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 Security related changes:
146 On ARM EABI (32-bit), generating a backtrace for execution contexts which
147 have been created with makecontext could fail to terminate due to a
148 missing .cantunwind annotation. This has been observed to lead to a hang
149 (denial of service) in some Go applications compiled with gccgo. Reported
150 by Andreas Schwab. (CVE-2016-6323)
152 The following bugs are resolved with this release:
154 [The release manager will add the list generated by
155 scripts/list-fixed-bugs.py just before the release.]
159 * The minimum Linux kernel version that this version of the GNU C Library
160 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
161 kernel version 2.6.32 or later suffices (on architectures that already
162 required kernel versions more recent than 3.2, those requirements remain
163 unchanged). Linux 3.2 or later kernel headers are required on all
166 * The pap_AN locale has been deleted. This has been deprecated for a long
167 time. It has been replaced by pap_AW & pap_CW, both of which have long
168 been included in previous releases.
170 * The readdir_r and readdir64_r functions have been deprecated. It is
171 recommended to use readdir and readdir64 instead.
173 * The type “union wait” has been removed. It was deprecated in the early
174 1990s and never part of POSIX. Application code should use the int type
175 instead of “union wait”.
177 * A new NSS action is added to facilitate large distributed system
178 administration. The action, MERGE, allows remote user stores like LDAP
179 to be merged into local user stores like /etc/groups in order to provide
180 easy to use, updated, and managed sets of merged credentials. The new
181 action can be used by configuring it in /etc/nsswitch.conf:
182 group: files [SUCCESS=merge] nis
183 Implemented by Stephen Gallagher (Red Hat).
185 * The deprecated __malloc_initialize_hook variable has been removed from the
188 * The long unused localedef --old-style option has been removed. It hasn't
189 done anything in over 16 years. Scripts using this option can safely
192 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
193 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
194 return the next representable value in the direction of positive infinity
195 and the nextdown functions return the next representable value in the
196 direction of negative infinity. These are currently enabled as GNU
199 Security related changes:
201 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
202 could result in a stack overflow when getnetbyname was called with an
203 overly long name. (CVE-2016-3075)
205 * Previously, getaddrinfo copied large amounts of address data to the stack,
206 even after the fix for CVE-2013-4458 has been applied, potentially
207 resulting in a stack overflow. getaddrinfo now uses a heap allocation
208 instead. Reported by Michael Petlan. (CVE-2016-3706)
210 * The glob function suffered from a stack-based buffer overflow when it was
211 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
212 Reported by Alexander Cherepanov. (CVE-2016-1234)
214 * The Sun RPC UDP client could exhaust all available stack space when
215 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
216 alloca plugin for GCC. (CVE-2016-4429)
218 * The IPv6 name server management code in libresolv could result in a memory
219 leak for each thread which is created, performs a failing naming lookup,
220 and exits. Over time, this could result in a denial of service due to
221 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
223 The following bugs are resolved with this release:
225 [1170] localedata: ne_NP: update Nepali locale definition file
226 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
228 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
229 [6796] math: fdim() does not set errno on overflow
230 [10354] libc: posix_spawn should use vfork() in more cases than presently
231 [11213] localedata: localedata: add copyright disclaimer to locale files
232 [12143] localedata: chr_US: new Cherokee locale
233 [12450] localedata: sgs_LT: new locale
234 [12676] localedata: ln_CD: new locale
235 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
237 [13304] math: fma, fmaf, fmal produce wrong results
238 [14259] build: --localedir arg to configure is ignored
239 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
240 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
241 [14934] localedata: es_CL: wrong first weekday chilean locale
242 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
244 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
246 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
247 [15368] nptl: raise() is not async-signal-safe
248 [15479] math: ceil, floor, round and trunc raise inexact exception
249 [15578] localedata: kk_KZ: various updates
250 [16003] localedata: pap_AN: punt old locale
251 [16137] localedata: iw_IL: punt old locale
252 [16190] localedata: eo: new esperanto locale
253 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
254 [16742] malloc: race condition: pthread_atfork() called before first
255 malloc() results in unexpected locking behaviour/deadlocks
256 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
258 [16983] localedata: postal_fmt does not allow %l and %n modifiers
259 [17565] localedata: pt_PT: wrong (work-)week start
260 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
261 [17950] build: Build fails with -msse
262 [18205] localedata: be_BY*: wrong first_weekday and first_workday
263 [18433] libc: posix_spawn does not return correctly upon failure to
265 [18453] localedata: charmaps/IBM875: incorrect codes
266 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
268 [18896] localedata: he_IL: improvements for currency
269 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
271 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
273 [19056] libc: Deprecate readdir_r
274 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
276 [19198] localedata: nl_NL: small improvements for Dutch locales
277 [19257] network: Per-thread memory leak in __res_vinit with IPv6
278 nameservers (CVE-2016-5417)
279 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
281 [19400] locale: Language missing in "iso-639.def", trivial fix in
283 [19431] malloc: Deadlock between fflush, getdelim, and fork
284 [19505] libc: Incorrect file descriptor validity checks in
285 posix_spawn_file_actions_add{open,close,dup2}
286 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
288 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
289 `intl/{gettextP,loadinfo}.h'
290 [19534] libc: execle, execlp may use malloc
291 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
292 [19573] network: res_nclose and __res_maybe_init disagree about name
293 server initialization, breaking Hesiod
294 [19575] localedata: Status of GB18030 tables
295 [19581] localedata: sr_* date_fmt string contains additional newline
296 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
298 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
299 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
300 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
301 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
303 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
305 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
307 [19610] dynamic-link: ldconfig -X removes stale symbolic links
308 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
309 [19633] locale: strfmon_l applies global locale to number formatting
310 [19642] network: Memory leak in getnameinfo
311 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
312 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
314 [19654] math: [x86_64] Need testcase for BZ #19590 fix
315 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
317 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
318 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
320 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
322 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
324 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
325 pointers and lengths in error-case.
326 [19727] locale: Converting from/to UTF-xx with iconv() does not always
327 report errors on UTF-16 surrogates values.
328 [19755] nscd: nscd assertion failure in gc
329 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
330 [19759] libc: mempcpy shouldn't be inlined
331 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
332 [19765] libc: s390 needs an optimized mempcpy
333 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
334 NAME_MAX limit assumption (CVE-2016-1234)
335 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
336 [19787] network: Missing and incorrect truncation checks in getnameinfo
337 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
339 [19791] network: Assertion failure in res_query.c with un-connectable name
341 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
342 [19822] math: libm.so install clobbers old version
343 [19825] network: resolv: send_vc can return uninitialized data in second
344 response to getaddrinfo
345 [19830] network: nss_dns: should check RDATA length against buffer length
346 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
347 confronted with A/AAAA records of invalid size
348 [19837] nss: nss_db: No retries for some long lines with a larger buffer
349 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
350 [19853] stdio: Printing IBM long double in decimal with high precision is
352 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
353 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
354 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
356 [19865] network: Assertion failure or memory leak in
357 _nss_dns_getcanonname_r
358 [19868] network: nss_dns: netent code does not skip over non-PTR records
359 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
361 [19881] string: Improve x86-64 memset
362 [19907] string: Incorrect memcpy tests
363 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
365 [19925] libc: termios.h XCASE namespace
366 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
367 [19929] libc: limits.h NL_NMAX namespace
368 [19931] stdio: Memory leak in vfprintf
369 [19957] libc: clone(CLONE_VM) access invalid parent memory
370 [19963] localedata: en_IL: New locale
371 [19989] stdio: stdio.h cuserid namespace
372 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
374 [19996] locale: langinfo.h nl_langinfo_l namespace
375 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
376 [20010] network: getaddrinfo: Stack overflow in hostent translation
378 [20012] stdio: libio: fmemopen append mode failure
379 [20014] stdio: stdio.h namespace for pre-threads POSIX
380 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
381 [20023] libc: fcntl.h timespec namespace
382 [20024] math: [x86_64] vectorized sincos trashes the stack
383 [20031] network: nss_hesiod: Heap overflow in get_txt_records
384 [20041] time: sys/time.h timespec namespace
385 [20043] libc: unistd.h missing cuserid for UNIX98 and before
386 [20044] libc: unistd.h missing pthread_atfork for UNIX98
387 [20051] libc: ttyslot in wrong header under wrong conditions
388 [20054] libc: gethostname not declared for XPG4
389 [20055] libc: termios.h missing tcgetsid for XPG4
390 [20072] dynamic-link: x86 init_cpu_features is called twice in static
392 [20073] libc: sys/stat.h fchmod namespace
393 [20074] libc: stdlib.h rand_r namespace
394 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
395 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
397 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
398 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
400 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
401 [20119] libc: Wrong mask for processors level type from CPUID
402 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
403 AS not supporting AVX512
404 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
406 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
407 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
409 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
410 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
411 [20178] libc: posix_spawn{p} should not call exit
412 [20191] stdio: libio: vtables hardening
413 [20195] string: FMA4 detection requires CPUID execution with register
415 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
416 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
418 [20212] math: acos (sNaN) returns sNaN
419 [20213] math: asin (sNaN) returns sNaN
420 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
421 [20218] math: [i386] asinhl (sNaN) returns sNaN
422 [20219] math: [i386] atanhl (sNaN) returns sNaN
423 [20222] stdio: fopencookie: Mangle function pointers
424 [20224] math: [i386] cbrtl (sNaN) returns sNaN
425 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
426 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
428 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
429 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
430 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
431 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
432 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
433 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
434 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
435 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
436 [20240] math: modf (sNaN) returns sNaN
437 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
438 [20250] math: frexp (sNaN) returns sNaN
439 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
440 [20255] math: [i386] fdim, fdimf return with excess range and precision /
442 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
443 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
444 subscript is below array bounds [-Werror=array-bounds]
445 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
446 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
448 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
449 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
451 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
453 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
454 [20316] localedata: id_ID: Februari instead of Pebruari
455 [20327] string: POWER8 strcasecmp returns incorrect result
456 [20347] math: Failure: Test: j0_downward (0xap+0)
457 [20348] libc: FAIL: misc/tst-preadvwritev64
458 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
459 [20350] libc: There is no test for p{read,write}64
460 [20357] math: Incorrect cos result for 1.5174239687223976
461 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
465 * Unicode 8.0.0 Support: Character encoding, character type info, and
466 transliteration tables are all updated to Unicode 8.0.0, using new
467 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
468 These updates cause user visible changes, such as the fixes for bugs
469 89, 16061, and 18568.
471 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
472 kernel-internal CPU set size. This means that requests that change the
473 CPU affinity which failed before (for example, an all-ones CPU mask) will
474 now succeed. Applications that need to determine the effective CPU
475 affinities need to call sched_getaffinity or pthread_getaffinity_np after
476 setting it because the kernel can adjust it (and the previous size check
477 would not detect this in the majority of cases).
479 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
480 the following new symbols are used: fts64_children, fts64_close,
481 fts64_open, fts64_read and fts64_set.
483 * getaddrinfo now detects certain invalid responses on an internal netlink
484 socket. If such responses are received, an affected process will
485 terminate with an error message of "Unexpected error <number> on netlink
486 descriptor <number>" or "Unexpected netlink response of size <number> on
487 descriptor <number>". The most likely cause for these errors is a
488 multi-threaded application which erroneously closes and reuses the netlink
489 file descriptor while it is used by getaddrinfo.
491 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
492 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
493 unnecessary serialization of memory allocation requests across threads.
494 The defect is now corrected. Users should see a substantial increase in
495 the concurent throughput of allocation requests for applications which
496 trigger this bug. Affected applications typically create and
497 destroy threads frequently. (Bug 19048 was reported and analyzed by
500 * There is now a --disable-timezone-tools configure option for disabling the
501 building and installing of the timezone related utilities (zic, zdump, and
502 tzselect). This is useful for people who build the timezone data and code
503 independent of the GNU C Library.
505 * The obsolete header <regexp.h> has been removed. Programs that require
506 this header must be updated to use <regex.h> instead.
508 * The obsolete functions bdflush, create_module, get_kernel_syms,
509 query_module and uselib are no longer available to newly linked binaries;
510 the header <sys/kdaemon.h> has been removed. These functions and header
511 were specific to systems using the Linux kernel and could not usefully be
512 used with the GNU C Library on systems with version 2.6 or later of the
515 * Optimized string, wcsmbs and memory functions for IBM z13.
516 Implemented by Stefan Liebler.
518 * Newly linked programs that define a variable called signgam will no longer
519 have it set by the lgamma, lgammaf and lgammal functions. Programs that
520 require signgam to be set by those functions must ensure that they use the
521 variable provided by the GNU C Library and declared in <math.h>, without
522 defining their own copy.
524 * The minimum GCC version that can be used to build this version of the GNU
525 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
526 still be used to compile programs using the GNU C Library.
528 Security related changes:
530 * An out-of-bounds value in a broken-out struct tm argument to strftime no
531 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
533 * The LD_POINTER_GUARD environment variable can no longer be used to disable
534 the pointer guard feature. It is always enabled. Previously,
535 LD_POINTER_GUARD could be used to disable security hardening in binaries
536 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
539 * An integer overflow in hcreate and hcreate_r could lead to an
540 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
542 * The catopen function no longer has unbounded stack usage. Reported by
545 * The nan, nanf and nanl functions no longer have unbounded stack usage
546 depending on the length of the string passed as an argument to the
547 functions. Reported by Joseph Myers. (CVE-2014-9761)
549 * A stack-based buffer overflow was found in libresolv when invoked from
550 libnss_dns, allowing specially crafted DNS responses to seize control
551 of execution flow in the DNS client. The buffer overflow occurs in
552 the functions send_dg (send datagram) and send_vc (send TCP) for the
553 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
554 family. The use of AF_UNSPEC triggers the low-level resolver code to
555 send out two parallel queries for A and AAAA. A mismanagement of the
556 buffers used for those queries could result in the response of a query
557 writing beyond the alloca allocated buffer created by
558 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
559 the overflow. Thanks to the Google Security Team and Red Hat for
560 reporting the security impact of this issue, and Robert Holiday of
561 Ciena for reporting the related bug 18665. (CVE-2015-7547)
563 The following bugs are resolved with this release:
565 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
566 [887] math: Math library function "logb" and "nextafter" inconsistent
567 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
568 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
569 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
570 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
572 [4404] localedata: German translation of "Alarm clock" is misleading
573 [6799] math: nextafter() and nexttoward() doen't set errno on
574 overflow/underflow errors
575 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
577 [10432] nis: _nss_nis_setnetgrent assertion failure
578 [11460] libc: fts has no LFS support
579 [12926] network: getaddrinfo()/make_request() may spin forever
580 [13065] nptl: Race condition in pthread barriers
581 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
582 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
584 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
585 [14912] libc: Rename non-installed bits/*.h headers
586 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
587 [15367] math: Let gcc use __builtin_isinf
588 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
589 [15421] math: lgamma wrongly sets signgam for ISO C
590 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
591 FE_INVALID with argument out of range
592 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
594 [15786] dynamic-link: ifunc resolver functions can smash function
596 [15918] math: Unnecessary check for equality in hypotf()
597 [16061] localedata: Review / update transliteration data
598 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
600 [16141] time: strptime %z offset restriction
601 [16171] math: drem should be alias of remainder
602 [16296] math: fegetround is pure?
603 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
604 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
606 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
607 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
609 [16422] math: [powerpc] math-float, math-double failing llrint tests with
610 "Exception "Inexact" set" on ppc32
611 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
612 [16517] math: Missing underflow exception from tanf/tan/tanl
613 [16519] math: Missing underflow exception from sinhf
614 [16520] math: Missing underflow exception from tanhf
615 [16521] math: Missing underflow exception from exp2
616 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
618 [16734] stdio: fopen calls mmap to allocate its buffer
619 [16961] math: nan function incorrect handling of bad sequences
620 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
621 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
622 [16985] locale: localedef: confusing error message when opening output
624 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
625 [17197] locale: Redundant shift character in iconv conversion output at
627 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
628 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
629 [17250] dynamic-link: static linking breaks nss loading
630 (getaddrinfo/getpwnam/etc...)
631 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
633 [17441] math: isnan() should use __builtin_isnan() in GCC
634 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
635 (related to lock elision)
636 [17787] manual: Exponent on page 324 of the PDF ends prematurely
637 [17886] time: strptime should be able to parse "Z" as a timezone with %z
638 [17887] time: strptime should be able to parse "+01:00" style timezones
639 [17905] libc: catopen() Multiple unbounded stack allocations
641 [18084] libc: backtrace (..., 0) dumps core on x86
642 [18086] libc: nice() sets errno to 0 on success
643 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
644 is too large (CVE-2015-8778)
645 [18251] dynamic-link: SONAME missing when audit modules provides path
646 [18265] libc: add attributes for wchar string and memory functions
647 [18370] math: csqrt missing underflows
648 [18421] libc: [hppa] read-only segment has dynamic relocations
649 [18472] libc: Obsolete syscall wrappers should be compat symbols
650 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
651 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
652 [18525] localedata: Remove locale timezone information
653 [18560] libc: [powerpc] spurious bits/ipc.h definitions
654 [18568] localedata: Update locale data to Unicode 8.0
655 [18589] locale: sort-test.sh fails at random
656 [18595] math: ctan, ctanh missing underflows
657 [18604] libc: assert macro-expands its argument
658 [18610] math: S390: fetestexcept() reports any exception if DXC-code
659 contains a vector instruction exception.
660 [18611] math: j1, jn missing errno setting on underflow
661 [18618] localedata: sync Chechen locale definitions with other *_RU
663 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
664 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
665 [18665] network: In send_dg, the recvfrom function is NOT always using the
666 buffer size of a newly created buffer (CVE-2015-7547)
667 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
669 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
671 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
672 [18699] math: tilegx cproj() for various complex infinities does not yield
674 [18724] libc: Harden put*ent functions against data injection
675 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
676 [18755] build: build errors with -DNDEBUG
677 [18757] stdio: fmemopen fails to set errno on failure
678 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
680 [18781] libc: openat64 lacks O_LARGEFILE
681 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
682 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
683 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
684 [18790] math: [ldbl-128ibm] tanhl inaccurate
685 [18795] libc: stpncpy fortification misses buffer lengths that are
687 [18796] build: build fails for --disable-mathvec
688 [18803] math: hypot missing underflows
689 [18820] stdio: fmemopen may leak memory on failure
690 [18823] math: csqrt spurious underflows
691 [18824] math: fma spurious underflows
692 [18825] math: pow missing underflows
693 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
694 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
695 [18870] build: sem_open.c fails to compile with missing symbol
697 [18872] stdio: Fix memory leak in printf_positional
698 [18873] libc: posix_fallocate overflow check ineffective
699 [18875] math: Excess precision leads incorrect libm
700 [18877] libc: arm: mmap offset regression
701 [18887] libc: memory corruption when using getmntent on blank lines
702 [18918] localedata: hu_HU: change time to HH:MM:SS format
703 [18921] libc: Regression: extraneous stat() and fstat() performed by
705 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
706 binaries (CVE-2015-8777)
707 [18951] math: tgamma missing underflows
708 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
710 [18953] localedata: lt_LT: change currency symbol to the euro
711 [18956] math: powf inaccuracy
712 [18961] math: [i386] exp missing underflows
713 [18966] math: [i386] exp10 missing underflows
714 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
715 [18969] build: multiple string test failures due to missing locale
717 [18970] libc: Reference of pthread_setcancelstate in libc.a
718 [18977] math: float / long double Bessel functions not in XSI POSIX
719 [18980] math: i386 libm functions return with excess range and precision
720 [18981] math: i386 scalb*, ldexp return with excess range and precision
721 [18982] stdio: va_list and vprintf
722 [18985] time: Passing out of range data to strftime() causes a segfault
724 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
725 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
726 [19012] locale: iconv_open leaks memory on error path
727 [19016] math: clog, clog10 inaccuracy
728 [19018] nptl: Mangle function pointers in tls_dtor_list
729 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
730 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
731 [19048] malloc: malloc: arena free list can become cyclic, increasing
733 [19049] math: [powerpc] erfc incorrect zero sign
734 [19050] math: [powerpc] log* incorrect zero sign
735 [19058] math: [x86_64] Link fail with -fopenmp and -flto
736 [19059] math: nexttoward overflow incorrect in non-default rounding modes
737 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
738 [19074] network: Data race in _res_hconf_reorder_addrs
739 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
740 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
741 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
743 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
745 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
746 [19086] manual: posix_fallocate64 documented argument order is wrong.
747 [19088] math: lround, llround missing exceptions close to overflow
749 [19094] math: lrint, llrint missing exceptions close to overflow threshold
750 [19095] math: dbl-64 lrint incorrect for 64-bit long
751 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
752 [19124] dynamic-link: ld.so failed to build with older assmebler
753 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
754 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
755 [19134] math: [powerpc32] lround, lroundf spurious exceptions
756 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
757 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
758 pthread_setaffinity_np
759 [19156] math: [ldbl-128] j0l spurious underflows
760 [19164] nptl: tst-getcpu fails with many possible CPUs
761 [19168] math: math/test-ildoubl and math/test-ldouble failure
762 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
763 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
765 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
766 (FE_NOMASK_ENV) do not clear SSE exceptions
767 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
768 _int_new_arena/reused_arena
769 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
770 [19201] math: dbl-64 remainder incorrect sign of zero result
771 [19205] math: bits/math-finite.h conditions do not match math.h and
773 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
774 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
775 for C99-based standards
776 [19212] libc: features.h not -Wundef clean
777 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
779 [19214] libc: Family and model identification for AMD CPU's are incorrect.
780 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
781 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
783 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
785 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
787 [19242] libc: strtol incorrect in Turkish locales
788 [19243] malloc: reused_arena can pick an arena on the free list, leading
789 to an assertion failure and reference count corruption
790 [19253] time: tzset() ineffective when temporary TZ did not include DST
792 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
793 [19270] math: [hppa] Shared libm missing __isnanl
794 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
795 MAP_HUGETLB and MAP_STACK defines
796 [19313] nptl: Wrong __cpu_mask for x32
797 [19347] libc: grantpt: try to force a specific gid even without pt_chown
798 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
799 [19350] math: [ldbl-128ibm] sinhl spurious overflows
800 [19351] math: [ldbl-128ibm] logl inaccurate near 1
801 [19363] time: x32: times() return value wrongly truncates/sign extends
803 [19367] dynamic-link: Improve branch prediction on Silvermont
804 [19369] network: Default domain name not reset by res_ninit when "search"
805 / "domain" entry is removed from resolv.conf
806 [19375] math: powerpc: incorrect results for POWER7 logb with negative
808 [19385] localedata: bg_BG: time separator should be colon, not comma
809 [19408] libc: linux personality syscall wrapper may erroneously return an
810 error on 32-bit architectures
811 [19415] libc: dladdr returns wrong names on hppa
812 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
813 IBM905, IBM907, and IBM909
814 [19439] math: Unix98 isinf and isnan functions conflict with C++11
815 [19443] build: build failures with -DDEBUG
816 [19451] build: Make check fails on test-double-vlen2
817 [19462] libc: Glibc failed to build with -Os
818 [19465] math: Wrong code with -Os
819 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
821 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
823 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
824 [19486] math: S390: Math tests fail with "Exception Inexact set".
825 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
826 [19550] libc: [mips] mmap negative offset handling inconsistent with other
828 [19590] math: Fail to build shared objects that use libmvec.so functions.
832 * The following bugs are resolved with this release:
834 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
835 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
836 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
837 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
838 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
839 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
840 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
841 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
842 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
843 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
844 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
845 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
846 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
847 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
848 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
849 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
850 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
851 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
852 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
853 18657, 18676, 18694, 18696, 18887.
855 * Cache information can be queried via sysconf() function on s390 e.g. with
856 _SC_LEVEL1_ICACHE_SIZE as argument.
858 * A buffer overflow in gethostbyname_r and related functions performing DNS
859 requests has been fixed. If the NSS functions were called with a
860 misaligned buffer, the buffer length change due to pointer alignment was
861 not taken into account. This could result in application crashes or,
862 potentially arbitrary code execution, using crafted, but syntactically
863 valid DNS responses. (CVE-2015-1781)
865 * The time zone file parser has been made more robust against crafted time
866 zone files, avoiding heap buffer overflows related to the processing of
867 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
868 large time zone data files. Overly long time zone specifiers in the TZ
869 variable no longer result in stack overflows and crashes.
871 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
872 for LD and GD on x86 and x86-64, has been implemented. You will need
873 binutils-2.24 or later to enable this optimization.
875 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
876 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
877 Hat). These updates cause user visible changes, such as the fix for bug
880 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
881 and getXXbyYY NSS calls for the same database, causing a denial-of-service
882 condition in some applications.
884 * Added vector math library named libmvec with the following vectorized x86_64
885 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
887 The library can be disabled with --disable-mathvec. Use of the functions is
888 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
889 Shared library libmvec.so is linked in as needed when using -lm (no need to
890 specify -lmvec explicitly for not static builds).
891 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
893 * A new fmemopen implementation has been added with the goal of POSIX
894 compliance. The new implementation fixes the following long-standing
895 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
896 old implementation is still present for use be by existing binaries.
898 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
899 releases. It has been fixed to match 2.19 and older, but binaries built
900 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
902 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
903 Contributed by Roland McGrath (Google).
905 * The header <regexp.h> is deprecated, and will be removed in a future
906 release. Use of this header will trigger a deprecation warning.
907 Application developers should update their code to use <regex.h> instead.
909 This header was formerly part of SUS, but was deprecated in 1994 and
910 removed from the standard in 2001. Also, the glibc implementation
911 leaks memory. See BZ#18681 for more details.
915 * The following bugs are resolved with this release:
917 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
918 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
919 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
920 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
921 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
922 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
923 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
924 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
925 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
926 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
929 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
930 too little memory for the to-be-scanned arguments and overflow the
931 allocated buffer. The implementation now correctly computes the required
932 buffer size when using malloc, and switches to malloc from alloca as
935 * A new semaphore algorithm has been implemented in generic C code for all
936 machines. Previous custom assembly implementations of semaphore were
937 difficult to reason about or ensure that they were safe. The new version
938 of semaphore supports machines with 64-bit or 32-bit atomic operations.
939 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
940 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
942 * Port to Altera Nios II has been contributed by Mentor Graphics.
944 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
945 implementations for powerpc64/powerpc64le.
946 Implemented by Adhemerval Zanella (IBM).
948 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
949 and powerpc64le. This may improve lock scaling of existing programs on
950 HTM capable systems. The lock elision code is only enabled with
951 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
952 powerpc will issue a transaction abort on every syscall to avoid side
953 effects being visible outside transactions.
955 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
956 AArch64. Contributed by ARM Ltd.
958 * i386 memcpy functions optimized with SSE2 unaligned load/store.
960 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
961 under certain input conditions resulting in the execution of a shell for
962 command substitution when the application did not request it. The
963 implementation now checks WRDE_NOCMD immediately before executing the
964 shell and returns the error WRDE_CMDSUB as expected.
966 * CVE-2012-3406 printf-style functions could run into a stack overflow when
967 processing format strings with a large number of format specifiers.
969 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
970 infinite loop if the DNS response contained a PTR record of an unexpected
973 * The minimum GCC version that can be used to build this version of the GNU
974 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
975 still be used to compile programs using the GNU C Library.
977 * The GNU C Library is now built with -Werror by default. This can be
978 disabled by configuring with --disable-werror.
980 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
982 * The obsolete sigvec function has been removed. This was the original
983 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
984 programs have been using instead for about 25 years. Of course, ABI
985 compatibility for old binaries using sigvec remains intact.
987 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
988 with newer versions of bison.
990 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
991 The original MIPS o32 hard-float ABI requires an FPU where double-precision
992 registers overlay two consecutive single-precision registers. MIPS32R2
993 introduced a new FPU mode (FR=1) where double-precision registers extend the
994 corresponding single-precision registers which is incompatible with the
995 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
996 require the use of FR=1 making a transition necessary. New o32 ABI
997 extensions enable users to migrate over time from the original o32 ABI
998 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
999 tracks the ABI of any loaded object and verifies that new objects are
1000 compatible. Mode transitions will also be requested as required and
1001 unsupportable objects will be rejected. The ABI checks include both soft and
1002 hard float ABIs for o32, n32 and n64.
1004 GCC 5 with GNU binutils 2.25 onwards:
1005 It is strongly recommended that all o32 system libraries are built using the
1006 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1007 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1008 compiler using --with-fp-32=xx to set this by default.
1012 * The following bugs are resolved with this release:
1014 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1015 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1016 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1017 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1018 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1019 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1020 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1021 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1022 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1023 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1024 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1025 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1026 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1027 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1028 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1030 * Reverted change of ABI data structures for s390 and s390x:
1031 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1032 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1033 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1034 are preserved pointing straight to the same implementation as the old ones.
1035 Given that, new callers will simply provide a too-big buffer to these
1036 functions. Any applications/libraries out there that embed jmp_buf or
1037 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1038 against 2.19 headers will have to rebuilt again. This is necessary in any
1039 case to revert the breakage in their ABI caused by the glibc change.
1041 * Support for file description locks is added to systems running the
1042 Linux kernel. The standard file locking interfaces are extended to
1043 operate on file descriptions, not file descriptors, via the use of
1044 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1045 are associated with an open file instead of a process.
1047 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1049 * The minimum Linux kernel version that this version of the GNU C Library
1050 can be used with is 2.6.32.
1052 * Running the testsuite no longer terminates as soon as a test fails.
1053 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1054 with PASS or FAIL lines for individual tests. A summary of the results is
1055 printed, including a list of failing lists, and "make check" exits with
1056 error status if there were any unexpected failures. "make check
1057 stop-on-test-failure=y" may be used to keep the old behavior.
1059 * The am33 port, which had not worked for several years, has been removed
1062 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1063 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1064 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1065 conflicted with POSIX (support for which was removed in 2.19), the
1066 interfaces those macros enabled remain available when compiling with
1067 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1068 test macros defined.
1070 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1072 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1073 This may improve lock scaling of existing programs on TX capable systems.
1074 The lock elision code is only built with --enable-lock-elision=yes and
1075 then requires a GCC version supporting the TX builtins. With lock elision
1076 default mutexes are elided via __builtin_tbegin, if the cpu supports
1077 transactions. By default lock elision is not enabled and the elision code
1080 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1081 copy the path argument. This allowed programs to cause posix_spawn to
1082 deference a dangling pointer, or use an unexpected pathname argument if
1083 the string was modified after the posix_spawn_file_actions_addopen
1086 * All supported architectures now use the main glibc sysdeps directory
1087 instead of some being in a separate "ports" directory (which was
1088 distributed separately before glibc 2.17).
1090 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1091 On configurations that support it (all Linux configurations), it's now
1092 used regardless of the --enable-add-ons switch to configure. It is no
1093 longer possible to build such configurations without pthreads support.
1095 * Locale names, including those obtained from environment variables (LANG
1096 and the LC_* variables), are more tightly checked for proper syntax.
1097 setlocale will now fail (with EINVAL) for locale names that are overly
1098 long, contain slashes without starting with a slash, or contain ".." path
1099 components. (CVE-2014-0475) Previously, some valid locale names were
1100 silently replaced with the "C" locale when running in AT_SECURE mode
1101 (e.g., in a SUID program). This is no longer necessary because of the
1104 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1105 with application code using Intel MPX instructions. (With all previous
1106 versions, the MPX register state could be clobbered when making calls
1107 into or out of a shared library.) Note that while the new dynamic
1108 linker is compatible with all known x86 hardware whether or not it
1109 supports Intel MPX, some x86 instruction-set emulators might fail to
1110 handle the new instruction encodings. This is known to affect Valgrind
1111 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1112 release), and might affect other tools that do instruction emulation.
1114 * Support for loadable gconv transliteration modules has been removed.
1115 The support for transliteration modules has been non-functional for
1116 over a decade, and the removal is prompted by security defects. The
1117 normal gconv conversion modules are still supported. Transliteration
1118 with //TRANSLIT is still possible, and the //IGNORE specifier
1119 continues to be supported. (CVE-2014-5119)
1121 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1122 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1123 resulting a denial-of-service security vulnerability in applications which
1124 use functions related to iconv. (CVE-2014-6040)
1128 * The following bugs are resolved with this release:
1130 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1131 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1132 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1133 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1134 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1135 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1136 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1137 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1138 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1139 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1140 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1141 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1142 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1143 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1144 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1145 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1146 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1147 16431, 16453, 16474, 16506, 16510, 16529
1149 * Slovenian translations for glibc messages have been contributed by the
1150 Translation Project's Slovenian team of translators.
1152 * The public headers no longer use __unused nor __block. This change is to
1153 support compiling programs that are derived from BSD sources and use
1154 __unused internally, and to support compiling with Clang's -fblock
1155 extension which uses __block.
1157 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1158 large collation sequences to optimize multiple passes. This cache
1159 computation may overflow for large collation sequences and may cause a
1160 stack or buffer overflow. This is now fixed to use a slower algorithm
1161 which does not use a cache if there is an integer overflow.
1163 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1164 rules for large collation sequences to optimize multiple passes and falls
1165 back to alloca if malloc fails, resulting in a possible stack overflow.
1166 The implementation now falls back to an uncached collation sequence lookup
1169 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1170 initialized for static applications resulting in the security feature
1171 being disabled. The pointer guard is now correctly initialized to a
1172 random value for static applications. Existing static applications need
1173 to be recompiled to take advantage of the fix (bug 15754).
1175 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1176 to the d_name member of struct dirent, or omit the terminating NUL
1177 character. (Bugzilla #14699).
1179 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1180 aligned_alloc functions could allocate too few bytes or corrupt the
1181 heap when passed very large allocation size values (Bugzilla #15855,
1184 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1185 for AF_INET6 has been fixed (Bugzilla #16072).
1187 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1188 pap_CW, quz_PE, the_NP.
1190 * Substantially revised locales: gd_GB, ht_HT
1192 * The LC_ADDRESS field was updated to support country_car for almost all
1195 * ISO 1427 definitions were updated.
1197 * ISO 3166 definitions were updated.
1199 * The localedef utility now supports --big-endian and --little-endian
1200 command-line options to generate locales for a different system from that
1201 for which the C library was built.
1203 * Binary locale files now only depend on the endianness of the system for
1204 which they are generated and not on other properties of that system. As a
1205 consequence, binary files generated with new localedef may be incompatible
1206 with old versions of the GNU C Library, and binary files generated with
1207 old localedef may be incompatible with this version of the GNU C Library,
1208 in the following circumstances:
1210 + Locale files may be incompatible on m68k systems.
1212 + Locale archive files (but not separate files for individual locales) may
1213 be incompatible on systems where plain "char" is signed.
1215 * The configure option --disable-versioning has been removed. Builds with
1216 --disable-versioning had not worked for several years.
1218 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1219 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1221 * SystemTap probes for malloc have been introduced.
1223 * SystemTap probes for slow multiple precision fallback paths of
1224 transcendental functions have been introduced.
1226 * Support for powerpc64le has been added.
1228 * The soft-float powerpc port now supports e500 processors.
1230 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1232 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1233 set of header declarations that are enabled by default, even when other
1234 feature test macros or compiler options such as -std=c99 would otherwise
1235 disable some of those declarations.
1237 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1238 conflict with POSIX. The libbsd-compat library (which was a dummy library
1239 that did nothing) has also been removed.
1241 * Preliminary documentation about Multi-Thread, Async-Signal and
1242 Async-Cancel Safety has been added.
1244 * Change of ABI data structures for s390 and s390x:
1245 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1246 allow for future hardware extensions. All library functions that accept or
1247 return these structures were versioned in order to provide backwards
1248 compatibility with existing code. However, not all cases can be handled
1249 with interface versioning. If an application embeds either structure into
1250 another structure and that structure is passed to another compilation unit
1251 which expects the newer larger-sized object then any access to the new
1252 fields will result in undefined behaviour. Similarly any access to fields
1253 that were moved by the enlarging of the structures will cause undefined
1254 behaviour. To fix the undefined behaviour all objects that are part of the
1255 final in-memory image must agree on the size of structures, and this may
1256 require recompilation.
1260 * The following bugs are resolved with this release:
1262 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1263 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1264 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1265 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1266 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1267 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1268 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1269 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1270 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1271 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1272 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1273 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1274 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1277 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1278 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1279 Distributions can re-enable building and using pt_chown via the new configure
1280 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1281 considerable security risks and should only be used if the distribution
1282 understands and accepts the risks.
1284 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1287 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1288 fixed (Bugzilla #15330).
1290 * Add support for calling C++11 thread_local object destructors on thread
1291 and program exit. This needs compiler support for offloading C++11
1292 destructor calls to glibc.
1294 * Improved worst case performance of libm functions with double inputs and
1297 * Support for priority inherited mutexes in pthread condition variables on
1298 non-x86 architectures.
1300 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1302 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1304 * Optimized string functions for ARM. Implemented by Will Newton and
1307 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1309 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1312 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1313 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1315 * Added a benchmark framework to track performance of functions in glibc.
1317 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1318 It is based on draft TS 18661 and currently enabled as a GNU extension.
1320 * On Linux, the clock function now uses the clock_gettime system call
1321 for improved precision, rather than old times system call.
1323 * Added support for version-3 tz format files. This is needed when using
1324 the tz database release 2013e or later, and affects a few unusual cases --
1325 currently only TZ='America/Godthab' for time stamps after 2037.
1327 * Added new API functions pthread_getattr_default_np and
1328 pthread_setattr_default_np to get and set the default pthread
1329 attributes of a process.
1331 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1332 This may improve lock scaling of existing programs on TSX capable systems.
1333 When the --enable-lock-elision=yes parameter is specified at configure
1334 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1337 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1338 availability). Implemented by Ryan S. Arnold.
1340 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1344 * The following bugs are resolved with this release:
1346 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1347 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1348 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1349 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1350 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1351 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1352 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1353 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1354 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1355 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1356 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1357 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1358 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1359 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1360 14879, 14889, 14893, 14898, 14914.
1362 * Optimization of memcpy for MIPS.
1364 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1365 EMFILE has been fixed (Bugzilla #14889).
1367 * The add-on ports collection is now distributed in the "ports" subdirectory
1368 of the main GNU C Library distribution, rather than separately.
1370 * Port to ARM AArch64 contributed by Linaro.
1372 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1373 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1375 Implemented by Andreas Krebbel.
1377 * The new function secure_getenv allows secure access to the environment,
1378 returning NULL if running in a SUID/SGID process. This function replaces
1379 the internal function __secure_getenv.
1381 * SystemTap static probes have been added into the dynamic linker.
1382 Implemented by Gary Benson.
1384 * Optimizations of string functions strstr, strcasestr and memmem.
1385 Implemented by Maxim Kuvyrkov.
1387 * The minimum Linux kernel version that this version of the GNU C Library
1388 can be used with is 2.6.16.
1390 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1391 powerpc POWER7. Implemented by Will Schmidt.
1393 * New configure option --disable-nscd builds the C library such that it
1394 never attempts to contact the Name Service Caching Daemon (nscd).
1395 New configure option --disable-build-nscd avoids building nscd itself;
1396 this is the default if --disable-nscd is used.
1398 * Improved support for cross-compilation, including cross-testing and
1399 bootstrap builds without a previously built glibc.
1401 * Several testsuite tests are now able to test multiple IFUNC variants of an
1402 interface, rather than just testing the one that would be chooen by
1405 * New configure options --with-bugurl and --with-pkgversion, for
1406 distributors to use to embed their bug-reporting and package version
1407 information in --help and --version output.
1409 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1410 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1411 allows creation of chroots without the procfs mounted on /proc.
1413 * The `crypt' function now fails if passed salt bytes that violate the
1414 specification for those values. On Linux, the `crypt' function will
1415 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1416 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1417 when the mode is enabled.
1419 * The `clock_*' suite of functions (declared in <time.h>) is now available
1420 directly in the main C library. Previously it was necessary to link with
1421 -lrt to use these functions. This change has the effect that a
1422 single-threaded program that uses a function such as `clock_gettime' (and
1423 is not linked with -lrt) will no longer implicitly load the pthreads
1424 library at runtime and so will not suffer the overheads associated with
1425 multi-thread support in other code such as the C++ runtime library.
1427 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1432 * The following bugs are resolved with this release:
1434 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1435 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1436 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1437 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1438 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1439 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1440 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1441 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1442 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1443 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1444 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1445 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1446 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1447 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1448 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1449 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1450 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1451 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1452 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1453 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1454 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1455 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1458 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1459 configuring glibc with:
1460 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1461 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1462 Implemented by H.J. Lu.
1466 + define static_assert
1468 + do not declare gets
1470 + declare at_quick_exit and quick_exit also for ISO C11
1472 + aligned_alloc. NB: The code is deliberately allows the size parameter
1473 to not be a multiple of the alignment. This is a moronic requirement
1474 in the standard but it is only a requirement on the caller, not the
1477 + timespec_get added
1479 + uchar.h support added
1481 + CMPLX, CMPLXF, CMPLXL added
1483 Implemented by Ulrich Drepper.
1485 * Support for the IA-64 has been moved to ports.
1487 * Remove support for anything but ELF binary format
1489 * Checking versions of poll, ppoll added.
1490 Implemented by Ulrich Drepper.
1492 * More generic and 64-bit performance optimizations to math functions.
1493 Implemented by Ulrich Drepper.
1495 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1496 headers and functions available at compile time as they were before
1497 version 2.14. This option will be removed at some time in the future
1498 after the TI-RPC library becomes fully sufficient for the needs of
1499 existing applications.
1501 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1502 Note that glibc is not expected to work with any Linux kernel version
1505 * New header <sys/auxv.h> and function getauxval allowing easy access to
1506 the AT_* key-value pairs passed from the Linux kernel. The header also
1507 defines the HWCAP_* bits associated with the AT_HWCAP key.
1509 * New locales: mag_IN
1511 * New configure option --enable-systemtap builds SystemTap static probes
1512 into libc for setjmp and longjmp and into libpthread for various operations.
1513 So far the setjmp/longjmp probes and some of the libpthread probes are
1514 provided only for i*86 and x86_64.
1515 Implemented by Roland McGrath and Rayson Ho.
1517 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1519 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1522 * More optimized functions for SPARC. Implemented by David S. Miller.
1524 * Improved support for cross-compilation, especially bootstrap builds
1525 without a previously built glibc.
1527 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1528 Chris Metcalf from Tilera.
1530 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1531 now supported for ARM processors.
1533 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1534 as the name of the dynamic linker, to distinguish it from the
1535 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1537 * Support for CRIS has been removed from ports.
1539 * A new class of installed header has been documented for low-level
1540 platform-specific functionality. PowerPC added the first instance with a
1541 function to provide time base register access. Contributed by Tulio
1542 Magno Quites Machado Filho.
1544 * ix86 configurations now install header files that are consistent with
1545 what x86-64 configurations install. These same header files can be used
1546 for -m32, -m64, or -mx32 builds.
1547 Contributed by H.J. Lu.
1549 * Math library bug fixes. A thorough audit of all open math library bugs was
1550 conducted by Joseph Myers. Significant progress was made on many math
1551 library bugs resulting in more accurate exceptions and function results.
1552 Many thanks to all those that contributed including Andreas Jaeger for his
1553 patch review and work on the x87 trigonometric instruction issues.
1555 * Timezone data is no longer installed. Timezone-related binaries and scripts
1556 will continue to be installed. Users should obtain their timezone data from
1557 their distribution provider or from the tzdata package at
1558 <ftp://munnari.oz.au/pub/>.
1562 * The following bugs are resolved with this release:
1564 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1565 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1566 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1567 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1568 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1569 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1570 13484, 13506, 13515, 13523, 13524, 13538, 13540
1572 * New program pldd to list loaded object of a process
1573 Implemented by Ulrich Drepper.
1575 * Add nss_db support back to glibc. No more dependency on Berkeley db
1576 and support for initgroups lookups.
1577 Implemented by Ulrich Drepper.
1579 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1580 Contributed by HJ Lu.
1582 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1583 Contributed by HJ Lu.
1585 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1586 on x86-32 and x86-64.
1587 Contributed by Liubov Dmitrieva.
1589 * Optimized strchr and strrchr for SSE on x86-32.
1590 Contributed by Liubov Dmitrieva.
1592 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1593 for x86-64 and x86-32.
1594 Contributed by Liubov Dmitrieva.
1596 * New interfaces: scandirat, scandirat64
1597 Implemented by Ulrich Drepper.
1599 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1600 Implemented by Ulrich Drepper.
1602 * nscd now also caches the netgroup database.
1603 Implemented by Ulrich Drepper.
1605 * Integrate libm with gcc's -ffinite-math-only option.
1606 Implemented by Ulrich Drepper.
1608 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1609 to math functions. Implemented by Ulrich Drepper.
1611 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1612 Implemented by Ulrich Drepper.
1614 * New Linux interfaces: process_vm_readv, process_vm_writev
1616 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1617 Implemented by Ulrich Drepper.
1619 * Optimized nearbyint and strcasecmp for PPC.
1620 Implemented by Adhemerval Zanella.
1622 * New locales: bho_IN, unm_US, es_CU, ta_LK
1626 * The following bugs are resolved with this release:
1628 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1629 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1630 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1631 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1632 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1633 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1634 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1635 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1636 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1637 12795, 12811, 12813, 12814, 12841
1639 * The RPC implementation in libc is obsoleted. Old programs keep working
1640 but new programs cannot be linked with the routines in libc anymore.
1641 Programs in need of RPC functionality must be linked against TI-RPC.
1642 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1644 Visible changes of this change include (obviously) the inability to link
1645 programs using RPC functions without referencing the TI-RPC library and the
1646 removal of the RPC headers from the glibc headers.
1647 Implemented by Ulrich Drepper.
1649 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1650 syncfs, setns, sendmmsg
1652 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1653 yue_HK, lij_IT, mhr_RU
1655 * New encodings: CP770, CP771, CP772, CP773, CP774
1657 * New program sotruss to trace calls through PLTs
1658 Implemented by Ulrich Drepper.
1660 * The malloc hook implementation is marked deprecated and will be removed
1661 from the default implementation in the next version. The design never
1662 worked ever since the introduction of threads. Even programs which do
1663 not create threads themselves can use multiple threads created internally.
1667 * The following bugs are resolved with this release:
1669 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1670 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1671 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1672 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1675 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1677 * POWER7 optimizations: memset, memcmp, strncmp
1679 * New optimized string functions for x86-64: strnlen (SSE2),
1680 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1681 Implemented by Ulrich Drepper.
1685 * The following bugs are resolved with this release:
1687 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1688 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1689 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1690 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1691 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1692 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1693 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1694 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1696 * New interfaces: pthread_getname_np, pthread_setname_np
1698 * New Linux interface: recvmmsg
1700 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1702 * The dynamic linker now recognizes supported ABI versions from the
1703 EI_ABIVERSION field in the ELF header.
1704 Implemented by Ulrich Drepper.
1706 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1707 will not be used to fill in password fields in the passwd.byname replies.
1708 Instead it is used to synthesize the shadow.byname table, should it be
1709 missing. This is a useful mode in some installations involving Solaris.
1710 Implemented by Ulrich Drepper.
1712 * New locales: kok_IN, sq_MK, cv_RU
1716 * The following bugs are resolved with this release:
1718 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1719 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1720 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1721 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1722 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1723 10789, 10817, 10823, 10840, 10847
1725 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1727 Implemented by Ulrich Drepper.
1729 * Checking version of longjmp added that fails if an uninitialized stack
1730 frame would be created. Implemented by Ulrich Drepper.
1732 * STT_GNU_IFUNC is now supported in static executables.
1733 Implemented by H.J. Lu.
1735 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1737 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1738 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1739 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1740 Contributed by H.J. Lu.
1742 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1743 Implemented by Ulrich Drepper.
1745 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1747 Contributed by H.J. Lu.
1749 * Support for fma instruction in AVX on x86-64.
1750 Implemented by H.J. Lu and Ulrich Drepper.
1752 * AVX support in x86-64 auditing support in ld.so.
1753 Implemented by H.J. Lu.
1755 * STB_GNU_UNIQUE support added.
1756 Implemented by Ulrich Drepper.
1758 * Implement second fallback mode for optimized DNS lookup for even more
1759 broken environments. If two requests from the same port are not
1760 handled correctly close the socket and open a new one before sending
1761 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1762 can be used to select this mode right away, instead of rediscovering the
1763 necessity is every process again.
1764 Implemented by Ulrich Drepper.
1766 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1767 Implemented by Adam Tkac.
1769 * Optimized iconv conversions for S390x.
1770 Implemented by Andreas Krebbel.
1772 * Using condvars with PI mutexes is now more efficient due to kernel
1773 support for requeueing to PI futexes. NPTL support added for x86-64.
1774 Implemented by Ulrich Drepper.
1776 * New locales: ps_AF, my_MM
1780 * The following bugs are resolved with this release:
1782 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1783 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1784 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1785 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1786 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1788 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1789 Implemented by Ulrich Drepper.
1791 * New Linux interfaces: accept4, fallocate, fallocate64.
1792 Implemented by Ulrich Drepper.
1794 * Correct declarations of string function when used in C++ code. This
1795 could lead to compile errors for invalid C++ code.
1797 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1800 * New POSIX 2008 interface: psiginfo
1801 Implemented by Ulrich Drepper.
1803 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1804 Implemented by Ulrich Drepper.
1806 * Support for selecting between multiple function definitions at runtime
1807 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1809 * The libcrypt library can now use the hash function implementations in
1810 NSS. Implemented by Ulrich Drepper.
1812 * The malloc implementation can be compiled to be less memory efficient
1813 but higher performing in multi-threaded programs.
1814 Implemented by Ulrich Drepper.
1816 * New locales: nan_TW@latin, ks_IN
1818 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1819 Implemented by Ulrich Drepper.
1821 * Extended printf hook support. It is possible to use user-defined types
1822 and extend existing format specifiers.
1823 Implemented by Ulrich Drepper.
1825 * Handling for group shadow files has been added.
1826 Implemented by Ulrich Drepper.
1828 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1829 servers (the case, e.g., for some people using the built-in DNS
1830 server in ADSL modems/routers). There is a once-per-process timeout
1831 in case of a broken server. To avoid it, users can run nscd or put
1832 'options single-request' in /etc/resolv.conf.
1833 Implemented by Ulrich Drepper.
1837 * The following bugs are resolved with this release:
1839 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1840 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1841 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1842 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1843 6968, 6974, 6980, 6995, 7008, 7009, 7029
1845 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1846 up at the same time. Implemented by Ulrich Drepper.
1848 * TLS descriptors for LD and GD on x86 and x86-64.
1849 Implemented by Alexandre Oliva.
1851 * getaddrinfo now handles DCCP and UDPlite.
1852 Implemented by Ulrich Drepper.
1854 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1855 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1856 Implemented by Ulrich Drepper.
1858 * New implementation of memmem, strstr, and strcasestr which is O(n).
1859 Implemented by Eric Blake.
1861 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1863 * Implement "e" option for popen to open file descriptor with the
1864 close-on-exec flag set. Implemented by Ulrich Drepper.
1866 * Implement "b" mode for fmemopen. In this mode writes writes don't
1867 implicitly add a NUL byte and seeks from the end of the buffer really
1868 use the buffer end, not the string length as the basis.
1869 Implemented by Ulrich Drepper.
1871 * Many functions, exported and internal, now atomically set the close-on-exec
1872 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1874 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1876 Implemented by Pravin Satpute.
1878 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1882 * The following bugs are resolved with this release:
1884 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1885 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1886 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1887 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1888 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1889 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1890 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1892 * New locales: bo_CN, bo_IN, shs_CA.
1894 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1896 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1897 Implemented by Pravin Satpute.
1899 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1901 * nscd caches now all timeouts for DNS entries
1902 Implemented by Ulrich Drepper.
1904 * nscd is more efficient and wakes up less often.
1905 Implemented by Ulrich Drepper.
1907 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1908 vdprintf, and obstack_vprintf.
1909 Implemented by Jakub Jelinek.
1911 * Faster memset for x86-64.
1912 Implemented by Harsha Jagasia and H.J. Lu.
1914 * Faster memcpy on x86.
1915 Implemented by Ulrich Drepper.
1917 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1918 Implemented by Ulrich Drepper.
1920 * Faster sqrt and sqrtf implemention for some PPC variants.
1921 Implemented by Steven Munroe.
1925 * The following bugs are resolved with this release:
1927 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1928 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1929 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1930 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1931 5071, 5103, 5104, 5112, 5113, 5184, 5186
1933 * More checking functions: fread, fread_unlocked, open*, mq_open.
1934 Implemented by Jakub Jelinek and Ulrich Drepper.
1936 * Extend fortification to C++. Implemented by Jakub Jelinek.
1938 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1939 by not recognizing 'a' as a modifier when those specs are requested.
1940 Implemented by Jakub Jelinek.
1942 * PPC optimizations to math and string functions.
1943 Implemented by Steven Munroe.
1945 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1946 options to be passed. Implemented by Ulrich Drepper.
1948 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1950 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1951 Implemented by Ulrich Drepper.
1953 * Handle private futexes in the NPTL implementation.
1954 Implemented by Jakub Jelinek and Ulrich Drepper.
1956 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1957 Implemented by Roland McGrath and Ulrich Drepper.
1959 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1961 * SHA-256 and SHA-512 based password encryption.
1962 Implemented by Ulrich Drepper.
1964 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1965 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1968 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1969 Implemented by Ulrich Drepper.
1973 * The following bugs are resolved with this release:
1975 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
1976 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
1977 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
1978 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
1979 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
1980 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
1981 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
1984 * New Linux interfaces: epoll_pwait, sched_getcpu.
1986 * New generic interfaces: strerror_l.
1988 * nscd can now cache the services database. Implemented by Ulrich Drepper.
1992 * The following bugs are resolved with this release:
1994 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
1995 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
1996 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
1997 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
1998 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
1999 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2000 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2001 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2002 3225, 3251, 3252, 3253, 3273
2004 * For Linux, the sorting of addresses returned by getaddrinfo now also
2005 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2006 Implemented by Ulrich Drepper.
2008 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2009 Implemented by Ulrich Drepper.
2011 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2013 * New iconv module for MIK. Contributed by Alexander Shopov.
2015 * For sites with broken group and/or passwd database, the auto-propagate
2016 option of nscd can prevent creating ID lookup entries from the results
2017 of a name lookup and vice versa. This usually is no problem but some
2018 site might have problems with the default behavior.
2019 Implemented by Ulrich Drepper.
2021 * Iterating over entire database in NIS can be slow. With the
2022 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2023 to trade time for memory. The entire database will be read at once.
2024 Implemented by Ulrich Drepper.
2026 * The interfaces introduced in RFC 3542 have been implemented by
2029 * Support for the new ELF hash table format was added by Ulrich Drepper.
2031 * Support for priority inheritance mutexes added by Jakub Jelinek and
2034 * Support for priority protected mutexes added by Jakub Jelinek.
2036 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2037 or_IN, csb_PL, fy_NL, sr_ME.
2041 * More overflow detection functions.
2043 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2044 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2046 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2047 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2048 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2049 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2050 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2051 by Masahide Washizawa.
2053 * It is now possible to install audit libraries for the dynamic linker using
2054 LD_AUDIT. Implemented by Ulrich Drepper.
2056 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2057 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2058 For a libc and libpthread that works well on Linux 2.4 kernels, we
2059 recommend using the stable 2.3 branch.
2061 * The new function open_wmemstream acts like open_memstream,
2062 but uses a wchar_t wide character buffer.
2064 * The new function ppoll is an improved version of poll, similar to pselect.
2066 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2067 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2068 renameat, unlinkat, linkat, symlinkat, readlinkat.
2070 * New Linux kernel system calls: unshare,
2071 inotify_init, inotify_add_watch, inotify_rm_watch.
2073 * The euidaccess function is now also known by the alias eaccess,
2074 for compatibility with some other systems.
2076 * Timezone data updated to 2006b version.
2080 * The following bugs are resolved with this release:
2082 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2083 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2084 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2085 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2086 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2087 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2089 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2091 * As of this release, GCC 4 can be used to compile the C Library.
2093 * Timezone data updated to 2005m version.
2097 * The following bugs are resolved with this release:
2099 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2100 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2101 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2102 777, 787, 821, 822, 823, 825
2104 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2108 * Support for RFC 3678. Real implementations exist only for Linux so far.
2109 Implemented by Ulrich Drepper.
2111 * nscd can now cache entries persistently. Expiring entries are reloaded.
2112 For speedups the cache can be shared in memory with client processes.
2113 Implemented by Ulrich Drepper.
2115 * nscd can now perform SELinux checks.
2116 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2118 * getaddrinfo queries are now cached. Canonical name lookup is performed
2120 Implemented by Ulrich Drepper.
2122 * The nothrow function attribute is used when headers are used by gcc when
2123 compiling C code. This can avoid significant amounts of exception
2126 * The malloc functions perform more error checking and are stricter when
2127 it comes to reacting on errors. The default action is to terminate
2128 the process after showing an error message. Implemented by Ulrich Drepper.
2130 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2131 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2133 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2134 namespaces using the new function dlmopen(). This feature is of course,
2135 like most other dynamic loading functionality, not available in statically
2136 linked applications. Implemented by Ulrich Drepper.
2138 * Low-overhead boundary checking variants of string and some stdio functions
2139 were added. These are to be used in conjunction with a gcc patch by
2140 Jakub Jelinek which adds calls to these functions if possible.
2141 Implemented by Jakub Jelinek and Ulrich Drepper.
2143 * Old code for several operating systems and machine architectures that
2144 have not been in working condition in a long time have been removed from
2145 the main source tree maintained by the GNU C Library's maintainers.
2146 These files are now reside in the separate `ports' source module
2147 that is usable as an add-on when building the library.
2151 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2152 interrogate the dynamic linker, compatible with the Solaris interface.
2154 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2155 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2157 * getifaddrs now uses the netlink interface on Linux to get its information.
2158 Implemented by Thorsten Kukuk.
2160 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2161 Implemented by Ulrich Drepper.
2163 * support for non-executable stacks on x86 has been added. Changes mostly
2166 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2169 * getaddrinfo now performs destination address selection according to
2174 * Thread-safe interfaces for many functions that access locale data
2175 were added in version 2.3, but these features were omitted from NEWS.
2176 Many functions have variants with an `_l' suffix that take a `locale_t'
2177 object as a parameter rather than consulting the current locale.
2178 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2179 create and maintain `locale_t' objects. Additionally, the new function
2180 `uselocale' sets "the current locale" (as used by functions not so
2181 parameterized) set for an individual thread. These features were added
2182 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2184 * The functions getresuid, getresgid, setresuid, and setresgid, which
2185 have long been available on Linux, are now declared in <unistd.h>
2186 and are now also available on the Hurd.
2188 * ELF thread-local storage support (TLS) now works on x86-64.
2190 * The new dynamic string token $LIB is expanded in shared library names.
2191 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2193 * Aldy Hernandez contributed complete software floating point support for
2194 PowerPC machines with no FPU.
2196 * fexecve is implemented on Linux.
2198 * The `btowc' function should work at least twice as fast due to
2199 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2201 * With approriate thread add-ons cancelable functions are now implemented
2202 in libc.so as well. No need to call the function in libpthread. This
2203 change allowed to finally disable the incorrect and expensive handling
2204 of weak definition in ld.so.
2206 * Yet more PLT entries in libc.so have been removed. We finally arrived
2207 at the bare minimum. Startup times improved appropriately.
2209 * Support for the new Linux/x86 system call interface was added. The
2210 AT_SYSINFO auxiliary vector entry is recognized and handled.
2214 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2217 * iconv (the program and the interface) now accepts empty names (excluding
2218 options like //TRANSLIT) to mean "use charset of current locale".
2220 * localedef can now transliterate characters in strings which are not in
2221 the provided charmap. The information from the input locale is used.
2223 * Prelinking support was added for ELF targets. This requires additional
2224 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2226 * Read-only stdio streams now use mmap to speed up operation by eliminating
2227 copying and buffer underflows. To use add 'm' to the mode string of
2228 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2230 * The malloc functions were completely rewritten by Wolfram Gloger based
2231 on Doug Lea's malloc-2.7.0.c.
2233 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2234 implementation of regex.
2236 * Bruno Haible upgraded the iconv and locale implementation to support
2239 * Contents of the LC_* and LANG environment variables in the CEN style are
2240 not recognized anymore. It never was used. Change by Ulrich Drepper.
2242 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2243 thread-local storage (TLS) ABI on some platforms.
2244 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2246 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2247 EUC-JISX0213, and TSCII.
2249 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2250 BSD-compatible interface for getting all network interface addresses.
2251 Implementation for IPv4 by Roland McGrath.
2253 * Loading of locale data is faster due to the introduction of a locale
2254 archive. Implemented by Roland McGrath and Ulrich Drepper.
2256 * Startup times are significantly reduced by not using exported functions
2257 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2260 * Steven Munroe contributed a port to PowerPC64/Linux.
2264 * The Hurd now uses the GNU libio implementation of stdio rather than the
2265 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2267 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2268 with the same behavior as the Linux system call of the same name.
2272 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2273 128-bit long double format.
2275 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2276 IBM1161, and IBM1162 charsets.
2278 * Andreas Jaeger contributed a port to x86-64/Linux.
2280 * Peter Bruin contributed a port to PowerPC/Hurd.
2282 * libc functions using I/O streams now can handle wide-oriented streams
2285 * optimizations in the dynamic linker. Binaries created by recent binutils
2286 versions start up quicker due to reduced time spent on relocations.
2288 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2292 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2293 asinh, atanh, j0 for the 128-bit long double format.
2295 * Bruno Haible updated all the code handling Unicode in some form to
2296 support Unicode 3.1.
2298 * Speed of regex for single-byte locales is back to previous levels.
2299 Patch by Isamu Hasegawa.
2301 * Alpha, SPARC, and IA-64 now also using floating stacks.
2303 * Startup time of internationalized applications greatly improved through
2304 iconv cache. Use iconvconfig to generate the cache file.
2305 Contributed by Ulrich Drepper.
2307 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2308 severe performance problems. Patch by David Mosberger.
2310 * The Hurd port got a lot more functionality like AIO, various stdio
2311 extensions, etc. Mainly done by Roland McGrath.
2313 * mtrace can now lookup symbols in shared libraries.
2317 * Intel's IA-64 math library is largely integrated. It provides fast and
2318 accurate implementatations for most basic and standard math functions
2319 in float, double, and long double format.
2321 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2322 and asin for the 96-bit long double format and asin, log, tan for the
2323 128-bit long double format.
2325 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2326 The basic double functions exist today. Contributed by Abraham Ziv
2327 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2328 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2330 * An asynchronous name lookup library was added. The interface is designed
2331 after POSIX AIO. The proposal was circulated beforehand to get comments.
2332 No negative ones came in. Implemented by Ulrich Drepper.
2334 * Port to S390/64bit contributed by Martin Schwidefsky
2335 <schwidefsky@de.ibm.com>.
2337 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2338 of functions for Linux/IA-64.
2340 * The RPC code is now thread safe. Threads can now use the same service
2341 of different services at the same time. Patch by Eric Norum
2342 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2344 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2345 family of functions for Linux/S390.
2347 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2348 of functions for Linux/x86.
2350 * Port to Linux/CRIS contributed by Axis Communications.
2354 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2355 we can now check for namespace violations and missing declarations. The
2356 result is that almost all headers are now Unix-compliant (as defined in
2357 the upcoming XPG6). The negative side is that some programs might need
2358 corrections, too, if they depend on the incorrect form of the headers in
2359 previous versions which defined too many symbols and included too many
2362 * regex now handles multibyte character sets correctly.
2363 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2365 * iconv (the program) does now conform to the upcoming XPG6 and handles
2366 charmaps. Instead of the charset names the path of charmaps can be
2367 provided and the conversion happens based on this data.
2368 Contributed by Ulrich Drepper.
2370 * The locale program now provides detailed information about the installed
2375 only lists the names of the supported locales
2379 provides details such as country, language, and codeset name.
2380 Contributed by Ulrich Drepper.
2384 * The gencat program now parses the input file according to the charset
2385 selected by the LC_CTYPE category. This is important for stateful
2386 character sets. To make generating catalogs easier there is a way
2387 to overwrite the charset selected by the locale: before the first
2388 message or $ quote line the catalog can contain a line like
2390 $ codeset=ISO-8859-2
2392 to select the charset (ISO-8859-2 in this case).
2394 Implemented by Shinya Hanataka and Ulrich Drepper.
2396 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2397 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2398 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2400 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2401 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2403 * The LANGUAGE environment variable is now ignored unless the locale is
2404 changed from the default "C" locale.
2406 * The usual bug fixes.
2410 * Greg McGary added runtime support for bounds checking using gcc's
2411 new -fbounded-pointers option. ix86 target is complete. PowerPC
2414 * Thorsten Kukuk added secure mode support to nscd.
2416 * The Berkeley db libraries have been removed.
2418 Related, the nss_db module is now in a separate package since it
2419 obviously requires a database library being available.
2421 * Wide character I/O streams implemented by Ulrich Drepper.
2423 * Functions from the extended socket API added by Ulrich Drepper.
2425 * Functions feenableexcept and fedisableexcept to control the
2426 behaviour of individual exceptions have been added by Andreas Jaeger.
2428 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2430 * The resolver code has been updated from bind 8.2.3-T5B which supports
2431 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2434 This change could in some situations effect backward compatibility. Since
2435 now `_res' is a thread-local instead of a global variable, modifying it
2436 in one thread does not have any effect in other threads.
2438 The resolver library was also extended to allow IPv6 as the transport
2439 protocol for the requests. This work was done by Stig Venaas.
2441 * Compatibility code for K&R C compilers has been removed from the
2442 header files. A ISO C compiler is needed to use the library
2443 (conforming to either C89 or C99 standard).
2445 * Complete rewrite of the localedef program to support multibyte character
2446 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2447 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2448 Implemented by Ulrich Drepper.
2450 Bruno Haible significantly improved the generation and use of the data
2451 structures for the wide character tables.
2453 * Plural handling in gettext implemented by Ulrich Drepper.
2455 * The utmp daemon has been removed.
2457 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2459 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2462 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2464 * POSIX barriers implemented by Kaz Kylheku.
2466 * POSIX spawn function family implemented by Ulrich Drepper.
2468 * POSIX spinlocks are now available.
2470 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2472 * the configure option --enable-kernel=X.Y.Z allows to strip out
2473 compatibility for kernel versions before X.Y.Z. This is currently only
2474 implemented for Linux.
2476 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2477 field sin6_scope_id. This means that all programs using IPv6 should be
2478 recompiled. Don't expect binary compatibility with previous glibc
2481 * various conversion modules for IBM character sets contributed by
2484 * IA-64 port by Jes Sorensen and HJ Lu.
2498 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2500 * New xdr functions are added; some rpc functions are now 64bit clean.
2502 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2504 * Fixed known incompatibilities with glibc 2.0.
2506 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2508 * Optimized a number of functions (especially the ELF dynamic loader).
2510 * Update timezone data files.
2512 * lots of charmaps corrections
2514 * some new locale definitions and charmaps
2519 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2520 conform to POSIX member type requirements. Symbol versions have been
2521 adjusted as needed within the library, and for direct use by applications,
2522 but there is potential for problems if third-party libraries use
2523 struct timeval as part of their interface. This does not present
2524 a problem for X and other "essential" system libraries.
2526 * An additional locale model to support C++ Standard Library locale
2527 model and probably more was implemented by Ulrich Drepper.
2529 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2532 * Miles Bader provided the `argp' function family to support hierachical
2533 command line argument parsing, layered on top of getopt.
2535 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2537 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2540 * scanf recognizes the %a and %A format for scanning floating point numbers.
2542 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2543 information and interfaces for the available integer types.
2545 * about 130 new math functions were added to implement the ISO C9x math
2548 * the new header <complex.h> contains definitions of the complex math
2549 functions from ISO C 9X.
2551 * the new header <tgmath.h> defines generic macros to use complex or
2552 real valued functions.
2554 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2556 * Andreas Jaeger provided a test suite for the math library.
2558 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2560 * Ulrich Drepper added character set conversion functions (iconv).
2562 * Optimized string functions have been added.
2564 * The localedata addon is now part of glibc.
2566 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2568 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2569 daemon for NSS (nscd).
2571 Missing a better place here are some numbers on improvements. Under
2572 Linux 2.1.125 un-tar-ing the kernel sources takes
2576 using local files 12.19s 6.88s 22.91s
2578 using NIS 13.92s 8.91s 26.34s
2580 using NIS & nscd 10.37s 7.34s 25.30s
2582 using NIS+ 27.57s 30.37s 640.46s
2584 using NIS+ & nscd 10.25s 7.83s 26.51s
2586 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2588 Keep in mind that non-namelookup related operations dominate above times.
2589 It was just a common complain that using NIS+ unpacking the kernel is
2592 [1] The old nscd implementation is not available anymore since it was
2593 distributed with glibc up to version 2.0.98 and thus is now replaced.
2595 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2597 * Mark Kettenis provided a Hesiod NSS module.
2599 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2600 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2602 * rcmd can now handle netgroups (Dick Streefland).
2604 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2607 * Support for the IPv6 protocol has been added to the socket API, as per the
2608 latest draft standards.
2610 * Support for Linux 2.2 has been added.
2612 * Interface changes relative to the latest 2.0.x release:
2613 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2614 addseverity NEW: Unix98
2615 alphasort64 NEW: LFS
2616 argp_err_exit_status NEW: argp, GNU ext
2617 argp_error NEW: argp, GNU ext
2618 argp_failure NEW: argp, GNU ext
2619 argp_help NEW: argp, GNU ext
2620 argp_parse NEW: argp, GNU ext
2621 argp_program_bug_address NEW: argp, GNU ext
2622 argp_program_version NEW: argp, GNU ext
2623 argp_program_version_hook NEW: argp, GNU ext
2624 argp_state_help NEW: argp, GNU ext
2625 argp_usage NEW: argp, GNU ext
2626 authdes_create NEW: Secure RPC
2627 authdes_getucred NEW: Secure RPC
2628 authdes_pk_create NEW: Secure RPC
2629 backtrace NEW: GNU ext.
2630 backtrace_symbols NEW: GNU ext.
2631 backtrace_symbols_fd NEW: GNU ext.
2633 cacosf NEW: ISO C 9x
2634 cacosh NEW: ISO C 9x
2635 cacoshf NEW: ISO C 9x
2636 cacoshl NEW: ISO C 9x
2637 cacosl NEW: ISO C 9x
2644 casinf NEW: ISO C 9x
2645 casinh NEW: ISO C 9x
2646 casinhf NEW: ISO C 9x
2647 casinhl NEW: ISO C 9x
2648 casinl NEW: ISO C 9x
2650 catanf NEW: ISO C 9x
2651 catanh NEW: ISO C 9x
2652 catanhf NEW: ISO C 9x
2653 catanhl NEW: ISO C 9x
2654 catanl NEW: ISO C 9x
2655 cbc_crypt NEW: Secure RPC
2659 ccoshf NEW: ISO C 9x
2660 ccoshl NEW: ISO C 9x
2666 cimagf NEW: ISO C 9x
2667 cimagl NEW: ISO C 9x
2668 clearerr_locked REMOVED
2669 clntunix_create NEW: sunrpc ext
2671 clog10 NEW: ISO C 9x
2672 clog10f NEW: ISO C 9x
2673 clog10l NEW: ISO C 9x
2683 cprojf NEW: ISO C 9x
2684 cprojl NEW: ISO C 9x
2686 crealf NEW: ISO C 9x
2687 creall NEW: ISO C 9x
2692 csinhf NEW: ISO C 9x
2693 csinhl NEW: ISO C 9x
2696 csqrtf NEW: ISO C 9x
2697 csqrtl NEW: ISO C 9x
2701 ctanhf NEW: ISO C 9x
2702 ctanhl NEW: ISO C 9x
2704 des_setparity NEW: Secure RPC
2705 ecb_crypt NEW: Secure RPC
2706 endutxent NEW: Unix98
2708 exp10f NEW: ISO C 9x
2709 exp10l NEW: ISO C 9x
2713 fattach NEW: STREAMS
2714 fdetach NEW: STREAMS
2718 feclearexcept NEW: ISO C 9x
2719 fegetenv NEW: ISO C 9x
2720 fegetexceptflag NEW: ISO C 9x
2721 fegetround NEW: ISO C 9x
2722 feholdexcept NEW: ISO C 9x
2724 feraiseexcept NEW: ISO C 9x
2725 ferror_locked REMOVED
2726 fesetenv NEW: ISO C 9x
2727 fesetexceptflag NEW: ISO C 9x
2728 fesetround NEW: ISO C 9x
2729 fetestexcept NEW: ISO C 9x
2730 feupdateenv NEW: ISO C 9x
2731 fflush_locked REMOVED
2735 fgets_unlocked NEW: GNU ext.
2736 fileno_locked REMOVED
2748 fputc_locked REMOVED
2749 fputs_unlocked NEW: GNU ext.
2750 fread_unlocked NEW: GNU ext.
2755 fstatvfs NEW: Unix98
2759 ftruncate64 NEW: LFS
2761 fwrite_unlocked NEW: GNU ext.
2762 gai_strerror NEW: IPv6
2766 getchar_locked REMOVED
2768 getdate_err NEW: Unix98
2769 getdate_r NEW: GNU ext.
2771 getnameinfo NEW: IPv6
2772 getnetname NEW: Secure RPC
2773 getpmsg NEW: STREAMS
2774 getpt NEW: Unix98 PTY
2775 getrlimit64 NEW: LFS
2776 getutxent NEW: Unix98
2777 getutxid NEW: Unix98
2778 getutxline NEW: Unix98
2779 glob64 NEW: GNU ext.
2780 globfree64 NEW: GNU ext.
2781 gnu_get_libc_release NEW: GNU ext.
2782 gnu_get_libc_version NEW: GNU ext.
2783 grantpt NEW: Unix98 PTY
2784 host2netname NEW: Secure RPC
2786 iconv_close NEW: iconv
2787 iconv_open NEW: iconv
2788 if_freenameindex NEW: IPv6
2789 if_indextoname NEW: IPv6
2790 if_nameindex NEW: IPv6
2791 if_nametoindex NEW: IPv6
2792 in6addr_any NEW: IPv6
2793 in6addr_loopback NEW: IPv6
2794 inet6_isipv4mapped NEW: IPv6
2795 isastream NEW: STREAMS
2796 iswblank NEW: GNU ext.
2797 key_decryptsession NEW: Secure RPC
2798 key_decryptsession_pk NEW: Secure RPC
2799 key_encryptsession NEW: Secure RPC
2800 key_encryptsession_pk NEW: Secure RPC
2801 key_gendes NEW: Secure RPC
2802 key_get_conv NEW: Secure RPC
2803 key_secretkey_is_set NEW: Secure RPC
2804 key_setnet NEW: Secure RPC
2805 key_setsecret NEW: Secure RPC
2806 llrint NEW: ISO C 9x
2807 llrintf NEW: ISO C 9x
2808 llrintl NEW: ISO C 9x
2809 llround NEW: ISO C 9x
2810 llroundf NEW: ISO C 9x
2811 llroundl NEW: ISO C 9x
2816 lrintf NEW: ISO C 9x
2817 lrintl NEW: ISO C 9x
2818 lround NEW: ISO C 9x
2819 lroundf NEW: ISO C 9x
2820 lroundl NEW: ISO C 9x
2822 makecontext NEW: Unix98
2823 mempcpy NEW: GNU ext.
2826 modify_ldt NEW: kernel
2830 nearbyint NEW: ISO C 9x
2831 nearbyintf NEW: ISO C 9x
2832 nearbyintl NEW: ISO C 9x
2833 netname2host NEW: Secure RPC
2834 netname2user NEW: Secure RPC
2835 nexttoward NEW: ISO C 9x
2836 nexttowardf NEW: ISO C 9x
2837 nexttowardl NEW: ISO C 9x
2841 passwd2des NEW: Secure RPC
2843 pow10f NEW: GNU ext.
2844 pow10l NEW: GNU ext.
2847 printf_size NEW: GNU ext.
2848 printf_size_info NEW: GNU ext.
2849 profil_counter REMOVED
2850 pthread_mutexattr_getkind_np REPLACED
2851 pthread_mutexattr_setkind_np REPLACED
2852 ptsname NEW: Unix98 PTY
2853 ptsname_r NEW: Unix98 PTY
2855 putchar_locked REMOVED
2856 putgrent NEW: GNU ext.
2858 putpmsg NEW: STREAMS
2859 pututxline NEW: Unix98
2863 readdir64_r NEW: LFS
2864 remquo NEW: ISO C 9x
2865 remquof NEW: ISO C 9x
2866 remquol NEW: ISO C 9x
2868 roundf NEW: ISO C 9x
2869 roundl NEW: ISO C 9x
2871 scalbln NEW: ISO C 9x
2872 scalblnf NEW: ISO C 9x
2873 scalblnl NEW: ISO C 9x
2875 sendfile NEW: kernel
2876 setcontext NEW: Unix98
2877 setrlimit64 NEW: LFS
2878 setutxent NEW: Unix98
2880 sigignore NEW: Unix98
2881 sigqueue NEW: POSIX.1b
2882 sigrelse NEW: Unix98
2883 sigset NEW: POSIX.1b
2884 sigtimedwait NEW: POSIX.1b
2885 sigwaitinfo NEW: POSIX.1b
2886 sincos NEW: GNU ext.
2887 sincosf NEW: GNU ext.
2888 sincosl NEW: GNU ext.
2892 strcasestr NEW: GNU ext.
2893 strtoimax NEW: ISO C 9x
2894 strtoumax NEW: ISO C 9x
2895 strverscmp NEW: GNU ext.
2896 svcauthdes_stats NEW: Secure RPC
2897 svcunix_create NEW: sunrpc etx
2898 svcunixfd_create NEW: sunrpc ext
2899 swapcontext NEW: Unix98
2900 tcgetsid NEW: Unix98 PTY
2901 tdestroy NEW: GNU ext.
2902 tgamma NEW: ISO C 9x
2903 tgammaf NEW: ISO C 9x
2904 tgammal NEW: ISO C 9x
2908 truncf NEW: ISO C 9x
2909 truncl NEW: ISO C 9x
2911 unlockpt NEW: Unix98 PTY
2912 updwtmpx NEW: Unix98
2913 user2netname NEW: Secure RPC
2914 utmpxname NEW: Unix98
2915 versionsort NEW: GNU ext.
2916 versionsort64 NEW: GNU ext.
2918 wcscasecmp NEW: GNU ext.
2919 wcsncasecmp NEW: GNU ext.
2920 wcsnlen NEW: GNU ext.
2921 wcstoimax NEW: ISO C 9x
2922 wcstoll NEW: ISO C 9x
2923 wcstoull NEW: ISO C 9x
2924 wcstoumax NEW: ISO C 9x
2926 wordexp NEW: POSIX.2
2927 wordfree NEW: POSIX.2
2928 write_profiling REMOVED
2929 xdecrypt NEW: Secure RPC
2930 xdr_authdes_cred NEW: Secure RPC
2931 xdr_authdes_verf NEW: Secure RPC
2932 xdr_cryptkeyarg NEW: Secure RPC
2933 xdr_cryptkeyarg2 NEW: Secure RPC
2934 xdr_cryptkeyres NEW: Secure RPC
2935 xdr_getcredres NEW: Secure RPC
2936 xdr_key_netstarg NEW: Secure RPC
2937 xdr_key_netstres NEW: Secure RPC
2938 xdr_keybuf NEW: Secure RPC
2939 xdr_keystatus NEW: Secure RPC
2940 xdr_netnamestr NEW: Secure RPC
2941 xdr_sizeof NEW: Secure RPC
2942 xdr_unixcred NEW: sunrpc ext
2943 xencrypt NEW: Secure RPC
2944 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2955 * inet_ntoa is thread-safe
2957 * updwtmp is moved from libutil to libc
2959 * rewrite of cbrt function
2961 * update of timezone data
2975 * add atoll function
2977 * fix complex problems in Berkeley DB code
2979 * fix math functions
2983 * fixed lots of header problems (especially Linux/GNU specific)
2985 * dynamic loader preserves all registers
2987 * Roland McGrath provided support for handling of auxiliary objects in
2988 the ELF dynamic loader.
2990 * support for parallel builds is improved
2994 * GNU extensions are no longer declared by default. To enable them you
2995 must define the macro `_GNU_SOURCE' in your program or compile with
2998 * The library has changed from using GNU ld symbol aliases to using weak
2999 symbols where available. The ELF object file format supports weak
3000 symbols; GNU ld also supports weak symbols in the a.out format. (There
3001 is also now support for other GNU ld extensions in ELF. Use the
3002 `--with-elf' option to configure to indicate you have ELF, and
3003 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3004 of many files which contained only symbol aliases, reducing the size of
3005 the source and the compiled library; many other files were renamed to
3006 less cryptic names previously occupied by the symbol alias files.
3007 There is a new header file <elf.h> for programs which operate on
3008 files in the ELF format.
3010 * Converted to Autoconf version 2, so `configure' has more options.
3011 Run `configure --help' to see the details.
3013 * The library can now be configured to build profiling, highly-optimized
3014 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3015 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3016 `configure' enable building these extra libraries. The shared library is
3017 built by default when using both ELF and GNU ld. When shared libraries
3018 are enabled, the new library `-ldl' is available for arbitrary run-time
3019 loading of shared objects; its interface is defined in <dlfcn.h>. The
3020 new header file <link.h> gives access to the internals of the run-time
3021 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3022 application of same name on other systems and it provides information
3023 about dynamically linked binaries.
3025 * The C library now provides the run-time support code for profiling
3026 executables compiled with `-pg'. Programs can control the profiling code
3027 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3028 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3029 the support for this file format was contributed by David Mosberger-Tang.
3031 * The math code has been replaced with a math library based on fdlibm from
3032 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3033 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3034 functions. The math functions now reside in a separate library, so
3035 programs using them will need to use `-lm' their linking commands.
3037 * John C. Bowman contributed optimized ix87 assembler inline functions.
3039 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3040 mechanism similar to that found in Solaris 2. This is now used for the
3041 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3042 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3043 file controls what services are used for each individual database. This
3044 works by loading shared libraries with names specified in `nsswitch.conf',
3045 so service modules can be changed or added at any time without even
3046 relinking any program. Currently there are the file, db, and NIS based
3047 NSS services available.
3049 * The new functions `strtoq' and `strtouq' parse integer values from
3050 strings, like `strtol' and `strtoul', but they return `long long int' and
3051 `unsigned long long int' values, respectively (64-bit quantities).
3053 * The new functions `strtof' and `strtold' parse floating-point values from
3054 strings, like `strtod', but they return `float' and `long double' values,
3055 respectively (on some machines `double' and `long double' are the same).
3057 * Ulrich Drepper has contributed new implementations of the floating-point
3058 printing and reading code used in the `printf' family of functions and
3059 `strtod', `strtof', and `strtold'. These new functions are perfectly
3060 accurate, and much faster than the old ones.
3062 * The implementation of the POSIX locale model was completely rewritten by
3063 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3064 to compile the POSIX locale definition.
3066 * The former dummy implementations of the strcoll and strxfrm function are
3067 now replaced by fully functional code contributed by Ulrich Drepper. The
3068 collation information comes from the POSIX locale definitions.
3070 * The new header <langinfo.h> defines an interface for accessing
3071 various locale-dependent data (using the locale chosen with `setlocale').
3073 * Ulrich Drepper has contributed a new suite of functions for operation on
3074 wide-character and multibyte-character strings, in <wchar.h>;
3075 and classification and case conversion of wide characters, in <wctype.h>.
3076 These new functions are conforming to the ISO C, Amendement 1 specification.
3078 * There is now a second implementation of the standard I/O library available.
3079 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3080 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3082 * You can now use positional parameter specifications in format strings
3083 for the `printf' and `scanf' families of functions. For example,
3084 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3085 ``Number 6, Mr Jones''. This is mainly useful when providing different
3086 format strings for different languages, whose grammars may dictate
3087 different orderings of the values being printed. To support this
3088 feature, the interface for `register_printf_handler' has changed; see
3089 the header file <printf.h> for details.
3091 * The `printf' and `scanf' families of functions now understand a new
3092 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3093 to group numbers as indicated by the locale; for `scanf' and friends, this
3094 says to accept as valid only a number with all the proper grouping
3095 separators in the right places. In the default "C" locale, numbers are
3096 not grouped; but locales for specific countries will define the usual
3097 conventions (i.e. separate thousands with `,' in the US locale).
3099 * The pgrp functions have been regularized, slightly incompatibly but much
3100 less confusingly. The core functions are now `getpgid' and `setpgid',
3101 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3102 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3103 provided for compatibility. There is no longer an incompatible `getpgrp'
3104 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3106 * The new header file <fts.h> and suite of functions simplify programs that
3107 operate on directory trees. This code comes from 4.4 BSD.
3109 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3110 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3111 NSS scheme used in glibc.
3113 * There is a new malloc debugging hook `__memalign_hook'.
3115 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3116 `unsigned int' in <sys/types.h>. These are for compatibility only and
3117 their use is discouraged.
3119 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3120 done differently, so that it works even without GNU ld.
3122 * New function `euidaccess' checks allowed access to a file like `access',
3123 but using the effective IDs instead of the real IDs.
3125 * The time zone data files have been updated for the latest and greatest
3126 local time conventions of the countries of the world.
3128 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3131 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3132 for formatting floating-point numbers. They are provided only for
3133 compatibility; new programs should use `sprintf' instead. There are
3134 also equivalent function for the `long double' floating-point type and
3135 all functions also exist in a reentrant form.
3137 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3138 functions for maintaining the login-record files (primarily of use to
3139 system programs such as `login'), and convenient functions for
3140 allocating and initializing a pseudo-terminal (pty) device.
3142 * Ulrich Drepper has contributed new support for System V style
3143 shared memory and IPC on systems that support it.
3145 * Ulrich Drepper has contributed several miscellaneous new functions found
3146 in System V: The `hsearch' family of functions provide an effective
3147 implementation of hash tables; `a64l' and `l64a' provide a very simple
3148 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3151 * Ulrich Drepper has contributed new reentrant counterparts for the
3152 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3154 * Ulrich Drepper has contributed new, highly-optimized versions of several
3155 string functions for the i486/Pentium family of processors.
3157 * Ulrich Drepper has updated the Linux-specific code, based largely
3158 on work done in Hongjiu Lu's version of GNU libc for Linux.
3159 The GNU library now supports Linux versions 2.0.10 and later,
3160 using the ELF object file format (i[3456]86-*-linux).
3162 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3164 * David Mosberger-Tang and Richard Henderson have ported the C library
3165 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3166 dynamic linking support for ELF/Alpha.
3168 * Richard Henderson contributed several Alpha optimized assembler function
3169 for arithmetic and string handling.
3171 * Ulrich Drepper has contributed a new set of message catalog functions to
3172 support multiple languages using the <libintl.h> interface, for use with
3173 his new package GNU gettext. Translation volunteers have contributed
3174 catalogs of the library's messages in Spanish, German, and Korean.
3176 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3177 program and the `catgets' function for reading the catalog files it
3178 creates. (The <libintl.h> interface is preferred; we include the
3179 <nl_types.h> interface using `catgets' only for source compatibility with
3180 programs already written to use it.)
3182 * New header file <values.h> gives SVID-compatible names for <limits.h>
3185 * Various new macros, declarations, and small header files for compatibility
3188 * New function `group_member' is a convenient way to check if a process has
3189 a given effective group ID.
3191 * When using GCC 2.7 and later, the socket functions are now declared in a
3192 special way so that passing an argument of type `struct sockaddr_in *',
3193 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3194 `struct sockaddr *' type, does not generate a type-clash warning.
3196 * New function `error' declared in header file <error.h> is a convenient
3197 function for printing error messages and optionally exiting; this is the
3198 canonical function used in GNU programs. The new functions `err', `warn',
3199 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3200 doing the same thing.
3202 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3203 POSIX.2 `glob' function to do ~ and {...} expansion.
3205 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3206 `clearenv' which is used in POSIX.9 is also available.
3208 * New function `getsid' returns session ID number on systems that support it.
3210 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3211 files <db.h> and <mpool.h> provide a rich set of functions for several
3212 types of simple databases stored in memory and in files, and <ndbm.h> is
3213 an old `ndbm'-compatible interface using the `db' functions. Link with
3214 `-ldb' to get these functions.
3216 * New macro `strdupa' copies a string like `strdup', but uses local stack
3217 space from `alloca' instead of dynamic heap space from `malloc'.
3219 * New function `strnlen' is like `strlen' but searches only a given maximum
3220 number of characters for the null terminator. `stpncpy', `strndup' and
3221 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3224 * New function `statfs' in header <sys/statfs.h>.
3226 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3227 provide convenient functions for operating on blocks of null-terminated
3230 * A new suite of functions in <utmp.h> handle all the details of reading
3231 and writing the utmp file.
3233 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3236 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3237 completely POSIX compliant and also implemented the extended functionality
3238 to handle alternate digit representation and alternate era date formats.
3240 * Ulrich Drepper provided an implementation of the `strptime' function
3241 defined in XPG4.2 which transforms a string into a `struct tm' value.
3243 * Paul Eggert provided the tzselect shell script as part of the timezone
3244 code. The shell script makes it easy to select the correct timezone
3247 * The implementation of the malloc family of functions is completely replaced
3248 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3249 The implementation uses the mmap function (if available) and it is
3250 optimized for the use in multi threaded programs.
3252 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3253 functions. This new functionality is usable by specifying a special
3254 salt string and it is compatible with implementation on *BSD systems.
3256 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3257 `getsubopt' to handle second level command line options, `bsd_signal'
3258 to access BSD style `signal' functionality, the obsolete `regexp' style
3261 * the `lchown' function is available on system which support this
3264 * The implementation of the shadow password handling function was contributed
3267 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3269 * POSIX.1g support was added. The <sys/select.h> header is available,
3270 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3271 implementation of `getaddrinfo'.
3275 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3277 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3278 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3280 * The new function `daemon' from 4.4 BSD is useful for server programs that
3281 want to put themselves in the background.
3283 * Joel Sherrill has contributed support for several standalone boards that
3284 run without an operating system.
3286 * `printf', `scanf' and friends now accept a `q' type modifier for long
3287 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3289 * All of the code taken from BSD (notably most of the math and networking
3290 routines) has been updated from the BSD 4.4-Lite release.
3292 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3294 * The new functions `getdomainname' and `setdomainname' fetch or change the
3295 YP/NIS domain name. These are system calls which exist on systems which
3298 * The time zone data files have been updated for the latest international
3301 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3302 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3306 * The C library now includes support for Sun RPC, from Sun's free
3307 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3308 are included. (There is still no support for YP.)
3310 * Tom Quinn has contributed a port of the C library to SGI machines running
3311 Irix 4 (mips-sgi-irix4).
3313 * The new `lockf' function is a simplified interface to the locking
3314 facilities of `fcntl', included for compatibility.
3316 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3318 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3320 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3323 * The error message printed by `assert' for a failed assertion now includes
3324 the name of the program (if using GNU ld) and the name of the calling
3325 function (with versions of GCC that support this).
3327 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3329 * The library now includes the <sys/mman.h> header file and memory
3330 management functions `mmap', `munmap', `mprotect', `msync', and
3331 `madvise', on systems that support those facilities.
3333 * The interface for `mcheck' has changed slightly: the function called to
3334 abort the program when an allocation inconsistency is detected now takes
3335 an argument that indicates the type of failure. The new function
3336 `mprobe' lets you request a consistency check for a particular block at
3337 any time (checks are normally done only when you call `free' or `realloc'
3340 * It is now possible to easily cross-compile the C library, building on one
3341 system a library to run on another machine and/or operating system. All
3342 you need to do is set the variable `HOST_CC' in `configparms' to the
3343 native compiler for programs to run on the machine you are building on (a
3344 few generator programs are used on Unix systems); set `CC' to the
3347 * The new function `fexecve' (only implemented on the GNU system) executes
3348 a program file given a file descriptor already open on the file.
3352 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3353 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3355 * You can set the variable `libprefix' in `configparms' to specify a prefix
3356 to be prepended to installed library files; this makes it easy to install
3357 the GNU C library to be linked as `-lgnuc' or whatever.
3359 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3360 copies a limited number of characters from a string, and returns the
3361 address of the last character written.
3363 * You no longer need to check for whether the installed `stddef.h' is
3364 compatible with the GNU C library. configure now checks for you.
3366 * You can now define a per-stream `fileno' function to convert the
3367 stream's cookie into an integral file descriptor.
3369 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3370 allocates zero bytes of storage, and returns a unique pointer which
3371 you can pass to `realloc' or `free'. The behavior is undefined if
3372 you dereference this pointer.
3374 * The C library now runs on Sony NEWS m68k machines running either
3375 NewsOS 3 or NewsOS 4.
3377 * The new `syscall' function is a system-dependent primitive function
3378 for invoking system calls. It has the canonical behavior on Unix
3379 systems, including unreliable return values for some calls (such as
3380 `pipe', `fork' and `getppid').
3382 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3383 to `EAGAIN', which is the preferred name. On systems whose kernels
3384 use two distinct codes, the C library now translates EWOULDBLOCK to
3385 EAGAIN in every system call function.
3389 * The GNU C Library Reference Manual is now distributed with the library.
3390 `make dvi' will produce a DVI file of the printed manual.
3391 `make info' will produce Info files that you can read on line using C-h i
3392 in Emacs or the `info' program.
3393 Please send comments on the manual to bug-glibc-manual@gnu.org.
3395 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3397 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3399 * Jason Merrill has contributed a port to the Sequent Symmetry running
3400 Dynix version 3 (i386-sequent-dynix).
3402 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3403 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3405 * New function `memory_warnings' lets you arrange to get warnings when
3406 malloc is running out of memory to allocate, like Emacs gives you.
3408 * The C library now contains the relocating allocator used in Emacs 19 for
3409 its editing buffers. This allocator (ralloc) minimizes allocation
3410 overhead and fragmentation by moving allocated regions around whenever it
3411 needs to. You always refer to a ralloc'd region with a "handle" (a
3412 pointer to a pointer--an object of type `void **').
3414 * There is a new `printf' format: `%m' gives you the string corresponding
3415 to the error code in `errno'.
3417 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3418 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3419 pass, the `a' modifier says to fill in a `char **' you pass with a
3422 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3423 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3424 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3426 * `mkstemp' is a traditional Unix function to atomically create and open a
3427 uniquely-named temporary file.
3431 * The standard location for the file that says what the local timezone is
3432 has changed again. It is now `/usr/local/etc/localtime' (or more
3433 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3435 * The distribution no longer contains any files with names longer than 14
3438 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3439 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3441 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3445 * The configuration process has changed quite a bit. The `configure'
3446 script is now used just like the configuration scripts for other GNU
3447 packages. The `sysdeps' directory hierarchy is much rearranged.
3448 The file `INSTALL' explains the new scheme in detail.
3450 * The header files no longer need to be processed into ANSI C and
3451 traditional C versions. There is just one set of files to install, and
3452 it will work with ANSI or old C compilers (including `gcc -traditional').
3454 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3455 MIPS DECStation running Ultrix 4.
3457 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3458 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3459 made itself into a shared library.
3461 * Yet further improved support for the i386, running 4.3 BSD-like systems
3462 (such as Mach 3 with the Unix single-server), or System V.
3464 * New function `strncasecmp' to do case-insensitive string comparison
3465 with limited length.
3467 * New function `strsep' is a reentrant alternative to `strtok'.
3469 * New functions `scandir' and `alphasort' for searching directories.
3471 * New function `setenv' is a better interface to `putenv'.
3473 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3474 function for traversing a directory tree.
3476 * The GNU obstack package is now also part of the C library.
3477 The new function `open_obstack_stream' creates a stdio stream that
3478 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3479 formatted output directly to an obstack.
3481 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3482 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3484 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3486 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3487 things to your strings.
3489 * There are some new test programs: test-fseek, testmb, and testrand.
3491 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3492 These ports are not finished, but are a good starting place for really
3493 supporting those systems.
3495 * `/etc/localtime' is now the standard location for the file that says what
3496 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3497 This follows the general principle that `/etc' is the place for all local
3498 configuration files.
3500 * The C library header files now use `extern "C"' when used by the C++
3501 compiler, so the C library should now work with C++ code.
3503 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3504 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3507 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3508 function which uses the `merge sort' algorithm, and is said to be
3509 significantly faster than the old GNU `qsort' function. Merge sort is
3510 now the standard `qsort' function. The new algorithm can require a lot
3511 of temporary storage; so, the old sorting function is called when the
3512 required storage is not available.
3514 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3515 provides the Unix `crypt' function, plus some other entry points.
3517 * The code and header files taken from 4.4 BSD have been updated with the
3518 latest files released from Berkeley.
3520 ----------------------------------------------------------------------
3521 Copyright information:
3523 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3525 Permission is granted to anyone to make or distribute verbatim copies
3526 of this document as received, in any medium, provided that the
3527 copyright notice and this permission notice are preserved,
3528 thus giving the recipient permission to redistribute in turn.
3530 Permission is granted to distribute modified versions
3531 of this document, or of portions of it,
3532 under the above conditions, provided also that they
3533 carry prominent notices stating who last changed them.
3536 version-control: never