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.
5 Please send GNU C library bug reports via <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin and tan
13 with FMA, contributed by Arjan van de Ven and H.J. Lu from Intel.
15 Deprecated and removed features, and other changes affecting compatibility:
17 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
18 defined by <sys/ptrace.h>.
20 * libm no longer supports SVID error handling (calling a user-provided
21 matherr function on error) or the _LIB_VERSION variable to control error
22 handling. (SVID error handling and the _LIB_VERSION variable still work
23 for binaries linked against older versions of the GNU C Library.) The
24 libieee.a library is no longer provided. math.h no longer defines struct
25 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
26 TLOSS, PLOSS and HUGE.
28 Changes to build and runtime requirements:
30 [Add changes to build and runtime requirements here]
32 Security related changes:
34 CVE-2009-5064: The ldd script would sometimes run the program under
35 examination directly, without preventing code execution through the
36 dynamic linker. (The glibc project disputes that this is a security
37 vulnerability; only trusted binaries must be examined using the ldd
40 The following bugs are resolved with this release:
42 [The release manager will add the list generated by
43 scripts/list-fixed-bugs.py just before the release.]
50 * A per-thread cache has been added to malloc. Access to the cache requires
51 no locks and therefore significantly accelerates the fast path to allocate
52 and free small amounts of memory. Refilling an empty cache requires locking
53 the underlying arena. Performance measurements show significant gains in a
54 wide variety of user workloads. Workloads were captured using a special
55 instrumented malloc and analyzed with a malloc simulator. Contributed by
56 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
58 * Unicode 10.0.0 Support: Character encoding, character type info, and
59 transliteration tables are all updated to Unicode 10.0.0, using
60 generator scripts contributed by Mike FABIAN (Red Hat).
61 These updates cause user visible changes, especially the changes in
62 wcwidth for many emoji characters cause problems when emoji sequences
63 are rendered with pango, see for example:
64 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
66 * Collation of Hungarian has been overhauled and is now consistent with "The
67 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
70 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
72 - The GNU C Library will now detect when /etc/resolv.conf has been
73 modified and reload the changed configuration. The new resolver option
74 “no-reload” (RES_NORELOAD) disables this behavior.
76 - The GNU C Library now supports an arbitrary number of search domains
77 (configured using the “search” directive in /etc/resolv.conf);
78 previously, there was a hard limit of six domains. For backward
79 compatibility, applications that directly modify the ‘_res’ global
80 object are still limited to six search domains.
82 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
83 Library will now randomly pick a name server from the configuration as a
84 starting point. (Previously, the second name server was always used.)
86 * The tunables feature is now enabled by default. This allows users to tweak
87 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
89 * New function reallocarray, which resizes an allocated block (like realloc)
90 to the product of two sizes, with a guaranteed clean failure upon integer
91 overflow in the multiplication. Originally from OpenBSD, contributed by
92 Dennis Wölfing and Rüdiger Sonderfeld.
94 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
95 These are extended versions of preadv and pwritev, respectively, taking an
96 additional flags argument. The set of supported flags depends on the
97 running kernel; full support currently requires kernel 4.7 or later.
99 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
100 create a new session ID for the spawned process. This feature is
101 scheduled to be added to the next major revision of POSIX; for the time
102 being, it is available under _GNU_SOURCE.
104 * errno.h is now safe to use from C-preprocessed assembly language on all
105 supported operating systems. In this context, it will only define the
106 Exxxx constants, as preprocessor macros expanding to integer literals.
108 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
109 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
110 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
111 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
113 To compile programs that use this feature, the compiler must support
114 128-bit floating point with the type name _Float128 (as defined by TS
115 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
116 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
117 must be defined to make the new interfaces visible.
119 The new functions and macros correspond to those present for other
120 floating-point types (except for a few obsolescent interfaces not
121 supported for the new type), with F128 or f128 suffixes; for example,
122 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
123 printf or scanf formats for the new type; the strfromf128 and strtof128
124 interfaces should be used instead.
126 Deprecated and removed features, and other changes affecting compatibility:
128 * The synchronization that pthread_spin_unlock performs has been changed to
129 now be equivalent to a C11 atomic store with release memory order to the
130 spin lock's memory location. Previously, several (but not all)
131 architectures used stronger synchronization (e.g., containing what is
132 often called a full barrier). This change can improve performance, but
133 may affect odd fringe uses of spin locks that depend on the previous
134 behavior (e.g., using spin locks as atomic variables to try to implement
135 Dekker's mutual exclusion algorithm).
137 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
140 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
141 will only be built and installed when the GNU C Library is configured with
142 --enable-obsolete-rpc. This allows alternative RPC implementations, such
143 as TIRPC or rpcsvc-proto, to be used.
145 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
146 libnss_compat, are deprecated, and will not be built or installed by
149 The NIS(+) support library, libnsl, is also deprecated. By default, a
150 compatibility shared library will be built and installed, but not headers
151 or development libraries. Only a few NIS-related programs require this
152 library. (In particular, the GNU C Library has never required programs
153 that use 'gethostbyname' to be linked with libnsl.)
155 Replacement implementations based on TIRPC, which additionally support
156 IPv6, are available from <https://github.com/thkukuk/>. The configure
157 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
158 name service modules, to be built and installed.
160 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
161 support is enabled, the configured recursive resolver must support EDNS.
162 (Responding to EDNS-enabled queries with responses which are not
163 EDNS-enabled is fine, but FORMERR responses are not.)
165 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
166 servers have not supported this opcode for a long time.
168 * The _res_opcodes variable has been removed from libresolv. It had been
169 exported by accident.
171 * <string.h> no longer includes inline versions of any string functions,
172 as this kind of optimization is better done by the compiler. The macros
173 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
175 * The nonstandard header <xlocale.h> has been removed. Most programs should
176 use <locale.h> instead. If you have a specific need for the definition of
177 locale_t with no other declarations, please contact
178 libc-alpha@sourceware.org and explain.
180 * The obsolete header <sys/ultrasound.h> has been removed.
182 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
184 * The obsolete function cfree has been removed. Applications should use
187 * The stack_t type no longer has the name struct sigaltstack. This changes
188 the C++ name mangling for interfaces involving this type.
190 * The ucontext_t type no longer has the name struct ucontext. This changes
191 the C++ name mangling for interfaces involving this type.
193 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
194 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
195 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
196 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
197 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
198 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
199 fpregset_t type no longer has the name struct fpu. This changes the C++
200 name mangling for interfaces involving those types.
202 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
203 synced with the kernel:
205 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
206 are not supported on this architecture and have been removed.
208 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
209 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
210 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
212 Programs that assume the GET/SETREGS ptrace requests are universally
213 available will now fail to build, instead of malfunctioning at runtime.
215 Changes to build and runtime requirements:
217 * Linux kernel 3.2 or later is required at runtime, on all architectures
218 supported by that kernel. (This is a change from version 2.25 only for
221 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
223 * On most architectures, GCC 4.9 or later is required to build the GNU C
224 Library. On powerpc64le, GCC 6.2 or later is required.
226 Older GCC versions and non-GNU compilers are still supported when
227 compiling programs that use the GNU C Library. (We do not know exactly
228 how old, and some GNU extensions to C may be _de facto_ required. If you
229 are interested in helping us make this statement less vague, please
230 contact libc-alpha@sourceware.org.)
232 Security related changes:
234 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
235 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
237 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
238 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
240 * Avoid printing a backtrace from the __stack_chk_fail function since it is
241 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
244 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
245 fixed (CVE-2017-12133).
247 The following bugs are resolved with this release:
249 [984] network: Respond to changed resolv.conf in gethostbyname
250 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
251 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
253 [12189] libc: __stack_chk_fail should not attempt a backtrace
255 [14096] time: Race condition on timezone/tst-timezone.out
256 [14172] localedata: az_IR: new locale
257 [14995] build: glibc fails to build if gold is the default linker, even if
259 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
260 [16637] network: inet_pton function is accepting IPv6 with bad format
261 [16640] string: string/strtok.c: undefined behaviour inconsistent between
262 x86 and other generic code
263 [16875] localedata: ko_KR: fix lang_name
264 [17225] localedata: ar_SY: localized month names for May and June are
266 [17297] localedata: da_DK: wrong date_fmt string
267 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
268 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
269 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
270 [19066] localedata: ar_SA abbreviated day and month names are in English
271 [19569] network: resolv: Support an arbitrary number of search domains
272 [19570] network: Implement random DNS server selection in the stub
274 [19838] locale: localedef fails on PA-RISC
275 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
276 order of 0D36 and 0D37
277 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
279 [20098] libc: FAIL: debug/backtrace-tst on hppa
280 [20257] network: sunrpc: clntudp_call does not enforce timeout when
282 [20275] localedata: locale day/abday/mon/abmon should not have trailing
284 [20313] localedata: Update locale data to Unicode 9.0
285 [20424] manual: Document how to provide a malloc replacement
286 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
288 [20686] locale: Add el_GR@euro to SUPPORTED.
289 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
290 failures consistently
291 [21015] dynamic-link: Document and fix --enable-bind-now
292 [21016] nptl: pthread_cond support is broken on hppa
293 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
294 frame-pointer on i386
295 [21049] libc: segfault in longjmp_chk() due to clobbered processor
297 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
298 [21088] libc: Build fails with --enable-static-nss
299 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
300 generic c code is used
301 [21109] libc: Tunables broken on big-endian
302 [21112] math: powf has large ulp errors with base close to 1 and exponent
304 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
306 [21120] malloc: glibc malloc is incompatible with GCC 7
307 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
309 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
310 (-0.0) when linking with -lieee
311 [21171] math: log10, log2 and lgamma return incorrect results
312 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
313 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
314 [21207] localedata: ce_RU: update weekdays from CLDR
315 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
316 [21217] localedata: Update months from CLDR-31
317 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
318 [21243] libc: support_delete_temp_file should issue warning for failed
320 [21244] libc: support resolv_test_start() socket fd close should be
322 [21253] libc: localedef randomly segfaults when using -fstack-check due to
323 new posix_spawn implementation
324 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
325 leads to lower CPU frequency
326 [21259] libc: [alpha] termios.h missing IXANY for POSIX
327 [21261] libc: [sparc64] bits/setjmp.h namespace
328 [21267] network: [mips] bits/socket.h IOC* namespace
329 [21268] libc: [alpha] termios.h NL2, NL3 namespace
330 [21270] libc: mmap64 silently truncates large offset values
331 [21275] libc: posix_spawn always crashes on ia64 now
332 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
333 [21280] math: [powerpc] logbl for POWER7 return incorrect results
334 [21289] libc: Incorrect declaration for 32-bit platforms with
335 _FILE_OFFSET_BITS=64 causes build error
336 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
337 not support gethostbyname4_r
338 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
340 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
342 [21340] libc: Support POSIX_SPAWN_SETSID
343 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
344 [21359] network: ns_name_pack needs additional byte in destination buffer
345 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
346 against fragmentation attacks (CVE-2017-12132)
347 [21369] network: resolv: Remove EDNS fallback
348 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
350 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
351 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
352 [21393] stdio: Missing dup3 error check in freopen, freopen64
353 [21396] libc: Use AVX2 memcpy/memset on Skylake server
354 [21399] localedata: Bad description for U00EC in
355 localedata/charmaps/CP1254
356 [21411] malloc: realloc documentation error
357 [21426] network: sys/socket.h uio.h namespace
358 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
359 [21445] libc: signal.h bsd_signal namespace
360 [21455] network: Network headers stdint.h namespace
361 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
363 [21475] network: resolv: Overlong search path is truncated mid-label
364 [21511] libc: sigstack namespace
365 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
366 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
367 bits/syscall.h: No such file or directory
368 [21517] libc: struct sigaltstack namespace
369 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
370 [21533] localedata: Update locale data to Unicode 10.0
372 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
373 junk at end of line, first unrecognized character is `@'
374 [21538] libc: SIG_HOLD missing for XPG4
375 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
376 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
377 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
378 [21543] libc: sigevent namespace
379 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
380 size for general purpose registers in mcontext_t structure
381 [21550] libc: sigwait namespace
382 [21552] libc: XPG4 bsd_signal namespace
383 [21554] libc: sigpause namespace
384 [21560] libc: sys/wait.h signal.h namespace
385 [21561] libc: waitid namespace
386 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
387 '/usr/include/stdlib.h': Permission denied
388 [21575] libc: sys/wait.h missing struct rusage definition
389 [21584] libc: sigaltstack etc namespace
390 [21597] libc: siginterrupt namespace
391 [21607] math: hppa: FAIL: math/test-tgmath
392 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
394 [21622] libc: [tile] missing SA_* for POSIX.1:2008
395 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
396 stack and heap (CVE-2017-1000366)
397 [21625] libc: wait3 namespace
398 [21654] nss: Incorrect pointer alignment in NSS group merge result
400 [21657] network: Parse interface zone id for node-local multicast
401 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
403 [21666] libc: .symver is used on common symbol
404 [21668] network: resolv: res_init cross-thread broadcast introduces race
406 [21687] math: tgmath.h totalorder, totalordermag return type
407 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
409 [21696] libc: Incorrect assumption of of __cpu_mask in
410 posix/sched_cpucount.c
411 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
412 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
414 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
415 leading to relocation crash
416 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
417 `resolv_conf_matches (resp, conf)' failed.
418 [21710] localedata: Added Samoan language locale for Samoa
419 [21711] localedata: Pashto yesstr/nostr locale are missing
420 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
421 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
422 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
423 [21723] localedata: yesstr/nostr missing for Chinese language locale
424 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
426 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
428 [21728] localedata: New Locale for Tongan language
429 [21729] localedata: incorrect LC_NAME fields for hi_IN
430 [21733] localedata: yesstr and nostr are missing for zh_HK
431 [21734] localedata: Missing yesstr and nostr are for kw_GB
432 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
433 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
434 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
435 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
437 [21744] libc: Tests failing on --enable-tunables --enable-stack-
439 [21749] localedata: Wrong abbreviated day name (“abday”) for
441 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
442 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
443 [21759] localedata: missing yesstr and nostr for Tigrinya
444 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
445 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
446 [21767] localedata: Missing Bislama locales
447 [21768] localedata: Missing yesstr and nostr for aa_ET
448 [21770] localedata: Missing Field in li_NL
449 [21778] nptl: Robust mutex may deadlock
450 [21779] libc: MicroBlaze segfaults when loading libpthread
451 [21783] localedata: Fix int_select international_call_prefixes
452 [21784] localedata: Inconsistency in country_isbn
453 [21788] localedata: Missing Country Postal Abbreviations
454 [21794] localedata: Added-country_isbn-for-Italy
455 [21795] localedata: Add/Fix country_isbn for France
456 [21796] localedata: Added country_isbn for Republic of Korea
457 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
458 [21799] localedata: Added int_select international_call_prefixes
459 [21801] localedata: Added int_select international_call_prefixes
460 [21804] nptl: Double semicolon in thread-shared-types.h
461 [21807] localedata: LC_ADDRESS fix for pap_CW
462 [21808] localedata: Fix LC_ADDRESS for pap_AW
463 [21821] localedata: Added country_name in mai_IN
464 [21822] localedata: Fix LC_TIME for mai_IN
465 [21823] localedata: missing yesstr, nostr for sa_IN
466 [21825] localedata: Fix name_mrs for mag_IN
467 [21828] localedata: 2.26 changelog should mention user visible changes
469 [21835] localedata: Added Maithili language locale for Nepal
470 [21838] localedata: Removed redundant data for the_NP
471 [21839] localedata: Fix LC_MONETARY for ta_LK
472 [21844] localedata: Fix Latin characters and Months Sequence.
473 [21848] localedata: Fix mai_NP Title Name
478 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
479 24731-2:2010, is supported to enable declarations of functions from that
480 TR. Note that not all functions from that TR are supported by the GNU C
483 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
484 18661-1:2014, is supported to enable declarations of functions and macros
485 from that TS. Note that not all features from that TS are supported by
488 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
489 18661-4:2015, is supported to enable declarations of functions and macros
490 from that TS. Note that most features from that TS are not supported by
493 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
494 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
495 Since the GNU C Library defaults to a much newer revision of POSIX, this
496 will only affect programs that specifically request an old conformance
497 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
498 see a change in the visible declarations, but a program compiled with
499 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
502 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
503 defined by all multithreaded code, but glibc has not required this for
506 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
507 means that in a future release, the macros “major”, “minor”, and “makedev”
508 will only be available from <sys/sysmacros.h>.
510 These macros are not part of POSIX nor XSI, and their names frequently
511 collide with user code; see for instance glibc bug 19239 and Red Hat bug
512 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
513 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
516 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
517 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
518 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
520 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
521 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
522 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
523 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
524 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
525 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
526 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
527 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
528 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
529 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
530 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
533 * New <math.h> features are added from TS 18661-1:2014:
535 - Signaling NaN macros: SNANF, SNAN, SNANL.
537 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
538 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
539 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
541 - llogb functions: the llogb, llogbf and llogbl functions, and the
542 FP_LLOGB0 and FP_LLOGBNAN macros.
544 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
547 - Comparison macros: iseqsig.
549 - Classification macros: iscanonical, issubnormal, iszero.
551 - Total order functions: totalorder, totalorderf, totalorderl,
552 totalordermag, totalordermagf, totalordermagl.
554 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
556 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
557 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
559 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
560 are added to libc. They convert a floating-point number into string.
562 * Most of glibc can now be built with the stack smashing protector enabled.
563 It is recommended to build glibc with --enable-stack-protector=strong.
564 Implemented by Nick Alcock (Oracle).
566 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
567 intended to be used instead of memset() to erase sensitive data after use;
568 the compiler will not optimize out calls to explicit_bzero even if they
569 are "unnecessary" (in the sense that no _correct_ program can observe the
570 effects of the memory clear).
572 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
573 to float instead of double. This does not affect the ABI of any libraries
574 that are part of the GNU C Library, but may affect the ABI of other
575 libraries that use this type in their interfaces.
577 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
578 float_t and double_t types are now defined to long double instead of float
579 and double. These options are not the default, and this does not affect
580 the ABI of any libraries that are part of the GNU C Library, but it may
581 affect the ABI of other libraries that use this type in their interfaces,
582 if they are compiled or used with those options.
584 * The getentropy and getrandom functions, and the <sys/random.h> header file
587 * The buffer size for byte-oriented stdio streams is now limited to 8192
588 bytes by default. Previously, on Linux, the default buffer size on most
589 file systems was 4096 bytes (and thus remains unchanged), except on
590 network file systems, where the buffer size was unpredictable and could be
591 as large as several megabytes.
593 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
594 for the Linux quota interface which predates kernel version 2.4.22 has
597 * The malloc_get_state and malloc_set_state functions have been removed.
598 Already-existing binaries that dynamically link to these functions will
599 get a hidden implementation in which malloc_get_state is a stub. As far
600 as we know, these functions are used only by GNU Emacs and this change
601 will not adversely affect already-built Emacs executables. Any undumped
602 Emacs executables, which normally exist only during an Emacs build, should
603 be rebuilt by re-running “./configure; make” in the Emacs build tree.
605 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
606 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
607 “no-ip6-dotint” had already been the default, and support for the
608 “ip6-dotint” option was removed from the Internet in 2006.
610 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
611 flag from <resolv.h> have been removed. The option relied on a
612 backwards-incompatible DNS extension which was never deployed on the
615 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
616 RES_BLAST defined in the <resolv.h> header file have been deprecated.
617 They were already unimplemented.
619 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
620 _res.flags are deprecated. The flag was standardized in RFC 2133, but
621 removed again from the IETF name lookup interface specification in RFC
622 2553. Applications should use getaddrinfo instead.
624 * DNSSEC-related declarations and definitions have been removed from the
625 <arpa/nameser.h> header file, and libresolv will no longer attempt to
626 decode the data part of DNSSEC record types. Previous versions of glibc
627 only implemented minimal support for the previous version of DNSSEC, which
628 is incompatible with the currently deployed version.
630 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
631 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
632 <arpa/nameser.h> header file because the distinction between RR types and
633 meta-RR types is not officially standardized, subject to revision, and
634 thus not suitable for encoding in a macro.
636 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
637 and rhook members of the res_state type in <resolv.h> have been removed.
638 The glibc stub resolver did not support these hooks, but the header file
639 did not reflect that.
641 * For multi-arch support it is recommended to use a GCC which has
642 been built with support for GNU indirect functions. This ensures
643 that correct debugging information is generated for functions
644 selected by IFUNC resolvers. This support can either be enabled by
645 configuring GCC with '--enable-gnu-indirect-function', or by
646 enabling it by default by setting 'default_gnu_indirect_function'
647 variable for a particular architecture in the GCC source file
650 * GDB pretty printers have been added for mutex and condition variable
651 structures in POSIX Threads. When installed and loaded in gdb these pretty
652 printers show various pthread variables in human-readable form when read
653 using the 'print' or 'display' commands in gdb.
655 * Tunables feature added to allow tweaking of the runtime for an application
656 program. This feature can be enabled with the '--enable-tunables' configure
657 flag. The GNU C Library manual has details on usage and README.tunables has
658 instructions on adding new tunables to the library.
660 * A new version of condition variables functions have been implemented in
661 the NPTL implementation of POSIX Threads to provide stronger ordering
664 * A new version of pthread_rwlock functions have been implemented to use a more
665 scalable algorithm primarily through not using a critical section anymore to
668 Security related changes:
670 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
671 have been created with makecontext could fail to terminate due to a
672 missing .cantunwind annotation. This has been observed to lead to a hang
673 (denial of service) in some Go applications compiled with gccgo. Reported
674 by Andreas Schwab. (CVE-2016-6323)
676 * The DNS stub resolver functions would crash due to a NULL pointer
677 dereference when processing a query with a valid DNS question type which
678 was used internally in the implementation. The stub resolver now uses a
679 question type which is outside the range of valid question type values.
682 The following bugs are resolved with this release:
684 [4099] stdio: Overly agressive caching by stream i/o functions.
685 [7065] build: Support building glibc with -fstack-protector or -fstack-
687 [9842] localedata: en_CA: incorrect date format
688 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
689 before it started waiting
690 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
691 [15765] nptl: sem_open is wrongly a cancellation point
692 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
693 [16458] libc: endian.h and netinet/in.h byte order macros return values of
695 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
696 library linked with pthread
697 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
699 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
700 [17252] libc: getrandom and getentropy syscall
701 [17730] malloc: thread-local storage is sometimes improperly free()'d
702 after being __libc_memalign()'d
703 [18241] stdio: failed fseek on memstream does not set errno and can fail
705 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
707 [18463] nptl: pthread_cond_broadcast issue when surrounded by
708 PTHREAD_PRIO_INHERIT mutex on ARM
709 [18784] network: res_query and related function crash for special record
710 type queries (CVE-2015-5180)
711 [19380] math: strtod does not raise "inexact"
712 [19387] string: Integer overflow in memchr
713 [19390] string: Integer overflow in strncat
714 [19398] build: linknamespace tests fail with massively parallel build
715 [19402] nptl: Deadlock with robust shared mutex and asynchronous
717 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
718 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
719 [19514] libc: [PATCH] Fix spelling errors in spelling
720 "implement"/"implementation" in several places
721 [19582] network: Deprecate RES_USE_INET6
722 [19673] manual: clog10 docs appear to be erroneous
723 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
724 causes a segmentation fault
725 [19826] libc: invalid pointer returned from __tls_get_addr with static
727 [20016] network: resolv: Remove hooks support from the API
728 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
730 [20033] math: [x86_64] vectorized math function don't call the __finite
732 [20116] nptl: use after free in pthread_create
733 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
734 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
736 [20311] nptl: please install proc_service.h
737 [20366] libc: Compilation errors in installed headers in strict-compliance
739 [20370] malloc: malloc: Arena free list management is still racy
740 (incorrect fix in bug 19243)
741 [20386] libc: assert (X = 0) does not result in GCC warning
742 [20432] malloc: malloc: Minimize interface required for interposition
743 [20435] libc: Missing unwind info in __startcontext causes infinite loop
744 in _Unwind_Backtrace (CVE-2016-6323)
745 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
746 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
748 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
749 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
750 [20477] network: resolv: incorrect double-checked locking related to
752 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
754 [20495] math: x86_64 performance degradation due to AVX/SSE transition
756 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
757 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
758 cause transition penalty
759 [20517] math: sparcv9 missing fdiml compat symbol
760 [20524] manual: strverscmp is inconsistent
761 [20525] libc: <sys/quota.h> should be based on kernel headers
762 [20539] math: GCC 7 -static -lm fails to link at -O3
763 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
765 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
767 [20591] network: Remove obsolete DNSSEC support
768 [20592] network: DNS resource record type classification macros in
769 <arpa/nameser.h> are incorrect
770 [20593] network: Update DNS RR type definitions
771 [20611] network: getaddrinfo accepts invalid numeric scope IDs
772 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
774 [20629] network: libresolv: Remove support for bitstring labels
776 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
777 assertion: "maxidx >= 2"
778 [20660] math: [arm] Use VSQRT
779 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
780 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
781 [20689] libc: Test for FMA should also check for AVX.
782 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
783 [20715] math: iszero macro breaks existing code
784 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
785 [20728] libc: powerpc: Missing TOC stub in clone
786 [20729] build: glibc-2.24 fails to build for i486 with -Os
787 [20750] build: Build fails with default PIE enabled gcc-6.x
788 [20768] math: [alpha] sqrt fegetenv namespace
789 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
791 [20787] math: float_t is defined as float incorrectly on x86_64 even if
792 __FLT_EVAL_METHOD__ is 2
793 [20790] network: rpcgen buffer overrun in get_prog_declaration
794 [20822] nptl: race condition in __lll_unlock_elision on powerpc
795 [20829] libc: crypt snprintf namespace
796 [20847] libc: tst-vfork3 failure
797 [20855] math: Default bits/mathdef.h has inappropriate float_t
798 [20859] math: [sh4] FP_ILOGB0 invalid
799 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
800 U+20AC), not same as GBK
801 [20915] dynamic-link: global-dynamic TLS broken on aarch64
802 [20916] math: pow handling of sNaN arguments
803 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
804 [20919] math: Bad pow (qNaN, 0) result with -lieee
805 [20940] math: hypot sNaN handling
806 [20947] math: fmax, fmin sNaN handling
807 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
808 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
810 [20971] string: powerpc64/power7 memchr overflows internal pointer check
811 [20973] nptl: robust mutexes: Lost wake-ups
812 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
813 [20978] nis: strlen on null pointer in nss_nisplus
814 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
816 [21014] string: i686 memchr overflows internal pointer check
817 [21019] libc: [mips] n32 lseek incorrect on overflow
818 [21022] libc: [microblaze] __backtrace get_frame_size namespace
819 [21026] libc: [MIPS] readahead syscall is broken on n64
820 [21028] math: Fallback fesetexceptflag should always succeed
821 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
822 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
824 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
825 [21061] librt: [microblaze] librt lost clock_* exports
826 [21073] libc: tunables: insecure environment variables passed to
827 subprocesses with AT_SECURE
828 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
832 * The minimum Linux kernel version that this version of the GNU C Library
833 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
834 kernel version 2.6.32 or later suffices (on architectures that already
835 required kernel versions more recent than 3.2, those requirements remain
836 unchanged). Linux 3.2 or later kernel headers are required on all
839 * The pap_AN locale has been deleted. This has been deprecated for a long
840 time. It has been replaced by pap_AW & pap_CW, both of which have long
841 been included in previous releases.
843 * The readdir_r and readdir64_r functions have been deprecated. It is
844 recommended to use readdir and readdir64 instead.
846 * The type “union wait” has been removed. It was deprecated in the early
847 1990s and never part of POSIX. Application code should use the int type
848 instead of “union wait”.
850 * A new NSS action is added to facilitate large distributed system
851 administration. The action, MERGE, allows remote user stores like LDAP
852 to be merged into local user stores like /etc/groups in order to provide
853 easy to use, updated, and managed sets of merged credentials. The new
854 action can be used by configuring it in /etc/nsswitch.conf:
855 group: files [SUCCESS=merge] nis
856 Implemented by Stephen Gallagher (Red Hat).
858 * The deprecated __malloc_initialize_hook variable has been removed from the
861 * The long unused localedef --old-style option has been removed. It hasn't
862 done anything in over 16 years. Scripts using this option can safely
865 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
866 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
867 return the next representable value in the direction of positive infinity
868 and the nextdown functions return the next representable value in the
869 direction of negative infinity. These are currently enabled as GNU
872 Security related changes:
874 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
875 could result in a stack overflow when getnetbyname was called with an
876 overly long name. (CVE-2016-3075)
878 * Previously, getaddrinfo copied large amounts of address data to the stack,
879 even after the fix for CVE-2013-4458 has been applied, potentially
880 resulting in a stack overflow. getaddrinfo now uses a heap allocation
881 instead. Reported by Michael Petlan. (CVE-2016-3706)
883 * The glob function suffered from a stack-based buffer overflow when it was
884 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
885 Reported by Alexander Cherepanov. (CVE-2016-1234)
887 * The Sun RPC UDP client could exhaust all available stack space when
888 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
889 alloca plugin for GCC. (CVE-2016-4429)
891 * The IPv6 name server management code in libresolv could result in a memory
892 leak for each thread which is created, performs a failing naming lookup,
893 and exits. Over time, this could result in a denial of service due to
894 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
896 The following bugs are resolved with this release:
898 [1170] localedata: ne_NP: update Nepali locale definition file
899 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
901 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
902 [6796] math: fdim() does not set errno on overflow
903 [10354] libc: posix_spawn should use vfork() in more cases than presently
904 [11213] localedata: localedata: add copyright disclaimer to locale files
905 [12143] localedata: chr_US: new Cherokee locale
906 [12450] localedata: sgs_LT: new locale
907 [12676] localedata: ln_CD: new locale
908 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
910 [13304] math: fma, fmaf, fmal produce wrong results
911 [14259] build: --localedir arg to configure is ignored
912 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
913 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
914 [14934] localedata: es_CL: wrong first weekday chilean locale
915 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
917 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
919 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
920 [15368] nptl: raise() is not async-signal-safe
921 [15479] math: ceil, floor, round and trunc raise inexact exception
922 [15578] localedata: kk_KZ: various updates
923 [16003] localedata: pap_AN: punt old locale
924 [16137] localedata: iw_IL: punt old locale
925 [16190] localedata: eo: new esperanto locale
926 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
927 [16742] malloc: race condition: pthread_atfork() called before first
928 malloc() results in unexpected locking behaviour/deadlocks
929 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
931 [16983] localedata: postal_fmt does not allow %l and %n modifiers
932 [17565] localedata: pt_PT: wrong (work-)week start
933 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
934 [17950] build: Build fails with -msse
935 [18205] localedata: be_BY*: wrong first_weekday and first_workday
936 [18433] libc: posix_spawn does not return correctly upon failure to
938 [18453] localedata: charmaps/IBM875: incorrect codes
939 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
941 [18896] localedata: he_IL: improvements for currency
942 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
944 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
946 [19056] libc: Deprecate readdir_r
947 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
949 [19198] localedata: nl_NL: small improvements for Dutch locales
950 [19257] network: Per-thread memory leak in __res_vinit with IPv6
951 nameservers (CVE-2016-5417)
952 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
954 [19400] locale: Language missing in "iso-639.def", trivial fix in
956 [19431] malloc: Deadlock between fflush, getdelim, and fork
957 [19505] libc: Incorrect file descriptor validity checks in
958 posix_spawn_file_actions_add{open,close,dup2}
959 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
961 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
962 `intl/{gettextP,loadinfo}.h'
963 [19534] libc: execle, execlp may use malloc
964 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
965 [19573] network: res_nclose and __res_maybe_init disagree about name
966 server initialization, breaking Hesiod
967 [19575] localedata: Status of GB18030 tables
968 [19581] localedata: sr_* date_fmt string contains additional newline
969 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
971 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
972 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
973 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
974 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
976 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
978 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
980 [19610] dynamic-link: ldconfig -X removes stale symbolic links
981 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
982 [19633] locale: strfmon_l applies global locale to number formatting
983 [19642] network: Memory leak in getnameinfo
984 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
985 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
987 [19654] math: [x86_64] Need testcase for BZ #19590 fix
988 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
990 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
991 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
993 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
995 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
997 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
998 pointers and lengths in error-case.
999 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1000 report errors on UTF-16 surrogates values.
1001 [19755] nscd: nscd assertion failure in gc
1002 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1003 [19759] libc: mempcpy shouldn't be inlined
1004 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1005 [19765] libc: s390 needs an optimized mempcpy
1006 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1007 NAME_MAX limit assumption (CVE-2016-1234)
1008 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1009 [19787] network: Missing and incorrect truncation checks in getnameinfo
1010 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1012 [19791] network: Assertion failure in res_query.c with un-connectable name
1014 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1015 [19822] math: libm.so install clobbers old version
1016 [19825] network: resolv: send_vc can return uninitialized data in second
1017 response to getaddrinfo
1018 [19830] network: nss_dns: should check RDATA length against buffer length
1019 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1020 confronted with A/AAAA records of invalid size
1021 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1022 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1023 [19853] stdio: Printing IBM long double in decimal with high precision is
1025 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1026 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1027 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1029 [19865] network: Assertion failure or memory leak in
1030 _nss_dns_getcanonname_r
1031 [19868] network: nss_dns: netent code does not skip over non-PTR records
1032 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1034 [19881] string: Improve x86-64 memset
1035 [19907] string: Incorrect memcpy tests
1036 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1038 [19925] libc: termios.h XCASE namespace
1039 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1040 [19929] libc: limits.h NL_NMAX namespace
1041 [19931] stdio: Memory leak in vfprintf
1042 [19957] libc: clone(CLONE_VM) access invalid parent memory
1043 [19963] localedata: en_IL: New locale
1044 [19989] stdio: stdio.h cuserid namespace
1045 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1047 [19996] locale: langinfo.h nl_langinfo_l namespace
1048 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1049 [20010] network: getaddrinfo: Stack overflow in hostent translation
1051 [20012] stdio: libio: fmemopen append mode failure
1052 [20014] stdio: stdio.h namespace for pre-threads POSIX
1053 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1054 [20023] libc: fcntl.h timespec namespace
1055 [20024] math: [x86_64] vectorized sincos trashes the stack
1056 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1057 [20041] time: sys/time.h timespec namespace
1058 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1059 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1060 [20051] libc: ttyslot in wrong header under wrong conditions
1061 [20054] libc: gethostname not declared for XPG4
1062 [20055] libc: termios.h missing tcgetsid for XPG4
1063 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1065 [20073] libc: sys/stat.h fchmod namespace
1066 [20074] libc: stdlib.h rand_r namespace
1067 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1068 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1070 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1071 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1073 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1074 [20119] libc: Wrong mask for processors level type from CPUID
1075 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1076 AS not supporting AVX512
1077 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1079 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1080 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1082 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1083 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1084 [20178] libc: posix_spawn{p} should not call exit
1085 [20191] stdio: libio: vtables hardening
1086 [20195] string: FMA4 detection requires CPUID execution with register
1088 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1089 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1091 [20212] math: acos (sNaN) returns sNaN
1092 [20213] math: asin (sNaN) returns sNaN
1093 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1094 [20218] math: [i386] asinhl (sNaN) returns sNaN
1095 [20219] math: [i386] atanhl (sNaN) returns sNaN
1096 [20222] stdio: fopencookie: Mangle function pointers
1097 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1098 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1099 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1101 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1102 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1103 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1104 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1105 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1106 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1107 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1108 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1109 [20240] math: modf (sNaN) returns sNaN
1110 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1111 [20250] math: frexp (sNaN) returns sNaN
1112 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1113 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1115 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1116 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1117 subscript is below array bounds [-Werror=array-bounds]
1118 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1119 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1121 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1122 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1124 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1125 "invalid" exceptions
1126 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1127 [20316] localedata: id_ID: Februari instead of Pebruari
1128 [20327] string: POWER8 strcasecmp returns incorrect result
1129 [20347] math: Failure: Test: j0_downward (0xap+0)
1130 [20348] libc: FAIL: misc/tst-preadvwritev64
1131 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1132 [20350] libc: There is no test for p{read,write}64
1133 [20357] math: Incorrect cos result for 1.5174239687223976
1134 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1138 * Unicode 8.0.0 Support: Character encoding, character type info, and
1139 transliteration tables are all updated to Unicode 8.0.0, using new
1140 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1141 These updates cause user visible changes, such as the fixes for bugs
1142 89, 16061, and 18568.
1144 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1145 kernel-internal CPU set size. This means that requests that change the
1146 CPU affinity which failed before (for example, an all-ones CPU mask) will
1147 now succeed. Applications that need to determine the effective CPU
1148 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1149 setting it because the kernel can adjust it (and the previous size check
1150 would not detect this in the majority of cases).
1152 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1153 the following new symbols are used: fts64_children, fts64_close,
1154 fts64_open, fts64_read and fts64_set.
1156 * getaddrinfo now detects certain invalid responses on an internal netlink
1157 socket. If such responses are received, an affected process will
1158 terminate with an error message of "Unexpected error <number> on netlink
1159 descriptor <number>" or "Unexpected netlink response of size <number> on
1160 descriptor <number>". The most likely cause for these errors is a
1161 multi-threaded application which erroneously closes and reuses the netlink
1162 file descriptor while it is used by getaddrinfo.
1164 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1165 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1166 unnecessary serialization of memory allocation requests across threads.
1167 The defect is now corrected. Users should see a substantial increase in
1168 the concurent throughput of allocation requests for applications which
1169 trigger this bug. Affected applications typically create and
1170 destroy threads frequently. (Bug 19048 was reported and analyzed by
1173 * There is now a --disable-timezone-tools configure option for disabling the
1174 building and installing of the timezone related utilities (zic, zdump, and
1175 tzselect). This is useful for people who build the timezone data and code
1176 independent of the GNU C Library.
1178 * The obsolete header <regexp.h> has been removed. Programs that require
1179 this header must be updated to use <regex.h> instead.
1181 * The obsolete functions bdflush, create_module, get_kernel_syms,
1182 query_module and uselib are no longer available to newly linked binaries;
1183 the header <sys/kdaemon.h> has been removed. These functions and header
1184 were specific to systems using the Linux kernel and could not usefully be
1185 used with the GNU C Library on systems with version 2.6 or later of the
1188 * Optimized string, wcsmbs and memory functions for IBM z13.
1189 Implemented by Stefan Liebler.
1191 * Newly linked programs that define a variable called signgam will no longer
1192 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1193 require signgam to be set by those functions must ensure that they use the
1194 variable provided by the GNU C Library and declared in <math.h>, without
1195 defining their own copy.
1197 * The minimum GCC version that can be used to build this version of the GNU
1198 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1199 still be used to compile programs using the GNU C Library.
1201 Security related changes:
1203 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1204 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1206 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1207 the pointer guard feature. It is always enabled. Previously,
1208 LD_POINTER_GUARD could be used to disable security hardening in binaries
1209 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1212 * An integer overflow in hcreate and hcreate_r could lead to an
1213 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1215 * The catopen function no longer has unbounded stack usage. Reported by
1216 Max. (CVE-2015-8779)
1218 * The nan, nanf and nanl functions no longer have unbounded stack usage
1219 depending on the length of the string passed as an argument to the
1220 functions. Reported by Joseph Myers. (CVE-2014-9761)
1222 * A stack-based buffer overflow was found in libresolv when invoked from
1223 libnss_dns, allowing specially crafted DNS responses to seize control
1224 of execution flow in the DNS client. The buffer overflow occurs in
1225 the functions send_dg (send datagram) and send_vc (send TCP) for the
1226 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1227 family. The use of AF_UNSPEC triggers the low-level resolver code to
1228 send out two parallel queries for A and AAAA. A mismanagement of the
1229 buffers used for those queries could result in the response of a query
1230 writing beyond the alloca allocated buffer created by
1231 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1232 the overflow. Thanks to the Google Security Team and Red Hat for
1233 reporting the security impact of this issue, and Robert Holiday of
1234 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1236 The following bugs are resolved with this release:
1238 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1239 [887] math: Math library function "logb" and "nextafter" inconsistent
1240 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1241 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1242 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1243 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1245 [4404] localedata: German translation of "Alarm clock" is misleading
1246 [6799] math: nextafter() and nexttoward() doen't set errno on
1247 overflow/underflow errors
1248 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1250 [10432] nis: _nss_nis_setnetgrent assertion failure
1251 [11460] libc: fts has no LFS support
1252 [12926] network: getaddrinfo()/make_request() may spin forever
1253 [13065] nptl: Race condition in pthread barriers
1254 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1255 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1257 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1258 [14912] libc: Rename non-installed bits/*.h headers
1259 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1260 [15367] math: Let gcc use __builtin_isinf
1261 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1262 [15421] math: lgamma wrongly sets signgam for ISO C
1263 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1264 FE_INVALID with argument out of range
1265 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1267 [15786] dynamic-link: ifunc resolver functions can smash function
1269 [15918] math: Unnecessary check for equality in hypotf()
1270 [16061] localedata: Review / update transliteration data
1271 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1273 [16141] time: strptime %z offset restriction
1274 [16171] math: drem should be alias of remainder
1275 [16296] math: fegetround is pure?
1276 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1277 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1279 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1280 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1282 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1283 "Exception "Inexact" set" on ppc32
1284 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1285 [16517] math: Missing underflow exception from tanf/tan/tanl
1286 [16519] math: Missing underflow exception from sinhf
1287 [16520] math: Missing underflow exception from tanhf
1288 [16521] math: Missing underflow exception from exp2
1289 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1291 [16734] stdio: fopen calls mmap to allocate its buffer
1292 [16961] math: nan function incorrect handling of bad sequences
1293 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1294 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1295 [16985] locale: localedef: confusing error message when opening output
1297 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1298 [17197] locale: Redundant shift character in iconv conversion output at
1300 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1301 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1302 [17250] dynamic-link: static linking breaks nss loading
1303 (getaddrinfo/getpwnam/etc...)
1304 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1306 [17441] math: isnan() should use __builtin_isnan() in GCC
1307 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1308 (related to lock elision)
1309 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1310 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1311 [17887] time: strptime should be able to parse "+01:00" style timezones
1312 [17905] libc: catopen() Multiple unbounded stack allocations
1314 [18084] libc: backtrace (..., 0) dumps core on x86
1315 [18086] libc: nice() sets errno to 0 on success
1316 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1317 is too large (CVE-2015-8778)
1318 [18251] dynamic-link: SONAME missing when audit modules provides path
1319 [18265] libc: add attributes for wchar string and memory functions
1320 [18370] math: csqrt missing underflows
1321 [18421] libc: [hppa] read-only segment has dynamic relocations
1322 [18472] libc: Obsolete syscall wrappers should be compat symbols
1323 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1324 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1325 [18525] localedata: Remove locale timezone information
1326 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1327 [18568] localedata: Update locale data to Unicode 8.0
1328 [18589] locale: sort-test.sh fails at random
1329 [18595] math: ctan, ctanh missing underflows
1330 [18604] libc: assert macro-expands its argument
1331 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1332 contains a vector instruction exception.
1333 [18611] math: j1, jn missing errno setting on underflow
1334 [18618] localedata: sync Chechen locale definitions with other *_RU
1336 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1337 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1338 [18665] network: In send_dg, the recvfrom function is NOT always using the
1339 buffer size of a newly created buffer (CVE-2015-7547)
1340 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1342 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1344 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1345 [18699] math: tilegx cproj() for various complex infinities does not yield
1347 [18724] libc: Harden put*ent functions against data injection
1348 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1349 [18755] build: build errors with -DNDEBUG
1350 [18757] stdio: fmemopen fails to set errno on failure
1351 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1353 [18781] libc: openat64 lacks O_LARGEFILE
1354 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1355 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1356 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1357 [18790] math: [ldbl-128ibm] tanhl inaccurate
1358 [18795] libc: stpncpy fortification misses buffer lengths that are
1359 statically too large
1360 [18796] build: build fails for --disable-mathvec
1361 [18803] math: hypot missing underflows
1362 [18820] stdio: fmemopen may leak memory on failure
1363 [18823] math: csqrt spurious underflows
1364 [18824] math: fma spurious underflows
1365 [18825] math: pow missing underflows
1366 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1367 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1368 [18870] build: sem_open.c fails to compile with missing symbol
1370 [18872] stdio: Fix memory leak in printf_positional
1371 [18873] libc: posix_fallocate overflow check ineffective
1372 [18875] math: Excess precision leads incorrect libm
1373 [18877] libc: arm: mmap offset regression
1374 [18887] libc: memory corruption when using getmntent on blank lines
1375 [18918] localedata: hu_HU: change time to HH:MM:SS format
1376 [18921] libc: Regression: extraneous stat() and fstat() performed by
1378 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1379 binaries (CVE-2015-8777)
1380 [18951] math: tgamma missing underflows
1381 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1383 [18953] localedata: lt_LT: change currency symbol to the euro
1384 [18956] math: powf inaccuracy
1385 [18961] math: [i386] exp missing underflows
1386 [18966] math: [i386] exp10 missing underflows
1387 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1388 [18969] build: multiple string test failures due to missing locale
1390 [18970] libc: Reference of pthread_setcancelstate in libc.a
1391 [18977] math: float / long double Bessel functions not in XSI POSIX
1392 [18980] math: i386 libm functions return with excess range and precision
1393 [18981] math: i386 scalb*, ldexp return with excess range and precision
1394 [18982] stdio: va_list and vprintf
1395 [18985] time: Passing out of range data to strftime() causes a segfault
1397 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1398 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1399 [19012] locale: iconv_open leaks memory on error path
1400 [19016] math: clog, clog10 inaccuracy
1401 [19018] nptl: Mangle function pointers in tls_dtor_list
1402 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1403 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1404 [19048] malloc: malloc: arena free list can become cyclic, increasing
1406 [19049] math: [powerpc] erfc incorrect zero sign
1407 [19050] math: [powerpc] log* incorrect zero sign
1408 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1409 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1410 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1411 [19074] network: Data race in _res_hconf_reorder_addrs
1412 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1413 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1414 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1416 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1418 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1419 [19086] manual: posix_fallocate64 documented argument order is wrong.
1420 [19088] math: lround, llround missing exceptions close to overflow
1422 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1423 [19095] math: dbl-64 lrint incorrect for 64-bit long
1424 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1425 [19124] dynamic-link: ld.so failed to build with older assmebler
1426 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1427 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1428 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1429 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1430 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1431 pthread_setaffinity_np
1432 [19156] math: [ldbl-128] j0l spurious underflows
1433 [19164] nptl: tst-getcpu fails with many possible CPUs
1434 [19168] math: math/test-ildoubl and math/test-ldouble failure
1435 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1436 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1438 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1439 (FE_NOMASK_ENV) do not clear SSE exceptions
1440 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1441 _int_new_arena/reused_arena
1442 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1443 [19201] math: dbl-64 remainder incorrect sign of zero result
1444 [19205] math: bits/math-finite.h conditions do not match math.h and
1446 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1447 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1448 for C99-based standards
1449 [19212] libc: features.h not -Wundef clean
1450 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1452 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1453 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1454 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1456 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1457 "inexact" exceptions
1458 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1460 [19242] libc: strtol incorrect in Turkish locales
1461 [19243] malloc: reused_arena can pick an arena on the free list, leading
1462 to an assertion failure and reference count corruption
1463 [19253] time: tzset() ineffective when temporary TZ did not include DST
1465 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1466 [19270] math: [hppa] Shared libm missing __isnanl
1467 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1468 MAP_HUGETLB and MAP_STACK defines
1469 [19313] nptl: Wrong __cpu_mask for x32
1470 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1471 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1472 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1473 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1474 [19363] time: x32: times() return value wrongly truncates/sign extends
1476 [19367] dynamic-link: Improve branch prediction on Silvermont
1477 [19369] network: Default domain name not reset by res_ninit when "search"
1478 / "domain" entry is removed from resolv.conf
1479 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1481 [19385] localedata: bg_BG: time separator should be colon, not comma
1482 [19408] libc: linux personality syscall wrapper may erroneously return an
1483 error on 32-bit architectures
1484 [19415] libc: dladdr returns wrong names on hppa
1485 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1486 IBM905, IBM907, and IBM909
1487 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1488 [19443] build: build failures with -DDEBUG
1489 [19451] build: Make check fails on test-double-vlen2
1490 [19462] libc: Glibc failed to build with -Os
1491 [19465] math: Wrong code with -Os
1492 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1494 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1496 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1497 [19486] math: S390: Math tests fail with "Exception Inexact set".
1498 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1499 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1501 [19590] math: Fail to build shared objects that use libmvec.so functions.
1505 * The following bugs are resolved with this release:
1507 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1508 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1509 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1510 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1511 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1512 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1513 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1514 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1515 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1516 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1517 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1518 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1519 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1520 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1521 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1522 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1523 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1524 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1525 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1526 18657, 18676, 18694, 18696, 18887.
1528 * Cache information can be queried via sysconf() function on s390 e.g. with
1529 _SC_LEVEL1_ICACHE_SIZE as argument.
1531 * A buffer overflow in gethostbyname_r and related functions performing DNS
1532 requests has been fixed. If the NSS functions were called with a
1533 misaligned buffer, the buffer length change due to pointer alignment was
1534 not taken into account. This could result in application crashes or,
1535 potentially arbitrary code execution, using crafted, but syntactically
1536 valid DNS responses. (CVE-2015-1781)
1538 * The time zone file parser has been made more robust against crafted time
1539 zone files, avoiding heap buffer overflows related to the processing of
1540 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1541 large time zone data files. Overly long time zone specifiers in the TZ
1542 variable no longer result in stack overflows and crashes.
1544 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1545 for LD and GD on x86 and x86-64, has been implemented. You will need
1546 binutils-2.24 or later to enable this optimization.
1548 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1549 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1550 Hat). These updates cause user visible changes, such as the fix for bug
1553 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1554 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1555 condition in some applications.
1557 * Added vector math library named libmvec with the following vectorized x86_64
1558 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1560 The library can be disabled with --disable-mathvec. Use of the functions is
1561 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1562 Shared library libmvec.so is linked in as needed when using -lm (no need to
1563 specify -lmvec explicitly for not static builds).
1564 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1566 * A new fmemopen implementation has been added with the goal of POSIX
1567 compliance. The new implementation fixes the following long-standing
1568 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1569 old implementation is still present for use be by existing binaries.
1571 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1572 releases. It has been fixed to match 2.19 and older, but binaries built
1573 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1575 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1576 Contributed by Roland McGrath (Google).
1578 * The header <regexp.h> is deprecated, and will be removed in a future
1579 release. Use of this header will trigger a deprecation warning.
1580 Application developers should update their code to use <regex.h> instead.
1582 This header was formerly part of SUS, but was deprecated in 1994 and
1583 removed from the standard in 2001. Also, the glibc implementation
1584 leaks memory. See BZ#18681 for more details.
1588 * The following bugs are resolved with this release:
1590 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1591 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1592 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1593 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1594 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1595 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1596 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1597 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1598 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1599 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1602 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1603 too little memory for the to-be-scanned arguments and overflow the
1604 allocated buffer. The implementation now correctly computes the required
1605 buffer size when using malloc, and switches to malloc from alloca as
1608 * A new semaphore algorithm has been implemented in generic C code for all
1609 machines. Previous custom assembly implementations of semaphore were
1610 difficult to reason about or ensure that they were safe. The new version
1611 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1612 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1613 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1615 * Port to Altera Nios II has been contributed by Mentor Graphics.
1617 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1618 implementations for powerpc64/powerpc64le.
1619 Implemented by Adhemerval Zanella (IBM).
1621 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1622 and powerpc64le. This may improve lock scaling of existing programs on
1623 HTM capable systems. The lock elision code is only enabled with
1624 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1625 powerpc will issue a transaction abort on every syscall to avoid side
1626 effects being visible outside transactions.
1628 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1629 AArch64. Contributed by ARM Ltd.
1631 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1633 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1634 under certain input conditions resulting in the execution of a shell for
1635 command substitution when the application did not request it. The
1636 implementation now checks WRDE_NOCMD immediately before executing the
1637 shell and returns the error WRDE_CMDSUB as expected.
1639 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1640 processing format strings with a large number of format specifiers.
1642 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1643 infinite loop if the DNS response contained a PTR record of an unexpected
1646 * The minimum GCC version that can be used to build this version of the GNU
1647 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1648 still be used to compile programs using the GNU C Library.
1650 * The GNU C Library is now built with -Werror by default. This can be
1651 disabled by configuring with --disable-werror.
1653 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1655 * The obsolete sigvec function has been removed. This was the original
1656 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1657 programs have been using instead for about 25 years. Of course, ABI
1658 compatibility for old binaries using sigvec remains intact.
1660 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1661 with newer versions of bison.
1663 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1664 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1665 registers overlay two consecutive single-precision registers. MIPS32R2
1666 introduced a new FPU mode (FR=1) where double-precision registers extend the
1667 corresponding single-precision registers which is incompatible with the
1668 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1669 require the use of FR=1 making a transition necessary. New o32 ABI
1670 extensions enable users to migrate over time from the original o32 ABI
1671 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1672 tracks the ABI of any loaded object and verifies that new objects are
1673 compatible. Mode transitions will also be requested as required and
1674 unsupportable objects will be rejected. The ABI checks include both soft and
1675 hard float ABIs for o32, n32 and n64.
1677 GCC 5 with GNU binutils 2.25 onwards:
1678 It is strongly recommended that all o32 system libraries are built using the
1679 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1680 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1681 compiler using --with-fp-32=xx to set this by default.
1685 * The following bugs are resolved with this release:
1687 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1688 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1689 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1690 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1691 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1692 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1693 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1694 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1695 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1696 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1697 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1698 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1699 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1700 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1701 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1703 * Reverted change of ABI data structures for s390 and s390x:
1704 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1705 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1706 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1707 are preserved pointing straight to the same implementation as the old ones.
1708 Given that, new callers will simply provide a too-big buffer to these
1709 functions. Any applications/libraries out there that embed jmp_buf or
1710 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1711 against 2.19 headers will have to rebuilt again. This is necessary in any
1712 case to revert the breakage in their ABI caused by the glibc change.
1714 * Support for file description locks is added to systems running the
1715 Linux kernel. The standard file locking interfaces are extended to
1716 operate on file descriptions, not file descriptors, via the use of
1717 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1718 are associated with an open file instead of a process.
1720 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1722 * The minimum Linux kernel version that this version of the GNU C Library
1723 can be used with is 2.6.32.
1725 * Running the testsuite no longer terminates as soon as a test fails.
1726 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1727 with PASS or FAIL lines for individual tests. A summary of the results is
1728 printed, including a list of failing lists, and "make check" exits with
1729 error status if there were any unexpected failures. "make check
1730 stop-on-test-failure=y" may be used to keep the old behavior.
1732 * The am33 port, which had not worked for several years, has been removed
1735 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1736 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1737 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1738 conflicted with POSIX (support for which was removed in 2.19), the
1739 interfaces those macros enabled remain available when compiling with
1740 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1741 test macros defined.
1743 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1745 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1746 This may improve lock scaling of existing programs on TX capable systems.
1747 The lock elision code is only built with --enable-lock-elision=yes and
1748 then requires a GCC version supporting the TX builtins. With lock elision
1749 default mutexes are elided via __builtin_tbegin, if the cpu supports
1750 transactions. By default lock elision is not enabled and the elision code
1753 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1754 copy the path argument. This allowed programs to cause posix_spawn to
1755 deference a dangling pointer, or use an unexpected pathname argument if
1756 the string was modified after the posix_spawn_file_actions_addopen
1759 * All supported architectures now use the main glibc sysdeps directory
1760 instead of some being in a separate "ports" directory (which was
1761 distributed separately before glibc 2.17).
1763 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1764 On configurations that support it (all Linux configurations), it's now
1765 used regardless of the --enable-add-ons switch to configure. It is no
1766 longer possible to build such configurations without pthreads support.
1768 * Locale names, including those obtained from environment variables (LANG
1769 and the LC_* variables), are more tightly checked for proper syntax.
1770 setlocale will now fail (with EINVAL) for locale names that are overly
1771 long, contain slashes without starting with a slash, or contain ".." path
1772 components. (CVE-2014-0475) Previously, some valid locale names were
1773 silently replaced with the "C" locale when running in AT_SECURE mode
1774 (e.g., in a SUID program). This is no longer necessary because of the
1777 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1778 with application code using Intel MPX instructions. (With all previous
1779 versions, the MPX register state could be clobbered when making calls
1780 into or out of a shared library.) Note that while the new dynamic
1781 linker is compatible with all known x86 hardware whether or not it
1782 supports Intel MPX, some x86 instruction-set emulators might fail to
1783 handle the new instruction encodings. This is known to affect Valgrind
1784 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1785 release), and might affect other tools that do instruction emulation.
1787 * Support for loadable gconv transliteration modules has been removed.
1788 The support for transliteration modules has been non-functional for
1789 over a decade, and the removal is prompted by security defects. The
1790 normal gconv conversion modules are still supported. Transliteration
1791 with //TRANSLIT is still possible, and the //IGNORE specifier
1792 continues to be supported. (CVE-2014-5119)
1794 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1795 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1796 resulting a denial-of-service security vulnerability in applications which
1797 use functions related to iconv. (CVE-2014-6040)
1801 * The following bugs are resolved with this release:
1803 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1804 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1805 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1806 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1807 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1808 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1809 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1810 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1811 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1812 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1813 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1814 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1815 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1816 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1817 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1818 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1819 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1820 16431, 16453, 16474, 16506, 16510, 16529
1822 * Slovenian translations for glibc messages have been contributed by the
1823 Translation Project's Slovenian team of translators.
1825 * The public headers no longer use __unused nor __block. This change is to
1826 support compiling programs that are derived from BSD sources and use
1827 __unused internally, and to support compiling with Clang's -fblock
1828 extension which uses __block.
1830 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1831 large collation sequences to optimize multiple passes. This cache
1832 computation may overflow for large collation sequences and may cause a
1833 stack or buffer overflow. This is now fixed to use a slower algorithm
1834 which does not use a cache if there is an integer overflow.
1836 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1837 rules for large collation sequences to optimize multiple passes and falls
1838 back to alloca if malloc fails, resulting in a possible stack overflow.
1839 The implementation now falls back to an uncached collation sequence lookup
1842 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1843 initialized for static applications resulting in the security feature
1844 being disabled. The pointer guard is now correctly initialized to a
1845 random value for static applications. Existing static applications need
1846 to be recompiled to take advantage of the fix (bug 15754).
1848 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1849 to the d_name member of struct dirent, or omit the terminating NUL
1850 character. (Bugzilla #14699).
1852 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1853 aligned_alloc functions could allocate too few bytes or corrupt the
1854 heap when passed very large allocation size values (Bugzilla #15855,
1857 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1858 for AF_INET6 has been fixed (Bugzilla #16072).
1860 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1861 pap_CW, quz_PE, the_NP.
1863 * Substantially revised locales: gd_GB, ht_HT
1865 * The LC_ADDRESS field was updated to support country_car for almost all
1868 * ISO 1427 definitions were updated.
1870 * ISO 3166 definitions were updated.
1872 * The localedef utility now supports --big-endian and --little-endian
1873 command-line options to generate locales for a different system from that
1874 for which the C library was built.
1876 * Binary locale files now only depend on the endianness of the system for
1877 which they are generated and not on other properties of that system. As a
1878 consequence, binary files generated with new localedef may be incompatible
1879 with old versions of the GNU C Library, and binary files generated with
1880 old localedef may be incompatible with this version of the GNU C Library,
1881 in the following circumstances:
1883 + Locale files may be incompatible on m68k systems.
1885 + Locale archive files (but not separate files for individual locales) may
1886 be incompatible on systems where plain "char" is signed.
1888 * The configure option --disable-versioning has been removed. Builds with
1889 --disable-versioning had not worked for several years.
1891 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1892 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1894 * SystemTap probes for malloc have been introduced.
1896 * SystemTap probes for slow multiple precision fallback paths of
1897 transcendental functions have been introduced.
1899 * Support for powerpc64le has been added.
1901 * The soft-float powerpc port now supports e500 processors.
1903 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1905 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1906 set of header declarations that are enabled by default, even when other
1907 feature test macros or compiler options such as -std=c99 would otherwise
1908 disable some of those declarations.
1910 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1911 conflict with POSIX. The libbsd-compat library (which was a dummy library
1912 that did nothing) has also been removed.
1914 * Preliminary documentation about Multi-Thread, Async-Signal and
1915 Async-Cancel Safety has been added.
1917 * Change of ABI data structures for s390 and s390x:
1918 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1919 allow for future hardware extensions. All library functions that accept or
1920 return these structures were versioned in order to provide backwards
1921 compatibility with existing code. However, not all cases can be handled
1922 with interface versioning. If an application embeds either structure into
1923 another structure and that structure is passed to another compilation unit
1924 which expects the newer larger-sized object then any access to the new
1925 fields will result in undefined behaviour. Similarly any access to fields
1926 that were moved by the enlarging of the structures will cause undefined
1927 behaviour. To fix the undefined behaviour all objects that are part of the
1928 final in-memory image must agree on the size of structures, and this may
1929 require recompilation.
1933 * The following bugs are resolved with this release:
1935 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1936 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1937 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1938 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1939 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1940 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1941 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1942 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1943 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1944 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1945 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1946 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1947 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1950 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1951 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1952 Distributions can re-enable building and using pt_chown via the new configure
1953 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1954 considerable security risks and should only be used if the distribution
1955 understands and accepts the risks.
1957 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1960 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1961 fixed (Bugzilla #15330).
1963 * Add support for calling C++11 thread_local object destructors on thread
1964 and program exit. This needs compiler support for offloading C++11
1965 destructor calls to glibc.
1967 * Improved worst case performance of libm functions with double inputs and
1970 * Support for priority inherited mutexes in pthread condition variables on
1971 non-x86 architectures.
1973 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1975 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1977 * Optimized string functions for ARM. Implemented by Will Newton and
1980 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1982 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1985 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1986 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1988 * Added a benchmark framework to track performance of functions in glibc.
1990 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1991 It is based on draft TS 18661 and currently enabled as a GNU extension.
1993 * On Linux, the clock function now uses the clock_gettime system call
1994 for improved precision, rather than old times system call.
1996 * Added support for version-3 tz format files. This is needed when using
1997 the tz database release 2013e or later, and affects a few unusual cases --
1998 currently only TZ='America/Godthab' for time stamps after 2037.
2000 * Added new API functions pthread_getattr_default_np and
2001 pthread_setattr_default_np to get and set the default pthread
2002 attributes of a process.
2004 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2005 This may improve lock scaling of existing programs on TSX capable systems.
2006 When the --enable-lock-elision=yes parameter is specified at configure
2007 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2010 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2011 availability). Implemented by Ryan S. Arnold.
2013 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2017 * The following bugs are resolved with this release:
2019 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2020 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2021 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2022 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2023 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2024 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2025 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2026 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2027 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2028 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2029 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2030 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2031 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2032 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2033 14879, 14889, 14893, 14898, 14914.
2035 * Optimization of memcpy for MIPS.
2037 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2038 EMFILE has been fixed (Bugzilla #14889).
2040 * The add-on ports collection is now distributed in the "ports" subdirectory
2041 of the main GNU C Library distribution, rather than separately.
2043 * Port to ARM AArch64 contributed by Linaro.
2045 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2046 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2048 Implemented by Andreas Krebbel.
2050 * The new function secure_getenv allows secure access to the environment,
2051 returning NULL if running in a SUID/SGID process. This function replaces
2052 the internal function __secure_getenv.
2054 * SystemTap static probes have been added into the dynamic linker.
2055 Implemented by Gary Benson.
2057 * Optimizations of string functions strstr, strcasestr and memmem.
2058 Implemented by Maxim Kuvyrkov.
2060 * The minimum Linux kernel version that this version of the GNU C Library
2061 can be used with is 2.6.16.
2063 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2064 powerpc POWER7. Implemented by Will Schmidt.
2066 * New configure option --disable-nscd builds the C library such that it
2067 never attempts to contact the Name Service Caching Daemon (nscd).
2068 New configure option --disable-build-nscd avoids building nscd itself;
2069 this is the default if --disable-nscd is used.
2071 * Improved support for cross-compilation, including cross-testing and
2072 bootstrap builds without a previously built glibc.
2074 * Several testsuite tests are now able to test multiple IFUNC variants of an
2075 interface, rather than just testing the one that would be chooen by
2078 * New configure options --with-bugurl and --with-pkgversion, for
2079 distributors to use to embed their bug-reporting and package version
2080 information in --help and --version output.
2082 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2083 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2084 allows creation of chroots without the procfs mounted on /proc.
2086 * The `crypt' function now fails if passed salt bytes that violate the
2087 specification for those values. On Linux, the `crypt' function will
2088 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2089 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2090 when the mode is enabled.
2092 * The `clock_*' suite of functions (declared in <time.h>) is now available
2093 directly in the main C library. Previously it was necessary to link with
2094 -lrt to use these functions. This change has the effect that a
2095 single-threaded program that uses a function such as `clock_gettime' (and
2096 is not linked with -lrt) will no longer implicitly load the pthreads
2097 library at runtime and so will not suffer the overheads associated with
2098 multi-thread support in other code such as the C++ runtime library.
2100 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2105 * The following bugs are resolved with this release:
2107 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2108 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2109 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2110 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2111 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2112 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2113 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2114 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2115 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2116 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2117 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2118 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2119 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2120 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2121 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2122 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2123 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2124 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2125 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2126 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2127 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2128 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2131 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2132 configuring glibc with:
2133 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2134 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2135 Implemented by H.J. Lu.
2139 + define static_assert
2141 + do not declare gets
2143 + declare at_quick_exit and quick_exit also for ISO C11
2145 + aligned_alloc. NB: The code is deliberately allows the size parameter
2146 to not be a multiple of the alignment. This is a moronic requirement
2147 in the standard but it is only a requirement on the caller, not the
2150 + timespec_get added
2152 + uchar.h support added
2154 + CMPLX, CMPLXF, CMPLXL added
2156 Implemented by Ulrich Drepper.
2158 * Support for the IA-64 has been moved to ports.
2160 * Remove support for anything but ELF binary format
2162 * Checking versions of poll, ppoll added.
2163 Implemented by Ulrich Drepper.
2165 * More generic and 64-bit performance optimizations to math functions.
2166 Implemented by Ulrich Drepper.
2168 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2169 headers and functions available at compile time as they were before
2170 version 2.14. This option will be removed at some time in the future
2171 after the TI-RPC library becomes fully sufficient for the needs of
2172 existing applications.
2174 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2175 Note that glibc is not expected to work with any Linux kernel version
2178 * New header <sys/auxv.h> and function getauxval allowing easy access to
2179 the AT_* key-value pairs passed from the Linux kernel. The header also
2180 defines the HWCAP_* bits associated with the AT_HWCAP key.
2182 * New locales: mag_IN
2184 * New configure option --enable-systemtap builds SystemTap static probes
2185 into libc for setjmp and longjmp and into libpthread for various operations.
2186 So far the setjmp/longjmp probes and some of the libpthread probes are
2187 provided only for i*86 and x86_64.
2188 Implemented by Roland McGrath and Rayson Ho.
2190 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2192 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2195 * More optimized functions for SPARC. Implemented by David S. Miller.
2197 * Improved support for cross-compilation, especially bootstrap builds
2198 without a previously built glibc.
2200 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2201 Chris Metcalf from Tilera.
2203 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2204 now supported for ARM processors.
2206 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2207 as the name of the dynamic linker, to distinguish it from the
2208 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2210 * Support for CRIS has been removed from ports.
2212 * A new class of installed header has been documented for low-level
2213 platform-specific functionality. PowerPC added the first instance with a
2214 function to provide time base register access. Contributed by Tulio
2215 Magno Quites Machado Filho.
2217 * ix86 configurations now install header files that are consistent with
2218 what x86-64 configurations install. These same header files can be used
2219 for -m32, -m64, or -mx32 builds.
2220 Contributed by H.J. Lu.
2222 * Math library bug fixes. A thorough audit of all open math library bugs was
2223 conducted by Joseph Myers. Significant progress was made on many math
2224 library bugs resulting in more accurate exceptions and function results.
2225 Many thanks to all those that contributed including Andreas Jaeger for his
2226 patch review and work on the x87 trigonometric instruction issues.
2228 * Timezone data is no longer installed. Timezone-related binaries and scripts
2229 will continue to be installed. Users should obtain their timezone data from
2230 their distribution provider or from the tzdata package at
2231 <ftp://munnari.oz.au/pub/>.
2235 * The following bugs are resolved with this release:
2237 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2238 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2239 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2240 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2241 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2242 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2243 13484, 13506, 13515, 13523, 13524, 13538, 13540
2245 * New program pldd to list loaded object of a process
2246 Implemented by Ulrich Drepper.
2248 * Add nss_db support back to glibc. No more dependency on Berkeley db
2249 and support for initgroups lookups.
2250 Implemented by Ulrich Drepper.
2252 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2253 Contributed by HJ Lu.
2255 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2256 Contributed by HJ Lu.
2258 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2259 on x86-32 and x86-64.
2260 Contributed by Liubov Dmitrieva.
2262 * Optimized strchr and strrchr for SSE on x86-32.
2263 Contributed by Liubov Dmitrieva.
2265 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2266 for x86-64 and x86-32.
2267 Contributed by Liubov Dmitrieva.
2269 * New interfaces: scandirat, scandirat64
2270 Implemented by Ulrich Drepper.
2272 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2273 Implemented by Ulrich Drepper.
2275 * nscd now also caches the netgroup database.
2276 Implemented by Ulrich Drepper.
2278 * Integrate libm with gcc's -ffinite-math-only option.
2279 Implemented by Ulrich Drepper.
2281 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2282 to math functions. Implemented by Ulrich Drepper.
2284 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2285 Implemented by Ulrich Drepper.
2287 * New Linux interfaces: process_vm_readv, process_vm_writev
2289 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2290 Implemented by Ulrich Drepper.
2292 * Optimized nearbyint and strcasecmp for PPC.
2293 Implemented by Adhemerval Zanella.
2295 * New locales: bho_IN, unm_US, es_CU, ta_LK
2299 * The following bugs are resolved with this release:
2301 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2302 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2303 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2304 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2305 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2306 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2307 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2308 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2309 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2310 12795, 12811, 12813, 12814, 12841
2312 * The RPC implementation in libc is obsoleted. Old programs keep working
2313 but new programs cannot be linked with the routines in libc anymore.
2314 Programs in need of RPC functionality must be linked against TI-RPC.
2315 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2317 Visible changes of this change include (obviously) the inability to link
2318 programs using RPC functions without referencing the TI-RPC library and the
2319 removal of the RPC headers from the glibc headers.
2320 Implemented by Ulrich Drepper.
2322 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2323 syncfs, setns, sendmmsg
2325 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2326 yue_HK, lij_IT, mhr_RU
2328 * New encodings: CP770, CP771, CP772, CP773, CP774
2330 * New program sotruss to trace calls through PLTs
2331 Implemented by Ulrich Drepper.
2333 * The malloc hook implementation is marked deprecated and will be removed
2334 from the default implementation in the next version. The design never
2335 worked ever since the introduction of threads. Even programs which do
2336 not create threads themselves can use multiple threads created internally.
2340 * The following bugs are resolved with this release:
2342 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2343 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2344 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2345 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2348 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2350 * POWER7 optimizations: memset, memcmp, strncmp
2352 * New optimized string functions for x86-64: strnlen (SSE2),
2353 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2354 Implemented by Ulrich Drepper.
2358 * The following bugs are resolved with this release:
2360 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2361 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2362 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2363 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2364 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2365 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2366 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2367 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2369 * New interfaces: pthread_getname_np, pthread_setname_np
2371 * New Linux interface: recvmmsg
2373 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2375 * The dynamic linker now recognizes supported ABI versions from the
2376 EI_ABIVERSION field in the ELF header.
2377 Implemented by Ulrich Drepper.
2379 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2380 will not be used to fill in password fields in the passwd.byname replies.
2381 Instead it is used to synthesize the shadow.byname table, should it be
2382 missing. This is a useful mode in some installations involving Solaris.
2383 Implemented by Ulrich Drepper.
2385 * New locales: kok_IN, sq_MK, cv_RU
2389 * The following bugs are resolved with this release:
2391 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2392 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2393 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2394 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2395 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2396 10789, 10817, 10823, 10840, 10847
2398 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2400 Implemented by Ulrich Drepper.
2402 * Checking version of longjmp added that fails if an uninitialized stack
2403 frame would be created. Implemented by Ulrich Drepper.
2405 * STT_GNU_IFUNC is now supported in static executables.
2406 Implemented by H.J. Lu.
2408 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2410 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2411 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2412 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2413 Contributed by H.J. Lu.
2415 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2416 Implemented by Ulrich Drepper.
2418 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2420 Contributed by H.J. Lu.
2422 * Support for fma instruction in AVX on x86-64.
2423 Implemented by H.J. Lu and Ulrich Drepper.
2425 * AVX support in x86-64 auditing support in ld.so.
2426 Implemented by H.J. Lu.
2428 * STB_GNU_UNIQUE support added.
2429 Implemented by Ulrich Drepper.
2431 * Implement second fallback mode for optimized DNS lookup for even more
2432 broken environments. If two requests from the same port are not
2433 handled correctly close the socket and open a new one before sending
2434 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2435 can be used to select this mode right away, instead of rediscovering the
2436 necessity is every process again.
2437 Implemented by Ulrich Drepper.
2439 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2440 Implemented by Adam Tkac.
2442 * Optimized iconv conversions for S390x.
2443 Implemented by Andreas Krebbel.
2445 * Using condvars with PI mutexes is now more efficient due to kernel
2446 support for requeueing to PI futexes. NPTL support added for x86-64.
2447 Implemented by Ulrich Drepper.
2449 * New locales: ps_AF, my_MM
2453 * The following bugs are resolved with this release:
2455 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2456 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2457 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2458 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2459 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2461 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2462 Implemented by Ulrich Drepper.
2464 * New Linux interfaces: accept4, fallocate, fallocate64.
2465 Implemented by Ulrich Drepper.
2467 * Correct declarations of string function when used in C++ code. This
2468 could lead to compile errors for invalid C++ code.
2470 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2473 * New POSIX 2008 interface: psiginfo
2474 Implemented by Ulrich Drepper.
2476 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2477 Implemented by Ulrich Drepper.
2479 * Support for selecting between multiple function definitions at runtime
2480 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2482 * The libcrypt library can now use the hash function implementations in
2483 NSS. Implemented by Ulrich Drepper.
2485 * The malloc implementation can be compiled to be less memory efficient
2486 but higher performing in multi-threaded programs.
2487 Implemented by Ulrich Drepper.
2489 * New locales: nan_TW@latin, ks_IN
2491 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2492 Implemented by Ulrich Drepper.
2494 * Extended printf hook support. It is possible to use user-defined types
2495 and extend existing format specifiers.
2496 Implemented by Ulrich Drepper.
2498 * Handling for group shadow files has been added.
2499 Implemented by Ulrich Drepper.
2501 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2502 servers (the case, e.g., for some people using the built-in DNS
2503 server in ADSL modems/routers). There is a once-per-process timeout
2504 in case of a broken server. To avoid it, users can run nscd or put
2505 'options single-request' in /etc/resolv.conf.
2506 Implemented by Ulrich Drepper.
2510 * The following bugs are resolved with this release:
2512 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2513 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2514 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2515 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2516 6968, 6974, 6980, 6995, 7008, 7009, 7029
2518 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2519 up at the same time. Implemented by Ulrich Drepper.
2521 * TLS descriptors for LD and GD on x86 and x86-64.
2522 Implemented by Alexandre Oliva.
2524 * getaddrinfo now handles DCCP and UDPlite.
2525 Implemented by Ulrich Drepper.
2527 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2528 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2529 Implemented by Ulrich Drepper.
2531 * New implementation of memmem, strstr, and strcasestr which is O(n).
2532 Implemented by Eric Blake.
2534 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2536 * Implement "e" option for popen to open file descriptor with the
2537 close-on-exec flag set. Implemented by Ulrich Drepper.
2539 * Implement "b" mode for fmemopen. In this mode writes writes don't
2540 implicitly add a NUL byte and seeks from the end of the buffer really
2541 use the buffer end, not the string length as the basis.
2542 Implemented by Ulrich Drepper.
2544 * Many functions, exported and internal, now atomically set the close-on-exec
2545 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2547 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2549 Implemented by Pravin Satpute.
2551 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2555 * The following bugs are resolved with this release:
2557 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2558 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2559 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2560 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2561 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2562 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2563 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2565 * New locales: bo_CN, bo_IN, shs_CA.
2567 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2569 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2570 Implemented by Pravin Satpute.
2572 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2574 * nscd caches now all timeouts for DNS entries
2575 Implemented by Ulrich Drepper.
2577 * nscd is more efficient and wakes up less often.
2578 Implemented by Ulrich Drepper.
2580 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2581 vdprintf, and obstack_vprintf.
2582 Implemented by Jakub Jelinek.
2584 * Faster memset for x86-64.
2585 Implemented by Harsha Jagasia and H.J. Lu.
2587 * Faster memcpy on x86.
2588 Implemented by Ulrich Drepper.
2590 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2591 Implemented by Ulrich Drepper.
2593 * Faster sqrt and sqrtf implemention for some PPC variants.
2594 Implemented by Steven Munroe.
2598 * The following bugs are resolved with this release:
2600 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2601 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2602 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2603 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2604 5071, 5103, 5104, 5112, 5113, 5184, 5186
2606 * More checking functions: fread, fread_unlocked, open*, mq_open.
2607 Implemented by Jakub Jelinek and Ulrich Drepper.
2609 * Extend fortification to C++. Implemented by Jakub Jelinek.
2611 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2612 by not recognizing 'a' as a modifier when those specs are requested.
2613 Implemented by Jakub Jelinek.
2615 * PPC optimizations to math and string functions.
2616 Implemented by Steven Munroe.
2618 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2619 options to be passed. Implemented by Ulrich Drepper.
2621 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2623 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2624 Implemented by Ulrich Drepper.
2626 * Handle private futexes in the NPTL implementation.
2627 Implemented by Jakub Jelinek and Ulrich Drepper.
2629 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2630 Implemented by Roland McGrath and Ulrich Drepper.
2632 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2634 * SHA-256 and SHA-512 based password encryption.
2635 Implemented by Ulrich Drepper.
2637 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2638 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2641 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2642 Implemented by Ulrich Drepper.
2646 * The following bugs are resolved with this release:
2648 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2649 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2650 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2651 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2652 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2653 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2654 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2657 * New Linux interfaces: epoll_pwait, sched_getcpu.
2659 * New generic interfaces: strerror_l.
2661 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2665 * The following bugs are resolved with this release:
2667 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2668 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2669 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2670 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2671 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2672 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2673 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2674 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2675 3225, 3251, 3252, 3253, 3273
2677 * For Linux, the sorting of addresses returned by getaddrinfo now also
2678 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2679 Implemented by Ulrich Drepper.
2681 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2682 Implemented by Ulrich Drepper.
2684 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2686 * New iconv module for MIK. Contributed by Alexander Shopov.
2688 * For sites with broken group and/or passwd database, the auto-propagate
2689 option of nscd can prevent creating ID lookup entries from the results
2690 of a name lookup and vice versa. This usually is no problem but some
2691 site might have problems with the default behavior.
2692 Implemented by Ulrich Drepper.
2694 * Iterating over entire database in NIS can be slow. With the
2695 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2696 to trade time for memory. The entire database will be read at once.
2697 Implemented by Ulrich Drepper.
2699 * The interfaces introduced in RFC 3542 have been implemented by
2702 * Support for the new ELF hash table format was added by Ulrich Drepper.
2704 * Support for priority inheritance mutexes added by Jakub Jelinek and
2707 * Support for priority protected mutexes added by Jakub Jelinek.
2709 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2710 or_IN, csb_PL, fy_NL, sr_ME.
2714 * More overflow detection functions.
2716 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2717 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2719 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2720 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2721 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2722 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2723 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2724 by Masahide Washizawa.
2726 * It is now possible to install audit libraries for the dynamic linker using
2727 LD_AUDIT. Implemented by Ulrich Drepper.
2729 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2730 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2731 For a libc and libpthread that works well on Linux 2.4 kernels, we
2732 recommend using the stable 2.3 branch.
2734 * The new function open_wmemstream acts like open_memstream,
2735 but uses a wchar_t wide character buffer.
2737 * The new function ppoll is an improved version of poll, similar to pselect.
2739 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2740 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2741 renameat, unlinkat, linkat, symlinkat, readlinkat.
2743 * New Linux kernel system calls: unshare,
2744 inotify_init, inotify_add_watch, inotify_rm_watch.
2746 * The euidaccess function is now also known by the alias eaccess,
2747 for compatibility with some other systems.
2749 * Timezone data updated to 2006b version.
2753 * The following bugs are resolved with this release:
2755 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2756 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2757 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2758 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2759 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2760 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2762 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2764 * As of this release, GCC 4 can be used to compile the C Library.
2766 * Timezone data updated to 2005m version.
2770 * The following bugs are resolved with this release:
2772 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2773 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2774 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2775 777, 787, 821, 822, 823, 825
2777 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2781 * Support for RFC 3678. Real implementations exist only for Linux so far.
2782 Implemented by Ulrich Drepper.
2784 * nscd can now cache entries persistently. Expiring entries are reloaded.
2785 For speedups the cache can be shared in memory with client processes.
2786 Implemented by Ulrich Drepper.
2788 * nscd can now perform SELinux checks.
2789 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2791 * getaddrinfo queries are now cached. Canonical name lookup is performed
2793 Implemented by Ulrich Drepper.
2795 * The nothrow function attribute is used when headers are used by gcc when
2796 compiling C code. This can avoid significant amounts of exception
2799 * The malloc functions perform more error checking and are stricter when
2800 it comes to reacting on errors. The default action is to terminate
2801 the process after showing an error message. Implemented by Ulrich Drepper.
2803 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2804 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2806 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2807 namespaces using the new function dlmopen(). This feature is of course,
2808 like most other dynamic loading functionality, not available in statically
2809 linked applications. Implemented by Ulrich Drepper.
2811 * Low-overhead boundary checking variants of string and some stdio functions
2812 were added. These are to be used in conjunction with a gcc patch by
2813 Jakub Jelinek which adds calls to these functions if possible.
2814 Implemented by Jakub Jelinek and Ulrich Drepper.
2816 * Old code for several operating systems and machine architectures that
2817 have not been in working condition in a long time have been removed from
2818 the main source tree maintained by the GNU C Library's maintainers.
2819 These files are now reside in the separate `ports' source module
2820 that is usable as an add-on when building the library.
2824 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2825 interrogate the dynamic linker, compatible with the Solaris interface.
2827 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2828 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2830 * getifaddrs now uses the netlink interface on Linux to get its information.
2831 Implemented by Thorsten Kukuk.
2833 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2834 Implemented by Ulrich Drepper.
2836 * support for non-executable stacks on x86 has been added. Changes mostly
2839 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2842 * getaddrinfo now performs destination address selection according to
2847 * Thread-safe interfaces for many functions that access locale data
2848 were added in version 2.3, but these features were omitted from NEWS.
2849 Many functions have variants with an `_l' suffix that take a `locale_t'
2850 object as a parameter rather than consulting the current locale.
2851 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2852 create and maintain `locale_t' objects. Additionally, the new function
2853 `uselocale' sets "the current locale" (as used by functions not so
2854 parameterized) set for an individual thread. These features were added
2855 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2857 * The functions getresuid, getresgid, setresuid, and setresgid, which
2858 have long been available on Linux, are now declared in <unistd.h>
2859 and are now also available on the Hurd.
2861 * ELF thread-local storage support (TLS) now works on x86-64.
2863 * The new dynamic string token $LIB is expanded in shared library names.
2864 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2866 * Aldy Hernandez contributed complete software floating point support for
2867 PowerPC machines with no FPU.
2869 * fexecve is implemented on Linux.
2871 * The `btowc' function should work at least twice as fast due to
2872 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2874 * With approriate thread add-ons cancelable functions are now implemented
2875 in libc.so as well. No need to call the function in libpthread. This
2876 change allowed to finally disable the incorrect and expensive handling
2877 of weak definition in ld.so.
2879 * Yet more PLT entries in libc.so have been removed. We finally arrived
2880 at the bare minimum. Startup times improved appropriately.
2882 * Support for the new Linux/x86 system call interface was added. The
2883 AT_SYSINFO auxiliary vector entry is recognized and handled.
2887 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2890 * iconv (the program and the interface) now accepts empty names (excluding
2891 options like //TRANSLIT) to mean "use charset of current locale".
2893 * localedef can now transliterate characters in strings which are not in
2894 the provided charmap. The information from the input locale is used.
2896 * Prelinking support was added for ELF targets. This requires additional
2897 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2899 * Read-only stdio streams now use mmap to speed up operation by eliminating
2900 copying and buffer underflows. To use add 'm' to the mode string of
2901 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2903 * The malloc functions were completely rewritten by Wolfram Gloger based
2904 on Doug Lea's malloc-2.7.0.c.
2906 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2907 implementation of regex.
2909 * Bruno Haible upgraded the iconv and locale implementation to support
2912 * Contents of the LC_* and LANG environment variables in the CEN style are
2913 not recognized anymore. It never was used. Change by Ulrich Drepper.
2915 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2916 thread-local storage (TLS) ABI on some platforms.
2917 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2919 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2920 EUC-JISX0213, and TSCII.
2922 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2923 BSD-compatible interface for getting all network interface addresses.
2924 Implementation for IPv4 by Roland McGrath.
2926 * Loading of locale data is faster due to the introduction of a locale
2927 archive. Implemented by Roland McGrath and Ulrich Drepper.
2929 * Startup times are significantly reduced by not using exported functions
2930 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2933 * Steven Munroe contributed a port to PowerPC64/Linux.
2937 * The Hurd now uses the GNU libio implementation of stdio rather than the
2938 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2940 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2941 with the same behavior as the Linux system call of the same name.
2945 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2946 128-bit long double format.
2948 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2949 IBM1161, and IBM1162 charsets.
2951 * Andreas Jaeger contributed a port to x86-64/Linux.
2953 * Peter Bruin contributed a port to PowerPC/Hurd.
2955 * libc functions using I/O streams now can handle wide-oriented streams
2958 * optimizations in the dynamic linker. Binaries created by recent binutils
2959 versions start up quicker due to reduced time spent on relocations.
2961 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2965 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2966 asinh, atanh, j0 for the 128-bit long double format.
2968 * Bruno Haible updated all the code handling Unicode in some form to
2969 support Unicode 3.1.
2971 * Speed of regex for single-byte locales is back to previous levels.
2972 Patch by Isamu Hasegawa.
2974 * Alpha, SPARC, and IA-64 now also using floating stacks.
2976 * Startup time of internationalized applications greatly improved through
2977 iconv cache. Use iconvconfig to generate the cache file.
2978 Contributed by Ulrich Drepper.
2980 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2981 severe performance problems. Patch by David Mosberger.
2983 * The Hurd port got a lot more functionality like AIO, various stdio
2984 extensions, etc. Mainly done by Roland McGrath.
2986 * mtrace can now lookup symbols in shared libraries.
2990 * Intel's IA-64 math library is largely integrated. It provides fast and
2991 accurate implementatations for most basic and standard math functions
2992 in float, double, and long double format.
2994 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2995 and asin for the 96-bit long double format and asin, log, tan for the
2996 128-bit long double format.
2998 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2999 The basic double functions exist today. Contributed by Abraham Ziv
3000 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3001 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3003 * An asynchronous name lookup library was added. The interface is designed
3004 after POSIX AIO. The proposal was circulated beforehand to get comments.
3005 No negative ones came in. Implemented by Ulrich Drepper.
3007 * Port to S390/64bit contributed by Martin Schwidefsky
3008 <schwidefsky@de.ibm.com>.
3010 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3011 of functions for Linux/IA-64.
3013 * The RPC code is now thread safe. Threads can now use the same service
3014 of different services at the same time. Patch by Eric Norum
3015 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3017 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3018 family of functions for Linux/S390.
3020 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3021 of functions for Linux/x86.
3023 * Port to Linux/CRIS contributed by Axis Communications.
3027 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3028 we can now check for namespace violations and missing declarations. The
3029 result is that almost all headers are now Unix-compliant (as defined in
3030 the upcoming XPG6). The negative side is that some programs might need
3031 corrections, too, if they depend on the incorrect form of the headers in
3032 previous versions which defined too many symbols and included too many
3035 * regex now handles multibyte character sets correctly.
3036 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3038 * iconv (the program) does now conform to the upcoming XPG6 and handles
3039 charmaps. Instead of the charset names the path of charmaps can be
3040 provided and the conversion happens based on this data.
3041 Contributed by Ulrich Drepper.
3043 * The locale program now provides detailed information about the installed
3048 only lists the names of the supported locales
3052 provides details such as country, language, and codeset name.
3053 Contributed by Ulrich Drepper.
3057 * The gencat program now parses the input file according to the charset
3058 selected by the LC_CTYPE category. This is important for stateful
3059 character sets. To make generating catalogs easier there is a way
3060 to overwrite the charset selected by the locale: before the first
3061 message or $ quote line the catalog can contain a line like
3063 $ codeset=ISO-8859-2
3065 to select the charset (ISO-8859-2 in this case).
3067 Implemented by Shinya Hanataka and Ulrich Drepper.
3069 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3070 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3071 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3073 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3074 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3076 * The LANGUAGE environment variable is now ignored unless the locale is
3077 changed from the default "C" locale.
3079 * The usual bug fixes.
3083 * Greg McGary added runtime support for bounds checking using gcc's
3084 new -fbounded-pointers option. ix86 target is complete. PowerPC
3087 * Thorsten Kukuk added secure mode support to nscd.
3089 * The Berkeley db libraries have been removed.
3091 Related, the nss_db module is now in a separate package since it
3092 obviously requires a database library being available.
3094 * Wide character I/O streams implemented by Ulrich Drepper.
3096 * Functions from the extended socket API added by Ulrich Drepper.
3098 * Functions feenableexcept and fedisableexcept to control the
3099 behaviour of individual exceptions have been added by Andreas Jaeger.
3101 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3103 * The resolver code has been updated from bind 8.2.3-T5B which supports
3104 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3107 This change could in some situations effect backward compatibility. Since
3108 now `_res' is a thread-local instead of a global variable, modifying it
3109 in one thread does not have any effect in other threads.
3111 The resolver library was also extended to allow IPv6 as the transport
3112 protocol for the requests. This work was done by Stig Venaas.
3114 * Compatibility code for K&R C compilers has been removed from the
3115 header files. A ISO C compiler is needed to use the library
3116 (conforming to either C89 or C99 standard).
3118 * Complete rewrite of the localedef program to support multibyte character
3119 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3120 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3121 Implemented by Ulrich Drepper.
3123 Bruno Haible significantly improved the generation and use of the data
3124 structures for the wide character tables.
3126 * Plural handling in gettext implemented by Ulrich Drepper.
3128 * The utmp daemon has been removed.
3130 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3132 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3135 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3137 * POSIX barriers implemented by Kaz Kylheku.
3139 * POSIX spawn function family implemented by Ulrich Drepper.
3141 * POSIX spinlocks are now available.
3143 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3145 * the configure option --enable-kernel=X.Y.Z allows to strip out
3146 compatibility for kernel versions before X.Y.Z. This is currently only
3147 implemented for Linux.
3149 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3150 field sin6_scope_id. This means that all programs using IPv6 should be
3151 recompiled. Don't expect binary compatibility with previous glibc
3154 * various conversion modules for IBM character sets contributed by
3157 * IA-64 port by Jes Sorensen and HJ Lu.
3171 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3173 * New xdr functions are added; some rpc functions are now 64bit clean.
3175 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3177 * Fixed known incompatibilities with glibc 2.0.
3179 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3181 * Optimized a number of functions (especially the ELF dynamic loader).
3183 * Update timezone data files.
3185 * lots of charmaps corrections
3187 * some new locale definitions and charmaps
3192 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3193 conform to POSIX member type requirements. Symbol versions have been
3194 adjusted as needed within the library, and for direct use by applications,
3195 but there is potential for problems if third-party libraries use
3196 struct timeval as part of their interface. This does not present
3197 a problem for X and other "essential" system libraries.
3199 * An additional locale model to support C++ Standard Library locale
3200 model and probably more was implemented by Ulrich Drepper.
3202 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3205 * Miles Bader provided the `argp' function family to support hierachical
3206 command line argument parsing, layered on top of getopt.
3208 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3210 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3213 * scanf recognizes the %a and %A format for scanning floating point numbers.
3215 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3216 information and interfaces for the available integer types.
3218 * about 130 new math functions were added to implement the ISO C9x math
3221 * the new header <complex.h> contains definitions of the complex math
3222 functions from ISO C 9X.
3224 * the new header <tgmath.h> defines generic macros to use complex or
3225 real valued functions.
3227 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3229 * Andreas Jaeger provided a test suite for the math library.
3231 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3233 * Ulrich Drepper added character set conversion functions (iconv).
3235 * Optimized string functions have been added.
3237 * The localedata addon is now part of glibc.
3239 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3241 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3242 daemon for NSS (nscd).
3244 Missing a better place here are some numbers on improvements. Under
3245 Linux 2.1.125 un-tar-ing the kernel sources takes
3249 using local files 12.19s 6.88s 22.91s
3251 using NIS 13.92s 8.91s 26.34s
3253 using NIS & nscd 10.37s 7.34s 25.30s
3255 using NIS+ 27.57s 30.37s 640.46s
3257 using NIS+ & nscd 10.25s 7.83s 26.51s
3259 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3261 Keep in mind that non-namelookup related operations dominate above times.
3262 It was just a common complain that using NIS+ unpacking the kernel is
3265 [1] The old nscd implementation is not available anymore since it was
3266 distributed with glibc up to version 2.0.98 and thus is now replaced.
3268 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3270 * Mark Kettenis provided a Hesiod NSS module.
3272 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3273 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3275 * rcmd can now handle netgroups (Dick Streefland).
3277 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3280 * Support for the IPv6 protocol has been added to the socket API, as per the
3281 latest draft standards.
3283 * Support for Linux 2.2 has been added.
3285 * Interface changes relative to the latest 2.0.x release:
3286 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3287 addseverity NEW: Unix98
3288 alphasort64 NEW: LFS
3289 argp_err_exit_status NEW: argp, GNU ext
3290 argp_error NEW: argp, GNU ext
3291 argp_failure NEW: argp, GNU ext
3292 argp_help NEW: argp, GNU ext
3293 argp_parse NEW: argp, GNU ext
3294 argp_program_bug_address NEW: argp, GNU ext
3295 argp_program_version NEW: argp, GNU ext
3296 argp_program_version_hook NEW: argp, GNU ext
3297 argp_state_help NEW: argp, GNU ext
3298 argp_usage NEW: argp, GNU ext
3299 authdes_create NEW: Secure RPC
3300 authdes_getucred NEW: Secure RPC
3301 authdes_pk_create NEW: Secure RPC
3302 backtrace NEW: GNU ext.
3303 backtrace_symbols NEW: GNU ext.
3304 backtrace_symbols_fd NEW: GNU ext.
3306 cacosf NEW: ISO C 9x
3307 cacosh NEW: ISO C 9x
3308 cacoshf NEW: ISO C 9x
3309 cacoshl NEW: ISO C 9x
3310 cacosl NEW: ISO C 9x
3317 casinf NEW: ISO C 9x
3318 casinh NEW: ISO C 9x
3319 casinhf NEW: ISO C 9x
3320 casinhl NEW: ISO C 9x
3321 casinl NEW: ISO C 9x
3323 catanf NEW: ISO C 9x
3324 catanh NEW: ISO C 9x
3325 catanhf NEW: ISO C 9x
3326 catanhl NEW: ISO C 9x
3327 catanl NEW: ISO C 9x
3328 cbc_crypt NEW: Secure RPC
3332 ccoshf NEW: ISO C 9x
3333 ccoshl NEW: ISO C 9x
3339 cimagf NEW: ISO C 9x
3340 cimagl NEW: ISO C 9x
3341 clearerr_locked REMOVED
3342 clntunix_create NEW: sunrpc ext
3344 clog10 NEW: ISO C 9x
3345 clog10f NEW: ISO C 9x
3346 clog10l NEW: ISO C 9x
3356 cprojf NEW: ISO C 9x
3357 cprojl NEW: ISO C 9x
3359 crealf NEW: ISO C 9x
3360 creall NEW: ISO C 9x
3365 csinhf NEW: ISO C 9x
3366 csinhl NEW: ISO C 9x
3369 csqrtf NEW: ISO C 9x
3370 csqrtl NEW: ISO C 9x
3374 ctanhf NEW: ISO C 9x
3375 ctanhl NEW: ISO C 9x
3377 des_setparity NEW: Secure RPC
3378 ecb_crypt NEW: Secure RPC
3379 endutxent NEW: Unix98
3381 exp10f NEW: ISO C 9x
3382 exp10l NEW: ISO C 9x
3386 fattach NEW: STREAMS
3387 fdetach NEW: STREAMS
3391 feclearexcept NEW: ISO C 9x
3392 fegetenv NEW: ISO C 9x
3393 fegetexceptflag NEW: ISO C 9x
3394 fegetround NEW: ISO C 9x
3395 feholdexcept NEW: ISO C 9x
3397 feraiseexcept NEW: ISO C 9x
3398 ferror_locked REMOVED
3399 fesetenv NEW: ISO C 9x
3400 fesetexceptflag NEW: ISO C 9x
3401 fesetround NEW: ISO C 9x
3402 fetestexcept NEW: ISO C 9x
3403 feupdateenv NEW: ISO C 9x
3404 fflush_locked REMOVED
3408 fgets_unlocked NEW: GNU ext.
3409 fileno_locked REMOVED
3421 fputc_locked REMOVED
3422 fputs_unlocked NEW: GNU ext.
3423 fread_unlocked NEW: GNU ext.
3428 fstatvfs NEW: Unix98
3432 ftruncate64 NEW: LFS
3434 fwrite_unlocked NEW: GNU ext.
3435 gai_strerror NEW: IPv6
3439 getchar_locked REMOVED
3441 getdate_err NEW: Unix98
3442 getdate_r NEW: GNU ext.
3444 getnameinfo NEW: IPv6
3445 getnetname NEW: Secure RPC
3446 getpmsg NEW: STREAMS
3447 getpt NEW: Unix98 PTY
3448 getrlimit64 NEW: LFS
3449 getutxent NEW: Unix98
3450 getutxid NEW: Unix98
3451 getutxline NEW: Unix98
3452 glob64 NEW: GNU ext.
3453 globfree64 NEW: GNU ext.
3454 gnu_get_libc_release NEW: GNU ext.
3455 gnu_get_libc_version NEW: GNU ext.
3456 grantpt NEW: Unix98 PTY
3457 host2netname NEW: Secure RPC
3459 iconv_close NEW: iconv
3460 iconv_open NEW: iconv
3461 if_freenameindex NEW: IPv6
3462 if_indextoname NEW: IPv6
3463 if_nameindex NEW: IPv6
3464 if_nametoindex NEW: IPv6
3465 in6addr_any NEW: IPv6
3466 in6addr_loopback NEW: IPv6
3467 inet6_isipv4mapped NEW: IPv6
3468 isastream NEW: STREAMS
3469 iswblank NEW: GNU ext.
3470 key_decryptsession NEW: Secure RPC
3471 key_decryptsession_pk NEW: Secure RPC
3472 key_encryptsession NEW: Secure RPC
3473 key_encryptsession_pk NEW: Secure RPC
3474 key_gendes NEW: Secure RPC
3475 key_get_conv NEW: Secure RPC
3476 key_secretkey_is_set NEW: Secure RPC
3477 key_setnet NEW: Secure RPC
3478 key_setsecret NEW: Secure RPC
3479 llrint NEW: ISO C 9x
3480 llrintf NEW: ISO C 9x
3481 llrintl NEW: ISO C 9x
3482 llround NEW: ISO C 9x
3483 llroundf NEW: ISO C 9x
3484 llroundl NEW: ISO C 9x
3489 lrintf NEW: ISO C 9x
3490 lrintl NEW: ISO C 9x
3491 lround NEW: ISO C 9x
3492 lroundf NEW: ISO C 9x
3493 lroundl NEW: ISO C 9x
3495 makecontext NEW: Unix98
3496 mempcpy NEW: GNU ext.
3499 modify_ldt NEW: kernel
3503 nearbyint NEW: ISO C 9x
3504 nearbyintf NEW: ISO C 9x
3505 nearbyintl NEW: ISO C 9x
3506 netname2host NEW: Secure RPC
3507 netname2user NEW: Secure RPC
3508 nexttoward NEW: ISO C 9x
3509 nexttowardf NEW: ISO C 9x
3510 nexttowardl NEW: ISO C 9x
3514 passwd2des NEW: Secure RPC
3516 pow10f NEW: GNU ext.
3517 pow10l NEW: GNU ext.
3520 printf_size NEW: GNU ext.
3521 printf_size_info NEW: GNU ext.
3522 profil_counter REMOVED
3523 pthread_mutexattr_getkind_np REPLACED
3524 pthread_mutexattr_setkind_np REPLACED
3525 ptsname NEW: Unix98 PTY
3526 ptsname_r NEW: Unix98 PTY
3528 putchar_locked REMOVED
3529 putgrent NEW: GNU ext.
3531 putpmsg NEW: STREAMS
3532 pututxline NEW: Unix98
3536 readdir64_r NEW: LFS
3537 remquo NEW: ISO C 9x
3538 remquof NEW: ISO C 9x
3539 remquol NEW: ISO C 9x
3541 roundf NEW: ISO C 9x
3542 roundl NEW: ISO C 9x
3544 scalbln NEW: ISO C 9x
3545 scalblnf NEW: ISO C 9x
3546 scalblnl NEW: ISO C 9x
3548 sendfile NEW: kernel
3549 setcontext NEW: Unix98
3550 setrlimit64 NEW: LFS
3551 setutxent NEW: Unix98
3553 sigignore NEW: Unix98
3554 sigqueue NEW: POSIX.1b
3555 sigrelse NEW: Unix98
3556 sigset NEW: POSIX.1b
3557 sigtimedwait NEW: POSIX.1b
3558 sigwaitinfo NEW: POSIX.1b
3559 sincos NEW: GNU ext.
3560 sincosf NEW: GNU ext.
3561 sincosl NEW: GNU ext.
3565 strcasestr NEW: GNU ext.
3566 strtoimax NEW: ISO C 9x
3567 strtoumax NEW: ISO C 9x
3568 strverscmp NEW: GNU ext.
3569 svcauthdes_stats NEW: Secure RPC
3570 svcunix_create NEW: sunrpc etx
3571 svcunixfd_create NEW: sunrpc ext
3572 swapcontext NEW: Unix98
3573 tcgetsid NEW: Unix98 PTY
3574 tdestroy NEW: GNU ext.
3575 tgamma NEW: ISO C 9x
3576 tgammaf NEW: ISO C 9x
3577 tgammal NEW: ISO C 9x
3581 truncf NEW: ISO C 9x
3582 truncl NEW: ISO C 9x
3584 unlockpt NEW: Unix98 PTY
3585 updwtmpx NEW: Unix98
3586 user2netname NEW: Secure RPC
3587 utmpxname NEW: Unix98
3588 versionsort NEW: GNU ext.
3589 versionsort64 NEW: GNU ext.
3591 wcscasecmp NEW: GNU ext.
3592 wcsncasecmp NEW: GNU ext.
3593 wcsnlen NEW: GNU ext.
3594 wcstoimax NEW: ISO C 9x
3595 wcstoll NEW: ISO C 9x
3596 wcstoull NEW: ISO C 9x
3597 wcstoumax NEW: ISO C 9x
3599 wordexp NEW: POSIX.2
3600 wordfree NEW: POSIX.2
3601 write_profiling REMOVED
3602 xdecrypt NEW: Secure RPC
3603 xdr_authdes_cred NEW: Secure RPC
3604 xdr_authdes_verf NEW: Secure RPC
3605 xdr_cryptkeyarg NEW: Secure RPC
3606 xdr_cryptkeyarg2 NEW: Secure RPC
3607 xdr_cryptkeyres NEW: Secure RPC
3608 xdr_getcredres NEW: Secure RPC
3609 xdr_key_netstarg NEW: Secure RPC
3610 xdr_key_netstres NEW: Secure RPC
3611 xdr_keybuf NEW: Secure RPC
3612 xdr_keystatus NEW: Secure RPC
3613 xdr_netnamestr NEW: Secure RPC
3614 xdr_sizeof NEW: Secure RPC
3615 xdr_unixcred NEW: sunrpc ext
3616 xencrypt NEW: Secure RPC
3617 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3628 * inet_ntoa is thread-safe
3630 * updwtmp is moved from libutil to libc
3632 * rewrite of cbrt function
3634 * update of timezone data
3648 * add atoll function
3650 * fix complex problems in Berkeley DB code
3652 * fix math functions
3656 * fixed lots of header problems (especially Linux/GNU specific)
3658 * dynamic loader preserves all registers
3660 * Roland McGrath provided support for handling of auxiliary objects in
3661 the ELF dynamic loader.
3663 * support for parallel builds is improved
3667 * GNU extensions are no longer declared by default. To enable them you
3668 must define the macro `_GNU_SOURCE' in your program or compile with
3671 * The library has changed from using GNU ld symbol aliases to using weak
3672 symbols where available. The ELF object file format supports weak
3673 symbols; GNU ld also supports weak symbols in the a.out format. (There
3674 is also now support for other GNU ld extensions in ELF. Use the
3675 `--with-elf' option to configure to indicate you have ELF, and
3676 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3677 of many files which contained only symbol aliases, reducing the size of
3678 the source and the compiled library; many other files were renamed to
3679 less cryptic names previously occupied by the symbol alias files.
3680 There is a new header file <elf.h> for programs which operate on
3681 files in the ELF format.
3683 * Converted to Autoconf version 2, so `configure' has more options.
3684 Run `configure --help' to see the details.
3686 * The library can now be configured to build profiling, highly-optimized
3687 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3688 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3689 `configure' enable building these extra libraries. The shared library is
3690 built by default when using both ELF and GNU ld. When shared libraries
3691 are enabled, the new library `-ldl' is available for arbitrary run-time
3692 loading of shared objects; its interface is defined in <dlfcn.h>. The
3693 new header file <link.h> gives access to the internals of the run-time
3694 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3695 application of same name on other systems and it provides information
3696 about dynamically linked binaries.
3698 * The C library now provides the run-time support code for profiling
3699 executables compiled with `-pg'. Programs can control the profiling code
3700 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3701 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3702 the support for this file format was contributed by David Mosberger-Tang.
3704 * The math code has been replaced with a math library based on fdlibm from
3705 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3706 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3707 functions. The math functions now reside in a separate library, so
3708 programs using them will need to use `-lm' their linking commands.
3710 * John C. Bowman contributed optimized ix87 assembler inline functions.
3712 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3713 mechanism similar to that found in Solaris 2. This is now used for the
3714 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3715 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3716 file controls what services are used for each individual database. This
3717 works by loading shared libraries with names specified in `nsswitch.conf',
3718 so service modules can be changed or added at any time without even
3719 relinking any program. Currently there are the file, db, and NIS based
3720 NSS services available.
3722 * The new functions `strtoq' and `strtouq' parse integer values from
3723 strings, like `strtol' and `strtoul', but they return `long long int' and
3724 `unsigned long long int' values, respectively (64-bit quantities).
3726 * The new functions `strtof' and `strtold' parse floating-point values from
3727 strings, like `strtod', but they return `float' and `long double' values,
3728 respectively (on some machines `double' and `long double' are the same).
3730 * Ulrich Drepper has contributed new implementations of the floating-point
3731 printing and reading code used in the `printf' family of functions and
3732 `strtod', `strtof', and `strtold'. These new functions are perfectly
3733 accurate, and much faster than the old ones.
3735 * The implementation of the POSIX locale model was completely rewritten by
3736 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3737 to compile the POSIX locale definition.
3739 * The former dummy implementations of the strcoll and strxfrm function are
3740 now replaced by fully functional code contributed by Ulrich Drepper. The
3741 collation information comes from the POSIX locale definitions.
3743 * The new header <langinfo.h> defines an interface for accessing
3744 various locale-dependent data (using the locale chosen with `setlocale').
3746 * Ulrich Drepper has contributed a new suite of functions for operation on
3747 wide-character and multibyte-character strings, in <wchar.h>;
3748 and classification and case conversion of wide characters, in <wctype.h>.
3749 These new functions are conforming to the ISO C, Amendement 1 specification.
3751 * There is now a second implementation of the standard I/O library available.
3752 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3753 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3755 * You can now use positional parameter specifications in format strings
3756 for the `printf' and `scanf' families of functions. For example,
3757 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3758 ``Number 6, Mr Jones''. This is mainly useful when providing different
3759 format strings for different languages, whose grammars may dictate
3760 different orderings of the values being printed. To support this
3761 feature, the interface for `register_printf_handler' has changed; see
3762 the header file <printf.h> for details.
3764 * The `printf' and `scanf' families of functions now understand a new
3765 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3766 to group numbers as indicated by the locale; for `scanf' and friends, this
3767 says to accept as valid only a number with all the proper grouping
3768 separators in the right places. In the default "C" locale, numbers are
3769 not grouped; but locales for specific countries will define the usual
3770 conventions (i.e. separate thousands with `,' in the US locale).
3772 * The pgrp functions have been regularized, slightly incompatibly but much
3773 less confusingly. The core functions are now `getpgid' and `setpgid',
3774 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3775 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3776 provided for compatibility. There is no longer an incompatible `getpgrp'
3777 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3779 * The new header file <fts.h> and suite of functions simplify programs that
3780 operate on directory trees. This code comes from 4.4 BSD.
3782 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3783 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3784 NSS scheme used in glibc.
3786 * There is a new malloc debugging hook `__memalign_hook'.
3788 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3789 `unsigned int' in <sys/types.h>. These are for compatibility only and
3790 their use is discouraged.
3792 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3793 done differently, so that it works even without GNU ld.
3795 * New function `euidaccess' checks allowed access to a file like `access',
3796 but using the effective IDs instead of the real IDs.
3798 * The time zone data files have been updated for the latest and greatest
3799 local time conventions of the countries of the world.
3801 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3804 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3805 for formatting floating-point numbers. They are provided only for
3806 compatibility; new programs should use `sprintf' instead. There are
3807 also equivalent function for the `long double' floating-point type and
3808 all functions also exist in a reentrant form.
3810 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3811 functions for maintaining the login-record files (primarily of use to
3812 system programs such as `login'), and convenient functions for
3813 allocating and initializing a pseudo-terminal (pty) device.
3815 * Ulrich Drepper has contributed new support for System V style
3816 shared memory and IPC on systems that support it.
3818 * Ulrich Drepper has contributed several miscellaneous new functions found
3819 in System V: The `hsearch' family of functions provide an effective
3820 implementation of hash tables; `a64l' and `l64a' provide a very simple
3821 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3824 * Ulrich Drepper has contributed new reentrant counterparts for the
3825 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3827 * Ulrich Drepper has contributed new, highly-optimized versions of several
3828 string functions for the i486/Pentium family of processors.
3830 * Ulrich Drepper has updated the Linux-specific code, based largely
3831 on work done in Hongjiu Lu's version of GNU libc for Linux.
3832 The GNU library now supports Linux versions 2.0.10 and later,
3833 using the ELF object file format (i[3456]86-*-linux).
3835 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3837 * David Mosberger-Tang and Richard Henderson have ported the C library
3838 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3839 dynamic linking support for ELF/Alpha.
3841 * Richard Henderson contributed several Alpha optimized assembler function
3842 for arithmetic and string handling.
3844 * Ulrich Drepper has contributed a new set of message catalog functions to
3845 support multiple languages using the <libintl.h> interface, for use with
3846 his new package GNU gettext. Translation volunteers have contributed
3847 catalogs of the library's messages in Spanish, German, and Korean.
3849 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3850 program and the `catgets' function for reading the catalog files it
3851 creates. (The <libintl.h> interface is preferred; we include the
3852 <nl_types.h> interface using `catgets' only for source compatibility with
3853 programs already written to use it.)
3855 * New header file <values.h> gives SVID-compatible names for <limits.h>
3858 * Various new macros, declarations, and small header files for compatibility
3861 * New function `group_member' is a convenient way to check if a process has
3862 a given effective group ID.
3864 * When using GCC 2.7 and later, the socket functions are now declared in a
3865 special way so that passing an argument of type `struct sockaddr_in *',
3866 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3867 `struct sockaddr *' type, does not generate a type-clash warning.
3869 * New function `error' declared in header file <error.h> is a convenient
3870 function for printing error messages and optionally exiting; this is the
3871 canonical function used in GNU programs. The new functions `err', `warn',
3872 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3873 doing the same thing.
3875 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3876 POSIX.2 `glob' function to do ~ and {...} expansion.
3878 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3879 `clearenv' which is used in POSIX.9 is also available.
3881 * New function `getsid' returns session ID number on systems that support it.
3883 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3884 files <db.h> and <mpool.h> provide a rich set of functions for several
3885 types of simple databases stored in memory and in files, and <ndbm.h> is
3886 an old `ndbm'-compatible interface using the `db' functions. Link with
3887 `-ldb' to get these functions.
3889 * New macro `strdupa' copies a string like `strdup', but uses local stack
3890 space from `alloca' instead of dynamic heap space from `malloc'.
3892 * New function `strnlen' is like `strlen' but searches only a given maximum
3893 number of characters for the null terminator. `stpncpy', `strndup' and
3894 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3897 * New function `statfs' in header <sys/statfs.h>.
3899 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3900 provide convenient functions for operating on blocks of null-terminated
3903 * A new suite of functions in <utmp.h> handle all the details of reading
3904 and writing the utmp file.
3906 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3909 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3910 completely POSIX compliant and also implemented the extended functionality
3911 to handle alternate digit representation and alternate era date formats.
3913 * Ulrich Drepper provided an implementation of the `strptime' function
3914 defined in XPG4.2 which transforms a string into a `struct tm' value.
3916 * Paul Eggert provided the tzselect shell script as part of the timezone
3917 code. The shell script makes it easy to select the correct timezone
3920 * The implementation of the malloc family of functions is completely replaced
3921 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3922 The implementation uses the mmap function (if available) and it is
3923 optimized for the use in multi threaded programs.
3925 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3926 functions. This new functionality is usable by specifying a special
3927 salt string and it is compatible with implementation on *BSD systems.
3929 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3930 `getsubopt' to handle second level command line options, `bsd_signal'
3931 to access BSD style `signal' functionality, the obsolete `regexp' style
3934 * the `lchown' function is available on system which support this
3937 * The implementation of the shadow password handling function was contributed
3940 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3942 * POSIX.1g support was added. The <sys/select.h> header is available,
3943 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3944 implementation of `getaddrinfo'.
3948 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3950 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3951 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3953 * The new function `daemon' from 4.4 BSD is useful for server programs that
3954 want to put themselves in the background.
3956 * Joel Sherrill has contributed support for several standalone boards that
3957 run without an operating system.
3959 * `printf', `scanf' and friends now accept a `q' type modifier for long
3960 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3962 * All of the code taken from BSD (notably most of the math and networking
3963 routines) has been updated from the BSD 4.4-Lite release.
3965 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3967 * The new functions `getdomainname' and `setdomainname' fetch or change the
3968 YP/NIS domain name. These are system calls which exist on systems which
3971 * The time zone data files have been updated for the latest international
3974 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3975 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3979 * The C library now includes support for Sun RPC, from Sun's free
3980 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3981 are included. (There is still no support for YP.)
3983 * Tom Quinn has contributed a port of the C library to SGI machines running
3984 Irix 4 (mips-sgi-irix4).
3986 * The new `lockf' function is a simplified interface to the locking
3987 facilities of `fcntl', included for compatibility.
3989 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3991 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3993 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3996 * The error message printed by `assert' for a failed assertion now includes
3997 the name of the program (if using GNU ld) and the name of the calling
3998 function (with versions of GCC that support this).
4000 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4002 * The library now includes the <sys/mman.h> header file and memory
4003 management functions `mmap', `munmap', `mprotect', `msync', and
4004 `madvise', on systems that support those facilities.
4006 * The interface for `mcheck' has changed slightly: the function called to
4007 abort the program when an allocation inconsistency is detected now takes
4008 an argument that indicates the type of failure. The new function
4009 `mprobe' lets you request a consistency check for a particular block at
4010 any time (checks are normally done only when you call `free' or `realloc'
4013 * It is now possible to easily cross-compile the C library, building on one
4014 system a library to run on another machine and/or operating system. All
4015 you need to do is set the variable `HOST_CC' in `configparms' to the
4016 native compiler for programs to run on the machine you are building on (a
4017 few generator programs are used on Unix systems); set `CC' to the
4020 * The new function `fexecve' (only implemented on the GNU system) executes
4021 a program file given a file descriptor already open on the file.
4025 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4026 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4028 * You can set the variable `libprefix' in `configparms' to specify a prefix
4029 to be prepended to installed library files; this makes it easy to install
4030 the GNU C library to be linked as `-lgnuc' or whatever.
4032 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4033 copies a limited number of characters from a string, and returns the
4034 address of the last character written.
4036 * You no longer need to check for whether the installed `stddef.h' is
4037 compatible with the GNU C library. configure now checks for you.
4039 * You can now define a per-stream `fileno' function to convert the
4040 stream's cookie into an integral file descriptor.
4042 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4043 allocates zero bytes of storage, and returns a unique pointer which
4044 you can pass to `realloc' or `free'. The behavior is undefined if
4045 you dereference this pointer.
4047 * The C library now runs on Sony NEWS m68k machines running either
4048 NewsOS 3 or NewsOS 4.
4050 * The new `syscall' function is a system-dependent primitive function
4051 for invoking system calls. It has the canonical behavior on Unix
4052 systems, including unreliable return values for some calls (such as
4053 `pipe', `fork' and `getppid').
4055 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4056 to `EAGAIN', which is the preferred name. On systems whose kernels
4057 use two distinct codes, the C library now translates EWOULDBLOCK to
4058 EAGAIN in every system call function.
4062 * The GNU C Library Reference Manual is now distributed with the library.
4063 `make dvi' will produce a DVI file of the printed manual.
4064 `make info' will produce Info files that you can read on line using C-h i
4065 in Emacs or the `info' program.
4066 Please send comments on the manual to bug-glibc-manual@gnu.org.
4068 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4070 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4072 * Jason Merrill has contributed a port to the Sequent Symmetry running
4073 Dynix version 3 (i386-sequent-dynix).
4075 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4076 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4078 * New function `memory_warnings' lets you arrange to get warnings when
4079 malloc is running out of memory to allocate, like Emacs gives you.
4081 * The C library now contains the relocating allocator used in Emacs 19 for
4082 its editing buffers. This allocator (ralloc) minimizes allocation
4083 overhead and fragmentation by moving allocated regions around whenever it
4084 needs to. You always refer to a ralloc'd region with a "handle" (a
4085 pointer to a pointer--an object of type `void **').
4087 * There is a new `printf' format: `%m' gives you the string corresponding
4088 to the error code in `errno'.
4090 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4091 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4092 pass, the `a' modifier says to fill in a `char **' you pass with a
4095 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4096 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4097 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4099 * `mkstemp' is a traditional Unix function to atomically create and open a
4100 uniquely-named temporary file.
4104 * The standard location for the file that says what the local timezone is
4105 has changed again. It is now `/usr/local/etc/localtime' (or more
4106 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4108 * The distribution no longer contains any files with names longer than 14
4111 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4112 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4114 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4118 * The configuration process has changed quite a bit. The `configure'
4119 script is now used just like the configuration scripts for other GNU
4120 packages. The `sysdeps' directory hierarchy is much rearranged.
4121 The file `INSTALL' explains the new scheme in detail.
4123 * The header files no longer need to be processed into ANSI C and
4124 traditional C versions. There is just one set of files to install, and
4125 it will work with ANSI or old C compilers (including `gcc -traditional').
4127 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4128 MIPS DECStation running Ultrix 4.
4130 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4131 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4132 made itself into a shared library.
4134 * Yet further improved support for the i386, running 4.3 BSD-like systems
4135 (such as Mach 3 with the Unix single-server), or System V.
4137 * New function `strncasecmp' to do case-insensitive string comparison
4138 with limited length.
4140 * New function `strsep' is a reentrant alternative to `strtok'.
4142 * New functions `scandir' and `alphasort' for searching directories.
4144 * New function `setenv' is a better interface to `putenv'.
4146 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4147 function for traversing a directory tree.
4149 * The GNU obstack package is now also part of the C library.
4150 The new function `open_obstack_stream' creates a stdio stream that
4151 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4152 formatted output directly to an obstack.
4154 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4155 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4157 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4159 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4160 things to your strings.
4162 * There are some new test programs: test-fseek, testmb, and testrand.
4164 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4165 These ports are not finished, but are a good starting place for really
4166 supporting those systems.
4168 * `/etc/localtime' is now the standard location for the file that says what
4169 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4170 This follows the general principle that `/etc' is the place for all local
4171 configuration files.
4173 * The C library header files now use `extern "C"' when used by the C++
4174 compiler, so the C library should now work with C++ code.
4176 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4177 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4180 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4181 function which uses the `merge sort' algorithm, and is said to be
4182 significantly faster than the old GNU `qsort' function. Merge sort is
4183 now the standard `qsort' function. The new algorithm can require a lot
4184 of temporary storage; so, the old sorting function is called when the
4185 required storage is not available.
4187 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4188 provides the Unix `crypt' function, plus some other entry points.
4190 * The code and header files taken from 4.4 BSD have been updated with the
4191 latest files released from Berkeley.
4193 ----------------------------------------------------------------------
4194 Copyright information:
4196 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4198 Permission is granted to anyone to make or distribute verbatim copies
4199 of this document as received, in any medium, provided that the
4200 copyright notice and this permission notice are preserved,
4201 thus giving the recipient permission to redistribute in turn.
4203 Permission is granted to distribute modified versions
4204 of this document, or of portions of it,
4205 under the above conditions, provided also that they
4206 carry prominent notices stating who last changed them.
4209 version-control: never