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 It has been replaced by pap_AW & pap_CW, both of which have long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