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.
68 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
69 to float instead of double. This does not affect the ABI of any libraries
70 that are part of the GNU C Library, but may affect the ABI of other
71 libraries that use this type in their interfaces.
73 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
74 are added to libc. They convert a floating-point number into string.
76 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
77 for the Linux quota interface which predates kernel version 2.4.22 has
80 * The malloc_get_state and malloc_set_state functions have been removed.
81 Already-existing binaries that dynamically link to these functions will
82 get a hidden implementation in which malloc_get_state is a stub. As far
83 as we know, these functions are used only by GNU Emacs and this change
84 will not adversely affect already-built Emacs executables. Any undumped
85 Emacs executables, which normally exist only during an Emacs build, should
86 be rebuilt by re-running “./configure; make” in the Emacs build tree.
88 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
89 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
90 “no-ip6-dotint” had already been the default, and support for the
91 “ip6-dotint” option was removed from the Internet in 2006.
93 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
94 flag from <resolv.h> have been removed. The option relied on a
95 backwards-incompatible DNS extension which was never deployed on the
98 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
99 in the <resolv.h> header file have been deprecated. They were already
102 * DNSSEC-related declarations and definitions have been removed from the
103 <arpa/nameser.h> header file, and libresolv will no longer attempt to
104 decode the data part of DNSSEC record types. Previous versions of glibc
105 only implemented minimal support for the previous version of DNSSEC, which
106 is incompatible with the currently deployed version.
108 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
109 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
110 <arpa/nameser.h> header file because the distinction between RR types and
111 meta-RR types is not officially standardized, subject to revision, and
112 thus not suitable for encoding in a macro.
114 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
115 and rhook members of the res_state type in <resolv.h> have been removed.
116 The glibc stub resolver did not support these hooks, but the header file
117 did not reflect that.
119 * For multi-arch support it is recommended to use a GCC which has
120 been built with support for GNU indirect functions. This ensures
121 that correct debugging information is generated for functions
122 selected by IFUNC resolvers. This support can either be enabled by
123 configuring GCC with '--enable-gnu-indirect-function', or by
124 enabling it by default by setting 'default_gnu_indirect_function'
125 variable for a particular architecture in the GCC source file
128 Security related changes:
130 On ARM EABI (32-bit), generating a backtrace for execution contexts which
131 have been created with makecontext could fail to terminate due to a
132 missing .cantunwind annotation. This has been observed to lead to a hang
133 (denial of service) in some Go applications compiled with gccgo. Reported
134 by Andreas Schwab. (CVE-2016-6323)
136 The following bugs are resolved with this release:
138 [The release manager will add the list generated by
139 scripts/list-fixed-bugs.py just before the release.]
143 * The minimum Linux kernel version that this version of the GNU C Library
144 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
145 kernel version 2.6.32 or later suffices (on architectures that already
146 required kernel versions more recent than 3.2, those requirements remain
147 unchanged). Linux 3.2 or later kernel headers are required on all
150 * The pap_AN locale has been deleted. This has been deprecated for a long
151 time. It has been replaced by pap_AW & pap_CW, both of which have long
152 been included in previous releases.
154 * The readdir_r and readdir64_r functions have been deprecated. It is
155 recommended to use readdir and readdir64 instead.
157 * The type “union wait” has been removed. It was deprecated in the early
158 1990s and never part of POSIX. Application code should use the int type
159 instead of “union wait”.
161 * A new NSS action is added to facilitate large distributed system
162 administration. The action, MERGE, allows remote user stores like LDAP
163 to be merged into local user stores like /etc/groups in order to provide
164 easy to use, updated, and managed sets of merged credentials. The new
165 action can be used by configuring it in /etc/nsswitch.conf:
166 group: files [SUCCESS=merge] nis
167 Implemented by Stephen Gallagher (Red Hat).
169 * The deprecated __malloc_initialize_hook variable has been removed from the
172 * The long unused localedef --old-style option has been removed. It hasn't
173 done anything in over 16 years. Scripts using this option can safely
176 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
177 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
178 return the next representable value in the direction of positive infinity
179 and the nextdown functions return the next representable value in the
180 direction of negative infinity. These are currently enabled as GNU
183 Security related changes:
185 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
186 could result in a stack overflow when getnetbyname was called with an
187 overly long name. (CVE-2016-3075)
189 * Previously, getaddrinfo copied large amounts of address data to the stack,
190 even after the fix for CVE-2013-4458 has been applied, potentially
191 resulting in a stack overflow. getaddrinfo now uses a heap allocation
192 instead. Reported by Michael Petlan. (CVE-2016-3706)
194 * The glob function suffered from a stack-based buffer overflow when it was
195 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
196 Reported by Alexander Cherepanov. (CVE-2016-1234)
198 * The Sun RPC UDP client could exhaust all available stack space when
199 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
200 alloca plugin for GCC. (CVE-2016-4429)
202 * The IPv6 name server management code in libresolv could result in a memory
203 leak for each thread which is created, performs a failing naming lookup,
204 and exits. Over time, this could result in a denial of service due to
205 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
207 The following bugs are resolved with this release:
209 [1170] localedata: ne_NP: update Nepali locale definition file
210 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
212 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
213 [6796] math: fdim() does not set errno on overflow
214 [10354] libc: posix_spawn should use vfork() in more cases than presently
215 [11213] localedata: localedata: add copyright disclaimer to locale files
216 [12143] localedata: chr_US: new Cherokee locale
217 [12450] localedata: sgs_LT: new locale
218 [12676] localedata: ln_CD: new locale
219 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
221 [13304] math: fma, fmaf, fmal produce wrong results
222 [14259] build: --localedir arg to configure is ignored
223 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
224 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
225 [14934] localedata: es_CL: wrong first weekday chilean locale
226 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
228 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
230 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
231 [15368] nptl: raise() is not async-signal-safe
232 [15479] math: ceil, floor, round and trunc raise inexact exception
233 [15578] localedata: kk_KZ: various updates
234 [16003] localedata: pap_AN: punt old locale
235 [16137] localedata: iw_IL: punt old locale
236 [16190] localedata: eo: new esperanto locale
237 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
238 [16742] malloc: race condition: pthread_atfork() called before first
239 malloc() results in unexpected locking behaviour/deadlocks
240 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
242 [16983] localedata: postal_fmt does not allow %l and %n modifiers
243 [17565] localedata: pt_PT: wrong (work-)week start
244 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
245 [17950] build: Build fails with -msse
246 [18205] localedata: be_BY*: wrong first_weekday and first_workday
247 [18433] libc: posix_spawn does not return correctly upon failure to
249 [18453] localedata: charmaps/IBM875: incorrect codes
250 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
252 [18896] localedata: he_IL: improvements for currency
253 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
255 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
257 [19056] libc: Deprecate readdir_r
258 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
260 [19198] localedata: nl_NL: small improvements for Dutch locales
261 [19257] network: Per-thread memory leak in __res_vinit with IPv6
262 nameservers (CVE-2016-5417)
263 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
265 [19400] locale: Language missing in "iso-639.def", trivial fix in
267 [19431] malloc: Deadlock between fflush, getdelim, and fork
268 [19505] libc: Incorrect file descriptor validity checks in
269 posix_spawn_file_actions_add{open,close,dup2}
270 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
272 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
273 `intl/{gettextP,loadinfo}.h'
274 [19534] libc: execle, execlp may use malloc
275 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
276 [19573] network: res_nclose and __res_maybe_init disagree about name
277 server initialization, breaking Hesiod
278 [19575] localedata: Status of GB18030 tables
279 [19581] localedata: sr_* date_fmt string contains additional newline
280 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
282 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
283 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
284 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
285 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
287 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
289 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
291 [19610] dynamic-link: ldconfig -X removes stale symbolic links
292 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
293 [19633] locale: strfmon_l applies global locale to number formatting
294 [19642] network: Memory leak in getnameinfo
295 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
296 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
298 [19654] math: [x86_64] Need testcase for BZ #19590 fix
299 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
301 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
302 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
304 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
306 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
308 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
309 pointers and lengths in error-case.
310 [19727] locale: Converting from/to UTF-xx with iconv() does not always
311 report errors on UTF-16 surrogates values.
312 [19755] nscd: nscd assertion failure in gc
313 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
314 [19759] libc: mempcpy shouldn't be inlined
315 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
316 [19765] libc: s390 needs an optimized mempcpy
317 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
318 NAME_MAX limit assumption (CVE-2016-1234)
319 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
320 [19787] network: Missing and incorrect truncation checks in getnameinfo
321 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
323 [19791] network: Assertion failure in res_query.c with un-connectable name
325 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
326 [19822] math: libm.so install clobbers old version
327 [19825] network: resolv: send_vc can return uninitialized data in second
328 response to getaddrinfo
329 [19830] network: nss_dns: should check RDATA length against buffer length
330 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
331 confronted with A/AAAA records of invalid size
332 [19837] nss: nss_db: No retries for some long lines with a larger buffer
333 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
334 [19853] stdio: Printing IBM long double in decimal with high precision is
336 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
337 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
338 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
340 [19865] network: Assertion failure or memory leak in
341 _nss_dns_getcanonname_r
342 [19868] network: nss_dns: netent code does not skip over non-PTR records
343 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
345 [19881] string: Improve x86-64 memset
346 [19907] string: Incorrect memcpy tests
347 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
349 [19925] libc: termios.h XCASE namespace
350 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
351 [19929] libc: limits.h NL_NMAX namespace
352 [19931] stdio: Memory leak in vfprintf
353 [19957] libc: clone(CLONE_VM) access invalid parent memory
354 [19963] localedata: en_IL: New locale
355 [19989] stdio: stdio.h cuserid namespace
356 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
358 [19996] locale: langinfo.h nl_langinfo_l namespace
359 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
360 [20010] network: getaddrinfo: Stack overflow in hostent translation
362 [20012] stdio: libio: fmemopen append mode failure
363 [20014] stdio: stdio.h namespace for pre-threads POSIX
364 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
365 [20023] libc: fcntl.h timespec namespace
366 [20024] math: [x86_64] vectorized sincos trashes the stack
367 [20031] network: nss_hesiod: Heap overflow in get_txt_records
368 [20041] time: sys/time.h timespec namespace
369 [20043] libc: unistd.h missing cuserid for UNIX98 and before
370 [20044] libc: unistd.h missing pthread_atfork for UNIX98
371 [20051] libc: ttyslot in wrong header under wrong conditions
372 [20054] libc: gethostname not declared for XPG4
373 [20055] libc: termios.h missing tcgetsid for XPG4
374 [20072] dynamic-link: x86 init_cpu_features is called twice in static
376 [20073] libc: sys/stat.h fchmod namespace
377 [20074] libc: stdlib.h rand_r namespace
378 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
379 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
381 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
382 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
384 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
385 [20119] libc: Wrong mask for processors level type from CPUID
386 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
387 AS not supporting AVX512
388 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
390 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
391 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
393 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
394 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
395 [20178] libc: posix_spawn{p} should not call exit
396 [20191] stdio: libio: vtables hardening
397 [20195] string: FMA4 detection requires CPUID execution with register
399 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
400 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
402 [20212] math: acos (sNaN) returns sNaN
403 [20213] math: asin (sNaN) returns sNaN
404 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
405 [20218] math: [i386] asinhl (sNaN) returns sNaN
406 [20219] math: [i386] atanhl (sNaN) returns sNaN
407 [20222] stdio: fopencookie: Mangle function pointers
408 [20224] math: [i386] cbrtl (sNaN) returns sNaN
409 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
410 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
412 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
413 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
414 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
415 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
416 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
417 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
418 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
419 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
420 [20240] math: modf (sNaN) returns sNaN
421 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
422 [20250] math: frexp (sNaN) returns sNaN
423 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
424 [20255] math: [i386] fdim, fdimf return with excess range and precision /
426 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
427 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
428 subscript is below array bounds [-Werror=array-bounds]
429 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
430 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
432 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
433 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
435 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
437 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
438 [20316] localedata: id_ID: Februari instead of Pebruari
439 [20327] string: POWER8 strcasecmp returns incorrect result
440 [20347] math: Failure: Test: j0_downward (0xap+0)
441 [20348] libc: FAIL: misc/tst-preadvwritev64
442 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
443 [20350] libc: There is no test for p{read,write}64
444 [20357] math: Incorrect cos result for 1.5174239687223976
445 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
449 * Unicode 8.0.0 Support: Character encoding, character type info, and
450 transliteration tables are all updated to Unicode 8.0.0, using new
451 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
452 These updates cause user visible changes, such as the fixes for bugs
453 89, 16061, and 18568.
455 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
456 kernel-internal CPU set size. This means that requests that change the
457 CPU affinity which failed before (for example, an all-ones CPU mask) will
458 now succeed. Applications that need to determine the effective CPU
459 affinities need to call sched_getaffinity or pthread_getaffinity_np after
460 setting it because the kernel can adjust it (and the previous size check
461 would not detect this in the majority of cases).
463 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
464 the following new symbols are used: fts64_children, fts64_close,
465 fts64_open, fts64_read and fts64_set.
467 * getaddrinfo now detects certain invalid responses on an internal netlink
468 socket. If such responses are received, an affected process will
469 terminate with an error message of "Unexpected error <number> on netlink
470 descriptor <number>" or "Unexpected netlink response of size <number> on
471 descriptor <number>". The most likely cause for these errors is a
472 multi-threaded application which erroneously closes and reuses the netlink
473 file descriptor while it is used by getaddrinfo.
475 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
476 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
477 unnecessary serialization of memory allocation requests across threads.
478 The defect is now corrected. Users should see a substantial increase in
479 the concurent throughput of allocation requests for applications which
480 trigger this bug. Affected applications typically create and
481 destroy threads frequently. (Bug 19048 was reported and analyzed by
484 * There is now a --disable-timezone-tools configure option for disabling the
485 building and installing of the timezone related utilities (zic, zdump, and
486 tzselect). This is useful for people who build the timezone data and code
487 independent of the GNU C Library.
489 * The obsolete header <regexp.h> has been removed. Programs that require
490 this header must be updated to use <regex.h> instead.
492 * The obsolete functions bdflush, create_module, get_kernel_syms,
493 query_module and uselib are no longer available to newly linked binaries;
494 the header <sys/kdaemon.h> has been removed. These functions and header
495 were specific to systems using the Linux kernel and could not usefully be
496 used with the GNU C Library on systems with version 2.6 or later of the
499 * Optimized string, wcsmbs and memory functions for IBM z13.
500 Implemented by Stefan Liebler.
502 * Newly linked programs that define a variable called signgam will no longer
503 have it set by the lgamma, lgammaf and lgammal functions. Programs that
504 require signgam to be set by those functions must ensure that they use the
505 variable provided by the GNU C Library and declared in <math.h>, without
506 defining their own copy.
508 * The minimum GCC version that can be used to build this version of the GNU
509 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
510 still be used to compile programs using the GNU C Library.
512 Security related changes:
514 * An out-of-bounds value in a broken-out struct tm argument to strftime no
515 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
517 * The LD_POINTER_GUARD environment variable can no longer be used to disable
518 the pointer guard feature. It is always enabled. Previously,
519 LD_POINTER_GUARD could be used to disable security hardening in binaries
520 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
523 * An integer overflow in hcreate and hcreate_r could lead to an
524 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
526 * The catopen function no longer has unbounded stack usage. Reported by
529 * The nan, nanf and nanl functions no longer have unbounded stack usage
530 depending on the length of the string passed as an argument to the
531 functions. Reported by Joseph Myers. (CVE-2014-9761)
533 * A stack-based buffer overflow was found in libresolv when invoked from
534 libnss_dns, allowing specially crafted DNS responses to seize control
535 of execution flow in the DNS client. The buffer overflow occurs in
536 the functions send_dg (send datagram) and send_vc (send TCP) for the
537 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
538 family. The use of AF_UNSPEC triggers the low-level resolver code to
539 send out two parallel queries for A and AAAA. A mismanagement of the
540 buffers used for those queries could result in the response of a query
541 writing beyond the alloca allocated buffer created by
542 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
543 the overflow. Thanks to the Google Security Team and Red Hat for
544 reporting the security impact of this issue, and Robert Holiday of
545 Ciena for reporting the related bug 18665. (CVE-2015-7547)
547 The following bugs are resolved with this release:
549 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
550 [887] math: Math library function "logb" and "nextafter" inconsistent
551 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
552 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
553 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
554 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
556 [4404] localedata: German translation of "Alarm clock" is misleading
557 [6799] math: nextafter() and nexttoward() doen't set errno on
558 overflow/underflow errors
559 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
561 [10432] nis: _nss_nis_setnetgrent assertion failure
562 [11460] libc: fts has no LFS support
563 [12926] network: getaddrinfo()/make_request() may spin forever
564 [13065] nptl: Race condition in pthread barriers
565 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
566 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
568 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
569 [14912] libc: Rename non-installed bits/*.h headers
570 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
571 [15367] math: Let gcc use __builtin_isinf
572 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
573 [15421] math: lgamma wrongly sets signgam for ISO C
574 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
575 FE_INVALID with argument out of range
576 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
578 [15786] dynamic-link: ifunc resolver functions can smash function
580 [15918] math: Unnecessary check for equality in hypotf()
581 [16061] localedata: Review / update transliteration data
582 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
584 [16141] time: strptime %z offset restriction
585 [16171] math: drem should be alias of remainder
586 [16296] math: fegetround is pure?
587 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
588 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
590 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
591 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
593 [16422] math: [powerpc] math-float, math-double failing llrint tests with
594 "Exception "Inexact" set" on ppc32
595 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
596 [16517] math: Missing underflow exception from tanf/tan/tanl
597 [16519] math: Missing underflow exception from sinhf
598 [16520] math: Missing underflow exception from tanhf
599 [16521] math: Missing underflow exception from exp2
600 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
602 [16734] stdio: fopen calls mmap to allocate its buffer
603 [16961] math: nan function incorrect handling of bad sequences
604 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
605 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
606 [16985] locale: localedef: confusing error message when opening output
608 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
609 [17197] locale: Redundant shift character in iconv conversion output at
611 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
612 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
613 [17250] dynamic-link: static linking breaks nss loading
614 (getaddrinfo/getpwnam/etc...)
615 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
617 [17441] math: isnan() should use __builtin_isnan() in GCC
618 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
619 (related to lock elision)
620 [17787] manual: Exponent on page 324 of the PDF ends prematurely
621 [17886] time: strptime should be able to parse "Z" as a timezone with %z
622 [17887] time: strptime should be able to parse "+01:00" style timezones
623 [17905] libc: catopen() Multiple unbounded stack allocations
625 [18084] libc: backtrace (..., 0) dumps core on x86
626 [18086] libc: nice() sets errno to 0 on success
627 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
628 is too large (CVE-2015-8778)
629 [18251] dynamic-link: SONAME missing when audit modules provides path
630 [18265] libc: add attributes for wchar string and memory functions
631 [18370] math: csqrt missing underflows
632 [18421] libc: [hppa] read-only segment has dynamic relocations
633 [18472] libc: Obsolete syscall wrappers should be compat symbols
634 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
635 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
636 [18525] localedata: Remove locale timezone information
637 [18560] libc: [powerpc] spurious bits/ipc.h definitions
638 [18568] localedata: Update locale data to Unicode 8.0
639 [18589] locale: sort-test.sh fails at random
640 [18595] math: ctan, ctanh missing underflows
641 [18604] libc: assert macro-expands its argument
642 [18610] math: S390: fetestexcept() reports any exception if DXC-code
643 contains a vector instruction exception.
644 [18611] math: j1, jn missing errno setting on underflow
645 [18618] localedata: sync Chechen locale definitions with other *_RU
647 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
648 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
649 [18665] network: In send_dg, the recvfrom function is NOT always using the
650 buffer size of a newly created buffer (CVE-2015-7547)
651 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
653 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
655 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
656 [18699] math: tilegx cproj() for various complex infinities does not yield
658 [18724] libc: Harden put*ent functions against data injection
659 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
660 [18755] build: build errors with -DNDEBUG
661 [18757] stdio: fmemopen fails to set errno on failure
662 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
664 [18781] libc: openat64 lacks O_LARGEFILE
665 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
666 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
667 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
668 [18790] math: [ldbl-128ibm] tanhl inaccurate
669 [18795] libc: stpncpy fortification misses buffer lengths that are
671 [18796] build: build fails for --disable-mathvec
672 [18803] math: hypot missing underflows
673 [18820] stdio: fmemopen may leak memory on failure
674 [18823] math: csqrt spurious underflows
675 [18824] math: fma spurious underflows
676 [18825] math: pow missing underflows
677 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
678 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
679 [18870] build: sem_open.c fails to compile with missing symbol
681 [18872] stdio: Fix memory leak in printf_positional
682 [18873] libc: posix_fallocate overflow check ineffective
683 [18875] math: Excess precision leads incorrect libm
684 [18877] libc: arm: mmap offset regression
685 [18887] libc: memory corruption when using getmntent on blank lines
686 [18918] localedata: hu_HU: change time to HH:MM:SS format
687 [18921] libc: Regression: extraneous stat() and fstat() performed by
689 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
690 binaries (CVE-2015-8777)
691 [18951] math: tgamma missing underflows
692 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
694 [18953] localedata: lt_LT: change currency symbol to the euro
695 [18956] math: powf inaccuracy
696 [18961] math: [i386] exp missing underflows
697 [18966] math: [i386] exp10 missing underflows
698 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
699 [18969] build: multiple string test failures due to missing locale
701 [18970] libc: Reference of pthread_setcancelstate in libc.a
702 [18977] math: float / long double Bessel functions not in XSI POSIX
703 [18980] math: i386 libm functions return with excess range and precision
704 [18981] math: i386 scalb*, ldexp return with excess range and precision
705 [18982] stdio: va_list and vprintf
706 [18985] time: Passing out of range data to strftime() causes a segfault
708 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
709 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
710 [19012] locale: iconv_open leaks memory on error path
711 [19016] math: clog, clog10 inaccuracy
712 [19018] nptl: Mangle function pointers in tls_dtor_list
713 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
714 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
715 [19048] malloc: malloc: arena free list can become cyclic, increasing
717 [19049] math: [powerpc] erfc incorrect zero sign
718 [19050] math: [powerpc] log* incorrect zero sign
719 [19058] math: [x86_64] Link fail with -fopenmp and -flto
720 [19059] math: nexttoward overflow incorrect in non-default rounding modes
721 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
722 [19074] network: Data race in _res_hconf_reorder_addrs
723 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
724 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
725 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
727 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
729 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
730 [19086] manual: posix_fallocate64 documented argument order is wrong.
731 [19088] math: lround, llround missing exceptions close to overflow
733 [19094] math: lrint, llrint missing exceptions close to overflow threshold
734 [19095] math: dbl-64 lrint incorrect for 64-bit long
735 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
736 [19124] dynamic-link: ld.so failed to build with older assmebler
737 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
738 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
739 [19134] math: [powerpc32] lround, lroundf spurious exceptions
740 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
741 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
742 pthread_setaffinity_np
743 [19156] math: [ldbl-128] j0l spurious underflows
744 [19164] nptl: tst-getcpu fails with many possible CPUs
745 [19168] math: math/test-ildoubl and math/test-ldouble failure
746 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
747 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
749 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
750 (FE_NOMASK_ENV) do not clear SSE exceptions
751 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
752 _int_new_arena/reused_arena
753 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
754 [19201] math: dbl-64 remainder incorrect sign of zero result
755 [19205] math: bits/math-finite.h conditions do not match math.h and
757 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
758 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
759 for C99-based standards
760 [19212] libc: features.h not -Wundef clean
761 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
763 [19214] libc: Family and model identification for AMD CPU's are incorrect.
764 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
765 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
767 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
769 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
771 [19242] libc: strtol incorrect in Turkish locales
772 [19243] malloc: reused_arena can pick an arena on the free list, leading
773 to an assertion failure and reference count corruption
774 [19253] time: tzset() ineffective when temporary TZ did not include DST
776 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
777 [19270] math: [hppa] Shared libm missing __isnanl
778 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
779 MAP_HUGETLB and MAP_STACK defines
780 [19313] nptl: Wrong __cpu_mask for x32
781 [19347] libc: grantpt: try to force a specific gid even without pt_chown
782 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
783 [19350] math: [ldbl-128ibm] sinhl spurious overflows
784 [19351] math: [ldbl-128ibm] logl inaccurate near 1
785 [19363] time: x32: times() return value wrongly truncates/sign extends
787 [19367] dynamic-link: Improve branch prediction on Silvermont
788 [19369] network: Default domain name not reset by res_ninit when "search"
789 / "domain" entry is removed from resolv.conf
790 [19375] math: powerpc: incorrect results for POWER7 logb with negative
792 [19385] localedata: bg_BG: time separator should be colon, not comma
793 [19408] libc: linux personality syscall wrapper may erroneously return an
794 error on 32-bit architectures
795 [19415] libc: dladdr returns wrong names on hppa
796 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
797 IBM905, IBM907, and IBM909
798 [19439] math: Unix98 isinf and isnan functions conflict with C++11
799 [19443] build: build failures with -DDEBUG
800 [19451] build: Make check fails on test-double-vlen2
801 [19462] libc: Glibc failed to build with -Os
802 [19465] math: Wrong code with -Os
803 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
805 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
807 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
808 [19486] math: S390: Math tests fail with "Exception Inexact set".
809 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
810 [19550] libc: [mips] mmap negative offset handling inconsistent with other
812 [19590] math: Fail to build shared objects that use libmvec.so functions.
816 * The following bugs are resolved with this release:
818 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
819 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
820 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
821 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
822 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
823 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
824 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
825 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
826 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
827 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
828 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
829 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
830 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
831 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
832 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
833 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
834 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
835 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
836 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
837 18657, 18676, 18694, 18696, 18887.
839 * Cache information can be queried via sysconf() function on s390 e.g. with
840 _SC_LEVEL1_ICACHE_SIZE as argument.
842 * A buffer overflow in gethostbyname_r and related functions performing DNS
843 requests has been fixed. If the NSS functions were called with a
844 misaligned buffer, the buffer length change due to pointer alignment was
845 not taken into account. This could result in application crashes or,
846 potentially arbitrary code execution, using crafted, but syntactically
847 valid DNS responses. (CVE-2015-1781)
849 * The time zone file parser has been made more robust against crafted time
850 zone files, avoiding heap buffer overflows related to the processing of
851 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
852 large time zone data files. Overly long time zone specifiers in the TZ
853 variable no longer result in stack overflows and crashes.
855 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
856 for LD and GD on x86 and x86-64, has been implemented. You will need
857 binutils-2.24 or later to enable this optimization.
859 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
860 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
861 Hat). These updates cause user visible changes, such as the fix for bug
864 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
865 and getXXbyYY NSS calls for the same database, causing a denial-of-service
866 condition in some applications.
868 * Added vector math library named libmvec with the following vectorized x86_64
869 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
871 The library can be disabled with --disable-mathvec. Use of the functions is
872 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
873 Shared library libmvec.so is linked in as needed when using -lm (no need to
874 specify -lmvec explicitly for not static builds).
875 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
877 * A new fmemopen implementation has been added with the goal of POSIX
878 compliance. The new implementation fixes the following long-standing
879 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
880 old implementation is still present for use be by existing binaries.
882 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
883 releases. It has been fixed to match 2.19 and older, but binaries built
884 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
886 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
887 Contributed by Roland McGrath (Google).
889 * The header <regexp.h> is deprecated, and will be removed in a future
890 release. Use of this header will trigger a deprecation warning.
891 Application developers should update their code to use <regex.h> instead.
893 This header was formerly part of SUS, but was deprecated in 1994 and
894 removed from the standard in 2001. Also, the glibc implementation
895 leaks memory. See BZ#18681 for more details.
899 * The following bugs are resolved with this release:
901 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
902 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
903 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
904 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
905 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
906 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
907 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
908 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
909 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
910 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
913 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
914 too little memory for the to-be-scanned arguments and overflow the
915 allocated buffer. The implementation now correctly computes the required
916 buffer size when using malloc, and switches to malloc from alloca as
919 * A new semaphore algorithm has been implemented in generic C code for all
920 machines. Previous custom assembly implementations of semaphore were
921 difficult to reason about or ensure that they were safe. The new version
922 of semaphore supports machines with 64-bit or 32-bit atomic operations.
923 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
924 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
926 * Port to Altera Nios II has been contributed by Mentor Graphics.
928 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
929 implementations for powerpc64/powerpc64le.
930 Implemented by Adhemerval Zanella (IBM).
932 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
933 and powerpc64le. This may improve lock scaling of existing programs on
934 HTM capable systems. The lock elision code is only enabled with
935 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
936 powerpc will issue a transaction abort on every syscall to avoid side
937 effects being visible outside transactions.
939 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
940 AArch64. Contributed by ARM Ltd.
942 * i386 memcpy functions optimized with SSE2 unaligned load/store.
944 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
945 under certain input conditions resulting in the execution of a shell for
946 command substitution when the application did not request it. The
947 implementation now checks WRDE_NOCMD immediately before executing the
948 shell and returns the error WRDE_CMDSUB as expected.
950 * CVE-2012-3406 printf-style functions could run into a stack overflow when
951 processing format strings with a large number of format specifiers.
953 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
954 infinite loop if the DNS response contained a PTR record of an unexpected
957 * The minimum GCC version that can be used to build this version of the GNU
958 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
959 still be used to compile programs using the GNU C Library.
961 * The GNU C Library is now built with -Werror by default. This can be
962 disabled by configuring with --disable-werror.
964 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
966 * The obsolete sigvec function has been removed. This was the original
967 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
968 programs have been using instead for about 25 years. Of course, ABI
969 compatibility for old binaries using sigvec remains intact.
971 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
972 with newer versions of bison.
974 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
975 The original MIPS o32 hard-float ABI requires an FPU where double-precision
976 registers overlay two consecutive single-precision registers. MIPS32R2
977 introduced a new FPU mode (FR=1) where double-precision registers extend the
978 corresponding single-precision registers which is incompatible with the
979 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
980 require the use of FR=1 making a transition necessary. New o32 ABI
981 extensions enable users to migrate over time from the original o32 ABI
982 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
983 tracks the ABI of any loaded object and verifies that new objects are
984 compatible. Mode transitions will also be requested as required and
985 unsupportable objects will be rejected. The ABI checks include both soft and
986 hard float ABIs for o32, n32 and n64.
988 GCC 5 with GNU binutils 2.25 onwards:
989 It is strongly recommended that all o32 system libraries are built using the
990 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
991 with the original and all new o32 ABI extensions. Configure a MIPS GCC
992 compiler using --with-fp-32=xx to set this by default.
996 * The following bugs are resolved with this release:
998 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
999 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1000 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1001 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1002 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1003 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1004 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1005 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1006 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1007 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1008 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1009 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1010 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1011 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1012 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1014 * Reverted change of ABI data structures for s390 and s390x:
1015 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1016 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1017 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1018 are preserved pointing straight to the same implementation as the old ones.
1019 Given that, new callers will simply provide a too-big buffer to these
1020 functions. Any applications/libraries out there that embed jmp_buf or
1021 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1022 against 2.19 headers will have to rebuilt again. This is necessary in any
1023 case to revert the breakage in their ABI caused by the glibc change.
1025 * Support for file description locks is added to systems running the
1026 Linux kernel. The standard file locking interfaces are extended to
1027 operate on file descriptions, not file descriptors, via the use of
1028 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1029 are associated with an open file instead of a process.
1031 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1033 * The minimum Linux kernel version that this version of the GNU C Library
1034 can be used with is 2.6.32.
1036 * Running the testsuite no longer terminates as soon as a test fails.
1037 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1038 with PASS or FAIL lines for individual tests. A summary of the results is
1039 printed, including a list of failing lists, and "make check" exits with
1040 error status if there were any unexpected failures. "make check
1041 stop-on-test-failure=y" may be used to keep the old behavior.
1043 * The am33 port, which had not worked for several years, has been removed
1046 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1047 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1048 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1049 conflicted with POSIX (support for which was removed in 2.19), the
1050 interfaces those macros enabled remain available when compiling with
1051 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1052 test macros defined.
1054 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1056 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1057 This may improve lock scaling of existing programs on TX capable systems.
1058 The lock elision code is only built with --enable-lock-elision=yes and
1059 then requires a GCC version supporting the TX builtins. With lock elision
1060 default mutexes are elided via __builtin_tbegin, if the cpu supports
1061 transactions. By default lock elision is not enabled and the elision code
1064 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1065 copy the path argument. This allowed programs to cause posix_spawn to
1066 deference a dangling pointer, or use an unexpected pathname argument if
1067 the string was modified after the posix_spawn_file_actions_addopen
1070 * All supported architectures now use the main glibc sysdeps directory
1071 instead of some being in a separate "ports" directory (which was
1072 distributed separately before glibc 2.17).
1074 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1075 On configurations that support it (all Linux configurations), it's now
1076 used regardless of the --enable-add-ons switch to configure. It is no
1077 longer possible to build such configurations without pthreads support.
1079 * Locale names, including those obtained from environment variables (LANG
1080 and the LC_* variables), are more tightly checked for proper syntax.
1081 setlocale will now fail (with EINVAL) for locale names that are overly
1082 long, contain slashes without starting with a slash, or contain ".." path
1083 components. (CVE-2014-0475) Previously, some valid locale names were
1084 silently replaced with the "C" locale when running in AT_SECURE mode
1085 (e.g., in a SUID program). This is no longer necessary because of the
1088 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1089 with application code using Intel MPX instructions. (With all previous
1090 versions, the MPX register state could be clobbered when making calls
1091 into or out of a shared library.) Note that while the new dynamic
1092 linker is compatible with all known x86 hardware whether or not it
1093 supports Intel MPX, some x86 instruction-set emulators might fail to
1094 handle the new instruction encodings. This is known to affect Valgrind
1095 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1096 release), and might affect other tools that do instruction emulation.
1098 * Support for loadable gconv transliteration modules has been removed.
1099 The support for transliteration modules has been non-functional for
1100 over a decade, and the removal is prompted by security defects. The
1101 normal gconv conversion modules are still supported. Transliteration
1102 with //TRANSLIT is still possible, and the //IGNORE specifier
1103 continues to be supported. (CVE-2014-5119)
1105 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1106 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1107 resulting a denial-of-service security vulnerability in applications which
1108 use functions related to iconv. (CVE-2014-6040)
1112 * The following bugs are resolved with this release:
1114 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1115 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1116 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1117 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1118 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1119 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1120 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1121 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1122 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1123 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1124 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1125 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1126 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1127 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1128 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1129 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1130 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1131 16431, 16453, 16474, 16506, 16510, 16529
1133 * Slovenian translations for glibc messages have been contributed by the
1134 Translation Project's Slovenian team of translators.
1136 * The public headers no longer use __unused nor __block. This change is to
1137 support compiling programs that are derived from BSD sources and use
1138 __unused internally, and to support compiling with Clang's -fblock
1139 extension which uses __block.
1141 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1142 large collation sequences to optimize multiple passes. This cache
1143 computation may overflow for large collation sequences and may cause a
1144 stack or buffer overflow. This is now fixed to use a slower algorithm
1145 which does not use a cache if there is an integer overflow.
1147 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1148 rules for large collation sequences to optimize multiple passes and falls
1149 back to alloca if malloc fails, resulting in a possible stack overflow.
1150 The implementation now falls back to an uncached collation sequence lookup
1153 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1154 initialized for static applications resulting in the security feature
1155 being disabled. The pointer guard is now correctly initialized to a
1156 random value for static applications. Existing static applications need
1157 to be recompiled to take advantage of the fix (bug 15754).
1159 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1160 to the d_name member of struct dirent, or omit the terminating NUL
1161 character. (Bugzilla #14699).
1163 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1164 aligned_alloc functions could allocate too few bytes or corrupt the
1165 heap when passed very large allocation size values (Bugzilla #15855,
1168 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1169 for AF_INET6 has been fixed (Bugzilla #16072).
1171 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1172 pap_CW, quz_PE, the_NP.
1174 * Substantially revised locales: gd_GB, ht_HT
1176 * The LC_ADDRESS field was updated to support country_car for almost all
1179 * ISO 1427 definitions were updated.
1181 * ISO 3166 definitions were updated.
1183 * The localedef utility now supports --big-endian and --little-endian
1184 command-line options to generate locales for a different system from that
1185 for which the C library was built.
1187 * Binary locale files now only depend on the endianness of the system for
1188 which they are generated and not on other properties of that system. As a
1189 consequence, binary files generated with new localedef may be incompatible
1190 with old versions of the GNU C Library, and binary files generated with
1191 old localedef may be incompatible with this version of the GNU C Library,
1192 in the following circumstances:
1194 + Locale files may be incompatible on m68k systems.
1196 + Locale archive files (but not separate files for individual locales) may
1197 be incompatible on systems where plain "char" is signed.
1199 * The configure option --disable-versioning has been removed. Builds with
1200 --disable-versioning had not worked for several years.
1202 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1203 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1205 * SystemTap probes for malloc have been introduced.
1207 * SystemTap probes for slow multiple precision fallback paths of
1208 transcendental functions have been introduced.
1210 * Support for powerpc64le has been added.
1212 * The soft-float powerpc port now supports e500 processors.
1214 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1216 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1217 set of header declarations that are enabled by default, even when other
1218 feature test macros or compiler options such as -std=c99 would otherwise
1219 disable some of those declarations.
1221 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1222 conflict with POSIX. The libbsd-compat library (which was a dummy library
1223 that did nothing) has also been removed.
1225 * Preliminary documentation about Multi-Thread, Async-Signal and
1226 Async-Cancel Safety has been added.
1228 * Change of ABI data structures for s390 and s390x:
1229 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1230 allow for future hardware extensions. All library functions that accept or
1231 return these structures were versioned in order to provide backwards
1232 compatibility with existing code. However, not all cases can be handled
1233 with interface versioning. If an application embeds either structure into
1234 another structure and that structure is passed to another compilation unit
1235 which expects the newer larger-sized object then any access to the new
1236 fields will result in undefined behaviour. Similarly any access to fields
1237 that were moved by the enlarging of the structures will cause undefined
1238 behaviour. To fix the undefined behaviour all objects that are part of the
1239 final in-memory image must agree on the size of structures, and this may
1240 require recompilation.
1244 * The following bugs are resolved with this release:
1246 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1247 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1248 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1249 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1250 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1251 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1252 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1253 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1254 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1255 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1256 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1257 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1258 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1261 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1262 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1263 Distributions can re-enable building and using pt_chown via the new configure
1264 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1265 considerable security risks and should only be used if the distribution
1266 understands and accepts the risks.
1268 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1271 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1272 fixed (Bugzilla #15330).
1274 * Add support for calling C++11 thread_local object destructors on thread
1275 and program exit. This needs compiler support for offloading C++11
1276 destructor calls to glibc.
1278 * Improved worst case performance of libm functions with double inputs and
1281 * Support for priority inherited mutexes in pthread condition variables on
1282 non-x86 architectures.
1284 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1286 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1288 * Optimized string functions for ARM. Implemented by Will Newton and
1291 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1293 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1296 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1297 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1299 * Added a benchmark framework to track performance of functions in glibc.
1301 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1302 It is based on draft TS 18661 and currently enabled as a GNU extension.
1304 * On Linux, the clock function now uses the clock_gettime system call
1305 for improved precision, rather than old times system call.
1307 * Added support for version-3 tz format files. This is needed when using
1308 the tz database release 2013e or later, and affects a few unusual cases --
1309 currently only TZ='America/Godthab' for time stamps after 2037.
1311 * Added new API functions pthread_getattr_default_np and
1312 pthread_setattr_default_np to get and set the default pthread
1313 attributes of a process.
1315 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1316 This may improve lock scaling of existing programs on TSX capable systems.
1317 When the --enable-lock-elision=yes parameter is specified at configure
1318 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1321 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1322 availability). Implemented by Ryan S. Arnold.
1324 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1328 * The following bugs are resolved with this release:
1330 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1331 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1332 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1333 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1334 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1335 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1336 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1337 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1338 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1339 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1340 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1341 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1342 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1343 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1344 14879, 14889, 14893, 14898, 14914.
1346 * Optimization of memcpy for MIPS.
1348 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1349 EMFILE has been fixed (Bugzilla #14889).
1351 * The add-on ports collection is now distributed in the "ports" subdirectory
1352 of the main GNU C Library distribution, rather than separately.
1354 * Port to ARM AArch64 contributed by Linaro.
1356 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1357 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1359 Implemented by Andreas Krebbel.
1361 * The new function secure_getenv allows secure access to the environment,
1362 returning NULL if running in a SUID/SGID process. This function replaces
1363 the internal function __secure_getenv.
1365 * SystemTap static probes have been added into the dynamic linker.
1366 Implemented by Gary Benson.
1368 * Optimizations of string functions strstr, strcasestr and memmem.
1369 Implemented by Maxim Kuvyrkov.
1371 * The minimum Linux kernel version that this version of the GNU C Library
1372 can be used with is 2.6.16.
1374 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1375 powerpc POWER7. Implemented by Will Schmidt.
1377 * New configure option --disable-nscd builds the C library such that it
1378 never attempts to contact the Name Service Caching Daemon (nscd).
1379 New configure option --disable-build-nscd avoids building nscd itself;
1380 this is the default if --disable-nscd is used.
1382 * Improved support for cross-compilation, including cross-testing and
1383 bootstrap builds without a previously built glibc.
1385 * Several testsuite tests are now able to test multiple IFUNC variants of an
1386 interface, rather than just testing the one that would be chooen by
1389 * New configure options --with-bugurl and --with-pkgversion, for
1390 distributors to use to embed their bug-reporting and package version
1391 information in --help and --version output.
1393 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1394 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1395 allows creation of chroots without the procfs mounted on /proc.
1397 * The `crypt' function now fails if passed salt bytes that violate the
1398 specification for those values. On Linux, the `crypt' function will
1399 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1400 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1401 when the mode is enabled.
1403 * The `clock_*' suite of functions (declared in <time.h>) is now available
1404 directly in the main C library. Previously it was necessary to link with
1405 -lrt to use these functions. This change has the effect that a
1406 single-threaded program that uses a function such as `clock_gettime' (and
1407 is not linked with -lrt) will no longer implicitly load the pthreads
1408 library at runtime and so will not suffer the overheads associated with
1409 multi-thread support in other code such as the C++ runtime library.
1411 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1416 * The following bugs are resolved with this release:
1418 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1419 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1420 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1421 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1422 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1423 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1424 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1425 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1426 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1427 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1428 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1429 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1430 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1431 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1432 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1433 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1434 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1435 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1436 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1437 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1438 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1439 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1442 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1443 configuring glibc with:
1444 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1445 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1446 Implemented by H.J. Lu.
1450 + define static_assert
1452 + do not declare gets
1454 + declare at_quick_exit and quick_exit also for ISO C11
1456 + aligned_alloc. NB: The code is deliberately allows the size parameter
1457 to not be a multiple of the alignment. This is a moronic requirement
1458 in the standard but it is only a requirement on the caller, not the
1461 + timespec_get added
1463 + uchar.h support added
1465 + CMPLX, CMPLXF, CMPLXL added
1467 Implemented by Ulrich Drepper.
1469 * Support for the IA-64 has been moved to ports.
1471 * Remove support for anything but ELF binary format
1473 * Checking versions of poll, ppoll added.
1474 Implemented by Ulrich Drepper.
1476 * More generic and 64-bit performance optimizations to math functions.
1477 Implemented by Ulrich Drepper.
1479 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1480 headers and functions available at compile time as they were before
1481 version 2.14. This option will be removed at some time in the future
1482 after the TI-RPC library becomes fully sufficient for the needs of
1483 existing applications.
1485 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1486 Note that glibc is not expected to work with any Linux kernel version
1489 * New header <sys/auxv.h> and function getauxval allowing easy access to
1490 the AT_* key-value pairs passed from the Linux kernel. The header also
1491 defines the HWCAP_* bits associated with the AT_HWCAP key.
1493 * New locales: mag_IN
1495 * New configure option --enable-systemtap builds SystemTap static probes
1496 into libc for setjmp and longjmp and into libpthread for various operations.
1497 So far the setjmp/longjmp probes and some of the libpthread probes are
1498 provided only for i*86 and x86_64.
1499 Implemented by Roland McGrath and Rayson Ho.
1501 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1503 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1506 * More optimized functions for SPARC. Implemented by David S. Miller.
1508 * Improved support for cross-compilation, especially bootstrap builds
1509 without a previously built glibc.
1511 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1512 Chris Metcalf from Tilera.
1514 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1515 now supported for ARM processors.
1517 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1518 as the name of the dynamic linker, to distinguish it from the
1519 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1521 * Support for CRIS has been removed from ports.
1523 * A new class of installed header has been documented for low-level
1524 platform-specific functionality. PowerPC added the first instance with a
1525 function to provide time base register access. Contributed by Tulio
1526 Magno Quites Machado Filho.
1528 * ix86 configurations now install header files that are consistent with
1529 what x86-64 configurations install. These same header files can be used
1530 for -m32, -m64, or -mx32 builds.
1531 Contributed by H.J. Lu.
1533 * Math library bug fixes. A thorough audit of all open math library bugs was
1534 conducted by Joseph Myers. Significant progress was made on many math
1535 library bugs resulting in more accurate exceptions and function results.
1536 Many thanks to all those that contributed including Andreas Jaeger for his
1537 patch review and work on the x87 trigonometric instruction issues.
1539 * Timezone data is no longer installed. Timezone-related binaries and scripts
1540 will continue to be installed. Users should obtain their timezone data from
1541 their distribution provider or from the tzdata package at
1542 <ftp://munnari.oz.au/pub/>.
1546 * The following bugs are resolved with this release:
1548 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1549 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1550 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1551 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1552 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1553 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1554 13484, 13506, 13515, 13523, 13524, 13538, 13540
1556 * New program pldd to list loaded object of a process
1557 Implemented by Ulrich Drepper.
1559 * Add nss_db support back to glibc. No more dependency on Berkeley db
1560 and support for initgroups lookups.
1561 Implemented by Ulrich Drepper.
1563 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1564 Contributed by HJ Lu.
1566 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1567 Contributed by HJ Lu.
1569 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1570 on x86-32 and x86-64.
1571 Contributed by Liubov Dmitrieva.
1573 * Optimized strchr and strrchr for SSE on x86-32.
1574 Contributed by Liubov Dmitrieva.
1576 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1577 for x86-64 and x86-32.
1578 Contributed by Liubov Dmitrieva.
1580 * New interfaces: scandirat, scandirat64
1581 Implemented by Ulrich Drepper.
1583 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1584 Implemented by Ulrich Drepper.
1586 * nscd now also caches the netgroup database.
1587 Implemented by Ulrich Drepper.
1589 * Integrate libm with gcc's -ffinite-math-only option.
1590 Implemented by Ulrich Drepper.
1592 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1593 to math functions. Implemented by Ulrich Drepper.
1595 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1596 Implemented by Ulrich Drepper.
1598 * New Linux interfaces: process_vm_readv, process_vm_writev
1600 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1601 Implemented by Ulrich Drepper.
1603 * Optimized nearbyint and strcasecmp for PPC.
1604 Implemented by Adhemerval Zanella.
1606 * New locales: bho_IN, unm_US, es_CU, ta_LK
1610 * The following bugs are resolved with this release:
1612 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1613 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1614 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1615 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1616 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1617 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1618 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1619 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1620 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1621 12795, 12811, 12813, 12814, 12841
1623 * The RPC implementation in libc is obsoleted. Old programs keep working
1624 but new programs cannot be linked with the routines in libc anymore.
1625 Programs in need of RPC functionality must be linked against TI-RPC.
1626 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1628 Visible changes of this change include (obviously) the inability to link
1629 programs using RPC functions without referencing the TI-RPC library and the
1630 removal of the RPC headers from the glibc headers.
1631 Implemented by Ulrich Drepper.
1633 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1634 syncfs, setns, sendmmsg
1636 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1637 yue_HK, lij_IT, mhr_RU
1639 * New encodings: CP770, CP771, CP772, CP773, CP774
1641 * New program sotruss to trace calls through PLTs
1642 Implemented by Ulrich Drepper.
1644 * The malloc hook implementation is marked deprecated and will be removed
1645 from the default implementation in the next version. The design never
1646 worked ever since the introduction of threads. Even programs which do
1647 not create threads themselves can use multiple threads created internally.
1651 * The following bugs are resolved with this release:
1653 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1654 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1655 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1656 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1659 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1661 * POWER7 optimizations: memset, memcmp, strncmp
1663 * New optimized string functions for x86-64: strnlen (SSE2),
1664 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1665 Implemented by Ulrich Drepper.
1669 * The following bugs are resolved with this release:
1671 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1672 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1673 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1674 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1675 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1676 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1677 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1678 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1680 * New interfaces: pthread_getname_np, pthread_setname_np
1682 * New Linux interface: recvmmsg
1684 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1686 * The dynamic linker now recognizes supported ABI versions from the
1687 EI_ABIVERSION field in the ELF header.
1688 Implemented by Ulrich Drepper.
1690 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1691 will not be used to fill in password fields in the passwd.byname replies.
1692 Instead it is used to synthesize the shadow.byname table, should it be
1693 missing. This is a useful mode in some installations involving Solaris.
1694 Implemented by Ulrich Drepper.
1696 * New locales: kok_IN, sq_MK, cv_RU
1700 * The following bugs are resolved with this release:
1702 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1703 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1704 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1705 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1706 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1707 10789, 10817, 10823, 10840, 10847
1709 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1711 Implemented by Ulrich Drepper.
1713 * Checking version of longjmp added that fails if an uninitialized stack
1714 frame would be created. Implemented by Ulrich Drepper.
1716 * STT_GNU_IFUNC is now supported in static executables.
1717 Implemented by H.J. Lu.
1719 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1721 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1722 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1723 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1724 Contributed by H.J. Lu.
1726 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1727 Implemented by Ulrich Drepper.
1729 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1731 Contributed by H.J. Lu.
1733 * Support for fma instruction in AVX on x86-64.
1734 Implemented by H.J. Lu and Ulrich Drepper.
1736 * AVX support in x86-64 auditing support in ld.so.
1737 Implemented by H.J. Lu.
1739 * STB_GNU_UNIQUE support added.
1740 Implemented by Ulrich Drepper.
1742 * Implement second fallback mode for optimized DNS lookup for even more
1743 broken environments. If two requests from the same port are not
1744 handled correctly close the socket and open a new one before sending
1745 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1746 can be used to select this mode right away, instead of rediscovering the
1747 necessity is every process again.
1748 Implemented by Ulrich Drepper.
1750 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1751 Implemented by Adam Tkac.
1753 * Optimized iconv conversions for S390x.
1754 Implemented by Andreas Krebbel.
1756 * Using condvars with PI mutexes is now more efficient due to kernel
1757 support for requeueing to PI futexes. NPTL support added for x86-64.
1758 Implemented by Ulrich Drepper.
1760 * New locales: ps_AF, my_MM
1764 * The following bugs are resolved with this release:
1766 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1767 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1768 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1769 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1770 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1772 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1773 Implemented by Ulrich Drepper.
1775 * New Linux interfaces: accept4, fallocate, fallocate64.
1776 Implemented by Ulrich Drepper.
1778 * Correct declarations of string function when used in C++ code. This
1779 could lead to compile errors for invalid C++ code.
1781 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1784 * New POSIX 2008 interface: psiginfo
1785 Implemented by Ulrich Drepper.
1787 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1788 Implemented by Ulrich Drepper.
1790 * Support for selecting between multiple function definitions at runtime
1791 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1793 * The libcrypt library can now use the hash function implementations in
1794 NSS. Implemented by Ulrich Drepper.
1796 * The malloc implementation can be compiled to be less memory efficient
1797 but higher performing in multi-threaded programs.
1798 Implemented by Ulrich Drepper.
1800 * New locales: nan_TW@latin, ks_IN
1802 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1803 Implemented by Ulrich Drepper.
1805 * Extended printf hook support. It is possible to use user-defined types
1806 and extend existing format specifiers.
1807 Implemented by Ulrich Drepper.
1809 * Handling for group shadow files has been added.
1810 Implemented by Ulrich Drepper.
1812 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1813 servers (the case, e.g., for some people using the built-in DNS
1814 server in ADSL modems/routers). There is a once-per-process timeout
1815 in case of a broken server. To avoid it, users can run nscd or put
1816 'options single-request' in /etc/resolv.conf.
1817 Implemented by Ulrich Drepper.
1821 * The following bugs are resolved with this release:
1823 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1824 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1825 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1826 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1827 6968, 6974, 6980, 6995, 7008, 7009, 7029
1829 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1830 up at the same time. Implemented by Ulrich Drepper.
1832 * TLS descriptors for LD and GD on x86 and x86-64.
1833 Implemented by Alexandre Oliva.
1835 * getaddrinfo now handles DCCP and UDPlite.
1836 Implemented by Ulrich Drepper.
1838 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1839 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1840 Implemented by Ulrich Drepper.
1842 * New implementation of memmem, strstr, and strcasestr which is O(n).
1843 Implemented by Eric Blake.
1845 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1847 * Implement "e" option for popen to open file descriptor with the
1848 close-on-exec flag set. Implemented by Ulrich Drepper.
1850 * Implement "b" mode for fmemopen. In this mode writes writes don't
1851 implicitly add a NUL byte and seeks from the end of the buffer really
1852 use the buffer end, not the string length as the basis.
1853 Implemented by Ulrich Drepper.
1855 * Many functions, exported and internal, now atomically set the close-on-exec
1856 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1858 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1860 Implemented by Pravin Satpute.
1862 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1866 * The following bugs are resolved with this release:
1868 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1869 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1870 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1871 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1872 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1873 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1874 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1876 * New locales: bo_CN, bo_IN, shs_CA.
1878 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1880 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1881 Implemented by Pravin Satpute.
1883 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1885 * nscd caches now all timeouts for DNS entries
1886 Implemented by Ulrich Drepper.
1888 * nscd is more efficient and wakes up less often.
1889 Implemented by Ulrich Drepper.
1891 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1892 vdprintf, and obstack_vprintf.
1893 Implemented by Jakub Jelinek.
1895 * Faster memset for x86-64.
1896 Implemented by Harsha Jagasia and H.J. Lu.
1898 * Faster memcpy on x86.
1899 Implemented by Ulrich Drepper.
1901 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1902 Implemented by Ulrich Drepper.
1904 * Faster sqrt and sqrtf implemention for some PPC variants.
1905 Implemented by Steven Munroe.
1909 * The following bugs are resolved with this release:
1911 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1912 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1913 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1914 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1915 5071, 5103, 5104, 5112, 5113, 5184, 5186
1917 * More checking functions: fread, fread_unlocked, open*, mq_open.
1918 Implemented by Jakub Jelinek and Ulrich Drepper.
1920 * Extend fortification to C++. Implemented by Jakub Jelinek.
1922 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1923 by not recognizing 'a' as a modifier when those specs are requested.
1924 Implemented by Jakub Jelinek.
1926 * PPC optimizations to math and string functions.
1927 Implemented by Steven Munroe.
1929 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1930 options to be passed. Implemented by Ulrich Drepper.
1932 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1934 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1935 Implemented by Ulrich Drepper.
1937 * Handle private futexes in the NPTL implementation.
1938 Implemented by Jakub Jelinek and Ulrich Drepper.
1940 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1941 Implemented by Roland McGrath and Ulrich Drepper.
1943 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1945 * SHA-256 and SHA-512 based password encryption.
1946 Implemented by Ulrich Drepper.
1948 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1949 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1952 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1953 Implemented by Ulrich Drepper.
1957 * The following bugs are resolved with this release:
1959 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
1960 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
1961 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
1962 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
1963 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
1964 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
1965 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
1968 * New Linux interfaces: epoll_pwait, sched_getcpu.
1970 * New generic interfaces: strerror_l.
1972 * nscd can now cache the services database. Implemented by Ulrich Drepper.
1976 * The following bugs are resolved with this release:
1978 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
1979 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
1980 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
1981 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
1982 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
1983 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
1984 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
1985 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
1986 3225, 3251, 3252, 3253, 3273
1988 * For Linux, the sorting of addresses returned by getaddrinfo now also
1989 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
1990 Implemented by Ulrich Drepper.
1992 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
1993 Implemented by Ulrich Drepper.
1995 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
1997 * New iconv module for MIK. Contributed by Alexander Shopov.
1999 * For sites with broken group and/or passwd database, the auto-propagate
2000 option of nscd can prevent creating ID lookup entries from the results
2001 of a name lookup and vice versa. This usually is no problem but some
2002 site might have problems with the default behavior.
2003 Implemented by Ulrich Drepper.
2005 * Iterating over entire database in NIS can be slow. With the
2006 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2007 to trade time for memory. The entire database will be read at once.
2008 Implemented by Ulrich Drepper.
2010 * The interfaces introduced in RFC 3542 have been implemented by
2013 * Support for the new ELF hash table format was added by Ulrich Drepper.
2015 * Support for priority inheritance mutexes added by Jakub Jelinek and
2018 * Support for priority protected mutexes added by Jakub Jelinek.
2020 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2021 or_IN, csb_PL, fy_NL, sr_ME.
2025 * More overflow detection functions.
2027 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2028 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2030 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2031 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2032 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2033 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2034 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2035 by Masahide Washizawa.
2037 * It is now possible to install audit libraries for the dynamic linker using
2038 LD_AUDIT. Implemented by Ulrich Drepper.
2040 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2041 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2042 For a libc and libpthread that works well on Linux 2.4 kernels, we
2043 recommend using the stable 2.3 branch.
2045 * The new function open_wmemstream acts like open_memstream,
2046 but uses a wchar_t wide character buffer.
2048 * The new function ppoll is an improved version of poll, similar to pselect.
2050 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2051 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2052 renameat, unlinkat, linkat, symlinkat, readlinkat.
2054 * New Linux kernel system calls: unshare,
2055 inotify_init, inotify_add_watch, inotify_rm_watch.
2057 * The euidaccess function is now also known by the alias eaccess,
2058 for compatibility with some other systems.
2060 * Timezone data updated to 2006b version.
2064 * The following bugs are resolved with this release:
2066 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2067 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2068 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2069 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2070 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2071 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2073 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2075 * As of this release, GCC 4 can be used to compile the C Library.
2077 * Timezone data updated to 2005m version.
2081 * The following bugs are resolved with this release:
2083 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2084 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2085 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2086 777, 787, 821, 822, 823, 825
2088 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2092 * Support for RFC 3678. Real implementations exist only for Linux so far.
2093 Implemented by Ulrich Drepper.
2095 * nscd can now cache entries persistently. Expiring entries are reloaded.
2096 For speedups the cache can be shared in memory with client processes.
2097 Implemented by Ulrich Drepper.
2099 * nscd can now perform SELinux checks.
2100 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2102 * getaddrinfo queries are now cached. Canonical name lookup is performed
2104 Implemented by Ulrich Drepper.
2106 * The nothrow function attribute is used when headers are used by gcc when
2107 compiling C code. This can avoid significant amounts of exception
2110 * The malloc functions perform more error checking and are stricter when
2111 it comes to reacting on errors. The default action is to terminate
2112 the process after showing an error message. Implemented by Ulrich Drepper.
2114 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2115 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2117 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2118 namespaces using the new function dlmopen(). This feature is of course,
2119 like most other dynamic loading functionality, not available in statically
2120 linked applications. Implemented by Ulrich Drepper.
2122 * Low-overhead boundary checking variants of string and some stdio functions
2123 were added. These are to be used in conjunction with a gcc patch by
2124 Jakub Jelinek which adds calls to these functions if possible.
2125 Implemented by Jakub Jelinek and Ulrich Drepper.
2127 * Old code for several operating systems and machine architectures that
2128 have not been in working condition in a long time have been removed from
2129 the main source tree maintained by the GNU C Library's maintainers.
2130 These files are now reside in the separate `ports' source module
2131 that is usable as an add-on when building the library.
2135 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2136 interrogate the dynamic linker, compatible with the Solaris interface.
2138 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2139 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2141 * getifaddrs now uses the netlink interface on Linux to get its information.
2142 Implemented by Thorsten Kukuk.
2144 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2145 Implemented by Ulrich Drepper.
2147 * support for non-executable stacks on x86 has been added. Changes mostly
2150 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2153 * getaddrinfo now performs destination address selection according to
2158 * Thread-safe interfaces for many functions that access locale data
2159 were added in version 2.3, but these features were omitted from NEWS.
2160 Many functions have variants with an `_l' suffix that take a `locale_t'
2161 object as a parameter rather than consulting the current locale.
2162 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2163 create and maintain `locale_t' objects. Additionally, the new function
2164 `uselocale' sets "the current locale" (as used by functions not so
2165 parameterized) set for an individual thread. These features were added
2166 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2168 * The functions getresuid, getresgid, setresuid, and setresgid, which
2169 have long been available on Linux, are now declared in <unistd.h>
2170 and are now also available on the Hurd.
2172 * ELF thread-local storage support (TLS) now works on x86-64.
2174 * The new dynamic string token $LIB is expanded in shared library names.
2175 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2177 * Aldy Hernandez contributed complete software floating point support for
2178 PowerPC machines with no FPU.
2180 * fexecve is implemented on Linux.
2182 * The `btowc' function should work at least twice as fast due to
2183 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2185 * With approriate thread add-ons cancelable functions are now implemented
2186 in libc.so as well. No need to call the function in libpthread. This
2187 change allowed to finally disable the incorrect and expensive handling
2188 of weak definition in ld.so.
2190 * Yet more PLT entries in libc.so have been removed. We finally arrived
2191 at the bare minimum. Startup times improved appropriately.
2193 * Support for the new Linux/x86 system call interface was added. The
2194 AT_SYSINFO auxiliary vector entry is recognized and handled.
2198 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2201 * iconv (the program and the interface) now accepts empty names (excluding
2202 options like //TRANSLIT) to mean "use charset of current locale".
2204 * localedef can now transliterate characters in strings which are not in
2205 the provided charmap. The information from the input locale is used.
2207 * Prelinking support was added for ELF targets. This requires additional
2208 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2210 * Read-only stdio streams now use mmap to speed up operation by eliminating
2211 copying and buffer underflows. To use add 'm' to the mode string of
2212 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2214 * The malloc functions were completely rewritten by Wolfram Gloger based
2215 on Doug Lea's malloc-2.7.0.c.
2217 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2218 implementation of regex.
2220 * Bruno Haible upgraded the iconv and locale implementation to support
2223 * Contents of the LC_* and LANG environment variables in the CEN style are
2224 not recognized anymore. It never was used. Change by Ulrich Drepper.
2226 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2227 thread-local storage (TLS) ABI on some platforms.
2228 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2230 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2231 EUC-JISX0213, and TSCII.
2233 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2234 BSD-compatible interface for getting all network interface addresses.
2235 Implementation for IPv4 by Roland McGrath.
2237 * Loading of locale data is faster due to the introduction of a locale
2238 archive. Implemented by Roland McGrath and Ulrich Drepper.
2240 * Startup times are significantly reduced by not using exported functions
2241 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2244 * Steven Munroe contributed a port to PowerPC64/Linux.
2248 * The Hurd now uses the GNU libio implementation of stdio rather than the
2249 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2251 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2252 with the same behavior as the Linux system call of the same name.
2256 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2257 128-bit long double format.
2259 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2260 IBM1161, and IBM1162 charsets.
2262 * Andreas Jaeger contributed a port to x86-64/Linux.
2264 * Peter Bruin contributed a port to PowerPC/Hurd.
2266 * libc functions using I/O streams now can handle wide-oriented streams
2269 * optimizations in the dynamic linker. Binaries created by recent binutils
2270 versions start up quicker due to reduced time spent on relocations.
2272 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2276 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2277 asinh, atanh, j0 for the 128-bit long double format.
2279 * Bruno Haible updated all the code handling Unicode in some form to
2280 support Unicode 3.1.
2282 * Speed of regex for single-byte locales is back to previous levels.
2283 Patch by Isamu Hasegawa.
2285 * Alpha, SPARC, and IA-64 now also using floating stacks.
2287 * Startup time of internationalized applications greatly improved through
2288 iconv cache. Use iconvconfig to generate the cache file.
2289 Contributed by Ulrich Drepper.
2291 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2292 severe performance problems. Patch by David Mosberger.
2294 * The Hurd port got a lot more functionality like AIO, various stdio
2295 extensions, etc. Mainly done by Roland McGrath.
2297 * mtrace can now lookup symbols in shared libraries.
2301 * Intel's IA-64 math library is largely integrated. It provides fast and
2302 accurate implementatations for most basic and standard math functions
2303 in float, double, and long double format.
2305 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2306 and asin for the 96-bit long double format and asin, log, tan for the
2307 128-bit long double format.
2309 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2310 The basic double functions exist today. Contributed by Abraham Ziv
2311 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2312 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2314 * An asynchronous name lookup library was added. The interface is designed
2315 after POSIX AIO. The proposal was circulated beforehand to get comments.
2316 No negative ones came in. Implemented by Ulrich Drepper.
2318 * Port to S390/64bit contributed by Martin Schwidefsky
2319 <schwidefsky@de.ibm.com>.
2321 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2322 of functions for Linux/IA-64.
2324 * The RPC code is now thread safe. Threads can now use the same service
2325 of different services at the same time. Patch by Eric Norum
2326 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2328 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2329 family of functions for Linux/S390.
2331 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2332 of functions for Linux/x86.
2334 * Port to Linux/CRIS contributed by Axis Communications.
2338 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2339 we can now check for namespace violations and missing declarations. The
2340 result is that almost all headers are now Unix-compliant (as defined in
2341 the upcoming XPG6). The negative side is that some programs might need
2342 corrections, too, if they depend on the incorrect form of the headers in
2343 previous versions which defined too many symbols and included too many
2346 * regex now handles multibyte character sets correctly.
2347 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2349 * iconv (the program) does now conform to the upcoming XPG6 and handles
2350 charmaps. Instead of the charset names the path of charmaps can be
2351 provided and the conversion happens based on this data.
2352 Contributed by Ulrich Drepper.
2354 * The locale program now provides detailed information about the installed
2359 only lists the names of the supported locales
2363 provides details such as country, language, and codeset name.
2364 Contributed by Ulrich Drepper.
2368 * The gencat program now parses the input file according to the charset
2369 selected by the LC_CTYPE category. This is important for stateful
2370 character sets. To make generating catalogs easier there is a way
2371 to overwrite the charset selected by the locale: before the first
2372 message or $ quote line the catalog can contain a line like
2374 $ codeset=ISO-8859-2
2376 to select the charset (ISO-8859-2 in this case).
2378 Implemented by Shinya Hanataka and Ulrich Drepper.
2380 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2381 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2382 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2384 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2385 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2387 * The LANGUAGE environment variable is now ignored unless the locale is
2388 changed from the default "C" locale.
2390 * The usual bug fixes.
2394 * Greg McGary added runtime support for bounds checking using gcc's
2395 new -fbounded-pointers option. ix86 target is complete. PowerPC
2398 * Thorsten Kukuk added secure mode support to nscd.
2400 * The Berkeley db libraries have been removed.
2402 Related, the nss_db module is now in a separate package since it
2403 obviously requires a database library being available.
2405 * Wide character I/O streams implemented by Ulrich Drepper.
2407 * Functions from the extended socket API added by Ulrich Drepper.
2409 * Functions feenableexcept and fedisableexcept to control the
2410 behaviour of individual exceptions have been added by Andreas Jaeger.
2412 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2414 * The resolver code has been updated from bind 8.2.3-T5B which supports
2415 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2418 This change could in some situations effect backward compatibility. Since
2419 now `_res' is a thread-local instead of a global variable, modifying it
2420 in one thread does not have any effect in other threads.
2422 The resolver library was also extended to allow IPv6 as the transport
2423 protocol for the requests. This work was done by Stig Venaas.
2425 * Compatibility code for K&R C compilers has been removed from the
2426 header files. A ISO C compiler is needed to use the library
2427 (conforming to either C89 or C99 standard).
2429 * Complete rewrite of the localedef program to support multibyte character
2430 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2431 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2432 Implemented by Ulrich Drepper.
2434 Bruno Haible significantly improved the generation and use of the data
2435 structures for the wide character tables.
2437 * Plural handling in gettext implemented by Ulrich Drepper.
2439 * The utmp daemon has been removed.
2441 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2443 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2446 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2448 * POSIX barriers implemented by Kaz Kylheku.
2450 * POSIX spawn function family implemented by Ulrich Drepper.
2452 * POSIX spinlocks are now available.
2454 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2456 * the configure option --enable-kernel=X.Y.Z allows to strip out
2457 compatibility for kernel versions before X.Y.Z. This is currently only
2458 implemented for Linux.
2460 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2461 field sin6_scope_id. This means that all programs using IPv6 should be
2462 recompiled. Don't expect binary compatibility with previous glibc
2465 * various conversion modules for IBM character sets contributed by
2468 * IA-64 port by Jes Sorensen and HJ Lu.
2482 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2484 * New xdr functions are added; some rpc functions are now 64bit clean.
2486 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2488 * Fixed known incompatibilities with glibc 2.0.
2490 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2492 * Optimized a number of functions (especially the ELF dynamic loader).
2494 * Update timezone data files.
2496 * lots of charmaps corrections
2498 * some new locale definitions and charmaps
2503 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2504 conform to POSIX member type requirements. Symbol versions have been
2505 adjusted as needed within the library, and for direct use by applications,
2506 but there is potential for problems if third-party libraries use
2507 struct timeval as part of their interface. This does not present
2508 a problem for X and other "essential" system libraries.
2510 * An additional locale model to support C++ Standard Library locale
2511 model and probably more was implemented by Ulrich Drepper.
2513 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2516 * Miles Bader provided the `argp' function family to support hierachical
2517 command line argument parsing, layered on top of getopt.
2519 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2521 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2524 * scanf recognizes the %a and %A format for scanning floating point numbers.
2526 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2527 information and interfaces for the available integer types.
2529 * about 130 new math functions were added to implement the ISO C9x math
2532 * the new header <complex.h> contains definitions of the complex math
2533 functions from ISO C 9X.
2535 * the new header <tgmath.h> defines generic macros to use complex or
2536 real valued functions.
2538 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2540 * Andreas Jaeger provided a test suite for the math library.
2542 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2544 * Ulrich Drepper added character set conversion functions (iconv).
2546 * Optimized string functions have been added.
2548 * The localedata addon is now part of glibc.
2550 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2552 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2553 daemon for NSS (nscd).
2555 Missing a better place here are some numbers on improvements. Under
2556 Linux 2.1.125 un-tar-ing the kernel sources takes
2560 using local files 12.19s 6.88s 22.91s
2562 using NIS 13.92s 8.91s 26.34s
2564 using NIS & nscd 10.37s 7.34s 25.30s
2566 using NIS+ 27.57s 30.37s 640.46s
2568 using NIS+ & nscd 10.25s 7.83s 26.51s
2570 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2572 Keep in mind that non-namelookup related operations dominate above times.
2573 It was just a common complain that using NIS+ unpacking the kernel is
2576 [1] The old nscd implementation is not available anymore since it was
2577 distributed with glibc up to version 2.0.98 and thus is now replaced.
2579 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2581 * Mark Kettenis provided a Hesiod NSS module.
2583 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2584 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2586 * rcmd can now handle netgroups (Dick Streefland).
2588 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2591 * Support for the IPv6 protocol has been added to the socket API, as per the
2592 latest draft standards.
2594 * Support for Linux 2.2 has been added.
2596 * Interface changes relative to the latest 2.0.x release:
2597 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2598 addseverity NEW: Unix98
2599 alphasort64 NEW: LFS
2600 argp_err_exit_status NEW: argp, GNU ext
2601 argp_error NEW: argp, GNU ext
2602 argp_failure NEW: argp, GNU ext
2603 argp_help NEW: argp, GNU ext
2604 argp_parse NEW: argp, GNU ext
2605 argp_program_bug_address NEW: argp, GNU ext
2606 argp_program_version NEW: argp, GNU ext
2607 argp_program_version_hook NEW: argp, GNU ext
2608 argp_state_help NEW: argp, GNU ext
2609 argp_usage NEW: argp, GNU ext
2610 authdes_create NEW: Secure RPC
2611 authdes_getucred NEW: Secure RPC
2612 authdes_pk_create NEW: Secure RPC
2613 backtrace NEW: GNU ext.
2614 backtrace_symbols NEW: GNU ext.
2615 backtrace_symbols_fd NEW: GNU ext.
2617 cacosf NEW: ISO C 9x
2618 cacosh NEW: ISO C 9x
2619 cacoshf NEW: ISO C 9x
2620 cacoshl NEW: ISO C 9x
2621 cacosl NEW: ISO C 9x
2628 casinf NEW: ISO C 9x
2629 casinh NEW: ISO C 9x
2630 casinhf NEW: ISO C 9x
2631 casinhl NEW: ISO C 9x
2632 casinl NEW: ISO C 9x
2634 catanf NEW: ISO C 9x
2635 catanh NEW: ISO C 9x
2636 catanhf NEW: ISO C 9x
2637 catanhl NEW: ISO C 9x
2638 catanl NEW: ISO C 9x
2639 cbc_crypt NEW: Secure RPC
2643 ccoshf NEW: ISO C 9x
2644 ccoshl NEW: ISO C 9x
2650 cimagf NEW: ISO C 9x
2651 cimagl NEW: ISO C 9x
2652 clearerr_locked REMOVED
2653 clntunix_create NEW: sunrpc ext
2655 clog10 NEW: ISO C 9x
2656 clog10f NEW: ISO C 9x
2657 clog10l NEW: ISO C 9x
2667 cprojf NEW: ISO C 9x
2668 cprojl NEW: ISO C 9x
2670 crealf NEW: ISO C 9x
2671 creall NEW: ISO C 9x
2676 csinhf NEW: ISO C 9x
2677 csinhl NEW: ISO C 9x
2680 csqrtf NEW: ISO C 9x
2681 csqrtl NEW: ISO C 9x
2685 ctanhf NEW: ISO C 9x
2686 ctanhl NEW: ISO C 9x
2688 des_setparity NEW: Secure RPC
2689 ecb_crypt NEW: Secure RPC
2690 endutxent NEW: Unix98
2692 exp10f NEW: ISO C 9x
2693 exp10l NEW: ISO C 9x
2697 fattach NEW: STREAMS
2698 fdetach NEW: STREAMS
2702 feclearexcept NEW: ISO C 9x
2703 fegetenv NEW: ISO C 9x
2704 fegetexceptflag NEW: ISO C 9x
2705 fegetround NEW: ISO C 9x
2706 feholdexcept NEW: ISO C 9x
2708 feraiseexcept NEW: ISO C 9x
2709 ferror_locked REMOVED
2710 fesetenv NEW: ISO C 9x
2711 fesetexceptflag NEW: ISO C 9x
2712 fesetround NEW: ISO C 9x
2713 fetestexcept NEW: ISO C 9x
2714 feupdateenv NEW: ISO C 9x
2715 fflush_locked REMOVED
2719 fgets_unlocked NEW: GNU ext.
2720 fileno_locked REMOVED
2732 fputc_locked REMOVED
2733 fputs_unlocked NEW: GNU ext.
2734 fread_unlocked NEW: GNU ext.
2739 fstatvfs NEW: Unix98
2743 ftruncate64 NEW: LFS
2745 fwrite_unlocked NEW: GNU ext.
2746 gai_strerror NEW: IPv6
2750 getchar_locked REMOVED
2752 getdate_err NEW: Unix98
2753 getdate_r NEW: GNU ext.
2755 getnameinfo NEW: IPv6
2756 getnetname NEW: Secure RPC
2757 getpmsg NEW: STREAMS
2758 getpt NEW: Unix98 PTY
2759 getrlimit64 NEW: LFS
2760 getutxent NEW: Unix98
2761 getutxid NEW: Unix98
2762 getutxline NEW: Unix98
2763 glob64 NEW: GNU ext.
2764 globfree64 NEW: GNU ext.
2765 gnu_get_libc_release NEW: GNU ext.
2766 gnu_get_libc_version NEW: GNU ext.
2767 grantpt NEW: Unix98 PTY
2768 host2netname NEW: Secure RPC
2770 iconv_close NEW: iconv
2771 iconv_open NEW: iconv
2772 if_freenameindex NEW: IPv6
2773 if_indextoname NEW: IPv6
2774 if_nameindex NEW: IPv6
2775 if_nametoindex NEW: IPv6
2776 in6addr_any NEW: IPv6
2777 in6addr_loopback NEW: IPv6
2778 inet6_isipv4mapped NEW: IPv6
2779 isastream NEW: STREAMS
2780 iswblank NEW: GNU ext.
2781 key_decryptsession NEW: Secure RPC
2782 key_decryptsession_pk NEW: Secure RPC
2783 key_encryptsession NEW: Secure RPC
2784 key_encryptsession_pk NEW: Secure RPC
2785 key_gendes NEW: Secure RPC
2786 key_get_conv NEW: Secure RPC
2787 key_secretkey_is_set NEW: Secure RPC
2788 key_setnet NEW: Secure RPC
2789 key_setsecret NEW: Secure RPC
2790 llrint NEW: ISO C 9x
2791 llrintf NEW: ISO C 9x
2792 llrintl NEW: ISO C 9x
2793 llround NEW: ISO C 9x
2794 llroundf NEW: ISO C 9x
2795 llroundl NEW: ISO C 9x
2800 lrintf NEW: ISO C 9x
2801 lrintl NEW: ISO C 9x
2802 lround NEW: ISO C 9x
2803 lroundf NEW: ISO C 9x
2804 lroundl NEW: ISO C 9x
2806 makecontext NEW: Unix98
2807 mempcpy NEW: GNU ext.
2810 modify_ldt NEW: kernel
2814 nearbyint NEW: ISO C 9x
2815 nearbyintf NEW: ISO C 9x
2816 nearbyintl NEW: ISO C 9x
2817 netname2host NEW: Secure RPC
2818 netname2user NEW: Secure RPC
2819 nexttoward NEW: ISO C 9x
2820 nexttowardf NEW: ISO C 9x
2821 nexttowardl NEW: ISO C 9x
2825 passwd2des NEW: Secure RPC
2827 pow10f NEW: GNU ext.
2828 pow10l NEW: GNU ext.
2831 printf_size NEW: GNU ext.
2832 printf_size_info NEW: GNU ext.
2833 profil_counter REMOVED
2834 pthread_mutexattr_getkind_np REPLACED
2835 pthread_mutexattr_setkind_np REPLACED
2836 ptsname NEW: Unix98 PTY
2837 ptsname_r NEW: Unix98 PTY
2839 putchar_locked REMOVED
2840 putgrent NEW: GNU ext.
2842 putpmsg NEW: STREAMS
2843 pututxline NEW: Unix98
2847 readdir64_r NEW: LFS
2848 remquo NEW: ISO C 9x
2849 remquof NEW: ISO C 9x
2850 remquol NEW: ISO C 9x
2852 roundf NEW: ISO C 9x
2853 roundl NEW: ISO C 9x
2855 scalbln NEW: ISO C 9x
2856 scalblnf NEW: ISO C 9x
2857 scalblnl NEW: ISO C 9x
2859 sendfile NEW: kernel
2860 setcontext NEW: Unix98
2861 setrlimit64 NEW: LFS
2862 setutxent NEW: Unix98
2864 sigignore NEW: Unix98
2865 sigqueue NEW: POSIX.1b
2866 sigrelse NEW: Unix98
2867 sigset NEW: POSIX.1b
2868 sigtimedwait NEW: POSIX.1b
2869 sigwaitinfo NEW: POSIX.1b
2870 sincos NEW: GNU ext.
2871 sincosf NEW: GNU ext.
2872 sincosl NEW: GNU ext.
2876 strcasestr NEW: GNU ext.
2877 strtoimax NEW: ISO C 9x
2878 strtoumax NEW: ISO C 9x
2879 strverscmp NEW: GNU ext.
2880 svcauthdes_stats NEW: Secure RPC
2881 svcunix_create NEW: sunrpc etx
2882 svcunixfd_create NEW: sunrpc ext
2883 swapcontext NEW: Unix98
2884 tcgetsid NEW: Unix98 PTY
2885 tdestroy NEW: GNU ext.
2886 tgamma NEW: ISO C 9x
2887 tgammaf NEW: ISO C 9x
2888 tgammal NEW: ISO C 9x
2892 truncf NEW: ISO C 9x
2893 truncl NEW: ISO C 9x
2895 unlockpt NEW: Unix98 PTY
2896 updwtmpx NEW: Unix98
2897 user2netname NEW: Secure RPC
2898 utmpxname NEW: Unix98
2899 versionsort NEW: GNU ext.
2900 versionsort64 NEW: GNU ext.
2902 wcscasecmp NEW: GNU ext.
2903 wcsncasecmp NEW: GNU ext.
2904 wcsnlen NEW: GNU ext.
2905 wcstoimax NEW: ISO C 9x
2906 wcstoll NEW: ISO C 9x
2907 wcstoull NEW: ISO C 9x
2908 wcstoumax NEW: ISO C 9x
2910 wordexp NEW: POSIX.2
2911 wordfree NEW: POSIX.2
2912 write_profiling REMOVED
2913 xdecrypt NEW: Secure RPC
2914 xdr_authdes_cred NEW: Secure RPC
2915 xdr_authdes_verf NEW: Secure RPC
2916 xdr_cryptkeyarg NEW: Secure RPC
2917 xdr_cryptkeyarg2 NEW: Secure RPC
2918 xdr_cryptkeyres NEW: Secure RPC
2919 xdr_getcredres NEW: Secure RPC
2920 xdr_key_netstarg NEW: Secure RPC
2921 xdr_key_netstres NEW: Secure RPC
2922 xdr_keybuf NEW: Secure RPC
2923 xdr_keystatus NEW: Secure RPC
2924 xdr_netnamestr NEW: Secure RPC
2925 xdr_sizeof NEW: Secure RPC
2926 xdr_unixcred NEW: sunrpc ext
2927 xencrypt NEW: Secure RPC
2928 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2939 * inet_ntoa is thread-safe
2941 * updwtmp is moved from libutil to libc
2943 * rewrite of cbrt function
2945 * update of timezone data
2959 * add atoll function
2961 * fix complex problems in Berkeley DB code
2963 * fix math functions
2967 * fixed lots of header problems (especially Linux/GNU specific)
2969 * dynamic loader preserves all registers
2971 * Roland McGrath provided support for handling of auxiliary objects in
2972 the ELF dynamic loader.
2974 * support for parallel builds is improved
2978 * GNU extensions are no longer declared by default. To enable them you
2979 must define the macro `_GNU_SOURCE' in your program or compile with
2982 * The library has changed from using GNU ld symbol aliases to using weak
2983 symbols where available. The ELF object file format supports weak
2984 symbols; GNU ld also supports weak symbols in the a.out format. (There
2985 is also now support for other GNU ld extensions in ELF. Use the
2986 `--with-elf' option to configure to indicate you have ELF, and
2987 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
2988 of many files which contained only symbol aliases, reducing the size of
2989 the source and the compiled library; many other files were renamed to
2990 less cryptic names previously occupied by the symbol alias files.
2991 There is a new header file <elf.h> for programs which operate on
2992 files in the ELF format.
2994 * Converted to Autoconf version 2, so `configure' has more options.
2995 Run `configure --help' to see the details.
2997 * The library can now be configured to build profiling, highly-optimized
2998 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
2999 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3000 `configure' enable building these extra libraries. The shared library is
3001 built by default when using both ELF and GNU ld. When shared libraries
3002 are enabled, the new library `-ldl' is available for arbitrary run-time
3003 loading of shared objects; its interface is defined in <dlfcn.h>. The
3004 new header file <link.h> gives access to the internals of the run-time
3005 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3006 application of same name on other systems and it provides information
3007 about dynamically linked binaries.
3009 * The C library now provides the run-time support code for profiling
3010 executables compiled with `-pg'. Programs can control the profiling code
3011 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3012 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3013 the support for this file format was contributed by David Mosberger-Tang.
3015 * The math code has been replaced with a math library based on fdlibm from
3016 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3017 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3018 functions. The math functions now reside in a separate library, so
3019 programs using them will need to use `-lm' their linking commands.
3021 * John C. Bowman contributed optimized ix87 assembler inline functions.
3023 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3024 mechanism similar to that found in Solaris 2. This is now used for the
3025 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3026 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3027 file controls what services are used for each individual database. This
3028 works by loading shared libraries with names specified in `nsswitch.conf',
3029 so service modules can be changed or added at any time without even
3030 relinking any program. Currently there are the file, db, and NIS based
3031 NSS services available.
3033 * The new functions `strtoq' and `strtouq' parse integer values from
3034 strings, like `strtol' and `strtoul', but they return `long long int' and
3035 `unsigned long long int' values, respectively (64-bit quantities).
3037 * The new functions `strtof' and `strtold' parse floating-point values from
3038 strings, like `strtod', but they return `float' and `long double' values,
3039 respectively (on some machines `double' and `long double' are the same).
3041 * Ulrich Drepper has contributed new implementations of the floating-point
3042 printing and reading code used in the `printf' family of functions and
3043 `strtod', `strtof', and `strtold'. These new functions are perfectly
3044 accurate, and much faster than the old ones.
3046 * The implementation of the POSIX locale model was completely rewritten by
3047 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3048 to compile the POSIX locale definition.
3050 * The former dummy implementations of the strcoll and strxfrm function are
3051 now replaced by fully functional code contributed by Ulrich Drepper. The
3052 collation information comes from the POSIX locale definitions.
3054 * The new header <langinfo.h> defines an interface for accessing
3055 various locale-dependent data (using the locale chosen with `setlocale').
3057 * Ulrich Drepper has contributed a new suite of functions for operation on
3058 wide-character and multibyte-character strings, in <wchar.h>;
3059 and classification and case conversion of wide characters, in <wctype.h>.
3060 These new functions are conforming to the ISO C, Amendement 1 specification.
3062 * There is now a second implementation of the standard I/O library available.
3063 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3064 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3066 * You can now use positional parameter specifications in format strings
3067 for the `printf' and `scanf' families of functions. For example,
3068 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3069 ``Number 6, Mr Jones''. This is mainly useful when providing different
3070 format strings for different languages, whose grammars may dictate
3071 different orderings of the values being printed. To support this
3072 feature, the interface for `register_printf_handler' has changed; see
3073 the header file <printf.h> for details.
3075 * The `printf' and `scanf' families of functions now understand a new
3076 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3077 to group numbers as indicated by the locale; for `scanf' and friends, this
3078 says to accept as valid only a number with all the proper grouping
3079 separators in the right places. In the default "C" locale, numbers are
3080 not grouped; but locales for specific countries will define the usual
3081 conventions (i.e. separate thousands with `,' in the US locale).
3083 * The pgrp functions have been regularized, slightly incompatibly but much
3084 less confusingly. The core functions are now `getpgid' and `setpgid',
3085 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3086 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3087 provided for compatibility. There is no longer an incompatible `getpgrp'
3088 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3090 * The new header file <fts.h> and suite of functions simplify programs that
3091 operate on directory trees. This code comes from 4.4 BSD.
3093 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3094 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3095 NSS scheme used in glibc.
3097 * There is a new malloc debugging hook `__memalign_hook'.
3099 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3100 `unsigned int' in <sys/types.h>. These are for compatibility only and
3101 their use is discouraged.
3103 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3104 done differently, so that it works even without GNU ld.
3106 * New function `euidaccess' checks allowed access to a file like `access',
3107 but using the effective IDs instead of the real IDs.
3109 * The time zone data files have been updated for the latest and greatest
3110 local time conventions of the countries of the world.
3112 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3115 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3116 for formatting floating-point numbers. They are provided only for
3117 compatibility; new programs should use `sprintf' instead. There are
3118 also equivalent function for the `long double' floating-point type and
3119 all functions also exist in a reentrant form.
3121 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3122 functions for maintaining the login-record files (primarily of use to
3123 system programs such as `login'), and convenient functions for
3124 allocating and initializing a pseudo-terminal (pty) device.
3126 * Ulrich Drepper has contributed new support for System V style
3127 shared memory and IPC on systems that support it.
3129 * Ulrich Drepper has contributed several miscellaneous new functions found
3130 in System V: The `hsearch' family of functions provide an effective
3131 implementation of hash tables; `a64l' and `l64a' provide a very simple
3132 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3135 * Ulrich Drepper has contributed new reentrant counterparts for the
3136 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3138 * Ulrich Drepper has contributed new, highly-optimized versions of several
3139 string functions for the i486/Pentium family of processors.
3141 * Ulrich Drepper has updated the Linux-specific code, based largely
3142 on work done in Hongjiu Lu's version of GNU libc for Linux.
3143 The GNU library now supports Linux versions 2.0.10 and later,
3144 using the ELF object file format (i[3456]86-*-linux).
3146 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3148 * David Mosberger-Tang and Richard Henderson have ported the C library
3149 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3150 dynamic linking support for ELF/Alpha.
3152 * Richard Henderson contributed several Alpha optimized assembler function
3153 for arithmetic and string handling.
3155 * Ulrich Drepper has contributed a new set of message catalog functions to
3156 support multiple languages using the <libintl.h> interface, for use with
3157 his new package GNU gettext. Translation volunteers have contributed
3158 catalogs of the library's messages in Spanish, German, and Korean.
3160 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3161 program and the `catgets' function for reading the catalog files it
3162 creates. (The <libintl.h> interface is preferred; we include the
3163 <nl_types.h> interface using `catgets' only for source compatibility with
3164 programs already written to use it.)
3166 * New header file <values.h> gives SVID-compatible names for <limits.h>
3169 * Various new macros, declarations, and small header files for compatibility
3172 * New function `group_member' is a convenient way to check if a process has
3173 a given effective group ID.
3175 * When using GCC 2.7 and later, the socket functions are now declared in a
3176 special way so that passing an argument of type `struct sockaddr_in *',
3177 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3178 `struct sockaddr *' type, does not generate a type-clash warning.
3180 * New function `error' declared in header file <error.h> is a convenient
3181 function for printing error messages and optionally exiting; this is the
3182 canonical function used in GNU programs. The new functions `err', `warn',
3183 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3184 doing the same thing.
3186 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3187 POSIX.2 `glob' function to do ~ and {...} expansion.
3189 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3190 `clearenv' which is used in POSIX.9 is also available.
3192 * New function `getsid' returns session ID number on systems that support it.
3194 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3195 files <db.h> and <mpool.h> provide a rich set of functions for several
3196 types of simple databases stored in memory and in files, and <ndbm.h> is
3197 an old `ndbm'-compatible interface using the `db' functions. Link with
3198 `-ldb' to get these functions.
3200 * New macro `strdupa' copies a string like `strdup', but uses local stack
3201 space from `alloca' instead of dynamic heap space from `malloc'.
3203 * New function `strnlen' is like `strlen' but searches only a given maximum
3204 number of characters for the null terminator. `stpncpy', `strndup' and
3205 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3208 * New function `statfs' in header <sys/statfs.h>.
3210 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3211 provide convenient functions for operating on blocks of null-terminated
3214 * A new suite of functions in <utmp.h> handle all the details of reading
3215 and writing the utmp file.
3217 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3220 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3221 completely POSIX compliant and also implemented the extended functionality
3222 to handle alternate digit representation and alternate era date formats.
3224 * Ulrich Drepper provided an implementation of the `strptime' function
3225 defined in XPG4.2 which transforms a string into a `struct tm' value.
3227 * Paul Eggert provided the tzselect shell script as part of the timezone
3228 code. The shell script makes it easy to select the correct timezone
3231 * The implementation of the malloc family of functions is completely replaced
3232 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3233 The implementation uses the mmap function (if available) and it is
3234 optimized for the use in multi threaded programs.
3236 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3237 functions. This new functionality is usable by specifying a special
3238 salt string and it is compatible with implementation on *BSD systems.
3240 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3241 `getsubopt' to handle second level command line options, `bsd_signal'
3242 to access BSD style `signal' functionality, the obsolete `regexp' style
3245 * the `lchown' function is available on system which support this
3248 * The implementation of the shadow password handling function was contributed
3251 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3253 * POSIX.1g support was added. The <sys/select.h> header is available,
3254 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3255 implementation of `getaddrinfo'.
3259 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3261 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3262 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3264 * The new function `daemon' from 4.4 BSD is useful for server programs that
3265 want to put themselves in the background.
3267 * Joel Sherrill has contributed support for several standalone boards that
3268 run without an operating system.
3270 * `printf', `scanf' and friends now accept a `q' type modifier for long
3271 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3273 * All of the code taken from BSD (notably most of the math and networking
3274 routines) has been updated from the BSD 4.4-Lite release.
3276 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3278 * The new functions `getdomainname' and `setdomainname' fetch or change the
3279 YP/NIS domain name. These are system calls which exist on systems which
3282 * The time zone data files have been updated for the latest international
3285 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3286 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3290 * The C library now includes support for Sun RPC, from Sun's free
3291 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3292 are included. (There is still no support for YP.)
3294 * Tom Quinn has contributed a port of the C library to SGI machines running
3295 Irix 4 (mips-sgi-irix4).
3297 * The new `lockf' function is a simplified interface to the locking
3298 facilities of `fcntl', included for compatibility.
3300 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3302 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3304 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3307 * The error message printed by `assert' for a failed assertion now includes
3308 the name of the program (if using GNU ld) and the name of the calling
3309 function (with versions of GCC that support this).
3311 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3313 * The library now includes the <sys/mman.h> header file and memory
3314 management functions `mmap', `munmap', `mprotect', `msync', and
3315 `madvise', on systems that support those facilities.
3317 * The interface for `mcheck' has changed slightly: the function called to
3318 abort the program when an allocation inconsistency is detected now takes
3319 an argument that indicates the type of failure. The new function
3320 `mprobe' lets you request a consistency check for a particular block at
3321 any time (checks are normally done only when you call `free' or `realloc'
3324 * It is now possible to easily cross-compile the C library, building on one
3325 system a library to run on another machine and/or operating system. All
3326 you need to do is set the variable `HOST_CC' in `configparms' to the
3327 native compiler for programs to run on the machine you are building on (a
3328 few generator programs are used on Unix systems); set `CC' to the
3331 * The new function `fexecve' (only implemented on the GNU system) executes
3332 a program file given a file descriptor already open on the file.
3336 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3337 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3339 * You can set the variable `libprefix' in `configparms' to specify a prefix
3340 to be prepended to installed library files; this makes it easy to install
3341 the GNU C library to be linked as `-lgnuc' or whatever.
3343 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3344 copies a limited number of characters from a string, and returns the
3345 address of the last character written.
3347 * You no longer need to check for whether the installed `stddef.h' is
3348 compatible with the GNU C library. configure now checks for you.
3350 * You can now define a per-stream `fileno' function to convert the
3351 stream's cookie into an integral file descriptor.
3353 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3354 allocates zero bytes of storage, and returns a unique pointer which
3355 you can pass to `realloc' or `free'. The behavior is undefined if
3356 you dereference this pointer.
3358 * The C library now runs on Sony NEWS m68k machines running either
3359 NewsOS 3 or NewsOS 4.
3361 * The new `syscall' function is a system-dependent primitive function
3362 for invoking system calls. It has the canonical behavior on Unix
3363 systems, including unreliable return values for some calls (such as
3364 `pipe', `fork' and `getppid').
3366 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3367 to `EAGAIN', which is the preferred name. On systems whose kernels
3368 use two distinct codes, the C library now translates EWOULDBLOCK to
3369 EAGAIN in every system call function.
3373 * The GNU C Library Reference Manual is now distributed with the library.
3374 `make dvi' will produce a DVI file of the printed manual.
3375 `make info' will produce Info files that you can read on line using C-h i
3376 in Emacs or the `info' program.
3377 Please send comments on the manual to bug-glibc-manual@gnu.org.
3379 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3381 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3383 * Jason Merrill has contributed a port to the Sequent Symmetry running
3384 Dynix version 3 (i386-sequent-dynix).
3386 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3387 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3389 * New function `memory_warnings' lets you arrange to get warnings when
3390 malloc is running out of memory to allocate, like Emacs gives you.
3392 * The C library now contains the relocating allocator used in Emacs 19 for
3393 its editing buffers. This allocator (ralloc) minimizes allocation
3394 overhead and fragmentation by moving allocated regions around whenever it
3395 needs to. You always refer to a ralloc'd region with a "handle" (a
3396 pointer to a pointer--an object of type `void **').
3398 * There is a new `printf' format: `%m' gives you the string corresponding
3399 to the error code in `errno'.
3401 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3402 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3403 pass, the `a' modifier says to fill in a `char **' you pass with a
3406 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3407 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3408 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3410 * `mkstemp' is a traditional Unix function to atomically create and open a
3411 uniquely-named temporary file.
3415 * The standard location for the file that says what the local timezone is
3416 has changed again. It is now `/usr/local/etc/localtime' (or more
3417 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3419 * The distribution no longer contains any files with names longer than 14
3422 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3423 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3425 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3429 * The configuration process has changed quite a bit. The `configure'
3430 script is now used just like the configuration scripts for other GNU
3431 packages. The `sysdeps' directory hierarchy is much rearranged.
3432 The file `INSTALL' explains the new scheme in detail.
3434 * The header files no longer need to be processed into ANSI C and
3435 traditional C versions. There is just one set of files to install, and
3436 it will work with ANSI or old C compilers (including `gcc -traditional').
3438 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3439 MIPS DECStation running Ultrix 4.
3441 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3442 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3443 made itself into a shared library.
3445 * Yet further improved support for the i386, running 4.3 BSD-like systems
3446 (such as Mach 3 with the Unix single-server), or System V.
3448 * New function `strncasecmp' to do case-insensitive string comparison
3449 with limited length.
3451 * New function `strsep' is a reentrant alternative to `strtok'.
3453 * New functions `scandir' and `alphasort' for searching directories.
3455 * New function `setenv' is a better interface to `putenv'.
3457 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3458 function for traversing a directory tree.
3460 * The GNU obstack package is now also part of the C library.
3461 The new function `open_obstack_stream' creates a stdio stream that
3462 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3463 formatted output directly to an obstack.
3465 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3466 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3468 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3470 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3471 things to your strings.
3473 * There are some new test programs: test-fseek, testmb, and testrand.
3475 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3476 These ports are not finished, but are a good starting place for really
3477 supporting those systems.
3479 * `/etc/localtime' is now the standard location for the file that says what
3480 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3481 This follows the general principle that `/etc' is the place for all local
3482 configuration files.
3484 * The C library header files now use `extern "C"' when used by the C++
3485 compiler, so the C library should now work with C++ code.
3487 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3488 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3491 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3492 function which uses the `merge sort' algorithm, and is said to be
3493 significantly faster than the old GNU `qsort' function. Merge sort is
3494 now the standard `qsort' function. The new algorithm can require a lot
3495 of temporary storage; so, the old sorting function is called when the
3496 required storage is not available.
3498 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3499 provides the Unix `crypt' function, plus some other entry points.
3501 * The code and header files taken from 4.4 BSD have been updated with the
3502 latest files released from Berkeley.
3504 ----------------------------------------------------------------------
3505 Copyright information:
3507 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3509 Permission is granted to anyone to make or distribute verbatim copies
3510 of this document as received, in any medium, provided that the
3511 copyright notice and this permission notice are preserved,
3512 thus giving the recipient permission to redistribute in turn.
3514 Permission is granted to distribute modified versions
3515 of this document, or of portions of it,
3516 under the above conditions, provided also that they
3517 carry prominent notices stating who last changed them.
3520 version-control: never