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 * In order to support faster and safer process termination the malloc API
16 family of functions will no longer print a failure address and stack
17 backtrace after detecting heap corruption. The goal is to minimize the
18 amount of work done after corruption is detected and to avoid potential
19 security issues in continued process execution. Reducing shutdown time
20 leads to lower overall process restart latency, so there is benefit both
21 from a security and performance perspective.
23 Deprecated and removed features, and other changes affecting compatibility:
25 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
26 defined by <sys/ptrace.h>.
28 * libm no longer supports SVID error handling (calling a user-provided
29 matherr function on error) or the _LIB_VERSION variable to control error
30 handling. (SVID error handling and the _LIB_VERSION variable still work
31 for binaries linked against older versions of the GNU C Library.) The
32 libieee.a library is no longer provided. math.h no longer defines struct
33 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
34 TLOSS, PLOSS and HUGE.
36 Changes to build and runtime requirements:
38 [Add changes to build and runtime requirements here]
40 Security related changes:
42 CVE-2009-5064: The ldd script would sometimes run the program under
43 examination directly, without preventing code execution through the
44 dynamic linker. (The glibc project disputes that this is a security
45 vulnerability; only trusted binaries must be examined using the ldd
48 The following bugs are resolved with this release:
50 [The release manager will add the list generated by
51 scripts/list-fixed-bugs.py just before the release.]
58 * A per-thread cache has been added to malloc. Access to the cache requires
59 no locks and therefore significantly accelerates the fast path to allocate
60 and free small amounts of memory. Refilling an empty cache requires locking
61 the underlying arena. Performance measurements show significant gains in a
62 wide variety of user workloads. Workloads were captured using a special
63 instrumented malloc and analyzed with a malloc simulator. Contributed by
64 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
66 * Unicode 10.0.0 Support: Character encoding, character type info, and
67 transliteration tables are all updated to Unicode 10.0.0, using
68 generator scripts contributed by Mike FABIAN (Red Hat).
69 These updates cause user visible changes, especially the changes in
70 wcwidth for many emoji characters cause problems when emoji sequences
71 are rendered with pango, see for example:
72 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
74 * Collation of Hungarian has been overhauled and is now consistent with "The
75 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
78 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
80 - The GNU C Library will now detect when /etc/resolv.conf has been
81 modified and reload the changed configuration. The new resolver option
82 “no-reload” (RES_NORELOAD) disables this behavior.
84 - The GNU C Library now supports an arbitrary number of search domains
85 (configured using the “search” directive in /etc/resolv.conf);
86 previously, there was a hard limit of six domains. For backward
87 compatibility, applications that directly modify the ‘_res’ global
88 object are still limited to six search domains.
90 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
91 Library will now randomly pick a name server from the configuration as a
92 starting point. (Previously, the second name server was always used.)
94 * The tunables feature is now enabled by default. This allows users to tweak
95 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
97 * New function reallocarray, which resizes an allocated block (like realloc)
98 to the product of two sizes, with a guaranteed clean failure upon integer
99 overflow in the multiplication. Originally from OpenBSD, contributed by
100 Dennis Wölfing and Rüdiger Sonderfeld.
102 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
103 These are extended versions of preadv and pwritev, respectively, taking an
104 additional flags argument. The set of supported flags depends on the
105 running kernel; full support currently requires kernel 4.7 or later.
107 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
108 create a new session ID for the spawned process. This feature is
109 scheduled to be added to the next major revision of POSIX; for the time
110 being, it is available under _GNU_SOURCE.
112 * errno.h is now safe to use from C-preprocessed assembly language on all
113 supported operating systems. In this context, it will only define the
114 Exxxx constants, as preprocessor macros expanding to integer literals.
116 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
117 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
118 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
119 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
121 To compile programs that use this feature, the compiler must support
122 128-bit floating point with the type name _Float128 (as defined by TS
123 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
124 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
125 must be defined to make the new interfaces visible.
127 The new functions and macros correspond to those present for other
128 floating-point types (except for a few obsolescent interfaces not
129 supported for the new type), with F128 or f128 suffixes; for example,
130 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
131 printf or scanf formats for the new type; the strfromf128 and strtof128
132 interfaces should be used instead.
134 Deprecated and removed features, and other changes affecting compatibility:
136 * The synchronization that pthread_spin_unlock performs has been changed to
137 now be equivalent to a C11 atomic store with release memory order to the
138 spin lock's memory location. Previously, several (but not all)
139 architectures used stronger synchronization (e.g., containing what is
140 often called a full barrier). This change can improve performance, but
141 may affect odd fringe uses of spin locks that depend on the previous
142 behavior (e.g., using spin locks as atomic variables to try to implement
143 Dekker's mutual exclusion algorithm).
145 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
148 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
149 will only be built and installed when the GNU C Library is configured with
150 --enable-obsolete-rpc. This allows alternative RPC implementations, such
151 as TIRPC or rpcsvc-proto, to be used.
153 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
154 libnss_compat, are deprecated, and will not be built or installed by
157 The NIS(+) support library, libnsl, is also deprecated. By default, a
158 compatibility shared library will be built and installed, but not headers
159 or development libraries. Only a few NIS-related programs require this
160 library. (In particular, the GNU C Library has never required programs
161 that use 'gethostbyname' to be linked with libnsl.)
163 Replacement implementations based on TIRPC, which additionally support
164 IPv6, are available from <https://github.com/thkukuk/>. The configure
165 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
166 name service modules, to be built and installed.
168 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
169 support is enabled, the configured recursive resolver must support EDNS.
170 (Responding to EDNS-enabled queries with responses which are not
171 EDNS-enabled is fine, but FORMERR responses are not.)
173 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
174 servers have not supported this opcode for a long time.
176 * The _res_opcodes variable has been removed from libresolv. It had been
177 exported by accident.
179 * <string.h> no longer includes inline versions of any string functions,
180 as this kind of optimization is better done by the compiler. The macros
181 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
183 * The nonstandard header <xlocale.h> has been removed. Most programs should
184 use <locale.h> instead. If you have a specific need for the definition of
185 locale_t with no other declarations, please contact
186 libc-alpha@sourceware.org and explain.
188 * The obsolete header <sys/ultrasound.h> has been removed.
190 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
192 * The obsolete function cfree has been removed. Applications should use
195 * The stack_t type no longer has the name struct sigaltstack. This changes
196 the C++ name mangling for interfaces involving this type.
198 * The ucontext_t type no longer has the name struct ucontext. This changes
199 the C++ name mangling for interfaces involving this type.
201 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
202 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
203 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
204 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
205 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
206 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
207 fpregset_t type no longer has the name struct fpu. This changes the C++
208 name mangling for interfaces involving those types.
210 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
211 synced with the kernel:
213 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
214 are not supported on this architecture and have been removed.
216 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
217 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
218 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
220 Programs that assume the GET/SETREGS ptrace requests are universally
221 available will now fail to build, instead of malfunctioning at runtime.
223 Changes to build and runtime requirements:
225 * Linux kernel 3.2 or later is required at runtime, on all architectures
226 supported by that kernel. (This is a change from version 2.25 only for
229 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
231 * On most architectures, GCC 4.9 or later is required to build the GNU C
232 Library. On powerpc64le, GCC 6.2 or later is required.
234 Older GCC versions and non-GNU compilers are still supported when
235 compiling programs that use the GNU C Library. (We do not know exactly
236 how old, and some GNU extensions to C may be _de facto_ required. If you
237 are interested in helping us make this statement less vague, please
238 contact libc-alpha@sourceware.org.)
240 Security related changes:
242 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
243 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
245 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
246 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
248 * Avoid printing a backtrace from the __stack_chk_fail function since it is
249 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
252 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
253 fixed (CVE-2017-12133).
255 The following bugs are resolved with this release:
257 [984] network: Respond to changed resolv.conf in gethostbyname
258 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
259 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
261 [12189] libc: __stack_chk_fail should not attempt a backtrace
263 [14096] time: Race condition on timezone/tst-timezone.out
264 [14172] localedata: az_IR: new locale
265 [14995] build: glibc fails to build if gold is the default linker, even if
267 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
268 [16637] network: inet_pton function is accepting IPv6 with bad format
269 [16640] string: string/strtok.c: undefined behaviour inconsistent between
270 x86 and other generic code
271 [16875] localedata: ko_KR: fix lang_name
272 [17225] localedata: ar_SY: localized month names for May and June are
274 [17297] localedata: da_DK: wrong date_fmt string
275 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
276 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
277 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
278 [19066] localedata: ar_SA abbreviated day and month names are in English
279 [19569] network: resolv: Support an arbitrary number of search domains
280 [19570] network: Implement random DNS server selection in the stub
282 [19838] locale: localedef fails on PA-RISC
283 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
284 order of 0D36 and 0D37
285 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
287 [20098] libc: FAIL: debug/backtrace-tst on hppa
288 [20257] network: sunrpc: clntudp_call does not enforce timeout when
290 [20275] localedata: locale day/abday/mon/abmon should not have trailing
292 [20313] localedata: Update locale data to Unicode 9.0
293 [20424] manual: Document how to provide a malloc replacement
294 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
296 [20686] locale: Add el_GR@euro to SUPPORTED.
297 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
298 failures consistently
299 [21015] dynamic-link: Document and fix --enable-bind-now
300 [21016] nptl: pthread_cond support is broken on hppa
301 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
302 frame-pointer on i386
303 [21049] libc: segfault in longjmp_chk() due to clobbered processor
305 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
306 [21088] libc: Build fails with --enable-static-nss
307 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
308 generic c code is used
309 [21109] libc: Tunables broken on big-endian
310 [21112] math: powf has large ulp errors with base close to 1 and exponent
312 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
314 [21120] malloc: glibc malloc is incompatible with GCC 7
315 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
317 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
318 (-0.0) when linking with -lieee
319 [21171] math: log10, log2 and lgamma return incorrect results
320 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
321 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
322 [21207] localedata: ce_RU: update weekdays from CLDR
323 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
324 [21217] localedata: Update months from CLDR-31
325 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
326 [21243] libc: support_delete_temp_file should issue warning for failed
328 [21244] libc: support resolv_test_start() socket fd close should be
330 [21253] libc: localedef randomly segfaults when using -fstack-check due to
331 new posix_spawn implementation
332 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
333 leads to lower CPU frequency
334 [21259] libc: [alpha] termios.h missing IXANY for POSIX
335 [21261] libc: [sparc64] bits/setjmp.h namespace
336 [21267] network: [mips] bits/socket.h IOC* namespace
337 [21268] libc: [alpha] termios.h NL2, NL3 namespace
338 [21270] libc: mmap64 silently truncates large offset values
339 [21275] libc: posix_spawn always crashes on ia64 now
340 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
341 [21280] math: [powerpc] logbl for POWER7 return incorrect results
342 [21289] libc: Incorrect declaration for 32-bit platforms with
343 _FILE_OFFSET_BITS=64 causes build error
344 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
345 not support gethostbyname4_r
346 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
348 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
350 [21340] libc: Support POSIX_SPAWN_SETSID
351 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
352 [21359] network: ns_name_pack needs additional byte in destination buffer
353 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
354 against fragmentation attacks (CVE-2017-12132)
355 [21369] network: resolv: Remove EDNS fallback
356 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
358 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
359 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
360 [21393] stdio: Missing dup3 error check in freopen, freopen64
361 [21396] libc: Use AVX2 memcpy/memset on Skylake server
362 [21399] localedata: Bad description for U00EC in
363 localedata/charmaps/CP1254
364 [21411] malloc: realloc documentation error
365 [21426] network: sys/socket.h uio.h namespace
366 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
367 [21445] libc: signal.h bsd_signal namespace
368 [21455] network: Network headers stdint.h namespace
369 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
371 [21475] network: resolv: Overlong search path is truncated mid-label
372 [21511] libc: sigstack namespace
373 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
374 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
375 bits/syscall.h: No such file or directory
376 [21517] libc: struct sigaltstack namespace
377 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
378 [21533] localedata: Update locale data to Unicode 10.0
380 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
381 junk at end of line, first unrecognized character is `@'
382 [21538] libc: SIG_HOLD missing for XPG4
383 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
384 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
385 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
386 [21543] libc: sigevent namespace
387 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
388 size for general purpose registers in mcontext_t structure
389 [21550] libc: sigwait namespace
390 [21552] libc: XPG4 bsd_signal namespace
391 [21554] libc: sigpause namespace
392 [21560] libc: sys/wait.h signal.h namespace
393 [21561] libc: waitid namespace
394 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
395 '/usr/include/stdlib.h': Permission denied
396 [21575] libc: sys/wait.h missing struct rusage definition
397 [21584] libc: sigaltstack etc namespace
398 [21597] libc: siginterrupt namespace
399 [21607] math: hppa: FAIL: math/test-tgmath
400 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
402 [21622] libc: [tile] missing SA_* for POSIX.1:2008
403 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
404 stack and heap (CVE-2017-1000366)
405 [21625] libc: wait3 namespace
406 [21654] nss: Incorrect pointer alignment in NSS group merge result
408 [21657] network: Parse interface zone id for node-local multicast
409 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
411 [21666] libc: .symver is used on common symbol
412 [21668] network: resolv: res_init cross-thread broadcast introduces race
414 [21687] math: tgmath.h totalorder, totalordermag return type
415 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
417 [21696] libc: Incorrect assumption of of __cpu_mask in
418 posix/sched_cpucount.c
419 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
420 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
422 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
423 leading to relocation crash
424 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
425 `resolv_conf_matches (resp, conf)' failed.
426 [21710] localedata: Added Samoan language locale for Samoa
427 [21711] localedata: Pashto yesstr/nostr locale are missing
428 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
429 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
430 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
431 [21723] localedata: yesstr/nostr missing for Chinese language locale
432 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
434 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
436 [21728] localedata: New Locale for Tongan language
437 [21729] localedata: incorrect LC_NAME fields for hi_IN
438 [21733] localedata: yesstr and nostr are missing for zh_HK
439 [21734] localedata: Missing yesstr and nostr are for kw_GB
440 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
441 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
442 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
443 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
445 [21744] libc: Tests failing on --enable-tunables --enable-stack-
447 [21749] localedata: Wrong abbreviated day name (“abday”) for
449 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
450 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
451 [21759] localedata: missing yesstr and nostr for Tigrinya
452 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
453 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
454 [21767] localedata: Missing Bislama locales
455 [21768] localedata: Missing yesstr and nostr for aa_ET
456 [21770] localedata: Missing Field in li_NL
457 [21778] nptl: Robust mutex may deadlock
458 [21779] libc: MicroBlaze segfaults when loading libpthread
459 [21783] localedata: Fix int_select international_call_prefixes
460 [21784] localedata: Inconsistency in country_isbn
461 [21788] localedata: Missing Country Postal Abbreviations
462 [21794] localedata: Added-country_isbn-for-Italy
463 [21795] localedata: Add/Fix country_isbn for France
464 [21796] localedata: Added country_isbn for Republic of Korea
465 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
466 [21799] localedata: Added int_select international_call_prefixes
467 [21801] localedata: Added int_select international_call_prefixes
468 [21804] nptl: Double semicolon in thread-shared-types.h
469 [21807] localedata: LC_ADDRESS fix for pap_CW
470 [21808] localedata: Fix LC_ADDRESS for pap_AW
471 [21821] localedata: Added country_name in mai_IN
472 [21822] localedata: Fix LC_TIME for mai_IN
473 [21823] localedata: missing yesstr, nostr for sa_IN
474 [21825] localedata: Fix name_mrs for mag_IN
475 [21828] localedata: 2.26 changelog should mention user visible changes
477 [21835] localedata: Added Maithili language locale for Nepal
478 [21838] localedata: Removed redundant data for the_NP
479 [21839] localedata: Fix LC_MONETARY for ta_LK
480 [21844] localedata: Fix Latin characters and Months Sequence.
481 [21848] localedata: Fix mai_NP Title Name
486 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
487 24731-2:2010, is supported to enable declarations of functions from that
488 TR. Note that not all functions from that TR are supported by the GNU C
491 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
492 18661-1:2014, is supported to enable declarations of functions and macros
493 from that TS. Note that not all features from that TS are supported by
496 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
497 18661-4:2015, is supported to enable declarations of functions and macros
498 from that TS. Note that most features from that TS are not supported by
501 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
502 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
503 Since the GNU C Library defaults to a much newer revision of POSIX, this
504 will only affect programs that specifically request an old conformance
505 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
506 see a change in the visible declarations, but a program compiled with
507 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
510 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
511 defined by all multithreaded code, but glibc has not required this for
514 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
515 means that in a future release, the macros “major”, “minor”, and “makedev”
516 will only be available from <sys/sysmacros.h>.
518 These macros are not part of POSIX nor XSI, and their names frequently
519 collide with user code; see for instance glibc bug 19239 and Red Hat bug
520 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
521 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
524 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
525 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
526 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
528 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
529 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
530 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
531 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
532 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
533 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
534 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
535 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
536 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
537 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
538 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
541 * New <math.h> features are added from TS 18661-1:2014:
543 - Signaling NaN macros: SNANF, SNAN, SNANL.
545 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
546 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
547 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
549 - llogb functions: the llogb, llogbf and llogbl functions, and the
550 FP_LLOGB0 and FP_LLOGBNAN macros.
552 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
555 - Comparison macros: iseqsig.
557 - Classification macros: iscanonical, issubnormal, iszero.
559 - Total order functions: totalorder, totalorderf, totalorderl,
560 totalordermag, totalordermagf, totalordermagl.
562 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
564 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
565 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
567 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
568 are added to libc. They convert a floating-point number into string.
570 * Most of glibc can now be built with the stack smashing protector enabled.
571 It is recommended to build glibc with --enable-stack-protector=strong.
572 Implemented by Nick Alcock (Oracle).
574 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
575 intended to be used instead of memset() to erase sensitive data after use;
576 the compiler will not optimize out calls to explicit_bzero even if they
577 are "unnecessary" (in the sense that no _correct_ program can observe the
578 effects of the memory clear).
580 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
581 to float instead of double. This does not affect the ABI of any libraries
582 that are part of the GNU C Library, but may affect the ABI of other
583 libraries that use this type in their interfaces.
585 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
586 float_t and double_t types are now defined to long double instead of float
587 and double. These options are not the default, and this does not affect
588 the ABI of any libraries that are part of the GNU C Library, but it may
589 affect the ABI of other libraries that use this type in their interfaces,
590 if they are compiled or used with those options.
592 * The getentropy and getrandom functions, and the <sys/random.h> header file
595 * The buffer size for byte-oriented stdio streams is now limited to 8192
596 bytes by default. Previously, on Linux, the default buffer size on most
597 file systems was 4096 bytes (and thus remains unchanged), except on
598 network file systems, where the buffer size was unpredictable and could be
599 as large as several megabytes.
601 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
602 for the Linux quota interface which predates kernel version 2.4.22 has
605 * The malloc_get_state and malloc_set_state functions have been removed.
606 Already-existing binaries that dynamically link to these functions will
607 get a hidden implementation in which malloc_get_state is a stub. As far
608 as we know, these functions are used only by GNU Emacs and this change
609 will not adversely affect already-built Emacs executables. Any undumped
610 Emacs executables, which normally exist only during an Emacs build, should
611 be rebuilt by re-running “./configure; make” in the Emacs build tree.
613 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
614 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
615 “no-ip6-dotint” had already been the default, and support for the
616 “ip6-dotint” option was removed from the Internet in 2006.
618 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
619 flag from <resolv.h> have been removed. The option relied on a
620 backwards-incompatible DNS extension which was never deployed on the
623 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
624 RES_BLAST defined in the <resolv.h> header file have been deprecated.
625 They were already unimplemented.
627 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
628 _res.flags are deprecated. The flag was standardized in RFC 2133, but
629 removed again from the IETF name lookup interface specification in RFC
630 2553. Applications should use getaddrinfo instead.
632 * DNSSEC-related declarations and definitions have been removed from the
633 <arpa/nameser.h> header file, and libresolv will no longer attempt to
634 decode the data part of DNSSEC record types. Previous versions of glibc
635 only implemented minimal support for the previous version of DNSSEC, which
636 is incompatible with the currently deployed version.
638 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
639 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
640 <arpa/nameser.h> header file because the distinction between RR types and
641 meta-RR types is not officially standardized, subject to revision, and
642 thus not suitable for encoding in a macro.
644 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
645 and rhook members of the res_state type in <resolv.h> have been removed.
646 The glibc stub resolver did not support these hooks, but the header file
647 did not reflect that.
649 * For multi-arch support it is recommended to use a GCC which has
650 been built with support for GNU indirect functions. This ensures
651 that correct debugging information is generated for functions
652 selected by IFUNC resolvers. This support can either be enabled by
653 configuring GCC with '--enable-gnu-indirect-function', or by
654 enabling it by default by setting 'default_gnu_indirect_function'
655 variable for a particular architecture in the GCC source file
658 * GDB pretty printers have been added for mutex and condition variable
659 structures in POSIX Threads. When installed and loaded in gdb these pretty
660 printers show various pthread variables in human-readable form when read
661 using the 'print' or 'display' commands in gdb.
663 * Tunables feature added to allow tweaking of the runtime for an application
664 program. This feature can be enabled with the '--enable-tunables' configure
665 flag. The GNU C Library manual has details on usage and README.tunables has
666 instructions on adding new tunables to the library.
668 * A new version of condition variables functions have been implemented in
669 the NPTL implementation of POSIX Threads to provide stronger ordering
672 * A new version of pthread_rwlock functions have been implemented to use a more
673 scalable algorithm primarily through not using a critical section anymore to
676 Security related changes:
678 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
679 have been created with makecontext could fail to terminate due to a
680 missing .cantunwind annotation. This has been observed to lead to a hang
681 (denial of service) in some Go applications compiled with gccgo. Reported
682 by Andreas Schwab. (CVE-2016-6323)
684 * The DNS stub resolver functions would crash due to a NULL pointer
685 dereference when processing a query with a valid DNS question type which
686 was used internally in the implementation. The stub resolver now uses a
687 question type which is outside the range of valid question type values.
690 The following bugs are resolved with this release:
692 [4099] stdio: Overly agressive caching by stream i/o functions.
693 [7065] build: Support building glibc with -fstack-protector or -fstack-
695 [9842] localedata: en_CA: incorrect date format
696 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
697 before it started waiting
698 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
699 [15765] nptl: sem_open is wrongly a cancellation point
700 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
701 [16458] libc: endian.h and netinet/in.h byte order macros return values of
703 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
704 library linked with pthread
705 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
707 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
708 [17252] libc: getrandom and getentropy syscall
709 [17730] malloc: thread-local storage is sometimes improperly free()'d
710 after being __libc_memalign()'d
711 [18241] stdio: failed fseek on memstream does not set errno and can fail
713 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
715 [18463] nptl: pthread_cond_broadcast issue when surrounded by
716 PTHREAD_PRIO_INHERIT mutex on ARM
717 [18784] network: res_query and related function crash for special record
718 type queries (CVE-2015-5180)
719 [19380] math: strtod does not raise "inexact"
720 [19387] string: Integer overflow in memchr
721 [19390] string: Integer overflow in strncat
722 [19398] build: linknamespace tests fail with massively parallel build
723 [19402] nptl: Deadlock with robust shared mutex and asynchronous
725 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
726 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
727 [19514] libc: [PATCH] Fix spelling errors in spelling
728 "implement"/"implementation" in several places
729 [19582] network: Deprecate RES_USE_INET6
730 [19673] manual: clog10 docs appear to be erroneous
731 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
732 causes a segmentation fault
733 [19826] libc: invalid pointer returned from __tls_get_addr with static
735 [20016] network: resolv: Remove hooks support from the API
736 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
738 [20033] math: [x86_64] vectorized math function don't call the __finite
740 [20116] nptl: use after free in pthread_create
741 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
742 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
744 [20311] nptl: please install proc_service.h
745 [20366] libc: Compilation errors in installed headers in strict-compliance
747 [20370] malloc: malloc: Arena free list management is still racy
748 (incorrect fix in bug 19243)
749 [20386] libc: assert (X = 0) does not result in GCC warning
750 [20432] malloc: malloc: Minimize interface required for interposition
751 [20435] libc: Missing unwind info in __startcontext causes infinite loop
752 in _Unwind_Backtrace (CVE-2016-6323)
753 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
754 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
756 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
757 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
758 [20477] network: resolv: incorrect double-checked locking related to
760 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
762 [20495] math: x86_64 performance degradation due to AVX/SSE transition
764 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
765 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
766 cause transition penalty
767 [20517] math: sparcv9 missing fdiml compat symbol
768 [20524] manual: strverscmp is inconsistent
769 [20525] libc: <sys/quota.h> should be based on kernel headers
770 [20539] math: GCC 7 -static -lm fails to link at -O3
771 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
773 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
775 [20591] network: Remove obsolete DNSSEC support
776 [20592] network: DNS resource record type classification macros in
777 <arpa/nameser.h> are incorrect
778 [20593] network: Update DNS RR type definitions
779 [20611] network: getaddrinfo accepts invalid numeric scope IDs
780 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
782 [20629] network: libresolv: Remove support for bitstring labels
784 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
785 assertion: "maxidx >= 2"
786 [20660] math: [arm] Use VSQRT
787 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
788 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
789 [20689] libc: Test for FMA should also check for AVX.
790 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
791 [20715] math: iszero macro breaks existing code
792 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
793 [20728] libc: powerpc: Missing TOC stub in clone
794 [20729] build: glibc-2.24 fails to build for i486 with -Os
795 [20750] build: Build fails with default PIE enabled gcc-6.x
796 [20768] math: [alpha] sqrt fegetenv namespace
797 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
799 [20787] math: float_t is defined as float incorrectly on x86_64 even if
800 __FLT_EVAL_METHOD__ is 2
801 [20790] network: rpcgen buffer overrun in get_prog_declaration
802 [20822] nptl: race condition in __lll_unlock_elision on powerpc
803 [20829] libc: crypt snprintf namespace
804 [20847] libc: tst-vfork3 failure
805 [20855] math: Default bits/mathdef.h has inappropriate float_t
806 [20859] math: [sh4] FP_ILOGB0 invalid
807 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
808 U+20AC), not same as GBK
809 [20915] dynamic-link: global-dynamic TLS broken on aarch64
810 [20916] math: pow handling of sNaN arguments
811 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
812 [20919] math: Bad pow (qNaN, 0) result with -lieee
813 [20940] math: hypot sNaN handling
814 [20947] math: fmax, fmin sNaN handling
815 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
816 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
818 [20971] string: powerpc64/power7 memchr overflows internal pointer check
819 [20973] nptl: robust mutexes: Lost wake-ups
820 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
821 [20978] nis: strlen on null pointer in nss_nisplus
822 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
824 [21014] string: i686 memchr overflows internal pointer check
825 [21019] libc: [mips] n32 lseek incorrect on overflow
826 [21022] libc: [microblaze] __backtrace get_frame_size namespace
827 [21026] libc: [MIPS] readahead syscall is broken on n64
828 [21028] math: Fallback fesetexceptflag should always succeed
829 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
830 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
832 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
833 [21061] librt: [microblaze] librt lost clock_* exports
834 [21073] libc: tunables: insecure environment variables passed to
835 subprocesses with AT_SECURE
836 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
840 * The minimum Linux kernel version that this version of the GNU C Library
841 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
842 kernel version 2.6.32 or later suffices (on architectures that already
843 required kernel versions more recent than 3.2, those requirements remain
844 unchanged). Linux 3.2 or later kernel headers are required on all
847 * The pap_AN locale has been deleted. This has been deprecated for a long
848 time. It has been replaced by pap_AW & pap_CW, both of which have long
849 been included in previous releases.
851 * The readdir_r and readdir64_r functions have been deprecated. It is
852 recommended to use readdir and readdir64 instead.
854 * The type “union wait” has been removed. It was deprecated in the early
855 1990s and never part of POSIX. Application code should use the int type
856 instead of “union wait”.
858 * A new NSS action is added to facilitate large distributed system
859 administration. The action, MERGE, allows remote user stores like LDAP
860 to be merged into local user stores like /etc/groups in order to provide
861 easy to use, updated, and managed sets of merged credentials. The new
862 action can be used by configuring it in /etc/nsswitch.conf:
863 group: files [SUCCESS=merge] nis
864 Implemented by Stephen Gallagher (Red Hat).
866 * The deprecated __malloc_initialize_hook variable has been removed from the
869 * The long unused localedef --old-style option has been removed. It hasn't
870 done anything in over 16 years. Scripts using this option can safely
873 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
874 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
875 return the next representable value in the direction of positive infinity
876 and the nextdown functions return the next representable value in the
877 direction of negative infinity. These are currently enabled as GNU
880 Security related changes:
882 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
883 could result in a stack overflow when getnetbyname was called with an
884 overly long name. (CVE-2016-3075)
886 * Previously, getaddrinfo copied large amounts of address data to the stack,
887 even after the fix for CVE-2013-4458 has been applied, potentially
888 resulting in a stack overflow. getaddrinfo now uses a heap allocation
889 instead. Reported by Michael Petlan. (CVE-2016-3706)
891 * The glob function suffered from a stack-based buffer overflow when it was
892 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
893 Reported by Alexander Cherepanov. (CVE-2016-1234)
895 * The Sun RPC UDP client could exhaust all available stack space when
896 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
897 alloca plugin for GCC. (CVE-2016-4429)
899 * The IPv6 name server management code in libresolv could result in a memory
900 leak for each thread which is created, performs a failing naming lookup,
901 and exits. Over time, this could result in a denial of service due to
902 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
904 The following bugs are resolved with this release:
906 [1170] localedata: ne_NP: update Nepali locale definition file
907 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
909 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
910 [6796] math: fdim() does not set errno on overflow
911 [10354] libc: posix_spawn should use vfork() in more cases than presently
912 [11213] localedata: localedata: add copyright disclaimer to locale files
913 [12143] localedata: chr_US: new Cherokee locale
914 [12450] localedata: sgs_LT: new locale
915 [12676] localedata: ln_CD: new locale
916 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
918 [13304] math: fma, fmaf, fmal produce wrong results
919 [14259] build: --localedir arg to configure is ignored
920 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
921 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
922 [14934] localedata: es_CL: wrong first weekday chilean locale
923 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
925 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
927 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
928 [15368] nptl: raise() is not async-signal-safe
929 [15479] math: ceil, floor, round and trunc raise inexact exception
930 [15578] localedata: kk_KZ: various updates
931 [16003] localedata: pap_AN: punt old locale
932 [16137] localedata: iw_IL: punt old locale
933 [16190] localedata: eo: new esperanto locale
934 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
935 [16742] malloc: race condition: pthread_atfork() called before first
936 malloc() results in unexpected locking behaviour/deadlocks
937 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
939 [16983] localedata: postal_fmt does not allow %l and %n modifiers
940 [17565] localedata: pt_PT: wrong (work-)week start
941 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
942 [17950] build: Build fails with -msse
943 [18205] localedata: be_BY*: wrong first_weekday and first_workday
944 [18433] libc: posix_spawn does not return correctly upon failure to
946 [18453] localedata: charmaps/IBM875: incorrect codes
947 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
949 [18896] localedata: he_IL: improvements for currency
950 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
952 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
954 [19056] libc: Deprecate readdir_r
955 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
957 [19198] localedata: nl_NL: small improvements for Dutch locales
958 [19257] network: Per-thread memory leak in __res_vinit with IPv6
959 nameservers (CVE-2016-5417)
960 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
962 [19400] locale: Language missing in "iso-639.def", trivial fix in
964 [19431] malloc: Deadlock between fflush, getdelim, and fork
965 [19505] libc: Incorrect file descriptor validity checks in
966 posix_spawn_file_actions_add{open,close,dup2}
967 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
969 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
970 `intl/{gettextP,loadinfo}.h'
971 [19534] libc: execle, execlp may use malloc
972 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
973 [19573] network: res_nclose and __res_maybe_init disagree about name
974 server initialization, breaking Hesiod
975 [19575] localedata: Status of GB18030 tables
976 [19581] localedata: sr_* date_fmt string contains additional newline
977 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
979 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
980 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
981 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
982 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
984 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
986 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
988 [19610] dynamic-link: ldconfig -X removes stale symbolic links
989 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
990 [19633] locale: strfmon_l applies global locale to number formatting
991 [19642] network: Memory leak in getnameinfo
992 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
993 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
995 [19654] math: [x86_64] Need testcase for BZ #19590 fix
996 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
998 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
999 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1001 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1003 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1005 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1006 pointers and lengths in error-case.
1007 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1008 report errors on UTF-16 surrogates values.
1009 [19755] nscd: nscd assertion failure in gc
1010 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1011 [19759] libc: mempcpy shouldn't be inlined
1012 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1013 [19765] libc: s390 needs an optimized mempcpy
1014 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1015 NAME_MAX limit assumption (CVE-2016-1234)
1016 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1017 [19787] network: Missing and incorrect truncation checks in getnameinfo
1018 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1020 [19791] network: Assertion failure in res_query.c with un-connectable name
1022 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1023 [19822] math: libm.so install clobbers old version
1024 [19825] network: resolv: send_vc can return uninitialized data in second
1025 response to getaddrinfo
1026 [19830] network: nss_dns: should check RDATA length against buffer length
1027 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1028 confronted with A/AAAA records of invalid size
1029 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1030 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1031 [19853] stdio: Printing IBM long double in decimal with high precision is
1033 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1034 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1035 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1037 [19865] network: Assertion failure or memory leak in
1038 _nss_dns_getcanonname_r
1039 [19868] network: nss_dns: netent code does not skip over non-PTR records
1040 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1042 [19881] string: Improve x86-64 memset
1043 [19907] string: Incorrect memcpy tests
1044 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1046 [19925] libc: termios.h XCASE namespace
1047 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1048 [19929] libc: limits.h NL_NMAX namespace
1049 [19931] stdio: Memory leak in vfprintf
1050 [19957] libc: clone(CLONE_VM) access invalid parent memory
1051 [19963] localedata: en_IL: New locale
1052 [19989] stdio: stdio.h cuserid namespace
1053 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1055 [19996] locale: langinfo.h nl_langinfo_l namespace
1056 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1057 [20010] network: getaddrinfo: Stack overflow in hostent translation
1059 [20012] stdio: libio: fmemopen append mode failure
1060 [20014] stdio: stdio.h namespace for pre-threads POSIX
1061 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1062 [20023] libc: fcntl.h timespec namespace
1063 [20024] math: [x86_64] vectorized sincos trashes the stack
1064 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1065 [20041] time: sys/time.h timespec namespace
1066 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1067 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1068 [20051] libc: ttyslot in wrong header under wrong conditions
1069 [20054] libc: gethostname not declared for XPG4
1070 [20055] libc: termios.h missing tcgetsid for XPG4
1071 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1073 [20073] libc: sys/stat.h fchmod namespace
1074 [20074] libc: stdlib.h rand_r namespace
1075 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1076 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1078 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1079 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1081 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1082 [20119] libc: Wrong mask for processors level type from CPUID
1083 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1084 AS not supporting AVX512
1085 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1087 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1088 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1090 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1091 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1092 [20178] libc: posix_spawn{p} should not call exit
1093 [20191] stdio: libio: vtables hardening
1094 [20195] string: FMA4 detection requires CPUID execution with register
1096 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1097 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1099 [20212] math: acos (sNaN) returns sNaN
1100 [20213] math: asin (sNaN) returns sNaN
1101 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1102 [20218] math: [i386] asinhl (sNaN) returns sNaN
1103 [20219] math: [i386] atanhl (sNaN) returns sNaN
1104 [20222] stdio: fopencookie: Mangle function pointers
1105 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1106 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1107 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1109 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1110 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1111 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1112 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1113 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1114 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1115 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1116 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1117 [20240] math: modf (sNaN) returns sNaN
1118 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1119 [20250] math: frexp (sNaN) returns sNaN
1120 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1121 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1123 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1124 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1125 subscript is below array bounds [-Werror=array-bounds]
1126 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1127 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1129 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1130 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1132 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1133 "invalid" exceptions
1134 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1135 [20316] localedata: id_ID: Februari instead of Pebruari
1136 [20327] string: POWER8 strcasecmp returns incorrect result
1137 [20347] math: Failure: Test: j0_downward (0xap+0)
1138 [20348] libc: FAIL: misc/tst-preadvwritev64
1139 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1140 [20350] libc: There is no test for p{read,write}64
1141 [20357] math: Incorrect cos result for 1.5174239687223976
1142 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1146 * Unicode 8.0.0 Support: Character encoding, character type info, and
1147 transliteration tables are all updated to Unicode 8.0.0, using new
1148 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1149 These updates cause user visible changes, such as the fixes for bugs
1150 89, 16061, and 18568.
1152 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1153 kernel-internal CPU set size. This means that requests that change the
1154 CPU affinity which failed before (for example, an all-ones CPU mask) will
1155 now succeed. Applications that need to determine the effective CPU
1156 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1157 setting it because the kernel can adjust it (and the previous size check
1158 would not detect this in the majority of cases).
1160 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1161 the following new symbols are used: fts64_children, fts64_close,
1162 fts64_open, fts64_read and fts64_set.
1164 * getaddrinfo now detects certain invalid responses on an internal netlink
1165 socket. If such responses are received, an affected process will
1166 terminate with an error message of "Unexpected error <number> on netlink
1167 descriptor <number>" or "Unexpected netlink response of size <number> on
1168 descriptor <number>". The most likely cause for these errors is a
1169 multi-threaded application which erroneously closes and reuses the netlink
1170 file descriptor while it is used by getaddrinfo.
1172 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1173 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1174 unnecessary serialization of memory allocation requests across threads.
1175 The defect is now corrected. Users should see a substantial increase in
1176 the concurent throughput of allocation requests for applications which
1177 trigger this bug. Affected applications typically create and
1178 destroy threads frequently. (Bug 19048 was reported and analyzed by
1181 * There is now a --disable-timezone-tools configure option for disabling the
1182 building and installing of the timezone related utilities (zic, zdump, and
1183 tzselect). This is useful for people who build the timezone data and code
1184 independent of the GNU C Library.
1186 * The obsolete header <regexp.h> has been removed. Programs that require
1187 this header must be updated to use <regex.h> instead.
1189 * The obsolete functions bdflush, create_module, get_kernel_syms,
1190 query_module and uselib are no longer available to newly linked binaries;
1191 the header <sys/kdaemon.h> has been removed. These functions and header
1192 were specific to systems using the Linux kernel and could not usefully be
1193 used with the GNU C Library on systems with version 2.6 or later of the
1196 * Optimized string, wcsmbs and memory functions for IBM z13.
1197 Implemented by Stefan Liebler.
1199 * Newly linked programs that define a variable called signgam will no longer
1200 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1201 require signgam to be set by those functions must ensure that they use the
1202 variable provided by the GNU C Library and declared in <math.h>, without
1203 defining their own copy.
1205 * The minimum GCC version that can be used to build this version of the GNU
1206 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1207 still be used to compile programs using the GNU C Library.
1209 Security related changes:
1211 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1212 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1214 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1215 the pointer guard feature. It is always enabled. Previously,
1216 LD_POINTER_GUARD could be used to disable security hardening in binaries
1217 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1220 * An integer overflow in hcreate and hcreate_r could lead to an
1221 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1223 * The catopen function no longer has unbounded stack usage. Reported by
1224 Max. (CVE-2015-8779)
1226 * The nan, nanf and nanl functions no longer have unbounded stack usage
1227 depending on the length of the string passed as an argument to the
1228 functions. Reported by Joseph Myers. (CVE-2014-9761)
1230 * A stack-based buffer overflow was found in libresolv when invoked from
1231 libnss_dns, allowing specially crafted DNS responses to seize control
1232 of execution flow in the DNS client. The buffer overflow occurs in
1233 the functions send_dg (send datagram) and send_vc (send TCP) for the
1234 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1235 family. The use of AF_UNSPEC triggers the low-level resolver code to
1236 send out two parallel queries for A and AAAA. A mismanagement of the
1237 buffers used for those queries could result in the response of a query
1238 writing beyond the alloca allocated buffer created by
1239 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1240 the overflow. Thanks to the Google Security Team and Red Hat for
1241 reporting the security impact of this issue, and Robert Holiday of
1242 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1244 The following bugs are resolved with this release:
1246 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1247 [887] math: Math library function "logb" and "nextafter" inconsistent
1248 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1249 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1250 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1251 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1253 [4404] localedata: German translation of "Alarm clock" is misleading
1254 [6799] math: nextafter() and nexttoward() doen't set errno on
1255 overflow/underflow errors
1256 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1258 [10432] nis: _nss_nis_setnetgrent assertion failure
1259 [11460] libc: fts has no LFS support
1260 [12926] network: getaddrinfo()/make_request() may spin forever
1261 [13065] nptl: Race condition in pthread barriers
1262 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1263 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1265 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1266 [14912] libc: Rename non-installed bits/*.h headers
1267 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1268 [15367] math: Let gcc use __builtin_isinf
1269 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1270 [15421] math: lgamma wrongly sets signgam for ISO C
1271 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1272 FE_INVALID with argument out of range
1273 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1275 [15786] dynamic-link: ifunc resolver functions can smash function
1277 [15918] math: Unnecessary check for equality in hypotf()
1278 [16061] localedata: Review / update transliteration data
1279 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1281 [16141] time: strptime %z offset restriction
1282 [16171] math: drem should be alias of remainder
1283 [16296] math: fegetround is pure?
1284 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1285 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1287 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1288 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1290 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1291 "Exception "Inexact" set" on ppc32
1292 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1293 [16517] math: Missing underflow exception from tanf/tan/tanl
1294 [16519] math: Missing underflow exception from sinhf
1295 [16520] math: Missing underflow exception from tanhf
1296 [16521] math: Missing underflow exception from exp2
1297 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1299 [16734] stdio: fopen calls mmap to allocate its buffer
1300 [16961] math: nan function incorrect handling of bad sequences
1301 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1302 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1303 [16985] locale: localedef: confusing error message when opening output
1305 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1306 [17197] locale: Redundant shift character in iconv conversion output at
1308 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1309 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1310 [17250] dynamic-link: static linking breaks nss loading
1311 (getaddrinfo/getpwnam/etc...)
1312 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1314 [17441] math: isnan() should use __builtin_isnan() in GCC
1315 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1316 (related to lock elision)
1317 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1318 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1319 [17887] time: strptime should be able to parse "+01:00" style timezones
1320 [17905] libc: catopen() Multiple unbounded stack allocations
1322 [18084] libc: backtrace (..., 0) dumps core on x86
1323 [18086] libc: nice() sets errno to 0 on success
1324 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1325 is too large (CVE-2015-8778)
1326 [18251] dynamic-link: SONAME missing when audit modules provides path
1327 [18265] libc: add attributes for wchar string and memory functions
1328 [18370] math: csqrt missing underflows
1329 [18421] libc: [hppa] read-only segment has dynamic relocations
1330 [18472] libc: Obsolete syscall wrappers should be compat symbols
1331 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1332 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1333 [18525] localedata: Remove locale timezone information
1334 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1335 [18568] localedata: Update locale data to Unicode 8.0
1336 [18589] locale: sort-test.sh fails at random
1337 [18595] math: ctan, ctanh missing underflows
1338 [18604] libc: assert macro-expands its argument
1339 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1340 contains a vector instruction exception.
1341 [18611] math: j1, jn missing errno setting on underflow
1342 [18618] localedata: sync Chechen locale definitions with other *_RU
1344 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1345 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1346 [18665] network: In send_dg, the recvfrom function is NOT always using the
1347 buffer size of a newly created buffer (CVE-2015-7547)
1348 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1350 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1352 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1353 [18699] math: tilegx cproj() for various complex infinities does not yield
1355 [18724] libc: Harden put*ent functions against data injection
1356 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1357 [18755] build: build errors with -DNDEBUG
1358 [18757] stdio: fmemopen fails to set errno on failure
1359 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1361 [18781] libc: openat64 lacks O_LARGEFILE
1362 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1363 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1364 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1365 [18790] math: [ldbl-128ibm] tanhl inaccurate
1366 [18795] libc: stpncpy fortification misses buffer lengths that are
1367 statically too large
1368 [18796] build: build fails for --disable-mathvec
1369 [18803] math: hypot missing underflows
1370 [18820] stdio: fmemopen may leak memory on failure
1371 [18823] math: csqrt spurious underflows
1372 [18824] math: fma spurious underflows
1373 [18825] math: pow missing underflows
1374 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1375 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1376 [18870] build: sem_open.c fails to compile with missing symbol
1378 [18872] stdio: Fix memory leak in printf_positional
1379 [18873] libc: posix_fallocate overflow check ineffective
1380 [18875] math: Excess precision leads incorrect libm
1381 [18877] libc: arm: mmap offset regression
1382 [18887] libc: memory corruption when using getmntent on blank lines
1383 [18918] localedata: hu_HU: change time to HH:MM:SS format
1384 [18921] libc: Regression: extraneous stat() and fstat() performed by
1386 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1387 binaries (CVE-2015-8777)
1388 [18951] math: tgamma missing underflows
1389 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1391 [18953] localedata: lt_LT: change currency symbol to the euro
1392 [18956] math: powf inaccuracy
1393 [18961] math: [i386] exp missing underflows
1394 [18966] math: [i386] exp10 missing underflows
1395 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1396 [18969] build: multiple string test failures due to missing locale
1398 [18970] libc: Reference of pthread_setcancelstate in libc.a
1399 [18977] math: float / long double Bessel functions not in XSI POSIX
1400 [18980] math: i386 libm functions return with excess range and precision
1401 [18981] math: i386 scalb*, ldexp return with excess range and precision
1402 [18982] stdio: va_list and vprintf
1403 [18985] time: Passing out of range data to strftime() causes a segfault
1405 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1406 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1407 [19012] locale: iconv_open leaks memory on error path
1408 [19016] math: clog, clog10 inaccuracy
1409 [19018] nptl: Mangle function pointers in tls_dtor_list
1410 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1411 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1412 [19048] malloc: malloc: arena free list can become cyclic, increasing
1414 [19049] math: [powerpc] erfc incorrect zero sign
1415 [19050] math: [powerpc] log* incorrect zero sign
1416 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1417 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1418 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1419 [19074] network: Data race in _res_hconf_reorder_addrs
1420 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1421 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1422 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1424 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1426 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1427 [19086] manual: posix_fallocate64 documented argument order is wrong.
1428 [19088] math: lround, llround missing exceptions close to overflow
1430 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1431 [19095] math: dbl-64 lrint incorrect for 64-bit long
1432 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1433 [19124] dynamic-link: ld.so failed to build with older assmebler
1434 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1435 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1436 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1437 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1438 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1439 pthread_setaffinity_np
1440 [19156] math: [ldbl-128] j0l spurious underflows
1441 [19164] nptl: tst-getcpu fails with many possible CPUs
1442 [19168] math: math/test-ildoubl and math/test-ldouble failure
1443 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1444 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1446 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1447 (FE_NOMASK_ENV) do not clear SSE exceptions
1448 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1449 _int_new_arena/reused_arena
1450 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1451 [19201] math: dbl-64 remainder incorrect sign of zero result
1452 [19205] math: bits/math-finite.h conditions do not match math.h and
1454 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1455 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1456 for C99-based standards
1457 [19212] libc: features.h not -Wundef clean
1458 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1460 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1461 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1462 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1464 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1465 "inexact" exceptions
1466 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1468 [19242] libc: strtol incorrect in Turkish locales
1469 [19243] malloc: reused_arena can pick an arena on the free list, leading
1470 to an assertion failure and reference count corruption
1471 [19253] time: tzset() ineffective when temporary TZ did not include DST
1473 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1474 [19270] math: [hppa] Shared libm missing __isnanl
1475 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1476 MAP_HUGETLB and MAP_STACK defines
1477 [19313] nptl: Wrong __cpu_mask for x32
1478 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1479 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1480 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1481 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1482 [19363] time: x32: times() return value wrongly truncates/sign extends
1484 [19367] dynamic-link: Improve branch prediction on Silvermont
1485 [19369] network: Default domain name not reset by res_ninit when "search"
1486 / "domain" entry is removed from resolv.conf
1487 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1489 [19385] localedata: bg_BG: time separator should be colon, not comma
1490 [19408] libc: linux personality syscall wrapper may erroneously return an
1491 error on 32-bit architectures
1492 [19415] libc: dladdr returns wrong names on hppa
1493 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1494 IBM905, IBM907, and IBM909
1495 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1496 [19443] build: build failures with -DDEBUG
1497 [19451] build: Make check fails on test-double-vlen2
1498 [19462] libc: Glibc failed to build with -Os
1499 [19465] math: Wrong code with -Os
1500 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1502 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1504 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1505 [19486] math: S390: Math tests fail with "Exception Inexact set".
1506 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1507 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1509 [19590] math: Fail to build shared objects that use libmvec.so functions.
1513 * The following bugs are resolved with this release:
1515 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1516 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1517 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1518 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1519 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1520 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1521 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1522 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1523 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1524 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1525 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1526 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1527 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1528 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1529 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1530 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1531 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1532 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1533 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1534 18657, 18676, 18694, 18696, 18887.
1536 * Cache information can be queried via sysconf() function on s390 e.g. with
1537 _SC_LEVEL1_ICACHE_SIZE as argument.
1539 * A buffer overflow in gethostbyname_r and related functions performing DNS
1540 requests has been fixed. If the NSS functions were called with a
1541 misaligned buffer, the buffer length change due to pointer alignment was
1542 not taken into account. This could result in application crashes or,
1543 potentially arbitrary code execution, using crafted, but syntactically
1544 valid DNS responses. (CVE-2015-1781)
1546 * The time zone file parser has been made more robust against crafted time
1547 zone files, avoiding heap buffer overflows related to the processing of
1548 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1549 large time zone data files. Overly long time zone specifiers in the TZ
1550 variable no longer result in stack overflows and crashes.
1552 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1553 for LD and GD on x86 and x86-64, has been implemented. You will need
1554 binutils-2.24 or later to enable this optimization.
1556 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1557 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1558 Hat). These updates cause user visible changes, such as the fix for bug
1561 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1562 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1563 condition in some applications.
1565 * Added vector math library named libmvec with the following vectorized x86_64
1566 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1568 The library can be disabled with --disable-mathvec. Use of the functions is
1569 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1570 Shared library libmvec.so is linked in as needed when using -lm (no need to
1571 specify -lmvec explicitly for not static builds).
1572 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1574 * A new fmemopen implementation has been added with the goal of POSIX
1575 compliance. The new implementation fixes the following long-standing
1576 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1577 old implementation is still present for use be by existing binaries.
1579 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1580 releases. It has been fixed to match 2.19 and older, but binaries built
1581 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1583 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1584 Contributed by Roland McGrath (Google).
1586 * The header <regexp.h> is deprecated, and will be removed in a future
1587 release. Use of this header will trigger a deprecation warning.
1588 Application developers should update their code to use <regex.h> instead.
1590 This header was formerly part of SUS, but was deprecated in 1994 and
1591 removed from the standard in 2001. Also, the glibc implementation
1592 leaks memory. See BZ#18681 for more details.
1596 * The following bugs are resolved with this release:
1598 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1599 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1600 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1601 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1602 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1603 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1604 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1605 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1606 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1607 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1610 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1611 too little memory for the to-be-scanned arguments and overflow the
1612 allocated buffer. The implementation now correctly computes the required
1613 buffer size when using malloc, and switches to malloc from alloca as
1616 * A new semaphore algorithm has been implemented in generic C code for all
1617 machines. Previous custom assembly implementations of semaphore were
1618 difficult to reason about or ensure that they were safe. The new version
1619 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1620 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1621 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1623 * Port to Altera Nios II has been contributed by Mentor Graphics.
1625 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1626 implementations for powerpc64/powerpc64le.
1627 Implemented by Adhemerval Zanella (IBM).
1629 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1630 and powerpc64le. This may improve lock scaling of existing programs on
1631 HTM capable systems. The lock elision code is only enabled with
1632 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1633 powerpc will issue a transaction abort on every syscall to avoid side
1634 effects being visible outside transactions.
1636 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1637 AArch64. Contributed by ARM Ltd.
1639 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1641 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1642 under certain input conditions resulting in the execution of a shell for
1643 command substitution when the application did not request it. The
1644 implementation now checks WRDE_NOCMD immediately before executing the
1645 shell and returns the error WRDE_CMDSUB as expected.
1647 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1648 processing format strings with a large number of format specifiers.
1650 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1651 infinite loop if the DNS response contained a PTR record of an unexpected
1654 * The minimum GCC version that can be used to build this version of the GNU
1655 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1656 still be used to compile programs using the GNU C Library.
1658 * The GNU C Library is now built with -Werror by default. This can be
1659 disabled by configuring with --disable-werror.
1661 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1663 * The obsolete sigvec function has been removed. This was the original
1664 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1665 programs have been using instead for about 25 years. Of course, ABI
1666 compatibility for old binaries using sigvec remains intact.
1668 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1669 with newer versions of bison.
1671 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1672 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1673 registers overlay two consecutive single-precision registers. MIPS32R2
1674 introduced a new FPU mode (FR=1) where double-precision registers extend the
1675 corresponding single-precision registers which is incompatible with the
1676 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1677 require the use of FR=1 making a transition necessary. New o32 ABI
1678 extensions enable users to migrate over time from the original o32 ABI
1679 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1680 tracks the ABI of any loaded object and verifies that new objects are
1681 compatible. Mode transitions will also be requested as required and
1682 unsupportable objects will be rejected. The ABI checks include both soft and
1683 hard float ABIs for o32, n32 and n64.
1685 GCC 5 with GNU binutils 2.25 onwards:
1686 It is strongly recommended that all o32 system libraries are built using the
1687 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1688 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1689 compiler using --with-fp-32=xx to set this by default.
1693 * The following bugs are resolved with this release:
1695 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1696 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1697 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1698 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1699 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1700 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1701 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1702 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1703 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1704 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1705 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1706 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1707 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1708 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1709 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1711 * Reverted change of ABI data structures for s390 and s390x:
1712 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1713 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1714 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1715 are preserved pointing straight to the same implementation as the old ones.
1716 Given that, new callers will simply provide a too-big buffer to these
1717 functions. Any applications/libraries out there that embed jmp_buf or
1718 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1719 against 2.19 headers will have to rebuilt again. This is necessary in any
1720 case to revert the breakage in their ABI caused by the glibc change.
1722 * Support for file description locks is added to systems running the
1723 Linux kernel. The standard file locking interfaces are extended to
1724 operate on file descriptions, not file descriptors, via the use of
1725 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1726 are associated with an open file instead of a process.
1728 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1730 * The minimum Linux kernel version that this version of the GNU C Library
1731 can be used with is 2.6.32.
1733 * Running the testsuite no longer terminates as soon as a test fails.
1734 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1735 with PASS or FAIL lines for individual tests. A summary of the results is
1736 printed, including a list of failing lists, and "make check" exits with
1737 error status if there were any unexpected failures. "make check
1738 stop-on-test-failure=y" may be used to keep the old behavior.
1740 * The am33 port, which had not worked for several years, has been removed
1743 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1744 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1745 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1746 conflicted with POSIX (support for which was removed in 2.19), the
1747 interfaces those macros enabled remain available when compiling with
1748 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1749 test macros defined.
1751 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1753 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1754 This may improve lock scaling of existing programs on TX capable systems.
1755 The lock elision code is only built with --enable-lock-elision=yes and
1756 then requires a GCC version supporting the TX builtins. With lock elision
1757 default mutexes are elided via __builtin_tbegin, if the cpu supports
1758 transactions. By default lock elision is not enabled and the elision code
1761 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1762 copy the path argument. This allowed programs to cause posix_spawn to
1763 deference a dangling pointer, or use an unexpected pathname argument if
1764 the string was modified after the posix_spawn_file_actions_addopen
1767 * All supported architectures now use the main glibc sysdeps directory
1768 instead of some being in a separate "ports" directory (which was
1769 distributed separately before glibc 2.17).
1771 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1772 On configurations that support it (all Linux configurations), it's now
1773 used regardless of the --enable-add-ons switch to configure. It is no
1774 longer possible to build such configurations without pthreads support.
1776 * Locale names, including those obtained from environment variables (LANG
1777 and the LC_* variables), are more tightly checked for proper syntax.
1778 setlocale will now fail (with EINVAL) for locale names that are overly
1779 long, contain slashes without starting with a slash, or contain ".." path
1780 components. (CVE-2014-0475) Previously, some valid locale names were
1781 silently replaced with the "C" locale when running in AT_SECURE mode
1782 (e.g., in a SUID program). This is no longer necessary because of the
1785 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1786 with application code using Intel MPX instructions. (With all previous
1787 versions, the MPX register state could be clobbered when making calls
1788 into or out of a shared library.) Note that while the new dynamic
1789 linker is compatible with all known x86 hardware whether or not it
1790 supports Intel MPX, some x86 instruction-set emulators might fail to
1791 handle the new instruction encodings. This is known to affect Valgrind
1792 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1793 release), and might affect other tools that do instruction emulation.
1795 * Support for loadable gconv transliteration modules has been removed.
1796 The support for transliteration modules has been non-functional for
1797 over a decade, and the removal is prompted by security defects. The
1798 normal gconv conversion modules are still supported. Transliteration
1799 with //TRANSLIT is still possible, and the //IGNORE specifier
1800 continues to be supported. (CVE-2014-5119)
1802 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1803 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1804 resulting a denial-of-service security vulnerability in applications which
1805 use functions related to iconv. (CVE-2014-6040)
1809 * The following bugs are resolved with this release:
1811 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1812 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1813 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1814 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1815 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1816 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1817 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1818 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1819 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1820 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1821 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1822 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1823 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1824 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1825 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1826 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1827 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1828 16431, 16453, 16474, 16506, 16510, 16529
1830 * Slovenian translations for glibc messages have been contributed by the
1831 Translation Project's Slovenian team of translators.
1833 * The public headers no longer use __unused nor __block. This change is to
1834 support compiling programs that are derived from BSD sources and use
1835 __unused internally, and to support compiling with Clang's -fblock
1836 extension which uses __block.
1838 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1839 large collation sequences to optimize multiple passes. This cache
1840 computation may overflow for large collation sequences and may cause a
1841 stack or buffer overflow. This is now fixed to use a slower algorithm
1842 which does not use a cache if there is an integer overflow.
1844 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1845 rules for large collation sequences to optimize multiple passes and falls
1846 back to alloca if malloc fails, resulting in a possible stack overflow.
1847 The implementation now falls back to an uncached collation sequence lookup
1850 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1851 initialized for static applications resulting in the security feature
1852 being disabled. The pointer guard is now correctly initialized to a
1853 random value for static applications. Existing static applications need
1854 to be recompiled to take advantage of the fix (bug 15754).
1856 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1857 to the d_name member of struct dirent, or omit the terminating NUL
1858 character. (Bugzilla #14699).
1860 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1861 aligned_alloc functions could allocate too few bytes or corrupt the
1862 heap when passed very large allocation size values (Bugzilla #15855,
1865 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1866 for AF_INET6 has been fixed (Bugzilla #16072).
1868 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1869 pap_CW, quz_PE, the_NP.
1871 * Substantially revised locales: gd_GB, ht_HT
1873 * The LC_ADDRESS field was updated to support country_car for almost all
1876 * ISO 1427 definitions were updated.
1878 * ISO 3166 definitions were updated.
1880 * The localedef utility now supports --big-endian and --little-endian
1881 command-line options to generate locales for a different system from that
1882 for which the C library was built.
1884 * Binary locale files now only depend on the endianness of the system for
1885 which they are generated and not on other properties of that system. As a
1886 consequence, binary files generated with new localedef may be incompatible
1887 with old versions of the GNU C Library, and binary files generated with
1888 old localedef may be incompatible with this version of the GNU C Library,
1889 in the following circumstances:
1891 + Locale files may be incompatible on m68k systems.
1893 + Locale archive files (but not separate files for individual locales) may
1894 be incompatible on systems where plain "char" is signed.
1896 * The configure option --disable-versioning has been removed. Builds with
1897 --disable-versioning had not worked for several years.
1899 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1900 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1902 * SystemTap probes for malloc have been introduced.
1904 * SystemTap probes for slow multiple precision fallback paths of
1905 transcendental functions have been introduced.
1907 * Support for powerpc64le has been added.
1909 * The soft-float powerpc port now supports e500 processors.
1911 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1913 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1914 set of header declarations that are enabled by default, even when other
1915 feature test macros or compiler options such as -std=c99 would otherwise
1916 disable some of those declarations.
1918 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1919 conflict with POSIX. The libbsd-compat library (which was a dummy library
1920 that did nothing) has also been removed.
1922 * Preliminary documentation about Multi-Thread, Async-Signal and
1923 Async-Cancel Safety has been added.
1925 * Change of ABI data structures for s390 and s390x:
1926 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1927 allow for future hardware extensions. All library functions that accept or
1928 return these structures were versioned in order to provide backwards
1929 compatibility with existing code. However, not all cases can be handled
1930 with interface versioning. If an application embeds either structure into
1931 another structure and that structure is passed to another compilation unit
1932 which expects the newer larger-sized object then any access to the new
1933 fields will result in undefined behaviour. Similarly any access to fields
1934 that were moved by the enlarging of the structures will cause undefined
1935 behaviour. To fix the undefined behaviour all objects that are part of the
1936 final in-memory image must agree on the size of structures, and this may
1937 require recompilation.
1941 * The following bugs are resolved with this release:
1943 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1944 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1945 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1946 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1947 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1948 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1949 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1950 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1951 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1952 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1953 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1954 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1955 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1958 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1959 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1960 Distributions can re-enable building and using pt_chown via the new configure
1961 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1962 considerable security risks and should only be used if the distribution
1963 understands and accepts the risks.
1965 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1968 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1969 fixed (Bugzilla #15330).
1971 * Add support for calling C++11 thread_local object destructors on thread
1972 and program exit. This needs compiler support for offloading C++11
1973 destructor calls to glibc.
1975 * Improved worst case performance of libm functions with double inputs and
1978 * Support for priority inherited mutexes in pthread condition variables on
1979 non-x86 architectures.
1981 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1983 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1985 * Optimized string functions for ARM. Implemented by Will Newton and
1988 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1990 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1993 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1994 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1996 * Added a benchmark framework to track performance of functions in glibc.
1998 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1999 It is based on draft TS 18661 and currently enabled as a GNU extension.
2001 * On Linux, the clock function now uses the clock_gettime system call
2002 for improved precision, rather than old times system call.
2004 * Added support for version-3 tz format files. This is needed when using
2005 the tz database release 2013e or later, and affects a few unusual cases --
2006 currently only TZ='America/Godthab' for time stamps after 2037.
2008 * Added new API functions pthread_getattr_default_np and
2009 pthread_setattr_default_np to get and set the default pthread
2010 attributes of a process.
2012 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2013 This may improve lock scaling of existing programs on TSX capable systems.
2014 When the --enable-lock-elision=yes parameter is specified at configure
2015 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2018 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2019 availability). Implemented by Ryan S. Arnold.
2021 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2025 * The following bugs are resolved with this release:
2027 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2028 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2029 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2030 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2031 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2032 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2033 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2034 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2035 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2036 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2037 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2038 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2039 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2040 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2041 14879, 14889, 14893, 14898, 14914.
2043 * Optimization of memcpy for MIPS.
2045 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2046 EMFILE has been fixed (Bugzilla #14889).
2048 * The add-on ports collection is now distributed in the "ports" subdirectory
2049 of the main GNU C Library distribution, rather than separately.
2051 * Port to ARM AArch64 contributed by Linaro.
2053 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2054 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2056 Implemented by Andreas Krebbel.
2058 * The new function secure_getenv allows secure access to the environment,
2059 returning NULL if running in a SUID/SGID process. This function replaces
2060 the internal function __secure_getenv.
2062 * SystemTap static probes have been added into the dynamic linker.
2063 Implemented by Gary Benson.
2065 * Optimizations of string functions strstr, strcasestr and memmem.
2066 Implemented by Maxim Kuvyrkov.
2068 * The minimum Linux kernel version that this version of the GNU C Library
2069 can be used with is 2.6.16.
2071 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2072 powerpc POWER7. Implemented by Will Schmidt.
2074 * New configure option --disable-nscd builds the C library such that it
2075 never attempts to contact the Name Service Caching Daemon (nscd).
2076 New configure option --disable-build-nscd avoids building nscd itself;
2077 this is the default if --disable-nscd is used.
2079 * Improved support for cross-compilation, including cross-testing and
2080 bootstrap builds without a previously built glibc.
2082 * Several testsuite tests are now able to test multiple IFUNC variants of an
2083 interface, rather than just testing the one that would be chooen by
2086 * New configure options --with-bugurl and --with-pkgversion, for
2087 distributors to use to embed their bug-reporting and package version
2088 information in --help and --version output.
2090 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2091 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2092 allows creation of chroots without the procfs mounted on /proc.
2094 * The `crypt' function now fails if passed salt bytes that violate the
2095 specification for those values. On Linux, the `crypt' function will
2096 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2097 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2098 when the mode is enabled.
2100 * The `clock_*' suite of functions (declared in <time.h>) is now available
2101 directly in the main C library. Previously it was necessary to link with
2102 -lrt to use these functions. This change has the effect that a
2103 single-threaded program that uses a function such as `clock_gettime' (and
2104 is not linked with -lrt) will no longer implicitly load the pthreads
2105 library at runtime and so will not suffer the overheads associated with
2106 multi-thread support in other code such as the C++ runtime library.
2108 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2113 * The following bugs are resolved with this release:
2115 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2116 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2117 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2118 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2119 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2120 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2121 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2122 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2123 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2124 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2125 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2126 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2127 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2128 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2129 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2130 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2131 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2132 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2133 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2134 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2135 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2136 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2139 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2140 configuring glibc with:
2141 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2142 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2143 Implemented by H.J. Lu.
2147 + define static_assert
2149 + do not declare gets
2151 + declare at_quick_exit and quick_exit also for ISO C11
2153 + aligned_alloc. NB: The code is deliberately allows the size parameter
2154 to not be a multiple of the alignment. This is a moronic requirement
2155 in the standard but it is only a requirement on the caller, not the
2158 + timespec_get added
2160 + uchar.h support added
2162 + CMPLX, CMPLXF, CMPLXL added
2164 Implemented by Ulrich Drepper.
2166 * Support for the IA-64 has been moved to ports.
2168 * Remove support for anything but ELF binary format
2170 * Checking versions of poll, ppoll added.
2171 Implemented by Ulrich Drepper.
2173 * More generic and 64-bit performance optimizations to math functions.
2174 Implemented by Ulrich Drepper.
2176 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2177 headers and functions available at compile time as they were before
2178 version 2.14. This option will be removed at some time in the future
2179 after the TI-RPC library becomes fully sufficient for the needs of
2180 existing applications.
2182 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2183 Note that glibc is not expected to work with any Linux kernel version
2186 * New header <sys/auxv.h> and function getauxval allowing easy access to
2187 the AT_* key-value pairs passed from the Linux kernel. The header also
2188 defines the HWCAP_* bits associated with the AT_HWCAP key.
2190 * New locales: mag_IN
2192 * New configure option --enable-systemtap builds SystemTap static probes
2193 into libc for setjmp and longjmp and into libpthread for various operations.
2194 So far the setjmp/longjmp probes and some of the libpthread probes are
2195 provided only for i*86 and x86_64.
2196 Implemented by Roland McGrath and Rayson Ho.
2198 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2200 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2203 * More optimized functions for SPARC. Implemented by David S. Miller.
2205 * Improved support for cross-compilation, especially bootstrap builds
2206 without a previously built glibc.
2208 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2209 Chris Metcalf from Tilera.
2211 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2212 now supported for ARM processors.
2214 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2215 as the name of the dynamic linker, to distinguish it from the
2216 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2218 * Support for CRIS has been removed from ports.
2220 * A new class of installed header has been documented for low-level
2221 platform-specific functionality. PowerPC added the first instance with a
2222 function to provide time base register access. Contributed by Tulio
2223 Magno Quites Machado Filho.
2225 * ix86 configurations now install header files that are consistent with
2226 what x86-64 configurations install. These same header files can be used
2227 for -m32, -m64, or -mx32 builds.
2228 Contributed by H.J. Lu.
2230 * Math library bug fixes. A thorough audit of all open math library bugs was
2231 conducted by Joseph Myers. Significant progress was made on many math
2232 library bugs resulting in more accurate exceptions and function results.
2233 Many thanks to all those that contributed including Andreas Jaeger for his
2234 patch review and work on the x87 trigonometric instruction issues.
2236 * Timezone data is no longer installed. Timezone-related binaries and scripts
2237 will continue to be installed. Users should obtain their timezone data from
2238 their distribution provider or from the tzdata package at
2239 <ftp://munnari.oz.au/pub/>.
2243 * The following bugs are resolved with this release:
2245 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2246 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2247 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2248 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2249 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2250 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2251 13484, 13506, 13515, 13523, 13524, 13538, 13540
2253 * New program pldd to list loaded object of a process
2254 Implemented by Ulrich Drepper.
2256 * Add nss_db support back to glibc. No more dependency on Berkeley db
2257 and support for initgroups lookups.
2258 Implemented by Ulrich Drepper.
2260 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2261 Contributed by HJ Lu.
2263 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2264 Contributed by HJ Lu.
2266 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2267 on x86-32 and x86-64.
2268 Contributed by Liubov Dmitrieva.
2270 * Optimized strchr and strrchr for SSE on x86-32.
2271 Contributed by Liubov Dmitrieva.
2273 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2274 for x86-64 and x86-32.
2275 Contributed by Liubov Dmitrieva.
2277 * New interfaces: scandirat, scandirat64
2278 Implemented by Ulrich Drepper.
2280 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2281 Implemented by Ulrich Drepper.
2283 * nscd now also caches the netgroup database.
2284 Implemented by Ulrich Drepper.
2286 * Integrate libm with gcc's -ffinite-math-only option.
2287 Implemented by Ulrich Drepper.
2289 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2290 to math functions. Implemented by Ulrich Drepper.
2292 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2293 Implemented by Ulrich Drepper.
2295 * New Linux interfaces: process_vm_readv, process_vm_writev
2297 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2298 Implemented by Ulrich Drepper.
2300 * Optimized nearbyint and strcasecmp for PPC.
2301 Implemented by Adhemerval Zanella.
2303 * New locales: bho_IN, unm_US, es_CU, ta_LK
2307 * The following bugs are resolved with this release:
2309 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2310 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2311 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2312 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2313 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2314 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2315 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2316 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2317 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2318 12795, 12811, 12813, 12814, 12841
2320 * The RPC implementation in libc is obsoleted. Old programs keep working
2321 but new programs cannot be linked with the routines in libc anymore.
2322 Programs in need of RPC functionality must be linked against TI-RPC.
2323 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2325 Visible changes of this change include (obviously) the inability to link
2326 programs using RPC functions without referencing the TI-RPC library and the
2327 removal of the RPC headers from the glibc headers.
2328 Implemented by Ulrich Drepper.
2330 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2331 syncfs, setns, sendmmsg
2333 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2334 yue_HK, lij_IT, mhr_RU
2336 * New encodings: CP770, CP771, CP772, CP773, CP774
2338 * New program sotruss to trace calls through PLTs
2339 Implemented by Ulrich Drepper.
2341 * The malloc hook implementation is marked deprecated and will be removed
2342 from the default implementation in the next version. The design never
2343 worked ever since the introduction of threads. Even programs which do
2344 not create threads themselves can use multiple threads created internally.
2348 * The following bugs are resolved with this release:
2350 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2351 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2352 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2353 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2356 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2358 * POWER7 optimizations: memset, memcmp, strncmp
2360 * New optimized string functions for x86-64: strnlen (SSE2),
2361 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2362 Implemented by Ulrich Drepper.
2366 * The following bugs are resolved with this release:
2368 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2369 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2370 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2371 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2372 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2373 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2374 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2375 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2377 * New interfaces: pthread_getname_np, pthread_setname_np
2379 * New Linux interface: recvmmsg
2381 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2383 * The dynamic linker now recognizes supported ABI versions from the
2384 EI_ABIVERSION field in the ELF header.
2385 Implemented by Ulrich Drepper.
2387 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2388 will not be used to fill in password fields in the passwd.byname replies.
2389 Instead it is used to synthesize the shadow.byname table, should it be
2390 missing. This is a useful mode in some installations involving Solaris.
2391 Implemented by Ulrich Drepper.
2393 * New locales: kok_IN, sq_MK, cv_RU
2397 * The following bugs are resolved with this release:
2399 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2400 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2401 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2402 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2403 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2404 10789, 10817, 10823, 10840, 10847
2406 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2408 Implemented by Ulrich Drepper.
2410 * Checking version of longjmp added that fails if an uninitialized stack
2411 frame would be created. Implemented by Ulrich Drepper.
2413 * STT_GNU_IFUNC is now supported in static executables.
2414 Implemented by H.J. Lu.
2416 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2418 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2419 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2420 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2421 Contributed by H.J. Lu.
2423 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2424 Implemented by Ulrich Drepper.
2426 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2428 Contributed by H.J. Lu.
2430 * Support for fma instruction in AVX on x86-64.
2431 Implemented by H.J. Lu and Ulrich Drepper.
2433 * AVX support in x86-64 auditing support in ld.so.
2434 Implemented by H.J. Lu.
2436 * STB_GNU_UNIQUE support added.
2437 Implemented by Ulrich Drepper.
2439 * Implement second fallback mode for optimized DNS lookup for even more
2440 broken environments. If two requests from the same port are not
2441 handled correctly close the socket and open a new one before sending
2442 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2443 can be used to select this mode right away, instead of rediscovering the
2444 necessity is every process again.
2445 Implemented by Ulrich Drepper.
2447 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2448 Implemented by Adam Tkac.
2450 * Optimized iconv conversions for S390x.
2451 Implemented by Andreas Krebbel.
2453 * Using condvars with PI mutexes is now more efficient due to kernel
2454 support for requeueing to PI futexes. NPTL support added for x86-64.
2455 Implemented by Ulrich Drepper.
2457 * New locales: ps_AF, my_MM
2461 * The following bugs are resolved with this release:
2463 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2464 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2465 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2466 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2467 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2469 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2470 Implemented by Ulrich Drepper.
2472 * New Linux interfaces: accept4, fallocate, fallocate64.
2473 Implemented by Ulrich Drepper.
2475 * Correct declarations of string function when used in C++ code. This
2476 could lead to compile errors for invalid C++ code.
2478 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2481 * New POSIX 2008 interface: psiginfo
2482 Implemented by Ulrich Drepper.
2484 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2485 Implemented by Ulrich Drepper.
2487 * Support for selecting between multiple function definitions at runtime
2488 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2490 * The libcrypt library can now use the hash function implementations in
2491 NSS. Implemented by Ulrich Drepper.
2493 * The malloc implementation can be compiled to be less memory efficient
2494 but higher performing in multi-threaded programs.
2495 Implemented by Ulrich Drepper.
2497 * New locales: nan_TW@latin, ks_IN
2499 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2500 Implemented by Ulrich Drepper.
2502 * Extended printf hook support. It is possible to use user-defined types
2503 and extend existing format specifiers.
2504 Implemented by Ulrich Drepper.
2506 * Handling for group shadow files has been added.
2507 Implemented by Ulrich Drepper.
2509 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2510 servers (the case, e.g., for some people using the built-in DNS
2511 server in ADSL modems/routers). There is a once-per-process timeout
2512 in case of a broken server. To avoid it, users can run nscd or put
2513 'options single-request' in /etc/resolv.conf.
2514 Implemented by Ulrich Drepper.
2518 * The following bugs are resolved with this release:
2520 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2521 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2522 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2523 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2524 6968, 6974, 6980, 6995, 7008, 7009, 7029
2526 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2527 up at the same time. Implemented by Ulrich Drepper.
2529 * TLS descriptors for LD and GD on x86 and x86-64.
2530 Implemented by Alexandre Oliva.
2532 * getaddrinfo now handles DCCP and UDPlite.
2533 Implemented by Ulrich Drepper.
2535 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2536 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2537 Implemented by Ulrich Drepper.
2539 * New implementation of memmem, strstr, and strcasestr which is O(n).
2540 Implemented by Eric Blake.
2542 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2544 * Implement "e" option for popen to open file descriptor with the
2545 close-on-exec flag set. Implemented by Ulrich Drepper.
2547 * Implement "b" mode for fmemopen. In this mode writes writes don't
2548 implicitly add a NUL byte and seeks from the end of the buffer really
2549 use the buffer end, not the string length as the basis.
2550 Implemented by Ulrich Drepper.
2552 * Many functions, exported and internal, now atomically set the close-on-exec
2553 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2555 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2557 Implemented by Pravin Satpute.
2559 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2563 * The following bugs are resolved with this release:
2565 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2566 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2567 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2568 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2569 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2570 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2571 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2573 * New locales: bo_CN, bo_IN, shs_CA.
2575 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2577 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2578 Implemented by Pravin Satpute.
2580 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2582 * nscd caches now all timeouts for DNS entries
2583 Implemented by Ulrich Drepper.
2585 * nscd is more efficient and wakes up less often.
2586 Implemented by Ulrich Drepper.
2588 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2589 vdprintf, and obstack_vprintf.
2590 Implemented by Jakub Jelinek.
2592 * Faster memset for x86-64.
2593 Implemented by Harsha Jagasia and H.J. Lu.
2595 * Faster memcpy on x86.
2596 Implemented by Ulrich Drepper.
2598 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2599 Implemented by Ulrich Drepper.
2601 * Faster sqrt and sqrtf implemention for some PPC variants.
2602 Implemented by Steven Munroe.
2606 * The following bugs are resolved with this release:
2608 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2609 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2610 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2611 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2612 5071, 5103, 5104, 5112, 5113, 5184, 5186
2614 * More checking functions: fread, fread_unlocked, open*, mq_open.
2615 Implemented by Jakub Jelinek and Ulrich Drepper.
2617 * Extend fortification to C++. Implemented by Jakub Jelinek.
2619 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2620 by not recognizing 'a' as a modifier when those specs are requested.
2621 Implemented by Jakub Jelinek.
2623 * PPC optimizations to math and string functions.
2624 Implemented by Steven Munroe.
2626 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2627 options to be passed. Implemented by Ulrich Drepper.
2629 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2631 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2632 Implemented by Ulrich Drepper.
2634 * Handle private futexes in the NPTL implementation.
2635 Implemented by Jakub Jelinek and Ulrich Drepper.
2637 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2638 Implemented by Roland McGrath and Ulrich Drepper.
2640 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2642 * SHA-256 and SHA-512 based password encryption.
2643 Implemented by Ulrich Drepper.
2645 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2646 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2649 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2650 Implemented by Ulrich Drepper.
2654 * The following bugs are resolved with this release:
2656 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2657 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2658 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2659 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2660 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2661 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2662 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2665 * New Linux interfaces: epoll_pwait, sched_getcpu.
2667 * New generic interfaces: strerror_l.
2669 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2673 * The following bugs are resolved with this release:
2675 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2676 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2677 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2678 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2679 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2680 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2681 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2682 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2683 3225, 3251, 3252, 3253, 3273
2685 * For Linux, the sorting of addresses returned by getaddrinfo now also
2686 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2687 Implemented by Ulrich Drepper.
2689 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2690 Implemented by Ulrich Drepper.
2692 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2694 * New iconv module for MIK. Contributed by Alexander Shopov.
2696 * For sites with broken group and/or passwd database, the auto-propagate
2697 option of nscd can prevent creating ID lookup entries from the results
2698 of a name lookup and vice versa. This usually is no problem but some
2699 site might have problems with the default behavior.
2700 Implemented by Ulrich Drepper.
2702 * Iterating over entire database in NIS can be slow. With the
2703 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2704 to trade time for memory. The entire database will be read at once.
2705 Implemented by Ulrich Drepper.
2707 * The interfaces introduced in RFC 3542 have been implemented by
2710 * Support for the new ELF hash table format was added by Ulrich Drepper.
2712 * Support for priority inheritance mutexes added by Jakub Jelinek and
2715 * Support for priority protected mutexes added by Jakub Jelinek.
2717 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2718 or_IN, csb_PL, fy_NL, sr_ME.
2722 * More overflow detection functions.
2724 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2725 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2727 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2728 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2729 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2730 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2731 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2732 by Masahide Washizawa.
2734 * It is now possible to install audit libraries for the dynamic linker using
2735 LD_AUDIT. Implemented by Ulrich Drepper.
2737 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2738 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2739 For a libc and libpthread that works well on Linux 2.4 kernels, we
2740 recommend using the stable 2.3 branch.
2742 * The new function open_wmemstream acts like open_memstream,
2743 but uses a wchar_t wide character buffer.
2745 * The new function ppoll is an improved version of poll, similar to pselect.
2747 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2748 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2749 renameat, unlinkat, linkat, symlinkat, readlinkat.
2751 * New Linux kernel system calls: unshare,
2752 inotify_init, inotify_add_watch, inotify_rm_watch.
2754 * The euidaccess function is now also known by the alias eaccess,
2755 for compatibility with some other systems.
2757 * Timezone data updated to 2006b version.
2761 * The following bugs are resolved with this release:
2763 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2764 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2765 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2766 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2767 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2768 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2770 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2772 * As of this release, GCC 4 can be used to compile the C Library.
2774 * Timezone data updated to 2005m version.
2778 * The following bugs are resolved with this release:
2780 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2781 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2782 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2783 777, 787, 821, 822, 823, 825
2785 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2789 * Support for RFC 3678. Real implementations exist only for Linux so far.
2790 Implemented by Ulrich Drepper.
2792 * nscd can now cache entries persistently. Expiring entries are reloaded.
2793 For speedups the cache can be shared in memory with client processes.
2794 Implemented by Ulrich Drepper.
2796 * nscd can now perform SELinux checks.
2797 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2799 * getaddrinfo queries are now cached. Canonical name lookup is performed
2801 Implemented by Ulrich Drepper.
2803 * The nothrow function attribute is used when headers are used by gcc when
2804 compiling C code. This can avoid significant amounts of exception
2807 * The malloc functions perform more error checking and are stricter when
2808 it comes to reacting on errors. The default action is to terminate
2809 the process after showing an error message. Implemented by Ulrich Drepper.
2811 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2812 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2814 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2815 namespaces using the new function dlmopen(). This feature is of course,
2816 like most other dynamic loading functionality, not available in statically
2817 linked applications. Implemented by Ulrich Drepper.
2819 * Low-overhead boundary checking variants of string and some stdio functions
2820 were added. These are to be used in conjunction with a gcc patch by
2821 Jakub Jelinek which adds calls to these functions if possible.
2822 Implemented by Jakub Jelinek and Ulrich Drepper.
2824 * Old code for several operating systems and machine architectures that
2825 have not been in working condition in a long time have been removed from
2826 the main source tree maintained by the GNU C Library's maintainers.
2827 These files are now reside in the separate `ports' source module
2828 that is usable as an add-on when building the library.
2832 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2833 interrogate the dynamic linker, compatible with the Solaris interface.
2835 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2836 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2838 * getifaddrs now uses the netlink interface on Linux to get its information.
2839 Implemented by Thorsten Kukuk.
2841 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2842 Implemented by Ulrich Drepper.
2844 * support for non-executable stacks on x86 has been added. Changes mostly
2847 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2850 * getaddrinfo now performs destination address selection according to
2855 * Thread-safe interfaces for many functions that access locale data
2856 were added in version 2.3, but these features were omitted from NEWS.
2857 Many functions have variants with an `_l' suffix that take a `locale_t'
2858 object as a parameter rather than consulting the current locale.
2859 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2860 create and maintain `locale_t' objects. Additionally, the new function
2861 `uselocale' sets "the current locale" (as used by functions not so
2862 parameterized) set for an individual thread. These features were added
2863 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2865 * The functions getresuid, getresgid, setresuid, and setresgid, which
2866 have long been available on Linux, are now declared in <unistd.h>
2867 and are now also available on the Hurd.
2869 * ELF thread-local storage support (TLS) now works on x86-64.
2871 * The new dynamic string token $LIB is expanded in shared library names.
2872 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2874 * Aldy Hernandez contributed complete software floating point support for
2875 PowerPC machines with no FPU.
2877 * fexecve is implemented on Linux.
2879 * The `btowc' function should work at least twice as fast due to
2880 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2882 * With approriate thread add-ons cancelable functions are now implemented
2883 in libc.so as well. No need to call the function in libpthread. This
2884 change allowed to finally disable the incorrect and expensive handling
2885 of weak definition in ld.so.
2887 * Yet more PLT entries in libc.so have been removed. We finally arrived
2888 at the bare minimum. Startup times improved appropriately.
2890 * Support for the new Linux/x86 system call interface was added. The
2891 AT_SYSINFO auxiliary vector entry is recognized and handled.
2895 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2898 * iconv (the program and the interface) now accepts empty names (excluding
2899 options like //TRANSLIT) to mean "use charset of current locale".
2901 * localedef can now transliterate characters in strings which are not in
2902 the provided charmap. The information from the input locale is used.
2904 * Prelinking support was added for ELF targets. This requires additional
2905 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2907 * Read-only stdio streams now use mmap to speed up operation by eliminating
2908 copying and buffer underflows. To use add 'm' to the mode string of
2909 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2911 * The malloc functions were completely rewritten by Wolfram Gloger based
2912 on Doug Lea's malloc-2.7.0.c.
2914 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2915 implementation of regex.
2917 * Bruno Haible upgraded the iconv and locale implementation to support
2920 * Contents of the LC_* and LANG environment variables in the CEN style are
2921 not recognized anymore. It never was used. Change by Ulrich Drepper.
2923 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2924 thread-local storage (TLS) ABI on some platforms.
2925 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2927 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2928 EUC-JISX0213, and TSCII.
2930 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2931 BSD-compatible interface for getting all network interface addresses.
2932 Implementation for IPv4 by Roland McGrath.
2934 * Loading of locale data is faster due to the introduction of a locale
2935 archive. Implemented by Roland McGrath and Ulrich Drepper.
2937 * Startup times are significantly reduced by not using exported functions
2938 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2941 * Steven Munroe contributed a port to PowerPC64/Linux.
2945 * The Hurd now uses the GNU libio implementation of stdio rather than the
2946 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2948 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2949 with the same behavior as the Linux system call of the same name.
2953 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2954 128-bit long double format.
2956 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2957 IBM1161, and IBM1162 charsets.
2959 * Andreas Jaeger contributed a port to x86-64/Linux.
2961 * Peter Bruin contributed a port to PowerPC/Hurd.
2963 * libc functions using I/O streams now can handle wide-oriented streams
2966 * optimizations in the dynamic linker. Binaries created by recent binutils
2967 versions start up quicker due to reduced time spent on relocations.
2969 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2973 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2974 asinh, atanh, j0 for the 128-bit long double format.
2976 * Bruno Haible updated all the code handling Unicode in some form to
2977 support Unicode 3.1.
2979 * Speed of regex for single-byte locales is back to previous levels.
2980 Patch by Isamu Hasegawa.
2982 * Alpha, SPARC, and IA-64 now also using floating stacks.
2984 * Startup time of internationalized applications greatly improved through
2985 iconv cache. Use iconvconfig to generate the cache file.
2986 Contributed by Ulrich Drepper.
2988 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2989 severe performance problems. Patch by David Mosberger.
2991 * The Hurd port got a lot more functionality like AIO, various stdio
2992 extensions, etc. Mainly done by Roland McGrath.
2994 * mtrace can now lookup symbols in shared libraries.
2998 * Intel's IA-64 math library is largely integrated. It provides fast and
2999 accurate implementatations for most basic and standard math functions
3000 in float, double, and long double format.
3002 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3003 and asin for the 96-bit long double format and asin, log, tan for the
3004 128-bit long double format.
3006 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3007 The basic double functions exist today. Contributed by Abraham Ziv
3008 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3009 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3011 * An asynchronous name lookup library was added. The interface is designed
3012 after POSIX AIO. The proposal was circulated beforehand to get comments.
3013 No negative ones came in. Implemented by Ulrich Drepper.
3015 * Port to S390/64bit contributed by Martin Schwidefsky
3016 <schwidefsky@de.ibm.com>.
3018 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3019 of functions for Linux/IA-64.
3021 * The RPC code is now thread safe. Threads can now use the same service
3022 of different services at the same time. Patch by Eric Norum
3023 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3025 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3026 family of functions for Linux/S390.
3028 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3029 of functions for Linux/x86.
3031 * Port to Linux/CRIS contributed by Axis Communications.
3035 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3036 we can now check for namespace violations and missing declarations. The
3037 result is that almost all headers are now Unix-compliant (as defined in
3038 the upcoming XPG6). The negative side is that some programs might need
3039 corrections, too, if they depend on the incorrect form of the headers in
3040 previous versions which defined too many symbols and included too many
3043 * regex now handles multibyte character sets correctly.
3044 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3046 * iconv (the program) does now conform to the upcoming XPG6 and handles
3047 charmaps. Instead of the charset names the path of charmaps can be
3048 provided and the conversion happens based on this data.
3049 Contributed by Ulrich Drepper.
3051 * The locale program now provides detailed information about the installed
3056 only lists the names of the supported locales
3060 provides details such as country, language, and codeset name.
3061 Contributed by Ulrich Drepper.
3065 * The gencat program now parses the input file according to the charset
3066 selected by the LC_CTYPE category. This is important for stateful
3067 character sets. To make generating catalogs easier there is a way
3068 to overwrite the charset selected by the locale: before the first
3069 message or $ quote line the catalog can contain a line like
3071 $ codeset=ISO-8859-2
3073 to select the charset (ISO-8859-2 in this case).
3075 Implemented by Shinya Hanataka and Ulrich Drepper.
3077 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3078 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3079 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3081 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3082 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3084 * The LANGUAGE environment variable is now ignored unless the locale is
3085 changed from the default "C" locale.
3087 * The usual bug fixes.
3091 * Greg McGary added runtime support for bounds checking using gcc's
3092 new -fbounded-pointers option. ix86 target is complete. PowerPC
3095 * Thorsten Kukuk added secure mode support to nscd.
3097 * The Berkeley db libraries have been removed.
3099 Related, the nss_db module is now in a separate package since it
3100 obviously requires a database library being available.
3102 * Wide character I/O streams implemented by Ulrich Drepper.
3104 * Functions from the extended socket API added by Ulrich Drepper.
3106 * Functions feenableexcept and fedisableexcept to control the
3107 behaviour of individual exceptions have been added by Andreas Jaeger.
3109 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3111 * The resolver code has been updated from bind 8.2.3-T5B which supports
3112 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3115 This change could in some situations effect backward compatibility. Since
3116 now `_res' is a thread-local instead of a global variable, modifying it
3117 in one thread does not have any effect in other threads.
3119 The resolver library was also extended to allow IPv6 as the transport
3120 protocol for the requests. This work was done by Stig Venaas.
3122 * Compatibility code for K&R C compilers has been removed from the
3123 header files. A ISO C compiler is needed to use the library
3124 (conforming to either C89 or C99 standard).
3126 * Complete rewrite of the localedef program to support multibyte character
3127 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3128 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3129 Implemented by Ulrich Drepper.
3131 Bruno Haible significantly improved the generation and use of the data
3132 structures for the wide character tables.
3134 * Plural handling in gettext implemented by Ulrich Drepper.
3136 * The utmp daemon has been removed.
3138 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3140 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3143 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3145 * POSIX barriers implemented by Kaz Kylheku.
3147 * POSIX spawn function family implemented by Ulrich Drepper.
3149 * POSIX spinlocks are now available.
3151 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3153 * the configure option --enable-kernel=X.Y.Z allows to strip out
3154 compatibility for kernel versions before X.Y.Z. This is currently only
3155 implemented for Linux.
3157 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3158 field sin6_scope_id. This means that all programs using IPv6 should be
3159 recompiled. Don't expect binary compatibility with previous glibc
3162 * various conversion modules for IBM character sets contributed by
3165 * IA-64 port by Jes Sorensen and HJ Lu.
3179 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3181 * New xdr functions are added; some rpc functions are now 64bit clean.
3183 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3185 * Fixed known incompatibilities with glibc 2.0.
3187 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3189 * Optimized a number of functions (especially the ELF dynamic loader).
3191 * Update timezone data files.
3193 * lots of charmaps corrections
3195 * some new locale definitions and charmaps
3200 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3201 conform to POSIX member type requirements. Symbol versions have been
3202 adjusted as needed within the library, and for direct use by applications,
3203 but there is potential for problems if third-party libraries use
3204 struct timeval as part of their interface. This does not present
3205 a problem for X and other "essential" system libraries.
3207 * An additional locale model to support C++ Standard Library locale
3208 model and probably more was implemented by Ulrich Drepper.
3210 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3213 * Miles Bader provided the `argp' function family to support hierachical
3214 command line argument parsing, layered on top of getopt.
3216 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3218 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3221 * scanf recognizes the %a and %A format for scanning floating point numbers.
3223 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3224 information and interfaces for the available integer types.
3226 * about 130 new math functions were added to implement the ISO C9x math
3229 * the new header <complex.h> contains definitions of the complex math
3230 functions from ISO C 9X.
3232 * the new header <tgmath.h> defines generic macros to use complex or
3233 real valued functions.
3235 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3237 * Andreas Jaeger provided a test suite for the math library.
3239 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3241 * Ulrich Drepper added character set conversion functions (iconv).
3243 * Optimized string functions have been added.
3245 * The localedata addon is now part of glibc.
3247 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3249 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3250 daemon for NSS (nscd).
3252 Missing a better place here are some numbers on improvements. Under
3253 Linux 2.1.125 un-tar-ing the kernel sources takes
3257 using local files 12.19s 6.88s 22.91s
3259 using NIS 13.92s 8.91s 26.34s
3261 using NIS & nscd 10.37s 7.34s 25.30s
3263 using NIS+ 27.57s 30.37s 640.46s
3265 using NIS+ & nscd 10.25s 7.83s 26.51s
3267 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3269 Keep in mind that non-namelookup related operations dominate above times.
3270 It was just a common complain that using NIS+ unpacking the kernel is
3273 [1] The old nscd implementation is not available anymore since it was
3274 distributed with glibc up to version 2.0.98 and thus is now replaced.
3276 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3278 * Mark Kettenis provided a Hesiod NSS module.
3280 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3281 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3283 * rcmd can now handle netgroups (Dick Streefland).
3285 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3288 * Support for the IPv6 protocol has been added to the socket API, as per the
3289 latest draft standards.
3291 * Support for Linux 2.2 has been added.
3293 * Interface changes relative to the latest 2.0.x release:
3294 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3295 addseverity NEW: Unix98
3296 alphasort64 NEW: LFS
3297 argp_err_exit_status NEW: argp, GNU ext
3298 argp_error NEW: argp, GNU ext
3299 argp_failure NEW: argp, GNU ext
3300 argp_help NEW: argp, GNU ext
3301 argp_parse NEW: argp, GNU ext
3302 argp_program_bug_address NEW: argp, GNU ext
3303 argp_program_version NEW: argp, GNU ext
3304 argp_program_version_hook NEW: argp, GNU ext
3305 argp_state_help NEW: argp, GNU ext
3306 argp_usage NEW: argp, GNU ext
3307 authdes_create NEW: Secure RPC
3308 authdes_getucred NEW: Secure RPC
3309 authdes_pk_create NEW: Secure RPC
3310 backtrace NEW: GNU ext.
3311 backtrace_symbols NEW: GNU ext.
3312 backtrace_symbols_fd NEW: GNU ext.
3314 cacosf NEW: ISO C 9x
3315 cacosh NEW: ISO C 9x
3316 cacoshf NEW: ISO C 9x
3317 cacoshl NEW: ISO C 9x
3318 cacosl NEW: ISO C 9x
3325 casinf NEW: ISO C 9x
3326 casinh NEW: ISO C 9x
3327 casinhf NEW: ISO C 9x
3328 casinhl NEW: ISO C 9x
3329 casinl NEW: ISO C 9x
3331 catanf NEW: ISO C 9x
3332 catanh NEW: ISO C 9x
3333 catanhf NEW: ISO C 9x
3334 catanhl NEW: ISO C 9x
3335 catanl NEW: ISO C 9x
3336 cbc_crypt NEW: Secure RPC
3340 ccoshf NEW: ISO C 9x
3341 ccoshl NEW: ISO C 9x
3347 cimagf NEW: ISO C 9x
3348 cimagl NEW: ISO C 9x
3349 clearerr_locked REMOVED
3350 clntunix_create NEW: sunrpc ext
3352 clog10 NEW: ISO C 9x
3353 clog10f NEW: ISO C 9x
3354 clog10l NEW: ISO C 9x
3364 cprojf NEW: ISO C 9x
3365 cprojl NEW: ISO C 9x
3367 crealf NEW: ISO C 9x
3368 creall NEW: ISO C 9x
3373 csinhf NEW: ISO C 9x
3374 csinhl NEW: ISO C 9x
3377 csqrtf NEW: ISO C 9x
3378 csqrtl NEW: ISO C 9x
3382 ctanhf NEW: ISO C 9x
3383 ctanhl NEW: ISO C 9x
3385 des_setparity NEW: Secure RPC
3386 ecb_crypt NEW: Secure RPC
3387 endutxent NEW: Unix98
3389 exp10f NEW: ISO C 9x
3390 exp10l NEW: ISO C 9x
3394 fattach NEW: STREAMS
3395 fdetach NEW: STREAMS
3399 feclearexcept NEW: ISO C 9x
3400 fegetenv NEW: ISO C 9x
3401 fegetexceptflag NEW: ISO C 9x
3402 fegetround NEW: ISO C 9x
3403 feholdexcept NEW: ISO C 9x
3405 feraiseexcept NEW: ISO C 9x
3406 ferror_locked REMOVED
3407 fesetenv NEW: ISO C 9x
3408 fesetexceptflag NEW: ISO C 9x
3409 fesetround NEW: ISO C 9x
3410 fetestexcept NEW: ISO C 9x
3411 feupdateenv NEW: ISO C 9x
3412 fflush_locked REMOVED
3416 fgets_unlocked NEW: GNU ext.
3417 fileno_locked REMOVED
3429 fputc_locked REMOVED
3430 fputs_unlocked NEW: GNU ext.
3431 fread_unlocked NEW: GNU ext.
3436 fstatvfs NEW: Unix98
3440 ftruncate64 NEW: LFS
3442 fwrite_unlocked NEW: GNU ext.
3443 gai_strerror NEW: IPv6
3447 getchar_locked REMOVED
3449 getdate_err NEW: Unix98
3450 getdate_r NEW: GNU ext.
3452 getnameinfo NEW: IPv6
3453 getnetname NEW: Secure RPC
3454 getpmsg NEW: STREAMS
3455 getpt NEW: Unix98 PTY
3456 getrlimit64 NEW: LFS
3457 getutxent NEW: Unix98
3458 getutxid NEW: Unix98
3459 getutxline NEW: Unix98
3460 glob64 NEW: GNU ext.
3461 globfree64 NEW: GNU ext.
3462 gnu_get_libc_release NEW: GNU ext.
3463 gnu_get_libc_version NEW: GNU ext.
3464 grantpt NEW: Unix98 PTY
3465 host2netname NEW: Secure RPC
3467 iconv_close NEW: iconv
3468 iconv_open NEW: iconv
3469 if_freenameindex NEW: IPv6
3470 if_indextoname NEW: IPv6
3471 if_nameindex NEW: IPv6
3472 if_nametoindex NEW: IPv6
3473 in6addr_any NEW: IPv6
3474 in6addr_loopback NEW: IPv6
3475 inet6_isipv4mapped NEW: IPv6
3476 isastream NEW: STREAMS
3477 iswblank NEW: GNU ext.
3478 key_decryptsession NEW: Secure RPC
3479 key_decryptsession_pk NEW: Secure RPC
3480 key_encryptsession NEW: Secure RPC
3481 key_encryptsession_pk NEW: Secure RPC
3482 key_gendes NEW: Secure RPC
3483 key_get_conv NEW: Secure RPC
3484 key_secretkey_is_set NEW: Secure RPC
3485 key_setnet NEW: Secure RPC
3486 key_setsecret NEW: Secure RPC
3487 llrint NEW: ISO C 9x
3488 llrintf NEW: ISO C 9x
3489 llrintl NEW: ISO C 9x
3490 llround NEW: ISO C 9x
3491 llroundf NEW: ISO C 9x
3492 llroundl NEW: ISO C 9x
3497 lrintf NEW: ISO C 9x
3498 lrintl NEW: ISO C 9x
3499 lround NEW: ISO C 9x
3500 lroundf NEW: ISO C 9x
3501 lroundl NEW: ISO C 9x
3503 makecontext NEW: Unix98
3504 mempcpy NEW: GNU ext.
3507 modify_ldt NEW: kernel
3511 nearbyint NEW: ISO C 9x
3512 nearbyintf NEW: ISO C 9x
3513 nearbyintl NEW: ISO C 9x
3514 netname2host NEW: Secure RPC
3515 netname2user NEW: Secure RPC
3516 nexttoward NEW: ISO C 9x
3517 nexttowardf NEW: ISO C 9x
3518 nexttowardl NEW: ISO C 9x
3522 passwd2des NEW: Secure RPC
3524 pow10f NEW: GNU ext.
3525 pow10l NEW: GNU ext.
3528 printf_size NEW: GNU ext.
3529 printf_size_info NEW: GNU ext.
3530 profil_counter REMOVED
3531 pthread_mutexattr_getkind_np REPLACED
3532 pthread_mutexattr_setkind_np REPLACED
3533 ptsname NEW: Unix98 PTY
3534 ptsname_r NEW: Unix98 PTY
3536 putchar_locked REMOVED
3537 putgrent NEW: GNU ext.
3539 putpmsg NEW: STREAMS
3540 pututxline NEW: Unix98
3544 readdir64_r NEW: LFS
3545 remquo NEW: ISO C 9x
3546 remquof NEW: ISO C 9x
3547 remquol NEW: ISO C 9x
3549 roundf NEW: ISO C 9x
3550 roundl NEW: ISO C 9x
3552 scalbln NEW: ISO C 9x
3553 scalblnf NEW: ISO C 9x
3554 scalblnl NEW: ISO C 9x
3556 sendfile NEW: kernel
3557 setcontext NEW: Unix98
3558 setrlimit64 NEW: LFS
3559 setutxent NEW: Unix98
3561 sigignore NEW: Unix98
3562 sigqueue NEW: POSIX.1b
3563 sigrelse NEW: Unix98
3564 sigset NEW: POSIX.1b
3565 sigtimedwait NEW: POSIX.1b
3566 sigwaitinfo NEW: POSIX.1b
3567 sincos NEW: GNU ext.
3568 sincosf NEW: GNU ext.
3569 sincosl NEW: GNU ext.
3573 strcasestr NEW: GNU ext.
3574 strtoimax NEW: ISO C 9x
3575 strtoumax NEW: ISO C 9x
3576 strverscmp NEW: GNU ext.
3577 svcauthdes_stats NEW: Secure RPC
3578 svcunix_create NEW: sunrpc etx
3579 svcunixfd_create NEW: sunrpc ext
3580 swapcontext NEW: Unix98
3581 tcgetsid NEW: Unix98 PTY
3582 tdestroy NEW: GNU ext.
3583 tgamma NEW: ISO C 9x
3584 tgammaf NEW: ISO C 9x
3585 tgammal NEW: ISO C 9x
3589 truncf NEW: ISO C 9x
3590 truncl NEW: ISO C 9x
3592 unlockpt NEW: Unix98 PTY
3593 updwtmpx NEW: Unix98
3594 user2netname NEW: Secure RPC
3595 utmpxname NEW: Unix98
3596 versionsort NEW: GNU ext.
3597 versionsort64 NEW: GNU ext.
3599 wcscasecmp NEW: GNU ext.
3600 wcsncasecmp NEW: GNU ext.
3601 wcsnlen NEW: GNU ext.
3602 wcstoimax NEW: ISO C 9x
3603 wcstoll NEW: ISO C 9x
3604 wcstoull NEW: ISO C 9x
3605 wcstoumax NEW: ISO C 9x
3607 wordexp NEW: POSIX.2
3608 wordfree NEW: POSIX.2
3609 write_profiling REMOVED
3610 xdecrypt NEW: Secure RPC
3611 xdr_authdes_cred NEW: Secure RPC
3612 xdr_authdes_verf NEW: Secure RPC
3613 xdr_cryptkeyarg NEW: Secure RPC
3614 xdr_cryptkeyarg2 NEW: Secure RPC
3615 xdr_cryptkeyres NEW: Secure RPC
3616 xdr_getcredres NEW: Secure RPC
3617 xdr_key_netstarg NEW: Secure RPC
3618 xdr_key_netstres NEW: Secure RPC
3619 xdr_keybuf NEW: Secure RPC
3620 xdr_keystatus NEW: Secure RPC
3621 xdr_netnamestr NEW: Secure RPC
3622 xdr_sizeof NEW: Secure RPC
3623 xdr_unixcred NEW: sunrpc ext
3624 xencrypt NEW: Secure RPC
3625 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3636 * inet_ntoa is thread-safe
3638 * updwtmp is moved from libutil to libc
3640 * rewrite of cbrt function
3642 * update of timezone data
3656 * add atoll function
3658 * fix complex problems in Berkeley DB code
3660 * fix math functions
3664 * fixed lots of header problems (especially Linux/GNU specific)
3666 * dynamic loader preserves all registers
3668 * Roland McGrath provided support for handling of auxiliary objects in
3669 the ELF dynamic loader.
3671 * support for parallel builds is improved
3675 * GNU extensions are no longer declared by default. To enable them you
3676 must define the macro `_GNU_SOURCE' in your program or compile with
3679 * The library has changed from using GNU ld symbol aliases to using weak
3680 symbols where available. The ELF object file format supports weak
3681 symbols; GNU ld also supports weak symbols in the a.out format. (There
3682 is also now support for other GNU ld extensions in ELF. Use the
3683 `--with-elf' option to configure to indicate you have ELF, and
3684 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3685 of many files which contained only symbol aliases, reducing the size of
3686 the source and the compiled library; many other files were renamed to
3687 less cryptic names previously occupied by the symbol alias files.
3688 There is a new header file <elf.h> for programs which operate on
3689 files in the ELF format.
3691 * Converted to Autoconf version 2, so `configure' has more options.
3692 Run `configure --help' to see the details.
3694 * The library can now be configured to build profiling, highly-optimized
3695 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3696 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3697 `configure' enable building these extra libraries. The shared library is
3698 built by default when using both ELF and GNU ld. When shared libraries
3699 are enabled, the new library `-ldl' is available for arbitrary run-time
3700 loading of shared objects; its interface is defined in <dlfcn.h>. The
3701 new header file <link.h> gives access to the internals of the run-time
3702 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3703 application of same name on other systems and it provides information
3704 about dynamically linked binaries.
3706 * The C library now provides the run-time support code for profiling
3707 executables compiled with `-pg'. Programs can control the profiling code
3708 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3709 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3710 the support for this file format was contributed by David Mosberger-Tang.
3712 * The math code has been replaced with a math library based on fdlibm from
3713 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3714 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3715 functions. The math functions now reside in a separate library, so
3716 programs using them will need to use `-lm' their linking commands.
3718 * John C. Bowman contributed optimized ix87 assembler inline functions.
3720 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3721 mechanism similar to that found in Solaris 2. This is now used for the
3722 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3723 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3724 file controls what services are used for each individual database. This
3725 works by loading shared libraries with names specified in `nsswitch.conf',
3726 so service modules can be changed or added at any time without even
3727 relinking any program. Currently there are the file, db, and NIS based
3728 NSS services available.
3730 * The new functions `strtoq' and `strtouq' parse integer values from
3731 strings, like `strtol' and `strtoul', but they return `long long int' and
3732 `unsigned long long int' values, respectively (64-bit quantities).
3734 * The new functions `strtof' and `strtold' parse floating-point values from
3735 strings, like `strtod', but they return `float' and `long double' values,
3736 respectively (on some machines `double' and `long double' are the same).
3738 * Ulrich Drepper has contributed new implementations of the floating-point
3739 printing and reading code used in the `printf' family of functions and
3740 `strtod', `strtof', and `strtold'. These new functions are perfectly
3741 accurate, and much faster than the old ones.
3743 * The implementation of the POSIX locale model was completely rewritten by
3744 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3745 to compile the POSIX locale definition.
3747 * The former dummy implementations of the strcoll and strxfrm function are
3748 now replaced by fully functional code contributed by Ulrich Drepper. The
3749 collation information comes from the POSIX locale definitions.
3751 * The new header <langinfo.h> defines an interface for accessing
3752 various locale-dependent data (using the locale chosen with `setlocale').
3754 * Ulrich Drepper has contributed a new suite of functions for operation on
3755 wide-character and multibyte-character strings, in <wchar.h>;
3756 and classification and case conversion of wide characters, in <wctype.h>.
3757 These new functions are conforming to the ISO C, Amendement 1 specification.
3759 * There is now a second implementation of the standard I/O library available.
3760 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3761 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3763 * You can now use positional parameter specifications in format strings
3764 for the `printf' and `scanf' families of functions. For example,
3765 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3766 ``Number 6, Mr Jones''. This is mainly useful when providing different
3767 format strings for different languages, whose grammars may dictate
3768 different orderings of the values being printed. To support this
3769 feature, the interface for `register_printf_handler' has changed; see
3770 the header file <printf.h> for details.
3772 * The `printf' and `scanf' families of functions now understand a new
3773 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3774 to group numbers as indicated by the locale; for `scanf' and friends, this
3775 says to accept as valid only a number with all the proper grouping
3776 separators in the right places. In the default "C" locale, numbers are
3777 not grouped; but locales for specific countries will define the usual
3778 conventions (i.e. separate thousands with `,' in the US locale).
3780 * The pgrp functions have been regularized, slightly incompatibly but much
3781 less confusingly. The core functions are now `getpgid' and `setpgid',
3782 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3783 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3784 provided for compatibility. There is no longer an incompatible `getpgrp'
3785 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3787 * The new header file <fts.h> and suite of functions simplify programs that
3788 operate on directory trees. This code comes from 4.4 BSD.
3790 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3791 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3792 NSS scheme used in glibc.
3794 * There is a new malloc debugging hook `__memalign_hook'.
3796 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3797 `unsigned int' in <sys/types.h>. These are for compatibility only and
3798 their use is discouraged.
3800 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3801 done differently, so that it works even without GNU ld.
3803 * New function `euidaccess' checks allowed access to a file like `access',
3804 but using the effective IDs instead of the real IDs.
3806 * The time zone data files have been updated for the latest and greatest
3807 local time conventions of the countries of the world.
3809 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3812 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3813 for formatting floating-point numbers. They are provided only for
3814 compatibility; new programs should use `sprintf' instead. There are
3815 also equivalent function for the `long double' floating-point type and
3816 all functions also exist in a reentrant form.
3818 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3819 functions for maintaining the login-record files (primarily of use to
3820 system programs such as `login'), and convenient functions for
3821 allocating and initializing a pseudo-terminal (pty) device.
3823 * Ulrich Drepper has contributed new support for System V style
3824 shared memory and IPC on systems that support it.
3826 * Ulrich Drepper has contributed several miscellaneous new functions found
3827 in System V: The `hsearch' family of functions provide an effective
3828 implementation of hash tables; `a64l' and `l64a' provide a very simple
3829 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3832 * Ulrich Drepper has contributed new reentrant counterparts for the
3833 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3835 * Ulrich Drepper has contributed new, highly-optimized versions of several
3836 string functions for the i486/Pentium family of processors.
3838 * Ulrich Drepper has updated the Linux-specific code, based largely
3839 on work done in Hongjiu Lu's version of GNU libc for Linux.
3840 The GNU library now supports Linux versions 2.0.10 and later,
3841 using the ELF object file format (i[3456]86-*-linux).
3843 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3845 * David Mosberger-Tang and Richard Henderson have ported the C library
3846 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3847 dynamic linking support for ELF/Alpha.
3849 * Richard Henderson contributed several Alpha optimized assembler function
3850 for arithmetic and string handling.
3852 * Ulrich Drepper has contributed a new set of message catalog functions to
3853 support multiple languages using the <libintl.h> interface, for use with
3854 his new package GNU gettext. Translation volunteers have contributed
3855 catalogs of the library's messages in Spanish, German, and Korean.
3857 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3858 program and the `catgets' function for reading the catalog files it
3859 creates. (The <libintl.h> interface is preferred; we include the
3860 <nl_types.h> interface using `catgets' only for source compatibility with
3861 programs already written to use it.)
3863 * New header file <values.h> gives SVID-compatible names for <limits.h>
3866 * Various new macros, declarations, and small header files for compatibility
3869 * New function `group_member' is a convenient way to check if a process has
3870 a given effective group ID.
3872 * When using GCC 2.7 and later, the socket functions are now declared in a
3873 special way so that passing an argument of type `struct sockaddr_in *',
3874 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3875 `struct sockaddr *' type, does not generate a type-clash warning.
3877 * New function `error' declared in header file <error.h> is a convenient
3878 function for printing error messages and optionally exiting; this is the
3879 canonical function used in GNU programs. The new functions `err', `warn',
3880 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3881 doing the same thing.
3883 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3884 POSIX.2 `glob' function to do ~ and {...} expansion.
3886 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3887 `clearenv' which is used in POSIX.9 is also available.
3889 * New function `getsid' returns session ID number on systems that support it.
3891 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3892 files <db.h> and <mpool.h> provide a rich set of functions for several
3893 types of simple databases stored in memory and in files, and <ndbm.h> is
3894 an old `ndbm'-compatible interface using the `db' functions. Link with
3895 `-ldb' to get these functions.
3897 * New macro `strdupa' copies a string like `strdup', but uses local stack
3898 space from `alloca' instead of dynamic heap space from `malloc'.
3900 * New function `strnlen' is like `strlen' but searches only a given maximum
3901 number of characters for the null terminator. `stpncpy', `strndup' and
3902 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3905 * New function `statfs' in header <sys/statfs.h>.
3907 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3908 provide convenient functions for operating on blocks of null-terminated
3911 * A new suite of functions in <utmp.h> handle all the details of reading
3912 and writing the utmp file.
3914 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3917 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3918 completely POSIX compliant and also implemented the extended functionality
3919 to handle alternate digit representation and alternate era date formats.
3921 * Ulrich Drepper provided an implementation of the `strptime' function
3922 defined in XPG4.2 which transforms a string into a `struct tm' value.
3924 * Paul Eggert provided the tzselect shell script as part of the timezone
3925 code. The shell script makes it easy to select the correct timezone
3928 * The implementation of the malloc family of functions is completely replaced
3929 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3930 The implementation uses the mmap function (if available) and it is
3931 optimized for the use in multi threaded programs.
3933 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3934 functions. This new functionality is usable by specifying a special
3935 salt string and it is compatible with implementation on *BSD systems.
3937 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3938 `getsubopt' to handle second level command line options, `bsd_signal'
3939 to access BSD style `signal' functionality, the obsolete `regexp' style
3942 * the `lchown' function is available on system which support this
3945 * The implementation of the shadow password handling function was contributed
3948 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3950 * POSIX.1g support was added. The <sys/select.h> header is available,
3951 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3952 implementation of `getaddrinfo'.
3956 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3958 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3959 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3961 * The new function `daemon' from 4.4 BSD is useful for server programs that
3962 want to put themselves in the background.
3964 * Joel Sherrill has contributed support for several standalone boards that
3965 run without an operating system.
3967 * `printf', `scanf' and friends now accept a `q' type modifier for long
3968 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3970 * All of the code taken from BSD (notably most of the math and networking
3971 routines) has been updated from the BSD 4.4-Lite release.
3973 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3975 * The new functions `getdomainname' and `setdomainname' fetch or change the
3976 YP/NIS domain name. These are system calls which exist on systems which
3979 * The time zone data files have been updated for the latest international
3982 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3983 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3987 * The C library now includes support for Sun RPC, from Sun's free
3988 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3989 are included. (There is still no support for YP.)
3991 * Tom Quinn has contributed a port of the C library to SGI machines running
3992 Irix 4 (mips-sgi-irix4).
3994 * The new `lockf' function is a simplified interface to the locking
3995 facilities of `fcntl', included for compatibility.
3997 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3999 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4001 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4004 * The error message printed by `assert' for a failed assertion now includes
4005 the name of the program (if using GNU ld) and the name of the calling
4006 function (with versions of GCC that support this).
4008 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4010 * The library now includes the <sys/mman.h> header file and memory
4011 management functions `mmap', `munmap', `mprotect', `msync', and
4012 `madvise', on systems that support those facilities.
4014 * The interface for `mcheck' has changed slightly: the function called to
4015 abort the program when an allocation inconsistency is detected now takes
4016 an argument that indicates the type of failure. The new function
4017 `mprobe' lets you request a consistency check for a particular block at
4018 any time (checks are normally done only when you call `free' or `realloc'
4021 * It is now possible to easily cross-compile the C library, building on one
4022 system a library to run on another machine and/or operating system. All
4023 you need to do is set the variable `HOST_CC' in `configparms' to the
4024 native compiler for programs to run on the machine you are building on (a
4025 few generator programs are used on Unix systems); set `CC' to the
4028 * The new function `fexecve' (only implemented on the GNU system) executes
4029 a program file given a file descriptor already open on the file.
4033 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4034 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4036 * You can set the variable `libprefix' in `configparms' to specify a prefix
4037 to be prepended to installed library files; this makes it easy to install
4038 the GNU C library to be linked as `-lgnuc' or whatever.
4040 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4041 copies a limited number of characters from a string, and returns the
4042 address of the last character written.
4044 * You no longer need to check for whether the installed `stddef.h' is
4045 compatible with the GNU C library. configure now checks for you.
4047 * You can now define a per-stream `fileno' function to convert the
4048 stream's cookie into an integral file descriptor.
4050 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4051 allocates zero bytes of storage, and returns a unique pointer which
4052 you can pass to `realloc' or `free'. The behavior is undefined if
4053 you dereference this pointer.
4055 * The C library now runs on Sony NEWS m68k machines running either
4056 NewsOS 3 or NewsOS 4.
4058 * The new `syscall' function is a system-dependent primitive function
4059 for invoking system calls. It has the canonical behavior on Unix
4060 systems, including unreliable return values for some calls (such as
4061 `pipe', `fork' and `getppid').
4063 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4064 to `EAGAIN', which is the preferred name. On systems whose kernels
4065 use two distinct codes, the C library now translates EWOULDBLOCK to
4066 EAGAIN in every system call function.
4070 * The GNU C Library Reference Manual is now distributed with the library.
4071 `make dvi' will produce a DVI file of the printed manual.
4072 `make info' will produce Info files that you can read on line using C-h i
4073 in Emacs or the `info' program.
4074 Please send comments on the manual to bug-glibc-manual@gnu.org.
4076 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4078 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4080 * Jason Merrill has contributed a port to the Sequent Symmetry running
4081 Dynix version 3 (i386-sequent-dynix).
4083 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4084 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4086 * New function `memory_warnings' lets you arrange to get warnings when
4087 malloc is running out of memory to allocate, like Emacs gives you.
4089 * The C library now contains the relocating allocator used in Emacs 19 for
4090 its editing buffers. This allocator (ralloc) minimizes allocation
4091 overhead and fragmentation by moving allocated regions around whenever it
4092 needs to. You always refer to a ralloc'd region with a "handle" (a
4093 pointer to a pointer--an object of type `void **').
4095 * There is a new `printf' format: `%m' gives you the string corresponding
4096 to the error code in `errno'.
4098 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4099 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4100 pass, the `a' modifier says to fill in a `char **' you pass with a
4103 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4104 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4105 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4107 * `mkstemp' is a traditional Unix function to atomically create and open a
4108 uniquely-named temporary file.
4112 * The standard location for the file that says what the local timezone is
4113 has changed again. It is now `/usr/local/etc/localtime' (or more
4114 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4116 * The distribution no longer contains any files with names longer than 14
4119 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4120 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4122 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4126 * The configuration process has changed quite a bit. The `configure'
4127 script is now used just like the configuration scripts for other GNU
4128 packages. The `sysdeps' directory hierarchy is much rearranged.
4129 The file `INSTALL' explains the new scheme in detail.
4131 * The header files no longer need to be processed into ANSI C and
4132 traditional C versions. There is just one set of files to install, and
4133 it will work with ANSI or old C compilers (including `gcc -traditional').
4135 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4136 MIPS DECStation running Ultrix 4.
4138 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4139 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4140 made itself into a shared library.
4142 * Yet further improved support for the i386, running 4.3 BSD-like systems
4143 (such as Mach 3 with the Unix single-server), or System V.
4145 * New function `strncasecmp' to do case-insensitive string comparison
4146 with limited length.
4148 * New function `strsep' is a reentrant alternative to `strtok'.
4150 * New functions `scandir' and `alphasort' for searching directories.
4152 * New function `setenv' is a better interface to `putenv'.
4154 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4155 function for traversing a directory tree.
4157 * The GNU obstack package is now also part of the C library.
4158 The new function `open_obstack_stream' creates a stdio stream that
4159 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4160 formatted output directly to an obstack.
4162 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4163 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4165 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4167 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4168 things to your strings.
4170 * There are some new test programs: test-fseek, testmb, and testrand.
4172 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4173 These ports are not finished, but are a good starting place for really
4174 supporting those systems.
4176 * `/etc/localtime' is now the standard location for the file that says what
4177 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4178 This follows the general principle that `/etc' is the place for all local
4179 configuration files.
4181 * The C library header files now use `extern "C"' when used by the C++
4182 compiler, so the C library should now work with C++ code.
4184 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4185 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4188 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4189 function which uses the `merge sort' algorithm, and is said to be
4190 significantly faster than the old GNU `qsort' function. Merge sort is
4191 now the standard `qsort' function. The new algorithm can require a lot
4192 of temporary storage; so, the old sorting function is called when the
4193 required storage is not available.
4195 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4196 provides the Unix `crypt' function, plus some other entry points.
4198 * The code and header files taken from 4.4 BSD have been updated with the
4199 latest files released from Berkeley.
4201 ----------------------------------------------------------------------
4202 Copyright information:
4204 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4206 Permission is granted to anyone to make or distribute verbatim copies
4207 of this document as received, in any medium, provided that the
4208 copyright notice and this permission notice are preserved,
4209 thus giving the recipient permission to redistribute in turn.
4211 Permission is granted to distribute modified versions
4212 of this document, or of portions of it,
4213 under the above conditions, provided also that they
4214 carry prominent notices stating who last changed them.
4217 version-control: never