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 - Comparison macros: iseqsig.
58 - Classification macros: iscanonical, issubnormal, iszero.
60 - Total order functions: totalorder, totalorderf, totalorderl,
61 totalordermag, totalordermagf, totalordermagl.
63 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
65 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
66 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
68 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
69 are added to libc. They convert a floating-point number into string.
71 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
72 to float instead of double. This does not affect the ABI of any libraries
73 that are part of the GNU C Library, but may affect the ABI of other
74 libraries that use this type in their interfaces.
76 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
77 float_t and double_t types are now defined to long double instead of float
78 and double. These options are not the default, and this does not affect
79 the ABI of any libraries that are part of the GNU C Library, but it may
80 affect the ABI of other libraries that use this type in their interfaces,
81 if they are compiled or used with those options.
83 * The buffer size for byte-oriented stdio streams is now limited to 8192
84 bytes by default. Previously, on Linux, the default buffer size on most
85 file systems was 4096 bytes (and thus remains unchanged), except on
86 network file systems, where the buffer size was unpredictable and could be
87 as large as several megabytes.
89 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
90 for the Linux quota interface which predates kernel version 2.4.22 has
93 * The malloc_get_state and malloc_set_state functions have been removed.
94 Already-existing binaries that dynamically link to these functions will
95 get a hidden implementation in which malloc_get_state is a stub. As far
96 as we know, these functions are used only by GNU Emacs and this change
97 will not adversely affect already-built Emacs executables. Any undumped
98 Emacs executables, which normally exist only during an Emacs build, should
99 be rebuilt by re-running “./configure; make” in the Emacs build tree.
101 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
102 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
103 “no-ip6-dotint” had already been the default, and support for the
104 “ip6-dotint” option was removed from the Internet in 2006.
106 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
107 flag from <resolv.h> have been removed. The option relied on a
108 backwards-incompatible DNS extension which was never deployed on the
111 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
112 in the <resolv.h> header file have been deprecated. They were already
115 * DNSSEC-related declarations and definitions have been removed from the
116 <arpa/nameser.h> header file, and libresolv will no longer attempt to
117 decode the data part of DNSSEC record types. Previous versions of glibc
118 only implemented minimal support for the previous version of DNSSEC, which
119 is incompatible with the currently deployed version.
121 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
122 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
123 <arpa/nameser.h> header file because the distinction between RR types and
124 meta-RR types is not officially standardized, subject to revision, and
125 thus not suitable for encoding in a macro.
127 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
128 and rhook members of the res_state type in <resolv.h> have been removed.
129 The glibc stub resolver did not support these hooks, but the header file
130 did not reflect that.
132 * For multi-arch support it is recommended to use a GCC which has
133 been built with support for GNU indirect functions. This ensures
134 that correct debugging information is generated for functions
135 selected by IFUNC resolvers. This support can either be enabled by
136 configuring GCC with '--enable-gnu-indirect-function', or by
137 enabling it by default by setting 'default_gnu_indirect_function'
138 variable for a particular architecture in the GCC source file
141 Security related changes:
143 On ARM EABI (32-bit), generating a backtrace for execution contexts which
144 have been created with makecontext could fail to terminate due to a
145 missing .cantunwind annotation. This has been observed to lead to a hang
146 (denial of service) in some Go applications compiled with gccgo. Reported
147 by Andreas Schwab. (CVE-2016-6323)
149 The following bugs are resolved with this release:
151 [The release manager will add the list generated by
152 scripts/list-fixed-bugs.py just before the release.]
156 * The minimum Linux kernel version that this version of the GNU C Library
157 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
158 kernel version 2.6.32 or later suffices (on architectures that already
159 required kernel versions more recent than 3.2, those requirements remain
160 unchanged). Linux 3.2 or later kernel headers are required on all
163 * The pap_AN locale has been deleted. This has been deprecated for a long
164 time. It has been replaced by pap_AW & pap_CW, both of which have long
165 been included in previous releases.
167 * The readdir_r and readdir64_r functions have been deprecated. It is
168 recommended to use readdir and readdir64 instead.
170 * The type “union wait” has been removed. It was deprecated in the early
171 1990s and never part of POSIX. Application code should use the int type
172 instead of “union wait”.
174 * A new NSS action is added to facilitate large distributed system
175 administration. The action, MERGE, allows remote user stores like LDAP
176 to be merged into local user stores like /etc/groups in order to provide
177 easy to use, updated, and managed sets of merged credentials. The new
178 action can be used by configuring it in /etc/nsswitch.conf:
179 group: files [SUCCESS=merge] nis
180 Implemented by Stephen Gallagher (Red Hat).
182 * The deprecated __malloc_initialize_hook variable has been removed from the
185 * The long unused localedef --old-style option has been removed. It hasn't
186 done anything in over 16 years. Scripts using this option can safely
189 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
190 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
191 return the next representable value in the direction of positive infinity
192 and the nextdown functions return the next representable value in the
193 direction of negative infinity. These are currently enabled as GNU
196 Security related changes:
198 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
199 could result in a stack overflow when getnetbyname was called with an
200 overly long name. (CVE-2016-3075)
202 * Previously, getaddrinfo copied large amounts of address data to the stack,
203 even after the fix for CVE-2013-4458 has been applied, potentially
204 resulting in a stack overflow. getaddrinfo now uses a heap allocation
205 instead. Reported by Michael Petlan. (CVE-2016-3706)
207 * The glob function suffered from a stack-based buffer overflow when it was
208 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
209 Reported by Alexander Cherepanov. (CVE-2016-1234)
211 * The Sun RPC UDP client could exhaust all available stack space when
212 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
213 alloca plugin for GCC. (CVE-2016-4429)
215 * The IPv6 name server management code in libresolv could result in a memory
216 leak for each thread which is created, performs a failing naming lookup,
217 and exits. Over time, this could result in a denial of service due to
218 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
220 The following bugs are resolved with this release:
222 [1170] localedata: ne_NP: update Nepali locale definition file
223 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
225 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
226 [6796] math: fdim() does not set errno on overflow
227 [10354] libc: posix_spawn should use vfork() in more cases than presently
228 [11213] localedata: localedata: add copyright disclaimer to locale files
229 [12143] localedata: chr_US: new Cherokee locale
230 [12450] localedata: sgs_LT: new locale
231 [12676] localedata: ln_CD: new locale
232 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
234 [13304] math: fma, fmaf, fmal produce wrong results
235 [14259] build: --localedir arg to configure is ignored
236 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
237 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
238 [14934] localedata: es_CL: wrong first weekday chilean locale
239 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
241 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
243 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
244 [15368] nptl: raise() is not async-signal-safe
245 [15479] math: ceil, floor, round and trunc raise inexact exception
246 [15578] localedata: kk_KZ: various updates
247 [16003] localedata: pap_AN: punt old locale
248 [16137] localedata: iw_IL: punt old locale
249 [16190] localedata: eo: new esperanto locale
250 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
251 [16742] malloc: race condition: pthread_atfork() called before first
252 malloc() results in unexpected locking behaviour/deadlocks
253 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
255 [16983] localedata: postal_fmt does not allow %l and %n modifiers
256 [17565] localedata: pt_PT: wrong (work-)week start
257 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
258 [17950] build: Build fails with -msse
259 [18205] localedata: be_BY*: wrong first_weekday and first_workday
260 [18433] libc: posix_spawn does not return correctly upon failure to
262 [18453] localedata: charmaps/IBM875: incorrect codes
263 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
265 [18896] localedata: he_IL: improvements for currency
266 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
268 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
270 [19056] libc: Deprecate readdir_r
271 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
273 [19198] localedata: nl_NL: small improvements for Dutch locales
274 [19257] network: Per-thread memory leak in __res_vinit with IPv6
275 nameservers (CVE-2016-5417)
276 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
278 [19400] locale: Language missing in "iso-639.def", trivial fix in
280 [19431] malloc: Deadlock between fflush, getdelim, and fork
281 [19505] libc: Incorrect file descriptor validity checks in
282 posix_spawn_file_actions_add{open,close,dup2}
283 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
285 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
286 `intl/{gettextP,loadinfo}.h'
287 [19534] libc: execle, execlp may use malloc
288 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
289 [19573] network: res_nclose and __res_maybe_init disagree about name
290 server initialization, breaking Hesiod
291 [19575] localedata: Status of GB18030 tables
292 [19581] localedata: sr_* date_fmt string contains additional newline
293 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
295 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
296 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
297 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
298 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
300 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
302 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
304 [19610] dynamic-link: ldconfig -X removes stale symbolic links
305 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
306 [19633] locale: strfmon_l applies global locale to number formatting
307 [19642] network: Memory leak in getnameinfo
308 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
309 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
311 [19654] math: [x86_64] Need testcase for BZ #19590 fix
312 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
314 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
315 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
317 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
319 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
321 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
322 pointers and lengths in error-case.
323 [19727] locale: Converting from/to UTF-xx with iconv() does not always
324 report errors on UTF-16 surrogates values.
325 [19755] nscd: nscd assertion failure in gc
326 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
327 [19759] libc: mempcpy shouldn't be inlined
328 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
329 [19765] libc: s390 needs an optimized mempcpy
330 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
331 NAME_MAX limit assumption (CVE-2016-1234)
332 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
333 [19787] network: Missing and incorrect truncation checks in getnameinfo
334 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
336 [19791] network: Assertion failure in res_query.c with un-connectable name
338 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
339 [19822] math: libm.so install clobbers old version
340 [19825] network: resolv: send_vc can return uninitialized data in second
341 response to getaddrinfo
342 [19830] network: nss_dns: should check RDATA length against buffer length
343 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
344 confronted with A/AAAA records of invalid size
345 [19837] nss: nss_db: No retries for some long lines with a larger buffer
346 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
347 [19853] stdio: Printing IBM long double in decimal with high precision is
349 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
350 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
351 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
353 [19865] network: Assertion failure or memory leak in
354 _nss_dns_getcanonname_r
355 [19868] network: nss_dns: netent code does not skip over non-PTR records
356 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
358 [19881] string: Improve x86-64 memset
359 [19907] string: Incorrect memcpy tests
360 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
362 [19925] libc: termios.h XCASE namespace
363 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
364 [19929] libc: limits.h NL_NMAX namespace
365 [19931] stdio: Memory leak in vfprintf
366 [19957] libc: clone(CLONE_VM) access invalid parent memory
367 [19963] localedata: en_IL: New locale
368 [19989] stdio: stdio.h cuserid namespace
369 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
371 [19996] locale: langinfo.h nl_langinfo_l namespace
372 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
373 [20010] network: getaddrinfo: Stack overflow in hostent translation
375 [20012] stdio: libio: fmemopen append mode failure
376 [20014] stdio: stdio.h namespace for pre-threads POSIX
377 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
378 [20023] libc: fcntl.h timespec namespace
379 [20024] math: [x86_64] vectorized sincos trashes the stack
380 [20031] network: nss_hesiod: Heap overflow in get_txt_records
381 [20041] time: sys/time.h timespec namespace
382 [20043] libc: unistd.h missing cuserid for UNIX98 and before
383 [20044] libc: unistd.h missing pthread_atfork for UNIX98
384 [20051] libc: ttyslot in wrong header under wrong conditions
385 [20054] libc: gethostname not declared for XPG4
386 [20055] libc: termios.h missing tcgetsid for XPG4
387 [20072] dynamic-link: x86 init_cpu_features is called twice in static
389 [20073] libc: sys/stat.h fchmod namespace
390 [20074] libc: stdlib.h rand_r namespace
391 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
392 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
394 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
395 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
397 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
398 [20119] libc: Wrong mask for processors level type from CPUID
399 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
400 AS not supporting AVX512
401 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
403 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
404 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
406 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
407 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
408 [20178] libc: posix_spawn{p} should not call exit
409 [20191] stdio: libio: vtables hardening
410 [20195] string: FMA4 detection requires CPUID execution with register
412 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
413 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
415 [20212] math: acos (sNaN) returns sNaN
416 [20213] math: asin (sNaN) returns sNaN
417 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
418 [20218] math: [i386] asinhl (sNaN) returns sNaN
419 [20219] math: [i386] atanhl (sNaN) returns sNaN
420 [20222] stdio: fopencookie: Mangle function pointers
421 [20224] math: [i386] cbrtl (sNaN) returns sNaN
422 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
423 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
425 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
426 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
427 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
428 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
429 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
430 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
431 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
432 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
433 [20240] math: modf (sNaN) returns sNaN
434 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
435 [20250] math: frexp (sNaN) returns sNaN
436 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
437 [20255] math: [i386] fdim, fdimf return with excess range and precision /
439 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
440 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
441 subscript is below array bounds [-Werror=array-bounds]
442 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
443 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
445 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
446 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
448 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
450 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
451 [20316] localedata: id_ID: Februari instead of Pebruari
452 [20327] string: POWER8 strcasecmp returns incorrect result
453 [20347] math: Failure: Test: j0_downward (0xap+0)
454 [20348] libc: FAIL: misc/tst-preadvwritev64
455 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
456 [20350] libc: There is no test for p{read,write}64
457 [20357] math: Incorrect cos result for 1.5174239687223976
458 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
462 * Unicode 8.0.0 Support: Character encoding, character type info, and
463 transliteration tables are all updated to Unicode 8.0.0, using new
464 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
465 These updates cause user visible changes, such as the fixes for bugs
466 89, 16061, and 18568.
468 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
469 kernel-internal CPU set size. This means that requests that change the
470 CPU affinity which failed before (for example, an all-ones CPU mask) will
471 now succeed. Applications that need to determine the effective CPU
472 affinities need to call sched_getaffinity or pthread_getaffinity_np after
473 setting it because the kernel can adjust it (and the previous size check
474 would not detect this in the majority of cases).
476 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
477 the following new symbols are used: fts64_children, fts64_close,
478 fts64_open, fts64_read and fts64_set.
480 * getaddrinfo now detects certain invalid responses on an internal netlink
481 socket. If such responses are received, an affected process will
482 terminate with an error message of "Unexpected error <number> on netlink
483 descriptor <number>" or "Unexpected netlink response of size <number> on
484 descriptor <number>". The most likely cause for these errors is a
485 multi-threaded application which erroneously closes and reuses the netlink
486 file descriptor while it is used by getaddrinfo.
488 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
489 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
490 unnecessary serialization of memory allocation requests across threads.
491 The defect is now corrected. Users should see a substantial increase in
492 the concurent throughput of allocation requests for applications which
493 trigger this bug. Affected applications typically create and
494 destroy threads frequently. (Bug 19048 was reported and analyzed by
497 * There is now a --disable-timezone-tools configure option for disabling the
498 building and installing of the timezone related utilities (zic, zdump, and
499 tzselect). This is useful for people who build the timezone data and code
500 independent of the GNU C Library.
502 * The obsolete header <regexp.h> has been removed. Programs that require
503 this header must be updated to use <regex.h> instead.
505 * The obsolete functions bdflush, create_module, get_kernel_syms,
506 query_module and uselib are no longer available to newly linked binaries;
507 the header <sys/kdaemon.h> has been removed. These functions and header
508 were specific to systems using the Linux kernel and could not usefully be
509 used with the GNU C Library on systems with version 2.6 or later of the
512 * Optimized string, wcsmbs and memory functions for IBM z13.
513 Implemented by Stefan Liebler.
515 * Newly linked programs that define a variable called signgam will no longer
516 have it set by the lgamma, lgammaf and lgammal functions. Programs that
517 require signgam to be set by those functions must ensure that they use the
518 variable provided by the GNU C Library and declared in <math.h>, without
519 defining their own copy.
521 * The minimum GCC version that can be used to build this version of the GNU
522 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
523 still be used to compile programs using the GNU C Library.
525 Security related changes:
527 * An out-of-bounds value in a broken-out struct tm argument to strftime no
528 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
530 * The LD_POINTER_GUARD environment variable can no longer be used to disable
531 the pointer guard feature. It is always enabled. Previously,
532 LD_POINTER_GUARD could be used to disable security hardening in binaries
533 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
536 * An integer overflow in hcreate and hcreate_r could lead to an
537 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
539 * The catopen function no longer has unbounded stack usage. Reported by
542 * The nan, nanf and nanl functions no longer have unbounded stack usage
543 depending on the length of the string passed as an argument to the
544 functions. Reported by Joseph Myers. (CVE-2014-9761)
546 * A stack-based buffer overflow was found in libresolv when invoked from
547 libnss_dns, allowing specially crafted DNS responses to seize control
548 of execution flow in the DNS client. The buffer overflow occurs in
549 the functions send_dg (send datagram) and send_vc (send TCP) for the
550 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
551 family. The use of AF_UNSPEC triggers the low-level resolver code to
552 send out two parallel queries for A and AAAA. A mismanagement of the
553 buffers used for those queries could result in the response of a query
554 writing beyond the alloca allocated buffer created by
555 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
556 the overflow. Thanks to the Google Security Team and Red Hat for
557 reporting the security impact of this issue, and Robert Holiday of
558 Ciena for reporting the related bug 18665. (CVE-2015-7547)
560 The following bugs are resolved with this release:
562 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
563 [887] math: Math library function "logb" and "nextafter" inconsistent
564 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
565 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
566 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
567 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
569 [4404] localedata: German translation of "Alarm clock" is misleading
570 [6799] math: nextafter() and nexttoward() doen't set errno on
571 overflow/underflow errors
572 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
574 [10432] nis: _nss_nis_setnetgrent assertion failure
575 [11460] libc: fts has no LFS support
576 [12926] network: getaddrinfo()/make_request() may spin forever
577 [13065] nptl: Race condition in pthread barriers
578 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
579 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
581 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
582 [14912] libc: Rename non-installed bits/*.h headers
583 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
584 [15367] math: Let gcc use __builtin_isinf
585 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
586 [15421] math: lgamma wrongly sets signgam for ISO C
587 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
588 FE_INVALID with argument out of range
589 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
591 [15786] dynamic-link: ifunc resolver functions can smash function
593 [15918] math: Unnecessary check for equality in hypotf()
594 [16061] localedata: Review / update transliteration data
595 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
597 [16141] time: strptime %z offset restriction
598 [16171] math: drem should be alias of remainder
599 [16296] math: fegetround is pure?
600 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
601 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
603 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
604 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
606 [16422] math: [powerpc] math-float, math-double failing llrint tests with
607 "Exception "Inexact" set" on ppc32
608 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
609 [16517] math: Missing underflow exception from tanf/tan/tanl
610 [16519] math: Missing underflow exception from sinhf
611 [16520] math: Missing underflow exception from tanhf
612 [16521] math: Missing underflow exception from exp2
613 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
615 [16734] stdio: fopen calls mmap to allocate its buffer
616 [16961] math: nan function incorrect handling of bad sequences
617 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
618 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
619 [16985] locale: localedef: confusing error message when opening output
621 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
622 [17197] locale: Redundant shift character in iconv conversion output at
624 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
625 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
626 [17250] dynamic-link: static linking breaks nss loading
627 (getaddrinfo/getpwnam/etc...)
628 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
630 [17441] math: isnan() should use __builtin_isnan() in GCC
631 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
632 (related to lock elision)
633 [17787] manual: Exponent on page 324 of the PDF ends prematurely
634 [17886] time: strptime should be able to parse "Z" as a timezone with %z
635 [17887] time: strptime should be able to parse "+01:00" style timezones
636 [17905] libc: catopen() Multiple unbounded stack allocations
638 [18084] libc: backtrace (..., 0) dumps core on x86
639 [18086] libc: nice() sets errno to 0 on success
640 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
641 is too large (CVE-2015-8778)
642 [18251] dynamic-link: SONAME missing when audit modules provides path
643 [18265] libc: add attributes for wchar string and memory functions
644 [18370] math: csqrt missing underflows
645 [18421] libc: [hppa] read-only segment has dynamic relocations
646 [18472] libc: Obsolete syscall wrappers should be compat symbols
647 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
648 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
649 [18525] localedata: Remove locale timezone information
650 [18560] libc: [powerpc] spurious bits/ipc.h definitions
651 [18568] localedata: Update locale data to Unicode 8.0
652 [18589] locale: sort-test.sh fails at random
653 [18595] math: ctan, ctanh missing underflows
654 [18604] libc: assert macro-expands its argument
655 [18610] math: S390: fetestexcept() reports any exception if DXC-code
656 contains a vector instruction exception.
657 [18611] math: j1, jn missing errno setting on underflow
658 [18618] localedata: sync Chechen locale definitions with other *_RU
660 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
661 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
662 [18665] network: In send_dg, the recvfrom function is NOT always using the
663 buffer size of a newly created buffer (CVE-2015-7547)
664 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
666 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
668 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
669 [18699] math: tilegx cproj() for various complex infinities does not yield
671 [18724] libc: Harden put*ent functions against data injection
672 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
673 [18755] build: build errors with -DNDEBUG
674 [18757] stdio: fmemopen fails to set errno on failure
675 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
677 [18781] libc: openat64 lacks O_LARGEFILE
678 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
679 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
680 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
681 [18790] math: [ldbl-128ibm] tanhl inaccurate
682 [18795] libc: stpncpy fortification misses buffer lengths that are
684 [18796] build: build fails for --disable-mathvec
685 [18803] math: hypot missing underflows
686 [18820] stdio: fmemopen may leak memory on failure
687 [18823] math: csqrt spurious underflows
688 [18824] math: fma spurious underflows
689 [18825] math: pow missing underflows
690 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
691 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
692 [18870] build: sem_open.c fails to compile with missing symbol
694 [18872] stdio: Fix memory leak in printf_positional
695 [18873] libc: posix_fallocate overflow check ineffective
696 [18875] math: Excess precision leads incorrect libm
697 [18877] libc: arm: mmap offset regression
698 [18887] libc: memory corruption when using getmntent on blank lines
699 [18918] localedata: hu_HU: change time to HH:MM:SS format
700 [18921] libc: Regression: extraneous stat() and fstat() performed by
702 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
703 binaries (CVE-2015-8777)
704 [18951] math: tgamma missing underflows
705 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
707 [18953] localedata: lt_LT: change currency symbol to the euro
708 [18956] math: powf inaccuracy
709 [18961] math: [i386] exp missing underflows
710 [18966] math: [i386] exp10 missing underflows
711 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
712 [18969] build: multiple string test failures due to missing locale
714 [18970] libc: Reference of pthread_setcancelstate in libc.a
715 [18977] math: float / long double Bessel functions not in XSI POSIX
716 [18980] math: i386 libm functions return with excess range and precision
717 [18981] math: i386 scalb*, ldexp return with excess range and precision
718 [18982] stdio: va_list and vprintf
719 [18985] time: Passing out of range data to strftime() causes a segfault
721 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
722 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
723 [19012] locale: iconv_open leaks memory on error path
724 [19016] math: clog, clog10 inaccuracy
725 [19018] nptl: Mangle function pointers in tls_dtor_list
726 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
727 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
728 [19048] malloc: malloc: arena free list can become cyclic, increasing
730 [19049] math: [powerpc] erfc incorrect zero sign
731 [19050] math: [powerpc] log* incorrect zero sign
732 [19058] math: [x86_64] Link fail with -fopenmp and -flto
733 [19059] math: nexttoward overflow incorrect in non-default rounding modes
734 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
735 [19074] network: Data race in _res_hconf_reorder_addrs
736 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
737 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
738 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
740 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
742 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
743 [19086] manual: posix_fallocate64 documented argument order is wrong.
744 [19088] math: lround, llround missing exceptions close to overflow
746 [19094] math: lrint, llrint missing exceptions close to overflow threshold
747 [19095] math: dbl-64 lrint incorrect for 64-bit long
748 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
749 [19124] dynamic-link: ld.so failed to build with older assmebler
750 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
751 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
752 [19134] math: [powerpc32] lround, lroundf spurious exceptions
753 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
754 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
755 pthread_setaffinity_np
756 [19156] math: [ldbl-128] j0l spurious underflows
757 [19164] nptl: tst-getcpu fails with many possible CPUs
758 [19168] math: math/test-ildoubl and math/test-ldouble failure
759 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
760 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
762 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
763 (FE_NOMASK_ENV) do not clear SSE exceptions
764 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
765 _int_new_arena/reused_arena
766 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
767 [19201] math: dbl-64 remainder incorrect sign of zero result
768 [19205] math: bits/math-finite.h conditions do not match math.h and
770 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
771 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
772 for C99-based standards
773 [19212] libc: features.h not -Wundef clean
774 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
776 [19214] libc: Family and model identification for AMD CPU's are incorrect.
777 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
778 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
780 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
782 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
784 [19242] libc: strtol incorrect in Turkish locales
785 [19243] malloc: reused_arena can pick an arena on the free list, leading
786 to an assertion failure and reference count corruption
787 [19253] time: tzset() ineffective when temporary TZ did not include DST
789 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
790 [19270] math: [hppa] Shared libm missing __isnanl
791 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
792 MAP_HUGETLB and MAP_STACK defines
793 [19313] nptl: Wrong __cpu_mask for x32
794 [19347] libc: grantpt: try to force a specific gid even without pt_chown
795 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
796 [19350] math: [ldbl-128ibm] sinhl spurious overflows
797 [19351] math: [ldbl-128ibm] logl inaccurate near 1
798 [19363] time: x32: times() return value wrongly truncates/sign extends
800 [19367] dynamic-link: Improve branch prediction on Silvermont
801 [19369] network: Default domain name not reset by res_ninit when "search"
802 / "domain" entry is removed from resolv.conf
803 [19375] math: powerpc: incorrect results for POWER7 logb with negative
805 [19385] localedata: bg_BG: time separator should be colon, not comma
806 [19408] libc: linux personality syscall wrapper may erroneously return an
807 error on 32-bit architectures
808 [19415] libc: dladdr returns wrong names on hppa
809 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
810 IBM905, IBM907, and IBM909
811 [19439] math: Unix98 isinf and isnan functions conflict with C++11
812 [19443] build: build failures with -DDEBUG
813 [19451] build: Make check fails on test-double-vlen2
814 [19462] libc: Glibc failed to build with -Os
815 [19465] math: Wrong code with -Os
816 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
818 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
820 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
821 [19486] math: S390: Math tests fail with "Exception Inexact set".
822 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
823 [19550] libc: [mips] mmap negative offset handling inconsistent with other
825 [19590] math: Fail to build shared objects that use libmvec.so functions.
829 * The following bugs are resolved with this release:
831 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
832 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
833 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
834 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
835 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
836 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
837 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
838 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
839 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
840 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
841 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
842 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
843 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
844 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
845 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
846 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
847 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
848 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
849 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
850 18657, 18676, 18694, 18696, 18887.
852 * Cache information can be queried via sysconf() function on s390 e.g. with
853 _SC_LEVEL1_ICACHE_SIZE as argument.
855 * A buffer overflow in gethostbyname_r and related functions performing DNS
856 requests has been fixed. If the NSS functions were called with a
857 misaligned buffer, the buffer length change due to pointer alignment was
858 not taken into account. This could result in application crashes or,
859 potentially arbitrary code execution, using crafted, but syntactically
860 valid DNS responses. (CVE-2015-1781)
862 * The time zone file parser has been made more robust against crafted time
863 zone files, avoiding heap buffer overflows related to the processing of
864 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
865 large time zone data files. Overly long time zone specifiers in the TZ
866 variable no longer result in stack overflows and crashes.
868 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
869 for LD and GD on x86 and x86-64, has been implemented. You will need
870 binutils-2.24 or later to enable this optimization.
872 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
873 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
874 Hat). These updates cause user visible changes, such as the fix for bug
877 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
878 and getXXbyYY NSS calls for the same database, causing a denial-of-service
879 condition in some applications.
881 * Added vector math library named libmvec with the following vectorized x86_64
882 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
884 The library can be disabled with --disable-mathvec. Use of the functions is
885 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
886 Shared library libmvec.so is linked in as needed when using -lm (no need to
887 specify -lmvec explicitly for not static builds).
888 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
890 * A new fmemopen implementation has been added with the goal of POSIX
891 compliance. The new implementation fixes the following long-standing
892 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
893 old implementation is still present for use be by existing binaries.
895 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
896 releases. It has been fixed to match 2.19 and older, but binaries built
897 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
899 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
900 Contributed by Roland McGrath (Google).
902 * The header <regexp.h> is deprecated, and will be removed in a future
903 release. Use of this header will trigger a deprecation warning.
904 Application developers should update their code to use <regex.h> instead.
906 This header was formerly part of SUS, but was deprecated in 1994 and
907 removed from the standard in 2001. Also, the glibc implementation
908 leaks memory. See BZ#18681 for more details.
912 * The following bugs are resolved with this release:
914 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
915 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
916 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
917 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
918 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
919 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
920 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
921 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
922 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
923 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
926 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
927 too little memory for the to-be-scanned arguments and overflow the
928 allocated buffer. The implementation now correctly computes the required
929 buffer size when using malloc, and switches to malloc from alloca as
932 * A new semaphore algorithm has been implemented in generic C code for all
933 machines. Previous custom assembly implementations of semaphore were
934 difficult to reason about or ensure that they were safe. The new version
935 of semaphore supports machines with 64-bit or 32-bit atomic operations.
936 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
937 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
939 * Port to Altera Nios II has been contributed by Mentor Graphics.
941 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
942 implementations for powerpc64/powerpc64le.
943 Implemented by Adhemerval Zanella (IBM).
945 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
946 and powerpc64le. This may improve lock scaling of existing programs on
947 HTM capable systems. The lock elision code is only enabled with
948 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
949 powerpc will issue a transaction abort on every syscall to avoid side
950 effects being visible outside transactions.
952 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
953 AArch64. Contributed by ARM Ltd.
955 * i386 memcpy functions optimized with SSE2 unaligned load/store.
957 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
958 under certain input conditions resulting in the execution of a shell for
959 command substitution when the application did not request it. The
960 implementation now checks WRDE_NOCMD immediately before executing the
961 shell and returns the error WRDE_CMDSUB as expected.
963 * CVE-2012-3406 printf-style functions could run into a stack overflow when
964 processing format strings with a large number of format specifiers.
966 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
967 infinite loop if the DNS response contained a PTR record of an unexpected
970 * The minimum GCC version that can be used to build this version of the GNU
971 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
972 still be used to compile programs using the GNU C Library.
974 * The GNU C Library is now built with -Werror by default. This can be
975 disabled by configuring with --disable-werror.
977 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
979 * The obsolete sigvec function has been removed. This was the original
980 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
981 programs have been using instead for about 25 years. Of course, ABI
982 compatibility for old binaries using sigvec remains intact.
984 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
985 with newer versions of bison.
987 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
988 The original MIPS o32 hard-float ABI requires an FPU where double-precision
989 registers overlay two consecutive single-precision registers. MIPS32R2
990 introduced a new FPU mode (FR=1) where double-precision registers extend the
991 corresponding single-precision registers which is incompatible with the
992 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
993 require the use of FR=1 making a transition necessary. New o32 ABI
994 extensions enable users to migrate over time from the original o32 ABI
995 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
996 tracks the ABI of any loaded object and verifies that new objects are
997 compatible. Mode transitions will also be requested as required and
998 unsupportable objects will be rejected. The ABI checks include both soft and
999 hard float ABIs for o32, n32 and n64.
1001 GCC 5 with GNU binutils 2.25 onwards:
1002 It is strongly recommended that all o32 system libraries are built using the
1003 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1004 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1005 compiler using --with-fp-32=xx to set this by default.
1009 * The following bugs are resolved with this release:
1011 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1012 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1013 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1014 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1015 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1016 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1017 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1018 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1019 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1020 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1021 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1022 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1023 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1024 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1025 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1027 * Reverted change of ABI data structures for s390 and s390x:
1028 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1029 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1030 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1031 are preserved pointing straight to the same implementation as the old ones.
1032 Given that, new callers will simply provide a too-big buffer to these
1033 functions. Any applications/libraries out there that embed jmp_buf or
1034 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1035 against 2.19 headers will have to rebuilt again. This is necessary in any
1036 case to revert the breakage in their ABI caused by the glibc change.
1038 * Support for file description locks is added to systems running the
1039 Linux kernel. The standard file locking interfaces are extended to
1040 operate on file descriptions, not file descriptors, via the use of
1041 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1042 are associated with an open file instead of a process.
1044 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1046 * The minimum Linux kernel version that this version of the GNU C Library
1047 can be used with is 2.6.32.
1049 * Running the testsuite no longer terminates as soon as a test fails.
1050 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1051 with PASS or FAIL lines for individual tests. A summary of the results is
1052 printed, including a list of failing lists, and "make check" exits with
1053 error status if there were any unexpected failures. "make check
1054 stop-on-test-failure=y" may be used to keep the old behavior.
1056 * The am33 port, which had not worked for several years, has been removed
1059 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1060 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1061 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1062 conflicted with POSIX (support for which was removed in 2.19), the
1063 interfaces those macros enabled remain available when compiling with
1064 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1065 test macros defined.
1067 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1069 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1070 This may improve lock scaling of existing programs on TX capable systems.
1071 The lock elision code is only built with --enable-lock-elision=yes and
1072 then requires a GCC version supporting the TX builtins. With lock elision
1073 default mutexes are elided via __builtin_tbegin, if the cpu supports
1074 transactions. By default lock elision is not enabled and the elision code
1077 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1078 copy the path argument. This allowed programs to cause posix_spawn to
1079 deference a dangling pointer, or use an unexpected pathname argument if
1080 the string was modified after the posix_spawn_file_actions_addopen
1083 * All supported architectures now use the main glibc sysdeps directory
1084 instead of some being in a separate "ports" directory (which was
1085 distributed separately before glibc 2.17).
1087 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1088 On configurations that support it (all Linux configurations), it's now
1089 used regardless of the --enable-add-ons switch to configure. It is no
1090 longer possible to build such configurations without pthreads support.
1092 * Locale names, including those obtained from environment variables (LANG
1093 and the LC_* variables), are more tightly checked for proper syntax.
1094 setlocale will now fail (with EINVAL) for locale names that are overly
1095 long, contain slashes without starting with a slash, or contain ".." path
1096 components. (CVE-2014-0475) Previously, some valid locale names were
1097 silently replaced with the "C" locale when running in AT_SECURE mode
1098 (e.g., in a SUID program). This is no longer necessary because of the
1101 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1102 with application code using Intel MPX instructions. (With all previous
1103 versions, the MPX register state could be clobbered when making calls
1104 into or out of a shared library.) Note that while the new dynamic
1105 linker is compatible with all known x86 hardware whether or not it
1106 supports Intel MPX, some x86 instruction-set emulators might fail to
1107 handle the new instruction encodings. This is known to affect Valgrind
1108 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1109 release), and might affect other tools that do instruction emulation.
1111 * Support for loadable gconv transliteration modules has been removed.
1112 The support for transliteration modules has been non-functional for
1113 over a decade, and the removal is prompted by security defects. The
1114 normal gconv conversion modules are still supported. Transliteration
1115 with //TRANSLIT is still possible, and the //IGNORE specifier
1116 continues to be supported. (CVE-2014-5119)
1118 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1119 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1120 resulting a denial-of-service security vulnerability in applications which
1121 use functions related to iconv. (CVE-2014-6040)
1125 * The following bugs are resolved with this release:
1127 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1128 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1129 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1130 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1131 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1132 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1133 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1134 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1135 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1136 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1137 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1138 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1139 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1140 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1141 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1142 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1143 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1144 16431, 16453, 16474, 16506, 16510, 16529
1146 * Slovenian translations for glibc messages have been contributed by the
1147 Translation Project's Slovenian team of translators.
1149 * The public headers no longer use __unused nor __block. This change is to
1150 support compiling programs that are derived from BSD sources and use
1151 __unused internally, and to support compiling with Clang's -fblock
1152 extension which uses __block.
1154 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1155 large collation sequences to optimize multiple passes. This cache
1156 computation may overflow for large collation sequences and may cause a
1157 stack or buffer overflow. This is now fixed to use a slower algorithm
1158 which does not use a cache if there is an integer overflow.
1160 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1161 rules for large collation sequences to optimize multiple passes and falls
1162 back to alloca if malloc fails, resulting in a possible stack overflow.
1163 The implementation now falls back to an uncached collation sequence lookup
1166 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1167 initialized for static applications resulting in the security feature
1168 being disabled. The pointer guard is now correctly initialized to a
1169 random value for static applications. Existing static applications need
1170 to be recompiled to take advantage of the fix (bug 15754).
1172 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1173 to the d_name member of struct dirent, or omit the terminating NUL
1174 character. (Bugzilla #14699).
1176 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1177 aligned_alloc functions could allocate too few bytes or corrupt the
1178 heap when passed very large allocation size values (Bugzilla #15855,
1181 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1182 for AF_INET6 has been fixed (Bugzilla #16072).
1184 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1185 pap_CW, quz_PE, the_NP.
1187 * Substantially revised locales: gd_GB, ht_HT
1189 * The LC_ADDRESS field was updated to support country_car for almost all
1192 * ISO 1427 definitions were updated.
1194 * ISO 3166 definitions were updated.
1196 * The localedef utility now supports --big-endian and --little-endian
1197 command-line options to generate locales for a different system from that
1198 for which the C library was built.
1200 * Binary locale files now only depend on the endianness of the system for
1201 which they are generated and not on other properties of that system. As a
1202 consequence, binary files generated with new localedef may be incompatible
1203 with old versions of the GNU C Library, and binary files generated with
1204 old localedef may be incompatible with this version of the GNU C Library,
1205 in the following circumstances:
1207 + Locale files may be incompatible on m68k systems.
1209 + Locale archive files (but not separate files for individual locales) may
1210 be incompatible on systems where plain "char" is signed.
1212 * The configure option --disable-versioning has been removed. Builds with
1213 --disable-versioning had not worked for several years.
1215 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1216 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1218 * SystemTap probes for malloc have been introduced.
1220 * SystemTap probes for slow multiple precision fallback paths of
1221 transcendental functions have been introduced.
1223 * Support for powerpc64le has been added.
1225 * The soft-float powerpc port now supports e500 processors.
1227 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1229 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1230 set of header declarations that are enabled by default, even when other
1231 feature test macros or compiler options such as -std=c99 would otherwise
1232 disable some of those declarations.
1234 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1235 conflict with POSIX. The libbsd-compat library (which was a dummy library
1236 that did nothing) has also been removed.
1238 * Preliminary documentation about Multi-Thread, Async-Signal and
1239 Async-Cancel Safety has been added.
1241 * Change of ABI data structures for s390 and s390x:
1242 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1243 allow for future hardware extensions. All library functions that accept or
1244 return these structures were versioned in order to provide backwards
1245 compatibility with existing code. However, not all cases can be handled
1246 with interface versioning. If an application embeds either structure into
1247 another structure and that structure is passed to another compilation unit
1248 which expects the newer larger-sized object then any access to the new
1249 fields will result in undefined behaviour. Similarly any access to fields
1250 that were moved by the enlarging of the structures will cause undefined
1251 behaviour. To fix the undefined behaviour all objects that are part of the
1252 final in-memory image must agree on the size of structures, and this may
1253 require recompilation.
1257 * The following bugs are resolved with this release:
1259 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1260 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1261 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1262 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1263 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1264 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1265 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1266 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1267 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1268 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1269 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1270 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1271 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1274 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1275 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1276 Distributions can re-enable building and using pt_chown via the new configure
1277 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1278 considerable security risks and should only be used if the distribution
1279 understands and accepts the risks.
1281 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1284 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1285 fixed (Bugzilla #15330).
1287 * Add support for calling C++11 thread_local object destructors on thread
1288 and program exit. This needs compiler support for offloading C++11
1289 destructor calls to glibc.
1291 * Improved worst case performance of libm functions with double inputs and
1294 * Support for priority inherited mutexes in pthread condition variables on
1295 non-x86 architectures.
1297 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1299 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1301 * Optimized string functions for ARM. Implemented by Will Newton and
1304 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1306 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1309 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1310 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1312 * Added a benchmark framework to track performance of functions in glibc.
1314 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1315 It is based on draft TS 18661 and currently enabled as a GNU extension.
1317 * On Linux, the clock function now uses the clock_gettime system call
1318 for improved precision, rather than old times system call.
1320 * Added support for version-3 tz format files. This is needed when using
1321 the tz database release 2013e or later, and affects a few unusual cases --
1322 currently only TZ='America/Godthab' for time stamps after 2037.
1324 * Added new API functions pthread_getattr_default_np and
1325 pthread_setattr_default_np to get and set the default pthread
1326 attributes of a process.
1328 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1329 This may improve lock scaling of existing programs on TSX capable systems.
1330 When the --enable-lock-elision=yes parameter is specified at configure
1331 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1334 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1335 availability). Implemented by Ryan S. Arnold.
1337 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1341 * The following bugs are resolved with this release:
1343 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1344 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1345 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1346 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1347 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1348 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1349 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1350 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1351 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1352 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1353 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1354 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1355 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1356 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1357 14879, 14889, 14893, 14898, 14914.
1359 * Optimization of memcpy for MIPS.
1361 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1362 EMFILE has been fixed (Bugzilla #14889).
1364 * The add-on ports collection is now distributed in the "ports" subdirectory
1365 of the main GNU C Library distribution, rather than separately.
1367 * Port to ARM AArch64 contributed by Linaro.
1369 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1370 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1372 Implemented by Andreas Krebbel.
1374 * The new function secure_getenv allows secure access to the environment,
1375 returning NULL if running in a SUID/SGID process. This function replaces
1376 the internal function __secure_getenv.
1378 * SystemTap static probes have been added into the dynamic linker.
1379 Implemented by Gary Benson.
1381 * Optimizations of string functions strstr, strcasestr and memmem.
1382 Implemented by Maxim Kuvyrkov.
1384 * The minimum Linux kernel version that this version of the GNU C Library
1385 can be used with is 2.6.16.
1387 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1388 powerpc POWER7. Implemented by Will Schmidt.
1390 * New configure option --disable-nscd builds the C library such that it
1391 never attempts to contact the Name Service Caching Daemon (nscd).
1392 New configure option --disable-build-nscd avoids building nscd itself;
1393 this is the default if --disable-nscd is used.
1395 * Improved support for cross-compilation, including cross-testing and
1396 bootstrap builds without a previously built glibc.
1398 * Several testsuite tests are now able to test multiple IFUNC variants of an
1399 interface, rather than just testing the one that would be chooen by
1402 * New configure options --with-bugurl and --with-pkgversion, for
1403 distributors to use to embed their bug-reporting and package version
1404 information in --help and --version output.
1406 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1407 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1408 allows creation of chroots without the procfs mounted on /proc.
1410 * The `crypt' function now fails if passed salt bytes that violate the
1411 specification for those values. On Linux, the `crypt' function will
1412 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1413 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1414 when the mode is enabled.
1416 * The `clock_*' suite of functions (declared in <time.h>) is now available
1417 directly in the main C library. Previously it was necessary to link with
1418 -lrt to use these functions. This change has the effect that a
1419 single-threaded program that uses a function such as `clock_gettime' (and
1420 is not linked with -lrt) will no longer implicitly load the pthreads
1421 library at runtime and so will not suffer the overheads associated with
1422 multi-thread support in other code such as the C++ runtime library.
1424 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1429 * The following bugs are resolved with this release:
1431 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1432 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1433 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1434 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1435 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1436 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1437 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1438 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1439 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1440 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1441 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1442 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1443 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1444 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1445 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1446 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1447 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1448 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1449 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1450 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1451 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1452 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1455 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1456 configuring glibc with:
1457 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1458 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1459 Implemented by H.J. Lu.
1463 + define static_assert
1465 + do not declare gets
1467 + declare at_quick_exit and quick_exit also for ISO C11
1469 + aligned_alloc. NB: The code is deliberately allows the size parameter
1470 to not be a multiple of the alignment. This is a moronic requirement
1471 in the standard but it is only a requirement on the caller, not the
1474 + timespec_get added
1476 + uchar.h support added
1478 + CMPLX, CMPLXF, CMPLXL added
1480 Implemented by Ulrich Drepper.
1482 * Support for the IA-64 has been moved to ports.
1484 * Remove support for anything but ELF binary format
1486 * Checking versions of poll, ppoll added.
1487 Implemented by Ulrich Drepper.
1489 * More generic and 64-bit performance optimizations to math functions.
1490 Implemented by Ulrich Drepper.
1492 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1493 headers and functions available at compile time as they were before
1494 version 2.14. This option will be removed at some time in the future
1495 after the TI-RPC library becomes fully sufficient for the needs of
1496 existing applications.
1498 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1499 Note that glibc is not expected to work with any Linux kernel version
1502 * New header <sys/auxv.h> and function getauxval allowing easy access to
1503 the AT_* key-value pairs passed from the Linux kernel. The header also
1504 defines the HWCAP_* bits associated with the AT_HWCAP key.
1506 * New locales: mag_IN
1508 * New configure option --enable-systemtap builds SystemTap static probes
1509 into libc for setjmp and longjmp and into libpthread for various operations.
1510 So far the setjmp/longjmp probes and some of the libpthread probes are
1511 provided only for i*86 and x86_64.
1512 Implemented by Roland McGrath and Rayson Ho.
1514 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1516 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1519 * More optimized functions for SPARC. Implemented by David S. Miller.
1521 * Improved support for cross-compilation, especially bootstrap builds
1522 without a previously built glibc.
1524 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1525 Chris Metcalf from Tilera.
1527 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1528 now supported for ARM processors.
1530 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1531 as the name of the dynamic linker, to distinguish it from the
1532 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1534 * Support for CRIS has been removed from ports.
1536 * A new class of installed header has been documented for low-level
1537 platform-specific functionality. PowerPC added the first instance with a
1538 function to provide time base register access. Contributed by Tulio
1539 Magno Quites Machado Filho.
1541 * ix86 configurations now install header files that are consistent with
1542 what x86-64 configurations install. These same header files can be used
1543 for -m32, -m64, or -mx32 builds.
1544 Contributed by H.J. Lu.
1546 * Math library bug fixes. A thorough audit of all open math library bugs was
1547 conducted by Joseph Myers. Significant progress was made on many math
1548 library bugs resulting in more accurate exceptions and function results.
1549 Many thanks to all those that contributed including Andreas Jaeger for his
1550 patch review and work on the x87 trigonometric instruction issues.
1552 * Timezone data is no longer installed. Timezone-related binaries and scripts
1553 will continue to be installed. Users should obtain their timezone data from
1554 their distribution provider or from the tzdata package at
1555 <ftp://munnari.oz.au/pub/>.
1559 * The following bugs are resolved with this release:
1561 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1562 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1563 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1564 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1565 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1566 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1567 13484, 13506, 13515, 13523, 13524, 13538, 13540
1569 * New program pldd to list loaded object of a process
1570 Implemented by Ulrich Drepper.
1572 * Add nss_db support back to glibc. No more dependency on Berkeley db
1573 and support for initgroups lookups.
1574 Implemented by Ulrich Drepper.
1576 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1577 Contributed by HJ Lu.
1579 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1580 Contributed by HJ Lu.
1582 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1583 on x86-32 and x86-64.
1584 Contributed by Liubov Dmitrieva.
1586 * Optimized strchr and strrchr for SSE on x86-32.
1587 Contributed by Liubov Dmitrieva.
1589 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1590 for x86-64 and x86-32.
1591 Contributed by Liubov Dmitrieva.
1593 * New interfaces: scandirat, scandirat64
1594 Implemented by Ulrich Drepper.
1596 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1597 Implemented by Ulrich Drepper.
1599 * nscd now also caches the netgroup database.
1600 Implemented by Ulrich Drepper.
1602 * Integrate libm with gcc's -ffinite-math-only option.
1603 Implemented by Ulrich Drepper.
1605 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1606 to math functions. Implemented by Ulrich Drepper.
1608 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1609 Implemented by Ulrich Drepper.
1611 * New Linux interfaces: process_vm_readv, process_vm_writev
1613 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1614 Implemented by Ulrich Drepper.
1616 * Optimized nearbyint and strcasecmp for PPC.
1617 Implemented by Adhemerval Zanella.
1619 * New locales: bho_IN, unm_US, es_CU, ta_LK
1623 * The following bugs are resolved with this release:
1625 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1626 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1627 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1628 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1629 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1630 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1631 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1632 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1633 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1634 12795, 12811, 12813, 12814, 12841
1636 * The RPC implementation in libc is obsoleted. Old programs keep working
1637 but new programs cannot be linked with the routines in libc anymore.
1638 Programs in need of RPC functionality must be linked against TI-RPC.
1639 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1641 Visible changes of this change include (obviously) the inability to link
1642 programs using RPC functions without referencing the TI-RPC library and the
1643 removal of the RPC headers from the glibc headers.
1644 Implemented by Ulrich Drepper.
1646 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1647 syncfs, setns, sendmmsg
1649 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1650 yue_HK, lij_IT, mhr_RU
1652 * New encodings: CP770, CP771, CP772, CP773, CP774
1654 * New program sotruss to trace calls through PLTs
1655 Implemented by Ulrich Drepper.
1657 * The malloc hook implementation is marked deprecated and will be removed
1658 from the default implementation in the next version. The design never
1659 worked ever since the introduction of threads. Even programs which do
1660 not create threads themselves can use multiple threads created internally.
1664 * The following bugs are resolved with this release:
1666 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1667 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1668 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1669 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1672 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1674 * POWER7 optimizations: memset, memcmp, strncmp
1676 * New optimized string functions for x86-64: strnlen (SSE2),
1677 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1678 Implemented by Ulrich Drepper.
1682 * The following bugs are resolved with this release:
1684 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1685 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1686 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1687 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1688 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1689 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1690 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1691 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1693 * New interfaces: pthread_getname_np, pthread_setname_np
1695 * New Linux interface: recvmmsg
1697 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1699 * The dynamic linker now recognizes supported ABI versions from the
1700 EI_ABIVERSION field in the ELF header.
1701 Implemented by Ulrich Drepper.
1703 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1704 will not be used to fill in password fields in the passwd.byname replies.
1705 Instead it is used to synthesize the shadow.byname table, should it be
1706 missing. This is a useful mode in some installations involving Solaris.
1707 Implemented by Ulrich Drepper.
1709 * New locales: kok_IN, sq_MK, cv_RU
1713 * The following bugs are resolved with this release:
1715 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1716 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1717 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1718 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1719 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1720 10789, 10817, 10823, 10840, 10847
1722 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1724 Implemented by Ulrich Drepper.
1726 * Checking version of longjmp added that fails if an uninitialized stack
1727 frame would be created. Implemented by Ulrich Drepper.
1729 * STT_GNU_IFUNC is now supported in static executables.
1730 Implemented by H.J. Lu.
1732 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1734 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1735 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1736 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1737 Contributed by H.J. Lu.
1739 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1740 Implemented by Ulrich Drepper.
1742 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1744 Contributed by H.J. Lu.
1746 * Support for fma instruction in AVX on x86-64.
1747 Implemented by H.J. Lu and Ulrich Drepper.
1749 * AVX support in x86-64 auditing support in ld.so.
1750 Implemented by H.J. Lu.
1752 * STB_GNU_UNIQUE support added.
1753 Implemented by Ulrich Drepper.
1755 * Implement second fallback mode for optimized DNS lookup for even more
1756 broken environments. If two requests from the same port are not
1757 handled correctly close the socket and open a new one before sending
1758 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1759 can be used to select this mode right away, instead of rediscovering the
1760 necessity is every process again.
1761 Implemented by Ulrich Drepper.
1763 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1764 Implemented by Adam Tkac.
1766 * Optimized iconv conversions for S390x.
1767 Implemented by Andreas Krebbel.
1769 * Using condvars with PI mutexes is now more efficient due to kernel
1770 support for requeueing to PI futexes. NPTL support added for x86-64.
1771 Implemented by Ulrich Drepper.
1773 * New locales: ps_AF, my_MM
1777 * The following bugs are resolved with this release:
1779 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1780 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1781 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1782 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1783 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1785 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1786 Implemented by Ulrich Drepper.
1788 * New Linux interfaces: accept4, fallocate, fallocate64.
1789 Implemented by Ulrich Drepper.
1791 * Correct declarations of string function when used in C++ code. This
1792 could lead to compile errors for invalid C++ code.
1794 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1797 * New POSIX 2008 interface: psiginfo
1798 Implemented by Ulrich Drepper.
1800 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1801 Implemented by Ulrich Drepper.
1803 * Support for selecting between multiple function definitions at runtime
1804 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1806 * The libcrypt library can now use the hash function implementations in
1807 NSS. Implemented by Ulrich Drepper.
1809 * The malloc implementation can be compiled to be less memory efficient
1810 but higher performing in multi-threaded programs.
1811 Implemented by Ulrich Drepper.
1813 * New locales: nan_TW@latin, ks_IN
1815 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1816 Implemented by Ulrich Drepper.
1818 * Extended printf hook support. It is possible to use user-defined types
1819 and extend existing format specifiers.
1820 Implemented by Ulrich Drepper.
1822 * Handling for group shadow files has been added.
1823 Implemented by Ulrich Drepper.
1825 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1826 servers (the case, e.g., for some people using the built-in DNS
1827 server in ADSL modems/routers). There is a once-per-process timeout
1828 in case of a broken server. To avoid it, users can run nscd or put
1829 'options single-request' in /etc/resolv.conf.
1830 Implemented by Ulrich Drepper.
1834 * The following bugs are resolved with this release:
1836 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1837 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1838 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1839 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1840 6968, 6974, 6980, 6995, 7008, 7009, 7029
1842 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1843 up at the same time. Implemented by Ulrich Drepper.
1845 * TLS descriptors for LD and GD on x86 and x86-64.
1846 Implemented by Alexandre Oliva.
1848 * getaddrinfo now handles DCCP and UDPlite.
1849 Implemented by Ulrich Drepper.
1851 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1852 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1853 Implemented by Ulrich Drepper.
1855 * New implementation of memmem, strstr, and strcasestr which is O(n).
1856 Implemented by Eric Blake.
1858 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1860 * Implement "e" option for popen to open file descriptor with the
1861 close-on-exec flag set. Implemented by Ulrich Drepper.
1863 * Implement "b" mode for fmemopen. In this mode writes writes don't
1864 implicitly add a NUL byte and seeks from the end of the buffer really
1865 use the buffer end, not the string length as the basis.
1866 Implemented by Ulrich Drepper.
1868 * Many functions, exported and internal, now atomically set the close-on-exec
1869 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1871 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1873 Implemented by Pravin Satpute.
1875 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1879 * The following bugs are resolved with this release:
1881 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1882 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1883 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1884 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1885 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1886 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1887 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1889 * New locales: bo_CN, bo_IN, shs_CA.
1891 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1893 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1894 Implemented by Pravin Satpute.
1896 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1898 * nscd caches now all timeouts for DNS entries
1899 Implemented by Ulrich Drepper.
1901 * nscd is more efficient and wakes up less often.
1902 Implemented by Ulrich Drepper.
1904 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1905 vdprintf, and obstack_vprintf.
1906 Implemented by Jakub Jelinek.
1908 * Faster memset for x86-64.
1909 Implemented by Harsha Jagasia and H.J. Lu.
1911 * Faster memcpy on x86.
1912 Implemented by Ulrich Drepper.
1914 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1915 Implemented by Ulrich Drepper.
1917 * Faster sqrt and sqrtf implemention for some PPC variants.
1918 Implemented by Steven Munroe.
1922 * The following bugs are resolved with this release:
1924 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1925 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1926 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1927 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1928 5071, 5103, 5104, 5112, 5113, 5184, 5186
1930 * More checking functions: fread, fread_unlocked, open*, mq_open.
1931 Implemented by Jakub Jelinek and Ulrich Drepper.
1933 * Extend fortification to C++. Implemented by Jakub Jelinek.
1935 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1936 by not recognizing 'a' as a modifier when those specs are requested.
1937 Implemented by Jakub Jelinek.
1939 * PPC optimizations to math and string functions.
1940 Implemented by Steven Munroe.
1942 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1943 options to be passed. Implemented by Ulrich Drepper.
1945 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1947 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1948 Implemented by Ulrich Drepper.
1950 * Handle private futexes in the NPTL implementation.
1951 Implemented by Jakub Jelinek and Ulrich Drepper.
1953 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1954 Implemented by Roland McGrath and Ulrich Drepper.
1956 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1958 * SHA-256 and SHA-512 based password encryption.
1959 Implemented by Ulrich Drepper.
1961 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1962 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1965 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1966 Implemented by Ulrich Drepper.
1970 * The following bugs are resolved with this release:
1972 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
1973 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
1974 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
1975 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
1976 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
1977 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
1978 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
1981 * New Linux interfaces: epoll_pwait, sched_getcpu.
1983 * New generic interfaces: strerror_l.
1985 * nscd can now cache the services database. Implemented by Ulrich Drepper.
1989 * The following bugs are resolved with this release:
1991 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
1992 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
1993 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
1994 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
1995 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
1996 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
1997 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
1998 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
1999 3225, 3251, 3252, 3253, 3273
2001 * For Linux, the sorting of addresses returned by getaddrinfo now also
2002 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2003 Implemented by Ulrich Drepper.
2005 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2006 Implemented by Ulrich Drepper.
2008 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2010 * New iconv module for MIK. Contributed by Alexander Shopov.
2012 * For sites with broken group and/or passwd database, the auto-propagate
2013 option of nscd can prevent creating ID lookup entries from the results
2014 of a name lookup and vice versa. This usually is no problem but some
2015 site might have problems with the default behavior.
2016 Implemented by Ulrich Drepper.
2018 * Iterating over entire database in NIS can be slow. With the
2019 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2020 to trade time for memory. The entire database will be read at once.
2021 Implemented by Ulrich Drepper.
2023 * The interfaces introduced in RFC 3542 have been implemented by
2026 * Support for the new ELF hash table format was added by Ulrich Drepper.
2028 * Support for priority inheritance mutexes added by Jakub Jelinek and
2031 * Support for priority protected mutexes added by Jakub Jelinek.
2033 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2034 or_IN, csb_PL, fy_NL, sr_ME.
2038 * More overflow detection functions.
2040 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2041 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2043 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2044 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2045 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2046 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2047 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2048 by Masahide Washizawa.
2050 * It is now possible to install audit libraries for the dynamic linker using
2051 LD_AUDIT. Implemented by Ulrich Drepper.
2053 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2054 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2055 For a libc and libpthread that works well on Linux 2.4 kernels, we
2056 recommend using the stable 2.3 branch.
2058 * The new function open_wmemstream acts like open_memstream,
2059 but uses a wchar_t wide character buffer.
2061 * The new function ppoll is an improved version of poll, similar to pselect.
2063 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2064 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2065 renameat, unlinkat, linkat, symlinkat, readlinkat.
2067 * New Linux kernel system calls: unshare,
2068 inotify_init, inotify_add_watch, inotify_rm_watch.
2070 * The euidaccess function is now also known by the alias eaccess,
2071 for compatibility with some other systems.
2073 * Timezone data updated to 2006b version.
2077 * The following bugs are resolved with this release:
2079 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2080 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2081 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2082 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2083 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2084 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2086 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2088 * As of this release, GCC 4 can be used to compile the C Library.
2090 * Timezone data updated to 2005m version.
2094 * The following bugs are resolved with this release:
2096 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2097 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2098 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2099 777, 787, 821, 822, 823, 825
2101 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2105 * Support for RFC 3678. Real implementations exist only for Linux so far.
2106 Implemented by Ulrich Drepper.
2108 * nscd can now cache entries persistently. Expiring entries are reloaded.
2109 For speedups the cache can be shared in memory with client processes.
2110 Implemented by Ulrich Drepper.
2112 * nscd can now perform SELinux checks.
2113 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2115 * getaddrinfo queries are now cached. Canonical name lookup is performed
2117 Implemented by Ulrich Drepper.
2119 * The nothrow function attribute is used when headers are used by gcc when
2120 compiling C code. This can avoid significant amounts of exception
2123 * The malloc functions perform more error checking and are stricter when
2124 it comes to reacting on errors. The default action is to terminate
2125 the process after showing an error message. Implemented by Ulrich Drepper.
2127 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2128 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2130 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2131 namespaces using the new function dlmopen(). This feature is of course,
2132 like most other dynamic loading functionality, not available in statically
2133 linked applications. Implemented by Ulrich Drepper.
2135 * Low-overhead boundary checking variants of string and some stdio functions
2136 were added. These are to be used in conjunction with a gcc patch by
2137 Jakub Jelinek which adds calls to these functions if possible.
2138 Implemented by Jakub Jelinek and Ulrich Drepper.
2140 * Old code for several operating systems and machine architectures that
2141 have not been in working condition in a long time have been removed from
2142 the main source tree maintained by the GNU C Library's maintainers.
2143 These files are now reside in the separate `ports' source module
2144 that is usable as an add-on when building the library.
2148 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2149 interrogate the dynamic linker, compatible with the Solaris interface.
2151 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2152 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2154 * getifaddrs now uses the netlink interface on Linux to get its information.
2155 Implemented by Thorsten Kukuk.
2157 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2158 Implemented by Ulrich Drepper.
2160 * support for non-executable stacks on x86 has been added. Changes mostly
2163 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2166 * getaddrinfo now performs destination address selection according to
2171 * Thread-safe interfaces for many functions that access locale data
2172 were added in version 2.3, but these features were omitted from NEWS.
2173 Many functions have variants with an `_l' suffix that take a `locale_t'
2174 object as a parameter rather than consulting the current locale.
2175 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2176 create and maintain `locale_t' objects. Additionally, the new function
2177 `uselocale' sets "the current locale" (as used by functions not so
2178 parameterized) set for an individual thread. These features were added
2179 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2181 * The functions getresuid, getresgid, setresuid, and setresgid, which
2182 have long been available on Linux, are now declared in <unistd.h>
2183 and are now also available on the Hurd.
2185 * ELF thread-local storage support (TLS) now works on x86-64.
2187 * The new dynamic string token $LIB is expanded in shared library names.
2188 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2190 * Aldy Hernandez contributed complete software floating point support for
2191 PowerPC machines with no FPU.
2193 * fexecve is implemented on Linux.
2195 * The `btowc' function should work at least twice as fast due to
2196 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2198 * With approriate thread add-ons cancelable functions are now implemented
2199 in libc.so as well. No need to call the function in libpthread. This
2200 change allowed to finally disable the incorrect and expensive handling
2201 of weak definition in ld.so.
2203 * Yet more PLT entries in libc.so have been removed. We finally arrived
2204 at the bare minimum. Startup times improved appropriately.
2206 * Support for the new Linux/x86 system call interface was added. The
2207 AT_SYSINFO auxiliary vector entry is recognized and handled.
2211 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2214 * iconv (the program and the interface) now accepts empty names (excluding
2215 options like //TRANSLIT) to mean "use charset of current locale".
2217 * localedef can now transliterate characters in strings which are not in
2218 the provided charmap. The information from the input locale is used.
2220 * Prelinking support was added for ELF targets. This requires additional
2221 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2223 * Read-only stdio streams now use mmap to speed up operation by eliminating
2224 copying and buffer underflows. To use add 'm' to the mode string of
2225 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2227 * The malloc functions were completely rewritten by Wolfram Gloger based
2228 on Doug Lea's malloc-2.7.0.c.
2230 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2231 implementation of regex.
2233 * Bruno Haible upgraded the iconv and locale implementation to support
2236 * Contents of the LC_* and LANG environment variables in the CEN style are
2237 not recognized anymore. It never was used. Change by Ulrich Drepper.
2239 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2240 thread-local storage (TLS) ABI on some platforms.
2241 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2243 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2244 EUC-JISX0213, and TSCII.
2246 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2247 BSD-compatible interface for getting all network interface addresses.
2248 Implementation for IPv4 by Roland McGrath.
2250 * Loading of locale data is faster due to the introduction of a locale
2251 archive. Implemented by Roland McGrath and Ulrich Drepper.
2253 * Startup times are significantly reduced by not using exported functions
2254 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2257 * Steven Munroe contributed a port to PowerPC64/Linux.
2261 * The Hurd now uses the GNU libio implementation of stdio rather than the
2262 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2264 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2265 with the same behavior as the Linux system call of the same name.
2269 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2270 128-bit long double format.
2272 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2273 IBM1161, and IBM1162 charsets.
2275 * Andreas Jaeger contributed a port to x86-64/Linux.
2277 * Peter Bruin contributed a port to PowerPC/Hurd.
2279 * libc functions using I/O streams now can handle wide-oriented streams
2282 * optimizations in the dynamic linker. Binaries created by recent binutils
2283 versions start up quicker due to reduced time spent on relocations.
2285 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2289 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2290 asinh, atanh, j0 for the 128-bit long double format.
2292 * Bruno Haible updated all the code handling Unicode in some form to
2293 support Unicode 3.1.
2295 * Speed of regex for single-byte locales is back to previous levels.
2296 Patch by Isamu Hasegawa.
2298 * Alpha, SPARC, and IA-64 now also using floating stacks.
2300 * Startup time of internationalized applications greatly improved through
2301 iconv cache. Use iconvconfig to generate the cache file.
2302 Contributed by Ulrich Drepper.
2304 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2305 severe performance problems. Patch by David Mosberger.
2307 * The Hurd port got a lot more functionality like AIO, various stdio
2308 extensions, etc. Mainly done by Roland McGrath.
2310 * mtrace can now lookup symbols in shared libraries.
2314 * Intel's IA-64 math library is largely integrated. It provides fast and
2315 accurate implementatations for most basic and standard math functions
2316 in float, double, and long double format.
2318 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2319 and asin for the 96-bit long double format and asin, log, tan for the
2320 128-bit long double format.
2322 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2323 The basic double functions exist today. Contributed by Abraham Ziv
2324 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2325 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2327 * An asynchronous name lookup library was added. The interface is designed
2328 after POSIX AIO. The proposal was circulated beforehand to get comments.
2329 No negative ones came in. Implemented by Ulrich Drepper.
2331 * Port to S390/64bit contributed by Martin Schwidefsky
2332 <schwidefsky@de.ibm.com>.
2334 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2335 of functions for Linux/IA-64.
2337 * The RPC code is now thread safe. Threads can now use the same service
2338 of different services at the same time. Patch by Eric Norum
2339 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2341 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2342 family of functions for Linux/S390.
2344 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2345 of functions for Linux/x86.
2347 * Port to Linux/CRIS contributed by Axis Communications.
2351 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2352 we can now check for namespace violations and missing declarations. The
2353 result is that almost all headers are now Unix-compliant (as defined in
2354 the upcoming XPG6). The negative side is that some programs might need
2355 corrections, too, if they depend on the incorrect form of the headers in
2356 previous versions which defined too many symbols and included too many
2359 * regex now handles multibyte character sets correctly.
2360 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2362 * iconv (the program) does now conform to the upcoming XPG6 and handles
2363 charmaps. Instead of the charset names the path of charmaps can be
2364 provided and the conversion happens based on this data.
2365 Contributed by Ulrich Drepper.
2367 * The locale program now provides detailed information about the installed
2372 only lists the names of the supported locales
2376 provides details such as country, language, and codeset name.
2377 Contributed by Ulrich Drepper.
2381 * The gencat program now parses the input file according to the charset
2382 selected by the LC_CTYPE category. This is important for stateful
2383 character sets. To make generating catalogs easier there is a way
2384 to overwrite the charset selected by the locale: before the first
2385 message or $ quote line the catalog can contain a line like
2387 $ codeset=ISO-8859-2
2389 to select the charset (ISO-8859-2 in this case).
2391 Implemented by Shinya Hanataka and Ulrich Drepper.
2393 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2394 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2395 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2397 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2398 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2400 * The LANGUAGE environment variable is now ignored unless the locale is
2401 changed from the default "C" locale.
2403 * The usual bug fixes.
2407 * Greg McGary added runtime support for bounds checking using gcc's
2408 new -fbounded-pointers option. ix86 target is complete. PowerPC
2411 * Thorsten Kukuk added secure mode support to nscd.
2413 * The Berkeley db libraries have been removed.
2415 Related, the nss_db module is now in a separate package since it
2416 obviously requires a database library being available.
2418 * Wide character I/O streams implemented by Ulrich Drepper.
2420 * Functions from the extended socket API added by Ulrich Drepper.
2422 * Functions feenableexcept and fedisableexcept to control the
2423 behaviour of individual exceptions have been added by Andreas Jaeger.
2425 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2427 * The resolver code has been updated from bind 8.2.3-T5B which supports
2428 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2431 This change could in some situations effect backward compatibility. Since
2432 now `_res' is a thread-local instead of a global variable, modifying it
2433 in one thread does not have any effect in other threads.
2435 The resolver library was also extended to allow IPv6 as the transport
2436 protocol for the requests. This work was done by Stig Venaas.
2438 * Compatibility code for K&R C compilers has been removed from the
2439 header files. A ISO C compiler is needed to use the library
2440 (conforming to either C89 or C99 standard).
2442 * Complete rewrite of the localedef program to support multibyte character
2443 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2444 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2445 Implemented by Ulrich Drepper.
2447 Bruno Haible significantly improved the generation and use of the data
2448 structures for the wide character tables.
2450 * Plural handling in gettext implemented by Ulrich Drepper.
2452 * The utmp daemon has been removed.
2454 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2456 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2459 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2461 * POSIX barriers implemented by Kaz Kylheku.
2463 * POSIX spawn function family implemented by Ulrich Drepper.
2465 * POSIX spinlocks are now available.
2467 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2469 * the configure option --enable-kernel=X.Y.Z allows to strip out
2470 compatibility for kernel versions before X.Y.Z. This is currently only
2471 implemented for Linux.
2473 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2474 field sin6_scope_id. This means that all programs using IPv6 should be
2475 recompiled. Don't expect binary compatibility with previous glibc
2478 * various conversion modules for IBM character sets contributed by
2481 * IA-64 port by Jes Sorensen and HJ Lu.
2495 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2497 * New xdr functions are added; some rpc functions are now 64bit clean.
2499 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2501 * Fixed known incompatibilities with glibc 2.0.
2503 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2505 * Optimized a number of functions (especially the ELF dynamic loader).
2507 * Update timezone data files.
2509 * lots of charmaps corrections
2511 * some new locale definitions and charmaps
2516 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2517 conform to POSIX member type requirements. Symbol versions have been
2518 adjusted as needed within the library, and for direct use by applications,
2519 but there is potential for problems if third-party libraries use
2520 struct timeval as part of their interface. This does not present
2521 a problem for X and other "essential" system libraries.
2523 * An additional locale model to support C++ Standard Library locale
2524 model and probably more was implemented by Ulrich Drepper.
2526 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2529 * Miles Bader provided the `argp' function family to support hierachical
2530 command line argument parsing, layered on top of getopt.
2532 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2534 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2537 * scanf recognizes the %a and %A format for scanning floating point numbers.
2539 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2540 information and interfaces for the available integer types.
2542 * about 130 new math functions were added to implement the ISO C9x math
2545 * the new header <complex.h> contains definitions of the complex math
2546 functions from ISO C 9X.
2548 * the new header <tgmath.h> defines generic macros to use complex or
2549 real valued functions.
2551 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2553 * Andreas Jaeger provided a test suite for the math library.
2555 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2557 * Ulrich Drepper added character set conversion functions (iconv).
2559 * Optimized string functions have been added.
2561 * The localedata addon is now part of glibc.
2563 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2565 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2566 daemon for NSS (nscd).
2568 Missing a better place here are some numbers on improvements. Under
2569 Linux 2.1.125 un-tar-ing the kernel sources takes
2573 using local files 12.19s 6.88s 22.91s
2575 using NIS 13.92s 8.91s 26.34s
2577 using NIS & nscd 10.37s 7.34s 25.30s
2579 using NIS+ 27.57s 30.37s 640.46s
2581 using NIS+ & nscd 10.25s 7.83s 26.51s
2583 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2585 Keep in mind that non-namelookup related operations dominate above times.
2586 It was just a common complain that using NIS+ unpacking the kernel is
2589 [1] The old nscd implementation is not available anymore since it was
2590 distributed with glibc up to version 2.0.98 and thus is now replaced.
2592 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2594 * Mark Kettenis provided a Hesiod NSS module.
2596 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2597 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2599 * rcmd can now handle netgroups (Dick Streefland).
2601 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2604 * Support for the IPv6 protocol has been added to the socket API, as per the
2605 latest draft standards.
2607 * Support for Linux 2.2 has been added.
2609 * Interface changes relative to the latest 2.0.x release:
2610 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2611 addseverity NEW: Unix98
2612 alphasort64 NEW: LFS
2613 argp_err_exit_status NEW: argp, GNU ext
2614 argp_error NEW: argp, GNU ext
2615 argp_failure NEW: argp, GNU ext
2616 argp_help NEW: argp, GNU ext
2617 argp_parse NEW: argp, GNU ext
2618 argp_program_bug_address NEW: argp, GNU ext
2619 argp_program_version NEW: argp, GNU ext
2620 argp_program_version_hook NEW: argp, GNU ext
2621 argp_state_help NEW: argp, GNU ext
2622 argp_usage NEW: argp, GNU ext
2623 authdes_create NEW: Secure RPC
2624 authdes_getucred NEW: Secure RPC
2625 authdes_pk_create NEW: Secure RPC
2626 backtrace NEW: GNU ext.
2627 backtrace_symbols NEW: GNU ext.
2628 backtrace_symbols_fd NEW: GNU ext.
2630 cacosf NEW: ISO C 9x
2631 cacosh NEW: ISO C 9x
2632 cacoshf NEW: ISO C 9x
2633 cacoshl NEW: ISO C 9x
2634 cacosl NEW: ISO C 9x
2641 casinf NEW: ISO C 9x
2642 casinh NEW: ISO C 9x
2643 casinhf NEW: ISO C 9x
2644 casinhl NEW: ISO C 9x
2645 casinl NEW: ISO C 9x
2647 catanf NEW: ISO C 9x
2648 catanh NEW: ISO C 9x
2649 catanhf NEW: ISO C 9x
2650 catanhl NEW: ISO C 9x
2651 catanl NEW: ISO C 9x
2652 cbc_crypt NEW: Secure RPC
2656 ccoshf NEW: ISO C 9x
2657 ccoshl NEW: ISO C 9x
2663 cimagf NEW: ISO C 9x
2664 cimagl NEW: ISO C 9x
2665 clearerr_locked REMOVED
2666 clntunix_create NEW: sunrpc ext
2668 clog10 NEW: ISO C 9x
2669 clog10f NEW: ISO C 9x
2670 clog10l NEW: ISO C 9x
2680 cprojf NEW: ISO C 9x
2681 cprojl NEW: ISO C 9x
2683 crealf NEW: ISO C 9x
2684 creall NEW: ISO C 9x
2689 csinhf NEW: ISO C 9x
2690 csinhl NEW: ISO C 9x
2693 csqrtf NEW: ISO C 9x
2694 csqrtl NEW: ISO C 9x
2698 ctanhf NEW: ISO C 9x
2699 ctanhl NEW: ISO C 9x
2701 des_setparity NEW: Secure RPC
2702 ecb_crypt NEW: Secure RPC
2703 endutxent NEW: Unix98
2705 exp10f NEW: ISO C 9x
2706 exp10l NEW: ISO C 9x
2710 fattach NEW: STREAMS
2711 fdetach NEW: STREAMS
2715 feclearexcept NEW: ISO C 9x
2716 fegetenv NEW: ISO C 9x
2717 fegetexceptflag NEW: ISO C 9x
2718 fegetround NEW: ISO C 9x
2719 feholdexcept NEW: ISO C 9x
2721 feraiseexcept NEW: ISO C 9x
2722 ferror_locked REMOVED
2723 fesetenv NEW: ISO C 9x
2724 fesetexceptflag NEW: ISO C 9x
2725 fesetround NEW: ISO C 9x
2726 fetestexcept NEW: ISO C 9x
2727 feupdateenv NEW: ISO C 9x
2728 fflush_locked REMOVED
2732 fgets_unlocked NEW: GNU ext.
2733 fileno_locked REMOVED
2745 fputc_locked REMOVED
2746 fputs_unlocked NEW: GNU ext.
2747 fread_unlocked NEW: GNU ext.
2752 fstatvfs NEW: Unix98
2756 ftruncate64 NEW: LFS
2758 fwrite_unlocked NEW: GNU ext.
2759 gai_strerror NEW: IPv6
2763 getchar_locked REMOVED
2765 getdate_err NEW: Unix98
2766 getdate_r NEW: GNU ext.
2768 getnameinfo NEW: IPv6
2769 getnetname NEW: Secure RPC
2770 getpmsg NEW: STREAMS
2771 getpt NEW: Unix98 PTY
2772 getrlimit64 NEW: LFS
2773 getutxent NEW: Unix98
2774 getutxid NEW: Unix98
2775 getutxline NEW: Unix98
2776 glob64 NEW: GNU ext.
2777 globfree64 NEW: GNU ext.
2778 gnu_get_libc_release NEW: GNU ext.
2779 gnu_get_libc_version NEW: GNU ext.
2780 grantpt NEW: Unix98 PTY
2781 host2netname NEW: Secure RPC
2783 iconv_close NEW: iconv
2784 iconv_open NEW: iconv
2785 if_freenameindex NEW: IPv6
2786 if_indextoname NEW: IPv6
2787 if_nameindex NEW: IPv6
2788 if_nametoindex NEW: IPv6
2789 in6addr_any NEW: IPv6
2790 in6addr_loopback NEW: IPv6
2791 inet6_isipv4mapped NEW: IPv6
2792 isastream NEW: STREAMS
2793 iswblank NEW: GNU ext.
2794 key_decryptsession NEW: Secure RPC
2795 key_decryptsession_pk NEW: Secure RPC
2796 key_encryptsession NEW: Secure RPC
2797 key_encryptsession_pk NEW: Secure RPC
2798 key_gendes NEW: Secure RPC
2799 key_get_conv NEW: Secure RPC
2800 key_secretkey_is_set NEW: Secure RPC
2801 key_setnet NEW: Secure RPC
2802 key_setsecret NEW: Secure RPC
2803 llrint NEW: ISO C 9x
2804 llrintf NEW: ISO C 9x
2805 llrintl NEW: ISO C 9x
2806 llround NEW: ISO C 9x
2807 llroundf NEW: ISO C 9x
2808 llroundl NEW: ISO C 9x
2813 lrintf NEW: ISO C 9x
2814 lrintl NEW: ISO C 9x
2815 lround NEW: ISO C 9x
2816 lroundf NEW: ISO C 9x
2817 lroundl NEW: ISO C 9x
2819 makecontext NEW: Unix98
2820 mempcpy NEW: GNU ext.
2823 modify_ldt NEW: kernel
2827 nearbyint NEW: ISO C 9x
2828 nearbyintf NEW: ISO C 9x
2829 nearbyintl NEW: ISO C 9x
2830 netname2host NEW: Secure RPC
2831 netname2user NEW: Secure RPC
2832 nexttoward NEW: ISO C 9x
2833 nexttowardf NEW: ISO C 9x
2834 nexttowardl NEW: ISO C 9x
2838 passwd2des NEW: Secure RPC
2840 pow10f NEW: GNU ext.
2841 pow10l NEW: GNU ext.
2844 printf_size NEW: GNU ext.
2845 printf_size_info NEW: GNU ext.
2846 profil_counter REMOVED
2847 pthread_mutexattr_getkind_np REPLACED
2848 pthread_mutexattr_setkind_np REPLACED
2849 ptsname NEW: Unix98 PTY
2850 ptsname_r NEW: Unix98 PTY
2852 putchar_locked REMOVED
2853 putgrent NEW: GNU ext.
2855 putpmsg NEW: STREAMS
2856 pututxline NEW: Unix98
2860 readdir64_r NEW: LFS
2861 remquo NEW: ISO C 9x
2862 remquof NEW: ISO C 9x
2863 remquol NEW: ISO C 9x
2865 roundf NEW: ISO C 9x
2866 roundl NEW: ISO C 9x
2868 scalbln NEW: ISO C 9x
2869 scalblnf NEW: ISO C 9x
2870 scalblnl NEW: ISO C 9x
2872 sendfile NEW: kernel
2873 setcontext NEW: Unix98
2874 setrlimit64 NEW: LFS
2875 setutxent NEW: Unix98
2877 sigignore NEW: Unix98
2878 sigqueue NEW: POSIX.1b
2879 sigrelse NEW: Unix98
2880 sigset NEW: POSIX.1b
2881 sigtimedwait NEW: POSIX.1b
2882 sigwaitinfo NEW: POSIX.1b
2883 sincos NEW: GNU ext.
2884 sincosf NEW: GNU ext.
2885 sincosl NEW: GNU ext.
2889 strcasestr NEW: GNU ext.
2890 strtoimax NEW: ISO C 9x
2891 strtoumax NEW: ISO C 9x
2892 strverscmp NEW: GNU ext.
2893 svcauthdes_stats NEW: Secure RPC
2894 svcunix_create NEW: sunrpc etx
2895 svcunixfd_create NEW: sunrpc ext
2896 swapcontext NEW: Unix98
2897 tcgetsid NEW: Unix98 PTY
2898 tdestroy NEW: GNU ext.
2899 tgamma NEW: ISO C 9x
2900 tgammaf NEW: ISO C 9x
2901 tgammal NEW: ISO C 9x
2905 truncf NEW: ISO C 9x
2906 truncl NEW: ISO C 9x
2908 unlockpt NEW: Unix98 PTY
2909 updwtmpx NEW: Unix98
2910 user2netname NEW: Secure RPC
2911 utmpxname NEW: Unix98
2912 versionsort NEW: GNU ext.
2913 versionsort64 NEW: GNU ext.
2915 wcscasecmp NEW: GNU ext.
2916 wcsncasecmp NEW: GNU ext.
2917 wcsnlen NEW: GNU ext.
2918 wcstoimax NEW: ISO C 9x
2919 wcstoll NEW: ISO C 9x
2920 wcstoull NEW: ISO C 9x
2921 wcstoumax NEW: ISO C 9x
2923 wordexp NEW: POSIX.2
2924 wordfree NEW: POSIX.2
2925 write_profiling REMOVED
2926 xdecrypt NEW: Secure RPC
2927 xdr_authdes_cred NEW: Secure RPC
2928 xdr_authdes_verf NEW: Secure RPC
2929 xdr_cryptkeyarg NEW: Secure RPC
2930 xdr_cryptkeyarg2 NEW: Secure RPC
2931 xdr_cryptkeyres NEW: Secure RPC
2932 xdr_getcredres NEW: Secure RPC
2933 xdr_key_netstarg NEW: Secure RPC
2934 xdr_key_netstres NEW: Secure RPC
2935 xdr_keybuf NEW: Secure RPC
2936 xdr_keystatus NEW: Secure RPC
2937 xdr_netnamestr NEW: Secure RPC
2938 xdr_sizeof NEW: Secure RPC
2939 xdr_unixcred NEW: sunrpc ext
2940 xencrypt NEW: Secure RPC
2941 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2952 * inet_ntoa is thread-safe
2954 * updwtmp is moved from libutil to libc
2956 * rewrite of cbrt function
2958 * update of timezone data
2972 * add atoll function
2974 * fix complex problems in Berkeley DB code
2976 * fix math functions
2980 * fixed lots of header problems (especially Linux/GNU specific)
2982 * dynamic loader preserves all registers
2984 * Roland McGrath provided support for handling of auxiliary objects in
2985 the ELF dynamic loader.
2987 * support for parallel builds is improved
2991 * GNU extensions are no longer declared by default. To enable them you
2992 must define the macro `_GNU_SOURCE' in your program or compile with
2995 * The library has changed from using GNU ld symbol aliases to using weak
2996 symbols where available. The ELF object file format supports weak
2997 symbols; GNU ld also supports weak symbols in the a.out format. (There
2998 is also now support for other GNU ld extensions in ELF. Use the
2999 `--with-elf' option to configure to indicate you have ELF, and
3000 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3001 of many files which contained only symbol aliases, reducing the size of
3002 the source and the compiled library; many other files were renamed to
3003 less cryptic names previously occupied by the symbol alias files.
3004 There is a new header file <elf.h> for programs which operate on
3005 files in the ELF format.
3007 * Converted to Autoconf version 2, so `configure' has more options.
3008 Run `configure --help' to see the details.
3010 * The library can now be configured to build profiling, highly-optimized
3011 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3012 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3013 `configure' enable building these extra libraries. The shared library is
3014 built by default when using both ELF and GNU ld. When shared libraries
3015 are enabled, the new library `-ldl' is available for arbitrary run-time
3016 loading of shared objects; its interface is defined in <dlfcn.h>. The
3017 new header file <link.h> gives access to the internals of the run-time
3018 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3019 application of same name on other systems and it provides information
3020 about dynamically linked binaries.
3022 * The C library now provides the run-time support code for profiling
3023 executables compiled with `-pg'. Programs can control the profiling code
3024 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3025 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3026 the support for this file format was contributed by David Mosberger-Tang.
3028 * The math code has been replaced with a math library based on fdlibm from
3029 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3030 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3031 functions. The math functions now reside in a separate library, so
3032 programs using them will need to use `-lm' their linking commands.
3034 * John C. Bowman contributed optimized ix87 assembler inline functions.
3036 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3037 mechanism similar to that found in Solaris 2. This is now used for the
3038 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3039 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3040 file controls what services are used for each individual database. This
3041 works by loading shared libraries with names specified in `nsswitch.conf',
3042 so service modules can be changed or added at any time without even
3043 relinking any program. Currently there are the file, db, and NIS based
3044 NSS services available.
3046 * The new functions `strtoq' and `strtouq' parse integer values from
3047 strings, like `strtol' and `strtoul', but they return `long long int' and
3048 `unsigned long long int' values, respectively (64-bit quantities).
3050 * The new functions `strtof' and `strtold' parse floating-point values from
3051 strings, like `strtod', but they return `float' and `long double' values,
3052 respectively (on some machines `double' and `long double' are the same).
3054 * Ulrich Drepper has contributed new implementations of the floating-point
3055 printing and reading code used in the `printf' family of functions and
3056 `strtod', `strtof', and `strtold'. These new functions are perfectly
3057 accurate, and much faster than the old ones.
3059 * The implementation of the POSIX locale model was completely rewritten by
3060 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3061 to compile the POSIX locale definition.
3063 * The former dummy implementations of the strcoll and strxfrm function are
3064 now replaced by fully functional code contributed by Ulrich Drepper. The
3065 collation information comes from the POSIX locale definitions.
3067 * The new header <langinfo.h> defines an interface for accessing
3068 various locale-dependent data (using the locale chosen with `setlocale').
3070 * Ulrich Drepper has contributed a new suite of functions for operation on
3071 wide-character and multibyte-character strings, in <wchar.h>;
3072 and classification and case conversion of wide characters, in <wctype.h>.
3073 These new functions are conforming to the ISO C, Amendement 1 specification.
3075 * There is now a second implementation of the standard I/O library available.
3076 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3077 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3079 * You can now use positional parameter specifications in format strings
3080 for the `printf' and `scanf' families of functions. For example,
3081 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3082 ``Number 6, Mr Jones''. This is mainly useful when providing different
3083 format strings for different languages, whose grammars may dictate
3084 different orderings of the values being printed. To support this
3085 feature, the interface for `register_printf_handler' has changed; see
3086 the header file <printf.h> for details.
3088 * The `printf' and `scanf' families of functions now understand a new
3089 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3090 to group numbers as indicated by the locale; for `scanf' and friends, this
3091 says to accept as valid only a number with all the proper grouping
3092 separators in the right places. In the default "C" locale, numbers are
3093 not grouped; but locales for specific countries will define the usual
3094 conventions (i.e. separate thousands with `,' in the US locale).
3096 * The pgrp functions have been regularized, slightly incompatibly but much
3097 less confusingly. The core functions are now `getpgid' and `setpgid',
3098 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3099 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3100 provided for compatibility. There is no longer an incompatible `getpgrp'
3101 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3103 * The new header file <fts.h> and suite of functions simplify programs that
3104 operate on directory trees. This code comes from 4.4 BSD.
3106 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3107 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3108 NSS scheme used in glibc.
3110 * There is a new malloc debugging hook `__memalign_hook'.
3112 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3113 `unsigned int' in <sys/types.h>. These are for compatibility only and
3114 their use is discouraged.
3116 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3117 done differently, so that it works even without GNU ld.
3119 * New function `euidaccess' checks allowed access to a file like `access',
3120 but using the effective IDs instead of the real IDs.
3122 * The time zone data files have been updated for the latest and greatest
3123 local time conventions of the countries of the world.
3125 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3128 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3129 for formatting floating-point numbers. They are provided only for
3130 compatibility; new programs should use `sprintf' instead. There are
3131 also equivalent function for the `long double' floating-point type and
3132 all functions also exist in a reentrant form.
3134 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3135 functions for maintaining the login-record files (primarily of use to
3136 system programs such as `login'), and convenient functions for
3137 allocating and initializing a pseudo-terminal (pty) device.
3139 * Ulrich Drepper has contributed new support for System V style
3140 shared memory and IPC on systems that support it.
3142 * Ulrich Drepper has contributed several miscellaneous new functions found
3143 in System V: The `hsearch' family of functions provide an effective
3144 implementation of hash tables; `a64l' and `l64a' provide a very simple
3145 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3148 * Ulrich Drepper has contributed new reentrant counterparts for the
3149 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3151 * Ulrich Drepper has contributed new, highly-optimized versions of several
3152 string functions for the i486/Pentium family of processors.
3154 * Ulrich Drepper has updated the Linux-specific code, based largely
3155 on work done in Hongjiu Lu's version of GNU libc for Linux.
3156 The GNU library now supports Linux versions 2.0.10 and later,
3157 using the ELF object file format (i[3456]86-*-linux).
3159 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3161 * David Mosberger-Tang and Richard Henderson have ported the C library
3162 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3163 dynamic linking support for ELF/Alpha.
3165 * Richard Henderson contributed several Alpha optimized assembler function
3166 for arithmetic and string handling.
3168 * Ulrich Drepper has contributed a new set of message catalog functions to
3169 support multiple languages using the <libintl.h> interface, for use with
3170 his new package GNU gettext. Translation volunteers have contributed
3171 catalogs of the library's messages in Spanish, German, and Korean.
3173 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3174 program and the `catgets' function for reading the catalog files it
3175 creates. (The <libintl.h> interface is preferred; we include the
3176 <nl_types.h> interface using `catgets' only for source compatibility with
3177 programs already written to use it.)
3179 * New header file <values.h> gives SVID-compatible names for <limits.h>
3182 * Various new macros, declarations, and small header files for compatibility
3185 * New function `group_member' is a convenient way to check if a process has
3186 a given effective group ID.
3188 * When using GCC 2.7 and later, the socket functions are now declared in a
3189 special way so that passing an argument of type `struct sockaddr_in *',
3190 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3191 `struct sockaddr *' type, does not generate a type-clash warning.
3193 * New function `error' declared in header file <error.h> is a convenient
3194 function for printing error messages and optionally exiting; this is the
3195 canonical function used in GNU programs. The new functions `err', `warn',
3196 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3197 doing the same thing.
3199 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3200 POSIX.2 `glob' function to do ~ and {...} expansion.
3202 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3203 `clearenv' which is used in POSIX.9 is also available.
3205 * New function `getsid' returns session ID number on systems that support it.
3207 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3208 files <db.h> and <mpool.h> provide a rich set of functions for several
3209 types of simple databases stored in memory and in files, and <ndbm.h> is
3210 an old `ndbm'-compatible interface using the `db' functions. Link with
3211 `-ldb' to get these functions.
3213 * New macro `strdupa' copies a string like `strdup', but uses local stack
3214 space from `alloca' instead of dynamic heap space from `malloc'.
3216 * New function `strnlen' is like `strlen' but searches only a given maximum
3217 number of characters for the null terminator. `stpncpy', `strndup' and
3218 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3221 * New function `statfs' in header <sys/statfs.h>.
3223 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3224 provide convenient functions for operating on blocks of null-terminated
3227 * A new suite of functions in <utmp.h> handle all the details of reading
3228 and writing the utmp file.
3230 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3233 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3234 completely POSIX compliant and also implemented the extended functionality
3235 to handle alternate digit representation and alternate era date formats.
3237 * Ulrich Drepper provided an implementation of the `strptime' function
3238 defined in XPG4.2 which transforms a string into a `struct tm' value.
3240 * Paul Eggert provided the tzselect shell script as part of the timezone
3241 code. The shell script makes it easy to select the correct timezone
3244 * The implementation of the malloc family of functions is completely replaced
3245 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3246 The implementation uses the mmap function (if available) and it is
3247 optimized for the use in multi threaded programs.
3249 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3250 functions. This new functionality is usable by specifying a special
3251 salt string and it is compatible with implementation on *BSD systems.
3253 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3254 `getsubopt' to handle second level command line options, `bsd_signal'
3255 to access BSD style `signal' functionality, the obsolete `regexp' style
3258 * the `lchown' function is available on system which support this
3261 * The implementation of the shadow password handling function was contributed
3264 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3266 * POSIX.1g support was added. The <sys/select.h> header is available,
3267 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3268 implementation of `getaddrinfo'.
3272 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3274 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3275 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3277 * The new function `daemon' from 4.4 BSD is useful for server programs that
3278 want to put themselves in the background.
3280 * Joel Sherrill has contributed support for several standalone boards that
3281 run without an operating system.
3283 * `printf', `scanf' and friends now accept a `q' type modifier for long
3284 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3286 * All of the code taken from BSD (notably most of the math and networking
3287 routines) has been updated from the BSD 4.4-Lite release.
3289 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3291 * The new functions `getdomainname' and `setdomainname' fetch or change the
3292 YP/NIS domain name. These are system calls which exist on systems which
3295 * The time zone data files have been updated for the latest international
3298 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3299 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3303 * The C library now includes support for Sun RPC, from Sun's free
3304 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3305 are included. (There is still no support for YP.)
3307 * Tom Quinn has contributed a port of the C library to SGI machines running
3308 Irix 4 (mips-sgi-irix4).
3310 * The new `lockf' function is a simplified interface to the locking
3311 facilities of `fcntl', included for compatibility.
3313 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3315 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3317 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3320 * The error message printed by `assert' for a failed assertion now includes
3321 the name of the program (if using GNU ld) and the name of the calling
3322 function (with versions of GCC that support this).
3324 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3326 * The library now includes the <sys/mman.h> header file and memory
3327 management functions `mmap', `munmap', `mprotect', `msync', and
3328 `madvise', on systems that support those facilities.
3330 * The interface for `mcheck' has changed slightly: the function called to
3331 abort the program when an allocation inconsistency is detected now takes
3332 an argument that indicates the type of failure. The new function
3333 `mprobe' lets you request a consistency check for a particular block at
3334 any time (checks are normally done only when you call `free' or `realloc'
3337 * It is now possible to easily cross-compile the C library, building on one
3338 system a library to run on another machine and/or operating system. All
3339 you need to do is set the variable `HOST_CC' in `configparms' to the
3340 native compiler for programs to run on the machine you are building on (a
3341 few generator programs are used on Unix systems); set `CC' to the
3344 * The new function `fexecve' (only implemented on the GNU system) executes
3345 a program file given a file descriptor already open on the file.
3349 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3350 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3352 * You can set the variable `libprefix' in `configparms' to specify a prefix
3353 to be prepended to installed library files; this makes it easy to install
3354 the GNU C library to be linked as `-lgnuc' or whatever.
3356 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3357 copies a limited number of characters from a string, and returns the
3358 address of the last character written.
3360 * You no longer need to check for whether the installed `stddef.h' is
3361 compatible with the GNU C library. configure now checks for you.
3363 * You can now define a per-stream `fileno' function to convert the
3364 stream's cookie into an integral file descriptor.
3366 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3367 allocates zero bytes of storage, and returns a unique pointer which
3368 you can pass to `realloc' or `free'. The behavior is undefined if
3369 you dereference this pointer.
3371 * The C library now runs on Sony NEWS m68k machines running either
3372 NewsOS 3 or NewsOS 4.
3374 * The new `syscall' function is a system-dependent primitive function
3375 for invoking system calls. It has the canonical behavior on Unix
3376 systems, including unreliable return values for some calls (such as
3377 `pipe', `fork' and `getppid').
3379 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3380 to `EAGAIN', which is the preferred name. On systems whose kernels
3381 use two distinct codes, the C library now translates EWOULDBLOCK to
3382 EAGAIN in every system call function.
3386 * The GNU C Library Reference Manual is now distributed with the library.
3387 `make dvi' will produce a DVI file of the printed manual.
3388 `make info' will produce Info files that you can read on line using C-h i
3389 in Emacs or the `info' program.
3390 Please send comments on the manual to bug-glibc-manual@gnu.org.
3392 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3394 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3396 * Jason Merrill has contributed a port to the Sequent Symmetry running
3397 Dynix version 3 (i386-sequent-dynix).
3399 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3400 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3402 * New function `memory_warnings' lets you arrange to get warnings when
3403 malloc is running out of memory to allocate, like Emacs gives you.
3405 * The C library now contains the relocating allocator used in Emacs 19 for
3406 its editing buffers. This allocator (ralloc) minimizes allocation
3407 overhead and fragmentation by moving allocated regions around whenever it
3408 needs to. You always refer to a ralloc'd region with a "handle" (a
3409 pointer to a pointer--an object of type `void **').
3411 * There is a new `printf' format: `%m' gives you the string corresponding
3412 to the error code in `errno'.
3414 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3415 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3416 pass, the `a' modifier says to fill in a `char **' you pass with a
3419 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3420 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3421 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3423 * `mkstemp' is a traditional Unix function to atomically create and open a
3424 uniquely-named temporary file.
3428 * The standard location for the file that says what the local timezone is
3429 has changed again. It is now `/usr/local/etc/localtime' (or more
3430 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3432 * The distribution no longer contains any files with names longer than 14
3435 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3436 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3438 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3442 * The configuration process has changed quite a bit. The `configure'
3443 script is now used just like the configuration scripts for other GNU
3444 packages. The `sysdeps' directory hierarchy is much rearranged.
3445 The file `INSTALL' explains the new scheme in detail.
3447 * The header files no longer need to be processed into ANSI C and
3448 traditional C versions. There is just one set of files to install, and
3449 it will work with ANSI or old C compilers (including `gcc -traditional').
3451 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3452 MIPS DECStation running Ultrix 4.
3454 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3455 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3456 made itself into a shared library.
3458 * Yet further improved support for the i386, running 4.3 BSD-like systems
3459 (such as Mach 3 with the Unix single-server), or System V.
3461 * New function `strncasecmp' to do case-insensitive string comparison
3462 with limited length.
3464 * New function `strsep' is a reentrant alternative to `strtok'.
3466 * New functions `scandir' and `alphasort' for searching directories.
3468 * New function `setenv' is a better interface to `putenv'.
3470 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3471 function for traversing a directory tree.
3473 * The GNU obstack package is now also part of the C library.
3474 The new function `open_obstack_stream' creates a stdio stream that
3475 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3476 formatted output directly to an obstack.
3478 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3479 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3481 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3483 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3484 things to your strings.
3486 * There are some new test programs: test-fseek, testmb, and testrand.
3488 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3489 These ports are not finished, but are a good starting place for really
3490 supporting those systems.
3492 * `/etc/localtime' is now the standard location for the file that says what
3493 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3494 This follows the general principle that `/etc' is the place for all local
3495 configuration files.
3497 * The C library header files now use `extern "C"' when used by the C++
3498 compiler, so the C library should now work with C++ code.
3500 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3501 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3504 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3505 function which uses the `merge sort' algorithm, and is said to be
3506 significantly faster than the old GNU `qsort' function. Merge sort is
3507 now the standard `qsort' function. The new algorithm can require a lot
3508 of temporary storage; so, the old sorting function is called when the
3509 required storage is not available.
3511 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3512 provides the Unix `crypt' function, plus some other entry points.
3514 * The code and header files taken from 4.4 BSD have been updated with the
3515 latest files released from Berkeley.
3517 ----------------------------------------------------------------------
3518 Copyright information:
3520 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3522 Permission is granted to anyone to make or distribute verbatim copies
3523 of this document as received, in any medium, provided that the
3524 copyright notice and this permission notice are preserved,
3525 thus giving the recipient permission to redistribute in turn.
3527 Permission is granted to distribute modified versions
3528 of this document, or of portions of it,
3529 under the above conditions, provided also that they
3530 carry prominent notices stating who last changed them.
3533 version-control: never