is much rearranged.
3453 The file `INSTALL' explains the new scheme in detail.
3455 * The header files no longer need to be processed into ANSI C and
3456 traditional C versions. There is just one set of files to install, and
3457 it will work with ANSI or old C compilers (including `gcc -traditional').
3459 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3460 MIPS DECStation running Ultrix 4.
3462 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3463 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3464 made itself into a shared library.
3466 * Yet further improved support for the i386, running 4.3 BSD-like systems
3467 (such as Mach 3 with the Unix single-server), or System V.
3469 * New function `strncasecmp' to do case-insensitive string comparison
3470 with limited length.
3472 * New function `strsep' is a reentrant alternative to `strtok'.
3474 * New functions `scandir' and `alphasort' for searching directories.
3476 * New function `setenv' is a better interface to `putenv'.
3478 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3479 function for traversing a directory tree.
3481 * The GNU obstack package is now also part of the C library.
3482 The new function `open_obstack_stream' creates a stdio stream that
3483 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3484 formatted output directly to an obstack.
3486 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3487 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3489 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3491 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3492 things to your strings.
3494 * There are some new test programs: test-fseek, testmb, and testrand.
3496 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3497 These ports are not finished, but are a good starting place for really
3498 supporting those systems.
3500 * `/etc/localtime' is now the standard location for the file that says what
3501 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3502 This follows the general principle that `/etc' is the place for all local
3503 configuration files.
3505 * The C library header files now use `extern "C"' when used by the C++
3506 compiler, so the C library should now work with C++ code.
3508 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3509 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3512 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3513 function which uses the `merge sort' algorithm, and is said to be
3514 significantly faster than the old GNU `qsort' function. Merge sort is
3515 now the standard `qsort' function. The new algorithm can require a lot
3516 of temporary storage; so, the old sorting function is called when the
3517 required storage is not available.
3519 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3520 provides the Unix `crypt' function, plus some other entry points.
3522 * The code and header files taken from 4.4 BSD have been updated with the
3523 latest files released from Berkeley.
3525 ----------------------------------------------------------------------
3526 Copyright information:
3528 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3530 Permission is granted to anyone to make or distribute verbatim copies
3531 of this document as received, in any medium, provided that the
3532 copyright notice and this permission notice are preserved,
3533 thus giving the recipient permission to redistribute in turn.
3535 Permission is granted to distribute modified versions
3536 of this document, or of portions of it,
3537 under the above conditions, provided also that they
3538 carry prominent notices stating who last changed them.
3541 version-control: never