]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
Enhance --enable-tunables to select tunables frontend at build time
[thirdparty/glibc.git] / NEWS
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.
4
5 Please send GNU C library bug reports via <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
7 \f
8 Version 2.25
9
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
13 Library.
14
15 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
16 18661-1:2014, is supported to enable declarations of functions and macros
17 from that TS. Note that not all features from that TS are supported by
18 the GNU C Library.
19
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
23 the GNU C Library.
24
25 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
26 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
27 Since the GNU C Library defaults to a much newer revision of POSIX, this
28 will only affect programs that specifically request an old conformance
29 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
30 see a change in the visible declarations, but a program compiled with
31 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
32 will not.
33
34 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
35 defined by all multithreaded code, but glibc has not required this for
36 many years.
37
38 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
39 means that in a future release, the macros “major”, “minor”, and “makedev”
40 will only be available from <sys/sysmacros.h>.
41
42 These macros are not part of POSIX nor XSI, and their names frequently
43 collide with user code; see for instance glibc bug 19239 and Red Hat bug
44 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
45 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
46 problem.
47
48 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
49 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
50 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
51
52 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
53 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
54 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
55 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
56 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
57 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
58 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
59 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
60 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
61 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
62 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
63 WINT_WIDTH.
64
65 * New <math.h> features are added from TS 18661-1:2014:
66
67 - Signaling NaN macros: SNANF, SNAN, SNANL.
68
69 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
70 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
71 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
72
73 - llogb functions: the llogb, llogbf and llogbl functions, and the
74 FP_LLOGB0 and FP_LLOGBNAN macros.
75
76 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
77 fminmagf, fminmagl.
78
79 - Comparison macros: iseqsig.
80
81 - Classification macros: iscanonical, issubnormal, iszero.
82
83 - Total order functions: totalorder, totalorderf, totalorderl,
84 totalordermag, totalordermagf, totalordermagl.
85
86 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
87
88 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
89 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
90
91 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
92 are added to libc. They convert a floating-point number into string.
93
94 * Most of glibc can now be built with the stack smashing protector enabled.
95 It is recommended to build glibc with --enable-stack-protector=strong.
96 Implemented by Nick Alcock (Oracle).
97
98 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
99 intended to be used instead of memset() to erase sensitive data after use;
100 the compiler will not optimize out calls to explicit_bzero even if they
101 are "unnecessary" (in the sense that no _correct_ program can observe the
102 effects of the memory clear).
103
104 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
105 to float instead of double. This does not affect the ABI of any libraries
106 that are part of the GNU C Library, but may affect the ABI of other
107 libraries that use this type in their interfaces.
108
109 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
110 float_t and double_t types are now defined to long double instead of float
111 and double. These options are not the default, and this does not affect
112 the ABI of any libraries that are part of the GNU C Library, but it may
113 affect the ABI of other libraries that use this type in their interfaces,
114 if they are compiled or used with those options.
115
116 * The getentropy and getrandom functions, and the <sys/random.h> header file
117 have been added.
118
119 * The buffer size for byte-oriented stdio streams is now limited to 8192
120 bytes by default. Previously, on Linux, the default buffer size on most
121 file systems was 4096 bytes (and thus remains unchanged), except on
122 network file systems, where the buffer size was unpredictable and could be
123 as large as several megabytes.
124
125 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
126 for the Linux quota interface which predates kernel version 2.4.22 has
127 been removed.
128
129 * The malloc_get_state and malloc_set_state functions have been removed.
130 Already-existing binaries that dynamically link to these functions will
131 get a hidden implementation in which malloc_get_state is a stub. As far
132 as we know, these functions are used only by GNU Emacs and this change
133 will not adversely affect already-built Emacs executables. Any undumped
134 Emacs executables, which normally exist only during an Emacs build, should
135 be rebuilt by re-running “./configure; make” in the Emacs build tree.
136
137 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
138 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
139 “no-ip6-dotint” had already been the default, and support for the
140 “ip6-dotint” option was removed from the Internet in 2006.
141
142 * The "ip6-bytestring" resolver option and the corresponding RES_NOIP6DOTINT
143 flag from <resolv.h> have been removed. The option relied on a
144 backwards-incompatible DNS extension which was never deployed on the
145 Internet.
146
147 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
148 RES_BLAST defined in the <resolv.h> header file have been deprecated.
149 They were already unimplemented.
150
151 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
152 _res.flags are deprecated. The flag was standardized in RFC 2133, but
153 removed again from the IETF name lookup interface specification in RFC
154 2553. Applications should use getaddrinfo instead.
155
156 * DNSSEC-related declarations and definitions have been removed from the
157 <arpa/nameser.h> header file, and libresolv will no longer attempt to
158 decode the data part of DNSSEC record types. Previous versions of glibc
159 only implemented minimal support for the previous version of DNSSEC, which
160 is incompatible with the currently deployed version.
161
162 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
163 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
164 <arpa/nameser.h> header file because the distinction between RR types and
165 meta-RR types is not officially standardized, subject to revision, and
166 thus not suitable for encoding in a macro.
167
168 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
169 and rhook members of the res_state type in <resolv.h> have been removed.
170 The glibc stub resolver did not support these hooks, but the header file
171 did not reflect that.
172
173 * For multi-arch support it is recommended to use a GCC which has
174 been built with support for GNU indirect functions. This ensures
175 that correct debugging information is generated for functions
176 selected by IFUNC resolvers. This support can either be enabled by
177 configuring GCC with '--enable-gnu-indirect-function', or by
178 enabling it by default by setting 'default_gnu_indirect_function'
179 variable for a particular architecture in the GCC source file
180 'gcc/config.gcc'.
181
182 * GDB pretty printers have been added for mutex and condition variable
183 structures in POSIX Threads. When installed and loaded in gdb these pretty
184 printers show various pthread variables in human-readable form when read
185 using the 'print' or 'display' commands in gdb.
186
187 Security related changes:
188
189 On ARM EABI (32-bit), generating a backtrace for execution contexts which
190 have been created with makecontext could fail to terminate due to a
191 missing .cantunwind annotation. This has been observed to lead to a hang
192 (denial of service) in some Go applications compiled with gccgo. Reported
193 by Andreas Schwab. (CVE-2016-6323)
194
195 The following bugs are resolved with this release:
196
197 [The release manager will add the list generated by
198 scripts/list-fixed-bugs.py just before the release.]
199 \f
200 Version 2.24
201
202 * The minimum Linux kernel version that this version of the GNU C Library
203 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
204 kernel version 2.6.32 or later suffices (on architectures that already
205 required kernel versions more recent than 3.2, those requirements remain
206 unchanged). Linux 3.2 or later kernel headers are required on all
207 architectures.
208
209 * The pap_AN locale has been deleted. This has been deprecated for a long
210 time. It has been replaced by pap_AW & pap_CW, both of which have long
211 been included in previous releases.
212
213 * The readdir_r and readdir64_r functions have been deprecated. It is
214 recommended to use readdir and readdir64 instead.
215
216 * The type “union wait” has been removed. It was deprecated in the early
217 1990s and never part of POSIX. Application code should use the int type
218 instead of “union wait”.
219
220 * A new NSS action is added to facilitate large distributed system
221 administration. The action, MERGE, allows remote user stores like LDAP
222 to be merged into local user stores like /etc/groups in order to provide
223 easy to use, updated, and managed sets of merged credentials. The new
224 action can be used by configuring it in /etc/nsswitch.conf:
225 group: files [SUCCESS=merge] nis
226 Implemented by Stephen Gallagher (Red Hat).
227
228 * The deprecated __malloc_initialize_hook variable has been removed from the
229 API.
230
231 * The long unused localedef --old-style option has been removed. It hasn't
232 done anything in over 16 years. Scripts using this option can safely
233 drop it.
234
235 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
236 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
237 return the next representable value in the direction of positive infinity
238 and the nextdown functions return the next representable value in the
239 direction of negative infinity. These are currently enabled as GNU
240 extensions.
241
242 Security related changes:
243
244 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
245 could result in a stack overflow when getnetbyname was called with an
246 overly long name. (CVE-2016-3075)
247
248 * Previously, getaddrinfo copied large amounts of address data to the stack,
249 even after the fix for CVE-2013-4458 has been applied, potentially
250 resulting in a stack overflow. getaddrinfo now uses a heap allocation
251 instead. Reported by Michael Petlan. (CVE-2016-3706)
252
253 * The glob function suffered from a stack-based buffer overflow when it was
254 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
255 Reported by Alexander Cherepanov. (CVE-2016-1234)
256
257 * The Sun RPC UDP client could exhaust all available stack space when
258 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
259 alloca plugin for GCC. (CVE-2016-4429)
260
261 * The IPv6 name server management code in libresolv could result in a memory
262 leak for each thread which is created, performs a failing naming lookup,
263 and exits. Over time, this could result in a denial of service due to
264 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
265
266 The following bugs are resolved with this release:
267
268 [1170] localedata: ne_NP: update Nepali locale definition file
269 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
270 of MS-DOS.
271 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
272 [6796] math: fdim() does not set errno on overflow
273 [10354] libc: posix_spawn should use vfork() in more cases than presently
274 [11213] localedata: localedata: add copyright disclaimer to locale files
275 [12143] localedata: chr_US: new Cherokee locale
276 [12450] localedata: sgs_LT: new locale
277 [12676] localedata: ln_CD: new locale
278 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
279 CLDR data
280 [13304] math: fma, fmaf, fmal produce wrong results
281 [14259] build: --localedir arg to configure is ignored
282 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
283 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
284 [14934] localedata: es_CL: wrong first weekday chilean locale
285 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
286 romanisation
287 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
288 and +/-
289 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
290 [15368] nptl: raise() is not async-signal-safe
291 [15479] math: ceil, floor, round and trunc raise inexact exception
292 [15578] localedata: kk_KZ: various updates
293 [16003] localedata: pap_AN: punt old locale
294 [16137] localedata: iw_IL: punt old locale
295 [16190] localedata: eo: new esperanto locale
296 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
297 [16742] malloc: race condition: pthread_atfork() called before first
298 malloc() results in unexpected locking behaviour/deadlocks
299 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
300 all locales
301 [16983] localedata: postal_fmt does not allow %l and %n modifiers
302 [17565] localedata: pt_PT: wrong (work-)week start
303 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
304 [17950] build: Build fails with -msse
305 [18205] localedata: be_BY*: wrong first_weekday and first_workday
306 [18433] libc: posix_spawn does not return correctly upon failure to
307 execute
308 [18453] localedata: charmaps/IBM875: incorrect codes
309 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
310 -Wsystem-headers
311 [18896] localedata: he_IL: improvements for currency
312 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
313 Romanian locale data
314 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
315 symbol
316 [19056] libc: Deprecate readdir_r
317 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
318 language
319 [19198] localedata: nl_NL: small improvements for Dutch locales
320 [19257] network: Per-thread memory leak in __res_vinit with IPv6
321 nameservers (CVE-2016-5417)
322 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
323 machine
324 [19400] locale: Language missing in "iso-639.def", trivial fix in
325 description
326 [19431] malloc: Deadlock between fflush, getdelim, and fork
327 [19505] libc: Incorrect file descriptor validity checks in
328 posix_spawn_file_actions_add{open,close,dup2}
329 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
330 when using RTLD_NEXT
331 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
332 `intl/{gettextP,loadinfo}.h'
333 [19534] libc: execle, execlp may use malloc
334 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
335 [19573] network: res_nclose and __res_maybe_init disagree about name
336 server initialization, breaking Hesiod
337 [19575] localedata: Status of GB18030 tables
338 [19581] localedata: sr_* date_fmt string contains additional newline
339 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
340 Excavator core
341 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
342 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
343 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
344 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
345 double range
346 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
347 part zero incorrect
348 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
349 equality tests
350 [19610] dynamic-link: ldconfig -X removes stale symbolic links
351 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
352 [19633] locale: strfmon_l applies global locale to number formatting
353 [19642] network: Memory leak in getnameinfo
354 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
355 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
356 glibc-2.22
357 [19654] math: [x86_64] Need testcase for BZ #19590 fix
358 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
359 'tst-numeric.c'
360 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
361 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
362 low part
363 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
364 result
365 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
366 unaligned stack
367 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
368 pointers and lengths in error-case.
369 [19727] locale: Converting from/to UTF-xx with iconv() does not always
370 report errors on UTF-16 surrogates values.
371 [19755] nscd: nscd assertion failure in gc
372 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
373 [19759] libc: mempcpy shouldn't be inlined
374 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
375 [19765] libc: s390 needs an optimized mempcpy
376 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
377 NAME_MAX limit assumption (CVE-2016-1234)
378 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
379 [19787] network: Missing and incorrect truncation checks in getnameinfo
380 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
381 modes
382 [19791] network: Assertion failure in res_query.c with un-connectable name
383 server addresses
384 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
385 [19822] math: libm.so install clobbers old version
386 [19825] network: resolv: send_vc can return uninitialized data in second
387 response to getaddrinfo
388 [19830] network: nss_dns: should check RDATA length against buffer length
389 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
390 confronted with A/AAAA records of invalid size
391 [19837] nss: nss_db: No retries for some long lines with a larger buffer
392 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
393 [19853] stdio: Printing IBM long double in decimal with high precision is
394 sometimes incorrect
395 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
396 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
397 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
398 record types
399 [19865] network: Assertion failure or memory leak in
400 _nss_dns_getcanonname_r
401 [19868] network: nss_dns: netent code does not skip over non-PTR records
402 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
403 (CVE-2016-3075)
404 [19881] string: Improve x86-64 memset
405 [19907] string: Incorrect memcpy tests
406 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
407 resolving symbols
408 [19925] libc: termios.h XCASE namespace
409 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
410 [19929] libc: limits.h NL_NMAX namespace
411 [19931] stdio: Memory leak in vfprintf
412 [19957] libc: clone(CLONE_VM) access invalid parent memory
413 [19963] localedata: en_IL: New locale
414 [19989] stdio: stdio.h cuserid namespace
415 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
416 gethosts
417 [19996] locale: langinfo.h nl_langinfo_l namespace
418 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
419 [20010] network: getaddrinfo: Stack overflow in hostent translation
420 (CVE-2016-3706)
421 [20012] stdio: libio: fmemopen append mode failure
422 [20014] stdio: stdio.h namespace for pre-threads POSIX
423 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
424 [20023] libc: fcntl.h timespec namespace
425 [20024] math: [x86_64] vectorized sincos trashes the stack
426 [20031] network: nss_hesiod: Heap overflow in get_txt_records
427 [20041] time: sys/time.h timespec namespace
428 [20043] libc: unistd.h missing cuserid for UNIX98 and before
429 [20044] libc: unistd.h missing pthread_atfork for UNIX98
430 [20051] libc: ttyslot in wrong header under wrong conditions
431 [20054] libc: gethostname not declared for XPG4
432 [20055] libc: termios.h missing tcgetsid for XPG4
433 [20072] dynamic-link: x86 init_cpu_features is called twice in static
434 executable
435 [20073] libc: sys/stat.h fchmod namespace
436 [20074] libc: stdlib.h rand_r namespace
437 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
438 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
439 XPG3
440 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
441 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
442 (CVE-2016-4429)
443 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
444 [20119] libc: Wrong mask for processors level type from CPUID
445 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
446 AS not supporting AVX512
447 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
448 sNaN argument
449 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
450 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
451 argument
452 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
453 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
454 [20178] libc: posix_spawn{p} should not call exit
455 [20191] stdio: libio: vtables hardening
456 [20195] string: FMA4 detection requires CPUID execution with register
457 eax=0x80000001
458 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
459 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
460 subnormals
461 [20212] math: acos (sNaN) returns sNaN
462 [20213] math: asin (sNaN) returns sNaN
463 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
464 [20218] math: [i386] asinhl (sNaN) returns sNaN
465 [20219] math: [i386] atanhl (sNaN) returns sNaN
466 [20222] stdio: fopencookie: Mangle function pointers
467 [20224] math: [i386] cbrtl (sNaN) returns sNaN
468 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
469 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
470 input
471 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
472 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
473 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
474 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
475 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
476 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
477 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
478 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
479 [20240] math: modf (sNaN) returns sNaN
480 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
481 [20250] math: frexp (sNaN) returns sNaN
482 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
483 [20255] math: [i386] fdim, fdimf return with excess range and precision /
484 double rounding
485 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
486 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
487 subscript is below array bounds [-Werror=array-bounds]
488 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
489 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
490 (Only arm/linux)
491 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
492 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
493 fallbacks
494 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
495 "invalid" exceptions
496 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
497 [20316] localedata: id_ID: Februari instead of Pebruari
498 [20327] string: POWER8 strcasecmp returns incorrect result
499 [20347] math: Failure: Test: j0_downward (0xap+0)
500 [20348] libc: FAIL: misc/tst-preadvwritev64
501 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
502 [20350] libc: There is no test for p{read,write}64
503 [20357] math: Incorrect cos result for 1.5174239687223976
504 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
505 \f
506 Version 2.23
507
508 * Unicode 8.0.0 Support: Character encoding, character type info, and
509 transliteration tables are all updated to Unicode 8.0.0, using new
510 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
511 These updates cause user visible changes, such as the fixes for bugs
512 89, 16061, and 18568.
513
514 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
515 kernel-internal CPU set size. This means that requests that change the
516 CPU affinity which failed before (for example, an all-ones CPU mask) will
517 now succeed. Applications that need to determine the effective CPU
518 affinities need to call sched_getaffinity or pthread_getaffinity_np after
519 setting it because the kernel can adjust it (and the previous size check
520 would not detect this in the majority of cases).
521
522 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
523 the following new symbols are used: fts64_children, fts64_close,
524 fts64_open, fts64_read and fts64_set.
525
526 * getaddrinfo now detects certain invalid responses on an internal netlink
527 socket. If such responses are received, an affected process will
528 terminate with an error message of "Unexpected error <number> on netlink
529 descriptor <number>" or "Unexpected netlink response of size <number> on
530 descriptor <number>". The most likely cause for these errors is a
531 multi-threaded application which erroneously closes and reuses the netlink
532 file descriptor while it is used by getaddrinfo.
533
534 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
535 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
536 unnecessary serialization of memory allocation requests across threads.
537 The defect is now corrected. Users should see a substantial increase in
538 the concurent throughput of allocation requests for applications which
539 trigger this bug. Affected applications typically create and
540 destroy threads frequently. (Bug 19048 was reported and analyzed by
541 Ericsson.)
542
543 * There is now a --disable-timezone-tools configure option for disabling the
544 building and installing of the timezone related utilities (zic, zdump, and
545 tzselect). This is useful for people who build the timezone data and code
546 independent of the GNU C Library.
547
548 * The obsolete header <regexp.h> has been removed. Programs that require
549 this header must be updated to use <regex.h> instead.
550
551 * The obsolete functions bdflush, create_module, get_kernel_syms,
552 query_module and uselib are no longer available to newly linked binaries;
553 the header <sys/kdaemon.h> has been removed. These functions and header
554 were specific to systems using the Linux kernel and could not usefully be
555 used with the GNU C Library on systems with version 2.6 or later of the
556 Linux kernel.
557
558 * Optimized string, wcsmbs and memory functions for IBM z13.
559 Implemented by Stefan Liebler.
560
561 * Newly linked programs that define a variable called signgam will no longer
562 have it set by the lgamma, lgammaf and lgammal functions. Programs that
563 require signgam to be set by those functions must ensure that they use the
564 variable provided by the GNU C Library and declared in <math.h>, without
565 defining their own copy.
566
567 * The minimum GCC version that can be used to build this version of the GNU
568 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
569 still be used to compile programs using the GNU C Library.
570
571 Security related changes:
572
573 * An out-of-bounds value in a broken-out struct tm argument to strftime no
574 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
575
576 * The LD_POINTER_GUARD environment variable can no longer be used to disable
577 the pointer guard feature. It is always enabled. Previously,
578 LD_POINTER_GUARD could be used to disable security hardening in binaries
579 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
580 (CVE-2015-8777)
581
582 * An integer overflow in hcreate and hcreate_r could lead to an
583 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
584
585 * The catopen function no longer has unbounded stack usage. Reported by
586 Max. (CVE-2015-8779)
587
588 * The nan, nanf and nanl functions no longer have unbounded stack usage
589 depending on the length of the string passed as an argument to the
590 functions. Reported by Joseph Myers. (CVE-2014-9761)
591
592 * A stack-based buffer overflow was found in libresolv when invoked from
593 libnss_dns, allowing specially crafted DNS responses to seize control
594 of execution flow in the DNS client. The buffer overflow occurs in
595 the functions send_dg (send datagram) and send_vc (send TCP) for the
596 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
597 family. The use of AF_UNSPEC triggers the low-level resolver code to
598 send out two parallel queries for A and AAAA. A mismanagement of the
599 buffers used for those queries could result in the response of a query
600 writing beyond the alloca allocated buffer created by
601 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
602 the overflow. Thanks to the Google Security Team and Red Hat for
603 reporting the security impact of this issue, and Robert Holiday of
604 Ciena for reporting the related bug 18665. (CVE-2015-7547)
605
606 The following bugs are resolved with this release:
607
608 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
609 [887] math: Math library function "logb" and "nextafter" inconsistent
610 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
611 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
612 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
613 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
614 use `mkstemp'
615 [4404] localedata: German translation of "Alarm clock" is misleading
616 [6799] math: nextafter() and nexttoward() doen't set errno on
617 overflow/underflow errors
618 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
619 overflow/underflow
620 [10432] nis: _nss_nis_setnetgrent assertion failure
621 [11460] libc: fts has no LFS support
622 [12926] network: getaddrinfo()/make_request() may spin forever
623 [13065] nptl: Race condition in pthread barriers
624 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
625 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
626 are not contiguous
627 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
628 [14912] libc: Rename non-installed bits/*.h headers
629 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
630 [15367] math: Let gcc use __builtin_isinf
631 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
632 [15421] math: lgamma wrongly sets signgam for ISO C
633 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
634 FE_INVALID with argument out of range
635 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
636 all exceptions
637 [15786] dynamic-link: ifunc resolver functions can smash function
638 arguments
639 [15918] math: Unnecessary check for equality in hypotf()
640 [16061] localedata: Review / update transliteration data
641 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
642 should include
643 [16141] time: strptime %z offset restriction
644 [16171] math: drem should be alias of remainder
645 [16296] math: fegetround is pure?
646 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
647 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
648 GNU/Linux
649 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
650 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
651 arguments
652 [16422] math: [powerpc] math-float, math-double failing llrint tests with
653 "Exception "Inexact" set" on ppc32
654 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
655 [16517] math: Missing underflow exception from tanf/tan/tanl
656 [16519] math: Missing underflow exception from sinhf
657 [16520] math: Missing underflow exception from tanhf
658 [16521] math: Missing underflow exception from exp2
659 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
660 rounding results
661 [16734] stdio: fopen calls mmap to allocate its buffer
662 [16961] math: nan function incorrect handling of bad sequences
663 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
664 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
665 [16985] locale: localedef: confusing error message when opening output
666 fails
667 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
668 [17197] locale: Redundant shift character in iconv conversion output at
669 block boundary
670 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
671 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
672 [17250] dynamic-link: static linking breaks nss loading
673 (getaddrinfo/getpwnam/etc...)
674 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
675 4.7?
676 [17441] math: isnan() should use __builtin_isnan() in GCC
677 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
678 (related to lock elision)
679 [17787] manual: Exponent on page 324 of the PDF ends prematurely
680 [17886] time: strptime should be able to parse "Z" as a timezone with %z
681 [17887] time: strptime should be able to parse "+01:00" style timezones
682 [17905] libc: catopen() Multiple unbounded stack allocations
683 (CVE-2015-8779)
684 [18084] libc: backtrace (..., 0) dumps core on x86
685 [18086] libc: nice() sets errno to 0 on success
686 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
687 is too large (CVE-2015-8778)
688 [18251] dynamic-link: SONAME missing when audit modules provides path
689 [18265] libc: add attributes for wchar string and memory functions
690 [18370] math: csqrt missing underflows
691 [18421] libc: [hppa] read-only segment has dynamic relocations
692 [18472] libc: Obsolete syscall wrappers should be compat symbols
693 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
694 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
695 [18525] localedata: Remove locale timezone information
696 [18560] libc: [powerpc] spurious bits/ipc.h definitions
697 [18568] localedata: Update locale data to Unicode 8.0
698 [18589] locale: sort-test.sh fails at random
699 [18595] math: ctan, ctanh missing underflows
700 [18604] libc: assert macro-expands its argument
701 [18610] math: S390: fetestexcept() reports any exception if DXC-code
702 contains a vector instruction exception.
703 [18611] math: j1, jn missing errno setting on underflow
704 [18618] localedata: sync Chechen locale definitions with other *_RU
705 locales
706 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
707 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
708 [18665] network: In send_dg, the recvfrom function is NOT always using the
709 buffer size of a newly created buffer (CVE-2015-7547)
710 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
711 missing break ?
712 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
713 32bit processes
714 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
715 [18699] math: tilegx cproj() for various complex infinities does not yield
716 infinity
717 [18724] libc: Harden put*ent functions against data injection
718 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
719 [18755] build: build errors with -DNDEBUG
720 [18757] stdio: fmemopen fails to set errno on failure
721 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
722 be forced unloaded
723 [18781] libc: openat64 lacks O_LARGEFILE
724 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
725 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
726 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
727 [18790] math: [ldbl-128ibm] tanhl inaccurate
728 [18795] libc: stpncpy fortification misses buffer lengths that are
729 statically too large
730 [18796] build: build fails for --disable-mathvec
731 [18803] math: hypot missing underflows
732 [18820] stdio: fmemopen may leak memory on failure
733 [18823] math: csqrt spurious underflows
734 [18824] math: fma spurious underflows
735 [18825] math: pow missing underflows
736 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
737 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
738 [18870] build: sem_open.c fails to compile with missing symbol
739 FUTEX_SHARED
740 [18872] stdio: Fix memory leak in printf_positional
741 [18873] libc: posix_fallocate overflow check ineffective
742 [18875] math: Excess precision leads incorrect libm
743 [18877] libc: arm: mmap offset regression
744 [18887] libc: memory corruption when using getmntent on blank lines
745 [18918] localedata: hu_HU: change time to HH:MM:SS format
746 [18921] libc: Regression: extraneous stat() and fstat() performed by
747 opendir()
748 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
749 binaries (CVE-2015-8777)
750 [18951] math: tgamma missing underflows
751 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
752 signgam
753 [18953] localedata: lt_LT: change currency symbol to the euro
754 [18956] math: powf inaccuracy
755 [18961] math: [i386] exp missing underflows
756 [18966] math: [i386] exp10 missing underflows
757 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
758 [18969] build: multiple string test failures due to missing locale
759 dependencies
760 [18970] libc: Reference of pthread_setcancelstate in libc.a
761 [18977] math: float / long double Bessel functions not in XSI POSIX
762 [18980] math: i386 libm functions return with excess range and precision
763 [18981] math: i386 scalb*, ldexp return with excess range and precision
764 [18982] stdio: va_list and vprintf
765 [18985] time: Passing out of range data to strftime() causes a segfault
766 (CVE-2015-8776)
767 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
768 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
769 [19012] locale: iconv_open leaks memory on error path
770 [19016] math: clog, clog10 inaccuracy
771 [19018] nptl: Mangle function pointers in tls_dtor_list
772 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
773 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
774 [19048] malloc: malloc: arena free list can become cyclic, increasing
775 contention
776 [19049] math: [powerpc] erfc incorrect zero sign
777 [19050] math: [powerpc] log* incorrect zero sign
778 [19058] math: [x86_64] Link fail with -fopenmp and -flto
779 [19059] math: nexttoward overflow incorrect in non-default rounding modes
780 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
781 [19074] network: Data race in _res_hconf_reorder_addrs
782 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
783 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
784 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
785 rounding modes
786 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
787 ILP32
788 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
789 [19086] manual: posix_fallocate64 documented argument order is wrong.
790 [19088] math: lround, llround missing exceptions close to overflow
791 threshold
792 [19094] math: lrint, llrint missing exceptions close to overflow threshold
793 [19095] math: dbl-64 lrint incorrect for 64-bit long
794 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
795 [19124] dynamic-link: ld.so failed to build with older assmebler
796 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
797 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
798 [19134] math: [powerpc32] lround, lroundf spurious exceptions
799 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
800 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
801 pthread_setaffinity_np
802 [19156] math: [ldbl-128] j0l spurious underflows
803 [19164] nptl: tst-getcpu fails with many possible CPUs
804 [19168] math: math/test-ildoubl and math/test-ldouble failure
805 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
806 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
807 prelink
808 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
809 (FE_NOMASK_ENV) do not clear SSE exceptions
810 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
811 _int_new_arena/reused_arena
812 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
813 [19201] math: dbl-64 remainder incorrect sign of zero result
814 [19205] math: bits/math-finite.h conditions do not match math.h and
815 bits/mathcalls.h
816 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
817 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
818 for C99-based standards
819 [19212] libc: features.h not -Wundef clean
820 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
821 math-only
822 [19214] libc: Family and model identification for AMD CPU's are incorrect.
823 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
824 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
825 disabled
826 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
827 "inexact" exceptions
828 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
829 arguments
830 [19242] libc: strtol incorrect in Turkish locales
831 [19243] malloc: reused_arena can pick an arena on the free list, leading
832 to an assertion failure and reference count corruption
833 [19253] time: tzset() ineffective when temporary TZ did not include DST
834 rules
835 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
836 [19270] math: [hppa] Shared libm missing __isnanl
837 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
838 MAP_HUGETLB and MAP_STACK defines
839 [19313] nptl: Wrong __cpu_mask for x32
840 [19347] libc: grantpt: try to force a specific gid even without pt_chown
841 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
842 [19350] math: [ldbl-128ibm] sinhl spurious overflows
843 [19351] math: [ldbl-128ibm] logl inaccurate near 1
844 [19363] time: x32: times() return value wrongly truncates/sign extends
845 from 32bit
846 [19367] dynamic-link: Improve branch prediction on Silvermont
847 [19369] network: Default domain name not reset by res_ninit when "search"
848 / "domain" entry is removed from resolv.conf
849 [19375] math: powerpc: incorrect results for POWER7 logb with negative
850 subnormals
851 [19385] localedata: bg_BG: time separator should be colon, not comma
852 [19408] libc: linux personality syscall wrapper may erroneously return an
853 error on 32-bit architectures
854 [19415] libc: dladdr returns wrong names on hppa
855 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
856 IBM905, IBM907, and IBM909
857 [19439] math: Unix98 isinf and isnan functions conflict with C++11
858 [19443] build: build failures with -DDEBUG
859 [19451] build: Make check fails on test-double-vlen2
860 [19462] libc: Glibc failed to build with -Os
861 [19465] math: Wrong code with -Os
862 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
863 -Os
864 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
865 CPU's.
866 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
867 [19486] math: S390: Math tests fail with "Exception Inexact set".
868 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
869 [19550] libc: [mips] mmap negative offset handling inconsistent with other
870 architectures
871 [19590] math: Fail to build shared objects that use libmvec.so functions.
872 \f
873 Version 2.22
874
875 * The following bugs are resolved with this release:
876
877 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
878 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
879 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
880 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
881 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
882 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
883 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
884 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
885 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
886 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
887 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
888 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
889 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
890 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
891 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
892 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
893 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
894 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
895 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
896 18657, 18676, 18694, 18696, 18887.
897
898 * Cache information can be queried via sysconf() function on s390 e.g. with
899 _SC_LEVEL1_ICACHE_SIZE as argument.
900
901 * A buffer overflow in gethostbyname_r and related functions performing DNS
902 requests has been fixed. If the NSS functions were called with a
903 misaligned buffer, the buffer length change due to pointer alignment was
904 not taken into account. This could result in application crashes or,
905 potentially arbitrary code execution, using crafted, but syntactically
906 valid DNS responses. (CVE-2015-1781)
907
908 * The time zone file parser has been made more robust against crafted time
909 zone files, avoiding heap buffer overflows related to the processing of
910 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
911 large time zone data files. Overly long time zone specifiers in the TZ
912 variable no longer result in stack overflows and crashes.
913
914 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
915 for LD and GD on x86 and x86-64, has been implemented. You will need
916 binutils-2.24 or later to enable this optimization.
917
918 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
919 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
920 Hat). These updates cause user visible changes, such as the fix for bug
921 17998.
922
923 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
924 and getXXbyYY NSS calls for the same database, causing a denial-of-service
925 condition in some applications.
926
927 * Added vector math library named libmvec with the following vectorized x86_64
928 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
929 pow, powf.
930 The library can be disabled with --disable-mathvec. Use of the functions is
931 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
932 Shared library libmvec.so is linked in as needed when using -lm (no need to
933 specify -lmvec explicitly for not static builds).
934 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
935
936 * A new fmemopen implementation has been added with the goal of POSIX
937 compliance. The new implementation fixes the following long-standing
938 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
939 old implementation is still present for use be by existing binaries.
940
941 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
942 releases. It has been fixed to match 2.19 and older, but binaries built
943 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
944
945 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
946 Contributed by Roland McGrath (Google).
947
948 * The header <regexp.h> is deprecated, and will be removed in a future
949 release. Use of this header will trigger a deprecation warning.
950 Application developers should update their code to use <regex.h> instead.
951
952 This header was formerly part of SUS, but was deprecated in 1994 and
953 removed from the standard in 2001. Also, the glibc implementation
954 leaks memory. See BZ#18681 for more details.
955 \f
956 Version 2.21
957
958 * The following bugs are resolved with this release:
959
960 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
961 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
962 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
963 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
964 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
965 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
966 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
967 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
968 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
969 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
970 17892.
971
972 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
973 too little memory for the to-be-scanned arguments and overflow the
974 allocated buffer. The implementation now correctly computes the required
975 buffer size when using malloc, and switches to malloc from alloca as
976 intended.
977
978 * A new semaphore algorithm has been implemented in generic C code for all
979 machines. Previous custom assembly implementations of semaphore were
980 difficult to reason about or ensure that they were safe. The new version
981 of semaphore supports machines with 64-bit or 32-bit atomic operations.
982 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
983 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
984
985 * Port to Altera Nios II has been contributed by Mentor Graphics.
986
987 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
988 implementations for powerpc64/powerpc64le.
989 Implemented by Adhemerval Zanella (IBM).
990
991 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
992 and powerpc64le. This may improve lock scaling of existing programs on
993 HTM capable systems. The lock elision code is only enabled with
994 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
995 powerpc will issue a transaction abort on every syscall to avoid side
996 effects being visible outside transactions.
997
998 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
999 AArch64. Contributed by ARM Ltd.
1000
1001 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1002
1003 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1004 under certain input conditions resulting in the execution of a shell for
1005 command substitution when the application did not request it. The
1006 implementation now checks WRDE_NOCMD immediately before executing the
1007 shell and returns the error WRDE_CMDSUB as expected.
1008
1009 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1010 processing format strings with a large number of format specifiers.
1011
1012 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1013 infinite loop if the DNS response contained a PTR record of an unexpected
1014 format.
1015
1016 * The minimum GCC version that can be used to build this version of the GNU
1017 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1018 still be used to compile programs using the GNU C Library.
1019
1020 * The GNU C Library is now built with -Werror by default. This can be
1021 disabled by configuring with --disable-werror.
1022
1023 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1024
1025 * The obsolete sigvec function has been removed. This was the original
1026 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1027 programs have been using instead for about 25 years. Of course, ABI
1028 compatibility for old binaries using sigvec remains intact.
1029
1030 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1031 with newer versions of bison.
1032
1033 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1034 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1035 registers overlay two consecutive single-precision registers. MIPS32R2
1036 introduced a new FPU mode (FR=1) where double-precision registers extend the
1037 corresponding single-precision registers which is incompatible with the
1038 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1039 require the use of FR=1 making a transition necessary. New o32 ABI
1040 extensions enable users to migrate over time from the original o32 ABI
1041 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1042 tracks the ABI of any loaded object and verifies that new objects are
1043 compatible. Mode transitions will also be requested as required and
1044 unsupportable objects will be rejected. The ABI checks include both soft and
1045 hard float ABIs for o32, n32 and n64.
1046
1047 GCC 5 with GNU binutils 2.25 onwards:
1048 It is strongly recommended that all o32 system libraries are built using the
1049 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1050 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1051 compiler using --with-fp-32=xx to set this by default.
1052 \f
1053 Version 2.20
1054
1055 * The following bugs are resolved with this release:
1056
1057 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1058 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1059 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1060 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1061 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1062 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1063 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1064 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1065 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1066 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1067 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1068 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1069 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1070 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1071 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1072
1073 * Reverted change of ABI data structures for s390 and s390x:
1074 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1075 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1076 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1077 are preserved pointing straight to the same implementation as the old ones.
1078 Given that, new callers will simply provide a too-big buffer to these
1079 functions. Any applications/libraries out there that embed jmp_buf or
1080 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1081 against 2.19 headers will have to rebuilt again. This is necessary in any
1082 case to revert the breakage in their ABI caused by the glibc change.
1083
1084 * Support for file description locks is added to systems running the
1085 Linux kernel. The standard file locking interfaces are extended to
1086 operate on file descriptions, not file descriptors, via the use of
1087 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1088 are associated with an open file instead of a process.
1089
1090 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1091
1092 * The minimum Linux kernel version that this version of the GNU C Library
1093 can be used with is 2.6.32.
1094
1095 * Running the testsuite no longer terminates as soon as a test fails.
1096 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1097 with PASS or FAIL lines for individual tests. A summary of the results is
1098 printed, including a list of failing lists, and "make check" exits with
1099 error status if there were any unexpected failures. "make check
1100 stop-on-test-failure=y" may be used to keep the old behavior.
1101
1102 * The am33 port, which had not worked for several years, has been removed
1103 from ports.
1104
1105 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1106 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1107 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1108 conflicted with POSIX (support for which was removed in 2.19), the
1109 interfaces those macros enabled remain available when compiling with
1110 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1111 test macros defined.
1112
1113 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1114
1115 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1116 This may improve lock scaling of existing programs on TX capable systems.
1117 The lock elision code is only built with --enable-lock-elision=yes and
1118 then requires a GCC version supporting the TX builtins. With lock elision
1119 default mutexes are elided via __builtin_tbegin, if the cpu supports
1120 transactions. By default lock elision is not enabled and the elision code
1121 is not built.
1122
1123 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1124 copy the path argument. This allowed programs to cause posix_spawn to
1125 deference a dangling pointer, or use an unexpected pathname argument if
1126 the string was modified after the posix_spawn_file_actions_addopen
1127 invocation.
1128
1129 * All supported architectures now use the main glibc sysdeps directory
1130 instead of some being in a separate "ports" directory (which was
1131 distributed separately before glibc 2.17).
1132
1133 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1134 On configurations that support it (all Linux configurations), it's now
1135 used regardless of the --enable-add-ons switch to configure. It is no
1136 longer possible to build such configurations without pthreads support.
1137
1138 * Locale names, including those obtained from environment variables (LANG
1139 and the LC_* variables), are more tightly checked for proper syntax.
1140 setlocale will now fail (with EINVAL) for locale names that are overly
1141 long, contain slashes without starting with a slash, or contain ".." path
1142 components. (CVE-2014-0475) Previously, some valid locale names were
1143 silently replaced with the "C" locale when running in AT_SECURE mode
1144 (e.g., in a SUID program). This is no longer necessary because of the
1145 additional checks.
1146
1147 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1148 with application code using Intel MPX instructions. (With all previous
1149 versions, the MPX register state could be clobbered when making calls
1150 into or out of a shared library.) Note that while the new dynamic
1151 linker is compatible with all known x86 hardware whether or not it
1152 supports Intel MPX, some x86 instruction-set emulators might fail to
1153 handle the new instruction encodings. This is known to affect Valgrind
1154 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1155 release), and might affect other tools that do instruction emulation.
1156
1157 * Support for loadable gconv transliteration modules has been removed.
1158 The support for transliteration modules has been non-functional for
1159 over a decade, and the removal is prompted by security defects. The
1160 normal gconv conversion modules are still supported. Transliteration
1161 with //TRANSLIT is still possible, and the //IGNORE specifier
1162 continues to be supported. (CVE-2014-5119)
1163
1164 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1165 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1166 resulting a denial-of-service security vulnerability in applications which
1167 use functions related to iconv. (CVE-2014-6040)
1168 \f
1169 Version 2.19
1170
1171 * The following bugs are resolved with this release:
1172
1173 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1174 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1175 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1176 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1177 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1178 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1179 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1180 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1181 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1182 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1183 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1184 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1185 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1186 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1187 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1188 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1189 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1190 16431, 16453, 16474, 16506, 16510, 16529
1191
1192 * Slovenian translations for glibc messages have been contributed by the
1193 Translation Project's Slovenian team of translators.
1194
1195 * The public headers no longer use __unused nor __block. This change is to
1196 support compiling programs that are derived from BSD sources and use
1197 __unused internally, and to support compiling with Clang's -fblock
1198 extension which uses __block.
1199
1200 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1201 large collation sequences to optimize multiple passes. This cache
1202 computation may overflow for large collation sequences and may cause a
1203 stack or buffer overflow. This is now fixed to use a slower algorithm
1204 which does not use a cache if there is an integer overflow.
1205
1206 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1207 rules for large collation sequences to optimize multiple passes and falls
1208 back to alloca if malloc fails, resulting in a possible stack overflow.
1209 The implementation now falls back to an uncached collation sequence lookup
1210 if malloc fails.
1211
1212 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1213 initialized for static applications resulting in the security feature
1214 being disabled. The pointer guard is now correctly initialized to a
1215 random value for static applications. Existing static applications need
1216 to be recompiled to take advantage of the fix (bug 15754).
1217
1218 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1219 to the d_name member of struct dirent, or omit the terminating NUL
1220 character. (Bugzilla #14699).
1221
1222 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1223 aligned_alloc functions could allocate too few bytes or corrupt the
1224 heap when passed very large allocation size values (Bugzilla #15855,
1225 #15856, #15857).
1226
1227 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1228 for AF_INET6 has been fixed (Bugzilla #16072).
1229
1230 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1231 pap_CW, quz_PE, the_NP.
1232
1233 * Substantially revised locales: gd_GB, ht_HT
1234
1235 * The LC_ADDRESS field was updated to support country_car for almost all
1236 supported locales.
1237
1238 * ISO 1427 definitions were updated.
1239
1240 * ISO 3166 definitions were updated.
1241
1242 * The localedef utility now supports --big-endian and --little-endian
1243 command-line options to generate locales for a different system from that
1244 for which the C library was built.
1245
1246 * Binary locale files now only depend on the endianness of the system for
1247 which they are generated and not on other properties of that system. As a
1248 consequence, binary files generated with new localedef may be incompatible
1249 with old versions of the GNU C Library, and binary files generated with
1250 old localedef may be incompatible with this version of the GNU C Library,
1251 in the following circumstances:
1252
1253 + Locale files may be incompatible on m68k systems.
1254
1255 + Locale archive files (but not separate files for individual locales) may
1256 be incompatible on systems where plain "char" is signed.
1257
1258 * The configure option --disable-versioning has been removed. Builds with
1259 --disable-versioning had not worked for several years.
1260
1261 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1262 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1263
1264 * SystemTap probes for malloc have been introduced.
1265
1266 * SystemTap probes for slow multiple precision fallback paths of
1267 transcendental functions have been introduced.
1268
1269 * Support for powerpc64le has been added.
1270
1271 * The soft-float powerpc port now supports e500 processors.
1272
1273 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1274
1275 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1276 set of header declarations that are enabled by default, even when other
1277 feature test macros or compiler options such as -std=c99 would otherwise
1278 disable some of those declarations.
1279
1280 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1281 conflict with POSIX. The libbsd-compat library (which was a dummy library
1282 that did nothing) has also been removed.
1283
1284 * Preliminary documentation about Multi-Thread, Async-Signal and
1285 Async-Cancel Safety has been added.
1286
1287 * Change of ABI data structures for s390 and s390x:
1288 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1289 allow for future hardware extensions. All library functions that accept or
1290 return these structures were versioned in order to provide backwards
1291 compatibility with existing code. However, not all cases can be handled
1292 with interface versioning. If an application embeds either structure into
1293 another structure and that structure is passed to another compilation unit
1294 which expects the newer larger-sized object then any access to the new
1295 fields will result in undefined behaviour. Similarly any access to fields
1296 that were moved by the enlarging of the structures will cause undefined
1297 behaviour. To fix the undefined behaviour all objects that are part of the
1298 final in-memory image must agree on the size of structures, and this may
1299 require recompilation.
1300 \f
1301 Version 2.18
1302
1303 * The following bugs are resolved with this release:
1304
1305 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1306 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1307 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1308 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1309 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1310 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1311 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1312 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1313 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1314 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1315 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1316 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1317 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1318 15755, 15759.
1319
1320 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1321 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1322 Distributions can re-enable building and using pt_chown via the new configure
1323 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1324 considerable security risks and should only be used if the distribution
1325 understands and accepts the risks.
1326
1327 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1328 #15078).
1329
1330 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1331 fixed (Bugzilla #15330).
1332
1333 * Add support for calling C++11 thread_local object destructors on thread
1334 and program exit. This needs compiler support for offloading C++11
1335 destructor calls to glibc.
1336
1337 * Improved worst case performance of libm functions with double inputs and
1338 output.
1339
1340 * Support for priority inherited mutexes in pthread condition variables on
1341 non-x86 architectures.
1342
1343 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1344
1345 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1346
1347 * Optimized string functions for ARM. Implemented by Will Newton and
1348 Richard Henderson.
1349
1350 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1351
1352 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1353 Richard Henderson.
1354
1355 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1356 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1357
1358 * Added a benchmark framework to track performance of functions in glibc.
1359
1360 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1361 It is based on draft TS 18661 and currently enabled as a GNU extension.
1362
1363 * On Linux, the clock function now uses the clock_gettime system call
1364 for improved precision, rather than old times system call.
1365
1366 * Added support for version-3 tz format files. This is needed when using
1367 the tz database release 2013e or later, and affects a few unusual cases --
1368 currently only TZ='America/Godthab' for time stamps after 2037.
1369
1370 * Added new API functions pthread_getattr_default_np and
1371 pthread_setattr_default_np to get and set the default pthread
1372 attributes of a process.
1373
1374 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1375 This may improve lock scaling of existing programs on TSX capable systems.
1376 When the --enable-lock-elision=yes parameter is specified at configure
1377 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1378 mutexes.
1379
1380 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1381 availability). Implemented by Ryan S. Arnold.
1382
1383 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1384 \f
1385 Version 2.17
1386
1387 * The following bugs are resolved with this release:
1388
1389 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1390 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1391 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1392 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1393 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1394 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1395 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1396 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1397 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1398 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1399 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1400 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1401 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1402 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1403 14879, 14889, 14893, 14898, 14914.
1404
1405 * Optimization of memcpy for MIPS.
1406
1407 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1408 EMFILE has been fixed (Bugzilla #14889).
1409
1410 * The add-on ports collection is now distributed in the "ports" subdirectory
1411 of the main GNU C Library distribution, rather than separately.
1412
1413 * Port to ARM AArch64 contributed by Linaro.
1414
1415 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1416 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1417 zEnterprise z196.
1418 Implemented by Andreas Krebbel.
1419
1420 * The new function secure_getenv allows secure access to the environment,
1421 returning NULL if running in a SUID/SGID process. This function replaces
1422 the internal function __secure_getenv.
1423
1424 * SystemTap static probes have been added into the dynamic linker.
1425 Implemented by Gary Benson.
1426
1427 * Optimizations of string functions strstr, strcasestr and memmem.
1428 Implemented by Maxim Kuvyrkov.
1429
1430 * The minimum Linux kernel version that this version of the GNU C Library
1431 can be used with is 2.6.16.
1432
1433 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1434 powerpc POWER7. Implemented by Will Schmidt.
1435
1436 * New configure option --disable-nscd builds the C library such that it
1437 never attempts to contact the Name Service Caching Daemon (nscd).
1438 New configure option --disable-build-nscd avoids building nscd itself;
1439 this is the default if --disable-nscd is used.
1440
1441 * Improved support for cross-compilation, including cross-testing and
1442 bootstrap builds without a previously built glibc.
1443
1444 * Several testsuite tests are now able to test multiple IFUNC variants of an
1445 interface, rather than just testing the one that would be chooen by
1446 default.
1447
1448 * New configure options --with-bugurl and --with-pkgversion, for
1449 distributors to use to embed their bug-reporting and package version
1450 information in --help and --version output.
1451
1452 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1453 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1454 allows creation of chroots without the procfs mounted on /proc.
1455
1456 * The `crypt' function now fails if passed salt bytes that violate the
1457 specification for those values. On Linux, the `crypt' function will
1458 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1459 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1460 when the mode is enabled.
1461
1462 * The `clock_*' suite of functions (declared in <time.h>) is now available
1463 directly in the main C library. Previously it was necessary to link with
1464 -lrt to use these functions. This change has the effect that a
1465 single-threaded program that uses a function such as `clock_gettime' (and
1466 is not linked with -lrt) will no longer implicitly load the pthreads
1467 library at runtime and so will not suffer the overheads associated with
1468 multi-thread support in other code such as the C++ runtime library.
1469
1470 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1471 sat_IN, and szl_PL.
1472 \f
1473 Version 2.16
1474
1475 * The following bugs are resolved with this release:
1476
1477 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1478 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1479 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1480 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1481 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1482 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1483 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1484 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1485 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1486 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1487 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1488 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1489 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1490 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1491 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1492 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1493 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1494 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1495 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1496 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1497 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1498 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1499 14277, 14278.
1500
1501 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1502 configuring glibc with:
1503 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1504 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1505 Implemented by H.J. Lu.
1506
1507 * ISO C11 support:
1508
1509 + define static_assert
1510
1511 + do not declare gets
1512
1513 + declare at_quick_exit and quick_exit also for ISO C11
1514
1515 + aligned_alloc. NB: The code is deliberately allows the size parameter
1516 to not be a multiple of the alignment. This is a moronic requirement
1517 in the standard but it is only a requirement on the caller, not the
1518 implementation.
1519
1520 + timespec_get added
1521
1522 + uchar.h support added
1523
1524 + CMPLX, CMPLXF, CMPLXL added
1525
1526 Implemented by Ulrich Drepper.
1527
1528 * Support for the IA-64 has been moved to ports.
1529
1530 * Remove support for anything but ELF binary format
1531
1532 * Checking versions of poll, ppoll added.
1533 Implemented by Ulrich Drepper.
1534
1535 * More generic and 64-bit performance optimizations to math functions.
1536 Implemented by Ulrich Drepper.
1537
1538 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1539 headers and functions available at compile time as they were before
1540 version 2.14. This option will be removed at some time in the future
1541 after the TI-RPC library becomes fully sufficient for the needs of
1542 existing applications.
1543
1544 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1545 Note that glibc is not expected to work with any Linux kernel version
1546 before 2.6.
1547
1548 * New header <sys/auxv.h> and function getauxval allowing easy access to
1549 the AT_* key-value pairs passed from the Linux kernel. The header also
1550 defines the HWCAP_* bits associated with the AT_HWCAP key.
1551
1552 * New locales: mag_IN
1553
1554 * New configure option --enable-systemtap builds SystemTap static probes
1555 into libc for setjmp and longjmp and into libpthread for various operations.
1556 So far the setjmp/longjmp probes and some of the libpthread probes are
1557 provided only for i*86 and x86_64.
1558 Implemented by Roland McGrath and Rayson Ho.
1559
1560 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1561
1562 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1563 and Will Schmidt.
1564
1565 * More optimized functions for SPARC. Implemented by David S. Miller.
1566
1567 * Improved support for cross-compilation, especially bootstrap builds
1568 without a previously built glibc.
1569
1570 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1571 Chris Metcalf from Tilera.
1572
1573 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1574 now supported for ARM processors.
1575
1576 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1577 as the name of the dynamic linker, to distinguish it from the
1578 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1579
1580 * Support for CRIS has been removed from ports.
1581
1582 * A new class of installed header has been documented for low-level
1583 platform-specific functionality. PowerPC added the first instance with a
1584 function to provide time base register access. Contributed by Tulio
1585 Magno Quites Machado Filho.
1586
1587 * ix86 configurations now install header files that are consistent with
1588 what x86-64 configurations install. These same header files can be used
1589 for -m32, -m64, or -mx32 builds.
1590 Contributed by H.J. Lu.
1591
1592 * Math library bug fixes. A thorough audit of all open math library bugs was
1593 conducted by Joseph Myers. Significant progress was made on many math
1594 library bugs resulting in more accurate exceptions and function results.
1595 Many thanks to all those that contributed including Andreas Jaeger for his
1596 patch review and work on the x87 trigonometric instruction issues.
1597
1598 * Timezone data is no longer installed. Timezone-related binaries and scripts
1599 will continue to be installed. Users should obtain their timezone data from
1600 their distribution provider or from the tzdata package at
1601 <ftp://munnari.oz.au/pub/>.
1602 \f
1603 Version 2.15
1604
1605 * The following bugs are resolved with this release:
1606
1607 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1608 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1609 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1610 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1611 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1612 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1613 13484, 13506, 13515, 13523, 13524, 13538, 13540
1614
1615 * New program pldd to list loaded object of a process
1616 Implemented by Ulrich Drepper.
1617
1618 * Add nss_db support back to glibc. No more dependency on Berkeley db
1619 and support for initgroups lookups.
1620 Implemented by Ulrich Drepper.
1621
1622 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1623 Contributed by HJ Lu.
1624
1625 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1626 Contributed by HJ Lu.
1627
1628 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1629 on x86-32 and x86-64.
1630 Contributed by Liubov Dmitrieva.
1631
1632 * Optimized strchr and strrchr for SSE on x86-32.
1633 Contributed by Liubov Dmitrieva.
1634
1635 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1636 for x86-64 and x86-32.
1637 Contributed by Liubov Dmitrieva.
1638
1639 * New interfaces: scandirat, scandirat64
1640 Implemented by Ulrich Drepper.
1641
1642 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
1643 Implemented by Ulrich Drepper.
1644
1645 * nscd now also caches the netgroup database.
1646 Implemented by Ulrich Drepper.
1647
1648 * Integrate libm with gcc's -ffinite-math-only option.
1649 Implemented by Ulrich Drepper.
1650
1651 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
1652 to math functions. Implemented by Ulrich Drepper.
1653
1654 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
1655 Implemented by Ulrich Drepper.
1656
1657 * New Linux interfaces: process_vm_readv, process_vm_writev
1658
1659 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
1660 Implemented by Ulrich Drepper.
1661
1662 * Optimized nearbyint and strcasecmp for PPC.
1663 Implemented by Adhemerval Zanella.
1664
1665 * New locales: bho_IN, unm_US, es_CU, ta_LK
1666 \f
1667 Version 2.14
1668
1669 * The following bugs are resolved with this release:
1670
1671 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
1672 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
1673 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
1674 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
1675 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
1676 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
1677 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
1678 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
1679 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
1680 12795, 12811, 12813, 12814, 12841
1681
1682 * The RPC implementation in libc is obsoleted. Old programs keep working
1683 but new programs cannot be linked with the routines in libc anymore.
1684 Programs in need of RPC functionality must be linked against TI-RPC.
1685 The TI-RPC implementation is IPv6 enabled and there are other benefits.
1686
1687 Visible changes of this change include (obviously) the inability to link
1688 programs using RPC functions without referencing the TI-RPC library and the
1689 removal of the RPC headers from the glibc headers.
1690 Implemented by Ulrich Drepper.
1691
1692 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
1693 syncfs, setns, sendmmsg
1694
1695 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
1696 yue_HK, lij_IT, mhr_RU
1697
1698 * New encodings: CP770, CP771, CP772, CP773, CP774
1699
1700 * New program sotruss to trace calls through PLTs
1701 Implemented by Ulrich Drepper.
1702
1703 * The malloc hook implementation is marked deprecated and will be removed
1704 from the default implementation in the next version. The design never
1705 worked ever since the introduction of threads. Even programs which do
1706 not create threads themselves can use multiple threads created internally.
1707 \f
1708 Version 2.13
1709
1710 * The following bugs are resolved with this release:
1711
1712 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
1713 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
1714 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
1715 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
1716 12378, 12394, 12397
1717
1718 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
1719
1720 * POWER7 optimizations: memset, memcmp, strncmp
1721
1722 * New optimized string functions for x86-64: strnlen (SSE2),
1723 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
1724 Implemented by Ulrich Drepper.
1725 \f
1726 Version 2.12
1727
1728 * The following bugs are resolved with this release:
1729
1730 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
1731 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
1732 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
1733 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
1734 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
1735 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
1736 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
1737 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
1738
1739 * New interfaces: pthread_getname_np, pthread_setname_np
1740
1741 * New Linux interface: recvmmsg
1742
1743 * STT_GNU_IFUNC implemented for Sparc by David Miller.
1744
1745 * The dynamic linker now recognizes supported ABI versions from the
1746 EI_ABIVERSION field in the ELF header.
1747 Implemented by Ulrich Drepper.
1748
1749 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
1750 will not be used to fill in password fields in the passwd.byname replies.
1751 Instead it is used to synthesize the shadow.byname table, should it be
1752 missing. This is a useful mode in some installations involving Solaris.
1753 Implemented by Ulrich Drepper.
1754
1755 * New locales: kok_IN, sq_MK, cv_RU
1756 \f
1757 Version 2.11
1758
1759 * The following bugs are resolved with this release:
1760
1761 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
1762 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
1763 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
1764 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
1765 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
1766 10789, 10817, 10823, 10840, 10847
1767
1768 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
1769 mkostemps64
1770 Implemented by Ulrich Drepper.
1771
1772 * Checking version of longjmp added that fails if an uninitialized stack
1773 frame would be created. Implemented by Ulrich Drepper.
1774
1775 * STT_GNU_IFUNC is now supported in static executables.
1776 Implemented by H.J. Lu.
1777
1778 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
1779
1780 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
1781 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
1782 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
1783 Contributed by H.J. Lu.
1784
1785 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
1786 Implemented by Ulrich Drepper.
1787
1788 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
1789 strstr, strcasestr.
1790 Contributed by H.J. Lu.
1791
1792 * Support for fma instruction in AVX on x86-64.
1793 Implemented by H.J. Lu and Ulrich Drepper.
1794
1795 * AVX support in x86-64 auditing support in ld.so.
1796 Implemented by H.J. Lu.
1797
1798 * STB_GNU_UNIQUE support added.
1799 Implemented by Ulrich Drepper.
1800
1801 * Implement second fallback mode for optimized DNS lookup for even more
1802 broken environments. If two requests from the same port are not
1803 handled correctly close the socket and open a new one before sending
1804 the second request. The 'single-request-reopen' option in /etc/resolv.conf
1805 can be used to select this mode right away, instead of rediscovering the
1806 necessity is every process again.
1807 Implemented by Ulrich Drepper.
1808
1809 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
1810 Implemented by Adam Tkac.
1811
1812 * Optimized iconv conversions for S390x.
1813 Implemented by Andreas Krebbel.
1814
1815 * Using condvars with PI mutexes is now more efficient due to kernel
1816 support for requeueing to PI futexes. NPTL support added for x86-64.
1817 Implemented by Ulrich Drepper.
1818
1819 * New locales: ps_AF, my_MM
1820 \f
1821 Version 2.10
1822
1823 * The following bugs are resolved with this release:
1824
1825 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
1826 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
1827 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
1828 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
1829 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
1830
1831 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
1832 Implemented by Ulrich Drepper.
1833
1834 * New Linux interfaces: accept4, fallocate, fallocate64.
1835 Implemented by Ulrich Drepper.
1836
1837 * Correct declarations of string function when used in C++ code. This
1838 could lead to compile errors for invalid C++ code.
1839
1840 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
1841 now in POSIX.
1842
1843 * New POSIX 2008 interface: psiginfo
1844 Implemented by Ulrich Drepper.
1845
1846 * New ISO C++1x interfaces: quick_exit, at_quick_exit
1847 Implemented by Ulrich Drepper.
1848
1849 * Support for selecting between multiple function definitions at runtime
1850 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
1851
1852 * The libcrypt library can now use the hash function implementations in
1853 NSS. Implemented by Ulrich Drepper.
1854
1855 * The malloc implementation can be compiled to be less memory efficient
1856 but higher performing in multi-threaded programs.
1857 Implemented by Ulrich Drepper.
1858
1859 * New locales: nan_TW@latin, ks_IN
1860
1861 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
1862 Implemented by Ulrich Drepper.
1863
1864 * Extended printf hook support. It is possible to use user-defined types
1865 and extend existing format specifiers.
1866 Implemented by Ulrich Drepper.
1867
1868 * Handling for group shadow files has been added.
1869 Implemented by Ulrich Drepper.
1870
1871 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
1872 servers (the case, e.g., for some people using the built-in DNS
1873 server in ADSL modems/routers). There is a once-per-process timeout
1874 in case of a broken server. To avoid it, users can run nscd or put
1875 'options single-request' in /etc/resolv.conf.
1876 Implemented by Ulrich Drepper.
1877 \f
1878 Version 2.9
1879
1880 * The following bugs are resolved with this release:
1881
1882 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
1883 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
1884 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
1885 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
1886 6968, 6974, 6980, 6995, 7008, 7009, 7029
1887
1888 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
1889 up at the same time. Implemented by Ulrich Drepper.
1890
1891 * TLS descriptors for LD and GD on x86 and x86-64.
1892 Implemented by Alexandre Oliva.
1893
1894 * getaddrinfo now handles DCCP and UDPlite.
1895 Implemented by Ulrich Drepper.
1896
1897 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
1898 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
1899 Implemented by Ulrich Drepper.
1900
1901 * New implementation of memmem, strstr, and strcasestr which is O(n).
1902 Implemented by Eric Blake.
1903
1904 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
1905
1906 * Implement "e" option for popen to open file descriptor with the
1907 close-on-exec flag set. Implemented by Ulrich Drepper.
1908
1909 * Implement "b" mode for fmemopen. In this mode writes writes don't
1910 implicitly add a NUL byte and seeks from the end of the buffer really
1911 use the buffer end, not the string length as the basis.
1912 Implemented by Ulrich Drepper.
1913
1914 * Many functions, exported and internal, now atomically set the close-on-exec
1915 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
1916
1917 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
1918 Sinhala)
1919 Implemented by Pravin Satpute.
1920
1921 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
1922 \f
1923 Version 2.8
1924
1925 * The following bugs are resolved with this release:
1926
1927 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
1928 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
1929 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
1930 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
1931 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
1932 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
1933 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
1934
1935 * New locales: bo_CN, bo_IN, shs_CA.
1936
1937 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
1938
1939 * Sorting rules for some Indian languages (Devanagari and Gujarati).
1940 Implemented by Pravin Satpute.
1941
1942 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
1943
1944 * nscd caches now all timeouts for DNS entries
1945 Implemented by Ulrich Drepper.
1946
1947 * nscd is more efficient and wakes up less often.
1948 Implemented by Ulrich Drepper.
1949
1950 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
1951 vdprintf, and obstack_vprintf.
1952 Implemented by Jakub Jelinek.
1953
1954 * Faster memset for x86-64.
1955 Implemented by Harsha Jagasia and H.J. Lu.
1956
1957 * Faster memcpy on x86.
1958 Implemented by Ulrich Drepper.
1959
1960 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
1961 Implemented by Ulrich Drepper.
1962
1963 * Faster sqrt and sqrtf implemention for some PPC variants.
1964 Implemented by Steven Munroe.
1965 \f
1966 Version 2.7
1967
1968 * The following bugs are resolved with this release:
1969
1970 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
1971 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
1972 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
1973 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
1974 5071, 5103, 5104, 5112, 5113, 5184, 5186
1975
1976 * More checking functions: fread, fread_unlocked, open*, mq_open.
1977 Implemented by Jakub Jelinek and Ulrich Drepper.
1978
1979 * Extend fortification to C++. Implemented by Jakub Jelinek.
1980
1981 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
1982 by not recognizing 'a' as a modifier when those specs are requested.
1983 Implemented by Jakub Jelinek.
1984
1985 * PPC optimizations to math and string functions.
1986 Implemented by Steven Munroe.
1987
1988 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
1989 options to be passed. Implemented by Ulrich Drepper.
1990
1991 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
1992
1993 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
1994 Implemented by Ulrich Drepper.
1995
1996 * Handle private futexes in the NPTL implementation.
1997 Implemented by Jakub Jelinek and Ulrich Drepper.
1998
1999 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2000 Implemented by Roland McGrath and Ulrich Drepper.
2001
2002 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2003
2004 * SHA-256 and SHA-512 based password encryption.
2005 Implemented by Ulrich Drepper.
2006
2007 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2008 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2009 yo_NG.
2010
2011 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2012 Implemented by Ulrich Drepper.
2013 \f
2014 Version 2.6
2015
2016 * The following bugs are resolved with this release:
2017
2018 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2019 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2020 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2021 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2022 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2023 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2024 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2025 4702, 4858
2026
2027 * New Linux interfaces: epoll_pwait, sched_getcpu.
2028
2029 * New generic interfaces: strerror_l.
2030
2031 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2032 \f
2033 Version 2.5
2034
2035 * The following bugs are resolved with this release:
2036
2037 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2038 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2039 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2040 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2041 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2042 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2043 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2044 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2045 3225, 3251, 3252, 3253, 3273
2046
2047 * For Linux, the sorting of addresses returned by getaddrinfo now also
2048 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2049 Implemented by Ulrich Drepper.
2050
2051 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2052 Implemented by Ulrich Drepper.
2053
2054 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2055
2056 * New iconv module for MIK. Contributed by Alexander Shopov.
2057
2058 * For sites with broken group and/or passwd database, the auto-propagate
2059 option of nscd can prevent creating ID lookup entries from the results
2060 of a name lookup and vice versa. This usually is no problem but some
2061 site might have problems with the default behavior.
2062 Implemented by Ulrich Drepper.
2063
2064 * Iterating over entire database in NIS can be slow. With the
2065 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2066 to trade time for memory. The entire database will be read at once.
2067 Implemented by Ulrich Drepper.
2068
2069 * The interfaces introduced in RFC 3542 have been implemented by
2070 Ulrich Drepper.
2071
2072 * Support for the new ELF hash table format was added by Ulrich Drepper.
2073
2074 * Support for priority inheritance mutexes added by Jakub Jelinek and
2075 Ulrich Drepper.
2076
2077 * Support for priority protected mutexes added by Jakub Jelinek.
2078
2079 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2080 or_IN, csb_PL, fy_NL, sr_ME.
2081 \f
2082 Version 2.4
2083
2084 * More overflow detection functions.
2085
2086 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2087 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2088
2089 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2090 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2091 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2092 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2093 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2094 by Masahide Washizawa.
2095
2096 * It is now possible to install audit libraries for the dynamic linker using
2097 LD_AUDIT. Implemented by Ulrich Drepper.
2098
2099 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2100 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2101 For a libc and libpthread that works well on Linux 2.4 kernels, we
2102 recommend using the stable 2.3 branch.
2103
2104 * The new function open_wmemstream acts like open_memstream,
2105 but uses a wchar_t wide character buffer.
2106
2107 * The new function ppoll is an improved version of poll, similar to pselect.
2108
2109 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2110 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2111 renameat, unlinkat, linkat, symlinkat, readlinkat.
2112
2113 * New Linux kernel system calls: unshare,
2114 inotify_init, inotify_add_watch, inotify_rm_watch.
2115
2116 * The euidaccess function is now also known by the alias eaccess,
2117 for compatibility with some other systems.
2118
2119 * Timezone data updated to 2006b version.
2120 \f
2121 Version 2.3.6
2122
2123 * The following bugs are resolved with this release:
2124
2125 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2126 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2127 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2128 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2129 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2130 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2131
2132 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2133
2134 * As of this release, GCC 4 can be used to compile the C Library.
2135
2136 * Timezone data updated to 2005m version.
2137 \f
2138 Version 2.3.5
2139
2140 * The following bugs are resolved with this release:
2141
2142 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2143 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2144 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2145 777, 787, 821, 822, 823, 825
2146
2147 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2148 \f
2149 Version 2.3.4
2150
2151 * Support for RFC 3678. Real implementations exist only for Linux so far.
2152 Implemented by Ulrich Drepper.
2153
2154 * nscd can now cache entries persistently. Expiring entries are reloaded.
2155 For speedups the cache can be shared in memory with client processes.
2156 Implemented by Ulrich Drepper.
2157
2158 * nscd can now perform SELinux checks.
2159 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2160
2161 * getaddrinfo queries are now cached. Canonical name lookup is performed
2162 efficiently.
2163 Implemented by Ulrich Drepper.
2164
2165 * The nothrow function attribute is used when headers are used by gcc when
2166 compiling C code. This can avoid significant amounts of exception
2167 handling data.
2168
2169 * The malloc functions perform more error checking and are stricter when
2170 it comes to reacting on errors. The default action is to terminate
2171 the process after showing an error message. Implemented by Ulrich Drepper.
2172
2173 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2174 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2175
2176 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2177 namespaces using the new function dlmopen(). This feature is of course,
2178 like most other dynamic loading functionality, not available in statically
2179 linked applications. Implemented by Ulrich Drepper.
2180
2181 * Low-overhead boundary checking variants of string and some stdio functions
2182 were added. These are to be used in conjunction with a gcc patch by
2183 Jakub Jelinek which adds calls to these functions if possible.
2184 Implemented by Jakub Jelinek and Ulrich Drepper.
2185
2186 * Old code for several operating systems and machine architectures that
2187 have not been in working condition in a long time have been removed from
2188 the main source tree maintained by the GNU C Library's maintainers.
2189 These files are now reside in the separate `ports' source module
2190 that is usable as an add-on when building the library.
2191 \f
2192 Version 2.3.3
2193
2194 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2195 interrogate the dynamic linker, compatible with the Solaris interface.
2196
2197 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2198 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2199
2200 * getifaddrs now uses the netlink interface on Linux to get its information.
2201 Implemented by Thorsten Kukuk.
2202
2203 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2204 Implemented by Ulrich Drepper.
2205
2206 * support for non-executable stacks on x86 has been added. Changes mostly
2207 by Roland McGrath.
2208
2209 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2210 and Ulrich Drepper.
2211
2212 * getaddrinfo now performs destination address selection according to
2213 RFC 3484.
2214 \f
2215 Version 2.3.2
2216
2217 * Thread-safe interfaces for many functions that access locale data
2218 were added in version 2.3, but these features were omitted from NEWS.
2219 Many functions have variants with an `_l' suffix that take a `locale_t'
2220 object as a parameter rather than consulting the current locale.
2221 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2222 create and maintain `locale_t' objects. Additionally, the new function
2223 `uselocale' sets "the current locale" (as used by functions not so
2224 parameterized) set for an individual thread. These features were added
2225 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2226
2227 * The functions getresuid, getresgid, setresuid, and setresgid, which
2228 have long been available on Linux, are now declared in <unistd.h>
2229 and are now also available on the Hurd.
2230
2231 * ELF thread-local storage support (TLS) now works on x86-64.
2232
2233 * The new dynamic string token $LIB is expanded in shared library names.
2234 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2235
2236 * Aldy Hernandez contributed complete software floating point support for
2237 PowerPC machines with no FPU.
2238
2239 * fexecve is implemented on Linux.
2240
2241 * The `btowc' function should work at least twice as fast due to
2242 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2243
2244 * With approriate thread add-ons cancelable functions are now implemented
2245 in libc.so as well. No need to call the function in libpthread. This
2246 change allowed to finally disable the incorrect and expensive handling
2247 of weak definition in ld.so.
2248
2249 * Yet more PLT entries in libc.so have been removed. We finally arrived
2250 at the bare minimum. Startup times improved appropriately.
2251
2252 * Support for the new Linux/x86 system call interface was added. The
2253 AT_SYSINFO auxiliary vector entry is recognized and handled.
2254 \f
2255 Version 2.3
2256
2257 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2258 charsets.
2259
2260 * iconv (the program and the interface) now accepts empty names (excluding
2261 options like //TRANSLIT) to mean "use charset of current locale".
2262
2263 * localedef can now transliterate characters in strings which are not in
2264 the provided charmap. The information from the input locale is used.
2265
2266 * Prelinking support was added for ELF targets. This requires additional
2267 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2268
2269 * Read-only stdio streams now use mmap to speed up operation by eliminating
2270 copying and buffer underflows. To use add 'm' to the mode string of
2271 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2272
2273 * The malloc functions were completely rewritten by Wolfram Gloger based
2274 on Doug Lea's malloc-2.7.0.c.
2275
2276 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2277 implementation of regex.
2278
2279 * Bruno Haible upgraded the iconv and locale implementation to support
2280 Unicode 3.2.
2281
2282 * Contents of the LC_* and LANG environment variables in the CEN style are
2283 not recognized anymore. It never was used. Change by Ulrich Drepper.
2284
2285 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2286 thread-local storage (TLS) ABI on some platforms.
2287 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2288
2289 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2290 EUC-JISX0213, and TSCII.
2291
2292 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2293 BSD-compatible interface for getting all network interface addresses.
2294 Implementation for IPv4 by Roland McGrath.
2295
2296 * Loading of locale data is faster due to the introduction of a locale
2297 archive. Implemented by Roland McGrath and Ulrich Drepper.
2298
2299 * Startup times are significantly reduced by not using exported functions
2300 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2301 and Ulrich Drepper.
2302
2303 * Steven Munroe contributed a port to PowerPC64/Linux.
2304 \f
2305 Version 2.2.6
2306
2307 * The Hurd now uses the GNU libio implementation of stdio rather than the
2308 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2309
2310 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2311 with the same behavior as the Linux system call of the same name.
2312 \f
2313 Version 2.2.5
2314
2315 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2316 128-bit long double format.
2317
2318 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2319 IBM1161, and IBM1162 charsets.
2320
2321 * Andreas Jaeger contributed a port to x86-64/Linux.
2322
2323 * Peter Bruin contributed a port to PowerPC/Hurd.
2324
2325 * libc functions using I/O streams now can handle wide-oriented streams
2326 as well.
2327
2328 * optimizations in the dynamic linker. Binaries created by recent binutils
2329 versions start up quicker due to reduced time spent on relocations.
2330
2331 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2332 \f
2333 Version 2.2.4
2334
2335 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2336 asinh, atanh, j0 for the 128-bit long double format.
2337
2338 * Bruno Haible updated all the code handling Unicode in some form to
2339 support Unicode 3.1.
2340
2341 * Speed of regex for single-byte locales is back to previous levels.
2342 Patch by Isamu Hasegawa.
2343
2344 * Alpha, SPARC, and IA-64 now also using floating stacks.
2345
2346 * Startup time of internationalized applications greatly improved through
2347 iconv cache. Use iconvconfig to generate the cache file.
2348 Contributed by Ulrich Drepper.
2349
2350 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2351 severe performance problems. Patch by David Mosberger.
2352
2353 * The Hurd port got a lot more functionality like AIO, various stdio
2354 extensions, etc. Mainly done by Roland McGrath.
2355
2356 * mtrace can now lookup symbols in shared libraries.
2357 \f
2358 Version 2.2.3
2359
2360 * Intel's IA-64 math library is largely integrated. It provides fast and
2361 accurate implementatations for most basic and standard math functions
2362 in float, double, and long double format.
2363
2364 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2365 and asin for the 96-bit long double format and asin, log, tan for the
2366 128-bit long double format.
2367
2368 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2369 The basic double functions exist today. Contributed by Abraham Ziv
2370 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2371 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2372
2373 * An asynchronous name lookup library was added. The interface is designed
2374 after POSIX AIO. The proposal was circulated beforehand to get comments.
2375 No negative ones came in. Implemented by Ulrich Drepper.
2376
2377 * Port to S390/64bit contributed by Martin Schwidefsky
2378 <schwidefsky@de.ibm.com>.
2379
2380 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2381 of functions for Linux/IA-64.
2382
2383 * The RPC code is now thread safe. Threads can now use the same service
2384 of different services at the same time. Patch by Eric Norum
2385 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2386
2387 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2388 family of functions for Linux/S390.
2389
2390 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2391 of functions for Linux/x86.
2392
2393 * Port to Linux/CRIS contributed by Axis Communications.
2394 \f
2395 Version 2.2.2
2396
2397 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2398 we can now check for namespace violations and missing declarations. The
2399 result is that almost all headers are now Unix-compliant (as defined in
2400 the upcoming XPG6). The negative side is that some programs might need
2401 corrections, too, if they depend on the incorrect form of the headers in
2402 previous versions which defined too many symbols and included too many
2403 other headers.
2404
2405 * regex now handles multibyte character sets correctly.
2406 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2407
2408 * iconv (the program) does now conform to the upcoming XPG6 and handles
2409 charmaps. Instead of the charset names the path of charmaps can be
2410 provided and the conversion happens based on this data.
2411 Contributed by Ulrich Drepper.
2412
2413 * The locale program now provides detailed information about the installed
2414 locales. While
2415
2416 locale -a
2417
2418 only lists the names of the supported locales
2419
2420 locale -a --verbose
2421
2422 provides details such as country, language, and codeset name.
2423 Contributed by Ulrich Drepper.
2424 \f
2425 Version 2.2.1
2426
2427 * The gencat program now parses the input file according to the charset
2428 selected by the LC_CTYPE category. This is important for stateful
2429 character sets. To make generating catalogs easier there is a way
2430 to overwrite the charset selected by the locale: before the first
2431 message or $ quote line the catalog can contain a line like
2432
2433 $ codeset=ISO-8859-2
2434
2435 to select the charset (ISO-8859-2 in this case).
2436
2437 Implemented by Shinya Hanataka and Ulrich Drepper.
2438
2439 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2440 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2441 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2442
2443 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2444 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2445
2446 * The LANGUAGE environment variable is now ignored unless the locale is
2447 changed from the default "C" locale.
2448
2449 * The usual bug fixes.
2450 \f
2451 Version 2.2
2452
2453 * Greg McGary added runtime support for bounds checking using gcc's
2454 new -fbounded-pointers option. ix86 target is complete. PowerPC
2455 is in progress.
2456
2457 * Thorsten Kukuk added secure mode support to nscd.
2458
2459 * The Berkeley db libraries have been removed.
2460
2461 Related, the nss_db module is now in a separate package since it
2462 obviously requires a database library being available.
2463
2464 * Wide character I/O streams implemented by Ulrich Drepper.
2465
2466 * Functions from the extended socket API added by Ulrich Drepper.
2467
2468 * Functions feenableexcept and fedisableexcept to control the
2469 behaviour of individual exceptions have been added by Andreas Jaeger.
2470
2471 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2472
2473 * The resolver code has been updated from bind 8.2.3-T5B which supports
2474 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2475 and Mark Kettenis.
2476
2477 This change could in some situations effect backward compatibility. Since
2478 now `_res' is a thread-local instead of a global variable, modifying it
2479 in one thread does not have any effect in other threads.
2480
2481 The resolver library was also extended to allow IPv6 as the transport
2482 protocol for the requests. This work was done by Stig Venaas.
2483
2484 * Compatibility code for K&R C compilers has been removed from the
2485 header files. A ISO C compiler is needed to use the library
2486 (conforming to either C89 or C99 standard).
2487
2488 * Complete rewrite of the localedef program to support multibyte character
2489 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2490 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2491 Implemented by Ulrich Drepper.
2492
2493 Bruno Haible significantly improved the generation and use of the data
2494 structures for the wide character tables.
2495
2496 * Plural handling in gettext implemented by Ulrich Drepper.
2497
2498 * The utmp daemon has been removed.
2499
2500 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2501
2502 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2503 and Yutaka Niibe.
2504
2505 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2506
2507 * POSIX barriers implemented by Kaz Kylheku.
2508
2509 * POSIX spawn function family implemented by Ulrich Drepper.
2510
2511 * POSIX spinlocks are now available.
2512
2513 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2514
2515 * the configure option --enable-kernel=X.Y.Z allows to strip out
2516 compatibility for kernel versions before X.Y.Z. This is currently only
2517 implemented for Linux.
2518
2519 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2520 field sin6_scope_id. This means that all programs using IPv6 should be
2521 recompiled. Don't expect binary compatibility with previous glibc
2522 versions.
2523
2524 * various conversion modules for IBM character sets contributed by
2525 Masahide Washizawa.
2526
2527 * IA-64 port by Jes Sorensen and HJ Lu.
2528 \f
2529 Version 2.1.3
2530
2531 * bug fixes
2532
2533 \f
2534 Version 2.1.2
2535
2536 * bug fixes
2537
2538 \f
2539 Version 2.1.1
2540
2541 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2542
2543 * New xdr functions are added; some rpc functions are now 64bit clean.
2544
2545 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2546
2547 * Fixed known incompatibilities with glibc 2.0.
2548
2549 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2550
2551 * Optimized a number of functions (especially the ELF dynamic loader).
2552
2553 * Update timezone data files.
2554
2555 * lots of charmaps corrections
2556
2557 * some new locale definitions and charmaps
2558
2559 \f
2560 Version 2.1
2561
2562 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2563 conform to POSIX member type requirements. Symbol versions have been
2564 adjusted as needed within the library, and for direct use by applications,
2565 but there is potential for problems if third-party libraries use
2566 struct timeval as part of their interface. This does not present
2567 a problem for X and other "essential" system libraries.
2568
2569 * An additional locale model to support C++ Standard Library locale
2570 model and probably more was implemented by Ulrich Drepper.
2571
2572 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2573 symbol level.
2574
2575 * Miles Bader provided the `argp' function family to support hierachical
2576 command line argument parsing, layered on top of getopt.
2577
2578 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2579
2580 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2581 numbers.
2582
2583 * scanf recognizes the %a and %A format for scanning floating point numbers.
2584
2585 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2586 information and interfaces for the available integer types.
2587
2588 * about 130 new math functions were added to implement the ISO C9x math
2589 library.
2590
2591 * the new header <complex.h> contains definitions of the complex math
2592 functions from ISO C 9X.
2593
2594 * the new header <tgmath.h> defines generic macros to use complex or
2595 real valued functions.
2596
2597 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2598
2599 * Andreas Jaeger provided a test suite for the math library.
2600
2601 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2602
2603 * Ulrich Drepper added character set conversion functions (iconv).
2604
2605 * Optimized string functions have been added.
2606
2607 * The localedata addon is now part of glibc.
2608
2609 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2610
2611 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2612 daemon for NSS (nscd).
2613
2614 Missing a better place here are some numbers on improvements. Under
2615 Linux 2.1.125 un-tar-ing the kernel sources takes
2616
2617 user system wall
2618
2619 using local files 12.19s 6.88s 22.91s
2620
2621 using NIS 13.92s 8.91s 26.34s
2622
2623 using NIS & nscd 10.37s 7.34s 25.30s
2624
2625 using NIS+ 27.57s 30.37s 640.46s
2626
2627 using NIS+ & nscd 10.25s 7.83s 26.51s
2628
2629 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2630
2631 Keep in mind that non-namelookup related operations dominate above times.
2632 It was just a common complain that using NIS+ unpacking the kernel is
2633 horribly slow.
2634
2635 [1] The old nscd implementation is not available anymore since it was
2636 distributed with glibc up to version 2.0.98 and thus is now replaced.
2637
2638 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2639
2640 * Mark Kettenis provided a Hesiod NSS module.
2641
2642 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
2643 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
2644
2645 * rcmd can now handle netgroups (Dick Streefland).
2646
2647 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
2648 Bambrough.
2649
2650 * Support for the IPv6 protocol has been added to the socket API, as per the
2651 latest draft standards.
2652
2653 * Support for Linux 2.2 has been added.
2654
2655 * Interface changes relative to the latest 2.0.x release:
2656 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2657 addseverity NEW: Unix98
2658 alphasort64 NEW: LFS
2659 argp_err_exit_status NEW: argp, GNU ext
2660 argp_error NEW: argp, GNU ext
2661 argp_failure NEW: argp, GNU ext
2662 argp_help NEW: argp, GNU ext
2663 argp_parse NEW: argp, GNU ext
2664 argp_program_bug_address NEW: argp, GNU ext
2665 argp_program_version NEW: argp, GNU ext
2666 argp_program_version_hook NEW: argp, GNU ext
2667 argp_state_help NEW: argp, GNU ext
2668 argp_usage NEW: argp, GNU ext
2669 authdes_create NEW: Secure RPC
2670 authdes_getucred NEW: Secure RPC
2671 authdes_pk_create NEW: Secure RPC
2672 backtrace NEW: GNU ext.
2673 backtrace_symbols NEW: GNU ext.
2674 backtrace_symbols_fd NEW: GNU ext.
2675 cacos NEW: ISO C 9x
2676 cacosf NEW: ISO C 9x
2677 cacosh NEW: ISO C 9x
2678 cacoshf NEW: ISO C 9x
2679 cacoshl NEW: ISO C 9x
2680 cacosl NEW: ISO C 9x
2681 capget NEW: kernel
2682 capset NEW: kernel
2683 carg NEW: ISO C 9x
2684 cargf NEW: ISO C 9x
2685 cargl NEW: ISO C 9x
2686 casin NEW: ISO C 9x
2687 casinf NEW: ISO C 9x
2688 casinh NEW: ISO C 9x
2689 casinhf NEW: ISO C 9x
2690 casinhl NEW: ISO C 9x
2691 casinl NEW: ISO C 9x
2692 catan NEW: ISO C 9x
2693 catanf NEW: ISO C 9x
2694 catanh NEW: ISO C 9x
2695 catanhf NEW: ISO C 9x
2696 catanhl NEW: ISO C 9x
2697 catanl NEW: ISO C 9x
2698 cbc_crypt NEW: Secure RPC
2699 ccos NEW: ISO C 9x
2700 ccosf NEW: ISO C 9x
2701 ccosh NEW: ISO C 9x
2702 ccoshf NEW: ISO C 9x
2703 ccoshl NEW: ISO C 9x
2704 ccosl NEW: ISO C 9x
2705 cexp NEW: ISO C 9x
2706 cexpf NEW: ISO C 9x
2707 cexpl NEW: ISO C 9x
2708 cimag NEW: ISO C 9x
2709 cimagf NEW: ISO C 9x
2710 cimagl NEW: ISO C 9x
2711 clearerr_locked REMOVED
2712 clntunix_create NEW: sunrpc ext
2713 clog NEW: ISO C 9x
2714 clog10 NEW: ISO C 9x
2715 clog10f NEW: ISO C 9x
2716 clog10l NEW: ISO C 9x
2717 clogf NEW: ISO C 9x
2718 clogl NEW: ISO C 9x
2719 conj NEW: ISO C 9x
2720 conjf NEW: ISO C 9x
2721 conjl NEW: ISO C 9x
2722 cpow NEW: ISO C 9x
2723 cpowf NEW: ISO C 9x
2724 cpowl NEW: ISO C 9x
2725 cproj NEW: ISO C 9x
2726 cprojf NEW: ISO C 9x
2727 cprojl NEW: ISO C 9x
2728 creal NEW: ISO C 9x
2729 crealf NEW: ISO C 9x
2730 creall NEW: ISO C 9x
2731 creat64 NEW: LFS
2732 csin NEW: ISO C 9x
2733 csinf NEW: ISO C 9x
2734 csinh NEW: ISO C 9x
2735 csinhf NEW: ISO C 9x
2736 csinhl NEW: ISO C 9x
2737 csinl NEW: ISO C 9x
2738 csqrt NEW: ISO C 9x
2739 csqrtf NEW: ISO C 9x
2740 csqrtl NEW: ISO C 9x
2741 ctan NEW: ISO C 9x
2742 ctanf NEW: ISO C 9x
2743 ctanh NEW: ISO C 9x
2744 ctanhf NEW: ISO C 9x
2745 ctanhl NEW: ISO C 9x
2746 ctanl NEW: ISO C 9x
2747 des_setparity NEW: Secure RPC
2748 ecb_crypt NEW: Secure RPC
2749 endutxent NEW: Unix98
2750 exp10 NEW: ISO C 9x
2751 exp10f NEW: ISO C 9x
2752 exp10l NEW: ISO C 9x
2753 exp2 NEW: ISO C 9x
2754 exp2f NEW: ISO C 9x
2755 exp2l NEW: ISO C 9x
2756 fattach NEW: STREAMS
2757 fdetach NEW: STREAMS
2758 fdim NEW: ISO C 9x
2759 fdimf NEW: ISO C 9x
2760 fdiml NEW: ISO C 9x
2761 feclearexcept NEW: ISO C 9x
2762 fegetenv NEW: ISO C 9x
2763 fegetexceptflag NEW: ISO C 9x
2764 fegetround NEW: ISO C 9x
2765 feholdexcept NEW: ISO C 9x
2766 feof_locked REMOVED
2767 feraiseexcept NEW: ISO C 9x
2768 ferror_locked REMOVED
2769 fesetenv NEW: ISO C 9x
2770 fesetexceptflag NEW: ISO C 9x
2771 fesetround NEW: ISO C 9x
2772 fetestexcept NEW: ISO C 9x
2773 feupdateenv NEW: ISO C 9x
2774 fflush_locked REMOVED
2775 ffsl NEW: GNU ext.
2776 ffsll NEW: GNU ext.
2777 fgetpos64 NEW: LFS
2778 fgets_unlocked NEW: GNU ext.
2779 fileno_locked REMOVED
2780 fma NEW: ISO C 9x
2781 fmaf NEW: ISO C 9x
2782 fmal NEW: ISO C 9x
2783 fmax NEW: ISO C 9x
2784 fmaxf NEW: ISO C 9x
2785 fmaxl NEW: ISO C 9x
2786 fmin NEW: ISO C 9x
2787 fminf NEW: ISO C 9x
2788 fminl NEW: ISO C 9x
2789 fmtmsg NEW: Unix98
2790 fopen64 NEW: LFS
2791 fputc_locked REMOVED
2792 fputs_unlocked NEW: GNU ext.
2793 fread_unlocked NEW: GNU ext.
2794 freopen64 NEW: LFS
2795 fseeko NEW: Unix98
2796 fsetpos64 NEW: LFS
2797 fstatfs64 NEW: LFS
2798 fstatvfs NEW: Unix98
2799 fstatvfs64 NEW: LFS
2800 ftello NEW: Unix98
2801 ftello64 NEW: LFS
2802 ftruncate64 NEW: LFS
2803 ftw64 NEW: LFS
2804 fwrite_unlocked NEW: GNU ext.
2805 gai_strerror NEW: IPv6
2806 gamma_r REMOVED
2807 gammaf_r REMOVED
2808 gammal_r REMOVED
2809 getchar_locked REMOVED
2810 getdate NEW: Unix98
2811 getdate_err NEW: Unix98
2812 getdate_r NEW: GNU ext.
2813 getmsg NEW: STREAMS
2814 getnameinfo NEW: IPv6
2815 getnetname NEW: Secure RPC
2816 getpmsg NEW: STREAMS
2817 getpt NEW: Unix98 PTY
2818 getrlimit64 NEW: LFS
2819 getutxent NEW: Unix98
2820 getutxid NEW: Unix98
2821 getutxline NEW: Unix98
2822 glob64 NEW: GNU ext.
2823 globfree64 NEW: GNU ext.
2824 gnu_get_libc_release NEW: GNU ext.
2825 gnu_get_libc_version NEW: GNU ext.
2826 grantpt NEW: Unix98 PTY
2827 host2netname NEW: Secure RPC
2828 iconv NEW: iconv
2829 iconv_close NEW: iconv
2830 iconv_open NEW: iconv
2831 if_freenameindex NEW: IPv6
2832 if_indextoname NEW: IPv6
2833 if_nameindex NEW: IPv6
2834 if_nametoindex NEW: IPv6
2835 in6addr_any NEW: IPv6
2836 in6addr_loopback NEW: IPv6
2837 inet6_isipv4mapped NEW: IPv6
2838 isastream NEW: STREAMS
2839 iswblank NEW: GNU ext.
2840 key_decryptsession NEW: Secure RPC
2841 key_decryptsession_pk NEW: Secure RPC
2842 key_encryptsession NEW: Secure RPC
2843 key_encryptsession_pk NEW: Secure RPC
2844 key_gendes NEW: Secure RPC
2845 key_get_conv NEW: Secure RPC
2846 key_secretkey_is_set NEW: Secure RPC
2847 key_setnet NEW: Secure RPC
2848 key_setsecret NEW: Secure RPC
2849 llrint NEW: ISO C 9x
2850 llrintf NEW: ISO C 9x
2851 llrintl NEW: ISO C 9x
2852 llround NEW: ISO C 9x
2853 llroundf NEW: ISO C 9x
2854 llroundl NEW: ISO C 9x
2855 log2 NEW: ISO C 9x
2856 log2f NEW: ISO C 9x
2857 log2l NEW: ISO C 9x
2858 lrint NEW: ISO C 9x
2859 lrintf NEW: ISO C 9x
2860 lrintl NEW: ISO C 9x
2861 lround NEW: ISO C 9x
2862 lroundf NEW: ISO C 9x
2863 lroundl NEW: ISO C 9x
2864 lseek64 NEW: LFS
2865 makecontext NEW: Unix98
2866 mempcpy NEW: GNU ext.
2867 mmap64 NEW: LFS
2868 moncontrol REMOVED
2869 modify_ldt NEW: kernel
2870 nan NEW: ISO C 9x
2871 nanf NEW: ISO C 9x
2872 nanl NEW: ISO C 9x
2873 nearbyint NEW: ISO C 9x
2874 nearbyintf NEW: ISO C 9x
2875 nearbyintl NEW: ISO C 9x
2876 netname2host NEW: Secure RPC
2877 netname2user NEW: Secure RPC
2878 nexttoward NEW: ISO C 9x
2879 nexttowardf NEW: ISO C 9x
2880 nexttowardl NEW: ISO C 9x
2881 nftw NEW: Unix98
2882 nftw64 NEW: LFS
2883 open64 NEW: LFS
2884 passwd2des NEW: Secure RPC
2885 pow10 NEW: GNU ext.
2886 pow10f NEW: GNU ext.
2887 pow10l NEW: GNU ext.
2888 pread NEW: Unix98
2889 pread64 NEW: LFS
2890 printf_size NEW: GNU ext.
2891 printf_size_info NEW: GNU ext.
2892 profil_counter REMOVED
2893 pthread_mutexattr_getkind_np REPLACED
2894 pthread_mutexattr_setkind_np REPLACED
2895 ptsname NEW: Unix98 PTY
2896 ptsname_r NEW: Unix98 PTY
2897 putc_locked REMOVED
2898 putchar_locked REMOVED
2899 putgrent NEW: GNU ext.
2900 putmsg NEW: STREAMS
2901 putpmsg NEW: STREAMS
2902 pututxline NEW: Unix98
2903 pwrite NEW: Unix98
2904 pwrite64 NEW: LFS
2905 readdir64 NEW: LFS
2906 readdir64_r NEW: LFS
2907 remquo NEW: ISO C 9x
2908 remquof NEW: ISO C 9x
2909 remquol NEW: ISO C 9x
2910 round NEW: ISO C 9x
2911 roundf NEW: ISO C 9x
2912 roundl NEW: ISO C 9x
2913 rtime NEW: GNU ext.
2914 scalbln NEW: ISO C 9x
2915 scalblnf NEW: ISO C 9x
2916 scalblnl NEW: ISO C 9x
2917 scandir64 NEW: LFS
2918 sendfile NEW: kernel
2919 setcontext NEW: Unix98
2920 setrlimit64 NEW: LFS
2921 setutxent NEW: Unix98
2922 sighold NEW: Unix98
2923 sigignore NEW: Unix98
2924 sigqueue NEW: POSIX.1b
2925 sigrelse NEW: Unix98
2926 sigset NEW: POSIX.1b
2927 sigtimedwait NEW: POSIX.1b
2928 sigwaitinfo NEW: POSIX.1b
2929 sincos NEW: GNU ext.
2930 sincosf NEW: GNU ext.
2931 sincosl NEW: GNU ext.
2932 statfs64 NEW: LFS
2933 statvfs NEW: Unix98
2934 statvfs64 NEW: LFS
2935 strcasestr NEW: GNU ext.
2936 strtoimax NEW: ISO C 9x
2937 strtoumax NEW: ISO C 9x
2938 strverscmp NEW: GNU ext.
2939 svcauthdes_stats NEW: Secure RPC
2940 svcunix_create NEW: sunrpc etx
2941 svcunixfd_create NEW: sunrpc ext
2942 swapcontext NEW: Unix98
2943 tcgetsid NEW: Unix98 PTY
2944 tdestroy NEW: GNU ext.
2945 tgamma NEW: ISO C 9x
2946 tgammaf NEW: ISO C 9x
2947 tgammal NEW: ISO C 9x
2948 tmpfile64 NEW: LFS
2949 trunc NEW: ISO C 9x
2950 truncate64 NEW: LFS
2951 truncf NEW: ISO C 9x
2952 truncl NEW: ISO C 9x
2953 umount2 NEW: kernel
2954 unlockpt NEW: Unix98 PTY
2955 updwtmpx NEW: Unix98
2956 user2netname NEW: Secure RPC
2957 utmpxname NEW: Unix98
2958 versionsort NEW: GNU ext.
2959 versionsort64 NEW: GNU ext.
2960 waitid NEW: Unix98
2961 wcscasecmp NEW: GNU ext.
2962 wcsncasecmp NEW: GNU ext.
2963 wcsnlen NEW: GNU ext.
2964 wcstoimax NEW: ISO C 9x
2965 wcstoll NEW: ISO C 9x
2966 wcstoull NEW: ISO C 9x
2967 wcstoumax NEW: ISO C 9x
2968 wcswcs NEW: Unix98
2969 wordexp NEW: POSIX.2
2970 wordfree NEW: POSIX.2
2971 write_profiling REMOVED
2972 xdecrypt NEW: Secure RPC
2973 xdr_authdes_cred NEW: Secure RPC
2974 xdr_authdes_verf NEW: Secure RPC
2975 xdr_cryptkeyarg NEW: Secure RPC
2976 xdr_cryptkeyarg2 NEW: Secure RPC
2977 xdr_cryptkeyres NEW: Secure RPC
2978 xdr_getcredres NEW: Secure RPC
2979 xdr_key_netstarg NEW: Secure RPC
2980 xdr_key_netstres NEW: Secure RPC
2981 xdr_keybuf NEW: Secure RPC
2982 xdr_keystatus NEW: Secure RPC
2983 xdr_netnamestr NEW: Secure RPC
2984 xdr_sizeof NEW: Secure RPC
2985 xdr_unixcred NEW: sunrpc ext
2986 xencrypt NEW: Secure RPC
2987 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
2988 \f
2989 Version 2.0.6
2990
2991 * more bug fixes
2992
2993 \f
2994 Version 2.0.5
2995
2996 * more bug fixes
2997
2998 * inet_ntoa is thread-safe
2999
3000 * updwtmp is moved from libutil to libc
3001
3002 * rewrite of cbrt function
3003
3004 * update of timezone data
3005 \f
3006 Version 2.0.4
3007
3008 * more bug fixes
3009 \f
3010 Version 2.0.3
3011
3012 * more bug fixes
3013 \f
3014 Version 2.0.2
3015
3016 * more bug fixes
3017
3018 * add atoll function
3019
3020 * fix complex problems in Berkeley DB code
3021
3022 * fix math functions
3023 \f
3024 Version 2.0.1
3025
3026 * fixed lots of header problems (especially Linux/GNU specific)
3027
3028 * dynamic loader preserves all registers
3029
3030 * Roland McGrath provided support for handling of auxiliary objects in
3031 the ELF dynamic loader.
3032
3033 * support for parallel builds is improved
3034 \f
3035 Version 2.0
3036
3037 * GNU extensions are no longer declared by default. To enable them you
3038 must define the macro `_GNU_SOURCE' in your program or compile with
3039 `-D_GNU_SOURCE'.
3040
3041 * The library has changed from using GNU ld symbol aliases to using weak
3042 symbols where available. The ELF object file format supports weak
3043 symbols; GNU ld also supports weak symbols in the a.out format. (There
3044 is also now support for other GNU ld extensions in ELF. Use the
3045 `--with-elf' option to configure to indicate you have ELF, and
3046 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3047 of many files which contained only symbol aliases, reducing the size of
3048 the source and the compiled library; many other files were renamed to
3049 less cryptic names previously occupied by the symbol alias files.
3050 There is a new header file <elf.h> for programs which operate on
3051 files in the ELF format.
3052
3053 * Converted to Autoconf version 2, so `configure' has more options.
3054 Run `configure --help' to see the details.
3055
3056 * The library can now be configured to build profiling, highly-optimized
3057 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3058 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3059 `configure' enable building these extra libraries. The shared library is
3060 built by default when using both ELF and GNU ld. When shared libraries
3061 are enabled, the new library `-ldl' is available for arbitrary run-time
3062 loading of shared objects; its interface is defined in <dlfcn.h>. The
3063 new header file <link.h> gives access to the internals of the run-time
3064 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3065 application of same name on other systems and it provides information
3066 about dynamically linked binaries.
3067
3068 * The C library now provides the run-time support code for profiling
3069 executables compiled with `-pg'. Programs can control the profiling code
3070 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3071 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3072 the support for this file format was contributed by David Mosberger-Tang.
3073
3074 * The math code has been replaced with a math library based on fdlibm from
3075 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3076 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3077 functions. The math functions now reside in a separate library, so
3078 programs using them will need to use `-lm' their linking commands.
3079
3080 * John C. Bowman contributed optimized ix87 assembler inline functions.
3081
3082 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3083 mechanism similar to that found in Solaris 2. This is now used for the
3084 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3085 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3086 file controls what services are used for each individual database. This
3087 works by loading shared libraries with names specified in `nsswitch.conf',
3088 so service modules can be changed or added at any time without even
3089 relinking any program. Currently there are the file, db, and NIS based
3090 NSS services available.
3091
3092 * The new functions `strtoq' and `strtouq' parse integer values from
3093 strings, like `strtol' and `strtoul', but they return `long long int' and
3094 `unsigned long long int' values, respectively (64-bit quantities).
3095
3096 * The new functions `strtof' and `strtold' parse floating-point values from
3097 strings, like `strtod', but they return `float' and `long double' values,
3098 respectively (on some machines `double' and `long double' are the same).
3099
3100 * Ulrich Drepper has contributed new implementations of the floating-point
3101 printing and reading code used in the `printf' family of functions and
3102 `strtod', `strtof', and `strtold'. These new functions are perfectly
3103 accurate, and much faster than the old ones.
3104
3105 * The implementation of the POSIX locale model was completely rewritten by
3106 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3107 to compile the POSIX locale definition.
3108
3109 * The former dummy implementations of the strcoll and strxfrm function are
3110 now replaced by fully functional code contributed by Ulrich Drepper. The
3111 collation information comes from the POSIX locale definitions.
3112
3113 * The new header <langinfo.h> defines an interface for accessing
3114 various locale-dependent data (using the locale chosen with `setlocale').
3115
3116 * Ulrich Drepper has contributed a new suite of functions for operation on
3117 wide-character and multibyte-character strings, in <wchar.h>;
3118 and classification and case conversion of wide characters, in <wctype.h>.
3119 These new functions are conforming to the ISO C, Amendement 1 specification.
3120
3121 * There is now a second implementation of the standard I/O library available.
3122 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3123 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3124
3125 * You can now use positional parameter specifications in format strings
3126 for the `printf' and `scanf' families of functions. For example,
3127 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3128 ``Number 6, Mr Jones''. This is mainly useful when providing different
3129 format strings for different languages, whose grammars may dictate
3130 different orderings of the values being printed. To support this
3131 feature, the interface for `register_printf_handler' has changed; see
3132 the header file <printf.h> for details.
3133
3134 * The `printf' and `scanf' families of functions now understand a new
3135 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3136 to group numbers as indicated by the locale; for `scanf' and friends, this
3137 says to accept as valid only a number with all the proper grouping
3138 separators in the right places. In the default "C" locale, numbers are
3139 not grouped; but locales for specific countries will define the usual
3140 conventions (i.e. separate thousands with `,' in the US locale).
3141
3142 * The pgrp functions have been regularized, slightly incompatibly but much
3143 less confusingly. The core functions are now `getpgid' and `setpgid',
3144 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3145 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3146 provided for compatibility. There is no longer an incompatible `getpgrp'
3147 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3148
3149 * The new header file <fts.h> and suite of functions simplify programs that
3150 operate on directory trees. This code comes from 4.4 BSD.
3151
3152 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3153 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3154 NSS scheme used in glibc.
3155
3156 * There is a new malloc debugging hook `__memalign_hook'.
3157
3158 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3159 `unsigned int' in <sys/types.h>. These are for compatibility only and
3160 their use is discouraged.
3161
3162 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3163 done differently, so that it works even without GNU ld.
3164
3165 * New function `euidaccess' checks allowed access to a file like `access',
3166 but using the effective IDs instead of the real IDs.
3167
3168 * The time zone data files have been updated for the latest and greatest
3169 local time conventions of the countries of the world.
3170
3171 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3172 see <dirent.h>.
3173
3174 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3175 for formatting floating-point numbers. They are provided only for
3176 compatibility; new programs should use `sprintf' instead. There are
3177 also equivalent function for the `long double' floating-point type and
3178 all functions also exist in a reentrant form.
3179
3180 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3181 functions for maintaining the login-record files (primarily of use to
3182 system programs such as `login'), and convenient functions for
3183 allocating and initializing a pseudo-terminal (pty) device.
3184
3185 * Ulrich Drepper has contributed new support for System V style
3186 shared memory and IPC on systems that support it.
3187
3188 * Ulrich Drepper has contributed several miscellaneous new functions found
3189 in System V: The `hsearch' family of functions provide an effective
3190 implementation of hash tables; `a64l' and `l64a' provide a very simple
3191 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3192 number generator.
3193
3194 * Ulrich Drepper has contributed new reentrant counterparts for the
3195 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3196
3197 * Ulrich Drepper has contributed new, highly-optimized versions of several
3198 string functions for the i486/Pentium family of processors.
3199
3200 * Ulrich Drepper has updated the Linux-specific code, based largely
3201 on work done in Hongjiu Lu's version of GNU libc for Linux.
3202 The GNU library now supports Linux versions 2.0.10 and later,
3203 using the ELF object file format (i[3456]86-*-linux).
3204
3205 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3206
3207 * David Mosberger-Tang and Richard Henderson have ported the C library
3208 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3209 dynamic linking support for ELF/Alpha.
3210
3211 * Richard Henderson contributed several Alpha optimized assembler function
3212 for arithmetic and string handling.
3213
3214 * Ulrich Drepper has contributed a new set of message catalog functions to
3215 support multiple languages using the <libintl.h> interface, for use with
3216 his new package GNU gettext. Translation volunteers have contributed
3217 catalogs of the library's messages in Spanish, German, and Korean.
3218
3219 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3220 program and the `catgets' function for reading the catalog files it
3221 creates. (The <libintl.h> interface is preferred; we include the
3222 <nl_types.h> interface using `catgets' only for source compatibility with
3223 programs already written to use it.)
3224
3225 * New header file <values.h> gives SVID-compatible names for <limits.h>
3226 constants.
3227
3228 * Various new macros, declarations, and small header files for compatibility
3229 with 4.4 BSD.
3230
3231 * New function `group_member' is a convenient way to check if a process has
3232 a given effective group ID.
3233
3234 * When using GCC 2.7 and later, the socket functions are now declared in a
3235 special way so that passing an argument of type `struct sockaddr_in *',
3236 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3237 `struct sockaddr *' type, does not generate a type-clash warning.
3238
3239 * New function `error' declared in header file <error.h> is a convenient
3240 function for printing error messages and optionally exiting; this is the
3241 canonical function used in GNU programs. The new functions `err', `warn',
3242 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3243 doing the same thing.
3244
3245 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3246 POSIX.2 `glob' function to do ~ and {...} expansion.
3247
3248 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3249 `clearenv' which is used in POSIX.9 is also available.
3250
3251 * New function `getsid' returns session ID number on systems that support it.
3252
3253 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3254 files <db.h> and <mpool.h> provide a rich set of functions for several
3255 types of simple databases stored in memory and in files, and <ndbm.h> is
3256 an old `ndbm'-compatible interface using the `db' functions. Link with
3257 `-ldb' to get these functions.
3258
3259 * New macro `strdupa' copies a string like `strdup', but uses local stack
3260 space from `alloca' instead of dynamic heap space from `malloc'.
3261
3262 * New function `strnlen' is like `strlen' but searches only a given maximum
3263 number of characters for the null terminator. `stpncpy', `strndup' and
3264 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3265 function.
3266
3267 * New function `statfs' in header <sys/statfs.h>.
3268
3269 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3270 provide convenient functions for operating on blocks of null-terminated
3271 strings.
3272
3273 * A new suite of functions in <utmp.h> handle all the details of reading
3274 and writing the utmp file.
3275
3276 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3277 Thorsten Kukuk.
3278
3279 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3280 completely POSIX compliant and also implemented the extended functionality
3281 to handle alternate digit representation and alternate era date formats.
3282
3283 * Ulrich Drepper provided an implementation of the `strptime' function
3284 defined in XPG4.2 which transforms a string into a `struct tm' value.
3285
3286 * Paul Eggert provided the tzselect shell script as part of the timezone
3287 code. The shell script makes it easy to select the correct timezone
3288 specification.
3289
3290 * The implementation of the malloc family of functions is completely replaced
3291 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3292 The implementation uses the mmap function (if available) and it is
3293 optimized for the use in multi threaded programs.
3294
3295 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3296 functions. This new functionality is usable by specifying a special
3297 salt string and it is compatible with implementation on *BSD systems.
3298
3299 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3300 `getsubopt' to handle second level command line options, `bsd_signal'
3301 to access BSD style `signal' functionality, the obsolete `regexp' style
3302 expression matcher.
3303
3304 * the `lchown' function is available on system which support this
3305 functionality.
3306
3307 * The implementation of the shadow password handling function was contributed
3308 by Ulrich Drepper.
3309
3310 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3311
3312 * POSIX.1g support was added. The <sys/select.h> header is available,
3313 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3314 implementation of `getaddrinfo'.
3315 \f
3316 Version 1.09
3317
3318 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3319
3320 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3321 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3322
3323 * The new function `daemon' from 4.4 BSD is useful for server programs that
3324 want to put themselves in the background.
3325
3326 * Joel Sherrill has contributed support for several standalone boards that
3327 run without an operating system.
3328
3329 * `printf', `scanf' and friends now accept a `q' type modifier for long
3330 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3331
3332 * All of the code taken from BSD (notably most of the math and networking
3333 routines) has been updated from the BSD 4.4-Lite release.
3334
3335 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3336
3337 * The new functions `getdomainname' and `setdomainname' fetch or change the
3338 YP/NIS domain name. These are system calls which exist on systems which
3339 have YP (aka NIS).
3340
3341 * The time zone data files have been updated for the latest international
3342 conventions.
3343
3344 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3345 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3346 \f
3347 Version 1.08
3348
3349 * The C library now includes support for Sun RPC, from Sun's free
3350 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3351 are included. (There is still no support for YP.)
3352
3353 * Tom Quinn has contributed a port of the C library to SGI machines running
3354 Irix 4 (mips-sgi-irix4).
3355
3356 * The new `lockf' function is a simplified interface to the locking
3357 facilities of `fcntl', included for compatibility.
3358
3359 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3360
3361 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3362
3363 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3364 compatibility.
3365
3366 * The error message printed by `assert' for a failed assertion now includes
3367 the name of the program (if using GNU ld) and the name of the calling
3368 function (with versions of GCC that support this).
3369
3370 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3371
3372 * The library now includes the <sys/mman.h> header file and memory
3373 management functions `mmap', `munmap', `mprotect', `msync', and
3374 `madvise', on systems that support those facilities.
3375
3376 * The interface for `mcheck' has changed slightly: the function called to
3377 abort the program when an allocation inconsistency is detected now takes
3378 an argument that indicates the type of failure. The new function
3379 `mprobe' lets you request a consistency check for a particular block at
3380 any time (checks are normally done only when you call `free' or `realloc'
3381 on a block).
3382
3383 * It is now possible to easily cross-compile the C library, building on one
3384 system a library to run on another machine and/or operating system. All
3385 you need to do is set the variable `HOST_CC' in `configparms' to the
3386 native compiler for programs to run on the machine you are building on (a
3387 few generator programs are used on Unix systems); set `CC' to the
3388 cross-compiler.
3389
3390 * The new function `fexecve' (only implemented on the GNU system) executes
3391 a program file given a file descriptor already open on the file.
3392 \f
3393 Version 1.07
3394
3395 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3396 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3397
3398 * You can set the variable `libprefix' in `configparms' to specify a prefix
3399 to be prepended to installed library files; this makes it easy to install
3400 the GNU C library to be linked as `-lgnuc' or whatever.
3401
3402 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3403 copies a limited number of characters from a string, and returns the
3404 address of the last character written.
3405
3406 * You no longer need to check for whether the installed `stddef.h' is
3407 compatible with the GNU C library. configure now checks for you.
3408
3409 * You can now define a per-stream `fileno' function to convert the
3410 stream's cookie into an integral file descriptor.
3411
3412 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3413 allocates zero bytes of storage, and returns a unique pointer which
3414 you can pass to `realloc' or `free'. The behavior is undefined if
3415 you dereference this pointer.
3416
3417 * The C library now runs on Sony NEWS m68k machines running either
3418 NewsOS 3 or NewsOS 4.
3419
3420 * The new `syscall' function is a system-dependent primitive function
3421 for invoking system calls. It has the canonical behavior on Unix
3422 systems, including unreliable return values for some calls (such as
3423 `pipe', `fork' and `getppid').
3424
3425 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3426 to `EAGAIN', which is the preferred name. On systems whose kernels
3427 use two distinct codes, the C library now translates EWOULDBLOCK to
3428 EAGAIN in every system call function.
3429 \f
3430 Version 1.06
3431
3432 * The GNU C Library Reference Manual is now distributed with the library.
3433 `make dvi' will produce a DVI file of the printed manual.
3434 `make info' will produce Info files that you can read on line using C-h i
3435 in Emacs or the `info' program.
3436 Please send comments on the manual to bug-glibc-manual@gnu.org.
3437
3438 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3439
3440 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3441
3442 * Jason Merrill has contributed a port to the Sequent Symmetry running
3443 Dynix version 3 (i386-sequent-dynix).
3444
3445 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3446 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3447
3448 * New function `memory_warnings' lets you arrange to get warnings when
3449 malloc is running out of memory to allocate, like Emacs gives you.
3450
3451 * The C library now contains the relocating allocator used in Emacs 19 for
3452 its editing buffers. This allocator (ralloc) minimizes allocation
3453 overhead and fragmentation by moving allocated regions around whenever it
3454 needs to. You always refer to a ralloc'd region with a "handle" (a
3455 pointer to a pointer--an object of type `void **').
3456
3457 * There is a new `printf' format: `%m' gives you the string corresponding
3458 to the error code in `errno'.
3459
3460 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3461 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3462 pass, the `a' modifier says to fill in a `char **' you pass with a
3463 malloc'd string.
3464
3465 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3466 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3467 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3468
3469 * `mkstemp' is a traditional Unix function to atomically create and open a
3470 uniquely-named temporary file.
3471 \f
3472 Version 1.05
3473
3474 * The standard location for the file that says what the local timezone is
3475 has changed again. It is now `/usr/local/etc/localtime' (or more
3476 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3477
3478 * The distribution no longer contains any files with names longer than 14
3479 characters.
3480
3481 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3482 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3483
3484 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3485 \f
3486 Version 1.04
3487
3488 * The configuration process has changed quite a bit. The `configure'
3489 script is now used just like the configuration scripts for other GNU
3490 packages. The `sysdeps' directory hierarchy is much rearranged.
3491 The file `INSTALL' explains the new scheme in detail.
3492
3493 * The header files no longer need to be processed into ANSI C and
3494 traditional C versions. There is just one set of files to install, and
3495 it will work with ANSI or old C compilers (including `gcc -traditional').
3496
3497 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3498 MIPS DECStation running Ultrix 4.
3499
3500 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3501 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3502 made itself into a shared library.
3503
3504 * Yet further improved support for the i386, running 4.3 BSD-like systems
3505 (such as Mach 3 with the Unix single-server), or System V.
3506
3507 * New function `strncasecmp' to do case-insensitive string comparison
3508 with limited length.
3509
3510 * New function `strsep' is a reentrant alternative to `strtok'.
3511
3512 * New functions `scandir' and `alphasort' for searching directories.
3513
3514 * New function `setenv' is a better interface to `putenv'.
3515
3516 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3517 function for traversing a directory tree.
3518
3519 * The GNU obstack package is now also part of the C library.
3520 The new function `open_obstack_stream' creates a stdio stream that
3521 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3522 formatted output directly to an obstack.
3523
3524 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3525 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3526
3527 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3528
3529 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3530 things to your strings.
3531
3532 * There are some new test programs: test-fseek, testmb, and testrand.
3533
3534 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3535 These ports are not finished, but are a good starting place for really
3536 supporting those systems.
3537
3538 * `/etc/localtime' is now the standard location for the file that says what
3539 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3540 This follows the general principle that `/etc' is the place for all local
3541 configuration files.
3542
3543 * The C library header files now use `extern "C"' when used by the C++
3544 compiler, so the C library should now work with C++ code.
3545
3546 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3547 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3548 in <strings.h>.)
3549
3550 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3551 function which uses the `merge sort' algorithm, and is said to be
3552 significantly faster than the old GNU `qsort' function. Merge sort is
3553 now the standard `qsort' function. The new algorithm can require a lot
3554 of temporary storage; so, the old sorting function is called when the
3555 required storage is not available.
3556
3557 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3558 provides the Unix `crypt' function, plus some other entry points.
3559
3560 * The code and header files taken from 4.4 BSD have been updated with the
3561 latest files released from Berkeley.
3562 \f
3563 ----------------------------------------------------------------------
3564 Copyright information:
3565
3566 Copyright (C) 1992-2016 Free Software Foundation, Inc.
3567
3568 Permission is granted to anyone to make or distribute verbatim copies
3569 of this document as received, in any medium, provided that the
3570 copyright notice and this permission notice are preserved,
3571 thus giving the recipient permission to redistribute in turn.
3572
3573 Permission is granted to distribute modified versions
3574 of this document, or of portions of it,
3575 under the above conditions, provided also that they
3576 carry prominent notices stating who last changed them.
3577 \f
3578 Local variables:
3579 version-control: never
3580 fill-column: 76
3581 End: