1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2017 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, fromfp,
70 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
71 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
73 - llogb functions: the llogb, llogbf and llogbl functions, and the
74 FP_LLOGB0 and FP_LLOGBNAN macros.
76 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
79 - Comparison macros: iseqsig.
81 - Classification macros: iscanonical, issubnormal, iszero.
83 - Total order functions: totalorder, totalorderf, totalorderl,
84 totalordermag, totalordermagf, totalordermagl.
86 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
88 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
89 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
91 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
92 are added to libc. They convert a floating-point number into string.
94 * Most of glibc can now be built with the stack smashing protector enabled.
95 It is recommended to build glibc with --enable-stack-protector=strong.
96 Implemented by Nick Alcock (Oracle).
98 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
99 intended to be used instead of memset() to erase sensitive data after use;
100 the compiler will not optimize out calls to explicit_bzero even if they
101 are "unnecessary" (in the sense that no _correct_ program can observe the
102 effects of the memory clear).
104 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
105 to float instead of double. This does not affect the ABI of any libraries
106 that are part of the GNU C Library, but may affect the ABI of other
107 libraries that use this type in their interfaces.
109 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
110 float_t and double_t types are now defined to long double instead of float
111 and double. These options are not the default, and this does not affect
112 the ABI of any libraries that are part of the GNU C Library, but it may
113 affect the ABI of other libraries that use this type in their interfaces,
114 if they are compiled or used with those options.
116 * The getentropy and getrandom functions, and the <sys/random.h> header file
119 * The buffer size for byte-oriented stdio streams is now limited to 8192
120 bytes by default. Previously, on Linux, the default buffer size on most
121 file systems was 4096 bytes (and thus remains unchanged), except on
122 network file systems, where the buffer size was unpredictable and could be
123 as large as several megabytes.
125 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
126 for the Linux quota interface which predates kernel version 2.4.22 has
129 * The malloc_get_state and malloc_set_state functions have been removed.
130 Already-existing binaries that dynamically link to these functions will
131 get a hidden implementation in which malloc_get_state is a stub. As far
132 as we know, these functions are used only by GNU Emacs and this change
133 will not adversely affect already-built Emacs executables. Any undumped
134 Emacs executables, which normally exist only during an Emacs build, should
135 be rebuilt by re-running “./configure; make” in the Emacs build tree.
137 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
138 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
139 “no-ip6-dotint” had already been the default, and support for the
140 “ip6-dotint” option was removed from the Internet in 2006.
142 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
143 flag from <resolv.h> have been removed. The option relied on a
144 backwards-incompatible DNS extension which was never deployed on the
147 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
148 RES_BLAST defined in the <resolv.h> header file have been deprecated.
149 They were already unimplemented.
151 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
152 _res.flags are deprecated. The flag was standardized in RFC 2133, but
153 removed again from the IETF name lookup interface specification in RFC
154 2553. Applications should use getaddrinfo instead.
156 * DNSSEC-related declarations and definitions have been removed from the
157 <arpa/nameser.h> header file, and libresolv will no longer attempt to
158 decode the data part of DNSSEC record types. Previous versions of glibc
159 only implemented minimal support for the previous version of DNSSEC, which
160 is incompatible with the currently deployed version.
162 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
163 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
164 <arpa/nameser.h> header file because the distinction between RR types and
165 meta-RR types is not officially standardized, subject to revision, and
166 thus not suitable for encoding in a macro.
168 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
169 and rhook members of the res_state type in <resolv.h> have been removed.
170 The glibc stub resolver did not support these hooks, but the header file
171 did not reflect that.
173 * For multi-arch support it is recommended to use a GCC which has
174 been built with support for GNU indirect functions. This ensures
175 that correct debugging information is generated for functions
176 selected by IFUNC resolvers. This support can either be enabled by
177 configuring GCC with '--enable-gnu-indirect-function', or by
178 enabling it by default by setting 'default_gnu_indirect_function'
179 variable for a particular architecture in the GCC source file
182 * GDB pretty printers have been added for mutex and condition variable
183 structures in POSIX Threads. When installed and loaded in gdb these pretty
184 printers show various pthread variables in human-readable form when read
185 using the 'print' or 'display' commands in gdb.
187 * Tunables feature added to allow tweaking of the runtime for an application
188 program. This feature can be enabled with the '--enable-tunables' configure
189 flag. The GNU C Library manual has details on usage and README.tunables has
190 instructions on adding new tunables to the library.
192 Security related changes:
194 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
195 have been created with makecontext could fail to terminate due to a
196 missing .cantunwind annotation. This has been observed to lead to a hang
197 (denial of service) in some Go applications compiled with gccgo. Reported
198 by Andreas Schwab. (CVE-2016-6323)
200 * The DNS stub resolver functions would crash due to a NULL pointer
201 dereference when processing a query with a valid DNS question type which
202 was used internally in the implementation. The stub resolver now uses a
203 question type which is outside the range of valid question type values.
206 The following bugs are resolved with this release:
208 [The release manager will add the list generated by
209 scripts/list-fixed-bugs.py just before the release.]
213 * The minimum Linux kernel version that this version of the GNU C Library
214 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
215 kernel version 2.6.32 or later suffices (on architectures that already
216 required kernel versions more recent than 3.2, those requirements remain
217 unchanged). Linux 3.2 or later kernel headers are required on all
220 * The pap_AN locale has been deleted. This has been deprecated for a long
221 time. It has been replaced by pap_AW & pap_CW, both of which have long
222 been included in previous releases.
224 * The readdir_r and readdir64_r functions have been deprecated. It is
225 recommended to use readdir and readdir64 instead.
227 * The type “union wait” has been removed. It was deprecated in the early
228 1990s and never part of POSIX. Application code should use the int type
229 instead of “union wait”.
231 * A new NSS action is added to facilitate large distributed system
232 administration. The action, MERGE, allows remote user stores like LDAP
233 to be merged into local user stores like /etc/groups in order to provide
234 easy to use, updated, and managed sets of merged credentials. The new
235 action can be used by configuring it in /etc/nsswitch.conf:
236 group: files [SUCCESS=merge] nis
237 Implemented by Stephen Gallagher (Red Hat).
239 * The deprecated __malloc_initialize_hook variable has been removed from the
242 * The long unused localedef --old-style option has been removed. It hasn't
243 done anything in over 16 years. Scripts using this option can safely
246 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
247 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
248 return the next representable value in the direction of positive infinity
249 and the nextdown functions return the next representable value in the
250 direction of negative infinity. These are currently enabled as GNU
253 Security related changes:
255 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
256 could result in a stack overflow when getnetbyname was called with an
257 overly long name. (CVE-2016-3075)
259 * Previously, getaddrinfo copied large amounts of address data to the stack,
260 even after the fix for CVE-2013-4458 has been applied, potentially
261 resulting in a stack overflow. getaddrinfo now uses a heap allocation
262 instead. Reported by Michael Petlan. (CVE-2016-3706)
264 * The glob function suffered from a stack-based buffer overflow when it was
265 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
266 Reported by Alexander Cherepanov. (CVE-2016-1234)
268 * The Sun RPC UDP client could exhaust all available stack space when
269 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
270 alloca plugin for GCC. (CVE-2016-4429)
272 * The IPv6 name server management code in libresolv could result in a memory
273 leak for each thread which is created, performs a failing naming lookup,
274 and exits. Over time, this could result in a denial of service due to
275 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
277 The following bugs are resolved with this release:
279 [1170] localedata: ne_NP: update Nepali locale definition file
280 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
282 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
283 [6796] math: fdim() does not set errno on overflow
284 [10354] libc: posix_spawn should use vfork() in more cases than presently
285 [11213] localedata: localedata: add copyright disclaimer to locale files
286 [12143] localedata: chr_US: new Cherokee locale
287 [12450] localedata: sgs_LT: new locale
288 [12676] localedata: ln_CD: new locale
289 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
291 [13304] math: fma, fmaf, fmal produce wrong results
292 [14259] build: --localedir arg to configure is ignored
293 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
294 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
295 [14934] localedata: es_CL: wrong first weekday chilean locale
296 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
298 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
300 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
301 [15368] nptl: raise() is not async-signal-safe
302 [15479] math: ceil, floor, round and trunc raise inexact exception
303 [15578] localedata: kk_KZ: various updates
304 [16003] localedata: pap_AN: punt old locale
305 [16137] localedata: iw_IL: punt old locale
306 [16190] localedata: eo: new esperanto locale
307 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
308 [16742] malloc: race condition: pthread_atfork() called before first
309 malloc() results in unexpected locking behaviour/deadlocks
310 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
312 [16983] localedata: postal_fmt does not allow %l and %n modifiers
313 [17565] localedata: pt_PT: wrong (work-)week start
314 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
315 [17950] build: Build fails with -msse
316 [18205] localedata: be_BY*: wrong first_weekday and first_workday
317 [18433] libc: posix_spawn does not return correctly upon failure to
319 [18453] localedata: charmaps/IBM875: incorrect codes
320 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
322 [18896] localedata: he_IL: improvements for currency
323 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
325 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
327 [19056] libc: Deprecate readdir_r
328 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
330 [19198] localedata: nl_NL: small improvements for Dutch locales
331 [19257] network: Per-thread memory leak in __res_vinit with IPv6
332 nameservers (CVE-2016-5417)
333 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
335 [19400] locale: Language missing in "iso-639.def", trivial fix in
337 [19431] malloc: Deadlock between fflush, getdelim, and fork
338 [19505] libc: Incorrect file descriptor validity checks in
339 posix_spawn_file_actions_add{open,close,dup2}
340 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
342 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
343 `intl/{gettextP,loadinfo}.h'
344 [19534] libc: execle, execlp may use malloc
345 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
346 [19573] network: res_nclose and __res_maybe_init disagree about name
347 server initialization, breaking Hesiod
348 [19575] localedata: Status of GB18030 tables
349 [19581] localedata: sr_* date_fmt string contains additional newline
350 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
352 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
353 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
354 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
355 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
357 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
359 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
361 [19610] dynamic-link: ldconfig -X removes stale symbolic links
362 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
363 [19633] locale: strfmon_l applies global locale to number formatting
364 [19642] network: Memory leak in getnameinfo
365 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
366 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
368 [19654] math: [x86_64] Need testcase for BZ #19590 fix
369 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
371 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
372 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
374 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
376 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
378 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
379 pointers and lengths in error-case.
380 [19727] locale: Converting from/to UTF-xx with iconv() does not always
381 report errors on UTF-16 surrogates values.
382 [19755] nscd: nscd assertion failure in gc
383 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
384 [19759] libc: mempcpy shouldn't be inlined
385 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
386 [19765] libc: s390 needs an optimized mempcpy
387 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
388 NAME_MAX limit assumption (CVE-2016-1234)
389 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
390 [19787] network: Missing and incorrect truncation checks in getnameinfo
391 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
393 [19791] network: Assertion failure in res_query.c with un-connectable name
395 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
396 [19822] math: libm.so install clobbers old version
397 [19825] network: resolv: send_vc can return uninitialized data in second
398 response to getaddrinfo
399 [19830] network: nss_dns: should check RDATA length against buffer length
400 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
401 confronted with A/AAAA records of invalid size
402 [19837] nss: nss_db: No retries for some long lines with a larger buffer
403 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
404 [19853] stdio: Printing IBM long double in decimal with high precision is
406 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
407 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
408 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
410 [19865] network: Assertion failure or memory leak in
411 _nss_dns_getcanonname_r
412 [19868] network: nss_dns: netent code does not skip over non-PTR records
413 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
415 [19881] string: Improve x86-64 memset
416 [19907] string: Incorrect memcpy tests
417 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
419 [19925] libc: termios.h XCASE namespace
420 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
421 [19929] libc: limits.h NL_NMAX namespace
422 [19931] stdio: Memory leak in vfprintf
423 [19957] libc: clone(CLONE_VM) access invalid parent memory
424 [19963] localedata: en_IL: New locale
425 [19989] stdio: stdio.h cuserid namespace
426 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
428 [19996] locale: langinfo.h nl_langinfo_l namespace
429 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
430 [20010] network: getaddrinfo: Stack overflow in hostent translation
432 [20012] stdio: libio: fmemopen append mode failure
433 [20014] stdio: stdio.h namespace for pre-threads POSIX
434 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
435 [20023] libc: fcntl.h timespec namespace
436 [20024] math: [x86_64] vectorized sincos trashes the stack
437 [20031] network: nss_hesiod: Heap overflow in get_txt_records
438 [20041] time: sys/time.h timespec namespace
439 [20043] libc: unistd.h missing cuserid for UNIX98 and before
440 [20044] libc: unistd.h missing pthread_atfork for UNIX98
441 [20051] libc: ttyslot in wrong header under wrong conditions
442 [20054] libc: gethostname not declared for XPG4
443 [20055] libc: termios.h missing tcgetsid for XPG4
444 [20072] dynamic-link: x86 init_cpu_features is called twice in static
446 [20073] libc: sys/stat.h fchmod namespace
447 [20074] libc: stdlib.h rand_r namespace
448 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
449 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
451 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
452 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
454 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
455 [20119] libc: Wrong mask for processors level type from CPUID
456 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
457 AS not supporting AVX512
458 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
460 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
461 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
463 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
464 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
465 [20178] libc: posix_spawn{p} should not call exit
466 [20191] stdio: libio: vtables hardening
467 [20195] string: FMA4 detection requires CPUID execution with register
469 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
470 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
472 [20212] math: acos (sNaN) returns sNaN
473 [20213] math: asin (sNaN) returns sNaN
474 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
475 [20218] math: [i386] asinhl (sNaN) returns sNaN
476 [20219] math: [i386] atanhl (sNaN) returns sNaN
477 [20222] stdio: fopencookie: Mangle function pointers
478 [20224] math: [i386] cbrtl (sNaN) returns sNaN
479 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
480 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
482 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
483 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
484 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
485 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
486 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
487 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
488 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
489 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
490 [20240] math: modf (sNaN) returns sNaN
491 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
492 [20250] math: frexp (sNaN) returns sNaN
493 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
494 [20255] math: [i386] fdim, fdimf return with excess range and precision /
496 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
497 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
498 subscript is below array bounds [-Werror=array-bounds]
499 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
500 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
502 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
503 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
505 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
507 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
508 [20316] localedata: id_ID: Februari instead of Pebruari
509 [20327] string: POWER8 strcasecmp returns incorrect result
510 [20347] math: Failure: Test: j0_downward (0xap+0)
511 [20348] libc: FAIL: misc/tst-preadvwritev64
512 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
513 [20350] libc: There is no test for p{read,write}64
514 [20357] math: Incorrect cos result for 1.5174239687223976
515 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
519 * Unicode 8.0.0 Support: Character encoding, character type info, and
520 transliteration tables are all updated to Unicode 8.0.0, using new
521 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
522 These updates cause user visible changes, such as the fixes for bugs
523 89, 16061, and 18568.
525 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
526 kernel-internal CPU set size. This means that requests that change the
527 CPU affinity which failed before (for example, an all-ones CPU mask) will
528 now succeed. Applications that need to determine the effective CPU
529 affinities need to call sched_getaffinity or pthread_getaffinity_np after
530 setting it because the kernel can adjust it (and the previous size check
531 would not detect this in the majority of cases).
533 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
534 the following new symbols are used: fts64_children, fts64_close,
535 fts64_open, fts64_read and fts64_set.
537 * getaddrinfo now detects certain invalid responses on an internal netlink
538 socket. If such responses are received, an affected process will
539 terminate with an error message of "Unexpected error <number> on netlink
540 descriptor <number>" or "Unexpected netlink response of size <number> on
541 descriptor <number>". The most likely cause for these errors is a
542 multi-threaded application which erroneously closes and reuses the netlink
543 file descriptor while it is used by getaddrinfo.
545 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
546 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
547 unnecessary serialization of memory allocation requests across threads.
548 The defect is now corrected. Users should see a substantial increase in
549 the concurent throughput of allocation requests for applications which
550 trigger this bug. Affected applications typically create and
551 destroy threads frequently. (Bug 19048 was reported and analyzed by
554 * There is now a --disable-timezone-tools configure option for disabling the
555 building and installing of the timezone related utilities (zic, zdump, and
556 tzselect). This is useful for people who build the timezone data and code
557 independent of the GNU C Library.
559 * The obsolete header <regexp.h> has been removed. Programs that require
560 this header must be updated to use <regex.h> instead.
562 * The obsolete functions bdflush, create_module, get_kernel_syms,
563 query_module and uselib are no longer available to newly linked binaries;
564 the header <sys/kdaemon.h> has been removed. These functions and header
565 were specific to systems using the Linux kernel and could not usefully be
566 used with the GNU C Library on systems with version 2.6 or later of the
569 * Optimized string, wcsmbs and memory functions for IBM z13.
570 Implemented by Stefan Liebler.
572 * Newly linked programs that define a variable called signgam will no longer
573 have it set by the lgamma, lgammaf and lgammal functions. Programs that
574 require signgam to be set by those functions must ensure that they use the
575 variable provided by the GNU C Library and declared in <math.h>, without
576 defining their own copy.
578 * The minimum GCC version that can be used to build this version of the GNU
579 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
580 still be used to compile programs using the GNU C Library.
582 Security related changes:
584 * An out-of-bounds value in a broken-out struct tm argument to strftime no
585 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
587 * The LD_POINTER_GUARD environment variable can no longer be used to disable
588 the pointer guard feature. It is always enabled. Previously,
589 LD_POINTER_GUARD could be used to disable security hardening in binaries
590 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
593 * An integer overflow in hcreate and hcreate_r could lead to an
594 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
596 * The catopen function no longer has unbounded stack usage. Reported by
599 * The nan, nanf and nanl functions no longer have unbounded stack usage
600 depending on the length of the string passed as an argument to the
601 functions. Reported by Joseph Myers. (CVE-2014-9761)
603 * A stack-based buffer overflow was found in libresolv when invoked from
604 libnss_dns, allowing specially crafted DNS responses to seize control
605 of execution flow in the DNS client. The buffer overflow occurs in
606 the functions send_dg (send datagram) and send_vc (send TCP) for the
607 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
608 family. The use of AF_UNSPEC triggers the low-level resolver code to
609 send out two parallel queries for A and AAAA. A mismanagement of the
610 buffers used for those queries could result in the response of a query
611 writing beyond the alloca allocated buffer created by
612 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
613 the overflow. Thanks to the Google Security Team and Red Hat for
614 reporting the security impact of this issue, and Robert Holiday of
615 Ciena for reporting the related bug 18665. (CVE-2015-7547)
617 The following bugs are resolved with this release:
619 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
620 [887] math: Math library function "logb" and "nextafter" inconsistent
621 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
622 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
623 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
624 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
626 [4404] localedata: German translation of "Alarm clock" is misleading
627 [6799] math: nextafter() and nexttoward() doen't set errno on
628 overflow/underflow errors
629 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
631 [10432] nis: _nss_nis_setnetgrent assertion failure
632 [11460] libc: fts has no LFS support
633 [12926] network: getaddrinfo()/make_request() may spin forever
634 [13065] nptl: Race condition in pthread barriers
635 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
636 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
638 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
639 [14912] libc: Rename non-installed bits/*.h headers
640 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
641 [15367] math: Let gcc use __builtin_isinf
642 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
643 [15421] math: lgamma wrongly sets signgam for ISO C
644 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
645 FE_INVALID with argument out of range
646 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
648 [15786] dynamic-link: ifunc resolver functions can smash function
650 [15918] math: Unnecessary check for equality in hypotf()
651 [16061] localedata: Review / update transliteration data
652 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
654 [16141] time: strptime %z offset restriction
655 [16171] math: drem should be alias of remainder
656 [16296] math: fegetround is pure?
657 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
658 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
660 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
661 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
663 [16422] math: [powerpc] math-float, math-double failing llrint tests with
664 "Exception "Inexact" set" on ppc32
665 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
666 [16517] math: Missing underflow exception from tanf/tan/tanl
667 [16519] math: Missing underflow exception from sinhf
668 [16520] math: Missing underflow exception from tanhf
669 [16521] math: Missing underflow exception from exp2
670 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
672 [16734] stdio: fopen calls mmap to allocate its buffer
673 [16961] math: nan function incorrect handling of bad sequences
674 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
675 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
676 [16985] locale: localedef: confusing error message when opening output
678 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
679 [17197] locale: Redundant shift character in iconv conversion output at
681 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
682 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
683 [17250] dynamic-link: static linking breaks nss loading
684 (getaddrinfo/getpwnam/etc...)
685 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
687 [17441] math: isnan() should use __builtin_isnan() in GCC
688 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
689 (related to lock elision)
690 [17787] manual: Exponent on page 324 of the PDF ends prematurely
691 [17886] time: strptime should be able to parse "Z" as a timezone with %z
692 [17887] time: strptime should be able to parse "+01:00" style timezones
693 [17905] libc: catopen() Multiple unbounded stack allocations
695 [18084] libc: backtrace (..., 0) dumps core on x86
696 [18086] libc: nice() sets errno to 0 on success
697 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
698 is too large (CVE-2015-8778)
699 [18251] dynamic-link: SONAME missing when audit modules provides path
700 [18265] libc: add attributes for wchar string and memory functions
701 [18370] math: csqrt missing underflows
702 [18421] libc: [hppa] read-only segment has dynamic relocations
703 [18472] libc: Obsolete syscall wrappers should be compat symbols
704 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
705 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
706 [18525] localedata: Remove locale timezone information
707 [18560] libc: [powerpc] spurious bits/ipc.h definitions
708 [18568] localedata: Update locale data to Unicode 8.0
709 [18589] locale: sort-test.sh fails at random
710 [18595] math: ctan, ctanh missing underflows
711 [18604] libc: assert macro-expands its argument
712 [18610] math: S390: fetestexcept() reports any exception if DXC-code
713 contains a vector instruction exception.
714 [18611] math: j1, jn missing errno setting on underflow
715 [18618] localedata: sync Chechen locale definitions with other *_RU
717 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
718 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
719 [18665] network: In send_dg, the recvfrom function is NOT always using the
720 buffer size of a newly created buffer (CVE-2015-7547)
721 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
723 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
725 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
726 [18699] math: tilegx cproj() for various complex infinities does not yield
728 [18724] libc: Harden put*ent functions against data injection
729 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
730 [18755] build: build errors with -DNDEBUG
731 [18757] stdio: fmemopen fails to set errno on failure
732 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
734 [18781] libc: openat64 lacks O_LARGEFILE
735 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
736 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
737 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
738 [18790] math: [ldbl-128ibm] tanhl inaccurate
739 [18795] libc: stpncpy fortification misses buffer lengths that are
741 [18796] build: build fails for --disable-mathvec
742 [18803] math: hypot missing underflows
743 [18820] stdio: fmemopen may leak memory on failure
744 [18823] math: csqrt spurious underflows
745 [18824] math: fma spurious underflows
746 [18825] math: pow missing underflows
747 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
748 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
749 [18870] build: sem_open.c fails to compile with missing symbol
751 [18872] stdio: Fix memory leak in printf_positional
752 [18873] libc: posix_fallocate overflow check ineffective
753 [18875] math: Excess precision leads incorrect libm
754 [18877] libc: arm: mmap offset regression
755 [18887] libc: memory corruption when using getmntent on blank lines
756 [18918] localedata: hu_HU: change time to HH:MM:SS format
757 [18921] libc: Regression: extraneous stat() and fstat() performed by
759 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
760 binaries (CVE-2015-8777)
761 [18951] math: tgamma missing underflows
762 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
764 [18953] localedata: lt_LT: change currency symbol to the euro
765 [18956] math: powf inaccuracy
766 [18961] math: [i386] exp missing underflows
767 [18966] math: [i386] exp10 missing underflows
768 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
769 [18969] build: multiple string test failures due to missing locale
771 [18970] libc: Reference of pthread_setcancelstate in libc.a
772 [18977] math: float / long double Bessel functions not in XSI POSIX
773 [18980] math: i386 libm functions return with excess range and precision
774 [18981] math: i386 scalb*, ldexp return with excess range and precision
775 [18982] stdio: va_list and vprintf
776 [18985] time: Passing out of range data to strftime() causes a segfault
778 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
779 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
780 [19012] locale: iconv_open leaks memory on error path
781 [19016] math: clog, clog10 inaccuracy
782 [19018] nptl: Mangle function pointers in tls_dtor_list
783 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
784 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
785 [19048] malloc: malloc: arena free list can become cyclic, increasing
787 [19049] math: [powerpc] erfc incorrect zero sign
788 [19050] math: [powerpc] log* incorrect zero sign
789 [19058] math: [x86_64] Link fail with -fopenmp and -flto
790 [19059] math: nexttoward overflow incorrect in non-default rounding modes
791 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
792 [19074] network: Data race in _res_hconf_reorder_addrs
793 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
794 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
795 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
797 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
799 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
800 [19086] manual: posix_fallocate64 documented argument order is wrong.
801 [19088] math: lround, llround missing exceptions close to overflow
803 [19094] math: lrint, llrint missing exceptions close to overflow threshold
804 [19095] math: dbl-64 lrint incorrect for 64-bit long
805 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
806 [19124] dynamic-link: ld.so failed to build with older assmebler
807 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
808 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
809 [19134] math: [powerpc32] lround, lroundf spurious exceptions
810 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
811 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
812 pthread_setaffinity_np
813 [19156] math: [ldbl-128] j0l spurious underflows
814 [19164] nptl: tst-getcpu fails with many possible CPUs
815 [19168] math: math/test-ildoubl and math/test-ldouble failure
816 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
817 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
819 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
820 (FE_NOMASK_ENV) do not clear SSE exceptions
821 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
822 _int_new_arena/reused_arena
823 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
824 [19201] math: dbl-64 remainder incorrect sign of zero result
825 [19205] math: bits/math-finite.h conditions do not match math.h and
827 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
828 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
829 for C99-based standards
830 [19212] libc: features.h not -Wundef clean
831 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
833 [19214] libc: Family and model identification for AMD CPU's are incorrect.
834 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
835 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
837 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
839 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
841 [19242] libc: strtol incorrect in Turkish locales
842 [19243] malloc: reused_arena can pick an arena on the free list, leading
843 to an assertion failure and reference count corruption
844 [19253] time: tzset() ineffective when temporary TZ did not include DST
846 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
847 [19270] math: [hppa] Shared libm missing __isnanl
848 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
849 MAP_HUGETLB and MAP_STACK defines
850 [19313] nptl: Wrong __cpu_mask for x32
851 [19347] libc: grantpt: try to force a specific gid even without pt_chown
852 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
853 [19350] math: [ldbl-128ibm] sinhl spurious overflows
854 [19351] math: [ldbl-128ibm] logl inaccurate near 1
855 [19363] time: x32: times() return value wrongly truncates/sign extends
857 [19367] dynamic-link: Improve branch prediction on Silvermont
858 [19369] network: Default domain name not reset by res_ninit when "search"
859 / "domain" entry is removed from resolv.conf
860 [19375] math: powerpc: incorrect results for POWER7 logb with negative
862 [19385] localedata: bg_BG: time separator should be colon, not comma
863 [19408] libc: linux personality syscall wrapper may erroneously return an
864 error on 32-bit architectures
865 [19415] libc: dladdr returns wrong names on hppa
866 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
867 IBM905, IBM907, and IBM909
868 [19439] math: Unix98 isinf and isnan functions conflict with C++11
869 [19443] build: build failures with -DDEBUG
870 [19451] build: Make check fails on test-double-vlen2
871 [19462] libc: Glibc failed to build with -Os
872 [19465] math: Wrong code with -Os
873 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
875 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
877 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
878 [19486] math: S390: Math tests fail with "Exception Inexact set".
879 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
880 [19550] libc: [mips] mmap negative offset handling inconsistent with other
882 [19590] math: Fail to build shared objects that use libmvec.so functions.
886 * The following bugs are resolved with this release:
888 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
889 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
890 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
891 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
892 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
893 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
894 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
895 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
896 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
897 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
898 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
899 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
900 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
901 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
902 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
903 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
904 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
905 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
906 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
907 18657, 18676, 18694, 18696, 18887.
909 * Cache information can be queried via sysconf() function on s390 e.g. with
910 _SC_LEVEL1_ICACHE_SIZE as argument.
912 * A buffer overflow in gethostbyname_r and related functions performing DNS
913 requests has been fixed. If the NSS functions were called with a
914 misaligned buffer, the buffer length change due to pointer alignment was
915 not taken into account. This could result in application crashes or,
916 potentially arbitrary code execution, using crafted, but syntactically
917 valid DNS responses. (CVE-2015-1781)
919 * The time zone file parser has been made more robust against crafted time
920 zone files, avoiding heap buffer overflows related to the processing of
921 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
922 large time zone data files. Overly long time zone specifiers in the TZ
923 variable no longer result in stack overflows and crashes.
925 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
926 for LD and GD on x86 and x86-64, has been implemented. You will need
927 binutils-2.24 or later to enable this optimization.
929 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
930 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
931 Hat). These updates cause user visible changes, such as the fix for bug
934 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
935 and getXXbyYY NSS calls for the same database, causing a denial-of-service
936 condition in some applications.
938 * Added vector math library named libmvec with the following vectorized x86_64
939 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
941 The library can be disabled with --disable-mathvec. Use of the functions is
942 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
943 Shared library libmvec.so is linked in as needed when using -lm (no need to
944 specify -lmvec explicitly for not static builds).
945 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
947 * A new fmemopen implementation has been added with the goal of POSIX
948 compliance. The new implementation fixes the following long-standing
949 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
950 old implementation is still present for use be by existing binaries.
952 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
953 releases. It has been fixed to match 2.19 and older, but binaries built
954 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
956 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
957 Contributed by Roland McGrath (Google).
959 * The header <regexp.h> is deprecated, and will be removed in a future
960 release. Use of this header will trigger a deprecation warning.
961 Application developers should update their code to use <regex.h> instead.
963 This header was formerly part of SUS, but was deprecated in 1994 and
964 removed from the standard in 2001. Also, the glibc implementation
965 leaks memory. See BZ#18681 for more details.
969 * The following bugs are resolved with this release:
971 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
972 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
973 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
974 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
975 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
976 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
977 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
978 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
979 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
980 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
983 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
984 too little memory for the to-be-scanned arguments and overflow the
985 allocated buffer. The implementation now correctly computes the required
986 buffer size when using malloc, and switches to malloc from alloca as
989 * A new semaphore algorithm has been implemented in generic C code for all
990 machines. Previous custom assembly implementations of semaphore were
991 difficult to reason about or ensure that they were safe. The new version
992 of semaphore supports machines with 64-bit or 32-bit atomic operations.
993 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
994 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
996 * Port to Altera Nios II has been contributed by Mentor Graphics.
998 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
999 implementations for powerpc64/powerpc64le.
1000 Implemented by Adhemerval Zanella (IBM).
1002 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1003 and powerpc64le. This may improve lock scaling of existing programs on
1004 HTM capable systems. The lock elision code is only enabled with
1005 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1006 powerpc will issue a transaction abort on every syscall to avoid side
1007 effects being visible outside transactions.
1009 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1010 AArch64. Contributed by ARM Ltd.
1012 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1014 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1015 under certain input conditions resulting in the execution of a shell for
1016 command substitution when the application did not request it. The
1017 implementation now checks WRDE_NOCMD immediately before executing the
1018 shell and returns the error WRDE_CMDSUB as expected.
1020 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1021 processing format strings with a large number of format specifiers.
1023 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1024 infinite loop if the DNS response contained a PTR record of an unexpected
1027 * The minimum GCC version that can be used to build this version of the GNU
1028 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1029 still be used to compile programs using the GNU C Library.
1031 * The GNU C Library is now built with -Werror by default. This can be
1032 disabled by configuring with --disable-werror.
1034 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1036 * The obsolete sigvec function has been removed. This was the original
1037 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1038 programs have been using instead for about 25 years. Of course, ABI
1039 compatibility for old binaries using sigvec remains intact.
1041 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1042 with newer versions of bison.
1044 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1045 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1046 registers overlay two consecutive single-precision registers. MIPS32R2
1047 introduced a new FPU mode (FR=1) where double-precision registers extend the
1048 corresponding single-precision registers which is incompatible with the
1049 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1050 require the use of FR=1 making a transition necessary. New o32 ABI
1051 extensions enable users to migrate over time from the original o32 ABI
1052 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1053 tracks the ABI of any loaded object and verifies that new objects are
1054 compatible. Mode transitions will also be requested as required and
1055 unsupportable objects will be rejected. The ABI checks include both soft and
1056 hard float ABIs for o32, n32 and n64.
1058 GCC 5 with GNU binutils 2.25 onwards:
1059 It is strongly recommended that all o32 system libraries are built using the
1060 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1061 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1062 compiler using --with-fp-32=xx to set this by default.
1066 * The following bugs are resolved with this release:
1068 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1069 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1070 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1071 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1072 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1073 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1074 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1075 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1076 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1077 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1078 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1079 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1080 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1081 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1082 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1084 * Reverted change of ABI data structures for s390 and s390x:
1085 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1086 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1087 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1088 are preserved pointing straight to the same implementation as the old ones.
1089 Given that, new callers will simply provide a too-big buffer to these
1090 functions. Any applications/libraries out there that embed jmp_buf or
1091 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1092 against 2.19 headers will have to rebuilt again. This is necessary in any
1093 case to revert the breakage in their ABI caused by the glibc change.
1095 * Support for file description locks is added to systems running the
1096 Linux kernel. The standard file locking interfaces are extended to
1097 operate on file descriptions, not file descriptors, via the use of
1098 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1099 are associated with an open file instead of a process.
1101 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1103 * The minimum Linux kernel version that this version of the GNU C Library
1104 can be used with is 2.6.32.
1106 * Running the testsuite no longer terminates as soon as a test fails.
1107 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1108 with PASS or FAIL lines for individual tests. A summary of the results is
1109 printed, including a list of failing lists, and "make check" exits with
1110 error status if there were any unexpected failures. "make check
1111 stop-on-test-failure=y" may be used to keep the old behavior.
1113 * The am33 port, which had not worked for several years, has been removed
1116 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1117 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1118 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1119 conflicted with POSIX (support for which was removed in 2.19), the
1120 interfaces those macros enabled remain available when compiling with
1121 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1122 test macros defined.
1124 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1126 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1127 This may improve lock scaling of existing programs on TX capable systems.
1128 The lock elision code is only built with --enable-lock-elision=yes and
1129 then requires a GCC version supporting the TX builtins. With lock elision
1130 default mutexes are elided via __builtin_tbegin, if the cpu supports
1131 transactions. By default lock elision is not enabled and the elision code
1134 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1135 copy the path argument. This allowed programs to cause posix_spawn to
1136 deference a dangling pointer, or use an unexpected pathname argument if
1137 the string was modified after the posix_spawn_file_actions_addopen
1140 * All supported architectures now use the main glibc sysdeps directory
1141 instead of some being in a separate "ports" directory (which was
1142 distributed separately before glibc 2.17).
1144 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1145 On configurations that support it (all Linux configurations), it's now
1146 used regardless of the --enable-add-ons switch to configure. It is no
1147 longer possible to build such configurations without pthreads support.
1149 * Locale names, including those obtained from environment variables (LANG
1150 and the LC_* variables), are more tightly checked for proper syntax.
1151 setlocale will now fail (with EINVAL) for locale names that are overly
1152 long, contain slashes without starting with a slash, or contain ".." path
1153 components. (CVE-2014-0475) Previously, some valid locale names were
1154 silently replaced with the "C" locale when running in AT_SECURE mode
1155 (e.g., in a SUID program). This is no longer necessary because of the
1158 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1159 with application code using Intel MPX instructions. (With all previous
1160 versions, the MPX register state could be clobbered when making calls
1161 into or out of a shared library.) Note that while the new dynamic
1162 linker is compatible with all known x86 hardware whether or not it
1163 supports Intel MPX, some x86 instruction-set emulators might fail to
1164 handle the new instruction encodings. This is known to affect Valgrind
1165 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1166 release), and might affect other tools that do instruction emulation.
1168 * Support for loadable gconv transliteration modules has been removed.
1169 The support for transliteration modules has been non-functional for
1170 over a decade, and the removal is prompted by security defects. The
1171 normal gconv conversion modules are still supported. Transliteration
1172 with //TRANSLIT is still possible, and the //IGNORE specifier
1173 continues to be supported. (CVE-2014-5119)
1175 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1176 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1177 resulting a denial-of-service security vulnerability in applications which
1178 use functions related to iconv. (CVE-2014-6040)
1182 * The following bugs are resolved with this release:
1184 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1185 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1186 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1187 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1188 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1189 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1190 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1191 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1192 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1193 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1194 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1195 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1196 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1197 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1198 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1199 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1200 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1201 16431, 16453, 16474, 16506, 16510, 16529
1203 * Slovenian translations for glibc messages have been contributed by the
1204 Translation Project's Slovenian team of translators.
1206 * The public headers no longer use __unused nor __block. This change is to
1207 support compiling programs that are derived from BSD sources and use
1208 __unused internally, and to support compiling with Clang's -fblock
1209 extension which uses __block.
1211 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1212 large collation sequences to optimize multiple passes. This cache
1213 computation may overflow for large collation sequences and may cause a
1214 stack or buffer overflow. This is now fixed to use a slower algorithm
1215 which does not use a cache if there is an integer overflow.
1217 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1218 rules for large collation sequences to optimize multiple passes and falls
1219 back to alloca if malloc fails, resulting in a possible stack overflow.
1220 The implementation now falls back to an uncached collation sequence lookup
1223 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1224 initialized for static applications resulting in the security feature
1225 being disabled. The pointer guard is now correctly initialized to a
1226 random value for static applications. Existing static applications need
1227 to be recompiled to take advantage of the fix (bug 15754).
1229 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1230 to the d_name member of struct dirent, or omit the terminating NUL
1231 character. (Bugzilla #14699).
1233 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1234 aligned_alloc functions could allocate too few bytes or corrupt the
1235 heap when passed very large allocation size values (Bugzilla #15855,
1238 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1239 for AF_INET6 has been fixed (Bugzilla #16072).
1241 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1242 pap_CW, quz_PE, the_NP.
1244 * Substantially revised locales: gd_GB, ht_HT
1246 * The LC_ADDRESS field was updated to support country_car for almost all
1249 * ISO 1427 definitions were updated.
1251 * ISO 3166 definitions were updated.
1253 * The localedef utility now supports --big-endian and --little-endian
1254 command-line options to generate locales for a different system from that
1255 for which the C library was built.
1257 * Binary locale files now only depend on the endianness of the system for
1258 which they are generated and not on other properties of that system. As a
1259 consequence, binary files generated with new localedef may be incompatible
1260 with old versions of the GNU C Library, and binary files generated with
1261 old localedef may be incompatible with this version of the GNU C Library,
1262 in the following circumstances:
1264 + Locale files may be incompatible on m68k systems.
1266 + Locale archive files (but not separate files for individual locales) may
1267 be incompatible on systems where plain "char" is signed.
1269 * The configure option --disable-versioning has been removed. Builds with
1270 --disable-versioning had not worked for several years.
1272 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1273 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1275 * SystemTap probes for malloc have been introduced.
1277 * SystemTap probes for slow multiple precision fallback paths of
1278 transcendental functions have been introduced.
1280 * Support for powerpc64le has been added.
1282 * The soft-float powerpc port now supports e500 processors.
1284 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1286 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1287 set of header declarations that are enabled by default, even when other
1288 feature test macros or compiler options such as -std=c99 would otherwise
1289 disable some of those declarations.
1291 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1292 conflict with POSIX. The libbsd-compat library (which was a dummy library
1293 that did nothing) has also been removed.
1295 * Preliminary documentation about Multi-Thread, Async-Signal and
1296 Async-Cancel Safety has been added.
1298 * Change of ABI data structures for s390 and s390x:
1299 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1300 allow for future hardware extensions. All library functions that accept or
1301 return these structures were versioned in order to provide backwards
1302 compatibility with existing code. However, not all cases can be handled
1303 with interface versioning. If an application embeds either structure into
1304 another structure and that structure is passed to another compilation unit
1305 which expects the newer larger-sized object then any access to the new
1306 fields will result in undefined behaviour. Similarly any access to fields
1307 that were moved by the enlarging of the structures will cause undefined
1308 behaviour. To fix the undefined behaviour all objects that are part of the
1309 final in-memory image must agree on the size of structures, and this may
1310 require recompilation.
1314 * The following bugs are resolved with this release:
1316 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1317 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1318 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1319 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1320 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1321 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1322 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1323 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1324 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1325 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1326 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1327 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1328 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1331 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1332 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1333 Distributions can re-enable building and using pt_chown via the new configure
1334 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1335 considerable security risks and should only be used if the distribution
1336 understands and accepts the risks.
1338 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1341 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1342 fixed (Bugzilla #15330).
1344 * Add support for calling C++11 thread_local object destructors on thread
1345 and program exit. This needs compiler support for offloading C++11
1346 destructor calls to glibc.
1348 * Improved worst case performance of libm functions with double inputs and
1351 * Support for priority inherited mutexes in pthread condition variables on
1352 non-x86 architectures.
1354 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1356 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1358 * Optimized string functions for ARM. Implemented by Will Newton and
1361 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1363 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1366 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1367 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1369 * Added a benchmark framework to track performance of functions in glibc.
1371 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1372 It is based on draft TS 18661 and currently enabled as a GNU extension.
1374 * On Linux, the clock function now uses the clock_gettime system call
1375 for improved precision, rather than old times system call.
1377 * Added support for version-3 tz format files. This is needed when using
1378 the tz database release 2013e or later, and affects a few unusual cases --
1379 currently only TZ='America/Godthab' for time stamps after 2037.
1381 * Added new API functions pthread_getattr_default_np and
1382 pthread_setattr_default_np to get and set the default pthread
1383 attributes of a process.
1385 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1386 This may improve lock scaling of existing programs on TSX capable systems.
1387 When the --enable-lock-elision=yes parameter is specified at configure
1388 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1391 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1392 availability). Implemented by Ryan S. Arnold.
1394 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1398 * The following bugs are resolved with this release:
1400 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1401 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1402 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1403 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1404 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1405 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1406 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1407 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1408 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1409 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1410 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1411 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1412 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1413 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1414 14879, 14889, 14893, 14898, 14914.
1416 * Optimization of memcpy for MIPS.
1418 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1419 EMFILE has been fixed (Bugzilla #14889).
1421 * The add-on ports collection is now distributed in the "ports" subdirectory
1422 of the main GNU C Library distribution, rather than separately.
1424 * Port to ARM AArch64 contributed by Linaro.
1426 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1427 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1429 Implemented by Andreas Krebbel.
1431 * The new function secure_getenv allows secure access to the environment,
1432 returning NULL if running in a SUID/SGID process. This function replaces
1433 the internal function __secure_getenv.
1435 * SystemTap static probes have been added into the dynamic linker.
1436 Implemented by Gary Benson.
1438 * Optimizations of string functions strstr, strcasestr and memmem.
1439 Implemented by Maxim Kuvyrkov.
1441 * The minimum Linux kernel version that this version of the GNU C Library
1442 can be used with is 2.6.16.
1444 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1445 powerpc POWER7. Implemented by Will Schmidt.
1447 * New configure option --disable-nscd builds the C library such that it
1448 never attempts to contact the Name Service Caching Daemon (nscd).
1449 New configure option --disable-build-nscd avoids building nscd itself;
1450 this is the default if --disable-nscd is used.
1452 * Improved support for cross-compilation, including cross-testing and
1453 bootstrap builds without a previously built glibc.
1455 * Several testsuite tests are now able to test multiple IFUNC variants of an
1456 interface, rather than just testing the one that would be chooen by
1459 * New configure options --with-bugurl and --with-pkgversion, for
1460 distributors to use to embed their bug-reporting and package version
1461 information in --help and --version output.
1463 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1464 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1465 allows creation of chroots without the procfs mounted on /proc.
1467 * The `crypt' function now fails if passed salt bytes that violate the
1468 specification for those values. On Linux, the `crypt' function will
1469 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1470 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1471 when the mode is enabled.
1473 * The `clock_*' suite of functions (declared in <time.h>) is now available
1474 directly in the main C library. Previously it was necessary to link with
1475 -lrt to use these functions. This change has the effect that a
1476 single-threaded program that uses a function such as `clock_gettime' (and
1477 is not linked with -lrt) will no longer implicitly load the pthreads
1478 library at runtime and so will not suffer the overheads associated with
1479 multi-thread support in other code such as the C++ runtime library.
1481 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1486 * The following bugs are resolved with this release:
1488 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1489 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1490 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1491 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1492 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1493 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1494 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1495 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1496 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1497 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1498 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1499 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1500 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1501 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1502 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1503 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1504 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1505 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1506 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1507 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1508 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1509 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1512 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1513 configuring glibc with:
1514 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1515 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1516 Implemented by H.J. Lu.
1520 + define static_assert
1522 + do not declare gets
1524 + declare at_quick_exit and quick_exit also for ISO C11
1526 + aligned_alloc. NB: The code is deliberately allows the size parameter
1527 to not be a multiple of the alignment. This is a moronic requirement
1528 in the standard but it is only a requirement on the caller, not the
1531 + timespec_get added
1533 + uchar.h support added
1535 + CMPLX, CMPLXF, CMPLXL added
1537 Implemented by Ulrich Drepper.
1539 * Support for the IA-64 has been moved to ports.
1541 * Remove support for anything but ELF binary format
1543 * Checking versions of poll, ppoll added.
1544 Implemented by Ulrich Drepper.
1546 * More generic and 64-bit performance optimizations to math functions.
1547 Implemented by Ulrich Drepper.
1549 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1550 headers and functions available at compile time as they were before
1551 version 2.14. This option will be removed at some time in the future
1552 after the TI-RPC library becomes fully sufficient for the needs of
1553 existing applications.
1555 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1556 Note that glibc is not expected to work with any Linux kernel version
1559 * New header <sys/auxv.h> and function getauxval allowing easy access to
1560 the AT_* key-value pairs passed from the Linux kernel. The header also
1561 defines the HWCAP_* bits associated with the AT_HWCAP key.
1563 * New locales: mag_IN
1565 * New configure option --enable-systemtap builds SystemTap static probes
1566 into libc for setjmp and longjmp and into libpthread for various operations.
1567 So far the setjmp/longjmp probes and some of the libpthread probes are
1568 provided only for i*86 and x86_64.
1569 Implemented by Roland McGrath and Rayson Ho.
1571 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1573 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1576 * More optimized functions for SPARC. Implemented by David S. Miller.
1578 * Improved support for cross-compilation, especially bootstrap builds
1579 without a previously built glibc.
1581 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1582 Chris Metcalf from Tilera.
1584 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1585 now supported for ARM processors.
1587 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1588 as the name of the dynamic linker, to distinguish it from the
1589 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1591 * Support for CRIS has been removed from ports.
1593 * A new class of installed header has been documented for low-level
1594 platform-specific functionality. PowerPC added the first instance with a
1595 function to provide time base register access. Contributed by Tulio
1596 Magno Quites Machado Filho.
1598 * ix86 configurations now install header files that are consistent with
1599 what x86-64 configurations install. These same header files can be used
1600 for -m32, -m64, or -mx32 builds.
1601 Contributed by H.J. Lu.
1603 * Math library bug fixes. A thorough audit of all open math library bugs was
1604 conducted by Joseph Myers. Significant progress was made on many math
1605 library bugs resulting in more accurate exceptions and function results.
1606 Many thanks to all those that contributed including Andreas Jaeger for his
1607 patch review and work on the x87 trigonometric instruction issues.
1609 * Timezone data is no longer installed. Timezone-related binaries and scripts
1610 will continue to be installed. Users should obtain their timezone data from
1611 their distribution provider or from the tzdata package at
1612 <ftp://munnari.oz.au/pub/>.
1616 * The following bugs are resolved with this release:
1618 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1619 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1620 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1621 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1622 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1623 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1624 13484, 13506, 13515, 13523, 13524, 13538, 13540
1626 * New program pldd to list loaded object of a process
1627 Implemented by Ulrich Drepper.
1629 * Add nss_db support back to glibc. No more dependency on Berkeley db
1630 and support for initgroups lookups.
1631 Implemented by Ulrich Drepper.
1633 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1634 Contributed by HJ Lu.
1636 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1637 Contributed by HJ Lu.
1639 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1640 on x86-32 and x86-64.
1641 Contributed by Liubov Dmitrieva.
1643 * Optimized strchr and strrchr for SSE on x86-32.
1644 Contributed by Liubov Dmitrieva.
1646 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1647 for x86-64 and x86-32.
1648 Contributed by Liubov Dmitrieva.
1650 * New interfaces: scandirat, scandirat64
1651 Implemented by Ulrich Drepper.
1653 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1654 Implemented by Ulrich Drepper.
1656 * nscd now also caches the netgroup database.
1657 Implemented by Ulrich Drepper.
1659 * Integrate libm with gcc's -ffinite-math-only option.
1660 Implemented by Ulrich Drepper.
1662 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1663 to math functions. Implemented by Ulrich Drepper.
1665 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1666 Implemented by Ulrich Drepper.
1668 * New Linux interfaces: process_vm_readv, process_vm_writev
1670 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1671 Implemented by Ulrich Drepper.
1673 * Optimized nearbyint and strcasecmp for PPC.
1674 Implemented by Adhemerval Zanella.
1676 * New locales: bho_IN, unm_US, es_CU, ta_LK
1680 * The following bugs are resolved with this release:
1682 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1683 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1684 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1685 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1686 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1687 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1688 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1689 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1690 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1691 12795, 12811, 12813, 12814, 12841
1693 * The RPC implementation in libc is obsoleted. Old programs keep working
1694 but new programs cannot be linked with the routines in libc anymore.
1695 Programs in need of RPC functionality must be linked against TI-RPC.
1696 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1698 Visible changes of this change include (obviously) the inability to link
1699 programs using RPC functions without referencing the TI-RPC library and the
1700 removal of the RPC headers from the glibc headers.
1701 Implemented by Ulrich Drepper.
1703 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1704 syncfs, setns, sendmmsg
1706 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1707 yue_HK, lij_IT, mhr_RU
1709 * New encodings: CP770, CP771, CP772, CP773, CP774
1711 * New program sotruss to trace calls through PLTs
1712 Implemented by Ulrich Drepper.
1714 * The malloc hook implementation is marked deprecated and will be removed
1715 from the default implementation in the next version. The design never
1716 worked ever since the introduction of threads. Even programs which do
1717 not create threads themselves can use multiple threads created internally.
1721 * The following bugs are resolved with this release:
1723 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1724 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1725 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1726 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1729 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1731 * POWER7 optimizations: memset, memcmp, strncmp
1733 * New optimized string functions for x86-64: strnlen (SSE2),
1734 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1735 Implemented by Ulrich Drepper.
1739 * The following bugs are resolved with this release:
1741 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1742 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1743 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1744 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1745 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1746 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1747 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1748 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1750 * New interfaces: pthread_getname_np, pthread_setname_np
1752 * New Linux interface: recvmmsg
1754 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1756 * The dynamic linker now recognizes supported ABI versions from the
1757 EI_ABIVERSION field in the ELF header.
1758 Implemented by Ulrich Drepper.
1760 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1761 will not be used to fill in password fields in the passwd.byname replies.
1762 Instead it is used to synthesize the shadow.byname table, should it be
1763 missing. This is a useful mode in some installations involving Solaris.
1764 Implemented by Ulrich Drepper.
1766 * New locales: kok_IN, sq_MK, cv_RU
1770 * The following bugs are resolved with this release:
1772 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1773 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1774 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1775 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1776 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1777 10789, 10817, 10823, 10840, 10847
1779 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1781 Implemented by Ulrich Drepper.
1783 * Checking version of longjmp added that fails if an uninitialized stack
1784 frame would be created. Implemented by Ulrich Drepper.
1786 * STT_GNU_IFUNC is now supported in static executables.
1787 Implemented by H.J. Lu.
1789 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1791 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1792 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1793 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1794 Contributed by H.J. Lu.
1796 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1797 Implemented by Ulrich Drepper.
1799 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1801 Contributed by H.J. Lu.
1803 * Support for fma instruction in AVX on x86-64.
1804 Implemented by H.J. Lu and Ulrich Drepper.
1806 * AVX support in x86-64 auditing support in ld.so.
1807 Implemented by H.J. Lu.
1809 * STB_GNU_UNIQUE support added.
1810 Implemented by Ulrich Drepper.
1812 * Implement second fallback mode for optimized DNS lookup for even more
1813 broken environments. If two requests from the same port are not
1814 handled correctly close the socket and open a new one before sending
1815 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1816 can be used to select this mode right away, instead of rediscovering the
1817 necessity is every process again.
1818 Implemented by Ulrich Drepper.
1820 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1821 Implemented by Adam Tkac.
1823 * Optimized iconv conversions for S390x.
1824 Implemented by Andreas Krebbel.
1826 * Using condvars with PI mutexes is now more efficient due to kernel
1827 support for requeueing to PI futexes. NPTL support added for x86-64.
1828 Implemented by Ulrich Drepper.
1830 * New locales: ps_AF, my_MM
1834 * The following bugs are resolved with this release:
1836 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1837 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1838 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1839 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1840 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1842 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1843 Implemented by Ulrich Drepper.
1845 * New Linux interfaces: accept4, fallocate, fallocate64.
1846 Implemented by Ulrich Drepper.
1848 * Correct declarations of string function when used in C++ code. This
1849 could lead to compile errors for invalid C++ code.
1851 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1854 * New POSIX 2008 interface: psiginfo
1855 Implemented by Ulrich Drepper.
1857 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1858 Implemented by Ulrich Drepper.
1860 * Support for selecting between multiple function definitions at runtime
1861 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1863 * The libcrypt library can now use the hash function implementations in
1864 NSS. Implemented by Ulrich Drepper.
1866 * The malloc implementation can be compiled to be less memory efficient
1867 but higher performing in multi-threaded programs.
1868 Implemented by Ulrich Drepper.
1870 * New locales: nan_TW@latin, ks_IN
1872 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1873 Implemented by Ulrich Drepper.
1875 * Extended printf hook support. It is possible to use user-defined types
1876 and extend existing format specifiers.
1877 Implemented by Ulrich Drepper.
1879 * Handling for group shadow files has been added.
1880 Implemented by Ulrich Drepper.
1882 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1883 servers (the case, e.g., for some people using the built-in DNS
1884 server in ADSL modems/routers). There is a once-per-process timeout
1885 in case of a broken server. To avoid it, users can run nscd or put
1886 'options single-request' in /etc/resolv.conf.
1887 Implemented by Ulrich Drepper.
1891 * The following bugs are resolved with this release:
1893 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1894 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1895 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1896 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1897 6968, 6974, 6980, 6995, 7008, 7009, 7029
1899 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1900 up at the same time. Implemented by Ulrich Drepper.
1902 * TLS descriptors for LD and GD on x86 and x86-64.
1903 Implemented by Alexandre Oliva.
1905 * getaddrinfo now handles DCCP and UDPlite.
1906 Implemented by Ulrich Drepper.
1908 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1909 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1910 Implemented by Ulrich Drepper.
1912 * New implementation of memmem, strstr, and strcasestr which is O(n).
1913 Implemented by Eric Blake.
1915 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1917 * Implement "e" option for popen to open file descriptor with the
1918 close-on-exec flag set. Implemented by Ulrich Drepper.
1920 * Implement "b" mode for fmemopen. In this mode writes writes don't
1921 implicitly add a NUL byte and seeks from the end of the buffer really
1922 use the buffer end, not the string length as the basis.
1923 Implemented by Ulrich Drepper.
1925 * Many functions, exported and internal, now atomically set the close-on-exec
1926 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1928 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1930 Implemented by Pravin Satpute.
1932 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1936 * The following bugs are resolved with this release:
1938 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1939 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1940 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1941 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1942 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1943 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1944 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1946 * New locales: bo_CN, bo_IN, shs_CA.
1948 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1950 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1951 Implemented by Pravin Satpute.
1953 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1955 * nscd caches now all timeouts for DNS entries
1956 Implemented by Ulrich Drepper.
1958 * nscd is more efficient and wakes up less often.
1959 Implemented by Ulrich Drepper.
1961 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1962 vdprintf, and obstack_vprintf.
1963 Implemented by Jakub Jelinek.
1965 * Faster memset for x86-64.
1966 Implemented by Harsha Jagasia and H.J. Lu.
1968 * Faster memcpy on x86.
1969 Implemented by Ulrich Drepper.
1971 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1972 Implemented by Ulrich Drepper.
1974 * Faster sqrt and sqrtf implemention for some PPC variants.
1975 Implemented by Steven Munroe.
1979 * The following bugs are resolved with this release:
1981 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1982 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1983 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1984 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1985 5071, 5103, 5104, 5112, 5113, 5184, 5186
1987 * More checking functions: fread, fread_unlocked, open*, mq_open.
1988 Implemented by Jakub Jelinek and Ulrich Drepper.
1990 * Extend fortification to C++. Implemented by Jakub Jelinek.
1992 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1993 by not recognizing 'a' as a modifier when those specs are requested.
1994 Implemented by Jakub Jelinek.
1996 * PPC optimizations to math and string functions.
1997 Implemented by Steven Munroe.
1999 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2000 options to be passed. Implemented by Ulrich Drepper.
2002 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2004 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2005 Implemented by Ulrich Drepper.
2007 * Handle private futexes in the NPTL implementation.
2008 Implemented by Jakub Jelinek and Ulrich Drepper.
2010 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2011 Implemented by Roland McGrath and Ulrich Drepper.
2013 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2015 * SHA-256 and SHA-512 based password encryption.
2016 Implemented by Ulrich Drepper.
2018 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2019 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2022 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2023 Implemented by Ulrich Drepper.
2027 * The following bugs are resolved with this release:
2029 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2030 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2031 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2032 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2033 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2034 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2035 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2038 * New Linux interfaces: epoll_pwait, sched_getcpu.
2040 * New generic interfaces: strerror_l.
2042 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2046 * The following bugs are resolved with this release:
2048 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2049 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2050 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2051 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2052 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2053 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2054 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2055 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2056 3225, 3251, 3252, 3253, 3273
2058 * For Linux, the sorting of addresses returned by getaddrinfo now also
2059 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2060 Implemented by Ulrich Drepper.
2062 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2063 Implemented by Ulrich Drepper.
2065 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2067 * New iconv module for MIK. Contributed by Alexander Shopov.
2069 * For sites with broken group and/or passwd database, the auto-propagate
2070 option of nscd can prevent creating ID lookup entries from the results
2071 of a name lookup and vice versa. This usually is no problem but some
2072 site might have problems with the default behavior.
2073 Implemented by Ulrich Drepper.
2075 * Iterating over entire database in NIS can be slow. With the
2076 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2077 to trade time for memory. The entire database will be read at once.
2078 Implemented by Ulrich Drepper.
2080 * The interfaces introduced in RFC 3542 have been implemented by
2083 * Support for the new ELF hash table format was added by Ulrich Drepper.
2085 * Support for priority inheritance mutexes added by Jakub Jelinek and
2088 * Support for priority protected mutexes added by Jakub Jelinek.
2090 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2091 or_IN, csb_PL, fy_NL, sr_ME.
2095 * More overflow detection functions.
2097 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2098 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2100 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2101 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2102 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2103 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2104 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2105 by Masahide Washizawa.
2107 * It is now possible to install audit libraries for the dynamic linker using
2108 LD_AUDIT. Implemented by Ulrich Drepper.
2110 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2111 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2112 For a libc and libpthread that works well on Linux 2.4 kernels, we
2113 recommend using the stable 2.3 branch.
2115 * The new function open_wmemstream acts like open_memstream,
2116 but uses a wchar_t wide character buffer.
2118 * The new function ppoll is an improved version of poll, similar to pselect.
2120 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2121 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2122 renameat, unlinkat, linkat, symlinkat, readlinkat.
2124 * New Linux kernel system calls: unshare,
2125 inotify_init, inotify_add_watch, inotify_rm_watch.
2127 * The euidaccess function is now also known by the alias eaccess,
2128 for compatibility with some other systems.
2130 * Timezone data updated to 2006b version.
2134 * The following bugs are resolved with this release:
2136 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2137 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2138 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2139 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2140 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2141 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2143 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2145 * As of this release, GCC 4 can be used to compile the C Library.
2147 * Timezone data updated to 2005m version.
2151 * The following bugs are resolved with this release:
2153 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2154 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2155 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2156 777, 787, 821, 822, 823, 825
2158 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2162 * Support for RFC 3678. Real implementations exist only for Linux so far.
2163 Implemented by Ulrich Drepper.
2165 * nscd can now cache entries persistently. Expiring entries are reloaded.
2166 For speedups the cache can be shared in memory with client processes.
2167 Implemented by Ulrich Drepper.
2169 * nscd can now perform SELinux checks.
2170 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2172 * getaddrinfo queries are now cached. Canonical name lookup is performed
2174 Implemented by Ulrich Drepper.
2176 * The nothrow function attribute is used when headers are used by gcc when
2177 compiling C code. This can avoid significant amounts of exception
2180 * The malloc functions perform more error checking and are stricter when
2181 it comes to reacting on errors. The default action is to terminate
2182 the process after showing an error message. Implemented by Ulrich Drepper.
2184 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2185 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2187 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2188 namespaces using the new function dlmopen(). This feature is of course,
2189 like most other dynamic loading functionality, not available in statically
2190 linked applications. Implemented by Ulrich Drepper.
2192 * Low-overhead boundary checking variants of string and some stdio functions
2193 were added. These are to be used in conjunction with a gcc patch by
2194 Jakub Jelinek which adds calls to these functions if possible.
2195 Implemented by Jakub Jelinek and Ulrich Drepper.
2197 * Old code for several operating systems and machine architectures that
2198 have not been in working condition in a long time have been removed from
2199 the main source tree maintained by the GNU C Library's maintainers.
2200 These files are now reside in the separate `ports' source module
2201 that is usable as an add-on when building the library.
2205 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2206 interrogate the dynamic linker, compatible with the Solaris interface.
2208 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2209 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2211 * getifaddrs now uses the netlink interface on Linux to get its information.
2212 Implemented by Thorsten Kukuk.
2214 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2215 Implemented by Ulrich Drepper.
2217 * support for non-executable stacks on x86 has been added. Changes mostly
2220 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2223 * getaddrinfo now performs destination address selection according to
2228 * Thread-safe interfaces for many functions that access locale data
2229 were added in version 2.3, but these features were omitted from NEWS.
2230 Many functions have variants with an `_l' suffix that take a `locale_t'
2231 object as a parameter rather than consulting the current locale.
2232 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2233 create and maintain `locale_t' objects. Additionally, the new function
2234 `uselocale' sets "the current locale" (as used by functions not so
2235 parameterized) set for an individual thread. These features were added
2236 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2238 * The functions getresuid, getresgid, setresuid, and setresgid, which
2239 have long been available on Linux, are now declared in <unistd.h>
2240 and are now also available on the Hurd.
2242 * ELF thread-local storage support (TLS) now works on x86-64.
2244 * The new dynamic string token $LIB is expanded in shared library names.
2245 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2247 * Aldy Hernandez contributed complete software floating point support for
2248 PowerPC machines with no FPU.
2250 * fexecve is implemented on Linux.
2252 * The `btowc' function should work at least twice as fast due to
2253 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2255 * With approriate thread add-ons cancelable functions are now implemented
2256 in libc.so as well. No need to call the function in libpthread. This
2257 change allowed to finally disable the incorrect and expensive handling
2258 of weak definition in ld.so.
2260 * Yet more PLT entries in libc.so have been removed. We finally arrived
2261 at the bare minimum. Startup times improved appropriately.
2263 * Support for the new Linux/x86 system call interface was added. The
2264 AT_SYSINFO auxiliary vector entry is recognized and handled.
2268 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2271 * iconv (the program and the interface) now accepts empty names (excluding
2272 options like //TRANSLIT) to mean "use charset of current locale".
2274 * localedef can now transliterate characters in strings which are not in
2275 the provided charmap. The information from the input locale is used.
2277 * Prelinking support was added for ELF targets. This requires additional
2278 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2280 * Read-only stdio streams now use mmap to speed up operation by eliminating
2281 copying and buffer underflows. To use add 'm' to the mode string of
2282 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2284 * The malloc functions were completely rewritten by Wolfram Gloger based
2285 on Doug Lea's malloc-2.7.0.c.
2287 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2288 implementation of regex.
2290 * Bruno Haible upgraded the iconv and locale implementation to support
2293 * Contents of the LC_* and LANG environment variables in the CEN style are
2294 not recognized anymore. It never was used. Change by Ulrich Drepper.
2296 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2297 thread-local storage (TLS) ABI on some platforms.
2298 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2300 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2301 EUC-JISX0213, and TSCII.
2303 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2304 BSD-compatible interface for getting all network interface addresses.
2305 Implementation for IPv4 by Roland McGrath.
2307 * Loading of locale data is faster due to the introduction of a locale
2308 archive. Implemented by Roland McGrath and Ulrich Drepper.
2310 * Startup times are significantly reduced by not using exported functions
2311 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2314 * Steven Munroe contributed a port to PowerPC64/Linux.
2318 * The Hurd now uses the GNU libio implementation of stdio rather than the
2319 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2321 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2322 with the same behavior as the Linux system call of the same name.
2326 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2327 128-bit long double format.
2329 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2330 IBM1161, and IBM1162 charsets.
2332 * Andreas Jaeger contributed a port to x86-64/Linux.
2334 * Peter Bruin contributed a port to PowerPC/Hurd.
2336 * libc functions using I/O streams now can handle wide-oriented streams
2339 * optimizations in the dynamic linker. Binaries created by recent binutils
2340 versions start up quicker due to reduced time spent on relocations.
2342 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2346 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2347 asinh, atanh, j0 for the 128-bit long double format.
2349 * Bruno Haible updated all the code handling Unicode in some form to
2350 support Unicode 3.1.
2352 * Speed of regex for single-byte locales is back to previous levels.
2353 Patch by Isamu Hasegawa.
2355 * Alpha, SPARC, and IA-64 now also using floating stacks.
2357 * Startup time of internationalized applications greatly improved through
2358 iconv cache. Use iconvconfig to generate the cache file.
2359 Contributed by Ulrich Drepper.
2361 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2362 severe performance problems. Patch by David Mosberger.
2364 * The Hurd port got a lot more functionality like AIO, various stdio
2365 extensions, etc. Mainly done by Roland McGrath.
2367 * mtrace can now lookup symbols in shared libraries.
2371 * Intel's IA-64 math library is largely integrated. It provides fast and
2372 accurate implementatations for most basic and standard math functions
2373 in float, double, and long double format.
2375 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2376 and asin for the 96-bit long double format and asin, log, tan for the
2377 128-bit long double format.
2379 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2380 The basic double functions exist today. Contributed by Abraham Ziv
2381 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2382 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2384 * An asynchronous name lookup library was added. The interface is designed
2385 after POSIX AIO. The proposal was circulated beforehand to get comments.
2386 No negative ones came in. Implemented by Ulrich Drepper.
2388 * Port to S390/64bit contributed by Martin Schwidefsky
2389 <schwidefsky@de.ibm.com>.
2391 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2392 of functions for Linux/IA-64.
2394 * The RPC code is now thread safe. Threads can now use the same service
2395 of different services at the same time. Patch by Eric Norum
2396 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2398 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2399 family of functions for Linux/S390.
2401 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2402 of functions for Linux/x86.
2404 * Port to Linux/CRIS contributed by Axis Communications.
2408 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2409 we can now check for namespace violations and missing declarations. The
2410 result is that almost all headers are now Unix-compliant (as defined in
2411 the upcoming XPG6). The negative side is that some programs might need
2412 corrections, too, if they depend on the incorrect form of the headers in
2413 previous versions which defined too many symbols and included too many
2416 * regex now handles multibyte character sets correctly.
2417 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2419 * iconv (the program) does now conform to the upcoming XPG6 and handles
2420 charmaps. Instead of the charset names the path of charmaps can be
2421 provided and the conversion happens based on this data.
2422 Contributed by Ulrich Drepper.
2424 * The locale program now provides detailed information about the installed
2429 only lists the names of the supported locales
2433 provides details such as country, language, and codeset name.
2434 Contributed by Ulrich Drepper.
2438 * The gencat program now parses the input file according to the charset
2439 selected by the LC_CTYPE category. This is important for stateful
2440 character sets. To make generating catalogs easier there is a way
2441 to overwrite the charset selected by the locale: before the first
2442 message or $ quote line the catalog can contain a line like
2444 $ codeset=ISO-8859-2
2446 to select the charset (ISO-8859-2 in this case).
2448 Implemented by Shinya Hanataka and Ulrich Drepper.
2450 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2451 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2452 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2454 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2455 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2457 * The LANGUAGE environment variable is now ignored unless the locale is
2458 changed from the default "C" locale.
2460 * The usual bug fixes.
2464 * Greg McGary added runtime support for bounds checking using gcc's
2465 new -fbounded-pointers option. ix86 target is complete. PowerPC
2468 * Thorsten Kukuk added secure mode support to nscd.
2470 * The Berkeley db libraries have been removed.
2472 Related, the nss_db module is now in a separate package since it
2473 obviously requires a database library being available.
2475 * Wide character I/O streams implemented by Ulrich Drepper.
2477 * Functions from the extended socket API added by Ulrich Drepper.
2479 * Functions feenableexcept and fedisableexcept to control the
2480 behaviour of individual exceptions have been added by Andreas Jaeger.
2482 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2484 * The resolver code has been updated from bind 8.2.3-T5B which supports
2485 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2488 This change could in some situations effect backward compatibility. Since
2489 now `_res' is a thread-local instead of a global variable, modifying it
2490 in one thread does not have any effect in other threads.
2492 The resolver library was also extended to allow IPv6 as the transport
2493 protocol for the requests. This work was done by Stig Venaas.
2495 * Compatibility code for K&R C compilers has been removed from the
2496 header files. A ISO C compiler is needed to use the library
2497 (conforming to either C89 or C99 standard).
2499 * Complete rewrite of the localedef program to support multibyte character
2500 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2501 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2502 Implemented by Ulrich Drepper.
2504 Bruno Haible significantly improved the generation and use of the data
2505 structures for the wide character tables.
2507 * Plural handling in gettext implemented by Ulrich Drepper.
2509 * The utmp daemon has been removed.
2511 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2513 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2516 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2518 * POSIX barriers implemented by Kaz Kylheku.
2520 * POSIX spawn function family implemented by Ulrich Drepper.
2522 * POSIX spinlocks are now available.
2524 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2526 * the configure option --enable-kernel=X.Y.Z allows to strip out
2527 compatibility for kernel versions before X.Y.Z. This is currently only
2528 implemented for Linux.
2530 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2531 field sin6_scope_id. This means that all programs using IPv6 should be
2532 recompiled. Don't expect binary compatibility with previous glibc
2535 * various conversion modules for IBM character sets contributed by
2538 * IA-64 port by Jes Sorensen and HJ Lu.
2552 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2554 * New xdr functions are added; some rpc functions are now 64bit clean.
2556 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2558 * Fixed known incompatibilities with glibc 2.0.
2560 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2562 * Optimized a number of functions (especially the ELF dynamic loader).
2564 * Update timezone data files.
2566 * lots of charmaps corrections
2568 * some new locale definitions and charmaps
2573 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2574 conform to POSIX member type requirements. Symbol versions have been
2575 adjusted as needed within the library, and for direct use by applications,
2576 but there is potential for problems if third-party libraries use
2577 struct timeval as part of their interface. This does not present
2578 a problem for X and other "essential" system libraries.
2580 * An additional locale model to support C++ Standard Library locale
2581 model and probably more was implemented by Ulrich Drepper.
2583 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2586 * Miles Bader provided the `argp' function family to support hierachical
2587 command line argument parsing, layered on top of getopt.
2589 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2591 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2594 * scanf recognizes the %a and %A format for scanning floating point numbers.
2596 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2597 information and interfaces for the available integer types.
2599 * about 130 new math functions were added to implement the ISO C9x math
2602 * the new header <complex.h> contains definitions of the complex math
2603 functions from ISO C 9X.
2605 * the new header <tgmath.h> defines generic macros to use complex or
2606 real valued functions.
2608 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2610 * Andreas Jaeger provided a test suite for the math library.
2612 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2614 * Ulrich Drepper added character set conversion functions (iconv).
2616 * Optimized string functions have been added.
2618 * The localedata addon is now part of glibc.
2620 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2622 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2623 daemon for NSS (nscd).
2625 Missing a better place here are some numbers on improvements. Under
2626 Linux 2.1.125 un-tar-ing the kernel sources takes
2630 using local files 12.19s 6.88s 22.91s
2632 using NIS 13.92s 8.91s 26.34s
2634 using NIS & nscd 10.37s 7.34s 25.30s
2636 using NIS+ 27.57s 30.37s 640.46s
2638 using NIS+ & nscd 10.25s 7.83s 26.51s
2640 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2642 Keep in mind that non-namelookup related operations dominate above times.
2643 It was just a common complain that using NIS+ unpacking the kernel is
2646 [1] The old nscd implementation is not available anymore since it was
2647 distributed with glibc up to version 2.0.98 and thus is now replaced.
2649 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2651 * Mark Kettenis provided a Hesiod NSS module.
2653 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2654 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2656 * rcmd can now handle netgroups (Dick Streefland).
2658 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2661 * Support for the IPv6 protocol has been added to the socket API, as per the
2662 latest draft standards.
2664 * Support for Linux 2.2 has been added.
2666 * Interface changes relative to the latest 2.0.x release:
2667 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2668 addseverity NEW: Unix98
2669 alphasort64 NEW: LFS
2670 argp_err_exit_status NEW: argp, GNU ext
2671 argp_error NEW: argp, GNU ext
2672 argp_failure NEW: argp, GNU ext
2673 argp_help NEW: argp, GNU ext
2674 argp_parse NEW: argp, GNU ext
2675 argp_program_bug_address NEW: argp, GNU ext
2676 argp_program_version NEW: argp, GNU ext
2677 argp_program_version_hook NEW: argp, GNU ext
2678 argp_state_help NEW: argp, GNU ext
2679 argp_usage NEW: argp, GNU ext
2680 authdes_create NEW: Secure RPC
2681 authdes_getucred NEW: Secure RPC
2682 authdes_pk_create NEW: Secure RPC
2683 backtrace NEW: GNU ext.
2684 backtrace_symbols NEW: GNU ext.
2685 backtrace_symbols_fd NEW: GNU ext.
2687 cacosf NEW: ISO C 9x
2688 cacosh NEW: ISO C 9x
2689 cacoshf NEW: ISO C 9x
2690 cacoshl NEW: ISO C 9x
2691 cacosl NEW: ISO C 9x
2698 casinf NEW: ISO C 9x
2699 casinh NEW: ISO C 9x
2700 casinhf NEW: ISO C 9x
2701 casinhl NEW: ISO C 9x
2702 casinl NEW: ISO C 9x
2704 catanf NEW: ISO C 9x
2705 catanh NEW: ISO C 9x
2706 catanhf NEW: ISO C 9x
2707 catanhl NEW: ISO C 9x
2708 catanl NEW: ISO C 9x
2709 cbc_crypt NEW: Secure RPC
2713 ccoshf NEW: ISO C 9x
2714 ccoshl NEW: ISO C 9x
2720 cimagf NEW: ISO C 9x
2721 cimagl NEW: ISO C 9x
2722 clearerr_locked REMOVED
2723 clntunix_create NEW: sunrpc ext
2725 clog10 NEW: ISO C 9x
2726 clog10f NEW: ISO C 9x
2727 clog10l NEW: ISO C 9x
2737 cprojf NEW: ISO C 9x
2738 cprojl NEW: ISO C 9x
2740 crealf NEW: ISO C 9x
2741 creall NEW: ISO C 9x
2746 csinhf NEW: ISO C 9x
2747 csinhl NEW: ISO C 9x
2750 csqrtf NEW: ISO C 9x
2751 csqrtl NEW: ISO C 9x
2755 ctanhf NEW: ISO C 9x
2756 ctanhl NEW: ISO C 9x
2758 des_setparity NEW: Secure RPC
2759 ecb_crypt NEW: Secure RPC
2760 endutxent NEW: Unix98
2762 exp10f NEW: ISO C 9x
2763 exp10l NEW: ISO C 9x
2767 fattach NEW: STREAMS
2768 fdetach NEW: STREAMS
2772 feclearexcept NEW: ISO C 9x
2773 fegetenv NEW: ISO C 9x
2774 fegetexceptflag NEW: ISO C 9x
2775 fegetround NEW: ISO C 9x
2776 feholdexcept NEW: ISO C 9x
2778 feraiseexcept NEW: ISO C 9x
2779 ferror_locked REMOVED
2780 fesetenv NEW: ISO C 9x
2781 fesetexceptflag NEW: ISO C 9x
2782 fesetround NEW: ISO C 9x
2783 fetestexcept NEW: ISO C 9x
2784 feupdateenv NEW: ISO C 9x
2785 fflush_locked REMOVED
2789 fgets_unlocked NEW: GNU ext.
2790 fileno_locked REMOVED
2802 fputc_locked REMOVED
2803 fputs_unlocked NEW: GNU ext.
2804 fread_unlocked NEW: GNU ext.
2809 fstatvfs NEW: Unix98
2813 ftruncate64 NEW: LFS
2815 fwrite_unlocked NEW: GNU ext.
2816 gai_strerror NEW: IPv6
2820 getchar_locked REMOVED
2822 getdate_err NEW: Unix98
2823 getdate_r NEW: GNU ext.
2825 getnameinfo NEW: IPv6
2826 getnetname NEW: Secure RPC
2827 getpmsg NEW: STREAMS
2828 getpt NEW: Unix98 PTY
2829 getrlimit64 NEW: LFS
2830 getutxent NEW: Unix98
2831 getutxid NEW: Unix98
2832 getutxline NEW: Unix98
2833 glob64 NEW: GNU ext.
2834 globfree64 NEW: GNU ext.
2835 gnu_get_libc_release NEW: GNU ext.
2836 gnu_get_libc_version NEW: GNU ext.
2837 grantpt NEW: Unix98 PTY
2838 host2netname NEW: Secure RPC
2840 iconv_close NEW: iconv
2841 iconv_open NEW: iconv
2842 if_freenameindex NEW: IPv6
2843 if_indextoname NEW: IPv6
2844 if_nameindex NEW: IPv6
2845 if_nametoindex NEW: IPv6
2846 in6addr_any NEW: IPv6
2847 in6addr_loopback NEW: IPv6
2848 inet6_isipv4mapped NEW: IPv6
2849 isastream NEW: STREAMS
2850 iswblank NEW: GNU ext.
2851 key_decryptsession NEW: Secure RPC
2852 key_decryptsession_pk NEW: Secure RPC
2853 key_encryptsession NEW: Secure RPC
2854 key_encryptsession_pk NEW: Secure RPC
2855 key_gendes NEW: Secure RPC
2856 key_get_conv NEW: Secure RPC
2857 key_secretkey_is_set NEW: Secure RPC
2858 key_setnet NEW: Secure RPC
2859 key_setsecret NEW: Secure RPC
2860 llrint NEW: ISO C 9x
2861 llrintf NEW: ISO C 9x
2862 llrintl NEW: ISO C 9x
2863 llround NEW: ISO C 9x
2864 llroundf NEW: ISO C 9x
2865 llroundl NEW: ISO C 9x
2870 lrintf NEW: ISO C 9x
2871 lrintl NEW: ISO C 9x
2872 lround NEW: ISO C 9x
2873 lroundf NEW: ISO C 9x
2874 lroundl NEW: ISO C 9x
2876 makecontext NEW: Unix98
2877 mempcpy NEW: GNU ext.
2880 modify_ldt NEW: kernel
2884 nearbyint NEW: ISO C 9x
2885 nearbyintf NEW: ISO C 9x
2886 nearbyintl NEW: ISO C 9x
2887 netname2host NEW: Secure RPC
2888 netname2user NEW: Secure RPC
2889 nexttoward NEW: ISO C 9x
2890 nexttowardf NEW: ISO C 9x
2891 nexttowardl NEW: ISO C 9x
2895 passwd2des NEW: Secure RPC
2897 pow10f NEW: GNU ext.
2898 pow10l NEW: GNU ext.
2901 printf_size NEW: GNU ext.
2902 printf_size_info NEW: GNU ext.
2903 profil_counter REMOVED
2904 pthread_mutexattr_getkind_np REPLACED
2905 pthread_mutexattr_setkind_np REPLACED
2906 ptsname NEW: Unix98 PTY
2907 ptsname_r NEW: Unix98 PTY
2909 putchar_locked REMOVED
2910 putgrent NEW: GNU ext.
2912 putpmsg NEW: STREAMS
2913 pututxline NEW: Unix98
2917 readdir64_r NEW: LFS
2918 remquo NEW: ISO C 9x
2919 remquof NEW: ISO C 9x
2920 remquol NEW: ISO C 9x
2922 roundf NEW: ISO C 9x
2923 roundl NEW: ISO C 9x
2925 scalbln NEW: ISO C 9x
2926 scalblnf NEW: ISO C 9x
2927 scalblnl NEW: ISO C 9x
2929 sendfile NEW: kernel
2930 setcontext NEW: Unix98
2931 setrlimit64 NEW: LFS
2932 setutxent NEW: Unix98
2934 sigignore NEW: Unix98
2935 sigqueue NEW: POSIX.1b
2936 sigrelse NEW: Unix98
2937 sigset NEW: POSIX.1b
2938 sigtimedwait NEW: POSIX.1b
2939 sigwaitinfo NEW: POSIX.1b
2940 sincos NEW: GNU ext.
2941 sincosf NEW: GNU ext.
2942 sincosl NEW: GNU ext.
2946 strcasestr NEW: GNU ext.
2947 strtoimax NEW: ISO C 9x
2948 strtoumax NEW: ISO C 9x
2949 strverscmp NEW: GNU ext.
2950 svcauthdes_stats NEW: Secure RPC
2951 svcunix_create NEW: sunrpc etx
2952 svcunixfd_create NEW: sunrpc ext
2953 swapcontext NEW: Unix98
2954 tcgetsid NEW: Unix98 PTY
2955 tdestroy NEW: GNU ext.
2956 tgamma NEW: ISO C 9x
2957 tgammaf NEW: ISO C 9x
2958 tgammal NEW: ISO C 9x
2962 truncf NEW: ISO C 9x
2963 truncl NEW: ISO C 9x
2965 unlockpt NEW: Unix98 PTY
2966 updwtmpx NEW: Unix98
2967 user2netname NEW: Secure RPC
2968 utmpxname NEW: Unix98
2969 versionsort NEW: GNU ext.
2970 versionsort64 NEW: GNU ext.
2972 wcscasecmp NEW: GNU ext.
2973 wcsncasecmp NEW: GNU ext.
2974 wcsnlen NEW: GNU ext.
2975 wcstoimax NEW: ISO C 9x
2976 wcstoll NEW: ISO C 9x
2977 wcstoull NEW: ISO C 9x
2978 wcstoumax NEW: ISO C 9x
2980 wordexp NEW: POSIX.2
2981 wordfree NEW: POSIX.2
2982 write_profiling REMOVED
2983 xdecrypt NEW: Secure RPC
2984 xdr_authdes_cred NEW: Secure RPC
2985 xdr_authdes_verf NEW: Secure RPC
2986 xdr_cryptkeyarg NEW: Secure RPC
2987 xdr_cryptkeyarg2 NEW: Secure RPC
2988 xdr_cryptkeyres NEW: Secure RPC
2989 xdr_getcredres NEW: Secure RPC
2990 xdr_key_netstarg NEW: Secure RPC
2991 xdr_key_netstres NEW: Secure RPC
2992 xdr_keybuf NEW: Secure RPC
2993 xdr_keystatus NEW: Secure RPC
2994 xdr_netnamestr NEW: Secure RPC
2995 xdr_sizeof NEW: Secure RPC
2996 xdr_unixcred NEW: sunrpc ext
2997 xencrypt NEW: Secure RPC
2998 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3009 * inet_ntoa is thread-safe
3011 * updwtmp is moved from libutil to libc
3013 * rewrite of cbrt function
3015 * update of timezone data
3029 * add atoll function
3031 * fix complex problems in Berkeley DB code
3033 * fix math functions
3037 * fixed lots of header problems (especially Linux/GNU specific)
3039 * dynamic loader preserves all registers
3041 * Roland McGrath provided support for handling of auxiliary objects in
3042 the ELF dynamic loader.
3044 * support for parallel builds is improved
3048 * GNU extensions are no longer declared by default. To enable them you
3049 must define the macro `_GNU_SOURCE' in your program or compile with
3052 * The library has changed from using GNU ld symbol aliases to using weak
3053 symbols where available. The ELF object file format supports weak
3054 symbols; GNU ld also supports weak symbols in the a.out format. (There
3055 is also now support for other GNU ld extensions in ELF. Use the
3056 `--with-elf' option to configure to indicate you have ELF, and
3057 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3058 of many files which contained only symbol aliases, reducing the size of
3059 the source and the compiled library; many other files were renamed to
3060 less cryptic names previously occupied by the symbol alias files.
3061 There is a new header file <elf.h> for programs which operate on
3062 files in the ELF format.
3064 * Converted to Autoconf version 2, so `configure' has more options.
3065 Run `configure --help' to see the details.
3067 * The library can now be configured to build profiling, highly-optimized
3068 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3069 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3070 `configure' enable building these extra libraries. The shared library is
3071 built by default when using both ELF and GNU ld. When shared libraries
3072 are enabled, the new library `-ldl' is available for arbitrary run-time
3073 loading of shared objects; its interface is defined in <dlfcn.h>. The
3074 new header file <link.h> gives access to the internals of the run-time
3075 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3076 application of same name on other systems and it provides information
3077 about dynamically linked binaries.
3079 * The C library now provides the run-time support code for profiling
3080 executables compiled with `-pg'. Programs can control the profiling code
3081 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3082 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3083 the support for this file format was contributed by David Mosberger-Tang.
3085 * The math code has been replaced with a math library based on fdlibm from
3086 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3087 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3088 functions. The math functions now reside in a separate library, so
3089 programs using them will need to use `-lm' their linking commands.
3091 * John C. Bowman contributed optimized ix87 assembler inline functions.
3093 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3094 mechanism similar to that found in Solaris 2. This is now used for the
3095 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3096 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3097 file controls what services are used for each individual database. This
3098 works by loading shared libraries with names specified in `nsswitch.conf',
3099 so service modules can be changed or added at any time without even
3100 relinking any program. Currently there are the file, db, and NIS based
3101 NSS services available.
3103 * The new functions `strtoq' and `strtouq' parse integer values from
3104 strings, like `strtol' and `strtoul', but they return `long long int' and
3105 `unsigned long long int' values, respectively (64-bit quantities).
3107 * The new functions `strtof' and `strtold' parse floating-point values from
3108 strings, like `strtod', but they return `float' and `long double' values,
3109 respectively (on some machines `double' and `long double' are the same).
3111 * Ulrich Drepper has contributed new implementations of the floating-point
3112 printing and reading code used in the `printf' family of functions and
3113 `strtod', `strtof', and `strtold'. These new functions are perfectly
3114 accurate, and much faster than the old ones.
3116 * The implementation of the POSIX locale model was completely rewritten by
3117 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3118 to compile the POSIX locale definition.
3120 * The former dummy implementations of the strcoll and strxfrm function are
3121 now replaced by fully functional code contributed by Ulrich Drepper. The
3122 collation information comes from the POSIX locale definitions.
3124 * The new header <langinfo.h> defines an interface for accessing
3125 various locale-dependent data (using the locale chosen with `setlocale').
3127 * Ulrich Drepper has contributed a new suite of functions for operation on
3128 wide-character and multibyte-character strings, in <wchar.h>;
3129 and classification and case conversion of wide characters, in <wctype.h>.
3130 These new functions are conforming to the ISO C, Amendement 1 specification.
3132 * There is now a second implementation of the standard I/O library available.
3133 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3134 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3136 * You can now use positional parameter specifications in format strings
3137 for the `printf' and `scanf' families of functions. For example,
3138 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3139 ``Number 6, Mr Jones''. This is mainly useful when providing different
3140 format strings for different languages, whose grammars may dictate
3141 different orderings of the values being printed. To support this
3142 feature, the interface for `register_printf_handler' has changed; see
3143 the header file <printf.h> for details.
3145 * The `printf' and `scanf' families of functions now understand a new
3146 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3147 to group numbers as indicated by the locale; for `scanf' and friends, this
3148 says to accept as valid only a number with all the proper grouping
3149 separators in the right places. In the default "C" locale, numbers are
3150 not grouped; but locales for specific countries will define the usual
3151 conventions (i.e. separate thousands with `,' in the US locale).
3153 * The pgrp functions have been regularized, slightly incompatibly but much
3154 less confusingly. The core functions are now `getpgid' and `setpgid',
3155 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3156 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3157 provided for compatibility. There is no longer an incompatible `getpgrp'
3158 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3160 * The new header file <fts.h> and suite of functions simplify programs that
3161 operate on directory trees. This code comes from 4.4 BSD.
3163 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3164 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3165 NSS scheme used in glibc.
3167 * There is a new malloc debugging hook `__memalign_hook'.
3169 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3170 `unsigned int' in <sys/types.h>. These are for compatibility only and
3171 their use is discouraged.
3173 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3174 done differently, so that it works even without GNU ld.
3176 * New function `euidaccess' checks allowed access to a file like `access',
3177 but using the effective IDs instead of the real IDs.
3179 * The time zone data files have been updated for the latest and greatest
3180 local time conventions of the countries of the world.
3182 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3185 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3186 for formatting floating-point numbers. They are provided only for
3187 compatibility; new programs should use `sprintf' instead. There are
3188 also equivalent function for the `long double' floating-point type and
3189 all functions also exist in a reentrant form.
3191 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3192 functions for maintaining the login-record files (primarily of use to
3193 system programs such as `login'), and convenient functions for
3194 allocating and initializing a pseudo-terminal (pty) device.
3196 * Ulrich Drepper has contributed new support for System V style
3197 shared memory and IPC on systems that support it.
3199 * Ulrich Drepper has contributed several miscellaneous new functions found
3200 in System V: The `hsearch' family of functions provide an effective
3201 implementation of hash tables; `a64l' and `l64a' provide a very simple
3202 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3205 * Ulrich Drepper has contributed new reentrant counterparts for the
3206 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3208 * Ulrich Drepper has contributed new, highly-optimized versions of several
3209 string functions for the i486/Pentium family of processors.
3211 * Ulrich Drepper has updated the Linux-specific code, based largely
3212 on work done in Hongjiu Lu's version of GNU libc for Linux.
3213 The GNU library now supports Linux versions 2.0.10 and later,
3214 using the ELF object file format (i[3456]86-*-linux).
3216 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3218 * David Mosberger-Tang and Richard Henderson have ported the C library
3219 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3220 dynamic linking support for ELF/Alpha.
3222 * Richard Henderson contributed several Alpha optimized assembler function
3223 for arithmetic and string handling.
3225 * Ulrich Drepper has contributed a new set of message catalog functions to
3226 support multiple languages using the <libintl.h> interface, for use with
3227 his new package GNU gettext. Translation volunteers have contributed
3228 catalogs of the library's messages in Spanish, German, and Korean.
3230 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3231 program and the `catgets' function for reading the catalog files it
3232 creates. (The <libintl.h> interface is preferred; we include the
3233 <nl_types.h> interface using `catgets' only for source compatibility with
3234 programs already written to use it.)
3236 * New header file <values.h> gives SVID-compatible names for <limits.h>
3239 * Various new macros, declarations, and small header files for compatibility
3242 * New function `group_member' is a convenient way to check if a process has
3243 a given effective group ID.
3245 * When using GCC 2.7 and later, the socket functions are now declared in a
3246 special way so that passing an argument of type `struct sockaddr_in *',
3247 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3248 `struct sockaddr *' type, does not generate a type-clash warning.
3250 * New function `error' declared in header file <error.h> is a convenient
3251 function for printing error messages and optionally exiting; this is the
3252 canonical function used in GNU programs. The new functions `err', `warn',
3253 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3254 doing the same thing.
3256 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3257 POSIX.2 `glob' function to do ~ and {...} expansion.
3259 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3260 `clearenv' which is used in POSIX.9 is also available.
3262 * New function `getsid' returns session ID number on systems that support it.
3264 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3265 files <db.h> and <mpool.h> provide a rich set of functions for several
3266 types of simple databases stored in memory and in files, and <ndbm.h> is
3267 an old `ndbm'-compatible interface using the `db' functions. Link with
3268 `-ldb' to get these functions.
3270 * New macro `strdupa' copies a string like `strdup', but uses local stack
3271 space from `alloca' instead of dynamic heap space from `malloc'.
3273 * New function `strnlen' is like `strlen' but searches only a given maximum
3274 number of characters for the null terminator. `stpncpy', `strndup' and
3275 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3278 * New function `statfs' in header <sys/statfs.h>.
3280 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3281 provide convenient functions for operating on blocks of null-terminated
3284 * A new suite of functions in <utmp.h> handle all the details of reading
3285 and writing the utmp file.
3287 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3290 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3291 completely POSIX compliant and also implemented the extended functionality
3292 to handle alternate digit representation and alternate era date formats.
3294 * Ulrich Drepper provided an implementation of the `strptime' function
3295 defined in XPG4.2 which transforms a string into a `struct tm' value.
3297 * Paul Eggert provided the tzselect shell script as part of the timezone
3298 code. The shell script makes it easy to select the correct timezone
3301 * The implementation of the malloc family of functions is completely replaced
3302 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3303 The implementation uses the mmap function (if available) and it is
3304 optimized for the use in multi threaded programs.
3306 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3307 functions. This new functionality is usable by specifying a special
3308 salt string and it is compatible with implementation on *BSD systems.
3310 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3311 `getsubopt' to handle second level command line options, `bsd_signal'
3312 to access BSD style `signal' functionality, the obsolete `regexp' style
3315 * the `lchown' function is available on system which support this
3318 * The implementation of the shadow password handling function was contributed
3321 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3323 * POSIX.1g support was added. The <sys/select.h> header is available,
3324 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3325 implementation of `getaddrinfo'.
3329 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3331 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3332 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3334 * The new function `daemon' from 4.4 BSD is useful for server programs that
3335 want to put themselves in the background.
3337 * Joel Sherrill has contributed support for several standalone boards that
3338 run without an operating system.
3340 * `printf', `scanf' and friends now accept a `q' type modifier for long
3341 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3343 * All of the code taken from BSD (notably most of the math and networking
3344 routines) has been updated from the BSD 4.4-Lite release.
3346 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3348 * The new functions `getdomainname' and `setdomainname' fetch or change the
3349 YP/NIS domain name. These are system calls which exist on systems which
3352 * The time zone data files have been updated for the latest international
3355 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3356 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3360 * The C library now includes support for Sun RPC, from Sun's free
3361 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3362 are included. (There is still no support for YP.)
3364 * Tom Quinn has contributed a port of the C library to SGI machines running
3365 Irix 4 (mips-sgi-irix4).
3367 * The new `lockf' function is a simplified interface to the locking
3368 facilities of `fcntl', included for compatibility.
3370 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3372 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3374 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3377 * The error message printed by `assert' for a failed assertion now includes
3378 the name of the program (if using GNU ld) and the name of the calling
3379 function (with versions of GCC that support this).
3381 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3383 * The library now includes the <sys/mman.h> header file and memory
3384 management functions `mmap', `munmap', `mprotect', `msync', and
3385 `madvise', on systems that support those facilities.
3387 * The interface for `mcheck' has changed slightly: the function called to
3388 abort the program when an allocation inconsistency is detected now takes
3389 an argument that indicates the type of failure. The new function
3390 `mprobe' lets you request a consistency check for a particular block at
3391 any time (checks are normally done only when you call `free' or `realloc'
3394 * It is now possible to easily cross-compile the C library, building on one
3395 system a library to run on another machine and/or operating system. All
3396 you need to do is set the variable `HOST_CC' in `configparms' to the
3397 native compiler for programs to run on the machine you are building on (a
3398 few generator programs are used on Unix systems); set `CC' to the
3401 * The new function `fexecve' (only implemented on the GNU system) executes
3402 a program file given a file descriptor already open on the file.
3406 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3407 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3409 * You can set the variable `libprefix' in `configparms' to specify a prefix
3410 to be prepended to installed library files; this makes it easy to install
3411 the GNU C library to be linked as `-lgnuc' or whatever.
3413 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3414 copies a limited number of characters from a string, and returns the
3415 address of the last character written.
3417 * You no longer need to check for whether the installed `stddef.h' is
3418 compatible with the GNU C library. configure now checks for you.
3420 * You can now define a per-stream `fileno' function to convert the
3421 stream's cookie into an integral file descriptor.
3423 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3424 allocates zero bytes of storage, and returns a unique pointer which
3425 you can pass to `realloc' or `free'. The behavior is undefined if
3426 you dereference this pointer.
3428 * The C library now runs on Sony NEWS m68k machines running either
3429 NewsOS 3 or NewsOS 4.
3431 * The new `syscall' function is a system-dependent primitive function
3432 for invoking system calls. It has the canonical behavior on Unix
3433 systems, including unreliable return values for some calls (such as
3434 `pipe', `fork' and `getppid').
3436 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3437 to `EAGAIN', which is the preferred name. On systems whose kernels
3438 use two distinct codes, the C library now translates EWOULDBLOCK to
3439 EAGAIN in every system call function.
3443 * The GNU C Library Reference Manual is now distributed with the library.
3444 `make dvi' will produce a DVI file of the printed manual.
3445 `make info' will produce Info files that you can read on line using C-h i
3446 in Emacs or the `info' program.
3447 Please send comments on the manual to bug-glibc-manual@gnu.org.
3449 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3451 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3453 * Jason Merrill has contributed a port to the Sequent Symmetry running
3454 Dynix version 3 (i386-sequent-dynix).
3456 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3457 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3459 * New function `memory_warnings' lets you arrange to get warnings when
3460 malloc is running out of memory to allocate, like Emacs gives you.
3462 * The C library now contains the relocating allocator used in Emacs 19 for
3463 its editing buffers. This allocator (ralloc) minimizes allocation
3464 overhead and fragmentation by moving allocated regions around whenever it
3465 needs to. You always refer to a ralloc'd region with a "handle" (a
3466 pointer to a pointer--an object of type `void **').
3468 * There is a new `printf' format: `%m' gives you the string corresponding
3469 to the error code in `errno'.
3471 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3472 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3473 pass, the `a' modifier says to fill in a `char **' you pass with a
3476 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3477 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3478 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3480 * `mkstemp' is a traditional Unix function to atomically create and open a
3481 uniquely-named temporary file.
3485 * The standard location for the file that says what the local timezone is
3486 has changed again. It is now `/usr/local/etc/localtime' (or more
3487 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3489 * The distribution no longer contains any files with names longer than 14
3492 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3493 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3495 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3499 * The configuration process has changed quite a bit. The `configure'
3500 script is now used just like the configuration scripts for other GNU
3501 packages. The `sysdeps' directory hierarchy is much rearranged.
3502 The file `INSTALL' explains the new scheme in detail.
3504 * The header files no longer need to be processed into ANSI C and
3505 traditional C versions. There is just one set of files to install, and
3506 it will work with ANSI or old C compilers (including `gcc -traditional').
3508 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3509 MIPS DECStation running Ultrix 4.
3511 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3512 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3513 made itself into a shared library.
3515 * Yet further improved support for the i386, running 4.3 BSD-like systems
3516 (such as Mach 3 with the Unix single-server), or System V.
3518 * New function `strncasecmp' to do case-insensitive string comparison
3519 with limited length.
3521 * New function `strsep' is a reentrant alternative to `strtok'.
3523 * New functions `scandir' and `alphasort' for searching directories.
3525 * New function `setenv' is a better interface to `putenv'.
3527 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3528 function for traversing a directory tree.
3530 * The GNU obstack package is now also part of the C library.
3531 The new function `open_obstack_stream' creates a stdio stream that
3532 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3533 formatted output directly to an obstack.
3535 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3536 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3538 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3540 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3541 things to your strings.
3543 * There are some new test programs: test-fseek, testmb, and testrand.
3545 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3546 These ports are not finished, but are a good starting place for really
3547 supporting those systems.
3549 * `/etc/localtime' is now the standard location for the file that says what
3550 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3551 This follows the general principle that `/etc' is the place for all local
3552 configuration files.
3554 * The C library header files now use `extern "C"' when used by the C++
3555 compiler, so the C library should now work with C++ code.
3557 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3558 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3561 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3562 function which uses the `merge sort' algorithm, and is said to be
3563 significantly faster than the old GNU `qsort' function. Merge sort is
3564 now the standard `qsort' function. The new algorithm can require a lot
3565 of temporary storage; so, the old sorting function is called when the
3566 required storage is not available.
3568 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3569 provides the Unix `crypt' function, plus some other entry points.
3571 * The code and header files taken from 4.4 BSD have been updated with the
3572 latest files released from Berkeley.
3574 ----------------------------------------------------------------------
3575 Copyright information:
3577 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3579 Permission is granted to anyone to make or distribute verbatim copies
3580 of this document as received, in any medium, provided that the
3581 copyright notice and this permission notice are preserved,
3582 thus giving the recipient permission to redistribute in turn.
3584 Permission is granted to distribute modified versions
3585 of this document, or of portions of it,
3586 under the above conditions, provided also that they
3587 carry prominent notices stating who last changed them.
3590 version-control: never