1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
10 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
11 24731-2:2010, is supported to enable declarations of functions from that
12 TR. Note that not all functions from that TR are supported by the GNU C
15 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
16 18661-1:2014, is supported to enable declarations of functions and macros
17 from that TS. Note that most features from that TS are not supported by
20 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
21 18661-4:2015, is supported to enable declarations of functions and macros
22 from that TS. Note that most features from that TS are not supported by
25 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
26 means that in a future release, the macros “major”, “minor”, and “makedev”
27 will only be available from <sys/sysmacros.h>.
29 These macros are not part of POSIX nor XSI, and their names frequently
30 collide with user code; see for instance glibc bug 19239 and Red Hat bug
31 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
32 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
35 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
36 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions,
37 the femode_t type and the FE_DFL_MODE macro.
39 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
40 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
41 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
42 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
43 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
44 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
45 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
46 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
47 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
48 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
49 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
52 * New <math.h> features are added from TS 18661-1:2014:
54 - Comparison macros: iseqsig.
56 - Classification macros: iscanonical, issubnormal, iszero.
58 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
59 for the Linux quota interface which predates kernel version 2.4.22 has
62 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
63 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
64 “no-ip6-dotint” had already been the default, and support for the
65 “ip6-dotint” option was removed from the Internet in 2006.
67 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG defined
68 in the <resolv.h> header file have been deprecated. They were already
71 * DNSSEC-related declarations and definitions have been removed from the
72 <arpa/nameser.h> header file, and libresolv will no longer attempt to
73 decode the data part of DNSSEC record types. Previous versions of glibc
74 only implemented minimal support for the previous version of DNSSEC, which
75 is incompatible with the currently deployed version.
77 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
78 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
79 <arpa/nameser.h> header file because the distinction between RR types and
80 meta-RR types is not officially standardized, subject to revision, and
81 thus not suitable for encoding in a macro.
83 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
84 and rhook members of the res_state type in <resolv.h> have been removed.
85 The glibc stub resolver did not support these hooks, but the header file
88 * For multi-arch support it is recommended to use a GCC which has
89 been built with support for GNU indirect functions. This ensures
90 that correct debugging information is generated for functions
91 selected by IFUNC resolvers. This support can either be enabled by
92 configuring GCC with '--enable-gnu-indirect-function', or by
93 enabling it by default by setting 'default_gnu_indirect_function'
94 variable for a particular architecture in the GCC source file
97 Security related changes:
99 On ARM EABI (32-bit), generating a backtrace for execution contexts which
100 have been created with makecontext could fail to terminate due to a
101 missing .cantunwind annotation. This has been observed to lead to a hang
102 (denial of service) in some Go applications compiled with gccgo. Reported
103 by Andreas Schwab. (CVE-2016-6323)
105 The following bugs are resolved with this release:
107 [The release manager will add the list generated by
108 scripts/list-fixed-bugs.py just before the release.]
112 * The minimum Linux kernel version that this version of the GNU C Library
113 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
114 kernel version 2.6.32 or later suffices (on architectures that already
115 required kernel versions more recent than 3.2, those requirements remain
116 unchanged). Linux 3.2 or later kernel headers are required on all
119 * The pap_AN locale has been deleted. This has been deprecated for a long
120 time. It has been replaced by pap_AW & pap_CW, both of which have long
121 been included in previous releases.
123 * The readdir_r and readdir64_r functions have been deprecated. It is
124 recommended to use readdir and readdir64 instead.
126 * The type “union wait” has been removed. It was deprecated in the early
127 1990s and never part of POSIX. Application code should use the int type
128 instead of “union wait”.
130 * A new NSS action is added to facilitate large distributed system
131 administration. The action, MERGE, allows remote user stores like LDAP
132 to be merged into local user stores like /etc/groups in order to provide
133 easy to use, updated, and managed sets of merged credentials. The new
134 action can be used by configuring it in /etc/nsswitch.conf:
135 group: files [SUCCESS=merge] nis
136 Implemented by Stephen Gallagher (Red Hat).
138 * The deprecated __malloc_initialize_hook variable has been removed from the
141 * The long unused localedef --old-style option has been removed. It hasn't
142 done anything in over 16 years. Scripts using this option can safely
145 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
146 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
147 return the next representable value in the direction of positive infinity
148 and the nextdown functions return the next representable value in the
149 direction of negative infinity. These are currently enabled as GNU
152 Security related changes:
154 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
155 could result in a stack overflow when getnetbyname was called with an
156 overly long name. (CVE-2016-3075)
158 * Previously, getaddrinfo copied large amounts of address data to the stack,
159 even after the fix for CVE-2013-4458 has been applied, potentially
160 resulting in a stack overflow. getaddrinfo now uses a heap allocation
161 instead. Reported by Michael Petlan. (CVE-2016-3706)
163 * The glob function suffered from a stack-based buffer overflow when it was
164 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
165 Reported by Alexander Cherepanov. (CVE-2016-1234)
167 * The Sun RPC UDP client could exhaust all available stack space when
168 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
169 alloca plugin for GCC. (CVE-2016-4429)
171 * The IPv6 name server management code in libresolv could result in a memory
172 leak for each thread which is created, performs a failing naming lookup,
173 and exits. Over time, this could result in a denial of service due to
174 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
176 The following bugs are resolved with this release:
178 [1170] localedata: ne_NP: update Nepali locale definition file
179 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
181 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
182 [6796] math: fdim() does not set errno on overflow
183 [10354] libc: posix_spawn should use vfork() in more cases than presently
184 [11213] localedata: localedata: add copyright disclaimer to locale files
185 [12143] localedata: chr_US: new Cherokee locale
186 [12450] localedata: sgs_LT: new locale
187 [12676] localedata: ln_CD: new locale
188 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
190 [13304] math: fma, fmaf, fmal produce wrong results
191 [14259] build: --localedir arg to configure is ignored
192 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
193 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
194 [14934] localedata: es_CL: wrong first weekday chilean locale
195 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
197 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
199 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
200 [15368] nptl: raise() is not async-signal-safe
201 [15479] math: ceil, floor, round and trunc raise inexact exception
202 [15578] localedata: kk_KZ: various updates
203 [16003] localedata: pap_AN: punt old locale
204 [16137] localedata: iw_IL: punt old locale
205 [16190] localedata: eo: new esperanto locale
206 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
207 [16742] malloc: race condition: pthread_atfork() called before first
208 malloc() results in unexpected locking behaviour/deadlocks
209 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
211 [16983] localedata: postal_fmt does not allow %l and %n modifiers
212 [17565] localedata: pt_PT: wrong (work-)week start
213 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
214 [17950] build: Build fails with -msse
215 [18205] localedata: be_BY*: wrong first_weekday and first_workday
216 [18433] libc: posix_spawn does not return correctly upon failure to
218 [18453] localedata: charmaps/IBM875: incorrect codes
219 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
221 [18896] localedata: he_IL: improvements for currency
222 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
224 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
226 [19056] libc: Deprecate readdir_r
227 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
229 [19198] localedata: nl_NL: small improvements for Dutch locales
230 [19257] network: Per-thread memory leak in __res_vinit with IPv6
231 nameservers (CVE-2016-5417)
232 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
234 [19400] locale: Language missing in "iso-639.def", trivial fix in
236 [19431] malloc: Deadlock between fflush, getdelim, and fork
237 [19505] libc: Incorrect file descriptor validity checks in
238 posix_spawn_file_actions_add{open,close,dup2}
239 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
241 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
242 `intl/{gettextP,loadinfo}.h'
243 [19534] libc: execle, execlp may use malloc
244 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
245 [19573] network: res_nclose and __res_maybe_init disagree about name
246 server initialization, breaking Hesiod
247 [19575] localedata: Status of GB18030 tables
248 [19581] localedata: sr_* date_fmt string contains additional newline
249 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
251 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
252 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
253 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
254 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
256 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
258 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
260 [19610] dynamic-link: ldconfig -X removes stale symbolic links
261 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
262 [19633] locale: strfmon_l applies global locale to number formatting
263 [19642] network: Memory leak in getnameinfo
264 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
265 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
267 [19654] math: [x86_64] Need testcase for BZ #19590 fix
268 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
270 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
271 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
273 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
275 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
277 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
278 pointers and lengths in error-case.
279 [19727] locale: Converting from/to UTF-xx with iconv() does not always
280 report errors on UTF-16 surrogates values.
281 [19755] nscd: nscd assertion failure in gc
282 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
283 [19759] libc: mempcpy shouldn't be inlined
284 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
285 [19765] libc: s390 needs an optimized mempcpy
286 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
287 NAME_MAX limit assumption (CVE-2016-1234)
288 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
289 [19787] network: Missing and incorrect truncation checks in getnameinfo
290 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
292 [19791] network: Assertion failure in res_query.c with un-connectable name
294 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
295 [19822] math: libm.so install clobbers old version
296 [19825] network: resolv: send_vc can return uninitialized data in second
297 response to getaddrinfo
298 [19830] network: nss_dns: should check RDATA length against buffer length
299 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
300 confronted with A/AAAA records of invalid size
301 [19837] nss: nss_db: No retries for some long lines with a larger buffer
302 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
303 [19853] stdio: Printing IBM long double in decimal with high precision is
305 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
306 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
307 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
309 [19865] network: Assertion failure or memory leak in
310 _nss_dns_getcanonname_r
311 [19868] network: nss_dns: netent code does not skip over non-PTR records
312 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
314 [19881] string: Improve x86-64 memset
315 [19907] string: Incorrect memcpy tests
316 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
318 [19925] libc: termios.h XCASE namespace
319 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
320 [19929] libc: limits.h NL_NMAX namespace
321 [19931] stdio: Memory leak in vfprintf
322 [19957] libc: clone(CLONE_VM) access invalid parent memory
323 [19963] localedata: en_IL: New locale
324 [19989] stdio: stdio.h cuserid namespace
325 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
327 [19996] locale: langinfo.h nl_langinfo_l namespace
328 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
329 [20010] network: getaddrinfo: Stack overflow in hostent translation
331 [20012] stdio: libio: fmemopen append mode failure
332 [20014] stdio: stdio.h namespace for pre-threads POSIX
333 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
334 [20023] libc: fcntl.h timespec namespace
335 [20024] math: [x86_64] vectorized sincos trashes the stack
336 [20031] network: nss_hesiod: Heap overflow in get_txt_records
337 [20041] time: sys/time.h timespec namespace
338 [20043] libc: unistd.h missing cuserid for UNIX98 and before
339 [20044] libc: unistd.h missing pthread_atfork for UNIX98
340 [20051] libc: ttyslot in wrong header under wrong conditions
341 [20054] libc: gethostname not declared for XPG4
342 [20055] libc: termios.h missing tcgetsid for XPG4
343 [20072] dynamic-link: x86 init_cpu_features is called twice in static
345 [20073] libc: sys/stat.h fchmod namespace
346 [20074] libc: stdlib.h rand_r namespace
347 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
348 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
350 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
351 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
353 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
354 [20119] libc: Wrong mask for processors level type from CPUID
355 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
356 AS not supporting AVX512
357 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
359 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
360 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
362 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
363 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
364 [20178] libc: posix_spawn{p} should not call exit
365 [20191] stdio: libio: vtables hardening
366 [20195] string: FMA4 detection requires CPUID execution with register
368 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
369 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
371 [20212] math: acos (sNaN) returns sNaN
372 [20213] math: asin (sNaN) returns sNaN
373 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
374 [20218] math: [i386] asinhl (sNaN) returns sNaN
375 [20219] math: [i386] atanhl (sNaN) returns sNaN
376 [20222] stdio: fopencookie: Mangle function pointers
377 [20224] math: [i386] cbrtl (sNaN) returns sNaN
378 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
379 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
381 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
382 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
383 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
384 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
385 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
386 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
387 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
388 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
389 [20240] math: modf (sNaN) returns sNaN
390 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
391 [20250] math: frexp (sNaN) returns sNaN
392 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
393 [20255] math: [i386] fdim, fdimf return with excess range and precision /
395 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
396 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
397 subscript is below array bounds [-Werror=array-bounds]
398 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
399 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
401 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
402 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
404 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
406 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
407 [20316] localedata: id_ID: Februari instead of Pebruari
408 [20327] string: POWER8 strcasecmp returns incorrect result
409 [20347] math: Failure: Test: j0_downward (0xap+0)
410 [20348] libc: FAIL: misc/tst-preadvwritev64
411 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
412 [20350] libc: There is no test for p{read,write}64
413 [20357] math: Incorrect cos result for 1.5174239687223976
414 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
418 * Unicode 8.0.0 Support: Character encoding, character type info, and
419 transliteration tables are all updated to Unicode 8.0.0, using new
420 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
421 These updates cause user visible changes, such as the fixes for bugs
422 89, 16061, and 18568.
424 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
425 kernel-internal CPU set size. This means that requests that change the
426 CPU affinity which failed before (for example, an all-ones CPU mask) will
427 now succeed. Applications that need to determine the effective CPU
428 affinities need to call sched_getaffinity or pthread_getaffinity_np after
429 setting it because the kernel can adjust it (and the previous size check
430 would not detect this in the majority of cases).
432 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
433 the following new symbols are used: fts64_children, fts64_close,
434 fts64_open, fts64_read and fts64_set.
436 * getaddrinfo now detects certain invalid responses on an internal netlink
437 socket. If such responses are received, an affected process will
438 terminate with an error message of "Unexpected error <number> on netlink
439 descriptor <number>" or "Unexpected netlink response of size <number> on
440 descriptor <number>". The most likely cause for these errors is a
441 multi-threaded application which erroneously closes and reuses the netlink
442 file descriptor while it is used by getaddrinfo.
444 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
445 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
446 unnecessary serialization of memory allocation requests across threads.
447 The defect is now corrected. Users should see a substantial increase in
448 the concurent throughput of allocation requests for applications which
449 trigger this bug. Affected applications typically create and
450 destroy threads frequently. (Bug 19048 was reported and analyzed by
453 * There is now a --disable-timezone-tools configure option for disabling the
454 building and installing of the timezone related utilities (zic, zdump, and
455 tzselect). This is useful for people who build the timezone data and code
456 independent of the GNU C Library.
458 * The obsolete header <regexp.h> has been removed. Programs that require
459 this header must be updated to use <regex.h> instead.
461 * The obsolete functions bdflush, create_module, get_kernel_syms,
462 query_module and uselib are no longer available to newly linked binaries;
463 the header <sys/kdaemon.h> has been removed. These functions and header
464 were specific to systems using the Linux kernel and could not usefully be
465 used with the GNU C Library on systems with version 2.6 or later of the
468 * Optimized string, wcsmbs and memory functions for IBM z13.
469 Implemented by Stefan Liebler.
471 * Newly linked programs that define a variable called signgam will no longer
472 have it set by the lgamma, lgammaf and lgammal functions. Programs that
473 require signgam to be set by those functions must ensure that they use the
474 variable provided by the GNU C Library and declared in <math.h>, without
475 defining their own copy.
477 * The minimum GCC version that can be used to build this version of the GNU
478 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
479 still be used to compile programs using the GNU C Library.
481 Security related changes:
483 * An out-of-bounds value in a broken-out struct tm argument to strftime no
484 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
486 * The LD_POINTER_GUARD environment variable can no longer be used to disable
487 the pointer guard feature. It is always enabled. Previously,
488 LD_POINTER_GUARD could be used to disable security hardening in binaries
489 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
492 * An integer overflow in hcreate and hcreate_r could lead to an
493 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
495 * The catopen function no longer has unbounded stack usage. Reported by
498 * The nan, nanf and nanl functions no longer have unbounded stack usage
499 depending on the length of the string passed as an argument to the
500 functions. Reported by Joseph Myers. (CVE-2014-9761)
502 * A stack-based buffer overflow was found in libresolv when invoked from
503 libnss_dns, allowing specially crafted DNS responses to seize control
504 of execution flow in the DNS client. The buffer overflow occurs in
505 the functions send_dg (send datagram) and send_vc (send TCP) for the
506 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
507 family. The use of AF_UNSPEC triggers the low-level resolver code to
508 send out two parallel queries for A and AAAA. A mismanagement of the
509 buffers used for those queries could result in the response of a query
510 writing beyond the alloca allocated buffer created by
511 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
512 the overflow. Thanks to the Google Security Team and Red Hat for
513 reporting the security impact of this issue, and Robert Holiday of
514 Ciena for reporting the related bug 18665. (CVE-2015-7547)
516 The following bugs are resolved with this release:
518 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
519 [887] math: Math library function "logb" and "nextafter" inconsistent
520 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
521 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
522 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
523 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
525 [4404] localedata: German translation of "Alarm clock" is misleading
526 [6799] math: nextafter() and nexttoward() doen't set errno on
527 overflow/underflow errors
528 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
530 [10432] nis: _nss_nis_setnetgrent assertion failure
531 [11460] libc: fts has no LFS support
532 [12926] network: getaddrinfo()/make_request() may spin forever
533 [13065] nptl: Race condition in pthread barriers
534 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
535 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
537 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
538 [14912] libc: Rename non-installed bits/*.h headers
539 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
540 [15367] math: Let gcc use __builtin_isinf
541 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
542 [15421] math: lgamma wrongly sets signgam for ISO C
543 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
544 FE_INVALID with argument out of range
545 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
547 [15786] dynamic-link: ifunc resolver functions can smash function
549 [15918] math: Unnecessary check for equality in hypotf()
550 [16061] localedata: Review / update transliteration data
551 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
553 [16141] time: strptime %z offset restriction
554 [16171] math: drem should be alias of remainder
555 [16296] math: fegetround is pure?
556 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
557 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
559 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
560 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
562 [16422] math: [powerpc] math-float, math-double failing llrint tests with
563 "Exception "Inexact" set" on ppc32
564 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
565 [16517] math: Missing underflow exception from tanf/tan/tanl
566 [16519] math: Missing underflow exception from sinhf
567 [16520] math: Missing underflow exception from tanhf
568 [16521] math: Missing underflow exception from exp2
569 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
571 [16734] stdio: fopen calls mmap to allocate its buffer
572 [16961] math: nan function incorrect handling of bad sequences
573 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
574 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
575 [16985] locale: localedef: confusing error message when opening output
577 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
578 [17197] locale: Redundant shift character in iconv conversion output at
580 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
581 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
582 [17250] dynamic-link: static linking breaks nss loading
583 (getaddrinfo/getpwnam/etc...)
584 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
586 [17441] math: isnan() should use __builtin_isnan() in GCC
587 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
588 (related to lock elision)
589 [17787] manual: Exponent on page 324 of the PDF ends prematurely
590 [17886] time: strptime should be able to parse "Z" as a timezone with %z
591 [17887] time: strptime should be able to parse "+01:00" style timezones
592 [17905] libc: catopen() Multiple unbounded stack allocations
594 [18084] libc: backtrace (..., 0) dumps core on x86
595 [18086] libc: nice() sets errno to 0 on success
596 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
597 is too large (CVE-2015-8778)
598 [18251] dynamic-link: SONAME missing when audit modules provides path
599 [18265] libc: add attributes for wchar string and memory functions
600 [18370] math: csqrt missing underflows
601 [18421] libc: [hppa] read-only segment has dynamic relocations
602 [18472] libc: Obsolete syscall wrappers should be compat symbols
603 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
604 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
605 [18525] localedata: Remove locale timezone information
606 [18560] libc: [powerpc] spurious bits/ipc.h definitions
607 [18568] localedata: Update locale data to Unicode 8.0
608 [18589] locale: sort-test.sh fails at random
609 [18595] math: ctan, ctanh missing underflows
610 [18604] libc: assert macro-expands its argument
611 [18610] math: S390: fetestexcept() reports any exception if DXC-code
612 contains a vector instruction exception.
613 [18611] math: j1, jn missing errno setting on underflow
614 [18618] localedata: sync Chechen locale definitions with other *_RU
616 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
617 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
618 [18665] network: In send_dg, the recvfrom function is NOT always using the
619 buffer size of a newly created buffer (CVE-2015-7547)
620 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
622 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
624 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
625 [18699] math: tilegx cproj() for various complex infinities does not yield
627 [18724] libc: Harden put*ent functions against data injection
628 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
629 [18755] build: build errors with -DNDEBUG
630 [18757] stdio: fmemopen fails to set errno on failure
631 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
633 [18781] libc: openat64 lacks O_LARGEFILE
634 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
635 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
636 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
637 [18790] math: [ldbl-128ibm] tanhl inaccurate
638 [18795] libc: stpncpy fortification misses buffer lengths that are
640 [18796] build: build fails for --disable-mathvec
641 [18803] math: hypot missing underflows
642 [18820] stdio: fmemopen may leak memory on failure
643 [18823] math: csqrt spurious underflows
644 [18824] math: fma spurious underflows
645 [18825] math: pow missing underflows
646 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
647 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
648 [18870] build: sem_open.c fails to compile with missing symbol
650 [18872] stdio: Fix memory leak in printf_positional
651 [18873] libc: posix_fallocate overflow check ineffective
652 [18875] math: Excess precision leads incorrect libm
653 [18877] libc: arm: mmap offset regression
654 [18887] libc: memory corruption when using getmntent on blank lines
655 [18918] localedata: hu_HU: change time to HH:MM:SS format
656 [18921] libc: Regression: extraneous stat() and fstat() performed by
658 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
659 binaries (CVE-2015-8777)
660 [18951] math: tgamma missing underflows
661 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
663 [18953] localedata: lt_LT: change currency symbol to the euro
664 [18956] math: powf inaccuracy
665 [18961] math: [i386] exp missing underflows
666 [18966] math: [i386] exp10 missing underflows
667 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
668 [18969] build: multiple string test failures due to missing locale
670 [18970] libc: Reference of pthread_setcancelstate in libc.a
671 [18977] math: float / long double Bessel functions not in XSI POSIX
672 [18980] math: i386 libm functions return with excess range and precision
673 [18981] math: i386 scalb*, ldexp return with excess range and precision
674 [18982] stdio: va_list and vprintf
675 [18985] time: Passing out of range data to strftime() causes a segfault
677 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
678 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
679 [19012] locale: iconv_open leaks memory on error path
680 [19016] math: clog, clog10 inaccuracy
681 [19018] nptl: Mangle function pointers in tls_dtor_list
682 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
683 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
684 [19048] malloc: malloc: arena free list can become cyclic, increasing
686 [19049] math: [powerpc] erfc incorrect zero sign
687 [19050] math: [powerpc] log* incorrect zero sign
688 [19058] math: [x86_64] Link fail with -fopenmp and -flto
689 [19059] math: nexttoward overflow incorrect in non-default rounding modes
690 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
691 [19074] network: Data race in _res_hconf_reorder_addrs
692 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
693 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
694 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
696 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
698 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
699 [19086] manual: posix_fallocate64 documented argument order is wrong.
700 [19088] math: lround, llround missing exceptions close to overflow
702 [19094] math: lrint, llrint missing exceptions close to overflow threshold
703 [19095] math: dbl-64 lrint incorrect for 64-bit long
704 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
705 [19124] dynamic-link: ld.so failed to build with older assmebler
706 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
707 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
708 [19134] math: [powerpc32] lround, lroundf spurious exceptions
709 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
710 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
711 pthread_setaffinity_np
712 [19156] math: [ldbl-128] j0l spurious underflows
713 [19164] nptl: tst-getcpu fails with many possible CPUs
714 [19168] math: math/test-ildoubl and math/test-ldouble failure
715 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
716 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
718 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
719 (FE_NOMASK_ENV) do not clear SSE exceptions
720 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
721 _int_new_arena/reused_arena
722 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
723 [19201] math: dbl-64 remainder incorrect sign of zero result
724 [19205] math: bits/math-finite.h conditions do not match math.h and
726 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
727 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
728 for C99-based standards
729 [19212] libc: features.h not -Wundef clean
730 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
732 [19214] libc: Family and model identification for AMD CPU's are incorrect.
733 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
734 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
736 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
738 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
740 [19242] libc: strtol incorrect in Turkish locales
741 [19243] malloc: reused_arena can pick an arena on the free list, leading
742 to an assertion failure and reference count corruption
743 [19253] time: tzset() ineffective when temporary TZ did not include DST
745 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
746 [19270] math: [hppa] Shared libm missing __isnanl
747 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
748 MAP_HUGETLB and MAP_STACK defines
749 [19313] nptl: Wrong __cpu_mask for x32
750 [19347] libc: grantpt: try to force a specific gid even without pt_chown
751 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
752 [19350] math: [ldbl-128ibm] sinhl spurious overflows
753 [19351] math: [ldbl-128ibm] logl inaccurate near 1
754 [19363] time: x32: times() return value wrongly truncates/sign extends
756 [19367] dynamic-link: Improve branch prediction on Silvermont
757 [19369] network: Default domain name not reset by res_ninit when "search"
758 / "domain" entry is removed from resolv.conf
759 [19375] math: powerpc: incorrect results for POWER7 logb with negative
761 [19385] localedata: bg_BG: time separator should be colon, not comma
762 [19408] libc: linux personality syscall wrapper may erroneously return an
763 error on 32-bit architectures
764 [19415] libc: dladdr returns wrong names on hppa
765 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
766 IBM905, IBM907, and IBM909
767 [19439] math: Unix98 isinf and isnan functions conflict with C++11
768 [19443] build: build failures with -DDEBUG
769 [19451] build: Make check fails on test-double-vlen2
770 [19462] libc: Glibc failed to build with -Os
771 [19465] math: Wrong code with -Os
772 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
774 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
776 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
777 [19486] math: S390: Math tests fail with "Exception Inexact set".
778 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
779 [19550] libc: [mips] mmap negative offset handling inconsistent with other
781 [19590] math: Fail to build shared objects that use libmvec.so functions.
785 * The following bugs are resolved with this release:
787 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
788 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
789 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
790 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
791 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
792 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
793 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
794 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
795 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
796 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
797 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
798 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
799 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
800 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
801 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
802 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
803 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
804 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
805 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
806 18657, 18676, 18694, 18696, 18887.
808 * Cache information can be queried via sysconf() function on s390 e.g. with
809 _SC_LEVEL1_ICACHE_SIZE as argument.
811 * A buffer overflow in gethostbyname_r and related functions performing DNS
812 requests has been fixed. If the NSS functions were called with a
813 misaligned buffer, the buffer length change due to pointer alignment was
814 not taken into account. This could result in application crashes or,
815 potentially arbitrary code execution, using crafted, but syntactically
816 valid DNS responses. (CVE-2015-1781)
818 * The time zone file parser has been made more robust against crafted time
819 zone files, avoiding heap buffer overflows related to the processing of
820 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
821 large time zone data files. Overly long time zone specifiers in the TZ
822 variable no longer result in stack overflows and crashes.
824 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
825 for LD and GD on x86 and x86-64, has been implemented. You will need
826 binutils-2.24 or later to enable this optimization.
828 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
829 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
830 Hat). These updates cause user visible changes, such as the fix for bug
833 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
834 and getXXbyYY NSS calls for the same database, causing a denial-of-service
835 condition in some applications.
837 * Added vector math library named libmvec with the following vectorized x86_64
838 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
840 The library can be disabled with --disable-mathvec. Use of the functions is
841 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
842 Shared library libmvec.so is linked in as needed when using -lm (no need to
843 specify -lmvec explicitly for not static builds).
844 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
846 * A new fmemopen implementation has been added with the goal of POSIX
847 compliance. The new implementation fixes the following long-standing
848 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
849 old implementation is still present for use be by existing binaries.
851 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
852 releases. It has been fixed to match 2.19 and older, but binaries built
853 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
855 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
856 Contributed by Roland McGrath (Google).
858 * The header <regexp.h> is deprecated, and will be removed in a future
859 release. Use of this header will trigger a deprecation warning.
860 Application developers should update their code to use <regex.h> instead.
862 This header was formerly part of SUS, but was deprecated in 1994 and
863 removed from the standard in 2001. Also, the glibc implementation
864 leaks memory. See BZ#18681 for more details.
868 * The following bugs are resolved with this release:
870 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
871 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
872 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
873 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
874 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
875 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
876 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
877 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
878 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
879 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
882 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
883 too little memory for the to-be-scanned arguments and overflow the
884 allocated buffer. The implementation now correctly computes the required
885 buffer size when using malloc, and switches to malloc from alloca as
888 * A new semaphore algorithm has been implemented in generic C code for all
889 machines. Previous custom assembly implementations of semaphore were
890 difficult to reason about or ensure that they were safe. The new version
891 of semaphore supports machines with 64-bit or 32-bit atomic operations.
892 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
893 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
895 * Port to Altera Nios II has been contributed by Mentor Graphics.
897 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
898 implementations for powerpc64/powerpc64le.
899 Implemented by Adhemerval Zanella (IBM).
901 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
902 and powerpc64le. This may improve lock scaling of existing programs on
903 HTM capable systems. The lock elision code is only enabled with
904 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
905 powerpc will issue a transaction abort on every syscall to avoid side
906 effects being visible outside transactions.
908 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
909 AArch64. Contributed by ARM Ltd.
911 * i386 memcpy functions optimized with SSE2 unaligned load/store.
913 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
914 under certain input conditions resulting in the execution of a shell for
915 command substitution when the application did not request it. The
916 implementation now checks WRDE_NOCMD immediately before executing the
917 shell and returns the error WRDE_CMDSUB as expected.
919 * CVE-2012-3406 printf-style functions could run into a stack overflow when
920 processing format strings with a large number of format specifiers.
922 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
923 infinite loop if the DNS response contained a PTR record of an unexpected
926 * The minimum GCC version that can be used to build this version of the GNU
927 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
928 still be used to compile programs using the GNU C Library.
930 * The GNU C Library is now built with -Werror by default. This can be
931 disabled by configuring with --disable-werror.
933 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
935 * The obsolete sigvec function has been removed. This was the original
936 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
937 programs have been using instead for about 25 years. Of course, ABI
938 compatibility for old binaries using sigvec remains intact.
940 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
941 with newer versions of bison.
943 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
944 The original MIPS o32 hard-float ABI requires an FPU where double-precision
945 registers overlay two consecutive single-precision registers. MIPS32R2
946 introduced a new FPU mode (FR=1) where double-precision registers extend the
947 corresponding single-precision registers which is incompatible with the
948 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
949 require the use of FR=1 making a transition necessary. New o32 ABI
950 extensions enable users to migrate over time from the original o32 ABI
951 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
952 tracks the ABI of any loaded object and verifies that new objects are
953 compatible. Mode transitions will also be requested as required and
954 unsupportable objects will be rejected. The ABI checks include both soft and
955 hard float ABIs for o32, n32 and n64.
957 GCC 5 with GNU binutils 2.25 onwards:
958 It is strongly recommended that all o32 system libraries are built using the
959 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
960 with the original and all new o32 ABI extensions. Configure a MIPS GCC
961 compiler using --with-fp-32=xx to set this by default.
965 * The following bugs are resolved with this release:
967 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
968 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
969 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
970 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
971 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
972 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
973 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
974 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
975 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
976 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
977 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
978 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
979 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
980 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
981 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
983 * Reverted change of ABI data structures for s390 and s390x:
984 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
985 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
986 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
987 are preserved pointing straight to the same implementation as the old ones.
988 Given that, new callers will simply provide a too-big buffer to these
989 functions. Any applications/libraries out there that embed jmp_buf or
990 ucontext_t in an ABI-relevant data structure that have already been rebuilt
991 against 2.19 headers will have to rebuilt again. This is necessary in any
992 case to revert the breakage in their ABI caused by the glibc change.
994 * Support for file description locks is added to systems running the
995 Linux kernel. The standard file locking interfaces are extended to
996 operate on file descriptions, not file descriptors, via the use of
997 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
998 are associated with an open file instead of a process.
1000 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1002 * The minimum Linux kernel version that this version of the GNU C Library
1003 can be used with is 2.6.32.
1005 * Running the testsuite no longer terminates as soon as a test fails.
1006 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1007 with PASS or FAIL lines for individual tests. A summary of the results is
1008 printed, including a list of failing lists, and "make check" exits with
1009 error status if there were any unexpected failures. "make check
1010 stop-on-test-failure=y" may be used to keep the old behavior.
1012 * The am33 port, which had not worked for several years, has been removed
1015 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1016 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1017 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1018 conflicted with POSIX (support for which was removed in 2.19), the
1019 interfaces those macros enabled remain available when compiling with
1020 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1021 test macros defined.
1023 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1025 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1026 This may improve lock scaling of existing programs on TX capable systems.
1027 The lock elision code is only built with --enable-lock-elision=yes and
1028 then requires a GCC version supporting the TX builtins. With lock elision
1029 default mutexes are elided via __builtin_tbegin, if the cpu supports
1030 transactions. By default lock elision is not enabled and the elision code
1033 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1034 copy the path argument. This allowed programs to cause posix_spawn to
1035 deference a dangling pointer, or use an unexpected pathname argument if
1036 the string was modified after the posix_spawn_file_actions_addopen
1039 * All supported architectures now use the main glibc sysdeps directory
1040 instead of some being in a separate "ports" directory (which was
1041 distributed separately before glibc 2.17).
1043 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1044 On configurations that support it (all Linux configurations), it's now
1045 used regardless of the --enable-add-ons switch to configure. It is no
1046 longer possible to build such configurations without pthreads support.
1048 * Locale names, including those obtained from environment variables (LANG
1049 and the LC_* variables), are more tightly checked for proper syntax.
1050 setlocale will now fail (with EINVAL) for locale names that are overly
1051 long, contain slashes without starting with a slash, or contain ".." path
1052 components. (CVE-2014-0475) Previously, some valid locale names were
1053 silently replaced with the "C" locale when running in AT_SECURE mode
1054 (e.g., in a SUID program). This is no longer necessary because of the
1057 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1058 with application code using Intel MPX instructions. (With all previous
1059 versions, the MPX register state could be clobbered when making calls
1060 into or out of a shared library.) Note that while the new dynamic
1061 linker is compatible with all known x86 hardware whether or not it
1062 supports Intel MPX, some x86 instruction-set emulators might fail to
1063 handle the new instruction encodings. This is known to affect Valgrind
1064 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1065 release), and might affect other tools that do instruction emulation.
1067 * Support for loadable gconv transliteration modules has been removed.
1068 The support for transliteration modules has been non-functional for
1069 over a decade, and the removal is prompted by security defects. The
1070 normal gconv conversion modules are still supported. Transliteration
1071 with //TRANSLIT is still possible, and the //IGNORE specifier
1072 continues to be supported. (CVE-2014-5119)
1074 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1075 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1076 resulting a denial-of-service security vulnerability in applications which
1077 use functions related to iconv. (CVE-2014-6040)
1081 * The following bugs are resolved with this release:
1083 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1084 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1085 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1086 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1087 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1088 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1089 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1090 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1091 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1092 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1093 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1094 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1095 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1096 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1097 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1098 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1099 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1100 16431, 16453, 16474, 16506, 16510, 16529
1102 * Slovenian translations for glibc messages have been contributed by the
1103 Translation Project's Slovenian team of translators.
1105 * The public headers no longer use __unused nor __block. This change is to
1106 support compiling programs that are derived from BSD sources and use
1107 __unused internally, and to support compiling with Clang's -fblock
1108 extension which uses __block.
1110 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1111 large collation sequences to optimize multiple passes. This cache
1112 computation may overflow for large collation sequences and may cause a
1113 stack or buffer overflow. This is now fixed to use a slower algorithm
1114 which does not use a cache if there is an integer overflow.
1116 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1117 rules for large collation sequences to optimize multiple passes and falls
1118 back to alloca if malloc fails, resulting in a possible stack overflow.
1119 The implementation now falls back to an uncached collation sequence lookup
1122 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1123 initialized for static applications resulting in the security feature
1124 being disabled. The pointer guard is now correctly initialized to a
1125 random value for static applications. Existing static applications need
1126 to be recompiled to take advantage of the fix (bug 15754).
1128 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1129 to the d_name member of struct dirent, or omit the terminating NUL
1130 character. (Bugzilla #14699).
1132 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1133 aligned_alloc functions could allocate too few bytes or corrupt the
1134 heap when passed very large allocation size values (Bugzilla #15855,
1137 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1138 for AF_INET6 has been fixed (Bugzilla #16072).
1140 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1141 pap_CW, quz_PE, the_NP.
1143 * Substantially revised locales: gd_GB, ht_HT
1145 * The LC_ADDRESS field was updated to support country_car for almost all
1148 * ISO 1427 definitions were updated.
1150 * ISO 3166 definitions were updated.
1152 * The localedef utility now supports --big-endian and --little-endian
1153 command-line options to generate locales for a different system from that
1154 for which the C library was built.
1156 * Binary locale files now only depend on the endianness of the system for
1157 which they are generated and not on other properties of that system. As a
1158 consequence, binary files generated with new localedef may be incompatible
1159 with old versions of the GNU C Library, and binary files generated with
1160 old localedef may be incompatible with this version of the GNU C Library,
1161 in the following circumstances:
1163 + Locale files may be incompatible on m68k systems.
1165 + Locale archive files (but not separate files for individual locales) may
1166 be incompatible on systems where plain "char" is signed.
1168 * The configure option --disable-versioning has been removed. Builds with
1169 --disable-versioning had not worked for several years.
1171 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1172 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1174 * SystemTap probes for malloc have been introduced.
1176 * SystemTap probes for slow multiple precision fallback paths of
1177 transcendental functions have been introduced.
1179 * Support for powerpc64le has been added.
1181 * The soft-float powerpc port now supports e500 processors.
1183 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1185 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1186 set of header declarations that are enabled by default, even when other
1187 feature test macros or compiler options such as -std=c99 would otherwise
1188 disable some of those declarations.
1190 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1191 conflict with POSIX. The libbsd-compat library (which was a dummy library
1192 that did nothing) has also been removed.
1194 * Preliminary documentation about Multi-Thread, Async-Signal and
1195 Async-Cancel Safety has been added.
1197 * Change of ABI data structures for s390 and s390x:
1198 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1199 allow for future hardware extensions. All library functions that accept or
1200 return these structures were versioned in order to provide backwards
1201 compatibility with existing code. However, not all cases can be handled
1202 with interface versioning. If an application embeds either structure into
1203 another structure and that structure is passed to another compilation unit
1204 which expects the newer larger-sized object then any access to the new
1205 fields will result in undefined behaviour. Similarly any access to fields
1206 that were moved by the enlarging of the structures will cause undefined
1207 behaviour. To fix the undefined behaviour all objects that are part of the
1208 final in-memory image must agree on the size of structures, and this may
1209 require recompilation.
1213 * The following bugs are resolved with this release:
1215 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1216 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1217 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1218 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1219 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1220 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1221 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1222 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1223 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1224 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1225 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1226 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1227 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1230 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1231 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1232 Distributions can re-enable building and using pt_chown via the new configure
1233 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1234 considerable security risks and should only be used if the distribution
1235 understands and accepts the risks.
1237 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1240 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1241 fixed (Bugzilla #15330).
1243 * Add support for calling C++11 thread_local object destructors on thread
1244 and program exit. This needs compiler support for offloading C++11
1245 destructor calls to glibc.
1247 * Improved worst case performance of libm functions with double inputs and
1250 * Support for priority inherited mutexes in pthread condition variables on
1251 non-x86 architectures.
1253 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1255 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1257 * Optimized string functions for ARM. Implemented by Will Newton and
1260 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1262 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1265 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1266 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1268 * Added a benchmark framework to track performance of functions in glibc.
1270 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1271 It is based on draft TS 18661 and currently enabled as a GNU extension.
1273 * On Linux, the clock function now uses the clock_gettime system call
1274 for improved precision, rather than old times system call.
1276 * Added support for version-3 tz format files. This is needed when using
1277 the tz database release 2013e or later, and affects a few unusual cases --
1278 currently only TZ='America/Godthab' for time stamps after 2037.
1280 * Added new API functions pthread_getattr_default_np and
1281 pthread_setattr_default_np to get and set the default pthread
1282 attributes of a process.
1284 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1285 This may improve lock scaling of existing programs on TSX capable systems.
1286 When the --enable-lock-elision=yes parameter is specified at configure
1287 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1290 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1291 availability). Implemented by Ryan S. Arnold.
1293 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1297 * The following bugs are resolved with this release:
1299 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1300 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1301 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1302 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1303 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1304 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1305 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1306 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1307 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1308 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1309 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1310 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1311 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1312 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1313 14879, 14889, 14893, 14898, 14914.
1315 * Optimization of memcpy for MIPS.
1317 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1318 EMFILE has been fixed (Bugzilla #14889).
1320 * The add-on ports collection is now distributed in the "ports" subdirectory
1321 of the main GNU C Library distribution, rather than separately.
1323 * Port to ARM AArch64 contributed by Linaro.
1325 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1326 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1328 Implemented by Andreas Krebbel.
1330 * The new function secure_getenv allows secure access to the environment,
1331 returning NULL if running in a SUID/SGID process. This function replaces
1332 the internal function __secure_getenv.
1334 * SystemTap static probes have been added into the dynamic linker.
1335 Implemented by Gary Benson.
1337 * Optimizations of string functions strstr, strcasestr and memmem.
1338 Implemented by Maxim Kuvyrkov.
1340 * The minimum Linux kernel version that this version of the GNU C Library
1341 can be used with is 2.6.16.
1343 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1344 powerpc POWER7. Implemented by Will Schmidt.
1346 * New configure option --disable-nscd builds the C library such that it
1347 never attempts to contact the Name Service Caching Daemon (nscd).
1348 New configure option --disable-build-nscd avoids building nscd itself;
1349 this is the default if --disable-nscd is used.
1351 * Improved support for cross-compilation, including cross-testing and
1352 bootstrap builds without a previously built glibc.
1354 * Several testsuite tests are now able to test multiple IFUNC variants of an
1355 interface, rather than just testing the one that would be chooen by
1358 * New configure options --with-bugurl and --with-pkgversion, for
1359 distributors to use to embed their bug-reporting and package version
1360 information in --help and --version output.
1362 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1363 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1364 allows creation of chroots without the procfs mounted on /proc.
1366 * The `crypt' function now fails if passed salt bytes that violate the
1367 specification for those values. On Linux, the `crypt' function will
1368 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1369 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1370 when the mode is enabled.
1372 * The `clock_*' suite of functions (declared in <time.h>) is now available
1373 directly in the main C library. Previously it was necessary to link with
1374 -lrt to use these functions. This change has the effect that a
1375 single-threaded program that uses a function such as `clock_gettime' (and
1376 is not linked with -lrt) will no longer implicitly load the pthreads
1377 library at runtime and so will not suffer the overheads associated with
1378 multi-thread support in other code such as the C++ runtime library.
1380 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1385 * The following bugs are resolved with this release:
1387 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1388 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1389 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1390 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1391 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1392 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1393 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1394 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1395 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1396 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1397 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1398 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1399 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1400 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1401 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1402 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1403 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1404 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1405 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1406 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1407 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1408 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1411 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1412 configuring glibc with:
1413 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1414 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1415 Implemented by H.J. Lu.
1419 + define static_assert
1421 + do not declare gets
1423 + declare at_quick_exit and quick_exit also for ISO C11
1425 + aligned_alloc. NB: The code is deliberately allows the size parameter
1426 to not be a multiple of the alignment. This is a moronic requirement
1427 in the standard but it is only a requirement on the caller, not the
1430 + timespec_get added
1432 + uchar.h support added
1434 + CMPLX, CMPLXF, CMPLXL added
1436 Implemented by Ulrich Drepper.
1438 * Support for the IA-64 has been moved to ports.
1440 * Remove support for anything but ELF binary format
1442 * Checking versions of poll, ppoll added.
1443 Implemented by Ulrich Drepper.
1445 * More generic and 64-bit performance optimizations to math functions.
1446 Implemented by Ulrich Drepper.
1448 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1449 headers and functions available at compile time as they were before
1450 version 2.14. This option will be removed at some time in the future
1451 after the TI-RPC library becomes fully sufficient for the needs of
1452 existing applications.
1454 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1455 Note that glibc is not expected to work with any Linux kernel version
1458 * New header <sys/auxv.h> and function getauxval allowing easy access to
1459 the AT_* key-value pairs passed from the Linux kernel. The header also
1460 defines the HWCAP_* bits associated with the AT_HWCAP key.
1462 * New locales: mag_IN
1464 * New configure option --enable-systemtap builds SystemTap static probes
1465 into libc for setjmp and longjmp and into libpthread for various operations.
1466 So far the setjmp/longjmp probes and some of the libpthread probes are
1467 provided only for i*86 and x86_64.
1468 Implemented by Roland McGrath and Rayson Ho.
1470 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1472 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1475 * More optimized functions for SPARC. Implemented by David S. Miller.
1477 * Improved support for cross-compilation, especially bootstrap builds
1478 without a previously built glibc.
1480 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1481 Chris Metcalf from Tilera.
1483 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1484 now supported for ARM processors.
1486 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1487 as the name of the dynamic linker, to distinguish it from the
1488 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1490 * Support for CRIS has been removed from ports.
1492 * A new class of installed header has been documented for low-level
1493 platform-specific functionality. PowerPC added the first instance with a
1494 function to provide time base register access. Contributed by Tulio
1495 Magno Quites Machado Filho.
1497 * ix86 configurations now install header files that are consistent with
1498 what x86-64 configurations install. These same header files can be used
1499 for -m32, -m64, or -mx32 builds.
1500 Contributed by H.J. Lu.
1502 * Math library bug fixes. A thorough audit of all open math library bugs was
1503 conducted by Joseph Myers. Significant progress was made on many math
1504 library bugs resulting in more accurate exceptions and function results.
1505 Many thanks to all those that contributed including Andreas Jaeger for his
1506 patch review and work on the x87 trigonometric instruction issues.
1508 * Timezone data is no longer installed. Timezone-related binaries and scripts
1509 will continue to be installed. Users should obtain their timezone data from
1510 their distribution provider or from the tzdata package at
1511 <ftp://munnari.oz.au/pub/>.
1515 * The following bugs are resolved with this release:
1517 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1518 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1519 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1520 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1521 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1522 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1523 13484, 13506, 13515, 13523, 13524, 13538, 13540
1525 * New program pldd to list loaded object of a process
1526 Implemented by Ulrich Drepper.
1528 * Add nss_db support back to glibc. No more dependency on Berkeley db
1529 and support for initgroups lookups.
1530 Implemented by Ulrich Drepper.
1532 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1533 Contributed by HJ Lu.
1535 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1536 Contributed by HJ Lu.
1538 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1539 on x86-32 and x86-64.
1540 Contributed by Liubov Dmitrieva.
1542 * Optimized strchr and strrchr for SSE on x86-32.
1543 Contributed by Liubov Dmitrieva.
1545 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1546 for x86-64 and x86-32.
1547 Contributed by Liubov Dmitrieva.
1549 * New interfaces: scandirat, scandirat64
1550 Implemented by Ulrich Drepper.
1552 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1553 Implemented by Ulrich Drepper.
1555 * nscd now also caches the netgroup database.
1556 Implemented by Ulrich Drepper.
1558 * Integrate libm with gcc's -ffinite-math-only option.
1559 Implemented by Ulrich Drepper.
1561 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1562 to math functions. Implemented by Ulrich Drepper.
1564 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1565 Implemented by Ulrich Drepper.
1567 * New Linux interfaces: process_vm_readv, process_vm_writev
1569 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1570 Implemented by Ulrich Drepper.
1572 * Optimized nearbyint and strcasecmp for PPC.
1573 Implemented by Adhemerval Zanella.
1575 * New locales: bho_IN, unm_US, es_CU, ta_LK
1579 * The following bugs are resolved with this release:
1581 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1582 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1583 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1584 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1585 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1586 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1587 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1588 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1589 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1590 12795, 12811, 12813, 12814, 12841
1592 * The RPC implementation in libc is obsoleted. Old programs keep working
1593 but new programs cannot be linked with the routines in libc anymore.
1594 Programs in need of RPC functionality must be linked against TI-RPC.
1595 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1597 Visible changes of this change include (obviously) the inability to link
1598 programs using RPC functions without referencing the TI-RPC library and the
1599 removal of the RPC headers from the glibc headers.
1600 Implemented by Ulrich Drepper.
1602 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1603 syncfs, setns, sendmmsg
1605 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1606 yue_HK, lij_IT, mhr_RU
1608 * New encodings: CP770, CP771, CP772, CP773, CP774
1610 * New program sotruss to trace calls through PLTs
1611 Implemented by Ulrich Drepper.
1613 * The malloc hook implementation is marked deprecated and will be removed
1614 from the default implementation in the next version. The design never
1615 worked ever since the introduction of threads. Even programs which do
1616 not create threads themselves can use multiple threads created internally.
1620 * The following bugs are resolved with this release:
1622 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1623 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1624 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1625 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1628 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1630 * POWER7 optimizations: memset, memcmp, strncmp
1632 * New optimized string functions for x86-64: strnlen (SSE2),
1633 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1634 Implemented by Ulrich Drepper.
1638 * The following bugs are resolved with this release:
1640 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1641 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1642 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1643 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1644 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1645 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1646 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1647 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1649 * New interfaces: pthread_getname_np, pthread_setname_np
1651 * New Linux interface: recvmmsg
1653 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1655 * The dynamic linker now recognizes supported ABI versions from the
1656 EI_ABIVERSION field in the ELF header.
1657 Implemented by Ulrich Drepper.
1659 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1660 will not be used to fill in password fields in the passwd.byname replies.
1661 Instead it is used to synthesize the shadow.byname table, should it be
1662 missing. This is a useful mode in some installations involving Solaris.
1663 Implemented by Ulrich Drepper.
1665 * New locales: kok_IN, sq_MK, cv_RU
1669 * The following bugs are resolved with this release:
1671 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1672 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1673 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1674 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1675 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1676 10789, 10817, 10823, 10840, 10847
1678 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1680 Implemented by Ulrich Drepper.
1682 * Checking version of longjmp added that fails if an uninitialized stack
1683 frame would be created. Implemented by Ulrich Drepper.
1685 * STT_GNU_IFUNC is now supported in static executables.
1686 Implemented by H.J. Lu.
1688 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1690 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1691 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1692 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1693 Contributed by H.J. Lu.
1695 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1696 Implemented by Ulrich Drepper.
1698 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1700 Contributed by H.J. Lu.
1702 * Support for fma instruction in AVX on x86-64.
1703 Implemented by H.J. Lu and Ulrich Drepper.
1705 * AVX support in x86-64 auditing support in ld.so.
1706 Implemented by H.J. Lu.
1708 * STB_GNU_UNIQUE support added.
1709 Implemented by Ulrich Drepper.
1711 * Implement second fallback mode for optimized DNS lookup for even more
1712 broken environments. If two requests from the same port are not
1713 handled correctly close the socket and open a new one before sending
1714 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1715 can be used to select this mode right away, instead of rediscovering the
1716 necessity is every process again.
1717 Implemented by Ulrich Drepper.
1719 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1720 Implemented by Adam Tkac.
1722 * Optimized iconv conversions for S390x.
1723 Implemented by Andreas Krebbel.
1725 * Using condvars with PI mutexes is now more efficient due to kernel
1726 support for requeueing to PI futexes. NPTL support added for x86-64.
1727 Implemented by Ulrich Drepper.
1729 * New locales: ps_AF, my_MM
1733 * The following bugs are resolved with this release:
1735 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1736 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1737 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1738 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1739 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1741 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1742 Implemented by Ulrich Drepper.
1744 * New Linux interfaces: accept4, fallocate, fallocate64.
1745 Implemented by Ulrich Drepper.
1747 * Correct declarations of string function when used in C++ code. This
1748 could lead to compile errors for invalid C++ code.
1750 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1753 * New POSIX 2008 interface: psiginfo
1754 Implemented by Ulrich Drepper.
1756 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1757 Implemented by Ulrich Drepper.
1759 * Support for selecting between multiple function definitions at runtime
1760 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1762 * The libcrypt library can now use the hash function implementations in
1763 NSS. Implemented by Ulrich Drepper.
1765 * The malloc implementation can be compiled to be less memory efficient
1766 but higher performing in multi-threaded programs.
1767 Implemented by Ulrich Drepper.
1769 * New locales: nan_TW@latin, ks_IN
1771 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1772 Implemented by Ulrich Drepper.
1774 * Extended printf hook support. It is possible to use user-defined types
1775 and extend existing format specifiers.
1776 Implemented by Ulrich Drepper.
1778 * Handling for group shadow files has been added.
1779 Implemented by Ulrich Drepper.
1781 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1782 servers (the case, e.g., for some people using the built-in DNS
1783 server in ADSL modems/routers). There is a once-per-process timeout
1784 in case of a broken server. To avoid it, users can run nscd or put
1785 'options single-request' in /etc/resolv.conf.
1786 Implemented by Ulrich Drepper.
1790 * The following bugs are resolved with this release:
1792 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1793 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1794 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1795 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1796 6968, 6974, 6980, 6995, 7008, 7009, 7029
1798 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1799 up at the same time. Implemented by Ulrich Drepper.
1801 * TLS descriptors for LD and GD on x86 and x86-64.
1802 Implemented by Alexandre Oliva.
1804 * getaddrinfo now handles DCCP and UDPlite.
1805 Implemented by Ulrich Drepper.
1807 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1808 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1809 Implemented by Ulrich Drepper.
1811 * New implementation of memmem, strstr, and strcasestr which is O(n).
1812 Implemented by Eric Blake.
1814 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1816 * Implement "e" option for popen to open file descriptor with the
1817 close-on-exec flag set. Implemented by Ulrich Drepper.
1819 * Implement "b" mode for fmemopen. In this mode writes writes don't
1820 implicitly add a NUL byte and seeks from the end of the buffer really
1821 use the buffer end, not the string length as the basis.
1822 Implemented by Ulrich Drepper.
1824 * Many functions, exported and internal, now atomically set the close-on-exec
1825 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1827 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1829 Implemented by Pravin Satpute.
1831 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1835 * The following bugs are resolved with this release:
1837 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1838 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1839 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1840 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1841 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1842 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1843 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1845 * New locales: bo_CN, bo_IN, shs_CA.
1847 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1849 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1850 Implemented by Pravin Satpute.
1852 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1854 * nscd caches now all timeouts for DNS entries
1855 Implemented by Ulrich Drepper.
1857 * nscd is more efficient and wakes up less often.
1858 Implemented by Ulrich Drepper.
1860 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1861 vdprintf, and obstack_vprintf.
1862 Implemented by Jakub Jelinek.
1864 * Faster memset for x86-64.
1865 Implemented by Harsha Jagasia and H.J. Lu.
1867 * Faster memcpy on x86.
1868 Implemented by Ulrich Drepper.
1870 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1871 Implemented by Ulrich Drepper.
1873 * Faster sqrt and sqrtf implemention for some PPC variants.
1874 Implemented by Steven Munroe.
1878 * The following bugs are resolved with this release:
1880 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1881 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1882 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1883 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1884 5071, 5103, 5104, 5112, 5113, 5184, 5186
1886 * More checking functions: fread, fread_unlocked, open*, mq_open.
1887 Implemented by Jakub Jelinek and Ulrich Drepper.
1889 * Extend fortification to C++. Implemented by Jakub Jelinek.
1891 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1892 by not recognizing 'a' as a modifier when those specs are requested.
1893 Implemented by Jakub Jelinek.
1895 * PPC optimizations to math and string functions.
1896 Implemented by Steven Munroe.
1898 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1899 options to be passed. Implemented by Ulrich Drepper.
1901 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1903 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1904 Implemented by Ulrich Drepper.
1906 * Handle private futexes in the NPTL implementation.
1907 Implemented by Jakub Jelinek and Ulrich Drepper.
1909 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
1910 Implemented by Roland McGrath and Ulrich Drepper.
1912 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
1914 * SHA-256 and SHA-512 based password encryption.
1915 Implemented by Ulrich Drepper.
1917 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
1918 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
1921 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
1922 Implemented by Ulrich Drepper.
1926 * The following bugs are resolved with this release:
1928 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
1929 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
1930 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
1931 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
1932 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
1933 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
1934 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
1937 * New Linux interfaces: epoll_pwait, sched_getcpu.
1939 * New generic interfaces: strerror_l.
1941 * nscd can now cache the services database. Implemented by Ulrich Drepper.
1945 * The following bugs are resolved with this release:
1947 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
1948 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
1949 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
1950 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
1951 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
1952 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
1953 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
1954 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
1955 3225, 3251, 3252, 3253, 3273
1957 * For Linux, the sorting of addresses returned by getaddrinfo now also
1958 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
1959 Implemented by Ulrich Drepper.
1961 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
1962 Implemented by Ulrich Drepper.
1964 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
1966 * New iconv module for MIK. Contributed by Alexander Shopov.
1968 * For sites with broken group and/or passwd database, the auto-propagate
1969 option of nscd can prevent creating ID lookup entries from the results
1970 of a name lookup and vice versa. This usually is no problem but some
1971 site might have problems with the default behavior.
1972 Implemented by Ulrich Drepper.
1974 * Iterating over entire database in NIS can be slow. With the
1975 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
1976 to trade time for memory. The entire database will be read at once.
1977 Implemented by Ulrich Drepper.
1979 * The interfaces introduced in RFC 3542 have been implemented by
1982 * Support for the new ELF hash table format was added by Ulrich Drepper.
1984 * Support for priority inheritance mutexes added by Jakub Jelinek and
1987 * Support for priority protected mutexes added by Jakub Jelinek.
1989 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
1990 or_IN, csb_PL, fy_NL, sr_ME.
1994 * More overflow detection functions.
1996 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
1997 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
1999 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2000 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2001 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2002 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2003 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2004 by Masahide Washizawa.
2006 * It is now possible to install audit libraries for the dynamic linker using
2007 LD_AUDIT. Implemented by Ulrich Drepper.
2009 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2010 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2011 For a libc and libpthread that works well on Linux 2.4 kernels, we
2012 recommend using the stable 2.3 branch.
2014 * The new function open_wmemstream acts like open_memstream,
2015 but uses a wchar_t wide character buffer.
2017 * The new function ppoll is an improved version of poll, similar to pselect.
2019 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2020 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2021 renameat, unlinkat, linkat, symlinkat, readlinkat.
2023 * New Linux kernel system calls: unshare,
2024 inotify_init, inotify_add_watch, inotify_rm_watch.
2026 * The euidaccess function is now also known by the alias eaccess,
2027 for compatibility with some other systems.
2029 * Timezone data updated to 2006b version.
2033 * The following bugs are resolved with this release:
2035 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2036 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2037 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2038 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2039 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2040 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2042 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2044 * As of this release, GCC 4 can be used to compile the C Library.
2046 * Timezone data updated to 2005m version.
2050 * The following bugs are resolved with this release:
2052 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2053 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2054 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2055 777, 787, 821, 822, 823, 825
2057 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2061 * Support for RFC 3678. Real implementations exist only for Linux so far.
2062 Implemented by Ulrich Drepper.
2064 * nscd can now cache entries persistently. Expiring entries are reloaded.
2065 For speedups the cache can be shared in memory with client processes.
2066 Implemented by Ulrich Drepper.
2068 * nscd can now perform SELinux checks.
2069 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2071 * getaddrinfo queries are now cached. Canonical name lookup is performed
2073 Implemented by Ulrich Drepper.
2075 * The nothrow function attribute is used when headers are used by gcc when
2076 compiling C code. This can avoid significant amounts of exception
2079 * The malloc functions perform more error checking and are stricter when
2080 it comes to reacting on errors. The default action is to terminate
2081 the process after showing an error message. Implemented by Ulrich Drepper.
2083 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2084 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2086 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2087 namespaces using the new function dlmopen(). This feature is of course,
2088 like most other dynamic loading functionality, not available in statically
2089 linked applications. Implemented by Ulrich Drepper.
2091 * Low-overhead boundary checking variants of string and some stdio functions
2092 were added. These are to be used in conjunction with a gcc patch by
2093 Jakub Jelinek which adds calls to these functions if possible.
2094 Implemented by Jakub Jelinek and Ulrich Drepper.
2096 * Old code for several operating systems and machine architectures that
2097 have not been in working condition in a long time have been removed from
2098 the main source tree maintained by the GNU C Library's maintainers.
2099 These files are now reside in the separate `ports' source module
2100 that is usable as an add-on when building the library.
2104 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2105 interrogate the dynamic linker, compatible with the Solaris interface.
2107 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2108 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2110 * getifaddrs now uses the netlink interface on Linux to get its information.
2111 Implemented by Thorsten Kukuk.
2113 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2114 Implemented by Ulrich Drepper.
2116 * support for non-executable stacks on x86 has been added. Changes mostly
2119 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2122 * getaddrinfo now performs destination address selection according to
2127 * Thread-safe interfaces for many functions that access locale data
2128 were added in version 2.3, but these features were omitted from NEWS.
2129 Many functions have variants with an `_l' suffix that take a `locale_t'
2130 object as a parameter rather than consulting the current locale.
2131 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2132 create and maintain `locale_t' objects. Additionally, the new function
2133 `uselocale' sets "the current locale" (as used by functions not so
2134 parameterized) set for an individual thread. These features were added
2135 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2137 * The functions getresuid, getresgid, setresuid, and setresgid, which
2138 have long been available on Linux, are now declared in <unistd.h>
2139 and are now also available on the Hurd.
2141 * ELF thread-local storage support (TLS) now works on x86-64.
2143 * The new dynamic string token $LIB is expanded in shared library names.
2144 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2146 * Aldy Hernandez contributed complete software floating point support for
2147 PowerPC machines with no FPU.
2149 * fexecve is implemented on Linux.
2151 * The `btowc' function should work at least twice as fast due to
2152 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2154 * With approriate thread add-ons cancelable functions are now implemented
2155 in libc.so as well. No need to call the function in libpthread. This
2156 change allowed to finally disable the incorrect and expensive handling
2157 of weak definition in ld.so.
2159 * Yet more PLT entries in libc.so have been removed. We finally arrived
2160 at the bare minimum. Startup times improved appropriately.
2162 * Support for the new Linux/x86 system call interface was added. The
2163 AT_SYSINFO auxiliary vector entry is recognized and handled.
2167 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2170 * iconv (the program and the interface) now accepts empty names (excluding
2171 options like //TRANSLIT) to mean "use charset of current locale".
2173 * localedef can now transliterate characters in strings which are not in
2174 the provided charmap. The information from the input locale is used.
2176 * Prelinking support was added for ELF targets. This requires additional
2177 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2179 * Read-only stdio streams now use mmap to speed up operation by eliminating
2180 copying and buffer underflows. To use add 'm' to the mode string of
2181 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2183 * The malloc functions were completely rewritten by Wolfram Gloger based
2184 on Doug Lea's malloc-2.7.0.c.
2186 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2187 implementation of regex.
2189 * Bruno Haible upgraded the iconv and locale implementation to support
2192 * Contents of the LC_* and LANG environment variables in the CEN style are
2193 not recognized anymore. It never was used. Change by Ulrich Drepper.
2195 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2196 thread-local storage (TLS) ABI on some platforms.
2197 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2199 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2200 EUC-JISX0213, and TSCII.
2202 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2203 BSD-compatible interface for getting all network interface addresses.
2204 Implementation for IPv4 by Roland McGrath.
2206 * Loading of locale data is faster due to the introduction of a locale
2207 archive. Implemented by Roland McGrath and Ulrich Drepper.
2209 * Startup times are significantly reduced by not using exported functions
2210 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2213 * Steven Munroe contributed a port to PowerPC64/Linux.
2217 * The Hurd now uses the GNU libio implementation of stdio rather than the
2218 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2220 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2221 with the same behavior as the Linux system call of the same name.
2225 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2226 128-bit long double format.
2228 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2229 IBM1161, and IBM1162 charsets.
2231 * Andreas Jaeger contributed a port to x86-64/Linux.
2233 * Peter Bruin contributed a port to PowerPC/Hurd.
2235 * libc functions using I/O streams now can handle wide-oriented streams
2238 * optimizations in the dynamic linker. Binaries created by recent binutils
2239 versions start up quicker due to reduced time spent on relocations.
2241 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2245 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2246 asinh, atanh, j0 for the 128-bit long double format.
2248 * Bruno Haible updated all the code handling Unicode in some form to
2249 support Unicode 3.1.
2251 * Speed of regex for single-byte locales is back to previous levels.
2252 Patch by Isamu Hasegawa.
2254 * Alpha, SPARC, and IA-64 now also using floating stacks.
2256 * Startup time of internationalized applications greatly improved through
2257 iconv cache. Use iconvconfig to generate the cache file.
2258 Contributed by Ulrich Drepper.
2260 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2261 severe performance problems. Patch by David Mosberger.
2263 * The Hurd port got a lot more functionality like AIO, various stdio
2264 extensions, etc. Mainly done by Roland McGrath.
2266 * mtrace can now lookup symbols in shared libraries.
2270 * Intel's IA-64 math library is largely integrated. It provides fast and
2271 accurate implementatations for most basic and standard math functions
2272 in float, double, and long double format.
2274 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2275 and asin for the 96-bit long double format and asin, log, tan for the
2276 128-bit long double format.
2278 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2279 The basic double functions exist today. Contributed by Abraham Ziv
2280 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2281 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2283 * An asynchronous name lookup library was added. The interface is designed
2284 after POSIX AIO. The proposal was circulated beforehand to get comments.
2285 No negative ones came in. Implemented by Ulrich Drepper.
2287 * Port to S390/64bit contributed by Martin Schwidefsky
2288 <schwidefsky@de.ibm.com>.
2290 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2291 of functions for Linux/IA-64.
2293 * The RPC code is now thread safe. Threads can now use the same service
2294 of different services at the same time. Patch by Eric Norum
2295 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2297 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2298 family of functions for Linux/S390.
2300 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2301 of functions for Linux/x86.
2303 * Port to Linux/CRIS contributed by Axis Communications.
2307 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2308 we can now check for namespace violations and missing declarations. The
2309 result is that almost all headers are now Unix-compliant (as defined in
2310 the upcoming XPG6). The negative side is that some programs might need
2311 corrections, too, if they depend on the incorrect form of the headers in
2312 previous versions which defined too many symbols and included too many
2315 * regex now handles multibyte character sets correctly.
2316 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2318 * iconv (the program) does now conform to the upcoming XPG6 and handles
2319 charmaps. Instead of the charset names the path of charmaps can be
2320 provided and the conversion happens based on this data.
2321 Contributed by Ulrich Drepper.
2323 * The locale program now provides detailed information about the installed
2328 only lists the names of the supported locales
2332 provides details such as country, language, and codeset name.
2333 Contributed by Ulrich Drepper.
2337 * The gencat program now parses the input file according to the charset
2338 selected by the LC_CTYPE category. This is important for stateful
2339 character sets. To make generating catalogs easier there is a way
2340 to overwrite the charset selected by the locale: before the first
2341 message or $ quote line the catalog can contain a line like
2343 $ codeset=ISO-8859-2
2345 to select the charset (ISO-8859-2 in this case).
2347 Implemented by Shinya Hanataka and Ulrich Drepper.
2349 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2350 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2351 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2353 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2354 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2356 * The LANGUAGE environment variable is now ignored unless the locale is
2357 changed from the default "C" locale.
2359 * The usual bug fixes.
2363 * Greg McGary added runtime support for bounds checking using gcc's
2364 new -fbounded-pointers option. ix86 target is complete. PowerPC
2367 * Thorsten Kukuk added secure mode support to nscd.
2369 * The Berkeley db libraries have been removed.
2371 Related, the nss_db module is now in a separate package since it
2372 obviously requires a database library being available.
2374 * Wide character I/O streams implemented by Ulrich Drepper.
2376 * Functions from the extended socket API added by Ulrich Drepper.
2378 * Functions feenableexcept and fedisableexcept to control the
2379 behaviour of individual exceptions have been added by Andreas Jaeger.
2381 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2383 * The resolver code has been updated from bind 8.2.3-T5B which supports
2384 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2387 This change could in some situations effect backward compatibility. Since
2388 now `_res' is a thread-local instead of a global variable, modifying it
2389 in one thread does not have any effect in other threads.
2391 The resolver library was also extended to allow IPv6 as the transport
2392 protocol for the requests. This work was done by Stig Venaas.
2394 * Compatibility code for K&R C compilers has been removed from the
2395 header files. A ISO C compiler is needed to use the library
2396 (conforming to either C89 or C99 standard).
2398 * Complete rewrite of the localedef program to support multibyte character
2399 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2400 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2401 Implemented by Ulrich Drepper.
2403 Bruno Haible significantly improved the generation and use of the data
2404 structures for the wide character tables.
2406 * Plural handling in gettext implemented by Ulrich Drepper.
2408 * The utmp daemon has been removed.
2410 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2412 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2415 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2417 * POSIX barriers implemented by Kaz Kylheku.
2419 * POSIX spawn function family implemented by Ulrich Drepper.
2421 * POSIX spinlocks are now available.
2423 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2425 * the configure option --enable-kernel=X.Y.Z allows to strip out
2426 compatibility for kernel versions before X.Y.Z. This is currently only
2427 implemented for Linux.
2429 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2430 field sin6_scope_id. This means that all programs using IPv6 should be
2431 recompiled. Don't expect binary compatibility with previous glibc
2434 * various conversion modules for IBM character sets contributed by
2437 * IA-64 port by Jes Sorensen and HJ Lu.
2451 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2453 * New xdr functions are added; some rpc functions are now 64bit clean.
2455 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2457 * Fixed known incompatibilities with glibc 2.0.
2459 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2461 * Optimized a number of functions (especially the ELF dynamic loader).
2463 * Update timezone data files.
2465 * lots of charmaps corrections
2467 * some new locale definitions and charmaps
2472 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2473 conform to POSIX member type requirements. Symbol versions have been
2474 adjusted as needed within the library, and for direct use by applications,
2475 but there is potential for problems if third-party libraries use
2476 struct timeval as part of their interface. This does not present
2477 a problem for X and other "essential" system libraries.
2479 * An additional locale model to support C++ Standard Library locale
2480 model and probably more was implemented by Ulrich Drepper.
2482 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2485 * Miles Bader provided the `argp' function family to support hierachical
2486 command line argument parsing, layered on top of getopt.
2488 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2490 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2493 * scanf recognizes the %a and %A format for scanning floating point numbers.
2495 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2496 information and interfaces for the available integer types.
2498 * about 130 new math functions were added to implement the ISO C9x math
2501 * the new header <complex.h> contains definitions of the complex math
2502 functions from ISO C 9X.
2504 * the new header <tgmath.h> defines generic macros to use complex or
2505 real valued functions.
2507 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2509 * Andreas Jaeger provided a test suite for the math library.
2511 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2513 * Ulrich Drepper added character set conversion functions (iconv).
2515 * Optimized string functions have been added.
2517 * The localedata addon is now part of glibc.
2519 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2521 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2522 daemon for NSS (nscd).
2524 Missing a better place here are some numbers on improvements. Under
2525 Linux 2.1.125 un-tar-ing the kernel sources takes
2529 using local files 12.19s 6.88s 22.91s
2531 using NIS 13.92s 8.91s 26.34s
2533 using NIS & nscd 10.37s 7.34s 25.30s
2535 using NIS+ 27.57s 30.37s 640.46s
2537 using NIS+ & nscd 10.25s 7.83s 26.51s
2539 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2541 Keep in mind that non-namelookup related operations dominate above times.
2542 It was just a common complain that using NIS+ unpacking the kernel is
2545 [1] The old nscd implementation is not available anymore since it was
2546 distributed with glibc up to version 2.0.98 and thus is now replaced.
2548 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2550 * Mark Kettenis provided a Hesiod NSS module.
2552 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2553 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2555 * rcmd can now handle netgroups (Dick Streefland).
2557 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2560 * Support for the IPv6 protocol has been added to the socket API, as per the
2561 latest draft standards.
2563 * Support for Linux 2.2 has been added.
2565 * Interface changes relative to the latest 2.0.x release:
2566 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2567 addseverity NEW: Unix98
2568 alphasort64 NEW: LFS
2569 argp_err_exit_status NEW: argp, GNU ext
2570 argp_error NEW: argp, GNU ext
2571 argp_failure NEW: argp, GNU ext
2572 argp_help NEW: argp, GNU ext
2573 argp_parse NEW: argp, GNU ext
2574 argp_program_bug_address NEW: argp, GNU ext
2575 argp_program_version NEW: argp, GNU ext
2576 argp_program_version_hook NEW: argp, GNU ext
2577 argp_state_help NEW: argp, GNU ext
2578 argp_usage NEW: argp, GNU ext
2579 authdes_create NEW: Secure RPC
2580 authdes_getucred NEW: Secure RPC
2581 authdes_pk_create NEW: Secure RPC
2582 backtrace NEW: GNU ext.
2583 backtrace_symbols NEW: GNU ext.
2584 backtrace_symbols_fd NEW: GNU ext.
2586 cacosf NEW: ISO C 9x
2587 cacosh NEW: ISO C 9x
2588 cacoshf NEW: ISO C 9x
2589 cacoshl NEW: ISO C 9x
2590 cacosl NEW: ISO C 9x
2597 casinf NEW: ISO C 9x
2598 casinh NEW: ISO C 9x
2599 casinhf NEW: ISO C 9x
2600 casinhl NEW: ISO C 9x
2601 casinl NEW: ISO C 9x
2603 catanf NEW: ISO C 9x
2604 catanh NEW: ISO C 9x
2605 catanhf NEW: ISO C 9x
2606 catanhl NEW: ISO C 9x
2607 catanl NEW: ISO C 9x
2608 cbc_crypt NEW: Secure RPC
2612 ccoshf NEW: ISO C 9x
2613 ccoshl NEW: ISO C 9x
2619 cimagf NEW: ISO C 9x
2620 cimagl NEW: ISO C 9x
2621 clearerr_locked REMOVED
2622 clntunix_create NEW: sunrpc ext
2624 clog10 NEW: ISO C 9x
2625 clog10f NEW: ISO C 9x
2626 clog10l NEW: ISO C 9x
2636 cprojf NEW: ISO C 9x
2637 cprojl NEW: ISO C 9x
2639 crealf NEW: ISO C 9x
2640 creall NEW: ISO C 9x
2645 csinhf NEW: ISO C 9x
2646 csinhl NEW: ISO C 9x
2649 csqrtf NEW: ISO C 9x
2650 csqrtl NEW: ISO C 9x
2654 ctanhf NEW: ISO C 9x
2655 ctanhl NEW: ISO C 9x
2657 des_setparity NEW: Secure RPC
2658 ecb_crypt NEW: Secure RPC
2659 endutxent NEW: Unix98
2661 exp10f NEW: ISO C 9x
2662 exp10l NEW: ISO C 9x
2666 fattach NEW: STREAMS
2667 fdetach NEW: STREAMS
2671 feclearexcept NEW: ISO C 9x
2672 fegetenv NEW: ISO C 9x
2673 fegetexceptflag NEW: ISO C 9x
2674 fegetround NEW: ISO C 9x
2675 feholdexcept NEW: ISO C 9x
2677 feraiseexcept NEW: ISO C 9x
2678 ferror_locked REMOVED
2679 fesetenv NEW: ISO C 9x
2680 fesetexceptflag NEW: ISO C 9x
2681 fesetround NEW: ISO C 9x
2682 fetestexcept NEW: ISO C 9x
2683 feupdateenv NEW: ISO C 9x
2684 fflush_locked REMOVED
2688 fgets_unlocked NEW: GNU ext.
2689 fileno_locked REMOVED
2701 fputc_locked REMOVED
2702 fputs_unlocked NEW: GNU ext.
2703 fread_unlocked NEW: GNU ext.
2708 fstatvfs NEW: Unix98
2712 ftruncate64 NEW: LFS
2714 fwrite_unlocked NEW: GNU ext.
2715 gai_strerror NEW: IPv6
2719 getchar_locked REMOVED
2721 getdate_err NEW: Unix98
2722 getdate_r NEW: GNU ext.
2724 getnameinfo NEW: IPv6
2725 getnetname NEW: Secure RPC
2726 getpmsg NEW: STREAMS
2727 getpt NEW: Unix98 PTY
2728 getrlimit64 NEW: LFS
2729 getutxent NEW: Unix98
2730 getutxid NEW: Unix98
2731 getutxline NEW: Unix98
2732 glob64 NEW: GNU ext.
2733 globfree64 NEW: GNU ext.
2734 gnu_get_libc_release NEW: GNU ext.
2735 gnu_get_libc_version NEW: GNU ext.
2736 grantpt NEW: Unix98 PTY
2737 host2netname NEW: Secure RPC
2739 iconv_close NEW: iconv
2740 iconv_open NEW: iconv
2741 if_freenameindex NEW: IPv6
2742 if_indextoname NEW: IPv6
2743 if_nameindex NEW: IPv6
2744 if_nametoindex NEW: IPv6
2745 in6addr_any NEW: IPv6
2746 in6addr_loopback NEW: IPv6
2747 inet6_isipv4mapped NEW: IPv6
2748 isastream NEW: STREAMS
2749 iswblank NEW: GNU ext.
2750 key_decryptsession NEW: Secure RPC
2751 key_decryptsession_pk NEW: Secure RPC
2752 key_encryptsession NEW: Secure RPC
2753 key_encryptsession_pk NEW: Secure RPC
2754 key_gendes NEW: Secure RPC
2755 key_get_conv NEW: Secure RPC
2756 key_secretkey_is_set NEW: Secure RPC
2757 key_setnet NEW: Secure RPC
2758 key_setsecret NEW: Secure RPC
2759 llrint NEW: ISO C 9x
2760 llrintf NEW: ISO C 9x
2761 llrintl NEW: ISO C 9x
2762 llround NEW: ISO C 9x
2763 llroundf NEW: ISO C 9x
2764 llroundl NEW: ISO C 9x
2769 lrintf NEW: ISO C 9x
2770 lrintl NEW: ISO C 9x
2771 lround NEW: ISO C 9x
2772 lroundf NEW: ISO C 9x
2773 lroundl NEW: ISO C 9x
2775 makecontext NEW: Unix98
2776 mempcpy NEW: GNU ext.
2779 modify_ldt NEW: kernel
2783 nearbyint NEW: ISO C 9x
2784 nearbyintf NEW: ISO C 9x
2785 nearbyintl NEW: ISO C 9x
2786 netname2host NEW: Secure RPC
2787 netname2user NEW: Secure RPC
2788 nexttoward NEW: ISO C 9x
2789 nexttowardf NEW: ISO C 9x
2790 nexttowardl NEW: ISO C 9x
2794 passwd2des NEW: Secure RPC
2796 pow10f NEW: GNU ext.
2797 pow10l NEW: GNU ext.
2800 printf_size NEW: GNU ext.
2801 printf_size_info NEW: GNU ext.
2802 profil_counter REMOVED
2803 pthread_mutexattr_getkind_np REPLACED
2804 pthread_mutexattr_setkind_np REPLACED
2805 ptsname NEW: Unix98 PTY
2806 ptsname_r NEW: Unix98 PTY
2808 putchar_locked REMOVED
2809 putgrent NEW: GNU ext.
2811 putpmsg NEW: STREAMS
2812 pututxline NEW: Unix98
2816 readdir64_r NEW: LFS
2817 remquo NEW: ISO C 9x
2818 remquof NEW: ISO C 9x
2819 remquol NEW: ISO C 9x
2821 roundf NEW: ISO C 9x
2822 roundl NEW: ISO C 9x
2824 scalbln NEW: ISO C 9x
2825 scalblnf NEW: ISO C 9x
2826 scalblnl NEW: ISO C 9x
2828 sendfile NEW: kernel
2829 setcontext NEW: Unix98
2830 setrlimit64 NEW: LFS
2831 setutxent NEW: Unix98
2833 sigignore NEW: Unix98
2834 sigqueue NEW: POSIX.1b
2835 sigrelse NEW: Unix98
2836 sigset NEW: POSIX.1b
2837 sigtimedwait NEW: POSIX.1b
2838 sigwaitinfo NEW: POSIX.1b
2839 sincos NEW: GNU ext.
2840 sincosf NEW: GNU ext.
2841 sincosl NEW: GNU ext.
2845 strcasestr NEW: GNU ext.
2846 strtoimax NEW: ISO C 9x
2847 strtoumax NEW: ISO C 9x
2848 strverscmp NEW: GNU ext.
2849 svcauthdes_stats NEW: Secure RPC
2850 svcunix_create NEW: sunrpc etx
2851 svcunixfd_create NEW: sunrpc ext
2852 swapcontext NEW: Unix98
2853 tcgetsid NEW: Unix98 PTY
2854 tdestroy NEW: GNU ext.
2855 tgamma NEW: ISO C 9x
2856 tgammaf NEW: ISO C 9x
2857 tgammal NEW: ISO C 9x
2861 truncf NEW: ISO C 9x
2862 truncl NEW: ISO C 9x
2864 unlockpt NEW: Unix98 PTY
2865 updwtmpx NEW: Unix98
2866 user2netname NEW: Secure RPC
2867 utmpxname NEW: Unix98
2868 versionsort NEW: GNU ext.
2869 versionsort64 NEW: GNU ext.
2871 wcscasecmp NEW: GNU ext.
2872 wcsncasecmp NEW: GNU ext.
2873 wcsnlen NEW: GNU ext.
2874 wcstoimax NEW: ISO C 9x
2875 wcstoll NEW: ISO C 9x
2876 wcstoull NEW: ISO C 9x
2877 wcstoumax NEW: ISO C 9x
2879 wordexp NEW: POSIX.2
2880 wordfree NEW: POSIX.2
2881 write_profiling REMOVED
2882 xdecrypt NEW: Secure RPC
2883 xdr_authdes_cred NEW: Secure RPC
2884 xdr_authdes_verf NEW: Secure RPC
2885 xdr_cryptkeyarg NEW: Secure RPC
2886 xdr_cryptkeyarg2 NEW: Secure RPC
2887 xdr_cryptkeyres NEW: Secure RPC
2888 xdr_getcredres NEW: Secure RPC
2889 xdr_key_netstarg NEW: Secure RPC
2890 xdr_key_netstres NEW: Secure RPC
2891 xdr_keybuf NEW: Secure RPC
2892 xdr_keystatus NEW: Secure RPC
2893 xdr_netnamestr NEW: Secure RPC
2894 xdr_sizeof NEW: Secure RPC
2895 xdr_unixcred NEW: sunrpc ext
2896 xencrypt NEW: Secure RPC
2897 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2908 * inet_ntoa is thread-safe
2910 * updwtmp is moved from libutil to libc
2912 * rewrite of cbrt function
2914 * update of timezone data
2928 * add atoll function
2930 * fix complex problems in Berkeley DB code
2932 * fix math functions
2936 * fixed lots of header problems (especially Linux/GNU specific)
2938 * dynamic loader preserves all registers
2940 * Roland McGrath provided support for handling of auxiliary objects in
2941 the ELF dynamic loader.
2943 * support for parallel builds is improved
2947 * GNU extensions are no longer declared by default. To enable them you
2948 must define the macro `_GNU_SOURCE' in your program or compile with
2951 * The library has changed from using GNU ld symbol aliases to using weak
2952 symbols where available. The ELF object file format supports weak
2953 symbols; GNU ld also supports weak symbols in the a.out format. (There
2954 is also now support for other GNU ld extensions in ELF. Use the
2955 `--with-elf' option to configure to indicate you have ELF, and
2956 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
2957 of many files which contained only symbol aliases, reducing the size of
2958 the source and the compiled library; many other files were renamed to
2959 less cryptic names previously occupied by the symbol alias files.
2960 There is a new header file <elf.h> for programs which operate on
2961 files in the ELF format.
2963 * Converted to Autoconf version 2, so `configure' has more options.
2964 Run `configure --help' to see the details.
2966 * The library can now be configured to build profiling, highly-optimized
2967 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
2968 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
2969 `configure' enable building these extra libraries. The shared library is
2970 built by default when using both ELF and GNU ld. When shared libraries
2971 are enabled, the new library `-ldl' is available for arbitrary run-time
2972 loading of shared objects; its interface is defined in <dlfcn.h>. The
2973 new header file <link.h> gives access to the internals of the run-time
2974 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
2975 application of same name on other systems and it provides information
2976 about dynamically linked binaries.
2978 * The C library now provides the run-time support code for profiling
2979 executables compiled with `-pg'. Programs can control the profiling code
2980 through the interface in <sys/gmon.h>. The `gmon.out' files written by
2981 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
2982 the support for this file format was contributed by David Mosberger-Tang.
2984 * The math code has been replaced with a math library based on fdlibm from
2985 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
2986 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
2987 functions. The math functions now reside in a separate library, so
2988 programs using them will need to use `-lm' their linking commands.
2990 * John C. Bowman contributed optimized ix87 assembler inline functions.
2992 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
2993 mechanism similar to that found in Solaris 2. This is now used for the
2994 group, passwd, hosts, networks, services, protocols, rpc, ethers,
2995 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
2996 file controls what services are used for each individual database. This
2997 works by loading shared libraries with names specified in `nsswitch.conf',
2998 so service modules can be changed or added at any time without even
2999 relinking any program. Currently there are the file, db, and NIS based
3000 NSS services available.
3002 * The new functions `strtoq' and `strtouq' parse integer values from
3003 strings, like `strtol' and `strtoul', but they return `long long int' and
3004 `unsigned long long int' values, respectively (64-bit quantities).
3006 * The new functions `strtof' and `strtold' parse floating-point values from
3007 strings, like `strtod', but they return `float' and `long double' values,
3008 respectively (on some machines `double' and `long double' are the same).
3010 * Ulrich Drepper has contributed new implementations of the floating-point
3011 printing and reading code used in the `printf' family of functions and
3012 `strtod', `strtof', and `strtold'. These new functions are perfectly
3013 accurate, and much faster than the old ones.
3015 * The implementation of the POSIX locale model was completely rewritten by
3016 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3017 to compile the POSIX locale definition.
3019 * The former dummy implementations of the strcoll and strxfrm function are
3020 now replaced by fully functional code contributed by Ulrich Drepper. The
3021 collation information comes from the POSIX locale definitions.
3023 * The new header <langinfo.h> defines an interface for accessing
3024 various locale-dependent data (using the locale chosen with `setlocale').
3026 * Ulrich Drepper has contributed a new suite of functions for operation on
3027 wide-character and multibyte-character strings, in <wchar.h>;
3028 and classification and case conversion of wide characters, in <wctype.h>.
3029 These new functions are conforming to the ISO C, Amendement 1 specification.
3031 * There is now a second implementation of the standard I/O library available.
3032 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3033 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3035 * You can now use positional parameter specifications in format strings
3036 for the `printf' and `scanf' families of functions. For example,
3037 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3038 ``Number 6, Mr Jones''. This is mainly useful when providing different
3039 format strings for different languages, whose grammars may dictate
3040 different orderings of the values being printed. To support this
3041 feature, the interface for `register_printf_handler' has changed; see
3042 the header file <printf.h> for details.
3044 * The `printf' and `scanf' families of functions now understand a new
3045 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3046 to group numbers as indicated by the locale; for `scanf' and friends, this
3047 says to accept as valid only a number with all the proper grouping
3048 separators in the right places. In the default "C" locale, numbers are
3049 not grouped; but locales for specific countries will define the usual
3050 conventions (i.e. separate thousands with `,' in the US locale).
3052 * The pgrp functions have been regularized, slightly incompatibly but much
3053 less confusingly. The core functions are now `getpgid' and `setpgid',
3054 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3055 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3056 provided for compatibility. There is no longer an incompatible `getpgrp'
3057 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3059 * The new header file <fts.h> and suite of functions simplify programs that
3060 operate on directory trees. This code comes from 4.4 BSD.
3062 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3063 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3064 NSS scheme used in glibc.
3066 * There is a new malloc debugging hook `__memalign_hook'.
3068 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3069 `unsigned int' in <sys/types.h>. These are for compatibility only and
3070 their use is discouraged.
3072 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3073 done differently, so that it works even without GNU ld.
3075 * New function `euidaccess' checks allowed access to a file like `access',
3076 but using the effective IDs instead of the real IDs.
3078 * The time zone data files have been updated for the latest and greatest
3079 local time conventions of the countries of the world.
3081 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3084 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3085 for formatting floating-point numbers. They are provided only for
3086 compatibility; new programs should use `sprintf' instead. There are
3087 also equivalent function for the `long double' floating-point type and
3088 all functions also exist in a reentrant form.
3090 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3091 functions for maintaining the login-record files (primarily of use to
3092 system programs such as `login'), and convenient functions for
3093 allocating and initializing a pseudo-terminal (pty) device.
3095 * Ulrich Drepper has contributed new support for System V style
3096 shared memory and IPC on systems that support it.
3098 * Ulrich Drepper has contributed several miscellaneous new functions found
3099 in System V: The `hsearch' family of functions provide an effective
3100 implementation of hash tables; `a64l' and `l64a' provide a very simple
3101 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3104 * Ulrich Drepper has contributed new reentrant counterparts for the
3105 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3107 * Ulrich Drepper has contributed new, highly-optimized versions of several
3108 string functions for the i486/Pentium family of processors.
3110 * Ulrich Drepper has updated the Linux-specific code, based largely
3111 on work done in Hongjiu Lu's version of GNU libc for Linux.
3112 The GNU library now supports Linux versions 2.0.10 and later,
3113 using the ELF object file format (i[3456]86-*-linux).
3115 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3117 * David Mosberger-Tang and Richard Henderson have ported the C library
3118 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3119 dynamic linking support for ELF/Alpha.
3121 * Richard Henderson contributed several Alpha optimized assembler function
3122 for arithmetic and string handling.
3124 * Ulrich Drepper has contributed a new set of message catalog functions to
3125 support multiple languages using the <libintl.h> interface, for use with
3126 his new package GNU gettext. Translation volunteers have contributed
3127 catalogs of the library's messages in Spanish, German, and Korean.
3129 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3130 program and the `catgets' function for reading the catalog files it
3131 creates. (The <libintl.h> interface is preferred; we include the
3132 <nl_types.h> interface using `catgets' only for source compatibility with
3133 programs already written to use it.)
3135 * New header file <values.h> gives SVID-compatible names for <limits.h>
3138 * Various new macros, declarations, and small header files for compatibility
3141 * New function `group_member' is a convenient way to check if a process has
3142 a given effective group ID.
3144 * When using GCC 2.7 and later, the socket functions are now declared in a
3145 special way so that passing an argument of type `struct sockaddr_in *',
3146 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3147 `struct sockaddr *' type, does not generate a type-clash warning.
3149 * New function `error' declared in header file <error.h> is a convenient
3150 function for printing error messages and optionally exiting; this is the
3151 canonical function used in GNU programs. The new functions `err', `warn',
3152 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3153 doing the same thing.
3155 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3156 POSIX.2 `glob' function to do ~ and {...} expansion.
3158 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3159 `clearenv' which is used in POSIX.9 is also available.
3161 * New function `getsid' returns session ID number on systems that support it.
3163 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3164 files <db.h> and <mpool.h> provide a rich set of functions for several
3165 types of simple databases stored in memory and in files, and <ndbm.h> is
3166 an old `ndbm'-compatible interface using the `db' functions. Link with
3167 `-ldb' to get these functions.
3169 * New macro `strdupa' copies a string like `strdup', but uses local stack
3170 space from `alloca' instead of dynamic heap space from `malloc'.
3172 * New function `strnlen' is like `strlen' but searches only a given maximum
3173 number of characters for the null terminator. `stpncpy', `strndup' and
3174 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3177 * New function `statfs' in header <sys/statfs.h>.
3179 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3180 provide convenient functions for operating on blocks of null-terminated
3183 * A new suite of functions in <utmp.h> handle all the details of reading
3184 and writing the utmp file.
3186 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3189 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3190 completely POSIX compliant and also implemented the extended functionality
3191 to handle alternate digit representation and alternate era date formats.
3193 * Ulrich Drepper provided an implementation of the `strptime' function
3194 defined in XPG4.2 which transforms a string into a `struct tm' value.
3196 * Paul Eggert provided the tzselect shell script as part of the timezone
3197 code. The shell script makes it easy to select the correct timezone
3200 * The implementation of the malloc family of functions is completely replaced
3201 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3202 The implementation uses the mmap function (if available) and it is
3203 optimized for the use in multi threaded programs.
3205 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3206 functions. This new functionality is usable by specifying a special
3207 salt string and it is compatible with implementation on *BSD systems.
3209 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3210 `getsubopt' to handle second level command line options, `bsd_signal'
3211 to access BSD style `signal' functionality, the obsolete `regexp' style
3214 * the `lchown' function is available on system which support this
3217 * The implementation of the shadow password handling function was contributed
3220 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3222 * POSIX.1g support was added. The <sys/select.h> header is available,
3223 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3224 implementation of `getaddrinfo'.
3228 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3230 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3231 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3233 * The new function `daemon' from 4.4 BSD is useful for server programs that
3234 want to put themselves in the background.
3236 * Joel Sherrill has contributed support for several standalone boards that
3237 run without an operating system.
3239 * `printf', `scanf' and friends now accept a `q' type modifier for long
3240 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3242 * All of the code taken from BSD (notably most of the math and networking
3243 routines) has been updated from the BSD 4.4-Lite release.
3245 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3247 * The new functions `getdomainname' and `setdomainname' fetch or change the
3248 YP/NIS domain name. These are system calls which exist on systems which
3251 * The time zone data files have been updated for the latest international
3254 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3255 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3259 * The C library now includes support for Sun RPC, from Sun's free
3260 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3261 are included. (There is still no support for YP.)
3263 * Tom Quinn has contributed a port of the C library to SGI machines running
3264 Irix 4 (mips-sgi-irix4).
3266 * The new `lockf' function is a simplified interface to the locking
3267 facilities of `fcntl', included for compatibility.
3269 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3271 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3273 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3276 * The error message printed by `assert' for a failed assertion now includes
3277 the name of the program (if using GNU ld) and the name of the calling
3278 function (with versions of GCC that support this).
3280 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3282 * The library now includes the <sys/mman.h> header file and memory
3283 management functions `mmap', `munmap', `mprotect', `msync', and
3284 `madvise', on systems that support those facilities.
3286 * The interface for `mcheck' has changed slightly: the function called to
3287 abort the program when an allocation inconsistency is detected now takes
3288 an argument that indicates the type of failure. The new function
3289 `mprobe' lets you request a consistency check for a particular block at
3290 any time (checks are normally done only when you call `free' or `realloc'
3293 * It is now possible to easily cross-compile the C library, building on one
3294 system a library to run on another machine and/or operating system. All
3295 you need to do is set the variable `HOST_CC' in `configparms' to the
3296 native compiler for programs to run on the machine you are building on (a
3297 few generator programs are used on Unix systems); set `CC' to the
3300 * The new function `fexecve' (only implemented on the GNU system) executes
3301 a program file given a file descriptor already open on the file.
3305 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3306 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3308 * You can set the variable `libprefix' in `configparms' to specify a prefix
3309 to be prepended to installed library files; this makes it easy to install
3310 the GNU C library to be linked as `-lgnuc' or whatever.
3312 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3313 copies a limited number of characters from a string, and returns the
3314 address of the last character written.
3316 * You no longer need to check for whether the installed `stddef.h' is
3317 compatible with the GNU C library. configure now checks for you.
3319 * You can now define a per-stream `fileno' function to convert the
3320 stream's cookie into an integral file descriptor.
3322 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3323 allocates zero bytes of storage, and returns a unique pointer which
3324 you can pass to `realloc' or `free'. The behavior is undefined if
3325 you dereference this pointer.
3327 * The C library now runs on Sony NEWS m68k machines running either
3328 NewsOS 3 or NewsOS 4.
3330 * The new `syscall' function is a system-dependent primitive function
3331 for invoking system calls. It has the canonical behavior on Unix
3332 systems, including unreliable return values for some calls (such as
3333 `pipe', `fork' and `getppid').
3335 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3336 to `EAGAIN', which is the preferred name. On systems whose kernels
3337 use two distinct codes, the C library now translates EWOULDBLOCK to
3338 EAGAIN in every system call function.
3342 * The GNU C Library Reference Manual is now distributed with the library.
3343 `make dvi' will produce a DVI file of the printed manual.
3344 `make info' will produce Info files that you can read on line using C-h i
3345 in Emacs or the `info' program.
3346 Please send comments on the manual to bug-glibc-manual@gnu.org.
3348 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3350 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3352 * Jason Merrill has contributed a port to the Sequent Symmetry running
3353 Dynix version 3 (i386-sequent-dynix).
3355 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3356 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3358 * New function `memory_warnings' lets you arrange to get warnings when
3359 malloc is running out of memory to allocate, like Emacs gives you.
3361 * The C library now contains the relocating allocator used in Emacs 19 for
3362 its editing buffers. This allocator (ralloc) minimizes allocation
3363 overhead and fragmentation by moving allocated regions around whenever it
3364 needs to. You always refer to a ralloc'd region with a "handle" (a
3365 pointer to a pointer--an object of type `void **').
3367 * There is a new `printf' format: `%m' gives you the string corresponding
3368 to the error code in `errno'.
3370 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3371 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3372 pass, the `a' modifier says to fill in a `char **' you pass with a
3375 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3376 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3377 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3379 * `mkstemp' is a traditional Unix function to atomically create and open a
3380 uniquely-named temporary file.
3384 * The standard location for the file that says what the local timezone is
3385 has changed again. It is now `/usr/local/etc/localtime' (or more
3386 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3388 * The distribution no longer contains any files with names longer than 14
3391 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3392 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3394 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3398 * The configuration process has changed quite a bit. The `configure'
3399 script is now used just like the configuration scripts for other GNU
3400 packages. The `sysdeps' directory hierarchy is much rearranged.
3401 The file `INSTALL' explains the new scheme in detail.
3403 * The header files no longer need to be processed into ANSI C and
3404 traditional C versions. There is just one set of files to install, and
3405 it will work with ANSI or old C compilers (including `gcc -traditional').
3407 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3408 MIPS DECStation running Ultrix 4.
3410 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3411 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3412 made itself into a shared library.
3414 * Yet further improved support for the i386, running 4.3 BSD-like systems
3415 (such as Mach 3 with the Unix single-server), or System V.
3417 * New function `strncasecmp' to do case-insensitive string comparison
3418 with limited length.
3420 * New function `strsep' is a reentrant alternative to `strtok'.
3422 * New functions `scandir' and `alphasort' for searching directories.
3424 * New function `setenv' is a better interface to `putenv'.
3426 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3427 function for traversing a directory tree.
3429 * The GNU obstack package is now also part of the C library.
3430 The new function `open_obstack_stream' creates a stdio stream that
3431 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3432 formatted output directly to an obstack.
3434 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3435 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3437 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3439 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3440 things to your strings.
3442 * There are some new test programs: test-fseek, testmb, and testrand.
3444 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3445 These ports are not finished, but are a good starting place for really
3446 supporting those systems.
3448 * `/etc/localtime' is now the standard location for the file that says what
3449 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3450 This follows the general principle that `/etc' is the place for all local
3451 configuration files.
3453 * The C library header files now use `extern "C"' when used by the C++
3454 compiler, so the C library should now work with C++ code.
3456 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3457 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3460 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3461 function which uses the `merge sort' algorithm, and is said to be
3462 significantly faster than the old GNU `qsort' function. Merge sort is
3463 now the standard `qsort' function. The new algorithm can require a lot
3464 of temporary storage; so, the old sorting function is called when the
3465 required storage is not available.
3467 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3468 provides the Unix `crypt' function, plus some other entry points.
3470 * The code and header files taken from 4.4 BSD have been updated with the
3471 latest files released from Berkeley.
3473 ----------------------------------------------------------------------
3474 Copyright information:
3476 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3478 Permission is granted to anyone to make or distribute verbatim copies
3479 of this document as received, in any medium, provided that the
3480 copyright notice and this permission notice are preserved,
3481 thus giving the recipient permission to redistribute in turn.
3483 Permission is granted to distribute modified versions
3484 of this document, or of portions of it,
3485 under the above conditions, provided also that they
3486 carry prominent notices stating who last changed them.
3489 version-control: never