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 not all features from that TS are supported by
20 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
21 18661-4:2015, is supported to enable declarations of functions and macros
22 from that TS. Note that most features from that TS are not supported by
25 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
26 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
27 Since the GNU C Library defaults to a much newer revision of POSIX, this
28 will only affect programs that specifically request an old conformance
29 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
30 see a change in the visible declarations, but a program compiled with
31 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
34 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
35 defined by all multithreaded code, but glibc has not required this for
38 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
39 means that in a future release, the macros “major”, “minor”, and “makedev”
40 will only be available from <sys/sysmacros.h>.
42 These macros are not part of POSIX nor XSI, and their names frequently
43 collide with user code; see for instance glibc bug 19239 and Red Hat bug
44 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
45 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
48 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
49 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
50 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
52 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
53 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
54 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
55 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
56 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
57 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
58 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
59 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
60 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
61 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
62 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
65 * New <math.h> features are added from TS 18661-1:2014:
67 - Signaling NaN macros: SNANF, SNAN, SNANL.
69 - Nearest integer functions: roundeven, roundevenf, roundevenl.
71 - llogb functions: the llogb, llogbf and llogbl functions, and the
72 FP_LLOGB0 and FP_LLOGBNAN macros.
74 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
77 - Comparison macros: iseqsig.
79 - Classification macros: iscanonical, issubnormal, iszero.
81 - Total order functions: totalorder, totalorderf, totalorderl,
82 totalordermag, totalordermagf, totalordermagl.
84 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
86 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
87 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
89 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
90 are added to libc. They convert a floating-point number into string.
92 * Most of glibc can now be built with the stack smashing protector enabled.
93 It is recommended to build glibc with --enable-stack-protector=strong.
94 Implemented by Nick Alcock (Oracle).
96 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
97 intended to be used instead of memset() to erase sensitive data after use;
98 the compiler will not optimize out calls to explicit_bzero even if they
99 are "unnecessary" (in the sense that no _correct_ program can observe the
100 effects of the memory clear).
102 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
103 to float instead of double. This does not affect the ABI of any libraries
104 that are part of the GNU C Library, but may affect the ABI of other
105 libraries that use this type in their interfaces.
107 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
108 float_t and double_t types are now defined to long double instead of float
109 and double. These options are not the default, and this does not affect
110 the ABI of any libraries that are part of the GNU C Library, but it may
111 affect the ABI of other libraries that use this type in their interfaces,
112 if they are compiled or used with those options.
114 * The getentropy and getrandom functions, and the <sys/random.h> header file
117 * The buffer size for byte-oriented stdio streams is now limited to 8192
118 bytes by default. Previously, on Linux, the default buffer size on most
119 file systems was 4096 bytes (and thus remains unchanged), except on
120 network file systems, where the buffer size was unpredictable and could be
121 as large as several megabytes.
123 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
124 for the Linux quota interface which predates kernel version 2.4.22 has
127 * The malloc_get_state and malloc_set_state functions have been removed.
128 Already-existing binaries that dynamically link to these functions will
129 get a hidden implementation in which malloc_get_state is a stub. As far
130 as we know, these functions are used only by GNU Emacs and this change
131 will not adversely affect already-built Emacs executables. Any undumped
132 Emacs executables, which normally exist only during an Emacs build, should
133 be rebuilt by re-running “./configure; make” in the Emacs build tree.
135 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
136 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
137 “no-ip6-dotint” had already been the default, and support for the
138 “ip6-dotint” option was removed from the Internet in 2006.
140 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
141 flag from <resolv.h> have been removed. The option relied on a
142 backwards-incompatible DNS extension which was never deployed on the
145 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
146 in the <resolv.h> header file have been deprecated. They were already
149 * DNSSEC-related declarations and definitions have been removed from the
150 <arpa/nameser.h> header file, and libresolv will no longer attempt to
151 decode the data part of DNSSEC record types. Previous versions of glibc
152 only implemented minimal support for the previous version of DNSSEC, which
153 is incompatible with the currently deployed version.
155 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
156 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
157 <arpa/nameser.h> header file because the distinction between RR types and
158 meta-RR types is not officially standardized, subject to revision, and
159 thus not suitable for encoding in a macro.
161 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
162 and rhook members of the res_state type in <resolv.h> have been removed.
163 The glibc stub resolver did not support these hooks, but the header file
164 did not reflect that.
166 * For multi-arch support it is recommended to use a GCC which has
167 been built with support for GNU indirect functions. This ensures
168 that correct debugging information is generated for functions
169 selected by IFUNC resolvers. This support can either be enabled by
170 configuring GCC with '--enable-gnu-indirect-function', or by
171 enabling it by default by setting 'default_gnu_indirect_function'
172 variable for a particular architecture in the GCC source file
175 * GDB pretty printers have been added for mutex and condition variable
176 structures in POSIX Threads. When installed and loaded in gdb these pretty
177 printers show various pthread variables in human-readable form when read
178 using the 'print' or 'display' commands in gdb.
180 Security related changes:
182 On ARM EABI (32-bit), generating a backtrace for execution contexts which
183 have been created with makecontext could fail to terminate due to a
184 missing .cantunwind annotation. This has been observed to lead to a hang
185 (denial of service) in some Go applications compiled with gccgo. Reported
186 by Andreas Schwab. (CVE-2016-6323)
188 The following bugs are resolved with this release:
190 [The release manager will add the list generated by
191 scripts/list-fixed-bugs.py just before the release.]
195 * The minimum Linux kernel version that this version of the GNU C Library
196 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
197 kernel version 2.6.32 or later suffices (on architectures that already
198 required kernel versions more recent than 3.2, those requirements remain
199 unchanged). Linux 3.2 or later kernel headers are required on all
202 * The pap_AN locale has been deleted. This has been deprecated for a long
203 time. It has been replaced by pap_AW & pap_CW, both of which have long
204 been included in previous releases.
206 * The readdir_r and readdir64_r functions have been deprecated. It is
207 recommended to use readdir and readdir64 instead.
209 * The type “union wait” has been removed. It was deprecated in the early
210 1990s and never part of POSIX. Application code should use the int type
211 instead of “union wait”.
213 * A new NSS action is added to facilitate large distributed system
214 administration. The action, MERGE, allows remote user stores like LDAP
215 to be merged into local user stores like /etc/groups in order to provide
216 easy to use, updated, and managed sets of merged credentials. The new
217 action can be used by configuring it in /etc/nsswitch.conf:
218 group: files [SUCCESS=merge] nis
219 Implemented by Stephen Gallagher (Red Hat).
221 * The deprecated __malloc_initialize_hook variable has been removed from the
224 * The long unused localedef --old-style option has been removed. It hasn't
225 done anything in over 16 years. Scripts using this option can safely
228 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
229 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
230 return the next representable value in the direction of positive infinity
231 and the nextdown functions return the next representable value in the
232 direction of negative infinity. These are currently enabled as GNU
235 Security related changes:
237 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
238 could result in a stack overflow when getnetbyname was called with an
239 overly long name. (CVE-2016-3075)
241 * Previously, getaddrinfo copied large amounts of address data to the stack,
242 even after the fix for CVE-2013-4458 has been applied, potentially
243 resulting in a stack overflow. getaddrinfo now uses a heap allocation
244 instead. Reported by Michael Petlan. (CVE-2016-3706)
246 * The glob function suffered from a stack-based buffer overflow when it was
247 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
248 Reported by Alexander Cherepanov. (CVE-2016-1234)
250 * The Sun RPC UDP client could exhaust all available stack space when
251 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
252 alloca plugin for GCC. (CVE-2016-4429)
254 * The IPv6 name server management code in libresolv could result in a memory
255 leak for each thread which is created, performs a failing naming lookup,
256 and exits. Over time, this could result in a denial of service due to
257 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
259 The following bugs are resolved with this release:
261 [1170] localedata: ne_NP: update Nepali locale definition file
262 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
264 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
265 [6796] math: fdim() does not set errno on overflow
266 [10354] libc: posix_spawn should use vfork() in more cases than presently
267 [11213] localedata: localedata: add copyright disclaimer to locale files
268 [12143] localedata: chr_US: new Cherokee locale
269 [12450] localedata: sgs_LT: new locale
270 [12676] localedata: ln_CD: new locale
271 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
273 [13304] math: fma, fmaf, fmal produce wrong results
274 [14259] build: --localedir arg to configure is ignored
275 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
276 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
277 [14934] localedata: es_CL: wrong first weekday chilean locale
278 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
280 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
282 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
283 [15368] nptl: raise() is not async-signal-safe
284 [15479] math: ceil, floor, round and trunc raise inexact exception
285 [15578] localedata: kk_KZ: various updates
286 [16003] localedata: pap_AN: punt old locale
287 [16137] localedata: iw_IL: punt old locale
288 [16190] localedata: eo: new esperanto locale
289 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
290 [16742] malloc: race condition: pthread_atfork() called before first
291 malloc() results in unexpected locking behaviour/deadlocks
292 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
294 [16983] localedata: postal_fmt does not allow %l and %n modifiers
295 [17565] localedata: pt_PT: wrong (work-)week start
296 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
297 [17950] build: Build fails with -msse
298 [18205] localedata: be_BY*: wrong first_weekday and first_workday
299 [18433] libc: posix_spawn does not return correctly upon failure to
301 [18453] localedata: charmaps/IBM875: incorrect codes
302 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
304 [18896] localedata: he_IL: improvements for currency
305 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
307 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
309 [19056] libc: Deprecate readdir_r
310 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
312 [19198] localedata: nl_NL: small improvements for Dutch locales
313 [19257] network: Per-thread memory leak in __res_vinit with IPv6
314 nameservers (CVE-2016-5417)
315 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
317 [19400] locale: Language missing in "iso-639.def", trivial fix in
319 [19431] malloc: Deadlock between fflush, getdelim, and fork
320 [19505] libc: Incorrect file descriptor validity checks in
321 posix_spawn_file_actions_add{open,close,dup2}
322 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
324 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
325 `intl/{gettextP,loadinfo}.h'
326 [19534] libc: execle, execlp may use malloc
327 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
328 [19573] network: res_nclose and __res_maybe_init disagree about name
329 server initialization, breaking Hesiod
330 [19575] localedata: Status of GB18030 tables
331 [19581] localedata: sr_* date_fmt string contains additional newline
332 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
334 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
335 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
336 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
337 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
339 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
341 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
343 [19610] dynamic-link: ldconfig -X removes stale symbolic links
344 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
345 [19633] locale: strfmon_l applies global locale to number formatting
346 [19642] network: Memory leak in getnameinfo
347 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
348 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
350 [19654] math: [x86_64] Need testcase for BZ #19590 fix
351 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
353 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
354 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
356 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
358 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
360 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
361 pointers and lengths in error-case.
362 [19727] locale: Converting from/to UTF-xx with iconv() does not always
363 report errors on UTF-16 surrogates values.
364 [19755] nscd: nscd assertion failure in gc
365 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
366 [19759] libc: mempcpy shouldn't be inlined
367 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
368 [19765] libc: s390 needs an optimized mempcpy
369 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
370 NAME_MAX limit assumption (CVE-2016-1234)
371 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
372 [19787] network: Missing and incorrect truncation checks in getnameinfo
373 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
375 [19791] network: Assertion failure in res_query.c with un-connectable name
377 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
378 [19822] math: libm.so install clobbers old version
379 [19825] network: resolv: send_vc can return uninitialized data in second
380 response to getaddrinfo
381 [19830] network: nss_dns: should check RDATA length against buffer length
382 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
383 confronted with A/AAAA records of invalid size
384 [19837] nss: nss_db: No retries for some long lines with a larger buffer
385 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
386 [19853] stdio: Printing IBM long double in decimal with high precision is
388 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
389 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
390 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
392 [19865] network: Assertion failure or memory leak in
393 _nss_dns_getcanonname_r
394 [19868] network: nss_dns: netent code does not skip over non-PTR records
395 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
397 [19881] string: Improve x86-64 memset
398 [19907] string: Incorrect memcpy tests
399 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
401 [19925] libc: termios.h XCASE namespace
402 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
403 [19929] libc: limits.h NL_NMAX namespace
404 [19931] stdio: Memory leak in vfprintf
405 [19957] libc: clone(CLONE_VM) access invalid parent memory
406 [19963] localedata: en_IL: New locale
407 [19989] stdio: stdio.h cuserid namespace
408 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
410 [19996] locale: langinfo.h nl_langinfo_l namespace
411 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
412 [20010] network: getaddrinfo: Stack overflow in hostent translation
414 [20012] stdio: libio: fmemopen append mode failure
415 [20014] stdio: stdio.h namespace for pre-threads POSIX
416 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
417 [20023] libc: fcntl.h timespec namespace
418 [20024] math: [x86_64] vectorized sincos trashes the stack
419 [20031] network: nss_hesiod: Heap overflow in get_txt_records
420 [20041] time: sys/time.h timespec namespace
421 [20043] libc: unistd.h missing cuserid for UNIX98 and before
422 [20044] libc: unistd.h missing pthread_atfork for UNIX98
423 [20051] libc: ttyslot in wrong header under wrong conditions
424 [20054] libc: gethostname not declared for XPG4
425 [20055] libc: termios.h missing tcgetsid for XPG4
426 [20072] dynamic-link: x86 init_cpu_features is called twice in static
428 [20073] libc: sys/stat.h fchmod namespace
429 [20074] libc: stdlib.h rand_r namespace
430 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
431 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
433 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
434 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
436 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
437 [20119] libc: Wrong mask for processors level type from CPUID
438 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
439 AS not supporting AVX512
440 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
442 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
443 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
445 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
446 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
447 [20178] libc: posix_spawn{p} should not call exit
448 [20191] stdio: libio: vtables hardening
449 [20195] string: FMA4 detection requires CPUID execution with register
451 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
452 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
454 [20212] math: acos (sNaN) returns sNaN
455 [20213] math: asin (sNaN) returns sNaN
456 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
457 [20218] math: [i386] asinhl (sNaN) returns sNaN
458 [20219] math: [i386] atanhl (sNaN) returns sNaN
459 [20222] stdio: fopencookie: Mangle function pointers
460 [20224] math: [i386] cbrtl (sNaN) returns sNaN
461 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
462 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
464 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
465 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
466 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
467 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
468 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
469 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
470 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
471 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
472 [20240] math: modf (sNaN) returns sNaN
473 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
474 [20250] math: frexp (sNaN) returns sNaN
475 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
476 [20255] math: [i386] fdim, fdimf return with excess range and precision /
478 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
479 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
480 subscript is below array bounds [-Werror=array-bounds]
481 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
482 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
484 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
485 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
487 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
489 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
490 [20316] localedata: id_ID: Februari instead of Pebruari
491 [20327] string: POWER8 strcasecmp returns incorrect result
492 [20347] math: Failure: Test: j0_downward (0xap+0)
493 [20348] libc: FAIL: misc/tst-preadvwritev64
494 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
495 [20350] libc: There is no test for p{read,write}64
496 [20357] math: Incorrect cos result for 1.5174239687223976
497 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
501 * Unicode 8.0.0 Support: Character encoding, character type info, and
502 transliteration tables are all updated to Unicode 8.0.0, using new
503 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
504 These updates cause user visible changes, such as the fixes for bugs
505 89, 16061, and 18568.
507 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
508 kernel-internal CPU set size. This means that requests that change the
509 CPU affinity which failed before (for example, an all-ones CPU mask) will
510 now succeed. Applications that need to determine the effective CPU
511 affinities need to call sched_getaffinity or pthread_getaffinity_np after
512 setting it because the kernel can adjust it (and the previous size check
513 would not detect this in the majority of cases).
515 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
516 the following new symbols are used: fts64_children, fts64_close,
517 fts64_open, fts64_read and fts64_set.
519 * getaddrinfo now detects certain invalid responses on an internal netlink
520 socket. If such responses are received, an affected process will
521 terminate with an error message of "Unexpected error <number> on netlink
522 descriptor <number>" or "Unexpected netlink response of size <number> on
523 descriptor <number>". The most likely cause for these errors is a
524 multi-threaded application which erroneously closes and reuses the netlink
525 file descriptor while it is used by getaddrinfo.
527 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
528 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
529 unnecessary serialization of memory allocation requests across threads.
530 The defect is now corrected. Users should see a substantial increase in
531 the concurent throughput of allocation requests for applications which
532 trigger this bug. Affected applications typically create and
533 destroy threads frequently. (Bug 19048 was reported and analyzed by
536 * There is now a --disable-timezone-tools configure option for disabling the
537 building and installing of the timezone related utilities (zic, zdump, and
538 tzselect). This is useful for people who build the timezone data and code
539 independent of the GNU C Library.
541 * The obsolete header <regexp.h> has been removed. Programs that require
542 this header must be updated to use <regex.h> instead.
544 * The obsolete functions bdflush, create_module, get_kernel_syms,
545 query_module and uselib are no longer available to newly linked binaries;
546 the header <sys/kdaemon.h> has been removed. These functions and header
547 were specific to systems using the Linux kernel and could not usefully be
548 used with the GNU C Library on systems with version 2.6 or later of the
551 * Optimized string, wcsmbs and memory functions for IBM z13.
552 Implemented by Stefan Liebler.
554 * Newly linked programs that define a variable called signgam will no longer
555 have it set by the lgamma, lgammaf and lgammal functions. Programs that
556 require signgam to be set by those functions must ensure that they use the
557 variable provided by the GNU C Library and declared in <math.h>, without
558 defining their own copy.
560 * The minimum GCC version that can be used to build this version of the GNU
561 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
562 still be used to compile programs using the GNU C Library.
564 Security related changes:
566 * An out-of-bounds value in a broken-out struct tm argument to strftime no
567 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
569 * The LD_POINTER_GUARD environment variable can no longer be used to disable
570 the pointer guard feature. It is always enabled. Previously,
571 LD_POINTER_GUARD could be used to disable security hardening in binaries
572 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
575 * An integer overflow in hcreate and hcreate_r could lead to an
576 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
578 * The catopen function no longer has unbounded stack usage. Reported by
581 * The nan, nanf and nanl functions no longer have unbounded stack usage
582 depending on the length of the string passed as an argument to the
583 functions. Reported by Joseph Myers. (CVE-2014-9761)
585 * A stack-based buffer overflow was found in libresolv when invoked from
586 libnss_dns, allowing specially crafted DNS responses to seize control
587 of execution flow in the DNS client. The buffer overflow occurs in
588 the functions send_dg (send datagram) and send_vc (send TCP) for the
589 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
590 family. The use of AF_UNSPEC triggers the low-level resolver code to
591 send out two parallel queries for A and AAAA. A mismanagement of the
592 buffers used for those queries could result in the response of a query
593 writing beyond the alloca allocated buffer created by
594 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
595 the overflow. Thanks to the Google Security Team and Red Hat for
596 reporting the security impact of this issue, and Robert Holiday of
597 Ciena for reporting the related bug 18665. (CVE-2015-7547)
599 The following bugs are resolved with this release:
601 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
602 [887] math: Math library function "logb" and "nextafter" inconsistent
603 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
604 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
605 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
606 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
608 [4404] localedata: German translation of "Alarm clock" is misleading
609 [6799] math: nextafter() and nexttoward() doen't set errno on
610 overflow/underflow errors
611 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
613 [10432] nis: _nss_nis_setnetgrent assertion failure
614 [11460] libc: fts has no LFS support
615 [12926] network: getaddrinfo()/make_request() may spin forever
616 [13065] nptl: Race condition in pthread barriers
617 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
618 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
620 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
621 [14912] libc: Rename non-installed bits/*.h headers
622 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
623 [15367] math: Let gcc use __builtin_isinf
624 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
625 [15421] math: lgamma wrongly sets signgam for ISO C
626 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
627 FE_INVALID with argument out of range
628 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
630 [15786] dynamic-link: ifunc resolver functions can smash function
632 [15918] math: Unnecessary check for equality in hypotf()
633 [16061] localedata: Review / update transliteration data
634 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
636 [16141] time: strptime %z offset restriction
637 [16171] math: drem should be alias of remainder
638 [16296] math: fegetround is pure?
639 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
640 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
642 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
643 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
645 [16422] math: [powerpc] math-float, math-double failing llrint tests with
646 "Exception "Inexact" set" on ppc32
647 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
648 [16517] math: Missing underflow exception from tanf/tan/tanl
649 [16519] math: Missing underflow exception from sinhf
650 [16520] math: Missing underflow exception from tanhf
651 [16521] math: Missing underflow exception from exp2
652 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
654 [16734] stdio: fopen calls mmap to allocate its buffer
655 [16961] math: nan function incorrect handling of bad sequences
656 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
657 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
658 [16985] locale: localedef: confusing error message when opening output
660 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
661 [17197] locale: Redundant shift character in iconv conversion output at
663 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
664 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
665 [17250] dynamic-link: static linking breaks nss loading
666 (getaddrinfo/getpwnam/etc...)
667 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
669 [17441] math: isnan() should use __builtin_isnan() in GCC
670 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
671 (related to lock elision)
672 [17787] manual: Exponent on page 324 of the PDF ends prematurely
673 [17886] time: strptime should be able to parse "Z" as a timezone with %z
674 [17887] time: strptime should be able to parse "+01:00" style timezones
675 [17905] libc: catopen() Multiple unbounded stack allocations
677 [18084] libc: backtrace (..., 0) dumps core on x86
678 [18086] libc: nice() sets errno to 0 on success
679 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
680 is too large (CVE-2015-8778)
681 [18251] dynamic-link: SONAME missing when audit modules provides path
682 [18265] libc: add attributes for wchar string and memory functions
683 [18370] math: csqrt missing underflows
684 [18421] libc: [hppa] read-only segment has dynamic relocations
685 [18472] libc: Obsolete syscall wrappers should be compat symbols
686 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
687 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
688 [18525] localedata: Remove locale timezone information
689 [18560] libc: [powerpc] spurious bits/ipc.h definitions
690 [18568] localedata: Update locale data to Unicode 8.0
691 [18589] locale: sort-test.sh fails at random
692 [18595] math: ctan, ctanh missing underflows
693 [18604] libc: assert macro-expands its argument
694 [18610] math: S390: fetestexcept() reports any exception if DXC-code
695 contains a vector instruction exception.
696 [18611] math: j1, jn missing errno setting on underflow
697 [18618] localedata: sync Chechen locale definitions with other *_RU
699 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
700 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
701 [18665] network: In send_dg, the recvfrom function is NOT always using the
702 buffer size of a newly created buffer (CVE-2015-7547)
703 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
705 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
707 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
708 [18699] math: tilegx cproj() for various complex infinities does not yield
710 [18724] libc: Harden put*ent functions against data injection
711 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
712 [18755] build: build errors with -DNDEBUG
713 [18757] stdio: fmemopen fails to set errno on failure
714 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
716 [18781] libc: openat64 lacks O_LARGEFILE
717 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
718 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
719 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
720 [18790] math: [ldbl-128ibm] tanhl inaccurate
721 [18795] libc: stpncpy fortification misses buffer lengths that are
723 [18796] build: build fails for --disable-mathvec
724 [18803] math: hypot missing underflows
725 [18820] stdio: fmemopen may leak memory on failure
726 [18823] math: csqrt spurious underflows
727 [18824] math: fma spurious underflows
728 [18825] math: pow missing underflows
729 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
730 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
731 [18870] build: sem_open.c fails to compile with missing symbol
733 [18872] stdio: Fix memory leak in printf_positional
734 [18873] libc: posix_fallocate overflow check ineffective
735 [18875] math: Excess precision leads incorrect libm
736 [18877] libc: arm: mmap offset regression
737 [18887] libc: memory corruption when using getmntent on blank lines
738 [18918] localedata: hu_HU: change time to HH:MM:SS format
739 [18921] libc: Regression: extraneous stat() and fstat() performed by
741 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
742 binaries (CVE-2015-8777)
743 [18951] math: tgamma missing underflows
744 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
746 [18953] localedata: lt_LT: change currency symbol to the euro
747 [18956] math: powf inaccuracy
748 [18961] math: [i386] exp missing underflows
749 [18966] math: [i386] exp10 missing underflows
750 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
751 [18969] build: multiple string test failures due to missing locale
753 [18970] libc: Reference of pthread_setcancelstate in libc.a
754 [18977] math: float / long double Bessel functions not in XSI POSIX
755 [18980] math: i386 libm functions return with excess range and precision
756 [18981] math: i386 scalb*, ldexp return with excess range and precision
757 [18982] stdio: va_list and vprintf
758 [18985] time: Passing out of range data to strftime() causes a segfault
760 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
761 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
762 [19012] locale: iconv_open leaks memory on error path
763 [19016] math: clog, clog10 inaccuracy
764 [19018] nptl: Mangle function pointers in tls_dtor_list
765 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
766 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
767 [19048] malloc: malloc: arena free list can become cyclic, increasing
769 [19049] math: [powerpc] erfc incorrect zero sign
770 [19050] math: [powerpc] log* incorrect zero sign
771 [19058] math: [x86_64] Link fail with -fopenmp and -flto
772 [19059] math: nexttoward overflow incorrect in non-default rounding modes
773 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
774 [19074] network: Data race in _res_hconf_reorder_addrs
775 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
776 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
777 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
779 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
781 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
782 [19086] manual: posix_fallocate64 documented argument order is wrong.
783 [19088] math: lround, llround missing exceptions close to overflow
785 [19094] math: lrint, llrint missing exceptions close to overflow threshold
786 [19095] math: dbl-64 lrint incorrect for 64-bit long
787 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
788 [19124] dynamic-link: ld.so failed to build with older assmebler
789 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
790 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
791 [19134] math: [powerpc32] lround, lroundf spurious exceptions
792 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
793 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
794 pthread_setaffinity_np
795 [19156] math: [ldbl-128] j0l spurious underflows
796 [19164] nptl: tst-getcpu fails with many possible CPUs
797 [19168] math: math/test-ildoubl and math/test-ldouble failure
798 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
799 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
801 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
802 (FE_NOMASK_ENV) do not clear SSE exceptions
803 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
804 _int_new_arena/reused_arena
805 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
806 [19201] math: dbl-64 remainder incorrect sign of zero result
807 [19205] math: bits/math-finite.h conditions do not match math.h and
809 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
810 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
811 for C99-based standards
812 [19212] libc: features.h not -Wundef clean
813 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
815 [19214] libc: Family and model identification for AMD CPU's are incorrect.
816 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
817 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
819 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
821 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
823 [19242] libc: strtol incorrect in Turkish locales
824 [19243] malloc: reused_arena can pick an arena on the free list, leading
825 to an assertion failure and reference count corruption
826 [19253] time: tzset() ineffective when temporary TZ did not include DST
828 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
829 [19270] math: [hppa] Shared libm missing __isnanl
830 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
831 MAP_HUGETLB and MAP_STACK defines
832 [19313] nptl: Wrong __cpu_mask for x32
833 [19347] libc: grantpt: try to force a specific gid even without pt_chown
834 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
835 [19350] math: [ldbl-128ibm] sinhl spurious overflows
836 [19351] math: [ldbl-128ibm] logl inaccurate near 1
837 [19363] time: x32: times() return value wrongly truncates/sign extends
839 [19367] dynamic-link: Improve branch prediction on Silvermont
840 [19369] network: Default domain name not reset by res_ninit when "search"
841 / "domain" entry is removed from resolv.conf
842 [19375] math: powerpc: incorrect results for POWER7 logb with negative
844 [19385] localedata: bg_BG: time separator should be colon, not comma
845 [19408] libc: linux personality syscall wrapper may erroneously return an
846 error on 32-bit architectures
847 [19415] libc: dladdr returns wrong names on hppa
848 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
849 IBM905, IBM907, and IBM909
850 [19439] math: Unix98 isinf and isnan functions conflict with C++11
851 [19443] build: build failures with -DDEBUG
852 [19451] build: Make check fails on test-double-vlen2
853 [19462] libc: Glibc failed to build with -Os
854 [19465] math: Wrong code with -Os
855 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
857 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
859 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
860 [19486] math: S390: Math tests fail with "Exception Inexact set".
861 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
862 [19550] libc: [mips] mmap negative offset handling inconsistent with other
864 [19590] math: Fail to build shared objects that use libmvec.so functions.
868 * The following bugs are resolved with this release:
870 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
871 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
872 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
873 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
874 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
875 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
876 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
877 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
878 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
879 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
880 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
881 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
882 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
883 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
884 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
885 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
886 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
887 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
888 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
889 18657, 18676, 18694, 18696, 18887.
891 * Cache information can be queried via sysconf() function on s390 e.g. with
892 _SC_LEVEL1_ICACHE_SIZE as argument.
894 * A buffer overflow in gethostbyname_r and related functions performing DNS
895 requests has been fixed. If the NSS functions were called with a
896 misaligned buffer, the buffer length change due to pointer alignment was
897 not taken into account. This could result in application crashes or,
898 potentially arbitrary code execution, using crafted, but syntactically
899 valid DNS responses. (CVE-2015-1781)
901 * The time zone file parser has been made more robust against crafted time
902 zone files, avoiding heap buffer overflows related to the processing of
903 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
904 large time zone data files. Overly long time zone specifiers in the TZ
905 variable no longer result in stack overflows and crashes.
907 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
908 for LD and GD on x86 and x86-64, has been implemented. You will need
909 binutils-2.24 or later to enable this optimization.
911 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
912 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
913 Hat). These updates cause user visible changes, such as the fix for bug
916 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
917 and getXXbyYY NSS calls for the same database, causing a denial-of-service
918 condition in some applications.
920 * Added vector math library named libmvec with the following vectorized x86_64
921 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
923 The library can be disabled with --disable-mathvec. Use of the functions is
924 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
925 Shared library libmvec.so is linked in as needed when using -lm (no need to
926 specify -lmvec explicitly for not static builds).
927 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
929 * A new fmemopen implementation has been added with the goal of POSIX
930 compliance. The new implementation fixes the following long-standing
931 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
932 old implementation is still present for use be by existing binaries.
934 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
935 releases. It has been fixed to match 2.19 and older, but binaries built
936 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
938 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
939 Contributed by Roland McGrath (Google).
941 * The header <regexp.h> is deprecated, and will be removed in a future
942 release. Use of this header will trigger a deprecation warning.
943 Application developers should update their code to use <regex.h> instead.
945 This header was formerly part of SUS, but was deprecated in 1994 and
946 removed from the standard in 2001. Also, the glibc implementation
947 leaks memory. See BZ#18681 for more details.
951 * The following bugs are resolved with this release:
953 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
954 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
955 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
956 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
957 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
958 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
959 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
960 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
961 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
962 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
965 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
966 too little memory for the to-be-scanned arguments and overflow the
967 allocated buffer. The implementation now correctly computes the required
968 buffer size when using malloc, and switches to malloc from alloca as
971 * A new semaphore algorithm has been implemented in generic C code for all
972 machines. Previous custom assembly implementations of semaphore were
973 difficult to reason about or ensure that they were safe. The new version
974 of semaphore supports machines with 64-bit or 32-bit atomic operations.
975 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
976 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
978 * Port to Altera Nios II has been contributed by Mentor Graphics.
980 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
981 implementations for powerpc64/powerpc64le.
982 Implemented by Adhemerval Zanella (IBM).
984 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
985 and powerpc64le. This may improve lock scaling of existing programs on
986 HTM capable systems. The lock elision code is only enabled with
987 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
988 powerpc will issue a transaction abort on every syscall to avoid side
989 effects being visible outside transactions.
991 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
992 AArch64. Contributed by ARM Ltd.
994 * i386 memcpy functions optimized with SSE2 unaligned load/store.
996 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
997 under certain input conditions resulting in the execution of a shell for
998 command substitution when the application did not request it. The
999 implementation now checks WRDE_NOCMD immediately before executing the
1000 shell and returns the error WRDE_CMDSUB as expected.
1002 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1003 processing format strings with a large number of format specifiers.
1005 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1006 infinite loop if the DNS response contained a PTR record of an unexpected
1009 * The minimum GCC version that can be used to build this version of the GNU
1010 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1011 still be used to compile programs using the GNU C Library.
1013 * The GNU C Library is now built with -Werror by default. This can be
1014 disabled by configuring with --disable-werror.
1016 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1018 * The obsolete sigvec function has been removed. This was the original
1019 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1020 programs have been using instead for about 25 years. Of course, ABI
1021 compatibility for old binaries using sigvec remains intact.
1023 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1024 with newer versions of bison.
1026 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1027 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1028 registers overlay two consecutive single-precision registers. MIPS32R2
1029 introduced a new FPU mode (FR=1) where double-precision registers extend the
1030 corresponding single-precision registers which is incompatible with the
1031 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1032 require the use of FR=1 making a transition necessary. New o32 ABI
1033 extensions enable users to migrate over time from the original o32 ABI
1034 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1035 tracks the ABI of any loaded object and verifies that new objects are
1036 compatible. Mode transitions will also be requested as required and
1037 unsupportable objects will be rejected. The ABI checks include both soft and
1038 hard float ABIs for o32, n32 and n64.
1040 GCC 5 with GNU binutils 2.25 onwards:
1041 It is strongly recommended that all o32 system libraries are built using the
1042 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1043 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1044 compiler using --with-fp-32=xx to set this by default.
1048 * The following bugs are resolved with this release:
1050 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1051 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1052 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1053 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1054 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1055 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1056 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1057 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1058 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1059 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1060 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1061 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1062 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1063 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1064 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1066 * Reverted change of ABI data structures for s390 and s390x:
1067 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1068 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1069 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1070 are preserved pointing straight to the same implementation as the old ones.
1071 Given that, new callers will simply provide a too-big buffer to these
1072 functions. Any applications/libraries out there that embed jmp_buf or
1073 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1074 against 2.19 headers will have to rebuilt again. This is necessary in any
1075 case to revert the breakage in their ABI caused by the glibc change.
1077 * Support for file description locks is added to systems running the
1078 Linux kernel. The standard file locking interfaces are extended to
1079 operate on file descriptions, not file descriptors, via the use of
1080 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1081 are associated with an open file instead of a process.
1083 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1085 * The minimum Linux kernel version that this version of the GNU C Library
1086 can be used with is 2.6.32.
1088 * Running the testsuite no longer terminates as soon as a test fails.
1089 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1090 with PASS or FAIL lines for individual tests. A summary of the results is
1091 printed, including a list of failing lists, and "make check" exits with
1092 error status if there were any unexpected failures. "make check
1093 stop-on-test-failure=y" may be used to keep the old behavior.
1095 * The am33 port, which had not worked for several years, has been removed
1098 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1099 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1100 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1101 conflicted with POSIX (support for which was removed in 2.19), the
1102 interfaces those macros enabled remain available when compiling with
1103 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1104 test macros defined.
1106 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1108 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1109 This may improve lock scaling of existing programs on TX capable systems.
1110 The lock elision code is only built with --enable-lock-elision=yes and
1111 then requires a GCC version supporting the TX builtins. With lock elision
1112 default mutexes are elided via __builtin_tbegin, if the cpu supports
1113 transactions. By default lock elision is not enabled and the elision code
1116 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1117 copy the path argument. This allowed programs to cause posix_spawn to
1118 deference a dangling pointer, or use an unexpected pathname argument if
1119 the string was modified after the posix_spawn_file_actions_addopen
1122 * All supported architectures now use the main glibc sysdeps directory
1123 instead of some being in a separate "ports" directory (which was
1124 distributed separately before glibc 2.17).
1126 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1127 On configurations that support it (all Linux configurations), it's now
1128 used regardless of the --enable-add-ons switch to configure. It is no
1129 longer possible to build such configurations without pthreads support.
1131 * Locale names, including those obtained from environment variables (LANG
1132 and the LC_* variables), are more tightly checked for proper syntax.
1133 setlocale will now fail (with EINVAL) for locale names that are overly
1134 long, contain slashes without starting with a slash, or contain ".." path
1135 components. (CVE-2014-0475) Previously, some valid locale names were
1136 silently replaced with the "C" locale when running in AT_SECURE mode
1137 (e.g., in a SUID program). This is no longer necessary because of the
1140 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1141 with application code using Intel MPX instructions. (With all previous
1142 versions, the MPX register state could be clobbered when making calls
1143 into or out of a shared library.) Note that while the new dynamic
1144 linker is compatible with all known x86 hardware whether or not it
1145 supports Intel MPX, some x86 instruction-set emulators might fail to
1146 handle the new instruction encodings. This is known to affect Valgrind
1147 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1148 release), and might affect other tools that do instruction emulation.
1150 * Support for loadable gconv transliteration modules has been removed.
1151 The support for transliteration modules has been non-functional for
1152 over a decade, and the removal is prompted by security defects. The
1153 normal gconv conversion modules are still supported. Transliteration
1154 with //TRANSLIT is still possible, and the //IGNORE specifier
1155 continues to be supported. (CVE-2014-5119)
1157 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1158 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1159 resulting a denial-of-service security vulnerability in applications which
1160 use functions related to iconv. (CVE-2014-6040)
1164 * The following bugs are resolved with this release:
1166 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1167 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1168 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1169 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1170 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1171 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1172 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1173 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1174 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1175 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1176 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1177 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1178 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1179 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1180 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1181 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1182 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1183 16431, 16453, 16474, 16506, 16510, 16529
1185 * Slovenian translations for glibc messages have been contributed by the
1186 Translation Project's Slovenian team of translators.
1188 * The public headers no longer use __unused nor __block. This change is to
1189 support compiling programs that are derived from BSD sources and use
1190 __unused internally, and to support compiling with Clang's -fblock
1191 extension which uses __block.
1193 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1194 large collation sequences to optimize multiple passes. This cache
1195 computation may overflow for large collation sequences and may cause a
1196 stack or buffer overflow. This is now fixed to use a slower algorithm
1197 which does not use a cache if there is an integer overflow.
1199 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1200 rules for large collation sequences to optimize multiple passes and falls
1201 back to alloca if malloc fails, resulting in a possible stack overflow.
1202 The implementation now falls back to an uncached collation sequence lookup
1205 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1206 initialized for static applications resulting in the security feature
1207 being disabled. The pointer guard is now correctly initialized to a
1208 random value for static applications. Existing static applications need
1209 to be recompiled to take advantage of the fix (bug 15754).
1211 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1212 to the d_name member of struct dirent, or omit the terminating NUL
1213 character. (Bugzilla #14699).
1215 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1216 aligned_alloc functions could allocate too few bytes or corrupt the
1217 heap when passed very large allocation size values (Bugzilla #15855,
1220 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1221 for AF_INET6 has been fixed (Bugzilla #16072).
1223 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1224 pap_CW, quz_PE, the_NP.
1226 * Substantially revised locales: gd_GB, ht_HT
1228 * The LC_ADDRESS field was updated to support country_car for almost all
1231 * ISO 1427 definitions were updated.
1233 * ISO 3166 definitions were updated.
1235 * The localedef utility now supports --big-endian and --little-endian
1236 command-line options to generate locales for a different system from that
1237 for which the C library was built.
1239 * Binary locale files now only depend on the endianness of the system for
1240 which they are generated and not on other properties of that system. As a
1241 consequence, binary files generated with new localedef may be incompatible
1242 with old versions of the GNU C Library, and binary files generated with
1243 old localedef may be incompatible with this version of the GNU C Library,
1244 in the following circumstances:
1246 + Locale files may be incompatible on m68k systems.
1248 + Locale archive files (but not separate files for individual locales) may
1249 be incompatible on systems where plain "char" is signed.
1251 * The configure option --disable-versioning has been removed. Builds with
1252 --disable-versioning had not worked for several years.
1254 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1255 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1257 * SystemTap probes for malloc have been introduced.
1259 * SystemTap probes for slow multiple precision fallback paths of
1260 transcendental functions have been introduced.
1262 * Support for powerpc64le has been added.
1264 * The soft-float powerpc port now supports e500 processors.
1266 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1268 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1269 set of header declarations that are enabled by default, even when other
1270 feature test macros or compiler options such as -std=c99 would otherwise
1271 disable some of those declarations.
1273 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1274 conflict with POSIX. The libbsd-compat library (which was a dummy library
1275 that did nothing) has also been removed.
1277 * Preliminary documentation about Multi-Thread, Async-Signal and
1278 Async-Cancel Safety has been added.
1280 * Change of ABI data structures for s390 and s390x:
1281 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1282 allow for future hardware extensions. All library functions that accept or
1283 return these structures were versioned in order to provide backwards
1284 compatibility with existing code. However, not all cases can be handled
1285 with interface versioning. If an application embeds either structure into
1286 another structure and that structure is passed to another compilation unit
1287 which expects the newer larger-sized object then any access to the new
1288 fields will result in undefined behaviour. Similarly any access to fields
1289 that were moved by the enlarging of the structures will cause undefined
1290 behaviour. To fix the undefined behaviour all objects that are part of the
1291 final in-memory image must agree on the size of structures, and this may
1292 require recompilation.
1296 * The following bugs are resolved with this release:
1298 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1299 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1300 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1301 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1302 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1303 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1304 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1305 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1306 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1307 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1308 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1309 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1310 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1313 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1314 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1315 Distributions can re-enable building and using pt_chown via the new configure
1316 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1317 considerable security risks and should only be used if the distribution
1318 understands and accepts the risks.
1320 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1323 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1324 fixed (Bugzilla #15330).
1326 * Add support for calling C++11 thread_local object destructors on thread
1327 and program exit. This needs compiler support for offloading C++11
1328 destructor calls to glibc.
1330 * Improved worst case performance of libm functions with double inputs and
1333 * Support for priority inherited mutexes in pthread condition variables on
1334 non-x86 architectures.
1336 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1338 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1340 * Optimized string functions for ARM. Implemented by Will Newton and
1343 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1345 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1348 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1349 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1351 * Added a benchmark framework to track performance of functions in glibc.
1353 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1354 It is based on draft TS 18661 and currently enabled as a GNU extension.
1356 * On Linux, the clock function now uses the clock_gettime system call
1357 for improved precision, rather than old times system call.
1359 * Added support for version-3 tz format files. This is needed when using
1360 the tz database release 2013e or later, and affects a few unusual cases --
1361 currently only TZ='America/Godthab' for time stamps after 2037.
1363 * Added new API functions pthread_getattr_default_np and
1364 pthread_setattr_default_np to get and set the default pthread
1365 attributes of a process.
1367 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1368 This may improve lock scaling of existing programs on TSX capable systems.
1369 When the --enable-lock-elision=yes parameter is specified at configure
1370 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1373 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1374 availability). Implemented by Ryan S. Arnold.
1376 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1380 * The following bugs are resolved with this release:
1382 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1383 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1384 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1385 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1386 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1387 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1388 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1389 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1390 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1391 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1392 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1393 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1394 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1395 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1396 14879, 14889, 14893, 14898, 14914.
1398 * Optimization of memcpy for MIPS.
1400 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1401 EMFILE has been fixed (Bugzilla #14889).
1403 * The add-on ports collection is now distributed in the "ports" subdirectory
1404 of the main GNU C Library distribution, rather than separately.
1406 * Port to ARM AArch64 contributed by Linaro.
1408 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1409 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1411 Implemented by Andreas Krebbel.
1413 * The new function secure_getenv allows secure access to the environment,
1414 returning NULL if running in a SUID/SGID process. This function replaces
1415 the internal function __secure_getenv.
1417 * SystemTap static probes have been added into the dynamic linker.
1418 Implemented by Gary Benson.
1420 * Optimizations of string functions strstr, strcasestr and memmem.
1421 Implemented by Maxim Kuvyrkov.
1423 * The minimum Linux kernel version that this version of the GNU C Library
1424 can be used with is 2.6.16.
1426 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1427 powerpc POWER7. Implemented by Will Schmidt.
1429 * New configure option --disable-nscd builds the C library such that it
1430 never attempts to contact the Name Service Caching Daemon (nscd).
1431 New configure option --disable-build-nscd avoids building nscd itself;
1432 this is the default if --disable-nscd is used.
1434 * Improved support for cross-compilation, including cross-testing and
1435 bootstrap builds without a previously built glibc.
1437 * Several testsuite tests are now able to test multiple IFUNC variants of an
1438 interface, rather than just testing the one that would be chooen by
1441 * New configure options --with-bugurl and --with-pkgversion, for
1442 distributors to use to embed their bug-reporting and package version
1443 information in --help and --version output.
1445 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1446 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1447 allows creation of chroots without the procfs mounted on /proc.
1449 * The `crypt' function now fails if passed salt bytes that violate the
1450 specification for those values. On Linux, the `crypt' function will
1451 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1452 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1453 when the mode is enabled.
1455 * The `clock_*' suite of functions (declared in <time.h>) is now available
1456 directly in the main C library. Previously it was necessary to link with
1457 -lrt to use these functions. This change has the effect that a
1458 single-threaded program that uses a function such as `clock_gettime' (and
1459 is not linked with -lrt) will no longer implicitly load the pthreads
1460 library at runtime and so will not suffer the overheads associated with
1461 multi-thread support in other code such as the C++ runtime library.
1463 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1468 * The following bugs are resolved with this release:
1470 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1471 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1472 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1473 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1474 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1475 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1476 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1477 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1478 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1479 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1480 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1481 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1482 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1483 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1484 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1485 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1486 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1487 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1488 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1489 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1490 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1491 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1494 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1495 configuring glibc with:
1496 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1497 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1498 Implemented by H.J. Lu.
1502 + define static_assert
1504 + do not declare gets
1506 + declare at_quick_exit and quick_exit also for ISO C11
1508 + aligned_alloc. NB: The code is deliberately allows the size parameter
1509 to not be a multiple of the alignment. This is a moronic requirement
1510 in the standard but it is only a requirement on the caller, not the
1513 + timespec_get added
1515 + uchar.h support added
1517 + CMPLX, CMPLXF, CMPLXL added
1519 Implemented by Ulrich Drepper.
1521 * Support for the IA-64 has been moved to ports.
1523 * Remove support for anything but ELF binary format
1525 * Checking versions of poll, ppoll added.
1526 Implemented by Ulrich Drepper.
1528 * More generic and 64-bit performance optimizations to math functions.
1529 Implemented by Ulrich Drepper.
1531 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1532 headers and functions available at compile time as they were before
1533 version 2.14. This option will be removed at some time in the future
1534 after the TI-RPC library becomes fully sufficient for the needs of
1535 existing applications.
1537 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1538 Note that glibc is not expected to work with any Linux kernel version
1541 * New header <sys/auxv.h> and function getauxval allowing easy access to
1542 the AT_* key-value pairs passed from the Linux kernel. The header also
1543 defines the HWCAP_* bits associated with the AT_HWCAP key.
1545 * New locales: mag_IN
1547 * New configure option --enable-systemtap builds SystemTap static probes
1548 into libc for setjmp and longjmp and into libpthread for various operations.
1549 So far the setjmp/longjmp probes and some of the libpthread probes are
1550 provided only for i*86 and x86_64.
1551 Implemented by Roland McGrath and Rayson Ho.
1553 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1555 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1558 * More optimized functions for SPARC. Implemented by David S. Miller.
1560 * Improved support for cross-compilation, especially bootstrap builds
1561 without a previously built glibc.
1563 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1564 Chris Metcalf from Tilera.
1566 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1567 now supported for ARM processors.
1569 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1570 as the name of the dynamic linker, to distinguish it from the
1571 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1573 * Support for CRIS has been removed from ports.
1575 * A new class of installed header has been documented for low-level
1576 platform-specific functionality. PowerPC added the first instance with a
1577 function to provide time base register access. Contributed by Tulio
1578 Magno Quites Machado Filho.
1580 * ix86 configurations now install header files that are consistent with
1581 what x86-64 configurations install. These same header files can be used
1582 for -m32, -m64, or -mx32 builds.
1583 Contributed by H.J. Lu.
1585 * Math library bug fixes. A thorough audit of all open math library bugs was
1586 conducted by Joseph Myers. Significant progress was made on many math
1587 library bugs resulting in more accurate exceptions and function results.
1588 Many thanks to all those that contributed including Andreas Jaeger for his
1589 patch review and work on the x87 trigonometric instruction issues.
1591 * Timezone data is no longer installed. Timezone-related binaries and scripts
1592 will continue to be installed. Users should obtain their timezone data from
1593 their distribution provider or from the tzdata package at
1594 <ftp://munnari.oz.au/pub/>.
1598 * The following bugs are resolved with this release:
1600 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1601 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1602 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1603 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1604 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1605 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1606 13484, 13506, 13515, 13523, 13524, 13538, 13540
1608 * New program pldd to list loaded object of a process
1609 Implemented by Ulrich Drepper.
1611 * Add nss_db support back to glibc. No more dependency on Berkeley db
1612 and support for initgroups lookups.
1613 Implemented by Ulrich Drepper.
1615 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1616 Contributed by HJ Lu.
1618 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1619 Contributed by HJ Lu.
1621 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1622 on x86-32 and x86-64.
1623 Contributed by Liubov Dmitrieva.
1625 * Optimized strchr and strrchr for SSE on x86-32.
1626 Contributed by Liubov Dmitrieva.
1628 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1629 for x86-64 and x86-32.
1630 Contributed by Liubov Dmitrieva.
1632 * New interfaces: scandirat, scandirat64
1633 Implemented by Ulrich Drepper.
1635 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1636 Implemented by Ulrich Drepper.
1638 * nscd now also caches the netgroup database.
1639 Implemented by Ulrich Drepper.
1641 * Integrate libm with gcc's -ffinite-math-only option.
1642 Implemented by Ulrich Drepper.
1644 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1645 to math functions. Implemented by Ulrich Drepper.
1647 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1648 Implemented by Ulrich Drepper.
1650 * New Linux interfaces: process_vm_readv, process_vm_writev
1652 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1653 Implemented by Ulrich Drepper.
1655 * Optimized nearbyint and strcasecmp for PPC.
1656 Implemented by Adhemerval Zanella.
1658 * New locales: bho_IN, unm_US, es_CU, ta_LK
1662 * The following bugs are resolved with this release:
1664 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1665 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1666 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1667 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1668 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1669 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1670 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1671 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1672 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1673 12795, 12811, 12813, 12814, 12841
1675 * The RPC implementation in libc is obsoleted. Old programs keep working
1676 but new programs cannot be linked with the routines in libc anymore.
1677 Programs in need of RPC functionality must be linked against TI-RPC.
1678 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1680 Visible changes of this change include (obviously) the inability to link
1681 programs using RPC functions without referencing the TI-RPC library and the
1682 removal of the RPC headers from the glibc headers.
1683 Implemented by Ulrich Drepper.
1685 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1686 syncfs, setns, sendmmsg
1688 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1689 yue_HK, lij_IT, mhr_RU
1691 * New encodings: CP770, CP771, CP772, CP773, CP774
1693 * New program sotruss to trace calls through PLTs
1694 Implemented by Ulrich Drepper.
1696 * The malloc hook implementation is marked deprecated and will be removed
1697 from the default implementation in the next version. The design never
1698 worked ever since the introduction of threads. Even programs which do
1699 not create threads themselves can use multiple threads created internally.
1703 * The following bugs are resolved with this release:
1705 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1706 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1707 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1708 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1711 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1713 * POWER7 optimizations: memset, memcmp, strncmp
1715 * New optimized string functions for x86-64: strnlen (SSE2),
1716 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1717 Implemented by Ulrich Drepper.
1721 * The following bugs are resolved with this release:
1723 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1724 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1725 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1726 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1727 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1728 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1729 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1730 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1732 * New interfaces: pthread_getname_np, pthread_setname_np
1734 * New Linux interface: recvmmsg
1736 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1738 * The dynamic linker now recognizes supported ABI versions from the
1739 EI_ABIVERSION field in the ELF header.
1740 Implemented by Ulrich Drepper.
1742 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1743 will not be used to fill in password fields in the passwd.byname replies.
1744 Instead it is used to synthesize the shadow.byname table, should it be
1745 missing. This is a useful mode in some installations involving Solaris.
1746 Implemented by Ulrich Drepper.
1748 * New locales: kok_IN, sq_MK, cv_RU
1752 * The following bugs are resolved with this release:
1754 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1755 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1756 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1757 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1758 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1759 10789, 10817, 10823, 10840, 10847
1761 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1763 Implemented by Ulrich Drepper.
1765 * Checking version of longjmp added that fails if an uninitialized stack
1766 frame would be created. Implemented by Ulrich Drepper.
1768 * STT_GNU_IFUNC is now supported in static executables.
1769 Implemented by H.J. Lu.
1771 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1773 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1774 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1775 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1776 Contributed by H.J. Lu.
1778 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1779 Implemented by Ulrich Drepper.
1781 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1783 Contributed by H.J. Lu.
1785 * Support for fma instruction in AVX on x86-64.
1786 Implemented by H.J. Lu and Ulrich Drepper.
1788 * AVX support in x86-64 auditing support in ld.so.
1789 Implemented by H.J. Lu.
1791 * STB_GNU_UNIQUE support added.
1792 Implemented by Ulrich Drepper.
1794 * Implement second fallback mode for optimized DNS lookup for even more
1795 broken environments. If two requests from the same port are not
1796 handled correctly close the socket and open a new one before sending
1797 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1798 can be used to select this mode right away, instead of rediscovering the
1799 necessity is every process again.
1800 Implemented by Ulrich Drepper.
1802 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1803 Implemented by Adam Tkac.
1805 * Optimized iconv conversions for S390x.
1806 Implemented by Andreas Krebbel.
1808 * Using condvars with PI mutexes is now more efficient due to kernel
1809 support for requeueing to PI futexes. NPTL support added for x86-64.
1810 Implemented by Ulrich Drepper.
1812 * New locales: ps_AF, my_MM
1816 * The following bugs are resolved with this release:
1818 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1819 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1820 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1821 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1822 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1824 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1825 Implemented by Ulrich Drepper.
1827 * New Linux interfaces: accept4, fallocate, fallocate64.
1828 Implemented by Ulrich Drepper.
1830 * Correct declarations of string function when used in C++ code. This
1831 could lead to compile errors for invalid C++ code.
1833 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1836 * New POSIX 2008 interface: psiginfo
1837 Implemented by Ulrich Drepper.
1839 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1840 Implemented by Ulrich Drepper.
1842 * Support for selecting between multiple function definitions at runtime
1843 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1845 * The libcrypt library can now use the hash function implementations in
1846 NSS. Implemented by Ulrich Drepper.
1848 * The malloc implementation can be compiled to be less memory efficient
1849 but higher performing in multi-threaded programs.
1850 Implemented by Ulrich Drepper.
1852 * New locales: nan_TW@latin, ks_IN
1854 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1855 Implemented by Ulrich Drepper.
1857 * Extended printf hook support. It is possible to use user-defined types
1858 and extend existing format specifiers.
1859 Implemented by Ulrich Drepper.
1861 * Handling for group shadow files has been added.
1862 Implemented by Ulrich Drepper.
1864 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1865 servers (the case, e.g., for some people using the built-in DNS
1866 server in ADSL modems/routers). There is a once-per-process timeout
1867 in case of a broken server. To avoid it, users can run nscd or put
1868 'options single-request' in /etc/resolv.conf.
1869 Implemented by Ulrich Drepper.
1873 * The following bugs are resolved with this release:
1875 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1876 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1877 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1878 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1879 6968, 6974, 6980, 6995, 7008, 7009, 7029
1881 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1882 up at the same time. Implemented by Ulrich Drepper.
1884 * TLS descriptors for LD and GD on x86 and x86-64.
1885 Implemented by Alexandre Oliva.
1887 * getaddrinfo now handles DCCP and UDPlite.
1888 Implemented by Ulrich Drepper.
1890 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1891 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1892 Implemented by Ulrich Drepper.
1894 * New implementation of memmem, strstr, and strcasestr which is O(n).
1895 Implemented by Eric Blake.
1897 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1899 * Implement "e" option for popen to open file descriptor with the
1900 close-on-exec flag set. Implemented by Ulrich Drepper.
1902 * Implement "b" mode for fmemopen. In this mode writes writes don't
1903 implicitly add a NUL byte and seeks from the end of the buffer really
1904 use the buffer end, not the string length as the basis.
1905 Implemented by Ulrich Drepper.
1907 * Many functions, exported and internal, now atomically set the close-on-exec
1908 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1910 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1912 Implemented by Pravin Satpute.
1914 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1918 * The following bugs are resolved with this release:
1920 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1921 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1922 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1923 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1924 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1925 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1926 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1928 * New locales: bo_CN, bo_IN, shs_CA.
1930 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1932 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1933 Implemented by Pravin Satpute.
1935 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1937 * nscd caches now all timeouts for DNS entries
1938 Implemented by Ulrich Drepper.
1940 * nscd is more efficient and wakes up less often.
1941 Implemented by Ulrich Drepper.
1943 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1944 vdprintf, and obstack_vprintf.
1945 Implemented by Jakub Jelinek.
1947 * Faster memset for x86-64.
1948 Implemented by Harsha Jagasia and H.J. Lu.
1950 * Faster memcpy on x86.
1951 Implemented by Ulrich Drepper.
1953 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1954 Implemented by Ulrich Drepper.
1956 * Faster sqrt and sqrtf implemention for some PPC variants.
1957 Implemented by Steven Munroe.
1961 * The following bugs are resolved with this release:
1963 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1964 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1965 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1966 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1967 5071, 5103, 5104, 5112, 5113, 5184, 5186
1969 * More checking functions: fread, fread_unlocked, open*, mq_open.
1970 Implemented by Jakub Jelinek and Ulrich Drepper.
1972 * Extend fortification to C++. Implemented by Jakub Jelinek.
1974 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1975 by not recognizing 'a' as a modifier when those specs are requested.
1976 Implemented by Jakub Jelinek.
1978 * PPC optimizations to math and string functions.
1979 Implemented by Steven Munroe.
1981 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1982 options to be passed. Implemented by Ulrich Drepper.
1984 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1986 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1987 Implemented by Ulrich Drepper.
1989 * Handle private futexes in the NPTL implementation.
1990 Implemented by Jakub Jelinek and Ulrich Drepper.
1992 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1993 Implemented by Roland McGrath and Ulrich Drepper.
1995 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1997 * SHA-256 and SHA-512 based password encryption.
1998 Implemented by Ulrich Drepper.
2000 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2001 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2004 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2005 Implemented by Ulrich Drepper.
2009 * The following bugs are resolved with this release:
2011 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2012 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2013 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2014 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2015 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2016 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2017 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2020 * New Linux interfaces: epoll_pwait, sched_getcpu.
2022 * New generic interfaces: strerror_l.
2024 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2028 * The following bugs are resolved with this release:
2030 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2031 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2032 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2033 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2034 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2035 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2036 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2037 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2038 3225, 3251, 3252, 3253, 3273
2040 * For Linux, the sorting of addresses returned by getaddrinfo now also
2041 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2042 Implemented by Ulrich Drepper.
2044 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2045 Implemented by Ulrich Drepper.
2047 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2049 * New iconv module for MIK. Contributed by Alexander Shopov.
2051 * For sites with broken group and/or passwd database, the auto-propagate
2052 option of nscd can prevent creating ID lookup entries from the results
2053 of a name lookup and vice versa. This usually is no problem but some
2054 site might have problems with the default behavior.
2055 Implemented by Ulrich Drepper.
2057 * Iterating over entire database in NIS can be slow. With the
2058 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2059 to trade time for memory. The entire database will be read at once.
2060 Implemented by Ulrich Drepper.
2062 * The interfaces introduced in RFC 3542 have been implemented by
2065 * Support for the new ELF hash table format was added by Ulrich Drepper.
2067 * Support for priority inheritance mutexes added by Jakub Jelinek and
2070 * Support for priority protected mutexes added by Jakub Jelinek.
2072 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2073 or_IN, csb_PL, fy_NL, sr_ME.
2077 * More overflow detection functions.
2079 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2080 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2082 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2083 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2084 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2085 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2086 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2087 by Masahide Washizawa.
2089 * It is now possible to install audit libraries for the dynamic linker using
2090 LD_AUDIT. Implemented by Ulrich Drepper.
2092 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2093 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2094 For a libc and libpthread that works well on Linux 2.4 kernels, we
2095 recommend using the stable 2.3 branch.
2097 * The new function open_wmemstream acts like open_memstream,
2098 but uses a wchar_t wide character buffer.
2100 * The new function ppoll is an improved version of poll, similar to pselect.
2102 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2103 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2104 renameat, unlinkat, linkat, symlinkat, readlinkat.
2106 * New Linux kernel system calls: unshare,
2107 inotify_init, inotify_add_watch, inotify_rm_watch.
2109 * The euidaccess function is now also known by the alias eaccess,
2110 for compatibility with some other systems.
2112 * Timezone data updated to 2006b version.
2116 * The following bugs are resolved with this release:
2118 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2119 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2120 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2121 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2122 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2123 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2125 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2127 * As of this release, GCC 4 can be used to compile the C Library.
2129 * Timezone data updated to 2005m version.
2133 * The following bugs are resolved with this release:
2135 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2136 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2137 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2138 777, 787, 821, 822, 823, 825
2140 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2144 * Support for RFC 3678. Real implementations exist only for Linux so far.
2145 Implemented by Ulrich Drepper.
2147 * nscd can now cache entries persistently. Expiring entries are reloaded.
2148 For speedups the cache can be shared in memory with client processes.
2149 Implemented by Ulrich Drepper.
2151 * nscd can now perform SELinux checks.
2152 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2154 * getaddrinfo queries are now cached. Canonical name lookup is performed
2156 Implemented by Ulrich Drepper.
2158 * The nothrow function attribute is used when headers are used by gcc when
2159 compiling C code. This can avoid significant amounts of exception
2162 * The malloc functions perform more error checking and are stricter when
2163 it comes to reacting on errors. The default action is to terminate
2164 the process after showing an error message. Implemented by Ulrich Drepper.
2166 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2167 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2169 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2170 namespaces using the new function dlmopen(). This feature is of course,
2171 like most other dynamic loading functionality, not available in statically
2172 linked applications. Implemented by Ulrich Drepper.
2174 * Low-overhead boundary checking variants of string and some stdio functions
2175 were added. These are to be used in conjunction with a gcc patch by
2176 Jakub Jelinek which adds calls to these functions if possible.
2177 Implemented by Jakub Jelinek and Ulrich Drepper.
2179 * Old code for several operating systems and machine architectures that
2180 have not been in working condition in a long time have been removed from
2181 the main source tree maintained by the GNU C Library's maintainers.
2182 These files are now reside in the separate `ports' source module
2183 that is usable as an add-on when building the library.
2187 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2188 interrogate the dynamic linker, compatible with the Solaris interface.
2190 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2191 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2193 * getifaddrs now uses the netlink interface on Linux to get its information.
2194 Implemented by Thorsten Kukuk.
2196 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2197 Implemented by Ulrich Drepper.
2199 * support for non-executable stacks on x86 has been added. Changes mostly
2202 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2205 * getaddrinfo now performs destination address selection according to
2210 * Thread-safe interfaces for many functions that access locale data
2211 were added in version 2.3, but these features were omitted from NEWS.
2212 Many functions have variants with an `_l' suffix that take a `locale_t'
2213 object as a parameter rather than consulting the current locale.
2214 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2215 create and maintain `locale_t' objects. Additionally, the new function
2216 `uselocale' sets "the current locale" (as used by functions not so
2217 parameterized) set for an individual thread. These features were added
2218 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2220 * The functions getresuid, getresgid, setresuid, and setresgid, which
2221 have long been available on Linux, are now declared in <unistd.h>
2222 and are now also available on the Hurd.
2224 * ELF thread-local storage support (TLS) now works on x86-64.
2226 * The new dynamic string token $LIB is expanded in shared library names.
2227 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2229 * Aldy Hernandez contributed complete software floating point support for
2230 PowerPC machines with no FPU.
2232 * fexecve is implemented on Linux.
2234 * The `btowc' function should work at least twice as fast due to
2235 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2237 * With approriate thread add-ons cancelable functions are now implemented
2238 in libc.so as well. No need to call the function in libpthread. This
2239 change allowed to finally disable the incorrect and expensive handling
2240 of weak definition in ld.so.
2242 * Yet more PLT entries in libc.so have been removed. We finally arrived
2243 at the bare minimum. Startup times improved appropriately.
2245 * Support for the new Linux/x86 system call interface was added. The
2246 AT_SYSINFO auxiliary vector entry is recognized and handled.
2250 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2253 * iconv (the program and the interface) now accepts empty names (excluding
2254 options like //TRANSLIT) to mean "use charset of current locale".
2256 * localedef can now transliterate characters in strings which are not in
2257 the provided charmap. The information from the input locale is used.
2259 * Prelinking support was added for ELF targets. This requires additional
2260 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2262 * Read-only stdio streams now use mmap to speed up operation by eliminating
2263 copying and buffer underflows. To use add 'm' to the mode string of
2264 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2266 * The malloc functions were completely rewritten by Wolfram Gloger based
2267 on Doug Lea's malloc-2.7.0.c.
2269 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2270 implementation of regex.
2272 * Bruno Haible upgraded the iconv and locale implementation to support
2275 * Contents of the LC_* and LANG environment variables in the CEN style are
2276 not recognized anymore. It never was used. Change by Ulrich Drepper.
2278 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2279 thread-local storage (TLS) ABI on some platforms.
2280 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2282 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2283 EUC-JISX0213, and TSCII.
2285 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2286 BSD-compatible interface for getting all network interface addresses.
2287 Implementation for IPv4 by Roland McGrath.
2289 * Loading of locale data is faster due to the introduction of a locale
2290 archive. Implemented by Roland McGrath and Ulrich Drepper.
2292 * Startup times are significantly reduced by not using exported functions
2293 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2296 * Steven Munroe contributed a port to PowerPC64/Linux.
2300 * The Hurd now uses the GNU libio implementation of stdio rather than the
2301 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2303 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2304 with the same behavior as the Linux system call of the same name.
2308 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2309 128-bit long double format.
2311 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2312 IBM1161, and IBM1162 charsets.
2314 * Andreas Jaeger contributed a port to x86-64/Linux.
2316 * Peter Bruin contributed a port to PowerPC/Hurd.
2318 * libc functions using I/O streams now can handle wide-oriented streams
2321 * optimizations in the dynamic linker. Binaries created by recent binutils
2322 versions start up quicker due to reduced time spent on relocations.
2324 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2328 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2329 asinh, atanh, j0 for the 128-bit long double format.
2331 * Bruno Haible updated all the code handling Unicode in some form to
2332 support Unicode 3.1.
2334 * Speed of regex for single-byte locales is back to previous levels.
2335 Patch by Isamu Hasegawa.
2337 * Alpha, SPARC, and IA-64 now also using floating stacks.
2339 * Startup time of internationalized applications greatly improved through
2340 iconv cache. Use iconvconfig to generate the cache file.
2341 Contributed by Ulrich Drepper.
2343 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2344 severe performance problems. Patch by David Mosberger.
2346 * The Hurd port got a lot more functionality like AIO, various stdio
2347 extensions, etc. Mainly done by Roland McGrath.
2349 * mtrace can now lookup symbols in shared libraries.
2353 * Intel's IA-64 math library is largely integrated. It provides fast and
2354 accurate implementatations for most basic and standard math functions
2355 in float, double, and long double format.
2357 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2358 and asin for the 96-bit long double format and asin, log, tan for the
2359 128-bit long double format.
2361 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2362 The basic double functions exist today. Contributed by Abraham Ziv
2363 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2364 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2366 * An asynchronous name lookup library was added. The interface is designed
2367 after POSIX AIO. The proposal was circulated beforehand to get comments.
2368 No negative ones came in. Implemented by Ulrich Drepper.
2370 * Port to S390/64bit contributed by Martin Schwidefsky
2371 <schwidefsky@de.ibm.com>.
2373 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2374 of functions for Linux/IA-64.
2376 * The RPC code is now thread safe. Threads can now use the same service
2377 of different services at the same time. Patch by Eric Norum
2378 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2380 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2381 family of functions for Linux/S390.
2383 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2384 of functions for Linux/x86.
2386 * Port to Linux/CRIS contributed by Axis Communications.
2390 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2391 we can now check for namespace violations and missing declarations. The
2392 result is that almost all headers are now Unix-compliant (as defined in
2393 the upcoming XPG6). The negative side is that some programs might need
2394 corrections, too, if they depend on the incorrect form of the headers in
2395 previous versions which defined too many symbols and included too many
2398 * regex now handles multibyte character sets correctly.
2399 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2401 * iconv (the program) does now conform to the upcoming XPG6 and handles
2402 charmaps. Instead of the charset names the path of charmaps can be
2403 provided and the conversion happens based on this data.
2404 Contributed by Ulrich Drepper.
2406 * The locale program now provides detailed information about the installed
2411 only lists the names of the supported locales
2415 provides details such as country, language, and codeset name.
2416 Contributed by Ulrich Drepper.
2420 * The gencat program now parses the input file according to the charset
2421 selected by the LC_CTYPE category. This is important for stateful
2422 character sets. To make generating catalogs easier there is a way
2423 to overwrite the charset selected by the locale: before the first
2424 message or $ quote line the catalog can contain a line like
2426 $ codeset=ISO-8859-2
2428 to select the charset (ISO-8859-2 in this case).
2430 Implemented by Shinya Hanataka and Ulrich Drepper.
2432 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2433 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2434 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2436 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2437 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2439 * The LANGUAGE environment variable is now ignored unless the locale is
2440 changed from the default "C" locale.
2442 * The usual bug fixes.
2446 * Greg McGary added runtime support for bounds checking using gcc's
2447 new -fbounded-pointers option. ix86 target is complete. PowerPC
2450 * Thorsten Kukuk added secure mode support to nscd.
2452 * The Berkeley db libraries have been removed.
2454 Related, the nss_db module is now in a separate package since it
2455 obviously requires a database library being available.
2457 * Wide character I/O streams implemented by Ulrich Drepper.
2459 * Functions from the extended socket API added by Ulrich Drepper.
2461 * Functions feenableexcept and fedisableexcept to control the
2462 behaviour of individual exceptions have been added by Andreas Jaeger.
2464 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2466 * The resolver code has been updated from bind 8.2.3-T5B which supports
2467 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2470 This change could in some situations effect backward compatibility. Since
2471 now `_res' is a thread-local instead of a global variable, modifying it
2472 in one thread does not have any effect in other threads.
2474 The resolver library was also extended to allow IPv6 as the transport
2475 protocol for the requests. This work was done by Stig Venaas.
2477 * Compatibility code for K&R C compilers has been removed from the
2478 header files. A ISO C compiler is needed to use the library
2479 (conforming to either C89 or C99 standard).
2481 * Complete rewrite of the localedef program to support multibyte character
2482 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2483 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2484 Implemented by Ulrich Drepper.
2486 Bruno Haible significantly improved the generation and use of the data
2487 structures for the wide character tables.
2489 * Plural handling in gettext implemented by Ulrich Drepper.
2491 * The utmp daemon has been removed.
2493 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2495 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2498 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2500 * POSIX barriers implemented by Kaz Kylheku.
2502 * POSIX spawn function family implemented by Ulrich Drepper.
2504 * POSIX spinlocks are now available.
2506 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2508 * the configure option --enable-kernel=X.Y.Z allows to strip out
2509 compatibility for kernel versions before X.Y.Z. This is currently only
2510 implemented for Linux.
2512 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2513 field sin6_scope_id. This means that all programs using IPv6 should be
2514 recompiled. Don't expect binary compatibility with previous glibc
2517 * various conversion modules for IBM character sets contributed by
2520 * IA-64 port by Jes Sorensen and HJ Lu.
2534 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2536 * New xdr functions are added; some rpc functions are now 64bit clean.
2538 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2540 * Fixed known incompatibilities with glibc 2.0.
2542 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2544 * Optimized a number of functions (especially the ELF dynamic loader).
2546 * Update timezone data files.
2548 * lots of charmaps corrections
2550 * some new locale definitions and charmaps
2555 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2556 conform to POSIX member type requirements. Symbol versions have been
2557 adjusted as needed within the library, and for direct use by applications,
2558 but there is potential for problems if third-party libraries use
2559 struct timeval as part of their interface. This does not present
2560 a problem for X and other "essential" system libraries.
2562 * An additional locale model to support C++ Standard Library locale
2563 model and probably more was implemented by Ulrich Drepper.
2565 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2568 * Miles Bader provided the `argp' function family to support hierachical
2569 command line argument parsing, layered on top of getopt.
2571 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2573 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2576 * scanf recognizes the %a and %A format for scanning floating point numbers.
2578 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2579 information and interfaces for the available integer types.
2581 * about 130 new math functions were added to implement the ISO C9x math
2584 * the new header <complex.h> contains definitions of the complex math
2585 functions from ISO C 9X.
2587 * the new header <tgmath.h> defines generic macros to use complex or
2588 real valued functions.
2590 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2592 * Andreas Jaeger provided a test suite for the math library.
2594 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2596 * Ulrich Drepper added character set conversion functions (iconv).
2598 * Optimized string functions have been added.
2600 * The localedata addon is now part of glibc.
2602 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2604 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2605 daemon for NSS (nscd).
2607 Missing a better place here are some numbers on improvements. Under
2608 Linux 2.1.125 un-tar-ing the kernel sources takes
2612 using local files 12.19s 6.88s 22.91s
2614 using NIS 13.92s 8.91s 26.34s
2616 using NIS & nscd 10.37s 7.34s 25.30s
2618 using NIS+ 27.57s 30.37s 640.46s
2620 using NIS+ & nscd 10.25s 7.83s 26.51s
2622 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2624 Keep in mind that non-namelookup related operations dominate above times.
2625 It was just a common complain that using NIS+ unpacking the kernel is
2628 [1] The old nscd implementation is not available anymore since it was
2629 distributed with glibc up to version 2.0.98 and thus is now replaced.
2631 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2633 * Mark Kettenis provided a Hesiod NSS module.
2635 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2636 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2638 * rcmd can now handle netgroups (Dick Streefland).
2640 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2643 * Support for the IPv6 protocol has been added to the socket API, as per the
2644 latest draft standards.
2646 * Support for Linux 2.2 has been added.
2648 * Interface changes relative to the latest 2.0.x release:
2649 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2650 addseverity NEW: Unix98
2651 alphasort64 NEW: LFS
2652 argp_err_exit_status NEW: argp, GNU ext
2653 argp_error NEW: argp, GNU ext
2654 argp_failure NEW: argp, GNU ext
2655 argp_help NEW: argp, GNU ext
2656 argp_parse NEW: argp, GNU ext
2657 argp_program_bug_address NEW: argp, GNU ext
2658 argp_program_version NEW: argp, GNU ext
2659 argp_program_version_hook NEW: argp, GNU ext
2660 argp_state_help NEW: argp, GNU ext
2661 argp_usage NEW: argp, GNU ext
2662 authdes_create NEW: Secure RPC
2663 authdes_getucred NEW: Secure RPC
2664 authdes_pk_create NEW: Secure RPC
2665 backtrace NEW: GNU ext.
2666 backtrace_symbols NEW: GNU ext.
2667 backtrace_symbols_fd NEW: GNU ext.
2669 cacosf NEW: ISO C 9x
2670 cacosh NEW: ISO C 9x
2671 cacoshf NEW: ISO C 9x
2672 cacoshl NEW: ISO C 9x
2673 cacosl NEW: ISO C 9x
2680 casinf NEW: ISO C 9x
2681 casinh NEW: ISO C 9x
2682 casinhf NEW: ISO C 9x
2683 casinhl NEW: ISO C 9x
2684 casinl NEW: ISO C 9x
2686 catanf NEW: ISO C 9x
2687 catanh NEW: ISO C 9x
2688 catanhf NEW: ISO C 9x
2689 catanhl NEW: ISO C 9x
2690 catanl NEW: ISO C 9x
2691 cbc_crypt NEW: Secure RPC
2695 ccoshf NEW: ISO C 9x
2696 ccoshl NEW: ISO C 9x
2702 cimagf NEW: ISO C 9x
2703 cimagl NEW: ISO C 9x
2704 clearerr_locked REMOVED
2705 clntunix_create NEW: sunrpc ext
2707 clog10 NEW: ISO C 9x
2708 clog10f NEW: ISO C 9x
2709 clog10l NEW: ISO C 9x
2719 cprojf NEW: ISO C 9x
2720 cprojl NEW: ISO C 9x
2722 crealf NEW: ISO C 9x
2723 creall NEW: ISO C 9x
2728 csinhf NEW: ISO C 9x
2729 csinhl NEW: ISO C 9x
2732 csqrtf NEW: ISO C 9x
2733 csqrtl NEW: ISO C 9x
2737 ctanhf NEW: ISO C 9x
2738 ctanhl NEW: ISO C 9x
2740 des_setparity NEW: Secure RPC
2741 ecb_crypt NEW: Secure RPC
2742 endutxent NEW: Unix98
2744 exp10f NEW: ISO C 9x
2745 exp10l NEW: ISO C 9x
2749 fattach NEW: STREAMS
2750 fdetach NEW: STREAMS
2754 feclearexcept NEW: ISO C 9x
2755 fegetenv NEW: ISO C 9x
2756 fegetexceptflag NEW: ISO C 9x
2757 fegetround NEW: ISO C 9x
2758 feholdexcept NEW: ISO C 9x
2760 feraiseexcept NEW: ISO C 9x
2761 ferror_locked REMOVED
2762 fesetenv NEW: ISO C 9x
2763 fesetexceptflag NEW: ISO C 9x
2764 fesetround NEW: ISO C 9x
2765 fetestexcept NEW: ISO C 9x
2766 feupdateenv NEW: ISO C 9x
2767 fflush_locked REMOVED
2771 fgets_unlocked NEW: GNU ext.
2772 fileno_locked REMOVED
2784 fputc_locked REMOVED
2785 fputs_unlocked NEW: GNU ext.
2786 fread_unlocked NEW: GNU ext.
2791 fstatvfs NEW: Unix98
2795 ftruncate64 NEW: LFS
2797 fwrite_unlocked NEW: GNU ext.
2798 gai_strerror NEW: IPv6
2802 getchar_locked REMOVED
2804 getdate_err NEW: Unix98
2805 getdate_r NEW: GNU ext.
2807 getnameinfo NEW: IPv6
2808 getnetname NEW: Secure RPC
2809 getpmsg NEW: STREAMS
2810 getpt NEW: Unix98 PTY
2811 getrlimit64 NEW: LFS
2812 getutxent NEW: Unix98
2813 getutxid NEW: Unix98
2814 getutxline NEW: Unix98
2815 glob64 NEW: GNU ext.
2816 globfree64 NEW: GNU ext.
2817 gnu_get_libc_release NEW: GNU ext.
2818 gnu_get_libc_version NEW: GNU ext.
2819 grantpt NEW: Unix98 PTY
2820 host2netname NEW: Secure RPC
2822 iconv_close NEW: iconv
2823 iconv_open NEW: iconv
2824 if_freenameindex NEW: IPv6
2825 if_indextoname NEW: IPv6
2826 if_nameindex NEW: IPv6
2827 if_nametoindex NEW: IPv6
2828 in6addr_any NEW: IPv6
2829 in6addr_loopback NEW: IPv6
2830 inet6_isipv4mapped NEW: IPv6
2831 isastream NEW: STREAMS
2832 iswblank NEW: GNU ext.
2833 key_decryptsession NEW: Secure RPC
2834 key_decryptsession_pk NEW: Secure RPC
2835 key_encryptsession NEW: Secure RPC
2836 key_encryptsession_pk NEW: Secure RPC
2837 key_gendes NEW: Secure RPC
2838 key_get_conv NEW: Secure RPC
2839 key_secretkey_is_set NEW: Secure RPC
2840 key_setnet NEW: Secure RPC
2841 key_setsecret NEW: Secure RPC
2842 llrint NEW: ISO C 9x
2843 llrintf NEW: ISO C 9x
2844 llrintl NEW: ISO C 9x
2845 llround NEW: ISO C 9x
2846 llroundf NEW: ISO C 9x
2847 llroundl NEW: ISO C 9x
2852 lrintf NEW: ISO C 9x
2853 lrintl NEW: ISO C 9x
2854 lround NEW: ISO C 9x
2855 lroundf NEW: ISO C 9x
2856 lroundl NEW: ISO C 9x
2858 makecontext NEW: Unix98
2859 mempcpy NEW: GNU ext.
2862 modify_ldt NEW: kernel
2866 nearbyint NEW: ISO C 9x
2867 nearbyintf NEW: ISO C 9x
2868 nearbyintl NEW: ISO C 9x
2869 netname2host NEW: Secure RPC
2870 netname2user NEW: Secure RPC
2871 nexttoward NEW: ISO C 9x
2872 nexttowardf NEW: ISO C 9x
2873 nexttowardl NEW: ISO C 9x
2877 passwd2des NEW: Secure RPC
2879 pow10f NEW: GNU ext.
2880 pow10l NEW: GNU ext.
2883 printf_size NEW: GNU ext.
2884 printf_size_info NEW: GNU ext.
2885 profil_counter REMOVED
2886 pthread_mutexattr_getkind_np REPLACED
2887 pthread_mutexattr_setkind_np REPLACED
2888 ptsname NEW: Unix98 PTY
2889 ptsname_r NEW: Unix98 PTY
2891 putchar_locked REMOVED
2892 putgrent NEW: GNU ext.
2894 putpmsg NEW: STREAMS
2895 pututxline NEW: Unix98
2899 readdir64_r NEW: LFS
2900 remquo NEW: ISO C 9x
2901 remquof NEW: ISO C 9x
2902 remquol NEW: ISO C 9x
2904 roundf NEW: ISO C 9x
2905 roundl NEW: ISO C 9x
2907 scalbln NEW: ISO C 9x
2908 scalblnf NEW: ISO C 9x
2909 scalblnl NEW: ISO C 9x
2911 sendfile NEW: kernel
2912 setcontext NEW: Unix98
2913 setrlimit64 NEW: LFS
2914 setutxent NEW: Unix98
2916 sigignore NEW: Unix98
2917 sigqueue NEW: POSIX.1b
2918 sigrelse NEW: Unix98
2919 sigset NEW: POSIX.1b
2920 sigtimedwait NEW: POSIX.1b
2921 sigwaitinfo NEW: POSIX.1b
2922 sincos NEW: GNU ext.
2923 sincosf NEW: GNU ext.
2924 sincosl NEW: GNU ext.
2928 strcasestr NEW: GNU ext.
2929 strtoimax NEW: ISO C 9x
2930 strtoumax NEW: ISO C 9x
2931 strverscmp NEW: GNU ext.
2932 svcauthdes_stats NEW: Secure RPC
2933 svcunix_create NEW: sunrpc etx
2934 svcunixfd_create NEW: sunrpc ext
2935 swapcontext NEW: Unix98
2936 tcgetsid NEW: Unix98 PTY
2937 tdestroy NEW: GNU ext.
2938 tgamma NEW: ISO C 9x
2939 tgammaf NEW: ISO C 9x
2940 tgammal NEW: ISO C 9x
2944 truncf NEW: ISO C 9x
2945 truncl NEW: ISO C 9x
2947 unlockpt NEW: Unix98 PTY
2948 updwtmpx NEW: Unix98
2949 user2netname NEW: Secure RPC
2950 utmpxname NEW: Unix98
2951 versionsort NEW: GNU ext.
2952 versionsort64 NEW: GNU ext.
2954 wcscasecmp NEW: GNU ext.
2955 wcsncasecmp NEW: GNU ext.
2956 wcsnlen NEW: GNU ext.
2957 wcstoimax NEW: ISO C 9x
2958 wcstoll NEW: ISO C 9x
2959 wcstoull NEW: ISO C 9x
2960 wcstoumax NEW: ISO C 9x
2962 wordexp NEW: POSIX.2
2963 wordfree NEW: POSIX.2
2964 write_profiling REMOVED
2965 xdecrypt NEW: Secure RPC
2966 xdr_authdes_cred NEW: Secure RPC
2967 xdr_authdes_verf NEW: Secure RPC
2968 xdr_cryptkeyarg NEW: Secure RPC
2969 xdr_cryptkeyarg2 NEW: Secure RPC
2970 xdr_cryptkeyres NEW: Secure RPC
2971 xdr_getcredres NEW: Secure RPC
2972 xdr_key_netstarg NEW: Secure RPC
2973 xdr_key_netstres NEW: Secure RPC
2974 xdr_keybuf NEW: Secure RPC
2975 xdr_keystatus NEW: Secure RPC
2976 xdr_netnamestr NEW: Secure RPC
2977 xdr_sizeof NEW: Secure RPC
2978 xdr_unixcred NEW: sunrpc ext
2979 xencrypt NEW: Secure RPC
2980 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2991 * inet_ntoa is thread-safe
2993 * updwtmp is moved from libutil to libc
2995 * rewrite of cbrt function
2997 * update of timezone data
3011 * add atoll function
3013 * fix complex problems in Berkeley DB code
3015 * fix math functions
3019 * fixed lots of header problems (especially Linux/GNU specific)
3021 * dynamic loader preserves all registers
3023 * Roland McGrath provided support for handling of auxiliary objects in
3024 the ELF dynamic loader.
3026 * support for parallel builds is improved
3030 * GNU extensions are no longer declared by default. To enable them you
3031 must define the macro `_GNU_SOURCE' in your program or compile with
3034 * The library has changed from using GNU ld symbol aliases to using weak
3035 symbols where available. The ELF object file format supports weak
3036 symbols; GNU ld also supports weak symbols in the a.out format. (There
3037 is also now support for other GNU ld extensions in ELF. Use the
3038 `--with-elf' option to configure to indicate you have ELF, and
3039 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3040 of many files which contained only symbol aliases, reducing the size of
3041 the source and the compiled library; many other files were renamed to
3042 less cryptic names previously occupied by the symbol alias files.
3043 There is a new header file <elf.h> for programs which operate on
3044 files in the ELF format.
3046 * Converted to Autoconf version 2, so `configure' has more options.
3047 Run `configure --help' to see the details.
3049 * The library can now be configured to build profiling, highly-optimized
3050 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3051 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3052 `configure' enable building these extra libraries. The shared library is
3053 built by default when using both ELF and GNU ld. When shared libraries
3054 are enabled, the new library `-ldl' is available for arbitrary run-time
3055 loading of shared objects; its interface is defined in <dlfcn.h>. The
3056 new header file <link.h> gives access to the internals of the run-time
3057 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3058 application of same name on other systems and it provides information
3059 about dynamically linked binaries.
3061 * The C library now provides the run-time support code for profiling
3062 executables compiled with `-pg'. Programs can control the profiling code
3063 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3064 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3065 the support for this file format was contributed by David Mosberger-Tang.
3067 * The math code has been replaced with a math library based on fdlibm from
3068 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3069 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3070 functions. The math functions now reside in a separate library, so
3071 programs using them will need to use `-lm' their linking commands.
3073 * John C. Bowman contributed optimized ix87 assembler inline functions.
3075 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3076 mechanism similar to that found in Solaris 2. This is now used for the
3077 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3078 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3079 file controls what services are used for each individual database. This
3080 works by loading shared libraries with names specified in `nsswitch.conf',
3081 so service modules can be changed or added at any time without even
3082 relinking any program. Currently there are the file, db, and NIS based
3083 NSS services available.
3085 * The new functions `strtoq' and `strtouq' parse integer values from
3086 strings, like `strtol' and `strtoul', but they return `long long int' and
3087 `unsigned long long int' values, respectively (64-bit quantities).
3089 * The new functions `strtof' and `strtold' parse floating-point values from
3090 strings, like `strtod', but they return `float' and `long double' values,
3091 respectively (on some machines `double' and `long double' are the same).
3093 * Ulrich Drepper has contributed new implementations of the floating-point
3094 printing and reading code used in the `printf' family of functions and
3095 `strtod', `strtof', and `strtold'. These new functions are perfectly
3096 accurate, and much faster than the old ones.
3098 * The implementation of the POSIX locale model was completely rewritten by
3099 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3100 to compile the POSIX locale definition.
3102 * The former dummy implementations of the strcoll and strxfrm function are
3103 now replaced by fully functional code contributed by Ulrich Drepper. The
3104 collation information comes from the POSIX locale definitions.
3106 * The new header <langinfo.h> defines an interface for accessing
3107 various locale-dependent data (using the locale chosen with `setlocale').
3109 * Ulrich Drepper has contributed a new suite of functions for operation on
3110 wide-character and multibyte-character strings, in <wchar.h>;
3111 and classification and case conversion of wide characters, in <wctype.h>.
3112 These new functions are conforming to the ISO C, Amendement 1 specification.
3114 * There is now a second implementation of the standard I/O library available.
3115 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3116 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3118 * You can now use positional parameter specifications in format strings
3119 for the `printf' and `scanf' families of functions. For example,
3120 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3121 ``Number 6, Mr Jones''. This is mainly useful when providing different
3122 format strings for different languages, whose grammars may dictate
3123 different orderings of the values being printed. To support this
3124 feature, the interface for `register_printf_handler' has changed; see
3125 the header file <printf.h> for details.
3127 * The `printf' and `scanf' families of functions now understand a new
3128 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3129 to group numbers as indicated by the locale; for `scanf' and friends, this
3130 says to accept as valid only a number with all the proper grouping
3131 separators in the right places. In the default "C" locale, numbers are
3132 not grouped; but locales for specific countries will define the usual
3133 conventions (i.e. separate thousands with `,' in the US locale).
3135 * The pgrp functions have been regularized, slightly incompatibly but much
3136 less confusingly. The core functions are now `getpgid' and `setpgid',
3137 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3138 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3139 provided for compatibility. There is no longer an incompatible `getpgrp'
3140 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3142 * The new header file <fts.h> and suite of functions simplify programs that
3143 operate on directory trees. This code comes from 4.4 BSD.
3145 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3146 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3147 NSS scheme used in glibc.
3149 * There is a new malloc debugging hook `__memalign_hook'.
3151 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3152 `unsigned int' in <sys/types.h>. These are for compatibility only and
3153 their use is discouraged.
3155 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3156 done differently, so that it works even without GNU ld.
3158 * New function `euidaccess' checks allowed access to a file like `access',
3159 but using the effective IDs instead of the real IDs.
3161 * The time zone data files have been updated for the latest and greatest
3162 local time conventions of the countries of the world.
3164 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3167 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3168 for formatting floating-point numbers. They are provided only for
3169 compatibility; new programs should use `sprintf' instead. There are
3170 also equivalent function for the `long double' floating-point type and
3171 all functions also exist in a reentrant form.
3173 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3174 functions for maintaining the login-record files (primarily of use to
3175 system programs such as `login'), and convenient functions for
3176 allocating and initializing a pseudo-terminal (pty) device.
3178 * Ulrich Drepper has contributed new support for System V style
3179 shared memory and IPC on systems that support it.
3181 * Ulrich Drepper has contributed several miscellaneous new functions found
3182 in System V: The `hsearch' family of functions provide an effective
3183 implementation of hash tables; `a64l' and `l64a' provide a very simple
3184 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3187 * Ulrich Drepper has contributed new reentrant counterparts for the
3188 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3190 * Ulrich Drepper has contributed new, highly-optimized versions of several
3191 string functions for the i486/Pentium family of processors.
3193 * Ulrich Drepper has updated the Linux-specific code, based largely
3194 on work done in Hongjiu Lu's version of GNU libc for Linux.
3195 The GNU library now supports Linux versions 2.0.10 and later,
3196 using the ELF object file format (i[3456]86-*-linux).
3198 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3200 * David Mosberger-Tang and Richard Henderson have ported the C library
3201 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3202 dynamic linking support for ELF/Alpha.
3204 * Richard Henderson contributed several Alpha optimized assembler function
3205 for arithmetic and string handling.
3207 * Ulrich Drepper has contributed a new set of message catalog functions to
3208 support multiple languages using the <libintl.h> interface, for use with
3209 his new package GNU gettext. Translation volunteers have contributed
3210 catalogs of the library's messages in Spanish, German, and Korean.
3212 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3213 program and the `catgets' function for reading the catalog files it
3214 creates. (The <libintl.h> interface is preferred; we include the
3215 <nl_types.h> interface using `catgets' only for source compatibility with
3216 programs already written to use it.)
3218 * New header file <values.h> gives SVID-compatible names for <limits.h>
3221 * Various new macros, declarations, and small header files for compatibility
3224 * New function `group_member' is a convenient way to check if a process has
3225 a given effective group ID.
3227 * When using GCC 2.7 and later, the socket functions are now declared in a
3228 special way so that passing an argument of type `struct sockaddr_in *',
3229 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3230 `struct sockaddr *' type, does not generate a type-clash warning.
3232 * New function `error' declared in header file <error.h> is a convenient
3233 function for printing error messages and optionally exiting; this is the
3234 canonical function used in GNU programs. The new functions `err', `warn',
3235 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3236 doing the same thing.
3238 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3239 POSIX.2 `glob' function to do ~ and {...} expansion.
3241 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3242 `clearenv' which is used in POSIX.9 is also available.
3244 * New function `getsid' returns session ID number on systems that support it.
3246 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3247 files <db.h> and <mpool.h> provide a rich set of functions for several
3248 types of simple databases stored in memory and in files, and <ndbm.h> is
3249 an old `ndbm'-compatible interface using the `db' functions. Link with
3250 `-ldb' to get these functions.
3252 * New macro `strdupa' copies a string like `strdup', but uses local stack
3253 space from `alloca' instead of dynamic heap space from `malloc'.
3255 * New function `strnlen' is like `strlen' but searches only a given maximum
3256 number of characters for the null terminator. `stpncpy', `strndup' and
3257 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3260 * New function `statfs' in header <sys/statfs.h>.
3262 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3263 provide convenient functions for operating on blocks of null-terminated
3266 * A new suite of functions in <utmp.h> handle all the details of reading
3267 and writing the utmp file.
3269 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3272 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3273 completely POSIX compliant and also implemented the extended functionality
3274 to handle alternate digit representation and alternate era date formats.
3276 * Ulrich Drepper provided an implementation of the `strptime' function
3277 defined in XPG4.2 which transforms a string into a `struct tm' value.
3279 * Paul Eggert provided the tzselect shell script as part of the timezone
3280 code. The shell script makes it easy to select the correct timezone
3283 * The implementation of the malloc family of functions is completely replaced
3284 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3285 The implementation uses the mmap function (if available) and it is
3286 optimized for the use in multi threaded programs.
3288 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3289 functions. This new functionality is usable by specifying a special
3290 salt string and it is compatible with implementation on *BSD systems.
3292 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3293 `getsubopt' to handle second level command line options, `bsd_signal'
3294 to access BSD style `signal' functionality, the obsolete `regexp' style
3297 * the `lchown' function is available on system which support this
3300 * The implementation of the shadow password handling function was contributed
3303 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3305 * POSIX.1g support was added. The <sys/select.h> header is available,
3306 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3307 implementation of `getaddrinfo'.
3311 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3313 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3314 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3316 * The new function `daemon' from 4.4 BSD is useful for server programs that
3317 want to put themselves in the background.
3319 * Joel Sherrill has contributed support for several standalone boards that
3320 run without an operating system.
3322 * `printf', `scanf' and friends now accept a `q' type modifier for long
3323 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3325 * All of the code taken from BSD (notably most of the math and networking
3326 routines) has been updated from the BSD 4.4-Lite release.
3328 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3330 * The new functions `getdomainname' and `setdomainname' fetch or change the
3331 YP/NIS domain name. These are system calls which exist on systems which
3334 * The time zone data files have been updated for the latest international
3337 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3338 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3342 * The C library now includes support for Sun RPC, from Sun's free
3343 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3344 are included. (There is still no support for YP.)
3346 * Tom Quinn has contributed a port of the C library to SGI machines running
3347 Irix 4 (mips-sgi-irix4).
3349 * The new `lockf' function is a simplified interface to the locking
3350 facilities of `fcntl', included for compatibility.
3352 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3354 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3356 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3359 * The error message printed by `assert' for a failed assertion now includes
3360 the name of the program (if using GNU ld) and the name of the calling
3361 function (with versions of GCC that support this).
3363 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3365 * The library now includes the <sys/mman.h> header file and memory
3366 management functions `mmap', `munmap', `mprotect', `msync', and
3367 `madvise', on systems that support those facilities.
3369 * The interface for `mcheck' has changed slightly: the function called to
3370 abort the program when an allocation inconsistency is detected now takes
3371 an argument that indicates the type of failure. The new function
3372 `mprobe' lets you request a consistency check for a particular block at
3373 any time (checks are normally done only when you call `free' or `realloc'
3376 * It is now possible to easily cross-compile the C library, building on one
3377 system a library to run on another machine and/or operating system. All
3378 you need to do is set the variable `HOST_CC' in `configparms' to the
3379 native compiler for programs to run on the machine you are building on (a
3380 few generator programs are used on Unix systems); set `CC' to the
3383 * The new function `fexecve' (only implemented on the GNU system) executes
3384 a program file given a file descriptor already open on the file.
3388 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3389 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3391 * You can set the variable `libprefix' in `configparms' to specify a prefix
3392 to be prepended to installed library files; this makes it easy to install
3393 the GNU C library to be linked as `-lgnuc' or whatever.
3395 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3396 copies a limited number of characters from a string, and returns the
3397 address of the last character written.
3399 * You no longer need to check for whether the installed `stddef.h' is
3400 compatible with the GNU C library. configure now checks for you.
3402 * You can now define a per-stream `fileno' function to convert the
3403 stream's cookie into an integral file descriptor.
3405 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3406 allocates zero bytes of storage, and returns a unique pointer which
3407 you can pass to `realloc' or `free'. The behavior is undefined if
3408 you dereference this pointer.
3410 * The C library now runs on Sony NEWS m68k machines running either
3411 NewsOS 3 or NewsOS 4.
3413 * The new `syscall' function is a system-dependent primitive function
3414 for invoking system calls. It has the canonical behavior on Unix
3415 systems, including unreliable return values for some calls (such as
3416 `pipe', `fork' and `getppid').
3418 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3419 to `EAGAIN', which is the preferred name. On systems whose kernels
3420 use two distinct codes, the C library now translates EWOULDBLOCK to
3421 EAGAIN in every system call function.
3425 * The GNU C Library Reference Manual is now distributed with the library.
3426 `make dvi' will produce a DVI file of the printed manual.
3427 `make info' will produce Info files that you can read on line using C-h i
3428 in Emacs or the `info' program.
3429 Please send comments on the manual to bug-glibc-manual@gnu.org.
3431 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3433 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3435 * Jason Merrill has contributed a port to the Sequent Symmetry running
3436 Dynix version 3 (i386-sequent-dynix).
3438 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3439 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3441 * New function `memory_warnings' lets you arrange to get warnings when
3442 malloc is running out of memory to allocate, like Emacs gives you.
3444 * The C library now contains the relocating allocator used in Emacs 19 for
3445 its editing buffers. This allocator (ralloc) minimizes allocation
3446 overhead and fragmentation by moving allocated regions around whenever it
3447 needs to. You always refer to a ralloc'd region with a "handle" (a
3448 pointer to a pointer--an object of type `void **').
3450 * There is a new `printf' format: `%m' gives you the string corresponding
3451 to the error code in `errno'.
3453 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3454 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3455 pass, the `a' modifier says to fill in a `char **' you pass with a
3458 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3459 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3460 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3462 * `mkstemp' is a traditional Unix function to atomically create and open a
3463 uniquely-named temporary file.
3467 * The standard location for the file that says what the local timezone is
3468 has changed again. It is now `/usr/local/etc/localtime' (or more
3469 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3471 * The distribution no longer contains any files with names longer than 14
3474 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3475 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3477 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3481 * The configuration process has changed quite a bit. The `configure'
3482 script is now used just like the configuration scripts for other GNU
3483 packages. The `sysdeps' directory hierarchy is much rearranged.
3484 The file `INSTALL' explains the new scheme in detail.
3486 * The header files no longer need to be processed into ANSI C and
3487 traditional C versions. There is just one set of files to install, and
3488 it will work with ANSI or old C compilers (including `gcc -traditional').
3490 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3491 MIPS DECStation running Ultrix 4.
3493 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3494 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3495 made itself into a shared library.
3497 * Yet further improved support for the i386, running 4.3 BSD-like systems
3498 (such as Mach 3 with the Unix single-server), or System V.
3500 * New function `strncasecmp' to do case-insensitive string comparison
3501 with limited length.
3503 * New function `strsep' is a reentrant alternative to `strtok'.
3505 * New functions `scandir' and `alphasort' for searching directories.
3507 * New function `setenv' is a better interface to `putenv'.
3509 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3510 function for traversing a directory tree.
3512 * The GNU obstack package is now also part of the C library.
3513 The new function `open_obstack_stream' creates a stdio stream that
3514 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3515 formatted output directly to an obstack.
3517 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3518 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3520 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3522 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3523 things to your strings.
3525 * There are some new test programs: test-fseek, testmb, and testrand.
3527 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3528 These ports are not finished, but are a good starting place for really
3529 supporting those systems.
3531 * `/etc/localtime' is now the standard location for the file that says what
3532 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3533 This follows the general principle that `/etc' is the place for all local
3534 configuration files.
3536 * The C library header files now use `extern "C"' when used by the C++
3537 compiler, so the C library should now work with C++ code.
3539 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3540 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3543 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3544 function which uses the `merge sort' algorithm, and is said to be
3545 significantly faster than the old GNU `qsort' function. Merge sort is
3546 now the standard `qsort' function. The new algorithm can require a lot
3547 of temporary storage; so, the old sorting function is called when the
3548 required storage is not available.
3550 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3551 provides the Unix `crypt' function, plus some other entry points.
3553 * The code and header files taken from 4.4 BSD have been updated with the
3554 latest files released from Berkeley.
3556 ----------------------------------------------------------------------
3557 Copyright information:
3559 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3561 Permission is granted to anyone to make or distribute verbatim copies
3562 of this document as received, in any medium, provided that the
3563 copyright notice and this permission notice are preserved,
3564 thus giving the recipient permission to redistribute in turn.
3566 Permission is granted to distribute modified versions
3567 of this document, or of portions of it,
3568 under the above conditions, provided also that they
3569 carry prominent notices stating who last changed them.
3572 version-control: never