]> git.ipfire.org Git - thirdparty/glibc.git/blob - NEWS
drop redundant ([eE][sS])? and [oO]? from yesexpr and noexpr in km_KH locale
[thirdparty/glibc.git] / NEWS
1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3 See the end for copying conditions.
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 2.27
9
10 Major new features:
11
12 [Add new features here]
13
14 Deprecated and removed features, and other changes affecting compatibility:
15
16 [Add deprecations, removals and changes affecting compatibility here]
17
18 Changes to build and runtime requirements:
19
20 [Add changes to build and runtime requirements here]
21
22 Security related changes:
23
24 [Add security related changes here]
25
26 The following bugs are resolved with this release:
27
28 [The release manager will add the list generated by
29 scripts/list-fixed-bugs.py just before the release.]
30
31 \f
32 Version 2.26
33
34 Major new features:
35
36 * A per-thread cache has been added to malloc. Access to the cache requires
37 no locks and therefore significantly accelerates the fast path to allocate
38 and free small amounts of memory. Refilling an empty cache requires locking
39 the underlying arena. Performance measurements show significant gains in a
40 wide variety of user workloads. Workloads were captured using a special
41 instrumented malloc and analyzed with a malloc simulator. Contributed by
42 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
43
44 * Unicode 10.0.0 Support: Character encoding, character type info, and
45 transliteration tables are all updated to Unicode 10.0.0, using
46 generator scripts contributed by Mike FABIAN (Red Hat).
47 These updates cause user visible changes, especially the changes in
48 wcwidth for many emoji characters cause problems when emoji sequences
49 are rendered with pango, see for example:
50 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
51
52 * Collation of Hungarian has been overhauled and is now consistent with "The
53 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
54 Egmont Koblinger.
55
56 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
57
58 - The GNU C Library will now detect when /etc/resolv.conf has been
59 modified and reload the changed configuration. The new resolver option
60 “no-reload” (RES_NORELOAD) disables this behavior.
61
62 - The GNU C Library now supports an arbitrary number of search domains
63 (configured using the “search” directive in /etc/resolv.conf);
64 previously, there was a hard limit of six domains. For backward
65 compatibility, applications that directly modify the ‘_res’ global
66 object are still limited to six search domains.
67
68 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
69 Library will now randomly pick a name server from the configuration as a
70 starting point. (Previously, the second name server was always used.)
71
72 * The tunables feature is now enabled by default. This allows users to tweak
73 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
74
75 * New function reallocarray, which resizes an allocated block (like realloc)
76 to the product of two sizes, with a guaranteed clean failure upon integer
77 overflow in the multiplication. Originally from OpenBSD, contributed by
78 Dennis Wölfing and Rüdiger Sonderfeld.
79
80 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
81 These are extended versions of preadv and pwritev, respectively, taking an
82 additional flags argument. The set of supported flags depends on the
83 running kernel; full support currently requires kernel 4.7 or later.
84
85 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
86 create a new session ID for the spawned process. This feature is
87 scheduled to be added to the next major revision of POSIX; for the time
88 being, it is available under _GNU_SOURCE.
89
90 * errno.h is now safe to use from C-preprocessed assembly language on all
91 supported operating systems. In this context, it will only define the
92 Exxxx constants, as preprocessor macros expanding to integer literals.
93
94 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
95 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
96 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
97 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
98
99 To compile programs that use this feature, the compiler must support
100 128-bit floating point with the type name _Float128 (as defined by TS
101 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
102 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
103 must be defined to make the new interfaces visible.
104
105 The new functions and macros correspond to those present for other
106 floating-point types (except for a few obsolescent interfaces not
107 supported for the new type), with F128 or f128 suffixes; for example,
108 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
109 printf or scanf formats for the new type; the strfromf128 and strtof128
110 interfaces should be used instead.
111
112 Deprecated and removed features, and other changes affecting compatibility:
113
114 * The synchronization that pthread_spin_unlock performs has been changed to
115 now be equivalent to a C11 atomic store with release memory order to the
116 spin lock's memory location. Previously, several (but not all)
117 architectures used stronger synchronization (e.g., containing what is
118 often called a full barrier). This change can improve performance, but
119 may affect odd fringe uses of spin locks that depend on the previous
120 behavior (e.g., using spin locks as atomic variables to try to implement
121 Dekker's mutual exclusion algorithm).
122
123 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
124 removed.
125
126 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
127 will only be built and installed when the GNU C Library is configured with
128 --enable-obsolete-rpc. This allows alternative RPC implementations, such
129 as TIRPC or rpcsvc-proto, to be used.
130
131 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
132 libnss_compat, are deprecated, and will not be built or installed by
133 default.
134
135 The NIS(+) support library, libnsl, is also deprecated. By default, a
136 compatibility shared library will be built and installed, but not headers
137 or development libraries. Only a few NIS-related programs require this
138 library. (In particular, the GNU C Library has never required programs
139 that use 'gethostbyname' to be linked with libnsl.)
140
141 Replacement implementations based on TIRPC, which additionally support
142 IPv6, are available from <https://github.com/thkukuk/>. The configure
143 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
144 name service modules, to be built and installed.
145
146 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
147 support is enabled, the configured recursive resolver must support EDNS.
148 (Responding to EDNS-enabled queries with responses which are not
149 EDNS-enabled is fine, but FORMERR responses are not.)
150
151 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
152 servers have not supported this opcode for a long time.
153
154 * The _res_opcodes variable has been removed from libresolv. It had been
155 exported by accident.
156
157 * <string.h> no longer includes inline versions of any string functions,
158 as this kind of optimization is better done by the compiler. The macros
159 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
160
161 * The nonstandard header <xlocale.h> has been removed. Most programs should
162 use <locale.h> instead. If you have a specific need for the definition of
163 locale_t with no other declarations, please contact
164 libc-alpha@sourceware.org and explain.
165
166 * The obsolete header <sys/ultrasound.h> has been removed.
167
168 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
169
170 * The obsolete function cfree has been removed. Applications should use
171 free instead.
172
173 * The stack_t type no longer has the name struct sigaltstack. This changes
174 the C++ name mangling for interfaces involving this type.
175
176 * The ucontext_t type no longer has the name struct ucontext. This changes
177 the C++ name mangling for interfaces involving this type.
178
179 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
180 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
181 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
182 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
183 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
184 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
185 fpregset_t type no longer has the name struct fpu. This changes the C++
186 name mangling for interfaces involving those types.
187
188 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
189 synced with the kernel:
190
191 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
192 are not supported on this architecture and have been removed.
193
194 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
195 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
196 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
197
198 Programs that assume the GET/SETREGS ptrace requests are universally
199 available will now fail to build, instead of malfunctioning at runtime.
200
201 Changes to build and runtime requirements:
202
203 * Linux kernel 3.2 or later is required at runtime, on all architectures
204 supported by that kernel. (This is a change from version 2.25 only for
205 x86-32 and x86-64.)
206
207 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
208
209 * On most architectures, GCC 4.9 or later is required to build the GNU C
210 Library. On powerpc64le, GCC 6.2 or later is required.
211
212 Older GCC versions and non-GNU compilers are still supported when
213 compiling programs that use the GNU C Library. (We do not know exactly
214 how old, and some GNU extensions to C may be _de facto_ required. If you
215 are interested in helping us make this statement less vague, please
216 contact libc-alpha@sourceware.org.)
217
218 Security related changes:
219
220 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
221 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
222
223 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
224 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
225
226 * Avoid printing a backtrace from the __stack_chk_fail function since it is
227 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
228 (CVE-2010-3192).
229
230 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
231 fixed (CVE-2017-12133).
232
233 The following bugs are resolved with this release:
234
235 [984] network: Respond to changed resolv.conf in gethostbyname
236 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
237 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
238 fields
239 [12189] libc: __stack_chk_fail should not attempt a backtrace
240 (CVE-2010-3192)
241 [14096] time: Race condition on timezone/tst-timezone.out
242 [14172] localedata: az_IR: new locale
243 [14995] build: glibc fails to build if gold is the default linker, even if
244 ld.bfd is available
245 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
246 [16637] network: inet_pton function is accepting IPv6 with bad format
247 [16640] string: string/strtok.c: undefined behaviour inconsistent between
248 x86 and other generic code
249 [16875] localedata: ko_KR: fix lang_name
250 [17225] localedata: ar_SY: localized month names for May and June are
251 incorrect
252 [17297] localedata: da_DK: wrong date_fmt string
253 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
254 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
255 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
256 [19066] localedata: ar_SA abbreviated day and month names are in English
257 [19569] network: resolv: Support an arbitrary number of search domains
258 [19570] network: Implement random DNS server selection in the stub
259 resolver
260 [19838] locale: localedef fails on PA-RISC
261 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
262 order of 0D36 and 0D37
263 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
264 chillu characters
265 [20098] libc: FAIL: debug/backtrace-tst on hppa
266 [20257] network: sunrpc: clntudp_call does not enforce timeout when
267 receiving data
268 [20275] localedata: locale day/abday/mon/abmon should not have trailing
269 whitespace
270 [20313] localedata: Update locale data to Unicode 9.0
271 [20424] manual: Document how to provide a malloc replacement
272 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
273 for Peru
274 [20686] locale: Add el_GR@euro to SUPPORTED.
275 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
276 failures consistently
277 [21015] dynamic-link: Document and fix --enable-bind-now
278 [21016] nptl: pthread_cond support is broken on hppa
279 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
280 frame-pointer on i386
281 [21049] libc: segfault in longjmp_chk() due to clobbered processor
282 register
283 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
284 [21088] libc: Build fails with --enable-static-nss
285 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
286 generic c code is used
287 [21109] libc: Tunables broken on big-endian
288 [21112] math: powf has large ulp errors with base close to 1 and exponent
289 around 4000
290 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
291 (CVE-2017-12133)
292 [21120] malloc: glibc malloc is incompatible with GCC 7
293 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
294 with -lieee
295 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
296 (-0.0) when linking with -lieee
297 [21171] math: log10, log2 and lgamma return incorrect results
298 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
299 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
300 [21207] localedata: ce_RU: update weekdays from CLDR
301 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
302 [21217] localedata: Update months from CLDR-31
303 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
304 [21243] libc: support_delete_temp_file should issue warning for failed
305 remove()
306 [21244] libc: support resolv_test_start() socket fd close should be
307 checked for errors.
308 [21253] libc: localedef randomly segfaults when using -fstack-check due to
309 new posix_spawn implementation
310 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
311 leads to lower CPU frequency
312 [21259] libc: [alpha] termios.h missing IXANY for POSIX
313 [21261] libc: [sparc64] bits/setjmp.h namespace
314 [21267] network: [mips] bits/socket.h IOC* namespace
315 [21268] libc: [alpha] termios.h NL2, NL3 namespace
316 [21270] libc: mmap64 silently truncates large offset values
317 [21275] libc: posix_spawn always crashes on ia64 now
318 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
319 [21280] math: [powerpc] logbl for POWER7 return incorrect results
320 [21289] libc: Incorrect declaration for 32-bit platforms with
321 _FILE_OFFSET_BITS=64 causes build error
322 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
323 not support gethostbyname4_r
324 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
325 switching
326 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
327 arenas
328 [21340] libc: Support POSIX_SPAWN_SETSID
329 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
330 [21359] network: ns_name_pack needs additional byte in destination buffer
331 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
332 against fragmentation attacks (CVE-2017-12132)
333 [21369] network: resolv: Remove EDNS fallback
334 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
335 and _POSIX_C_SOURCE
336 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
337 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
338 [21393] stdio: Missing dup3 error check in freopen, freopen64
339 [21396] libc: Use AVX2 memcpy/memset on Skylake server
340 [21399] localedata: Bad description for U00EC in
341 localedata/charmaps/CP1254
342 [21411] malloc: realloc documentation error
343 [21426] network: sys/socket.h uio.h namespace
344 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
345 [21445] libc: signal.h bsd_signal namespace
346 [21455] network: Network headers stdint.h namespace
347 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
348 retry value
349 [21475] network: resolv: Overlong search path is truncated mid-label
350 [21511] libc: sigstack namespace
351 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
352 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
353 bits/syscall.h: No such file or directory
354 [21517] libc: struct sigaltstack namespace
355 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
356 [21533] localedata: Update locale data to Unicode 10.0
357 [21537] libc:
358 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
359 junk at end of line, first unrecognized character is `@'
360 [21538] libc: SIG_HOLD missing for XPG4
361 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
362 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
363 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
364 [21543] libc: sigevent namespace
365 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
366 size for general purpose registers in mcontext_t structure
367 [21550] libc: sigwait namespace
368 [21552] libc: XPG4 bsd_signal namespace
369 [21554] libc: sigpause namespace
370 [21560] libc: sys/wait.h signal.h namespace
371 [21561] libc: waitid namespace
372 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
373 '/usr/include/stdlib.h': Permission denied
374 [21575] libc: sys/wait.h missing struct rusage definition
375 [21584] libc: sigaltstack etc namespace
376 [21597] libc: siginterrupt namespace
377 [21607] math: hppa: FAIL: math/test-tgmath
378 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
379 issue on x86-64
380 [21622] libc: [tile] missing SA_* for POSIX.1:2008
381 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
382 stack and heap (CVE-2017-1000366)
383 [21625] libc: wait3 namespace
384 [21654] nss: Incorrect pointer alignment in NSS group merge result
385 construction
386 [21657] network: Parse interface zone id for node-local multicast
387 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
388 between_2_3
389 [21666] libc: .symver is used on common symbol
390 [21668] network: resolv: res_init cross-thread broadcast introduces race
391 conditions
392 [21687] math: tgmath.h totalorder, totalordermag return type
393 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
394 Hindi Locale
395 [21696] libc: Incorrect assumption of of __cpu_mask in
396 posix/sched_cpucount.c
397 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
398 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
399 locale
400 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
401 leading to relocation crash
402 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
403 `resolv_conf_matches (resp, conf)' failed.
404 [21710] localedata: Added Samoan language locale for Samoa
405 [21711] localedata: Pashto yesstr/nostr locale are missing
406 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
407 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
408 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
409 [21723] localedata: yesstr/nostr missing for Chinese language locale
410 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
411 locale
412 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
413 locale
414 [21728] localedata: New Locale for Tongan language
415 [21729] localedata: incorrect LC_NAME fields for hi_IN
416 [21733] localedata: yesstr and nostr are missing for zh_HK
417 [21734] localedata: Missing yesstr and nostr are for kw_GB
418 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
419 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
420 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
421 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
422 strings
423 [21744] libc: Tests failing on --enable-tunables --enable-stack-
424 protector=all
425 [21749] localedata: Wrong abbreviated day name (“abday”) for
426 ar_JO/ar_LB/ar_SY
427 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
428 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
429 [21759] localedata: missing yesstr and nostr for Tigrinya
430 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
431 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
432 [21767] localedata: Missing Bislama locales
433 [21768] localedata: Missing yesstr and nostr for aa_ET
434 [21770] localedata: Missing Field in li_NL
435 [21778] nptl: Robust mutex may deadlock
436 [21779] libc: MicroBlaze segfaults when loading libpthread
437 [21783] localedata: Fix int_select international_call_prefixes
438 [21784] localedata: Inconsistency in country_isbn
439 [21788] localedata: Missing Country Postal Abbreviations
440 [21794] localedata: Added-country_isbn-for-Italy
441 [21795] localedata: Add/Fix country_isbn for France
442 [21796] localedata: Added country_isbn for Republic of Korea
443 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
444 [21799] localedata: Added int_select international_call_prefixes
445 [21801] localedata: Added int_select international_call_prefixes
446 [21804] nptl: Double semicolon in thread-shared-types.h
447 [21807] localedata: LC_ADDRESS fix for pap_CW
448 [21808] localedata: Fix LC_ADDRESS for pap_AW
449 [21821] localedata: Added country_name in mai_IN
450 [21822] localedata: Fix LC_TIME for mai_IN
451 [21823] localedata: missing yesstr, nostr for sa_IN
452 [21825] localedata: Fix name_mrs for mag_IN
453 [21828] localedata: 2.26 changelog should mention user visible changes
454 with unicode 9.0
455 [21835] localedata: Added Maithili language locale for Nepal
456 [21838] localedata: Removed redundant data for the_NP
457 [21839] localedata: Fix LC_MONETARY for ta_LK
458 [21844] localedata: Fix Latin characters and Months Sequence.
459 [21848] localedata: Fix mai_NP Title Name
460
461 \f
462 Version 2.25
463
464 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
465 24731-2:2010, is supported to enable declarations of functions from that
466 TR. Note that not all functions from that TR are supported by the GNU C
467 Library.
468
469 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
470 18661-1:2014, is supported to enable declarations of functions and macros
471 from that TS. Note that not all features from that TS are supported by
472 the GNU C Library.
473
474 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
475 18661-4:2015, is supported to enable declarations of functions and macros
476 from that TS. Note that most features from that TS are not supported by
477 the GNU C Library.
478
479 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
480 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
481 Since the GNU C Library defaults to a much newer revision of POSIX, this
482 will only affect programs that specifically request an old conformance
483 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
484 see a change in the visible declarations, but a program compiled with
485 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
486 will not.
487
488 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
489 defined by all multithreaded code, but glibc has not required this for
490 many years.
491
492 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
493 means that in a future release, the macros “major”, “minor”, and “makedev”
494 will only be available from <sys/sysmacros.h>.
495
496 These macros are not part of POSIX nor XSI, and their names frequently
497 collide with user code; see for instance glibc bug 19239 and Red Hat bug
498 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
499 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
500 problem.
501
502 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
503 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
504 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
505
506 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
507 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
508 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
509 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
510 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
511 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
512 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
513 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
514 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
515 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
516 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
517 WINT_WIDTH.
518
519 * New <math.h> features are added from TS 18661-1:2014:
520
521 - Signaling NaN macros: SNANF, SNAN, SNANL.
522
523 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
524 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
525 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
526
527 - llogb functions: the llogb, llogbf and llogbl functions, and the
528 FP_LLOGB0 and FP_LLOGBNAN macros.
529
530 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
531 fminmagf, fminmagl.
532
533 - Comparison macros: iseqsig.
534
535 - Classification macros: iscanonical, issubnormal, iszero.
536
537 - Total order functions: totalorder, totalorderf, totalorderl,
538 totalordermag, totalordermagf, totalordermagl.
539
540 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
541
542 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
543 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
544
545 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
546 are added to libc. They convert a floating-point number into string.
547
548 * Most of glibc can now be built with the stack smashing protector enabled.
549 It is recommended to build glibc with --enable-stack-protector=strong.
550 Implemented by Nick Alcock (Oracle).
551
552 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
553 intended to be used instead of memset() to erase sensitive data after use;
554 the compiler will not optimize out calls to explicit_bzero even if they
555 are "unnecessary" (in the sense that no _correct_ program can observe the
556 effects of the memory clear).
557
558 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
559 to float instead of double. This does not affect the ABI of any libraries
560 that are part of the GNU C Library, but may affect the ABI of other
561 libraries that use this type in their interfaces.
562
563 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
564 float_t and double_t types are now defined to long double instead of float
565 and double. These options are not the default, and this does not affect
566 the ABI of any libraries that are part of the GNU C Library, but it may
567 affect the ABI of other libraries that use this type in their interfaces,
568 if they are compiled or used with those options.
569
570 * The getentropy and getrandom functions, and the <sys/random.h> header file
571 have been added.
572
573 * The buffer size for byte-oriented stdio streams is now limited to 8192
574 bytes by default. Previously, on Linux, the default buffer size on most
575 file systems was 4096 bytes (and thus remains unchanged), except on
576 network file systems, where the buffer size was unpredictable and could be
577 as large as several megabytes.
578
579 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
580 for the Linux quota interface which predates kernel version 2.4.22 has
581 been removed.
582
583 * The malloc_get_state and malloc_set_state functions have been removed.
584 Already-existing binaries that dynamically link to these functions will
585 get a hidden implementation in which malloc_get_state is a stub. As far
586 as we know, these functions are used only by GNU Emacs and this change
587 will not adversely affect already-built Emacs executables. Any undumped
588 Emacs executables, which normally exist only during an Emacs build, should
589 be rebuilt by re-running “./configure; make” in the Emacs build tree.
590
591 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
592 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
593 “no-ip6-dotint” had already been the default, and support for the
594 “ip6-dotint” option was removed from the Internet in 2006.
595
596 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
597 flag from <resolv.h> have been removed. The option relied on a
598 backwards-incompatible DNS extension which was never deployed on the
599 Internet.
600
601 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
602 RES_BLAST defined in the <resolv.h> header file have been deprecated.
603 They were already unimplemented.
604
605 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
606 _res.flags are deprecated. The flag was standardized in RFC 2133, but
607 removed again from the IETF name lookup interface specification in RFC
608 2553. Applications should use getaddrinfo instead.
609
610 * DNSSEC-related declarations and definitions have been removed from the
611 <arpa/nameser.h> header file, and libresolv will no longer attempt to
612 decode the data part of DNSSEC record types. Previous versions of glibc
613 only implemented minimal support for the previous version of DNSSEC, which
614 is incompatible with the currently deployed version.
615
616 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
617 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
618 <arpa/nameser.h> header file because the distinction between RR types and
619 meta-RR types is not officially standardized, subject to revision, and
620 thus not suitable for encoding in a macro.
621
622 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
623 and rhook members of the res_state type in <resolv.h> have been removed.
624 The glibc stub resolver did not support these hooks, but the header file
625 did not reflect that.
626
627 * For multi-arch support it is recommended to use a GCC which has
628 been built with support for GNU indirect functions. This ensures
629 that correct debugging information is generated for functions
630 selected by IFUNC resolvers. This support can either be enabled by
631 configuring GCC with '--enable-gnu-indirect-function', or by
632 enabling it by default by setting 'default_gnu_indirect_function'
633 variable for a particular architecture in the GCC source file
634 'gcc/config.gcc'.
635
636 * GDB pretty printers have been added for mutex and condition variable
637 structures in POSIX Threads. When installed and loaded in gdb these pretty
638 printers show various pthread variables in human-readable form when read
639 using the 'print' or 'display' commands in gdb.
640
641 * Tunables feature added to allow tweaking of the runtime for an application
642 program. This feature can be enabled with the '--enable-tunables' configure
643 flag. The GNU C Library manual has details on usage and README.tunables has
644 instructions on adding new tunables to the library.
645
646 * A new version of condition variables functions have been implemented in
647 the NPTL implementation of POSIX Threads to provide stronger ordering
648 guarantees.
649
650 * A new version of pthread_rwlock functions have been implemented to use a more
651 scalable algorithm primarily through not using a critical section anymore to
652 make state changes.
653
654 Security related changes:
655
656 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
657 have been created with makecontext could fail to terminate due to a
658 missing .cantunwind annotation. This has been observed to lead to a hang
659 (denial of service) in some Go applications compiled with gccgo. Reported
660 by Andreas Schwab. (CVE-2016-6323)
661
662 * The DNS stub resolver functions would crash due to a NULL pointer
663 dereference when processing a query with a valid DNS question type which
664 was used internally in the implementation. The stub resolver now uses a
665 question type which is outside the range of valid question type values.
666 (CVE-2015-5180)
667
668 The following bugs are resolved with this release:
669
670 [4099] stdio: Overly agressive caching by stream i/o functions.
671 [7065] build: Support building glibc with -fstack-protector or -fstack-
672 protector-all
673 [9842] localedata: en_CA: incorrect date format
674 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
675 before it started waiting
676 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
677 [15765] nptl: sem_open is wrongly a cancellation point
678 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
679 [16458] libc: endian.h and netinet/in.h byte order macros return values of
680 wrong type
681 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
682 library linked with pthread
683 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
684 "FIXME: Ingo" issue)
685 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
686 [17252] libc: getrandom and getentropy syscall
687 [17730] malloc: thread-local storage is sometimes improperly free()'d
688 after being __libc_memalign()'d
689 [18241] stdio: failed fseek on memstream does not set errno and can fail
690 when it shouldnt
691 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
692 not
693 [18463] nptl: pthread_cond_broadcast issue when surrounded by
694 PTHREAD_PRIO_INHERIT mutex on ARM
695 [18784] network: res_query and related function crash for special record
696 type queries (CVE-2015-5180)
697 [19380] math: strtod does not raise "inexact"
698 [19387] string: Integer overflow in memchr
699 [19390] string: Integer overflow in strncat
700 [19398] build: linknamespace tests fail with massively parallel build
701 [19402] nptl: Deadlock with robust shared mutex and asynchronous
702 termination
703 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
704 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
705 [19514] libc: [PATCH] Fix spelling errors in spelling
706 "implement"/"implementation" in several places
707 [19582] network: Deprecate RES_USE_INET6
708 [19673] manual: clog10 docs appear to be erroneous
709 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
710 causes a segmentation fault
711 [19826] libc: invalid pointer returned from __tls_get_addr with static
712 linking
713 [20016] network: resolv: Remove hooks support from the API
714 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
715 uninitialized GOT
716 [20033] math: [x86_64] vectorized math function don't call the __finite
717 versions
718 [20116] nptl: use after free in pthread_create
719 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
720 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
721 is always true.
722 [20311] nptl: please install proc_service.h
723 [20366] libc: Compilation errors in installed headers in strict-compliance
724 modes
725 [20370] malloc: malloc: Arena free list management is still racy
726 (incorrect fix in bug 19243)
727 [20386] libc: assert (X = 0) does not result in GCC warning
728 [20432] malloc: malloc: Minimize interface required for interposition
729 [20435] libc: Missing unwind info in __startcontext causes infinite loop
730 in _Unwind_Backtrace (CVE-2016-6323)
731 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
732 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
733 libpthread.a
734 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
735 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
736 [20477] network: resolv: incorrect double-checked locking related to
737 _res_hconf
738 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
739 information.
740 [20495] math: x86_64 performance degradation due to AVX/SSE transition
741 penalty
742 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
743 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
744 cause transition penalty
745 [20517] math: sparcv9 missing fdiml compat symbol
746 [20524] manual: strverscmp is inconsistent
747 [20525] libc: <sys/quota.h> should be based on kernel headers
748 [20539] math: GCC 7 -static -lm fails to link at -O3
749 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
750 mcount.oS)
751 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
752 Checking
753 [20591] network: Remove obsolete DNSSEC support
754 [20592] network: DNS resource record type classification macros in
755 <arpa/nameser.h> are incorrect
756 [20593] network: Update DNS RR type definitions
757 [20611] network: getaddrinfo accepts invalid numeric scope IDs
758 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
759 multi-arch
760 [20629] network: libresolv: Remove support for bitstring labels
761 (RES_USEBSTRING)
762 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
763 assertion: "maxidx >= 2"
764 [20660] math: [arm] Use VSQRT
765 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
766 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
767 [20689] libc: Test for FMA should also check for AVX.
768 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
769 [20715] math: iszero macro breaks existing code
770 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
771 [20728] libc: powerpc: Missing TOC stub in clone
772 [20729] build: glibc-2.24 fails to build for i486 with -Os
773 [20750] build: Build fails with default PIE enabled gcc-6.x
774 [20768] math: [alpha] sqrt fegetenv namespace
775 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
776 glibc
777 [20787] math: float_t is defined as float incorrectly on x86_64 even if
778 __FLT_EVAL_METHOD__ is 2
779 [20790] network: rpcgen buffer overrun in get_prog_declaration
780 [20822] nptl: race condition in __lll_unlock_elision on powerpc
781 [20829] libc: crypt snprintf namespace
782 [20847] libc: tst-vfork3 failure
783 [20855] math: Default bits/mathdef.h has inappropriate float_t
784 [20859] math: [sh4] FP_ILOGB0 invalid
785 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
786 U+20AC), not same as GBK
787 [20915] dynamic-link: global-dynamic TLS broken on aarch64
788 [20916] math: pow handling of sNaN arguments
789 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
790 [20919] math: Bad pow (qNaN, 0) result with -lieee
791 [20940] math: hypot sNaN handling
792 [20947] math: fmax, fmin sNaN handling
793 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
794 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
795 RES_USE_INET6
796 [20971] string: powerpc64/power7 memchr overflows internal pointer check
797 [20973] nptl: robust mutexes: Lost wake-ups
798 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
799 [20978] nis: strlen on null pointer in nss_nisplus
800 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
801 wrong condition
802 [21014] string: i686 memchr overflows internal pointer check
803 [21019] libc: [mips] n32 lseek incorrect on overflow
804 [21022] libc: [microblaze] __backtrace get_frame_size namespace
805 [21026] libc: [MIPS] readahead syscall is broken on n64
806 [21028] math: Fallback fesetexceptflag should always succeed
807 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
808 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
809 clang
810 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
811 [21061] librt: [microblaze] librt lost clock_* exports
812 [21073] libc: tunables: insecure environment variables passed to
813 subprocesses with AT_SECURE
814 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
815 \f
816 Version 2.24
817
818 * The minimum Linux kernel version that this version of the GNU C Library
819 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
820 kernel version 2.6.32 or later suffices (on architectures that already
821 required kernel versions more recent than 3.2, those requirements remain
822 unchanged). Linux 3.2 or later kernel headers are required on all
823 architectures.
824
825 * The pap_AN locale has been deleted. This has been deprecated for a long
826 time. It has been replaced by pap_AW & pap_CW, both of which have long
827 been included in previous releases.
828
829 * The readdir_r and readdir64_r functions have been deprecated. It is
830 recommended to use readdir and readdir64 instead.
831
832 * The type “union wait” has been removed. It was deprecated in the early
833 1990s and never part of POSIX. Application code should use the int type
834 instead of “union wait”.
835
836 * A new NSS action is added to facilitate large distributed system
837 administration. The action, MERGE, allows remote user stores like LDAP
838 to be merged into local user stores like /etc/groups in order to provide
839 easy to use, updated, and managed sets of merged credentials. The new
840 action can be used by configuring it in /etc/nsswitch.conf:
841 group: files [SUCCESS=merge] nis
842 Implemented by Stephen Gallagher (Red Hat).
843
844 * The deprecated __malloc_initialize_hook variable has been removed from the
845 API.
846
847 * The long unused localedef --old-style option has been removed. It hasn't
848 done anything in over 16 years. Scripts using this option can safely
849 drop it.
850
851 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
852 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
853 return the next representable value in the direction of positive infinity
854 and the nextdown functions return the next representable value in the
855 direction of negative infinity. These are currently enabled as GNU
856 extensions.
857
858 Security related changes:
859
860 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
861 could result in a stack overflow when getnetbyname was called with an
862 overly long name. (CVE-2016-3075)
863
864 * Previously, getaddrinfo copied large amounts of address data to the stack,
865 even after the fix for CVE-2013-4458 has been applied, potentially
866 resulting in a stack overflow. getaddrinfo now uses a heap allocation
867 instead. Reported by Michael Petlan. (CVE-2016-3706)
868
869 * The glob function suffered from a stack-based buffer overflow when it was
870 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
871 Reported by Alexander Cherepanov. (CVE-2016-1234)
872
873 * The Sun RPC UDP client could exhaust all available stack space when
874 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
875 alloca plugin for GCC. (CVE-2016-4429)
876
877 * The IPv6 name server management code in libresolv could result in a memory
878 leak for each thread which is created, performs a failing naming lookup,
879 and exits. Over time, this could result in a denial of service due to
880 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
881
882 The following bugs are resolved with this release:
883
884 [1170] localedata: ne_NP: update Nepali locale definition file
885 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
886 of MS-DOS.
887 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
888 [6796] math: fdim() does not set errno on overflow
889 [10354] libc: posix_spawn should use vfork() in more cases than presently
890 [11213] localedata: localedata: add copyright disclaimer to locale files
891 [12143] localedata: chr_US: new Cherokee locale
892 [12450] localedata: sgs_LT: new locale
893 [12676] localedata: ln_CD: new locale
894 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
895 CLDR data
896 [13304] math: fma, fmaf, fmal produce wrong results
897 [14259] build: --localedir arg to configure is ignored
898 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
899 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
900 [14934] localedata: es_CL: wrong first weekday chilean locale
901 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
902 romanisation
903 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
904 and +/-
905 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
906 [15368] nptl: raise() is not async-signal-safe
907 [15479] math: ceil, floor, round and trunc raise inexact exception
908 [15578] localedata: kk_KZ: various updates
909 [16003] localedata: pap_AN: punt old locale
910 [16137] localedata: iw_IL: punt old locale
911 [16190] localedata: eo: new esperanto locale
912 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
913 [16742] malloc: race condition: pthread_atfork() called before first
914 malloc() results in unexpected locking behaviour/deadlocks
915 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
916 all locales
917 [16983] localedata: postal_fmt does not allow %l and %n modifiers
918 [17565] localedata: pt_PT: wrong (work-)week start
919 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
920 [17950] build: Build fails with -msse
921 [18205] localedata: be_BY*: wrong first_weekday and first_workday
922 [18433] libc: posix_spawn does not return correctly upon failure to
923 execute
924 [18453] localedata: charmaps/IBM875: incorrect codes
925 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
926 -Wsystem-headers
927 [18896] localedata: he_IL: improvements for currency
928 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
929 Romanian locale data
930 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
931 symbol
932 [19056] libc: Deprecate readdir_r
933 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
934 language
935 [19198] localedata: nl_NL: small improvements for Dutch locales
936 [19257] network: Per-thread memory leak in __res_vinit with IPv6
937 nameservers (CVE-2016-5417)
938 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
939 machine
940 [19400] locale: Language missing in "iso-639.def", trivial fix in
941 description
942 [19431] malloc: Deadlock between fflush, getdelim, and fork
943 [19505] libc: Incorrect file descriptor validity checks in
944 posix_spawn_file_actions_add{open,close,dup2}
945 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
946 when using RTLD_NEXT
947 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
948 `intl/{gettextP,loadinfo}.h'
949 [19534] libc: execle, execlp may use malloc
950 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
951 [19573] network: res_nclose and __res_maybe_init disagree about name
952 server initialization, breaking Hesiod
953 [19575] localedata: Status of GB18030 tables
954 [19581] localedata: sr_* date_fmt string contains additional newline
955 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
956 Excavator core
957 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
958 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
959 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
960 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
961 double range
962 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
963 part zero incorrect
964 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
965 equality tests
966 [19610] dynamic-link: ldconfig -X removes stale symbolic links
967 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
968 [19633] locale: strfmon_l applies global locale to number formatting
969 [19642] network: Memory leak in getnameinfo
970 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
971 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
972 glibc-2.22
973 [19654] math: [x86_64] Need testcase for BZ #19590 fix
974 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
975 'tst-numeric.c'
976 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
977 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
978 low part
979 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
980 result
981 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
982 unaligned stack
983 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
984 pointers and lengths in error-case.
985 [19727] locale: Converting from/to UTF-xx with iconv() does not always
986 report errors on UTF-16 surrogates values.
987 [19755] nscd: nscd assertion failure in gc
988 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
989 [19759] libc: mempcpy shouldn't be inlined
990 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
991 [19765] libc: s390 needs an optimized mempcpy
992 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
993 NAME_MAX limit assumption (CVE-2016-1234)
994 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
995 [19787] network: Missing and incorrect truncation checks in getnameinfo
996 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
997 modes
998 [19791] network: Assertion failure in res_query.c with un-connectable name
999 server addresses
1000 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1001 [19822] math: libm.so install clobbers old version
1002 [19825] network: resolv: send_vc can return uninitialized data in second
1003 response to getaddrinfo
1004 [19830] network: nss_dns: should check RDATA length against buffer length
1005 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1006 confronted with A/AAAA records of invalid size
1007 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1008 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1009 [19853] stdio: Printing IBM long double in decimal with high precision is
1010 sometimes incorrect
1011 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1012 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1013 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1014 record types
1015 [19865] network: Assertion failure or memory leak in
1016 _nss_dns_getcanonname_r
1017 [19868] network: nss_dns: netent code does not skip over non-PTR records
1018 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1019 (CVE-2016-3075)
1020 [19881] string: Improve x86-64 memset
1021 [19907] string: Incorrect memcpy tests
1022 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1023 resolving symbols
1024 [19925] libc: termios.h XCASE namespace
1025 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1026 [19929] libc: limits.h NL_NMAX namespace
1027 [19931] stdio: Memory leak in vfprintf
1028 [19957] libc: clone(CLONE_VM) access invalid parent memory
1029 [19963] localedata: en_IL: New locale
1030 [19989] stdio: stdio.h cuserid namespace
1031 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1032 gethosts
1033 [19996] locale: langinfo.h nl_langinfo_l namespace
1034 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1035 [20010] network: getaddrinfo: Stack overflow in hostent translation
1036 (CVE-2016-3706)
1037 [20012] stdio: libio: fmemopen append mode failure
1038 [20014] stdio: stdio.h namespace for pre-threads POSIX
1039 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1040 [20023] libc: fcntl.h timespec namespace
1041 [20024] math: [x86_64] vectorized sincos trashes the stack
1042 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1043 [20041] time: sys/time.h timespec namespace
1044 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1045 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1046 [20051] libc: ttyslot in wrong header under wrong conditions
1047 [20054] libc: gethostname not declared for XPG4
1048 [20055] libc: termios.h missing tcgetsid for XPG4
1049 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1050 executable
1051 [20073] libc: sys/stat.h fchmod namespace
1052 [20074] libc: stdlib.h rand_r namespace
1053 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1054 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1055 XPG3
1056 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1057 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1058 (CVE-2016-4429)
1059 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1060 [20119] libc: Wrong mask for processors level type from CPUID
1061 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1062 AS not supporting AVX512
1063 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1064 sNaN argument
1065 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1066 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1067 argument
1068 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1069 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1070 [20178] libc: posix_spawn{p} should not call exit
1071 [20191] stdio: libio: vtables hardening
1072 [20195] string: FMA4 detection requires CPUID execution with register
1073 eax=0x80000001
1074 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1075 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1076 subnormals
1077 [20212] math: acos (sNaN) returns sNaN
1078 [20213] math: asin (sNaN) returns sNaN
1079 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1080 [20218] math: [i386] asinhl (sNaN) returns sNaN
1081 [20219] math: [i386] atanhl (sNaN) returns sNaN
1082 [20222] stdio: fopencookie: Mangle function pointers
1083 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1084 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1085 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1086 input
1087 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1088 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1089 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1090 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1091 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1092 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1093 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1094 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1095 [20240] math: modf (sNaN) returns sNaN
1096 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1097 [20250] math: frexp (sNaN) returns sNaN
1098 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1099 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1100 double rounding
1101 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1102 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1103 subscript is below array bounds [-Werror=array-bounds]
1104 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1105 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1106 (Only arm/linux)
1107 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1108 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1109 fallbacks
1110 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1111 "invalid" exceptions
1112 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1113 [20316] localedata: id_ID: Februari instead of Pebruari
1114 [20327] string: POWER8 strcasecmp returns incorrect result
1115 [20347] math: Failure: Test: j0_downward (0xap+0)
1116 [20348] libc: FAIL: misc/tst-preadvwritev64
1117 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1118 [20350] libc: There is no test for p{read,write}64
1119 [20357] math: Incorrect cos result for 1.5174239687223976
1120 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1121 \f
1122 Version 2.23
1123
1124 * Unicode 8.0.0 Support: Character encoding, character type info, and
1125 transliteration tables are all updated to Unicode 8.0.0, using new
1126 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1127 These updates cause user visible changes, such as the fixes for bugs
1128 89, 16061, and 18568.
1129
1130 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1131 kernel-internal CPU set size. This means that requests that change the
1132 CPU affinity which failed before (for example, an all-ones CPU mask) will
1133 now succeed. Applications that need to determine the effective CPU
1134 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1135 setting it because the kernel can adjust it (and the previous size check
1136 would not detect this in the majority of cases).
1137
1138 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1139 the following new symbols are used: fts64_children, fts64_close,
1140 fts64_open, fts64_read and fts64_set.
1141
1142 * getaddrinfo now detects certain invalid responses on an internal netlink
1143 socket. If such responses are received, an affected process will
1144 terminate with an error message of "Unexpected error <number> on netlink
1145 descriptor <number>" or "Unexpected netlink response of size <number> on
1146 descriptor <number>". The most likely cause for these errors is a
1147 multi-threaded application which erroneously closes and reuses the netlink
1148 file descriptor while it is used by getaddrinfo.
1149
1150 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1151 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1152 unnecessary serialization of memory allocation requests across threads.
1153 The defect is now corrected. Users should see a substantial increase in
1154 the concurent throughput of allocation requests for applications which
1155 trigger this bug. Affected applications typically create and
1156 destroy threads frequently. (Bug 19048 was reported and analyzed by
1157 Ericsson.)
1158
1159 * There is now a --disable-timezone-tools configure option for disabling the
1160 building and installing of the timezone related utilities (zic, zdump, and
1161 tzselect). This is useful for people who build the timezone data and code
1162 independent of the GNU C Library.
1163
1164 * The obsolete header <regexp.h> has been removed. Programs that require
1165 this header must be updated to use <regex.h> instead.
1166
1167 * The obsolete functions bdflush, create_module, get_kernel_syms,
1168 query_module and uselib are no longer available to newly linked binaries;
1169 the header <sys/kdaemon.h> has been removed. These functions and header
1170 were specific to systems using the Linux kernel and could not usefully be
1171 used with the GNU C Library on systems with version 2.6 or later of the
1172 Linux kernel.
1173
1174 * Optimized string, wcsmbs and memory functions for IBM z13.
1175 Implemented by Stefan Liebler.
1176
1177 * Newly linked programs that define a variable called signgam will no longer
1178 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1179 require signgam to be set by those functions must ensure that they use the
1180 variable provided by the GNU C Library and declared in <math.h>, without
1181 defining their own copy.
1182
1183 * The minimum GCC version that can be used to build this version of the GNU
1184 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1185 still be used to compile programs using the GNU C Library.
1186
1187 Security related changes:
1188
1189 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1190 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1191
1192 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1193 the pointer guard feature. It is always enabled. Previously,
1194 LD_POINTER_GUARD could be used to disable security hardening in binaries
1195 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1196 (CVE-2015-8777)
1197
1198 * An integer overflow in hcreate and hcreate_r could lead to an
1199 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1200
1201 * The catopen function no longer has unbounded stack usage. Reported by
1202 Max. (CVE-2015-8779)
1203
1204 * The nan, nanf and nanl functions no longer have unbounded stack usage
1205 depending on the length of the string passed as an argument to the
1206 functions. Reported by Joseph Myers. (CVE-2014-9761)
1207
1208 * A stack-based buffer overflow was found in libresolv when invoked from
1209 libnss_dns, allowing specially crafted DNS responses to seize control
1210 of execution flow in the DNS client. The buffer overflow occurs in
1211 the functions send_dg (send datagram) and send_vc (send TCP) for the
1212 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1213 family. The use of AF_UNSPEC triggers the low-level resolver code to
1214 send out two parallel queries for A and AAAA. A mismanagement of the
1215 buffers used for those queries could result in the response of a query
1216 writing beyond the alloca allocated buffer created by
1217 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1218 the overflow. Thanks to the Google Security Team and Red Hat for
1219 reporting the security impact of this issue, and Robert Holiday of
1220 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1221
1222 The following bugs are resolved with this release:
1223
1224 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1225 [887] math: Math library function "logb" and "nextafter" inconsistent
1226 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1227 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1228 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1229 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1230 use `mkstemp'
1231 [4404] localedata: German translation of "Alarm clock" is misleading
1232 [6799] math: nextafter() and nexttoward() doen't set errno on
1233 overflow/underflow errors
1234 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1235 overflow/underflow
1236 [10432] nis: _nss_nis_setnetgrent assertion failure
1237 [11460] libc: fts has no LFS support
1238 [12926] network: getaddrinfo()/make_request() may spin forever
1239 [13065] nptl: Race condition in pthread barriers
1240 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1241 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1242 are not contiguous
1243 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1244 [14912] libc: Rename non-installed bits/*.h headers
1245 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1246 [15367] math: Let gcc use __builtin_isinf
1247 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1248 [15421] math: lgamma wrongly sets signgam for ISO C
1249 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1250 FE_INVALID with argument out of range
1251 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1252 all exceptions
1253 [15786] dynamic-link: ifunc resolver functions can smash function
1254 arguments
1255 [15918] math: Unnecessary check for equality in hypotf()
1256 [16061] localedata: Review / update transliteration data
1257 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1258 should include
1259 [16141] time: strptime %z offset restriction
1260 [16171] math: drem should be alias of remainder
1261 [16296] math: fegetround is pure?
1262 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1263 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1264 GNU/Linux
1265 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1266 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1267 arguments
1268 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1269 "Exception "Inexact" set" on ppc32
1270 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1271 [16517] math: Missing underflow exception from tanf/tan/tanl
1272 [16519] math: Missing underflow exception from sinhf
1273 [16520] math: Missing underflow exception from tanhf
1274 [16521] math: Missing underflow exception from exp2
1275 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1276 rounding results
1277 [16734] stdio: fopen calls mmap to allocate its buffer
1278 [16961] math: nan function incorrect handling of bad sequences
1279 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1280 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1281 [16985] locale: localedef: confusing error message when opening output
1282 fails
1283 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1284 [17197] locale: Redundant shift character in iconv conversion output at
1285 block boundary
1286 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1287 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1288 [17250] dynamic-link: static linking breaks nss loading
1289 (getaddrinfo/getpwnam/etc...)
1290 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1291 4.7?
1292 [17441] math: isnan() should use __builtin_isnan() in GCC
1293 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1294 (related to lock elision)
1295 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1296 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1297 [17887] time: strptime should be able to parse "+01:00" style timezones
1298 [17905] libc: catopen() Multiple unbounded stack allocations
1299 (CVE-2015-8779)
1300 [18084] libc: backtrace (..., 0) dumps core on x86
1301 [18086] libc: nice() sets errno to 0 on success
1302 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1303 is too large (CVE-2015-8778)
1304 [18251] dynamic-link: SONAME missing when audit modules provides path
1305 [18265] libc: add attributes for wchar string and memory functions
1306 [18370] math: csqrt missing underflows
1307 [18421] libc: [hppa] read-only segment has dynamic relocations
1308 [18472] libc: Obsolete syscall wrappers should be compat symbols
1309 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1310 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1311 [18525] localedata: Remove locale timezone information
1312 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1313 [18568] localedata: Update locale data to Unicode 8.0
1314 [18589] locale: sort-test.sh fails at random
1315 [18595] math: ctan, ctanh missing underflows
1316 [18604] libc: assert macro-expands its argument
1317 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1318 contains a vector instruction exception.
1319 [18611] math: j1, jn missing errno setting on underflow
1320 [18618] localedata: sync Chechen locale definitions with other *_RU
1321 locales
1322 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1323 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1324 [18665] network: In send_dg, the recvfrom function is NOT always using the
1325 buffer size of a newly created buffer (CVE-2015-7547)
1326 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1327 missing break ?
1328 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1329 32bit processes
1330 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1331 [18699] math: tilegx cproj() for various complex infinities does not yield
1332 infinity
1333 [18724] libc: Harden put*ent functions against data injection
1334 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1335 [18755] build: build errors with -DNDEBUG
1336 [18757] stdio: fmemopen fails to set errno on failure
1337 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1338 be forced unloaded
1339 [18781] libc: openat64 lacks O_LARGEFILE
1340 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1341 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1342 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1343 [18790] math: [ldbl-128ibm] tanhl inaccurate
1344 [18795] libc: stpncpy fortification misses buffer lengths that are
1345 statically too large
1346 [18796] build: build fails for --disable-mathvec
1347 [18803] math: hypot missing underflows
1348 [18820] stdio: fmemopen may leak memory on failure
1349 [18823] math: csqrt spurious underflows
1350 [18824] math: fma spurious underflows
1351 [18825] math: pow missing underflows
1352 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1353 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1354 [18870] build: sem_open.c fails to compile with missing symbol
1355 FUTEX_SHARED
1356 [18872] stdio: Fix memory leak in printf_positional
1357 [18873] libc: posix_fallocate overflow check ineffective
1358 [18875] math: Excess precision leads incorrect libm
1359 [18877] libc: arm: mmap offset regression
1360 [18887] libc: memory corruption when using getmntent on blank lines
1361 [18918] localedata: hu_HU: change time to HH:MM:SS format
1362 [18921] libc: Regression: extraneous stat() and fstat() performed by
1363 opendir()
1364 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1365 binaries (CVE-2015-8777)
1366 [18951] math: tgamma missing underflows
1367 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1368 signgam
1369 [18953] localedata: lt_LT: change currency symbol to the euro
1370 [18956] math: powf inaccuracy
1371 [18961] math: [i386] exp missing underflows
1372 [18966] math: [i386] exp10 missing underflows
1373 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1374 [18969] build: multiple string test failures due to missing locale
1375 dependencies
1376 [18970] libc: Reference of pthread_setcancelstate in libc.a
1377 [18977] math: float / long double Bessel functions not in XSI POSIX
1378 [18980] math: i386 libm functions return with excess range and precision
1379 [18981] math: i386 scalb*, ldexp return with excess range and precision
1380 [18982] stdio: va_list and vprintf
1381 [18985] time: Passing out of range data to strftime() causes a segfault
1382 (CVE-2015-8776)
1383 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1384 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1385 [19012] locale: iconv_open leaks memory on error path
1386 [19016] math: clog, clog10 inaccuracy
1387 [19018] nptl: Mangle function pointers in tls_dtor_list
1388 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1389 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1390 [19048] malloc: malloc: arena free list can become cyclic, increasing
1391 contention
1392 [19049] math: [powerpc] erfc incorrect zero sign
1393 [19050] math: [powerpc] log* incorrect zero sign
1394 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1395 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1396 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1397 [19074] network: Data race in _res_hconf_reorder_addrs
1398 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1399 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1400 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1401 rounding modes
1402 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1403 ILP32
1404 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1405 [19086] manual: posix_fallocate64 documented argument order is wrong.
1406 [19088] math: lround, llround missing exceptions close to overflow
1407 threshold
1408 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1409 [19095] math: dbl-64 lrint incorrect for 64-bit long
1410 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1411 [19124] dynamic-link: ld.so failed to build with older assmebler
1412 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1413 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1414 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1415 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1416 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1417 pthread_setaffinity_np
1418 [19156] math: [ldbl-128] j0l spurious underflows
1419 [19164] nptl: tst-getcpu fails with many possible CPUs
1420 [19168] math: math/test-ildoubl and math/test-ldouble failure
1421 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1422 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1423 prelink
1424 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1425 (FE_NOMASK_ENV) do not clear SSE exceptions
1426 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1427 _int_new_arena/reused_arena
1428 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1429 [19201] math: dbl-64 remainder incorrect sign of zero result
1430 [19205] math: bits/math-finite.h conditions do not match math.h and
1431 bits/mathcalls.h
1432 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1433 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1434 for C99-based standards
1435 [19212] libc: features.h not -Wundef clean
1436 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1437 math-only
1438 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1439 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1440 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1441 disabled
1442 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1443 "inexact" exceptions
1444 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1445 arguments
1446 [19242] libc: strtol incorrect in Turkish locales
1447 [19243] malloc: reused_arena can pick an arena on the free list, leading
1448 to an assertion failure and reference count corruption
1449 [19253] time: tzset() ineffective when temporary TZ did not include DST
1450 rules
1451 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1452 [19270] math: [hppa] Shared libm missing __isnanl
1453 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1454 MAP_HUGETLB and MAP_STACK defines
1455 [19313] nptl: Wrong __cpu_mask for x32
1456 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1457 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1458 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1459 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1460 [19363] time: x32: times() return value wrongly truncates/sign extends
1461 from 32bit
1462 [19367] dynamic-link: Improve branch prediction on Silvermont
1463 [19369] network: Default domain name not reset by res_ninit when "search"
1464 / "domain" entry is removed from resolv.conf
1465 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1466 subnormals
1467 [19385] localedata: bg_BG: time separator should be colon, not comma
1468 [19408] libc: linux personality syscall wrapper may erroneously return an
1469 error on 32-bit architectures
1470 [19415] libc: dladdr returns wrong names on hppa
1471 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1472 IBM905, IBM907, and IBM909
1473 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1474 [19443] build: build failures with -DDEBUG
1475 [19451] build: Make check fails on test-double-vlen2
1476 [19462] libc: Glibc failed to build with -Os
1477 [19465] math: Wrong code with -Os
1478 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1479 -Os
1480 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1481 CPU's.
1482 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1483 [19486] math: S390: Math tests fail with "Exception Inexact set".
1484 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1485 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1486 architectures
1487 [19590] math: Fail to build shared objects that use libmvec.so functions.
1488 \f
1489 Version 2.22
1490
1491 * The following bugs are resolved with this release:
1492
1493 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1494 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1495 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1496 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1497 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1498 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1499 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1500 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1501 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1502 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1503 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1504 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1505 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1506 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1507 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1508 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1509 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1510 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1511 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1512 18657, 18676, 18694, 18696, 18887.
1513
1514 * Cache information can be queried via sysconf() function on s390 e.g. with
1515 _SC_LEVEL1_ICACHE_SIZE as argument.
1516
1517 * A buffer overflow in gethostbyname_r and related functions performing DNS
1518 requests has been fixed. If the NSS functions were called with a
1519 misaligned buffer, the buffer length change due to pointer alignment was
1520 not taken into account. This could result in application crashes or,
1521 potentially arbitrary code execution, using crafted, but syntactically
1522 valid DNS responses. (CVE-2015-1781)
1523
1524 * The time zone file parser has been made more robust against crafted time
1525 zone files, avoiding heap buffer overflows related to the processing of
1526 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1527 large time zone data files. Overly long time zone specifiers in the TZ
1528 variable no longer result in stack overflows and crashes.
1529
1530 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1531 for LD and GD on x86 and x86-64, has been implemented. You will need
1532 binutils-2.24 or later to enable this optimization.
1533
1534 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1535 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1536 Hat). These updates cause user visible changes, such as the fix for bug
1537 17998.
1538
1539 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1540 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1541 condition in some applications.
1542
1543 * Added vector math library named libmvec with the following vectorized x86_64
1544 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1545 pow, powf.
1546 The library can be disabled with --disable-mathvec. Use of the functions is
1547 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1548 Shared library libmvec.so is linked in as needed when using -lm (no need to
1549 specify -lmvec explicitly for not static builds).
1550 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1551
1552 * A new fmemopen implementation has been added with the goal of POSIX
1553 compliance. The new implementation fixes the following long-standing
1554 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1555 old implementation is still present for use be by existing binaries.
1556
1557 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1558 releases. It has been fixed to match 2.19 and older, but binaries built
1559 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1560
1561 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1562 Contributed by Roland McGrath (Google).
1563
1564 * The header <regexp.h> is deprecated, and will be removed in a future
1565 release. Use of this header will trigger a deprecation warning.
1566 Application developers should update their code to use <regex.h> instead.
1567
1568 This header was formerly part of SUS, but was deprecated in 1994 and
1569 removed from the standard in 2001. Also, the glibc implementation
1570 leaks memory. See BZ#18681 for more details.
1571 \f
1572 Version 2.21
1573
1574 * The following bugs are resolved with this release:
1575
1576 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1577 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1578 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1579 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1580 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1581 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1582 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1583 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1584 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1585 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1586 17892.
1587
1588 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1589 too little memory for the to-be-scanned arguments and overflow the
1590 allocated buffer. The implementation now correctly computes the required
1591 buffer size when using malloc, and switches to malloc from alloca as
1592 intended.
1593
1594 * A new semaphore algorithm has been implemented in generic C code for all
1595 machines. Previous custom assembly implementations of semaphore were
1596 difficult to reason about or ensure that they were safe. The new version
1597 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1598 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1599 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1600
1601 * Port to Altera Nios II has been contributed by Mentor Graphics.
1602
1603 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1604 implementations for powerpc64/powerpc64le.
1605 Implemented by Adhemerval Zanella (IBM).
1606
1607 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1608 and powerpc64le. This may improve lock scaling of existing programs on
1609 HTM capable systems. The lock elision code is only enabled with
1610 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1611 powerpc will issue a transaction abort on every syscall to avoid side
1612 effects being visible outside transactions.
1613
1614 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1615 AArch64. Contributed by ARM Ltd.
1616
1617 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1618
1619 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1620 under certain input conditions resulting in the execution of a shell for
1621 command substitution when the application did not request it. The
1622 implementation now checks WRDE_NOCMD immediately before executing the
1623 shell and returns the error WRDE_CMDSUB as expected.
1624
1625 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1626 processing format strings with a large number of format specifiers.
1627
1628 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1629 infinite loop if the DNS response contained a PTR record of an unexpected
1630 format.
1631
1632 * The minimum GCC version that can be used to build this version of the GNU
1633 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1634 still be used to compile programs using the GNU C Library.
1635
1636 * The GNU C Library is now built with -Werror by default. This can be
1637 disabled by configuring with --disable-werror.
1638
1639 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1640
1641 * The obsolete sigvec function has been removed. This was the original
1642 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1643 programs have been using instead for about 25 years. Of course, ABI
1644 compatibility for old binaries using sigvec remains intact.
1645
1646 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1647 with newer versions of bison.
1648
1649 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1650 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1651 registers overlay two consecutive single-precision registers. MIPS32R2
1652 introduced a new FPU mode (FR=1) where double-precision registers extend the
1653 corresponding single-precision registers which is incompatible with the
1654 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1655 require the use of FR=1 making a transition necessary. New o32 ABI
1656 extensions enable users to migrate over time from the original o32 ABI
1657 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1658 tracks the ABI of any loaded object and verifies that new objects are
1659 compatible. Mode transitions will also be requested as required and
1660 unsupportable objects will be rejected. The ABI checks include both soft and
1661 hard float ABIs for o32, n32 and n64.
1662
1663 GCC 5 with GNU binutils 2.25 onwards:
1664 It is strongly recommended that all o32 system libraries are built using the
1665 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1666 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1667 compiler using --with-fp-32=xx to set this by default.
1668 \f
1669 Version 2.20
1670
1671 * The following bugs are resolved with this release:
1672
1673 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1674 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1675 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1676 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1677 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1678 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1679 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1680 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1681 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1682 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1683 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1684 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1685 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1686 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1687 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1688
1689 * Reverted change of ABI data structures for s390 and s390x:
1690 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1691 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1692 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1693 are preserved pointing straight to the same implementation as the old ones.
1694 Given that, new callers will simply provide a too-big buffer to these
1695 functions. Any applications/libraries out there that embed jmp_buf or
1696 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1697 against 2.19 headers will have to rebuilt again. This is necessary in any
1698 case to revert the breakage in their ABI caused by the glibc change.
1699
1700 * Support for file description locks is added to systems running the
1701 Linux kernel. The standard file locking interfaces are extended to
1702 operate on file descriptions, not file descriptors, via the use of
1703 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1704 are associated with an open file instead of a process.
1705
1706 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1707
1708 * The minimum Linux kernel version that this version of the GNU C Library
1709 can be used with is 2.6.32.
1710
1711 * Running the testsuite no longer terminates as soon as a test fails.
1712 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1713 with PASS or FAIL lines for individual tests. A summary of the results is
1714 printed, including a list of failing lists, and "make check" exits with
1715 error status if there were any unexpected failures. "make check
1716 stop-on-test-failure=y" may be used to keep the old behavior.
1717
1718 * The am33 port, which had not worked for several years, has been removed
1719 from ports.
1720
1721 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1722 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1723 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1724 conflicted with POSIX (support for which was removed in 2.19), the
1725 interfaces those macros enabled remain available when compiling with
1726 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1727 test macros defined.
1728
1729 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1730
1731 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1732 This may improve lock scaling of existing programs on TX capable systems.
1733 The lock elision code is only built with --enable-lock-elision=yes and
1734 then requires a GCC version supporting the TX builtins. With lock elision
1735 default mutexes are elided via __builtin_tbegin, if the cpu supports
1736 transactions. By default lock elision is not enabled and the elision code
1737 is not built.
1738
1739 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1740 copy the path argument. This allowed programs to cause posix_spawn to
1741 deference a dangling pointer, or use an unexpected pathname argument if
1742 the string was modified after the posix_spawn_file_actions_addopen
1743 invocation.
1744
1745 * All supported architectures now use the main glibc sysdeps directory
1746 instead of some being in a separate "ports" directory (which was
1747 distributed separately before glibc 2.17).
1748
1749 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1750 On configurations that support it (all Linux configurations), it's now
1751 used regardless of the --enable-add-ons switch to configure. It is no
1752 longer possible to build such configurations without pthreads support.
1753
1754 * Locale names, including those obtained from environment variables (LANG
1755 and the LC_* variables), are more tightly checked for proper syntax.
1756 setlocale will now fail (with EINVAL) for locale names that are overly
1757 long, contain slashes without starting with a slash, or contain ".." path
1758 components. (CVE-2014-0475) Previously, some valid locale names were
1759 silently replaced with the "C" locale when running in AT_SECURE mode
1760 (e.g., in a SUID program). This is no longer necessary because of the
1761 additional checks.
1762
1763 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1764 with application code using Intel MPX instructions. (With all previous
1765 versions, the MPX register state could be clobbered when making calls
1766 into or out of a shared library.) Note that while the new dynamic
1767 linker is compatible with all known x86 hardware whether or not it
1768 supports Intel MPX, some x86 instruction-set emulators might fail to
1769 handle the new instruction encodings. This is known to affect Valgrind
1770 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1771 release), and might affect other tools that do instruction emulation.
1772
1773 * Support for loadable gconv transliteration modules has been removed.
1774 The support for transliteration modules has been non-functional for
1775 over a decade, and the removal is prompted by security defects. The
1776 normal gconv conversion modules are still supported. Transliteration
1777 with //TRANSLIT is still possible, and the //IGNORE specifier
1778 continues to be supported. (CVE-2014-5119)
1779
1780 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1781 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1782 resulting a denial-of-service security vulnerability in applications which
1783 use functions related to iconv. (CVE-2014-6040)
1784 \f
1785 Version 2.19
1786
1787 * The following bugs are resolved with this release:
1788
1789 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1790 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1791 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1792 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1793 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1794 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1795 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1796 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1797 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1798 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1799 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1800 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1801 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1802 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1803 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1804 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1805 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1806 16431, 16453, 16474, 16506, 16510, 16529
1807
1808 * Slovenian translations for glibc messages have been contributed by the
1809 Translation Project's Slovenian team of translators.
1810
1811 * The public headers no longer use __unused nor __block. This change is to
1812 support compiling programs that are derived from BSD sources and use
1813 __unused internally, and to support compiling with Clang's -fblock
1814 extension which uses __block.
1815
1816 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1817 large collation sequences to optimize multiple passes. This cache
1818 computation may overflow for large collation sequences and may cause a
1819 stack or buffer overflow. This is now fixed to use a slower algorithm
1820 which does not use a cache if there is an integer overflow.
1821
1822 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1823 rules for large collation sequences to optimize multiple passes and falls
1824 back to alloca if malloc fails, resulting in a possible stack overflow.
1825 The implementation now falls back to an uncached collation sequence lookup
1826 if malloc fails.
1827
1828 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1829 initialized for static applications resulting in the security feature
1830 being disabled. The pointer guard is now correctly initialized to a
1831 random value for static applications. Existing static applications need
1832 to be recompiled to take advantage of the fix (bug 15754).
1833
1834 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1835 to the d_name member of struct dirent, or omit the terminating NUL
1836 character. (Bugzilla #14699).
1837
1838 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1839 aligned_alloc functions could allocate too few bytes or corrupt the
1840 heap when passed very large allocation size values (Bugzilla #15855,
1841 #15856, #15857).
1842
1843 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1844 for AF_INET6 has been fixed (Bugzilla #16072).
1845
1846 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1847 pap_CW, quz_PE, the_NP.
1848
1849 * Substantially revised locales: gd_GB, ht_HT
1850
1851 * The LC_ADDRESS field was updated to support country_car for almost all
1852 supported locales.
1853
1854 * ISO 1427 definitions were updated.
1855
1856 * ISO 3166 definitions were updated.
1857
1858 * The localedef utility now supports --big-endian and --little-endian
1859 command-line options to generate locales for a different system from that
1860 for which the C library was built.
1861
1862 * Binary locale files now only depend on the endianness of the system for
1863 which they are generated and not on other properties of that system. As a
1864 consequence, binary files generated with new localedef may be incompatible
1865 with old versions of the GNU C Library, and binary files generated with
1866 old localedef may be incompatible with this version of the GNU C Library,
1867 in the following circumstances:
1868
1869 + Locale files may be incompatible on m68k systems.
1870
1871 + Locale archive files (but not separate files for individual locales) may
1872 be incompatible on systems where plain "char" is signed.
1873
1874 * The configure option --disable-versioning has been removed. Builds with
1875 --disable-versioning had not worked for several years.
1876
1877 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1878 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1879
1880 * SystemTap probes for malloc have been introduced.
1881
1882 * SystemTap probes for slow multiple precision fallback paths of
1883 transcendental functions have been introduced.
1884
1885 * Support for powerpc64le has been added.
1886
1887 * The soft-float powerpc port now supports e500 processors.
1888
1889 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1890
1891 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1892 set of header declarations that are enabled by default, even when other
1893 feature test macros or compiler options such as -std=c99 would otherwise
1894 disable some of those declarations.
1895
1896 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1897 conflict with POSIX. The libbsd-compat library (which was a dummy library
1898 that did nothing) has also been removed.
1899
1900 * Preliminary documentation about Multi-Thread, Async-Signal and
1901 Async-Cancel Safety has been added.
1902
1903 * Change of ABI data structures for s390 and s390x:
1904 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1905 allow for future hardware extensions. All library functions that accept or
1906 return these structures were versioned in order to provide backwards
1907 compatibility with existing code. However, not all cases can be handled
1908 with interface versioning. If an application embeds either structure into
1909 another structure and that structure is passed to another compilation unit
1910 which expects the newer larger-sized object then any access to the new
1911 fields will result in undefined behaviour. Similarly any access to fields
1912 that were moved by the enlarging of the structures will cause undefined
1913 behaviour. To fix the undefined behaviour all objects that are part of the
1914 final in-memory image must agree on the size of structures, and this may
1915 require recompilation.
1916 \f
1917 Version 2.18
1918
1919 * The following bugs are resolved with this release:
1920
1921 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1922 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1923 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1924 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1925 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1926 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1927 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1928 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1929 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1930 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1931 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1932 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1933 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1934 15755, 15759.
1935
1936 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1937 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1938 Distributions can re-enable building and using pt_chown via the new configure
1939 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1940 considerable security risks and should only be used if the distribution
1941 understands and accepts the risks.
1942
1943 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1944 #15078).
1945
1946 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1947 fixed (Bugzilla #15330).
1948
1949 * Add support for calling C++11 thread_local object destructors on thread
1950 and program exit. This needs compiler support for offloading C++11
1951 destructor calls to glibc.
1952
1953 * Improved worst case performance of libm functions with double inputs and
1954 output.
1955
1956 * Support for priority inherited mutexes in pthread condition variables on
1957 non-x86 architectures.
1958
1959 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1960
1961 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1962
1963 * Optimized string functions for ARM. Implemented by Will Newton and
1964 Richard Henderson.
1965
1966 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1967
1968 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1969 Richard Henderson.
1970
1971 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1972 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1973
1974 * Added a benchmark framework to track performance of functions in glibc.
1975
1976 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1977 It is based on draft TS 18661 and currently enabled as a GNU extension.
1978
1979 * On Linux, the clock function now uses the clock_gettime system call
1980 for improved precision, rather than old times system call.
1981
1982 * Added support for version-3 tz format files. This is needed when using
1983 the tz database release 2013e or later, and affects a few unusual cases --
1984 currently only TZ='America/Godthab' for time stamps after 2037.
1985
1986 * Added new API functions pthread_getattr_default_np and
1987 pthread_setattr_default_np to get and set the default pthread
1988 attributes of a process.
1989
1990 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1991 This may improve lock scaling of existing programs on TSX capable systems.
1992 When the --enable-lock-elision=yes parameter is specified at configure
1993 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1994 mutexes.
1995
1996 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1997 availability). Implemented by Ryan S. Arnold.
1998
1999 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2000 \f
2001 Version 2.17
2002
2003 * The following bugs are resolved with this release:
2004
2005 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2006 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2007 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2008 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2009 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2010 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2011 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2012 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2013 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2014 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2015 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2016 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2017 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2018 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2019 14879, 14889, 14893, 14898, 14914.
2020
2021 * Optimization of memcpy for MIPS.
2022
2023 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2024 EMFILE has been fixed (Bugzilla #14889).
2025
2026 * The add-on ports collection is now distributed in the "ports" subdirectory
2027 of the main GNU C Library distribution, rather than separately.
2028
2029 * Port to ARM AArch64 contributed by Linaro.
2030
2031 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2032 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2033 zEnterprise z196.
2034 Implemented by Andreas Krebbel.
2035
2036 * The new function secure_getenv allows secure access to the environment,
2037 returning NULL if running in a SUID/SGID process. This function replaces
2038 the internal function __secure_getenv.
2039
2040 * SystemTap static probes have been added into the dynamic linker.
2041 Implemented by Gary Benson.
2042
2043 * Optimizations of string functions strstr, strcasestr and memmem.
2044 Implemented by Maxim Kuvyrkov.
2045
2046 * The minimum Linux kernel version that this version of the GNU C Library
2047 can be used with is 2.6.16.
2048
2049 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2050 powerpc POWER7. Implemented by Will Schmidt.
2051
2052 * New configure option --disable-nscd builds the C library such that it
2053 never attempts to contact the Name Service Caching Daemon (nscd).
2054 New configure option --disable-build-nscd avoids building nscd itself;
2055 this is the default if --disable-nscd is used.
2056
2057 * Improved support for cross-compilation, including cross-testing and
2058 bootstrap builds without a previously built glibc.
2059
2060 * Several testsuite tests are now able to test multiple IFUNC variants of an
2061 interface, rather than just testing the one that would be chooen by
2062 default.
2063
2064 * New configure options --with-bugurl and --with-pkgversion, for
2065 distributors to use to embed their bug-reporting and package version
2066 information in --help and --version output.
2067
2068 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2069 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2070 allows creation of chroots without the procfs mounted on /proc.
2071
2072 * The `crypt' function now fails if passed salt bytes that violate the
2073 specification for those values. On Linux, the `crypt' function will
2074 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2075 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2076 when the mode is enabled.
2077
2078 * The `clock_*' suite of functions (declared in <time.h>) is now available
2079 directly in the main C library. Previously it was necessary to link with
2080 -lrt to use these functions. This change has the effect that a
2081 single-threaded program that uses a function such as `clock_gettime' (and
2082 is not linked with -lrt) will no longer implicitly load the pthreads
2083 library at runtime and so will not suffer the overheads associated with
2084 multi-thread support in other code such as the C++ runtime library.
2085
2086 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2087 sat_IN, and szl_PL.
2088 \f
2089 Version 2.16
2090
2091 * The following bugs are resolved with this release:
2092
2093 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2094 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2095 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2096 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2097 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2098 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2099 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2100 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2101 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2102 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2103 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2104 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2105 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2106 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2107 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2108 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2109 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2110 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2111 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2112 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2113 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2114 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2115 14277, 14278.
2116
2117 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2118 configuring glibc with:
2119 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2120 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2121 Implemented by H.J. Lu.
2122
2123 * ISO C11 support:
2124
2125 + define static_assert
2126
2127 + do not declare gets
2128
2129 + declare at_quick_exit and quick_exit also for ISO C11
2130
2131 + aligned_alloc. NB: The code is deliberately allows the size parameter
2132 to not be a multiple of the alignment. This is a moronic requirement
2133 in the standard but it is only a requirement on the caller, not the
2134 implementation.
2135
2136 + timespec_get added
2137
2138 + uchar.h support added
2139
2140 + CMPLX, CMPLXF, CMPLXL added
2141
2142 Implemented by Ulrich Drepper.
2143
2144 * Support for the IA-64 has been moved to ports.
2145
2146 * Remove support for anything but ELF binary format
2147
2148 * Checking versions of poll, ppoll added.
2149 Implemented by Ulrich Drepper.
2150
2151 * More generic and 64-bit performance optimizations to math functions.
2152 Implemented by Ulrich Drepper.
2153
2154 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2155 headers and functions available at compile time as they were before
2156 version 2.14. This option will be removed at some time in the future
2157 after the TI-RPC library becomes fully sufficient for the needs of
2158 existing applications.
2159
2160 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2161 Note that glibc is not expected to work with any Linux kernel version
2162 before 2.6.
2163
2164 * New header <sys/auxv.h> and function getauxval allowing easy access to
2165 the AT_* key-value pairs passed from the Linux kernel. The header also
2166 defines the HWCAP_* bits associated with the AT_HWCAP key.
2167
2168 * New locales: mag_IN
2169
2170 * New configure option --enable-systemtap builds SystemTap static probes
2171 into libc for setjmp and longjmp and into libpthread for various operations.
2172 So far the setjmp/longjmp probes and some of the libpthread probes are
2173 provided only for i*86 and x86_64.
2174 Implemented by Roland McGrath and Rayson Ho.
2175
2176 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2177
2178 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2179 and Will Schmidt.
2180
2181 * More optimized functions for SPARC. Implemented by David S. Miller.
2182
2183 * Improved support for cross-compilation, especially bootstrap builds
2184 without a previously built glibc.
2185
2186 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2187 Chris Metcalf from Tilera.
2188
2189 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2190 now supported for ARM processors.
2191
2192 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2193 as the name of the dynamic linker, to distinguish it from the
2194 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2195
2196 * Support for CRIS has been removed from ports.
2197
2198 * A new class of installed header has been documented for low-level
2199 platform-specific functionality. PowerPC added the first instance with a
2200 function to provide time base register access. Contributed by Tulio
2201 Magno Quites Machado Filho.
2202
2203 * ix86 configurations now install header files that are consistent with
2204 what x86-64 configurations install. These same header files can be used
2205 for -m32, -m64, or -mx32 builds.
2206 Contributed by H.J. Lu.
2207
2208 * Math library bug fixes. A thorough audit of all open math library bugs was
2209 conducted by Joseph Myers. Significant progress was made on many math
2210 library bugs resulting in more accurate exceptions and function results.
2211 Many thanks to all those that contributed including Andreas Jaeger for his
2212 patch review and work on the x87 trigonometric instruction issues.
2213
2214 * Timezone data is no longer installed. Timezone-related binaries and scripts
2215 will continue to be installed. Users should obtain their timezone data from
2216 their distribution provider or from the tzdata package at
2217 <ftp://munnari.oz.au/pub/>.
2218 \f
2219 Version 2.15
2220
2221 * The following bugs are resolved with this release:
2222
2223 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2224 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2225 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2226 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2227 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2228 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2229 13484, 13506, 13515, 13523, 13524, 13538, 13540
2230
2231 * New program pldd to list loaded object of a process
2232 Implemented by Ulrich Drepper.
2233
2234 * Add nss_db support back to glibc. No more dependency on Berkeley db
2235 and support for initgroups lookups.
2236 Implemented by Ulrich Drepper.
2237
2238 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2239 Contributed by HJ Lu.
2240
2241 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2242 Contributed by HJ Lu.
2243
2244 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2245 on x86-32 and x86-64.
2246 Contributed by Liubov Dmitrieva.
2247
2248 * Optimized strchr and strrchr for SSE on x86-32.
2249 Contributed by Liubov Dmitrieva.
2250
2251 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2252 for x86-64 and x86-32.
2253 Contributed by Liubov Dmitrieva.
2254
2255 * New interfaces: scandirat, scandirat64
2256 Implemented by Ulrich Drepper.
2257
2258 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2259 Implemented by Ulrich Drepper.
2260
2261 * nscd now also caches the netgroup database.
2262 Implemented by Ulrich Drepper.
2263
2264 * Integrate libm with gcc's -ffinite-math-only option.
2265 Implemented by Ulrich Drepper.
2266
2267 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2268 to math functions. Implemented by Ulrich Drepper.
2269
2270 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2271 Implemented by Ulrich Drepper.
2272
2273 * New Linux interfaces: process_vm_readv, process_vm_writev
2274
2275 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2276 Implemented by Ulrich Drepper.
2277
2278 * Optimized nearbyint and strcasecmp for PPC.
2279 Implemented by Adhemerval Zanella.
2280
2281 * New locales: bho_IN, unm_US, es_CU, ta_LK
2282 \f
2283 Version 2.14
2284
2285 * The following bugs are resolved with this release:
2286
2287 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2288 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2289 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2290 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2291 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2292 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2293 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2294 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2295 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2296 12795, 12811, 12813, 12814, 12841
2297
2298 * The RPC implementation in libc is obsoleted. Old programs keep working
2299 but new programs cannot be linked with the routines in libc anymore.
2300 Programs in need of RPC functionality must be linked against TI-RPC.
2301 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2302
2303 Visible changes of this change include (obviously) the inability to link
2304 programs using RPC functions without referencing the TI-RPC library and the
2305 removal of the RPC headers from the glibc headers.
2306 Implemented by Ulrich Drepper.
2307
2308 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2309 syncfs, setns, sendmmsg
2310
2311 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2312 yue_HK, lij_IT, mhr_RU
2313
2314 * New encodings: CP770, CP771, CP772, CP773, CP774
2315
2316 * New program sotruss to trace calls through PLTs
2317 Implemented by Ulrich Drepper.
2318
2319 * The malloc hook implementation is marked deprecated and will be removed
2320 from the default implementation in the next version. The design never
2321 worked ever since the introduction of threads. Even programs which do
2322 not create threads themselves can use multiple threads created internally.
2323 \f
2324 Version 2.13
2325
2326 * The following bugs are resolved with this release:
2327
2328 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2329 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2330 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2331 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2332 12378, 12394, 12397
2333
2334 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2335
2336 * POWER7 optimizations: memset, memcmp, strncmp
2337
2338 * New optimized string functions for x86-64: strnlen (SSE2),
2339 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2340 Implemented by Ulrich Drepper.
2341 \f
2342 Version 2.12
2343
2344 * The following bugs are resolved with this release:
2345
2346 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2347 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2348 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2349 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2350 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2351 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2352 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2353 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2354
2355 * New interfaces: pthread_getname_np, pthread_setname_np
2356
2357 * New Linux interface: recvmmsg
2358
2359 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2360
2361 * The dynamic linker now recognizes supported ABI versions from the
2362 EI_ABIVERSION field in the ELF header.
2363 Implemented by Ulrich Drepper.
2364
2365 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2366 will not be used to fill in password fields in the passwd.byname replies.
2367 Instead it is used to synthesize the shadow.byname table, should it be
2368 missing. This is a useful mode in some installations involving Solaris.
2369 Implemented by Ulrich Drepper.
2370
2371 * New locales: kok_IN, sq_MK, cv_RU
2372 \f
2373 Version 2.11
2374
2375 * The following bugs are resolved with this release:
2376
2377 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2378 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2379 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2380 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2381 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2382 10789, 10817, 10823, 10840, 10847
2383
2384 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2385 mkostemps64
2386 Implemented by Ulrich Drepper.
2387
2388 * Checking version of longjmp added that fails if an uninitialized stack
2389 frame would be created. Implemented by Ulrich Drepper.
2390
2391 * STT_GNU_IFUNC is now supported in static executables.
2392 Implemented by H.J. Lu.
2393
2394 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2395
2396 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2397 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2398 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2399 Contributed by H.J. Lu.
2400
2401 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2402 Implemented by Ulrich Drepper.
2403
2404 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2405 strstr, strcasestr.
2406 Contributed by H.J. Lu.
2407
2408 * Support for fma instruction in AVX on x86-64.
2409 Implemented by H.J. Lu and Ulrich Drepper.
2410
2411 * AVX support in x86-64 auditing support in ld.so.
2412 Implemented by H.J. Lu.
2413
2414 * STB_GNU_UNIQUE support added.
2415 Implemented by Ulrich Drepper.
2416
2417 * Implement second fallback mode for optimized DNS lookup for even more
2418 broken environments. If two requests from the same port are not
2419 handled correctly close the socket and open a new one before sending
2420 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2421 can be used to select this mode right away, instead of rediscovering the
2422 necessity is every process again.
2423 Implemented by Ulrich Drepper.
2424
2425 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2426 Implemented by Adam Tkac.
2427
2428 * Optimized iconv conversions for S390x.
2429 Implemented by Andreas Krebbel.
2430
2431 * Using condvars with PI mutexes is now more efficient due to kernel
2432 support for requeueing to PI futexes. NPTL support added for x86-64.
2433 Implemented by Ulrich Drepper.
2434
2435 * New locales: ps_AF, my_MM
2436 \f
2437 Version 2.10
2438
2439 * The following bugs are resolved with this release:
2440
2441 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2442 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2443 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2444 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2445 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2446
2447 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2448 Implemented by Ulrich Drepper.
2449
2450 * New Linux interfaces: accept4, fallocate, fallocate64.
2451 Implemented by Ulrich Drepper.
2452
2453 * Correct declarations of string function when used in C++ code. This
2454 could lead to compile errors for invalid C++ code.
2455
2456 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2457 now in POSIX.
2458
2459 * New POSIX 2008 interface: psiginfo
2460 Implemented by Ulrich Drepper.
2461
2462 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2463 Implemented by Ulrich Drepper.
2464
2465 * Support for selecting between multiple function definitions at runtime
2466 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2467
2468 * The libcrypt library can now use the hash function implementations in
2469 NSS. Implemented by Ulrich Drepper.
2470
2471 * The malloc implementation can be compiled to be less memory efficient
2472 but higher performing in multi-threaded programs.
2473 Implemented by Ulrich Drepper.
2474
2475 * New locales: nan_TW@latin, ks_IN
2476
2477 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2478 Implemented by Ulrich Drepper.
2479
2480 * Extended printf hook support. It is possible to use user-defined types
2481 and extend existing format specifiers.
2482 Implemented by Ulrich Drepper.
2483
2484 * Handling for group shadow files has been added.
2485 Implemented by Ulrich Drepper.
2486
2487 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2488 servers (the case, e.g., for some people using the built-in DNS
2489 server in ADSL modems/routers). There is a once-per-process timeout
2490 in case of a broken server. To avoid it, users can run nscd or put
2491 'options single-request' in /etc/resolv.conf.
2492 Implemented by Ulrich Drepper.
2493 \f
2494 Version 2.9
2495
2496 * The following bugs are resolved with this release:
2497
2498 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2499 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2500 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2501 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2502 6968, 6974, 6980, 6995, 7008, 7009, 7029
2503
2504 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2505 up at the same time. Implemented by Ulrich Drepper.
2506
2507 * TLS descriptors for LD and GD on x86 and x86-64.
2508 Implemented by Alexandre Oliva.
2509
2510 * getaddrinfo now handles DCCP and UDPlite.
2511 Implemented by Ulrich Drepper.
2512
2513 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2514 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2515 Implemented by Ulrich Drepper.
2516
2517 * New implementation of memmem, strstr, and strcasestr which is O(n).
2518 Implemented by Eric Blake.
2519
2520 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2521
2522 * Implement "e" option for popen to open file descriptor with the
2523 close-on-exec flag set. Implemented by Ulrich Drepper.
2524
2525 * Implement "b" mode for fmemopen. In this mode writes writes don't
2526 implicitly add a NUL byte and seeks from the end of the buffer really
2527 use the buffer end, not the string length as the basis.
2528 Implemented by Ulrich Drepper.
2529
2530 * Many functions, exported and internal, now atomically set the close-on-exec
2531 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2532
2533 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2534 Sinhala)
2535 Implemented by Pravin Satpute.
2536
2537 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2538 \f
2539 Version 2.8
2540
2541 * The following bugs are resolved with this release:
2542
2543 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2544 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2545 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2546 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2547 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2548 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2549 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2550
2551 * New locales: bo_CN, bo_IN, shs_CA.
2552
2553 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2554
2555 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2556 Implemented by Pravin Satpute.
2557
2558 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2559
2560 * nscd caches now all timeouts for DNS entries
2561 Implemented by Ulrich Drepper.
2562
2563 * nscd is more efficient and wakes up less often.
2564 Implemented by Ulrich Drepper.
2565
2566 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2567 vdprintf, and obstack_vprintf.
2568 Implemented by Jakub Jelinek.
2569
2570 * Faster memset for x86-64.
2571 Implemented by Harsha Jagasia and H.J. Lu.
2572
2573 * Faster memcpy on x86.
2574 Implemented by Ulrich Drepper.
2575
2576 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2577 Implemented by Ulrich Drepper.
2578
2579 * Faster sqrt and sqrtf implemention for some PPC variants.
2580 Implemented by Steven Munroe.
2581 \f
2582 Version 2.7
2583
2584 * The following bugs are resolved with this release:
2585
2586 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2587 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2588 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2589 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2590 5071, 5103, 5104, 5112, 5113, 5184, 5186
2591
2592 * More checking functions: fread, fread_unlocked, open*, mq_open.
2593 Implemented by Jakub Jelinek and Ulrich Drepper.
2594
2595 * Extend fortification to C++. Implemented by Jakub Jelinek.
2596
2597 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2598 by not recognizing 'a' as a modifier when those specs are requested.
2599 Implemented by Jakub Jelinek.
2600
2601 * PPC optimizations to math and string functions.
2602 Implemented by Steven Munroe.
2603
2604 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2605 options to be passed. Implemented by Ulrich Drepper.
2606
2607 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2608
2609 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2610 Implemented by Ulrich Drepper.
2611
2612 * Handle private futexes in the NPTL implementation.
2613 Implemented by Jakub Jelinek and Ulrich Drepper.
2614
2615 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2616 Implemented by Roland McGrath and Ulrich Drepper.
2617
2618 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2619
2620 * SHA-256 and SHA-512 based password encryption.
2621 Implemented by Ulrich Drepper.
2622
2623 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2624 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2625 yo_NG.
2626
2627 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2628 Implemented by Ulrich Drepper.
2629 \f
2630 Version 2.6
2631
2632 * The following bugs are resolved with this release:
2633
2634 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2635 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2636 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2637 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2638 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2639 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2640 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2641 4702, 4858
2642
2643 * New Linux interfaces: epoll_pwait, sched_getcpu.
2644
2645 * New generic interfaces: strerror_l.
2646
2647 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2648 \f
2649 Version 2.5
2650
2651 * The following bugs are resolved with this release:
2652
2653 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2654 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2655 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2656 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2657 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2658 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2659 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2660 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2661 3225, 3251, 3252, 3253, 3273
2662
2663 * For Linux, the sorting of addresses returned by getaddrinfo now also
2664 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2665 Implemented by Ulrich Drepper.
2666
2667 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2668 Implemented by Ulrich Drepper.
2669
2670 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2671
2672 * New iconv module for MIK. Contributed by Alexander Shopov.
2673
2674 * For sites with broken group and/or passwd database, the auto-propagate
2675 option of nscd can prevent creating ID lookup entries from the results
2676 of a name lookup and vice versa. This usually is no problem but some
2677 site might have problems with the default behavior.
2678 Implemented by Ulrich Drepper.
2679
2680 * Iterating over entire database in NIS can be slow. With the
2681 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2682 to trade time for memory. The entire database will be read at once.
2683 Implemented by Ulrich Drepper.
2684
2685 * The interfaces introduced in RFC 3542 have been implemented by
2686 Ulrich Drepper.
2687
2688 * Support for the new ELF hash table format was added by Ulrich Drepper.
2689
2690 * Support for priority inheritance mutexes added by Jakub Jelinek and
2691 Ulrich Drepper.
2692
2693 * Support for priority protected mutexes added by Jakub Jelinek.
2694
2695 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2696 or_IN, csb_PL, fy_NL, sr_ME.
2697 \f
2698 Version 2.4
2699
2700 * More overflow detection functions.
2701
2702 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2703 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2704
2705 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2706 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2707 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2708 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2709 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2710 by Masahide Washizawa.
2711
2712 * It is now possible to install audit libraries for the dynamic linker using
2713 LD_AUDIT. Implemented by Ulrich Drepper.
2714
2715 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2716 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2717 For a libc and libpthread that works well on Linux 2.4 kernels, we
2718 recommend using the stable 2.3 branch.
2719
2720 * The new function open_wmemstream acts like open_memstream,
2721 but uses a wchar_t wide character buffer.
2722
2723 * The new function ppoll is an improved version of poll, similar to pselect.
2724
2725 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2726 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2727 renameat, unlinkat, linkat, symlinkat, readlinkat.
2728
2729 * New Linux kernel system calls: unshare,
2730 inotify_init, inotify_add_watch, inotify_rm_watch.
2731
2732 * The euidaccess function is now also known by the alias eaccess,
2733 for compatibility with some other systems.
2734
2735 * Timezone data updated to 2006b version.
2736 \f
2737 Version 2.3.6
2738
2739 * The following bugs are resolved with this release:
2740
2741 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2742 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2743 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2744 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2745 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2746 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2747
2748 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2749
2750 * As of this release, GCC 4 can be used to compile the C Library.
2751
2752 * Timezone data updated to 2005m version.
2753 \f
2754 Version 2.3.5
2755
2756 * The following bugs are resolved with this release:
2757
2758 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2759 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2760 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2761 777, 787, 821, 822, 823, 825
2762
2763 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2764 \f
2765 Version 2.3.4
2766
2767 * Support for RFC 3678. Real implementations exist only for Linux so far.
2768 Implemented by Ulrich Drepper.
2769
2770 * nscd can now cache entries persistently. Expiring entries are reloaded.
2771 For speedups the cache can be shared in memory with client processes.
2772 Implemented by Ulrich Drepper.
2773
2774 * nscd can now perform SELinux checks.
2775 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2776
2777 * getaddrinfo queries are now cached. Canonical name lookup is performed
2778 efficiently.
2779 Implemented by Ulrich Drepper.
2780
2781 * The nothrow function attribute is used when headers are used by gcc when
2782 compiling C code. This can avoid significant amounts of exception
2783 handling data.
2784
2785 * The malloc functions perform more error checking and are stricter when
2786 it comes to reacting on errors. The default action is to terminate
2787 the process after showing an error message. Implemented by Ulrich Drepper.
2788
2789 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2790 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2791
2792 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2793 namespaces using the new function dlmopen(). This feature is of course,
2794 like most other dynamic loading functionality, not available in statically
2795 linked applications. Implemented by Ulrich Drepper.
2796
2797 * Low-overhead boundary checking variants of string and some stdio functions
2798 were added. These are to be used in conjunction with a gcc patch by
2799 Jakub Jelinek which adds calls to these functions if possible.
2800 Implemented by Jakub Jelinek and Ulrich Drepper.
2801
2802 * Old code for several operating systems and machine architectures that
2803 have not been in working condition in a long time have been removed from
2804 the main source tree maintained by the GNU C Library's maintainers.
2805 These files are now reside in the separate `ports' source module
2806 that is usable as an add-on when building the library.
2807 \f
2808 Version 2.3.3
2809
2810 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2811 interrogate the dynamic linker, compatible with the Solaris interface.
2812
2813 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2814 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2815
2816 * getifaddrs now uses the netlink interface on Linux to get its information.
2817 Implemented by Thorsten Kukuk.
2818
2819 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2820 Implemented by Ulrich Drepper.
2821
2822 * support for non-executable stacks on x86 has been added. Changes mostly
2823 by Roland McGrath.
2824
2825 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2826 and Ulrich Drepper.
2827
2828 * getaddrinfo now performs destination address selection according to
2829 RFC 3484.
2830 \f
2831 Version 2.3.2
2832
2833 * Thread-safe interfaces for many functions that access locale data
2834 were added in version 2.3, but these features were omitted from NEWS.
2835 Many functions have variants with an `_l' suffix that take a `locale_t'
2836 object as a parameter rather than consulting the current locale.
2837 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2838 create and maintain `locale_t' objects. Additionally, the new function
2839 `uselocale' sets "the current locale" (as used by functions not so
2840 parameterized) set for an individual thread. These features were added
2841 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2842
2843 * The functions getresuid, getresgid, setresuid, and setresgid, which
2844 have long been available on Linux, are now declared in <unistd.h>
2845 and are now also available on the Hurd.
2846
2847 * ELF thread-local storage support (TLS) now works on x86-64.
2848
2849 * The new dynamic string token $LIB is expanded in shared library names.
2850 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2851
2852 * Aldy Hernandez contributed complete software floating point support for
2853 PowerPC machines with no FPU.
2854
2855 * fexecve is implemented on Linux.
2856
2857 * The `btowc' function should work at least twice as fast due to
2858 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2859
2860 * With approriate thread add-ons cancelable functions are now implemented
2861 in libc.so as well. No need to call the function in libpthread. This
2862 change allowed to finally disable the incorrect and expensive handling
2863 of weak definition in ld.so.
2864
2865 * Yet more PLT entries in libc.so have been removed. We finally arrived
2866 at the bare minimum. Startup times improved appropriately.
2867
2868 * Support for the new Linux/x86 system call interface was added. The
2869 AT_SYSINFO auxiliary vector entry is recognized and handled.
2870 \f
2871 Version 2.3
2872
2873 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2874 charsets.
2875
2876 * iconv (the program and the interface) now accepts empty names (excluding
2877 options like //TRANSLIT) to mean "use charset of current locale".
2878
2879 * localedef can now transliterate characters in strings which are not in
2880 the provided charmap. The information from the input locale is used.
2881
2882 * Prelinking support was added for ELF targets. This requires additional
2883 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2884
2885 * Read-only stdio streams now use mmap to speed up operation by eliminating
2886 copying and buffer underflows. To use add 'm' to the mode string of
2887 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2888
2889 * The malloc functions were completely rewritten by Wolfram Gloger based
2890 on Doug Lea's malloc-2.7.0.c.
2891
2892 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2893 implementation of regex.
2894
2895 * Bruno Haible upgraded the iconv and locale implementation to support
2896 Unicode 3.2.
2897
2898 * Contents of the LC_* and LANG environment variables in the CEN style are
2899 not recognized anymore. It never was used. Change by Ulrich Drepper.
2900
2901 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2902 thread-local storage (TLS) ABI on some platforms.
2903 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2904
2905 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2906 EUC-JISX0213, and TSCII.
2907
2908 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2909 BSD-compatible interface for getting all network interface addresses.
2910 Implementation for IPv4 by Roland McGrath.
2911
2912 * Loading of locale data is faster due to the introduction of a locale
2913 archive. Implemented by Roland McGrath and Ulrich Drepper.
2914
2915 * Startup times are significantly reduced by not using exported functions
2916 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2917 and Ulrich Drepper.
2918
2919 * Steven Munroe contributed a port to PowerPC64/Linux.
2920 \f
2921 Version 2.2.6
2922
2923 * The Hurd now uses the GNU libio implementation of stdio rather than the
2924 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2925
2926 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2927 with the same behavior as the Linux system call of the same name.
2928 \f
2929 Version 2.2.5
2930
2931 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2932 128-bit long double format.
2933
2934 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2935 IBM1161, and IBM1162 charsets.
2936
2937 * Andreas Jaeger contributed a port to x86-64/Linux.
2938
2939 * Peter Bruin contributed a port to PowerPC/Hurd.
2940
2941 * libc functions using I/O streams now can handle wide-oriented streams
2942 as well.
2943
2944 * optimizations in the dynamic linker. Binaries created by recent binutils
2945 versions start up quicker due to reduced time spent on relocations.
2946
2947 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2948 \f
2949 Version 2.2.4
2950
2951 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2952 asinh, atanh, j0 for the 128-bit long double format.
2953
2954 * Bruno Haible updated all the code handling Unicode in some form to
2955 support Unicode 3.1.
2956
2957 * Speed of regex for single-byte locales is back to previous levels.
2958 Patch by Isamu Hasegawa.
2959
2960 * Alpha, SPARC, and IA-64 now also using floating stacks.
2961
2962 * Startup time of internationalized applications greatly improved through
2963 iconv cache. Use iconvconfig to generate the cache file.
2964 Contributed by Ulrich Drepper.
2965
2966 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2967 severe performance problems. Patch by David Mosberger.
2968
2969 * The Hurd port got a lot more functionality like AIO, various stdio
2970 extensions, etc. Mainly done by Roland McGrath.
2971
2972 * mtrace can now lookup symbols in shared libraries.
2973 \f
2974 Version 2.2.3
2975
2976 * Intel's IA-64 math library is largely integrated. It provides fast and
2977 accurate implementatations for most basic and standard math functions
2978 in float, double, and long double format.
2979
2980 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2981 and asin for the 96-bit long double format and asin, log, tan for the
2982 128-bit long double format.
2983
2984 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2985 The basic double functions exist today. Contributed by Abraham Ziv
2986 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2987 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2988
2989 * An asynchronous name lookup library was added. The interface is designed
2990 after POSIX AIO. The proposal was circulated beforehand to get comments.
2991 No negative ones came in. Implemented by Ulrich Drepper.
2992
2993 * Port to S390/64bit contributed by Martin Schwidefsky
2994 <schwidefsky@de.ibm.com>.
2995
2996 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2997 of functions for Linux/IA-64.
2998
2999 * The RPC code is now thread safe. Threads can now use the same service
3000 of different services at the same time. Patch by Eric Norum
3001 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3002
3003 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3004 family of functions for Linux/S390.
3005
3006 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3007 of functions for Linux/x86.
3008
3009 * Port to Linux/CRIS contributed by Axis Communications.
3010 \f
3011 Version 2.2.2
3012
3013 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3014 we can now check for namespace violations and missing declarations. The
3015 result is that almost all headers are now Unix-compliant (as defined in
3016 the upcoming XPG6). The negative side is that some programs might need
3017 corrections, too, if they depend on the incorrect form of the headers in
3018 previous versions which defined too many symbols and included too many
3019 other headers.
3020
3021 * regex now handles multibyte character sets correctly.
3022 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3023
3024 * iconv (the program) does now conform to the upcoming XPG6 and handles
3025 charmaps. Instead of the charset names the path of charmaps can be
3026 provided and the conversion happens based on this data.
3027 Contributed by Ulrich Drepper.
3028
3029 * The locale program now provides detailed information about the installed
3030 locales. While
3031
3032 locale -a
3033
3034 only lists the names of the supported locales
3035
3036 locale -a --verbose
3037
3038 provides details such as country, language, and codeset name.
3039 Contributed by Ulrich Drepper.
3040 \f
3041 Version 2.2.1
3042
3043 * The gencat program now parses the input file according to the charset
3044 selected by the LC_CTYPE category. This is important for stateful
3045 character sets. To make generating catalogs easier there is a way
3046 to overwrite the charset selected by the locale: before the first
3047 message or $ quote line the catalog can contain a line like
3048
3049 $ codeset=ISO-8859-2
3050
3051 to select the charset (ISO-8859-2 in this case).
3052
3053 Implemented by Shinya Hanataka and Ulrich Drepper.
3054
3055 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3056 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3057 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3058
3059 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3060 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3061
3062 * The LANGUAGE environment variable is now ignored unless the locale is
3063 changed from the default "C" locale.
3064
3065 * The usual bug fixes.
3066 \f
3067 Version 2.2
3068
3069 * Greg McGary added runtime support for bounds checking using gcc's
3070 new -fbounded-pointers option. ix86 target is complete. PowerPC
3071 is in progress.
3072
3073 * Thorsten Kukuk added secure mode support to nscd.
3074
3075 * The Berkeley db libraries have been removed.
3076
3077 Related, the nss_db module is now in a separate package since it
3078 obviously requires a database library being available.
3079
3080 * Wide character I/O streams implemented by Ulrich Drepper.
3081
3082 * Functions from the extended socket API added by Ulrich Drepper.
3083
3084 * Functions feenableexcept and fedisableexcept to control the
3085 behaviour of individual exceptions have been added by Andreas Jaeger.
3086
3087 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3088
3089 * The resolver code has been updated from bind 8.2.3-T5B which supports
3090 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3091 and Mark Kettenis.
3092
3093 This change could in some situations effect backward compatibility. Since
3094 now `_res' is a thread-local instead of a global variable, modifying it
3095 in one thread does not have any effect in other threads.
3096
3097 The resolver library was also extended to allow IPv6 as the transport
3098 protocol for the requests. This work was done by Stig Venaas.
3099
3100 * Compatibility code for K&R C compilers has been removed from the
3101 header files. A ISO C compiler is needed to use the library
3102 (conforming to either C89 or C99 standard).
3103
3104 * Complete rewrite of the localedef program to support multibyte character
3105 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3106 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3107 Implemented by Ulrich Drepper.
3108
3109 Bruno Haible significantly improved the generation and use of the data
3110 structures for the wide character tables.
3111
3112 * Plural handling in gettext implemented by Ulrich Drepper.
3113
3114 * The utmp daemon has been removed.
3115
3116 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3117
3118 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3119 and Yutaka Niibe.
3120
3121 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3122
3123 * POSIX barriers implemented by Kaz Kylheku.
3124
3125 * POSIX spawn function family implemented by Ulrich Drepper.
3126
3127 * POSIX spinlocks are now available.
3128
3129 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3130
3131 * the configure option --enable-kernel=X.Y.Z allows to strip out
3132 compatibility for kernel versions before X.Y.Z. This is currently only
3133 implemented for Linux.
3134
3135 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3136 field sin6_scope_id. This means that all programs using IPv6 should be
3137 recompiled. Don't expect binary compatibility with previous glibc
3138 versions.
3139
3140 * various conversion modules for IBM character sets contributed by
3141 Masahide Washizawa.
3142
3143 * IA-64 port by Jes Sorensen and HJ Lu.
3144 \f
3145 Version 2.1.3
3146
3147 * bug fixes
3148
3149 \f
3150 Version 2.1.2
3151
3152 * bug fixes
3153
3154 \f
3155 Version 2.1.1
3156
3157 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3158
3159 * New xdr functions are added; some rpc functions are now 64bit clean.
3160
3161 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3162
3163 * Fixed known incompatibilities with glibc 2.0.
3164
3165 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3166
3167 * Optimized a number of functions (especially the ELF dynamic loader).
3168
3169 * Update timezone data files.
3170
3171 * lots of charmaps corrections
3172
3173 * some new locale definitions and charmaps
3174
3175 \f
3176 Version 2.1
3177
3178 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3179 conform to POSIX member type requirements. Symbol versions have been
3180 adjusted as needed within the library, and for direct use by applications,
3181 but there is potential for problems if third-party libraries use
3182 struct timeval as part of their interface. This does not present
3183 a problem for X and other "essential" system libraries.
3184
3185 * An additional locale model to support C++ Standard Library locale
3186 model and probably more was implemented by Ulrich Drepper.
3187
3188 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3189 symbol level.
3190
3191 * Miles Bader provided the `argp' function family to support hierachical
3192 command line argument parsing, layered on top of getopt.
3193
3194 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3195
3196 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3197 numbers.
3198
3199 * scanf recognizes the %a and %A format for scanning floating point numbers.
3200
3201 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3202 information and interfaces for the available integer types.
3203
3204 * about 130 new math functions were added to implement the ISO C9x math
3205 library.
3206
3207 * the new header <complex.h> contains definitions of the complex math
3208 functions from ISO C 9X.
3209
3210 * the new header <tgmath.h> defines generic macros to use complex or
3211 real valued functions.
3212
3213 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3214
3215 * Andreas Jaeger provided a test suite for the math library.
3216
3217 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3218
3219 * Ulrich Drepper added character set conversion functions (iconv).
3220
3221 * Optimized string functions have been added.
3222
3223 * The localedata addon is now part of glibc.
3224
3225 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3226
3227 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3228 daemon for NSS (nscd).
3229
3230 Missing a better place here are some numbers on improvements. Under
3231 Linux 2.1.125 un-tar-ing the kernel sources takes
3232
3233 user system wall
3234
3235 using local files 12.19s 6.88s 22.91s
3236
3237 using NIS 13.92s 8.91s 26.34s
3238
3239 using NIS & nscd 10.37s 7.34s 25.30s
3240
3241 using NIS+ 27.57s 30.37s 640.46s
3242
3243 using NIS+ & nscd 10.25s 7.83s 26.51s
3244
3245 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3246
3247 Keep in mind that non-namelookup related operations dominate above times.
3248 It was just a common complain that using NIS+ unpacking the kernel is
3249 horribly slow.
3250
3251 [1] The old nscd implementation is not available anymore since it was
3252 distributed with glibc up to version 2.0.98 and thus is now replaced.
3253
3254 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3255
3256 * Mark Kettenis provided a Hesiod NSS module.
3257
3258 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3259 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3260
3261 * rcmd can now handle netgroups (Dick Streefland).
3262
3263 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3264 Bambrough.
3265
3266 * Support for the IPv6 protocol has been added to the socket API, as per the
3267 latest draft standards.
3268
3269 * Support for Linux 2.2 has been added.
3270
3271 * Interface changes relative to the latest 2.0.x release:
3272 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3273 addseverity NEW: Unix98
3274 alphasort64 NEW: LFS
3275 argp_err_exit_status NEW: argp, GNU ext
3276 argp_error NEW: argp, GNU ext
3277 argp_failure NEW: argp, GNU ext
3278 argp_help NEW: argp, GNU ext
3279 argp_parse NEW: argp, GNU ext
3280 argp_program_bug_address NEW: argp, GNU ext
3281 argp_program_version NEW: argp, GNU ext
3282 argp_program_version_hook NEW: argp, GNU ext
3283 argp_state_help NEW: argp, GNU ext
3284 argp_usage NEW: argp, GNU ext
3285 authdes_create NEW: Secure RPC
3286 authdes_getucred NEW: Secure RPC
3287 authdes_pk_create NEW: Secure RPC
3288 backtrace NEW: GNU ext.
3289 backtrace_symbols NEW: GNU ext.
3290 backtrace_symbols_fd NEW: GNU ext.
3291 cacos NEW: ISO C 9x
3292 cacosf NEW: ISO C 9x
3293 cacosh NEW: ISO C 9x
3294 cacoshf NEW: ISO C 9x
3295 cacoshl NEW: ISO C 9x
3296 cacosl NEW: ISO C 9x
3297 capget NEW: kernel
3298 capset NEW: kernel
3299 carg NEW: ISO C 9x
3300 cargf NEW: ISO C 9x
3301 cargl NEW: ISO C 9x
3302 casin NEW: ISO C 9x
3303 casinf NEW: ISO C 9x
3304 casinh NEW: ISO C 9x
3305 casinhf NEW: ISO C 9x
3306 casinhl NEW: ISO C 9x
3307 casinl NEW: ISO C 9x
3308 catan NEW: ISO C 9x
3309 catanf NEW: ISO C 9x
3310 catanh NEW: ISO C 9x
3311 catanhf NEW: ISO C 9x
3312 catanhl NEW: ISO C 9x
3313 catanl NEW: ISO C 9x
3314 cbc_crypt NEW: Secure RPC
3315 ccos NEW: ISO C 9x
3316 ccosf NEW: ISO C 9x
3317 ccosh NEW: ISO C 9x
3318 ccoshf NEW: ISO C 9x
3319 ccoshl NEW: ISO C 9x
3320 ccosl NEW: ISO C 9x
3321 cexp NEW: ISO C 9x
3322 cexpf NEW: ISO C 9x
3323 cexpl NEW: ISO C 9x
3324 cimag NEW: ISO C 9x
3325 cimagf NEW: ISO C 9x
3326 cimagl NEW: ISO C 9x
3327 clearerr_locked REMOVED
3328 clntunix_create NEW: sunrpc ext
3329 clog NEW: ISO C 9x
3330 clog10 NEW: ISO C 9x
3331 clog10f NEW: ISO C 9x
3332 clog10l NEW: ISO C 9x
3333 clogf NEW: ISO C 9x
3334 clogl NEW: ISO C 9x
3335 conj NEW: ISO C 9x
3336 conjf NEW: ISO C 9x
3337 conjl NEW: ISO C 9x
3338 cpow NEW: ISO C 9x
3339 cpowf NEW: ISO C 9x
3340 cpowl NEW: ISO C 9x
3341 cproj NEW: ISO C 9x
3342 cprojf NEW: ISO C 9x
3343 cprojl NEW: ISO C 9x
3344 creal NEW: ISO C 9x
3345 crealf NEW: ISO C 9x
3346 creall NEW: ISO C 9x
3347 creat64 NEW: LFS
3348 csin NEW: ISO C 9x
3349 csinf NEW: ISO C 9x
3350 csinh NEW: ISO C 9x
3351 csinhf NEW: ISO C 9x
3352 csinhl NEW: ISO C 9x
3353 csinl NEW: ISO C 9x
3354 csqrt NEW: ISO C 9x
3355 csqrtf NEW: ISO C 9x
3356 csqrtl NEW: ISO C 9x
3357 ctan NEW: ISO C 9x
3358 ctanf NEW: ISO C 9x
3359 ctanh NEW: ISO C 9x
3360 ctanhf NEW: ISO C 9x
3361 ctanhl NEW: ISO C 9x
3362 ctanl NEW: ISO C 9x
3363 des_setparity NEW: Secure RPC
3364 ecb_crypt NEW: Secure RPC
3365 endutxent NEW: Unix98
3366 exp10 NEW: ISO C 9x
3367 exp10f NEW: ISO C 9x
3368 exp10l NEW: ISO C 9x
3369 exp2 NEW: ISO C 9x
3370 exp2f NEW: ISO C 9x
3371 exp2l NEW: ISO C 9x
3372 fattach NEW: STREAMS
3373 fdetach NEW: STREAMS
3374 fdim NEW: ISO C 9x
3375 fdimf NEW: ISO C 9x
3376 fdiml NEW: ISO C 9x
3377 feclearexcept NEW: ISO C 9x
3378 fegetenv NEW: ISO C 9x
3379 fegetexceptflag NEW: ISO C 9x
3380 fegetround NEW: ISO C 9x
3381 feholdexcept NEW: ISO C 9x
3382 feof_locked REMOVED
3383 feraiseexcept NEW: ISO C 9x
3384 ferror_locked REMOVED
3385 fesetenv NEW: ISO C 9x
3386 fesetexceptflag NEW: ISO C 9x
3387 fesetround NEW: ISO C 9x
3388 fetestexcept NEW: ISO C 9x
3389 feupdateenv NEW: ISO C 9x
3390 fflush_locked REMOVED
3391 ffsl NEW: GNU ext.
3392 ffsll NEW: GNU ext.
3393 fgetpos64 NEW: LFS
3394 fgets_unlocked NEW: GNU ext.
3395 fileno_locked REMOVED
3396 fma NEW: ISO C 9x
3397 fmaf NEW: ISO C 9x
3398 fmal NEW: ISO C 9x
3399 fmax NEW: ISO C 9x
3400 fmaxf NEW: ISO C 9x
3401 fmaxl NEW: ISO C 9x
3402 fmin NEW: ISO C 9x
3403 fminf NEW: ISO C 9x
3404 fminl NEW: ISO C 9x
3405 fmtmsg NEW: Unix98
3406 fopen64 NEW: LFS
3407 fputc_locked REMOVED
3408 fputs_unlocked NEW: GNU ext.
3409 fread_unlocked NEW: GNU ext.
3410 freopen64 NEW: LFS
3411 fseeko NEW: Unix98
3412 fsetpos64 NEW: LFS
3413 fstatfs64 NEW: LFS
3414 fstatvfs NEW: Unix98
3415 fstatvfs64 NEW: LFS
3416 ftello NEW: Unix98
3417 ftello64 NEW: LFS
3418 ftruncate64 NEW: LFS
3419 ftw64 NEW: LFS
3420 fwrite_unlocked NEW: GNU ext.
3421 gai_strerror NEW: IPv6
3422 gamma_r REMOVED
3423 gammaf_r REMOVED
3424 gammal_r REMOVED
3425 getchar_locked REMOVED
3426 getdate NEW: Unix98
3427 getdate_err NEW: Unix98
3428 getdate_r NEW: GNU ext.
3429 getmsg NEW: STREAMS
3430 getnameinfo NEW: IPv6
3431 getnetname NEW: Secure RPC
3432 getpmsg NEW: STREAMS
3433 getpt NEW: Unix98 PTY
3434 getrlimit64 NEW: LFS
3435 getutxent NEW: Unix98
3436 getutxid NEW: Unix98
3437 getutxline NEW: Unix98
3438 glob64 NEW: GNU ext.
3439 globfree64 NEW: GNU ext.
3440 gnu_get_libc_release NEW: GNU ext.
3441 gnu_get_libc_version NEW: GNU ext.
3442 grantpt NEW: Unix98 PTY
3443 host2netname NEW: Secure RPC
3444 iconv NEW: iconv
3445 iconv_close NEW: iconv
3446 iconv_open NEW: iconv
3447 if_freenameindex NEW: IPv6
3448 if_indextoname NEW: IPv6
3449 if_nameindex NEW: IPv6
3450 if_nametoindex NEW: IPv6
3451 in6addr_any NEW: IPv6
3452 in6addr_loopback NEW: IPv6
3453 inet6_isipv4mapped NEW: IPv6
3454 isastream NEW: STREAMS
3455 iswblank NEW: GNU ext.
3456 key_decryptsession NEW: Secure RPC
3457 key_decryptsession_pk NEW: Secure RPC
3458 key_encryptsession NEW: Secure RPC
3459 key_encryptsession_pk NEW: Secure RPC
3460 key_gendes NEW: Secure RPC
3461 key_get_conv NEW: Secure RPC
3462 key_secretkey_is_set NEW: Secure RPC
3463 key_setnet NEW: Secure RPC
3464 key_setsecret NEW: Secure RPC
3465 llrint NEW: ISO C 9x
3466 llrintf NEW: ISO C 9x
3467 llrintl NEW: ISO C 9x
3468 llround NEW: ISO C 9x
3469 llroundf NEW: ISO C 9x
3470 llroundl NEW: ISO C 9x
3471 log2 NEW: ISO C 9x
3472 log2f NEW: ISO C 9x
3473 log2l NEW: ISO C 9x
3474 lrint NEW: ISO C 9x
3475 lrintf NEW: ISO C 9x
3476 lrintl NEW: ISO C 9x
3477 lround NEW: ISO C 9x
3478 lroundf NEW: ISO C 9x
3479 lroundl NEW: ISO C 9x
3480 lseek64 NEW: LFS
3481 makecontext NEW: Unix98
3482 mempcpy NEW: GNU ext.
3483 mmap64 NEW: LFS
3484 moncontrol REMOVED
3485 modify_ldt NEW: kernel
3486 nan NEW: ISO C 9x
3487 nanf NEW: ISO C 9x
3488 nanl NEW: ISO C 9x
3489 nearbyint NEW: ISO C 9x
3490 nearbyintf NEW: ISO C 9x
3491 nearbyintl NEW: ISO C 9x
3492 netname2host NEW: Secure RPC
3493 netname2user NEW: Secure RPC
3494 nexttoward NEW: ISO C 9x
3495 nexttowardf NEW: ISO C 9x
3496 nexttowardl NEW: ISO C 9x
3497 nftw NEW: Unix98
3498 nftw64 NEW: LFS
3499 open64 NEW: LFS
3500 passwd2des NEW: Secure RPC
3501 pow10 NEW: GNU ext.
3502 pow10f NEW: GNU ext.
3503 pow10l NEW: GNU ext.
3504 pread NEW: Unix98
3505 pread64 NEW: LFS
3506 printf_size NEW: GNU ext.
3507 printf_size_info NEW: GNU ext.
3508 profil_counter REMOVED
3509 pthread_mutexattr_getkind_np REPLACED
3510 pthread_mutexattr_setkind_np REPLACED
3511 ptsname NEW: Unix98 PTY
3512 ptsname_r NEW: Unix98 PTY
3513 putc_locked REMOVED
3514 putchar_locked REMOVED
3515 putgrent NEW: GNU ext.
3516 putmsg NEW: STREAMS
3517 putpmsg NEW: STREAMS
3518 pututxline NEW: Unix98
3519 pwrite NEW: Unix98
3520 pwrite64 NEW: LFS
3521 readdir64 NEW: LFS
3522 readdir64_r NEW: LFS
3523 remquo NEW: ISO C 9x
3524 remquof NEW: ISO C 9x
3525 remquol NEW: ISO C 9x
3526 round NEW: ISO C 9x
3527 roundf NEW: ISO C 9x
3528 roundl NEW: ISO C 9x
3529 rtime NEW: GNU ext.
3530 scalbln NEW: ISO C 9x
3531 scalblnf NEW: ISO C 9x
3532 scalblnl NEW: ISO C 9x
3533 scandir64 NEW: LFS
3534 sendfile NEW: kernel
3535 setcontext NEW: Unix98
3536 setrlimit64 NEW: LFS
3537 setutxent NEW: Unix98
3538 sighold NEW: Unix98
3539 sigignore NEW: Unix98
3540 sigqueue NEW: POSIX.1b
3541 sigrelse NEW: Unix98
3542 sigset NEW: POSIX.1b
3543 sigtimedwait NEW: POSIX.1b
3544 sigwaitinfo NEW: POSIX.1b
3545 sincos NEW: GNU ext.
3546 sincosf NEW: GNU ext.
3547 sincosl NEW: GNU ext.
3548 statfs64 NEW: LFS
3549 statvfs NEW: Unix98
3550 statvfs64 NEW: LFS
3551 strcasestr NEW: GNU ext.
3552 strtoimax NEW: ISO C 9x
3553 strtoumax NEW: ISO C 9x
3554 strverscmp NEW: GNU ext.
3555 svcauthdes_stats NEW: Secure RPC
3556 svcunix_create NEW: sunrpc etx
3557 svcunixfd_create NEW: sunrpc ext
3558 swapcontext NEW: Unix98
3559 tcgetsid NEW: Unix98 PTY
3560 tdestroy NEW: GNU ext.
3561 tgamma NEW: ISO C 9x
3562 tgammaf NEW: ISO C 9x
3563 tgammal NEW: ISO C 9x
3564 tmpfile64 NEW: LFS
3565 trunc NEW: ISO C 9x
3566 truncate64 NEW: LFS
3567 truncf NEW: ISO C 9x
3568 truncl NEW: ISO C 9x
3569 umount2 NEW: kernel
3570 unlockpt NEW: Unix98 PTY
3571 updwtmpx NEW: Unix98
3572 user2netname NEW: Secure RPC
3573 utmpxname NEW: Unix98
3574 versionsort NEW: GNU ext.
3575 versionsort64 NEW: GNU ext.
3576 waitid NEW: Unix98
3577 wcscasecmp NEW: GNU ext.
3578 wcsncasecmp NEW: GNU ext.
3579 wcsnlen NEW: GNU ext.
3580 wcstoimax NEW: ISO C 9x
3581 wcstoll NEW: ISO C 9x
3582 wcstoull NEW: ISO C 9x
3583 wcstoumax NEW: ISO C 9x
3584 wcswcs NEW: Unix98
3585 wordexp NEW: POSIX.2
3586 wordfree NEW: POSIX.2
3587 write_profiling REMOVED
3588 xdecrypt NEW: Secure RPC
3589 xdr_authdes_cred NEW: Secure RPC
3590 xdr_authdes_verf NEW: Secure RPC
3591 xdr_cryptkeyarg NEW: Secure RPC
3592 xdr_cryptkeyarg2 NEW: Secure RPC
3593 xdr_cryptkeyres NEW: Secure RPC
3594 xdr_getcredres NEW: Secure RPC
3595 xdr_key_netstarg NEW: Secure RPC
3596 xdr_key_netstres NEW: Secure RPC
3597 xdr_keybuf NEW: Secure RPC
3598 xdr_keystatus NEW: Secure RPC
3599 xdr_netnamestr NEW: Secure RPC
3600 xdr_sizeof NEW: Secure RPC
3601 xdr_unixcred NEW: sunrpc ext
3602 xencrypt NEW: Secure RPC
3603 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3604 \f
3605 Version 2.0.6
3606
3607 * more bug fixes
3608
3609 \f
3610 Version 2.0.5
3611
3612 * more bug fixes
3613
3614 * inet_ntoa is thread-safe
3615
3616 * updwtmp is moved from libutil to libc
3617
3618 * rewrite of cbrt function
3619
3620 * update of timezone data
3621 \f
3622 Version 2.0.4
3623
3624 * more bug fixes
3625 \f
3626 Version 2.0.3
3627
3628 * more bug fixes
3629 \f
3630 Version 2.0.2
3631
3632 * more bug fixes
3633
3634 * add atoll function
3635
3636 * fix complex problems in Berkeley DB code
3637
3638 * fix math functions
3639 \f
3640 Version 2.0.1
3641
3642 * fixed lots of header problems (especially Linux/GNU specific)
3643
3644 * dynamic loader preserves all registers
3645
3646 * Roland McGrath provided support for handling of auxiliary objects in
3647 the ELF dynamic loader.
3648
3649 * support for parallel builds is improved
3650 \f
3651 Version 2.0
3652
3653 * GNU extensions are no longer declared by default. To enable them you
3654 must define the macro `_GNU_SOURCE' in your program or compile with
3655 `-D_GNU_SOURCE'.
3656
3657 * The library has changed from using GNU ld symbol aliases to using weak
3658 symbols where available. The ELF object file format supports weak
3659 symbols; GNU ld also supports weak symbols in the a.out format. (There
3660 is also now support for other GNU ld extensions in ELF. Use the
3661 `--with-elf' option to configure to indicate you have ELF, and
3662 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3663 of many files which contained only symbol aliases, reducing the size of
3664 the source and the compiled library; many other files were renamed to
3665 less cryptic names previously occupied by the symbol alias files.
3666 There is a new header file <elf.h> for programs which operate on
3667 files in the ELF format.
3668
3669 * Converted to Autoconf version 2, so `configure' has more options.
3670 Run `configure --help' to see the details.
3671
3672 * The library can now be configured to build profiling, highly-optimized
3673 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3674 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3675 `configure' enable building these extra libraries. The shared library is
3676 built by default when using both ELF and GNU ld. When shared libraries
3677 are enabled, the new library `-ldl' is available for arbitrary run-time
3678 loading of shared objects; its interface is defined in <dlfcn.h>. The
3679 new header file <link.h> gives access to the internals of the run-time
3680 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3681 application of same name on other systems and it provides information
3682 about dynamically linked binaries.
3683
3684 * The C library now provides the run-time support code for profiling
3685 executables compiled with `-pg'. Programs can control the profiling code
3686 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3687 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3688 the support for this file format was contributed by David Mosberger-Tang.
3689
3690 * The math code has been replaced with a math library based on fdlibm from
3691 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3692 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3693 functions. The math functions now reside in a separate library, so
3694 programs using them will need to use `-lm' their linking commands.
3695
3696 * John C. Bowman contributed optimized ix87 assembler inline functions.
3697
3698 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3699 mechanism similar to that found in Solaris 2. This is now used for the
3700 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3701 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3702 file controls what services are used for each individual database. This
3703 works by loading shared libraries with names specified in `nsswitch.conf',
3704 so service modules can be changed or added at any time without even
3705 relinking any program. Currently there are the file, db, and NIS based
3706 NSS services available.
3707
3708 * The new functions `strtoq' and `strtouq' parse integer values from
3709 strings, like `strtol' and `strtoul', but they return `long long int' and
3710 `unsigned long long int' values, respectively (64-bit quantities).
3711
3712 * The new functions `strtof' and `strtold' parse floating-point values from
3713 strings, like `strtod', but they return `float' and `long double' values,
3714 respectively (on some machines `double' and `long double' are the same).
3715
3716 * Ulrich Drepper has contributed new implementations of the floating-point
3717 printing and reading code used in the `printf' family of functions and
3718 `strtod', `strtof', and `strtold'. These new functions are perfectly
3719 accurate, and much faster than the old ones.
3720
3721 * The implementation of the POSIX locale model was completely rewritten by
3722 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3723 to compile the POSIX locale definition.
3724
3725 * The former dummy implementations of the strcoll and strxfrm function are
3726 now replaced by fully functional code contributed by Ulrich Drepper. The
3727 collation information comes from the POSIX locale definitions.
3728
3729 * The new header <langinfo.h> defines an interface for accessing
3730 various locale-dependent data (using the locale chosen with `setlocale').
3731
3732 * Ulrich Drepper has contributed a new suite of functions for operation on
3733 wide-character and multibyte-character strings, in <wchar.h>;
3734 and classification and case conversion of wide characters, in <wctype.h>.
3735 These new functions are conforming to the ISO C, Amendement 1 specification.
3736
3737 * There is now a second implementation of the standard I/O library available.
3738 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3739 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3740
3741 * You can now use positional parameter specifications in format strings
3742 for the `printf' and `scanf' families of functions. For example,
3743 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3744 ``Number 6, Mr Jones''. This is mainly useful when providing different
3745 format strings for different languages, whose grammars may dictate
3746 different orderings of the values being printed. To support this
3747 feature, the interface for `register_printf_handler' has changed; see
3748 the header file <printf.h> for details.
3749
3750 * The `printf' and `scanf' families of functions now understand a new
3751 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3752 to group numbers as indicated by the locale; for `scanf' and friends, this
3753 says to accept as valid only a number with all the proper grouping
3754 separators in the right places. In the default "C" locale, numbers are
3755 not grouped; but locales for specific countries will define the usual
3756 conventions (i.e. separate thousands with `,' in the US locale).
3757
3758 * The pgrp functions have been regularized, slightly incompatibly but much
3759 less confusingly. The core functions are now `getpgid' and `setpgid',
3760 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3761 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3762 provided for compatibility. There is no longer an incompatible `getpgrp'
3763 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3764
3765 * The new header file <fts.h> and suite of functions simplify programs that
3766 operate on directory trees. This code comes from 4.4 BSD.
3767
3768 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3769 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3770 NSS scheme used in glibc.
3771
3772 * There is a new malloc debugging hook `__memalign_hook'.
3773
3774 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3775 `unsigned int' in <sys/types.h>. These are for compatibility only and
3776 their use is discouraged.
3777
3778 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3779 done differently, so that it works even without GNU ld.
3780
3781 * New function `euidaccess' checks allowed access to a file like `access',
3782 but using the effective IDs instead of the real IDs.
3783
3784 * The time zone data files have been updated for the latest and greatest
3785 local time conventions of the countries of the world.
3786
3787 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3788 see <dirent.h>.
3789
3790 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3791 for formatting floating-point numbers. They are provided only for
3792 compatibility; new programs should use `sprintf' instead. There are
3793 also equivalent function for the `long double' floating-point type and
3794 all functions also exist in a reentrant form.
3795
3796 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3797 functions for maintaining the login-record files (primarily of use to
3798 system programs such as `login'), and convenient functions for
3799 allocating and initializing a pseudo-terminal (pty) device.
3800
3801 * Ulrich Drepper has contributed new support for System V style
3802 shared memory and IPC on systems that support it.
3803
3804 * Ulrich Drepper has contributed several miscellaneous new functions found
3805 in System V: The `hsearch' family of functions provide an effective
3806 implementation of hash tables; `a64l' and `l64a' provide a very simple
3807 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3808 number generator.
3809
3810 * Ulrich Drepper has contributed new reentrant counterparts for the
3811 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3812
3813 * Ulrich Drepper has contributed new, highly-optimized versions of several
3814 string functions for the i486/Pentium family of processors.
3815
3816 * Ulrich Drepper has updated the Linux-specific code, based largely
3817 on work done in Hongjiu Lu's version of GNU libc for Linux.
3818 The GNU library now supports Linux versions 2.0.10 and later,
3819 using the ELF object file format (i[3456]86-*-linux).
3820
3821 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3822
3823 * David Mosberger-Tang and Richard Henderson have ported the C library
3824 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3825 dynamic linking support for ELF/Alpha.
3826
3827 * Richard Henderson contributed several Alpha optimized assembler function
3828 for arithmetic and string handling.
3829
3830 * Ulrich Drepper has contributed a new set of message catalog functions to
3831 support multiple languages using the <libintl.h> interface, for use with
3832 his new package GNU gettext. Translation volunteers have contributed
3833 catalogs of the library's messages in Spanish, German, and Korean.
3834
3835 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3836 program and the `catgets' function for reading the catalog files it
3837 creates. (The <libintl.h> interface is preferred; we include the
3838 <nl_types.h> interface using `catgets' only for source compatibility with
3839 programs already written to use it.)
3840
3841 * New header file <values.h> gives SVID-compatible names for <limits.h>
3842 constants.
3843
3844 * Various new macros, declarations, and small header files for compatibility
3845 with 4.4 BSD.
3846
3847 * New function `group_member' is a convenient way to check if a process has
3848 a given effective group ID.
3849
3850 * When using GCC 2.7 and later, the socket functions are now declared in a
3851 special way so that passing an argument of type `struct sockaddr_in *',
3852 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3853 `struct sockaddr *' type, does not generate a type-clash warning.
3854
3855 * New function `error' declared in header file <error.h> is a convenient
3856 function for printing error messages and optionally exiting; this is the
3857 canonical function used in GNU programs. The new functions `err', `warn',
3858 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3859 doing the same thing.
3860
3861 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3862 POSIX.2 `glob' function to do ~ and {...} expansion.
3863
3864 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3865 `clearenv' which is used in POSIX.9 is also available.
3866
3867 * New function `getsid' returns session ID number on systems that support it.
3868
3869 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3870 files <db.h> and <mpool.h> provide a rich set of functions for several
3871 types of simple databases stored in memory and in files, and <ndbm.h> is
3872 an old `ndbm'-compatible interface using the `db' functions. Link with
3873 `-ldb' to get these functions.
3874
3875 * New macro `strdupa' copies a string like `strdup', but uses local stack
3876 space from `alloca' instead of dynamic heap space from `malloc'.
3877
3878 * New function `strnlen' is like `strlen' but searches only a given maximum
3879 number of characters for the null terminator. `stpncpy', `strndup' and
3880 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3881 function.
3882
3883 * New function `statfs' in header <sys/statfs.h>.
3884
3885 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3886 provide convenient functions for operating on blocks of null-terminated
3887 strings.
3888
3889 * A new suite of functions in <utmp.h> handle all the details of reading
3890 and writing the utmp file.
3891
3892 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3893 Thorsten Kukuk.
3894
3895 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3896 completely POSIX compliant and also implemented the extended functionality
3897 to handle alternate digit representation and alternate era date formats.
3898
3899 * Ulrich Drepper provided an implementation of the `strptime' function
3900 defined in XPG4.2 which transforms a string into a `struct tm' value.
3901
3902 * Paul Eggert provided the tzselect shell script as part of the timezone
3903 code. The shell script makes it easy to select the correct timezone
3904 specification.
3905
3906 * The implementation of the malloc family of functions is completely replaced
3907 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3908 The implementation uses the mmap function (if available) and it is
3909 optimized for the use in multi threaded programs.
3910
3911 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3912 functions. This new functionality is usable by specifying a special
3913 salt string and it is compatible with implementation on *BSD systems.
3914
3915 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3916 `getsubopt' to handle second level command line options, `bsd_signal'
3917 to access BSD style `signal' functionality, the obsolete `regexp' style
3918 expression matcher.
3919
3920 * the `lchown' function is available on system which support this
3921 functionality.
3922
3923 * The implementation of the shadow password handling function was contributed
3924 by Ulrich Drepper.
3925
3926 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3927
3928 * POSIX.1g support was added. The <sys/select.h> header is available,
3929 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3930 implementation of `getaddrinfo'.
3931 \f
3932 Version 1.09
3933
3934 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3935
3936 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3937 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3938
3939 * The new function `daemon' from 4.4 BSD is useful for server programs that
3940 want to put themselves in the background.
3941
3942 * Joel Sherrill has contributed support for several standalone boards that
3943 run without an operating system.
3944
3945 * `printf', `scanf' and friends now accept a `q' type modifier for long
3946 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3947
3948 * All of the code taken from BSD (notably most of the math and networking
3949 routines) has been updated from the BSD 4.4-Lite release.
3950
3951 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3952
3953 * The new functions `getdomainname' and `setdomainname' fetch or change the
3954 YP/NIS domain name. These are system calls which exist on systems which
3955 have YP (aka NIS).
3956
3957 * The time zone data files have been updated for the latest international
3958 conventions.
3959
3960 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3961 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3962 \f
3963 Version 1.08
3964
3965 * The C library now includes support for Sun RPC, from Sun's free
3966 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3967 are included. (There is still no support for YP.)
3968
3969 * Tom Quinn has contributed a port of the C library to SGI machines running
3970 Irix 4 (mips-sgi-irix4).
3971
3972 * The new `lockf' function is a simplified interface to the locking
3973 facilities of `fcntl', included for compatibility.
3974
3975 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3976
3977 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3978
3979 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3980 compatibility.
3981
3982 * The error message printed by `assert' for a failed assertion now includes
3983 the name of the program (if using GNU ld) and the name of the calling
3984 function (with versions of GCC that support this).
3985
3986 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3987
3988 * The library now includes the <sys/mman.h> header file and memory
3989 management functions `mmap', `munmap', `mprotect', `msync', and
3990 `madvise', on systems that support those facilities.
3991
3992 * The interface for `mcheck' has changed slightly: the function called to
3993 abort the program when an allocation inconsistency is detected now takes
3994 an argument that indicates the type of failure. The new function
3995 `mprobe' lets you request a consistency check for a particular block at
3996 any time (checks are normally done only when you call `free' or `realloc'
3997 on a block).
3998
3999 * It is now possible to easily cross-compile the C library, building on one
4000 system a library to run on another machine and/or operating system. All
4001 you need to do is set the variable `HOST_CC' in `configparms' to the
4002 native compiler for programs to run on the machine you are building on (a
4003 few generator programs are used on Unix systems); set `CC' to the
4004 cross-compiler.
4005
4006 * The new function `fexecve' (only implemented on the GNU system) executes
4007 a program file given a file descriptor already open on the file.
4008 \f
4009 Version 1.07
4010
4011 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4012 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4013
4014 * You can set the variable `libprefix' in `configparms' to specify a prefix
4015 to be prepended to installed library files; this makes it easy to install
4016 the GNU C library to be linked as `-lgnuc' or whatever.
4017
4018 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4019 copies a limited number of characters from a string, and returns the
4020 address of the last character written.
4021
4022 * You no longer need to check for whether the installed `stddef.h' is
4023 compatible with the GNU C library. configure now checks for you.
4024
4025 * You can now define a per-stream `fileno' function to convert the
4026 stream's cookie into an integral file descriptor.
4027
4028 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4029 allocates zero bytes of storage, and returns a unique pointer which
4030 you can pass to `realloc' or `free'. The behavior is undefined if
4031 you dereference this pointer.
4032
4033 * The C library now runs on Sony NEWS m68k machines running either
4034 NewsOS 3 or NewsOS 4.
4035
4036 * The new `syscall' function is a system-dependent primitive function
4037 for invoking system calls. It has the canonical behavior on Unix
4038 systems, including unreliable return values for some calls (such as
4039 `pipe', `fork' and `getppid').
4040
4041 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4042 to `EAGAIN', which is the preferred name. On systems whose kernels
4043 use two distinct codes, the C library now translates EWOULDBLOCK to
4044 EAGAIN in every system call function.
4045 \f
4046 Version 1.06
4047
4048 * The GNU C Library Reference Manual is now distributed with the library.
4049 `make dvi' will produce a DVI file of the printed manual.
4050 `make info' will produce Info files that you can read on line using C-h i
4051 in Emacs or the `info' program.
4052 Please send comments on the manual to bug-glibc-manual@gnu.org.
4053
4054 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4055
4056 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4057
4058 * Jason Merrill has contributed a port to the Sequent Symmetry running
4059 Dynix version 3 (i386-sequent-dynix).
4060
4061 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4062 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4063
4064 * New function `memory_warnings' lets you arrange to get warnings when
4065 malloc is running out of memory to allocate, like Emacs gives you.
4066
4067 * The C library now contains the relocating allocator used in Emacs 19 for
4068 its editing buffers. This allocator (ralloc) minimizes allocation
4069 overhead and fragmentation by moving allocated regions around whenever it
4070 needs to. You always refer to a ralloc'd region with a "handle" (a
4071 pointer to a pointer--an object of type `void **').
4072
4073 * There is a new `printf' format: `%m' gives you the string corresponding
4074 to the error code in `errno'.
4075
4076 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4077 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4078 pass, the `a' modifier says to fill in a `char **' you pass with a
4079 malloc'd string.
4080
4081 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4082 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4083 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4084
4085 * `mkstemp' is a traditional Unix function to atomically create and open a
4086 uniquely-named temporary file.
4087 \f
4088 Version 1.05
4089
4090 * The standard location for the file that says what the local timezone is
4091 has changed again. It is now `/usr/local/etc/localtime' (or more
4092 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4093
4094 * The distribution no longer contains any files with names longer than 14
4095 characters.
4096
4097 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4098 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4099
4100 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4101 \f
4102 Version 1.04
4103
4104 * The configuration process has changed quite a bit. The `configure'
4105 script is now used just like the configuration scripts for other GNU
4106 packages. The `sysdeps' directory hierarchy is much rearranged.
4107 The file `INSTALL' explains the new scheme in detail.
4108
4109 * The header files no longer need to be processed into ANSI C and
4110 traditional C versions. There is just one set of files to install, and
4111 it will work with ANSI or old C compilers (including `gcc -traditional').
4112
4113 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4114 MIPS DECStation running Ultrix 4.
4115
4116 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4117 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4118 made itself into a shared library.
4119
4120 * Yet further improved support for the i386, running 4.3 BSD-like systems
4121 (such as Mach 3 with the Unix single-server), or System V.
4122
4123 * New function `strncasecmp' to do case-insensitive string comparison
4124 with limited length.
4125
4126 * New function `strsep' is a reentrant alternative to `strtok'.
4127
4128 * New functions `scandir' and `alphasort' for searching directories.
4129
4130 * New function `setenv' is a better interface to `putenv'.
4131
4132 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4133 function for traversing a directory tree.
4134
4135 * The GNU obstack package is now also part of the C library.
4136 The new function `open_obstack_stream' creates a stdio stream that
4137 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4138 formatted output directly to an obstack.
4139
4140 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4141 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4142
4143 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4144
4145 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4146 things to your strings.
4147
4148 * There are some new test programs: test-fseek, testmb, and testrand.
4149
4150 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4151 These ports are not finished, but are a good starting place for really
4152 supporting those systems.
4153
4154 * `/etc/localtime' is now the standard location for the file that says what
4155 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4156 This follows the general principle that `/etc' is the place for all local
4157 configuration files.
4158
4159 * The C library header files now use `extern "C"' when used by the C++
4160 compiler, so the C library should now work with C++ code.
4161
4162 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4163 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4164 in <strings.h>.)
4165
4166 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4167 function which uses the `merge sort' algorithm, and is said to be
4168 significantly faster than the old GNU `qsort' function. Merge sort is
4169 now the standard `qsort' function. The new algorithm can require a lot
4170 of temporary storage; so, the old sorting function is called when the
4171 required storage is not available.
4172
4173 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4174 provides the Unix `crypt' function, plus some other entry points.
4175
4176 * The code and header files taken from 4.4 BSD have been updated with the
4177 latest files released from Berkeley.
4178 \f
4179 ----------------------------------------------------------------------
4180 Copyright information:
4181
4182 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4183
4184 Permission is granted to anyone to make or distribute verbatim copies
4185 of this document as received, in any medium, provided that the
4186 copyright notice and this permission notice are preserved,
4187 thus giving the recipient permission to redistribute in turn.
4188
4189 Permission is granted to distribute modified versions
4190 of this document, or of portions of it,
4191 under the above conditions, provided also that they
4192 carry prominent notices stating who last changed them.
4193 \f
4194 Local variables:
4195 version-control: never
4196 fill-column: 76
4197 End: