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 It has been replaced by pap_AW & pap_CW, both of which have long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