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 <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, sinf and
13 tan with FMA, contributed by Arjan van de Ven and H.J. Lu from Intel.
15 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
17 * Optimized generic expf, exp2f, logf, log2f, powf and sinf.
19 * In order to support faster and safer process termination the malloc API
20 family of functions will no longer print a failure address and stack
21 backtrace after detecting heap corruption. The goal is to minimize the
22 amount of work done after corruption is detected and to avoid potential
23 security issues in continued process execution. Reducing shutdown time
24 leads to lower overall process restart latency, so there is benefit both
25 from a security and performance perspective.
27 * The abort function terminates the process immediately, without flushing
28 stdio streams. Previous glibc versions used to flush streams, resulting
29 in deadlocks and further data corruption. This change also affects
30 process aborts as the result of assertion failures.
32 * On platforms where long double has the IEEE binary128 format (aarch64,
33 alpha, mips64, s390 and sparc), the math library now implements _Float128
34 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015. These
35 are the same interfaces added in version 2.26 for some platforms where
36 this format is supported but is not the format of long double.
38 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
39 mips64, powerpc64le, s390, sparc and x86_64), the math library now
40 implements interfaces for that type, as defined by ISO/IEC TS
41 18661-3:2015. These are corresponding interfaces to those supported for
44 * The math library now implements interfaces for the _Float32, _Float64 and
45 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
46 corresponding interfaces to those supported for _Float128.
48 * glibc now implements the memfd_create and mlock2 functions on Linux.
50 * Support for memory protection keys was added. The <sys/mman.h> header now
51 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
54 Deprecated and removed features, and other changes affecting compatibility:
56 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
57 defined by <sys/ptrace.h>.
59 * libm no longer supports SVID error handling (calling a user-provided
60 matherr function on error) or the _LIB_VERSION variable to control error
61 handling. (SVID error handling and the _LIB_VERSION variable still work
62 for binaries linked against older versions of the GNU C Library.) The
63 libieee.a library is no longer provided. math.h no longer defines struct
64 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
65 TLOSS, PLOSS and HUGE.
67 * The libm functions pow10, pow10f and pow10l are no longer supported for
68 new programs. Programs should use the standard names exp10, exp10f and
69 exp10l for these functions instead.
71 * The mcontext_t type is no longer the same as struct sigcontext. On
72 platforms where it was previously the same, this changes the C++ name
73 mangling for interfaces involving this type.
75 * The add-ons mechanism for building additional packages at the same time as
76 glibc has been removed. The --enable-add-ons configure option is now
79 * The --without-fp configure option is now ignored. Whether hardware
80 floating-point instructions are used is now configured based on whether
81 the compiler used at configure time (without any options implied by a
82 --with-cpu= configure option) uses such instructions.
84 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
85 the specified string can be parsed as a domain name.
87 * In the malloc_info output, the <heap> element may contain another <aspace>
88 element, "subheaps", which contains the number of sub-heaps.
90 * The libresolv function p_secstodate is no longer supported for new
93 Changes to build and runtime requirements:
95 * bison version 2.7 or later is required to generate code in the 'intl'
98 Security related changes:
100 CVE-2009-5064: The ldd script would sometimes run the program under
101 examination directly, without preventing code execution through the
102 dynamic linker. (The glibc project disputes that this is a security
103 vulnerability; only trusted binaries must be examined using the ldd
106 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
107 suffered from a one-byte overflow during ~ operator processing (either
108 on the stack or the heap, depending on the length of the user name).
109 Reported by Tim Rühsen.
111 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
112 would sometimes fail to free memory allocated during ~ operator
113 processing, leading to a memory leak and, potentially, to a denial
116 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
117 without GLOB_NOESCAPE, could write past the end of a buffer while
118 unescaping user names. Reported by Tim Rühsen.
120 CVE-2017-17426: The malloc function, when called with an object size near
121 the value SIZE_MAX, would return a pointer to a buffer which is too small,
122 instead of NULL. This was a regression introduced with the new malloc
123 thread cache in glibc 2.26. Reported by Iain Buclaw.
125 The following bugs are resolved with this release:
127 [The release manager will add the list generated by
128 scripts/list-fixed-bugs.py just before the release.]
135 * A per-thread cache has been added to malloc. Access to the cache requires
136 no locks and therefore significantly accelerates the fast path to allocate
137 and free small amounts of memory. Refilling an empty cache requires locking
138 the underlying arena. Performance measurements show significant gains in a
139 wide variety of user workloads. Workloads were captured using a special
140 instrumented malloc and analyzed with a malloc simulator. Contributed by
141 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
143 * Unicode 10.0.0 Support: Character encoding, character type info, and
144 transliteration tables are all updated to Unicode 10.0.0, using
145 generator scripts contributed by Mike FABIAN (Red Hat).
146 These updates cause user visible changes, especially the changes in
147 wcwidth for many emoji characters cause problems when emoji sequences
148 are rendered with pango, see for example:
149 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
151 * Collation of Hungarian has been overhauled and is now consistent with "The
152 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
155 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
157 - The GNU C Library will now detect when /etc/resolv.conf has been
158 modified and reload the changed configuration. The new resolver option
159 “no-reload” (RES_NORELOAD) disables this behavior.
161 - The GNU C Library now supports an arbitrary number of search domains
162 (configured using the “search” directive in /etc/resolv.conf);
163 previously, there was a hard limit of six domains. For backward
164 compatibility, applications that directly modify the ‘_res’ global
165 object are still limited to six search domains.
167 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
168 Library will now randomly pick a name server from the configuration as a
169 starting point. (Previously, the second name server was always used.)
171 * The tunables feature is now enabled by default. This allows users to tweak
172 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
174 * New function reallocarray, which resizes an allocated block (like realloc)
175 to the product of two sizes, with a guaranteed clean failure upon integer
176 overflow in the multiplication. Originally from OpenBSD, contributed by
177 Dennis Wölfing and Rüdiger Sonderfeld.
179 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
180 These are extended versions of preadv and pwritev, respectively, taking an
181 additional flags argument. The set of supported flags depends on the
182 running kernel; full support currently requires kernel 4.7 or later.
184 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
185 create a new session ID for the spawned process. This feature is
186 scheduled to be added to the next major revision of POSIX; for the time
187 being, it is available under _GNU_SOURCE.
189 * errno.h is now safe to use from C-preprocessed assembly language on all
190 supported operating systems. In this context, it will only define the
191 Exxxx constants, as preprocessor macros expanding to integer literals.
193 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
194 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
195 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
196 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
198 To compile programs that use this feature, the compiler must support
199 128-bit floating point with the type name _Float128 (as defined by TS
200 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
201 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
202 must be defined to make the new interfaces visible.
204 The new functions and macros correspond to those present for other
205 floating-point types (except for a few obsolescent interfaces not
206 supported for the new type), with F128 or f128 suffixes; for example,
207 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
208 printf or scanf formats for the new type; the strfromf128 and strtof128
209 interfaces should be used instead.
211 Deprecated and removed features, and other changes affecting compatibility:
213 * The synchronization that pthread_spin_unlock performs has been changed to
214 now be equivalent to a C11 atomic store with release memory order to the
215 spin lock's memory location. Previously, several (but not all)
216 architectures used stronger synchronization (e.g., containing what is
217 often called a full barrier). This change can improve performance, but
218 may affect odd fringe uses of spin locks that depend on the previous
219 behavior (e.g., using spin locks as atomic variables to try to implement
220 Dekker's mutual exclusion algorithm).
222 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
225 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
226 will only be built and installed when the GNU C Library is configured with
227 --enable-obsolete-rpc. This allows alternative RPC implementations, such
228 as TIRPC or rpcsvc-proto, to be used.
230 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
231 libnss_compat, are deprecated, and will not be built or installed by
234 The NIS(+) support library, libnsl, is also deprecated. By default, a
235 compatibility shared library will be built and installed, but not headers
236 or development libraries. Only a few NIS-related programs require this
237 library. (In particular, the GNU C Library has never required programs
238 that use 'gethostbyname' to be linked with libnsl.)
240 Replacement implementations based on TIRPC, which additionally support
241 IPv6, are available from <https://github.com/thkukuk/>. The configure
242 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
243 name service modules, to be built and installed.
245 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
246 support is enabled, the configured recursive resolver must support EDNS.
247 (Responding to EDNS-enabled queries with responses which are not
248 EDNS-enabled is fine, but FORMERR responses are not.)
250 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
251 servers have not supported this opcode for a long time.
253 * The _res_opcodes variable has been removed from libresolv. It had been
254 exported by accident.
256 * <string.h> no longer includes inline versions of any string functions,
257 as this kind of optimization is better done by the compiler. The macros
258 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
260 * The nonstandard header <xlocale.h> has been removed. Most programs should
261 use <locale.h> instead. If you have a specific need for the definition of
262 locale_t with no other declarations, please contact
263 libc-alpha@sourceware.org and explain.
265 * The obsolete header <sys/ultrasound.h> has been removed.
267 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
269 * The obsolete function cfree has been removed. Applications should use
272 * The stack_t type no longer has the name struct sigaltstack. This changes
273 the C++ name mangling for interfaces involving this type.
275 * The ucontext_t type no longer has the name struct ucontext. This changes
276 the C++ name mangling for interfaces involving this type.
278 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
279 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
280 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
281 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
282 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
283 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
284 fpregset_t type no longer has the name struct fpu. This changes the C++
285 name mangling for interfaces involving those types.
287 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
288 synced with the kernel:
290 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
291 are not supported on this architecture and have been removed.
293 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
294 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
295 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
297 Programs that assume the GET/SETREGS ptrace requests are universally
298 available will now fail to build, instead of malfunctioning at runtime.
300 Changes to build and runtime requirements:
302 * Linux kernel 3.2 or later is required at runtime, on all architectures
303 supported by that kernel. (This is a change from version 2.25 only for
306 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
308 * On most architectures, GCC 4.9 or later is required to build the GNU C
309 Library. On powerpc64le, GCC 6.2 or later is required.
311 Older GCC versions and non-GNU compilers are still supported when
312 compiling programs that use the GNU C Library. (We do not know exactly
313 how old, and some GNU extensions to C may be _de facto_ required. If you
314 are interested in helping us make this statement less vague, please
315 contact libc-alpha@sourceware.org.)
317 Security related changes:
319 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
320 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
322 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
323 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
325 * Avoid printing a backtrace from the __stack_chk_fail function since it is
326 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
329 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
330 fixed (CVE-2017-12133).
332 The following bugs are resolved with this release:
334 [984] network: Respond to changed resolv.conf in gethostbyname
335 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
336 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
338 [12189] libc: __stack_chk_fail should not attempt a backtrace
340 [14096] time: Race condition on timezone/tst-timezone.out
341 [14172] localedata: az_IR: new locale
342 [14995] build: glibc fails to build if gold is the default linker, even if
344 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
345 [16637] network: inet_pton function is accepting IPv6 with bad format
346 [16640] string: string/strtok.c: undefined behaviour inconsistent between
347 x86 and other generic code
348 [16875] localedata: ko_KR: fix lang_name
349 [17225] localedata: ar_SY: localized month names for May and June are
351 [17297] localedata: da_DK: wrong date_fmt string
352 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
353 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
354 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
355 [19066] localedata: ar_SA abbreviated day and month names are in English
356 [19569] network: resolv: Support an arbitrary number of search domains
357 [19570] network: Implement random DNS server selection in the stub
359 [19838] locale: localedef fails on PA-RISC
360 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
361 order of 0D36 and 0D37
362 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
364 [20098] libc: FAIL: debug/backtrace-tst on hppa
365 [20257] network: sunrpc: clntudp_call does not enforce timeout when
367 [20275] localedata: locale day/abday/mon/abmon should not have trailing
369 [20313] localedata: Update locale data to Unicode 9.0
370 [20424] manual: Document how to provide a malloc replacement
371 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
373 [20686] locale: Add el_GR@euro to SUPPORTED.
374 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
375 failures consistently
376 [21015] dynamic-link: Document and fix --enable-bind-now
377 [21016] nptl: pthread_cond support is broken on hppa
378 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
379 frame-pointer on i386
380 [21049] libc: segfault in longjmp_chk() due to clobbered processor
382 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
383 [21088] libc: Build fails with --enable-static-nss
384 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
385 generic c code is used
386 [21109] libc: Tunables broken on big-endian
387 [21112] math: powf has large ulp errors with base close to 1 and exponent
389 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
391 [21120] malloc: glibc malloc is incompatible with GCC 7
392 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
394 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
395 (-0.0) when linking with -lieee
396 [21171] math: log10, log2 and lgamma return incorrect results
397 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
398 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
399 [21207] localedata: ce_RU: update weekdays from CLDR
400 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
401 [21217] localedata: Update months from CLDR-31
402 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
403 [21243] libc: support_delete_temp_file should issue warning for failed
405 [21244] libc: support resolv_test_start() socket fd close should be
407 [21253] libc: localedef randomly segfaults when using -fstack-check due to
408 new posix_spawn implementation
409 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
410 leads to lower CPU frequency
411 [21259] libc: [alpha] termios.h missing IXANY for POSIX
412 [21261] libc: [sparc64] bits/setjmp.h namespace
413 [21267] network: [mips] bits/socket.h IOC* namespace
414 [21268] libc: [alpha] termios.h NL2, NL3 namespace
415 [21270] libc: mmap64 silently truncates large offset values
416 [21275] libc: posix_spawn always crashes on ia64 now
417 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
418 [21280] math: [powerpc] logbl for POWER7 return incorrect results
419 [21289] libc: Incorrect declaration for 32-bit platforms with
420 _FILE_OFFSET_BITS=64 causes build error
421 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
422 not support gethostbyname4_r
423 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
425 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
427 [21340] libc: Support POSIX_SPAWN_SETSID
428 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
429 [21359] network: ns_name_pack needs additional byte in destination buffer
430 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
431 against fragmentation attacks (CVE-2017-12132)
432 [21369] network: resolv: Remove EDNS fallback
433 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
435 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
436 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
437 [21393] stdio: Missing dup3 error check in freopen, freopen64
438 [21396] libc: Use AVX2 memcpy/memset on Skylake server
439 [21399] localedata: Bad description for U00EC in
440 localedata/charmaps/CP1254
441 [21411] malloc: realloc documentation error
442 [21426] network: sys/socket.h uio.h namespace
443 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
444 [21445] libc: signal.h bsd_signal namespace
445 [21455] network: Network headers stdint.h namespace
446 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
448 [21475] network: resolv: Overlong search path is truncated mid-label
449 [21511] libc: sigstack namespace
450 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
451 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
452 bits/syscall.h: No such file or directory
453 [21517] libc: struct sigaltstack namespace
454 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
455 [21533] localedata: Update locale data to Unicode 10.0
457 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
458 junk at end of line, first unrecognized character is `@'
459 [21538] libc: SIG_HOLD missing for XPG4
460 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
461 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
462 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
463 [21543] libc: sigevent namespace
464 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
465 size for general purpose registers in mcontext_t structure
466 [21550] libc: sigwait namespace
467 [21552] libc: XPG4 bsd_signal namespace
468 [21554] libc: sigpause namespace
469 [21560] libc: sys/wait.h signal.h namespace
470 [21561] libc: waitid namespace
471 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
472 '/usr/include/stdlib.h': Permission denied
473 [21575] libc: sys/wait.h missing struct rusage definition
474 [21584] libc: sigaltstack etc namespace
475 [21597] libc: siginterrupt namespace
476 [21607] math: hppa: FAIL: math/test-tgmath
477 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
479 [21622] libc: [tile] missing SA_* for POSIX.1:2008
480 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
481 stack and heap (CVE-2017-1000366)
482 [21625] libc: wait3 namespace
483 [21654] nss: Incorrect pointer alignment in NSS group merge result
485 [21657] network: Parse interface zone id for node-local multicast
486 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
488 [21666] libc: .symver is used on common symbol
489 [21668] network: resolv: res_init cross-thread broadcast introduces race
491 [21687] math: tgmath.h totalorder, totalordermag return type
492 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
494 [21696] libc: Incorrect assumption of of __cpu_mask in
495 posix/sched_cpucount.c
496 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
497 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
499 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
500 leading to relocation crash
501 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
502 `resolv_conf_matches (resp, conf)' failed.
503 [21710] localedata: Added Samoan language locale for Samoa
504 [21711] localedata: Pashto yesstr/nostr locale are missing
505 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
506 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
507 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
508 [21723] localedata: yesstr/nostr missing for Chinese language locale
509 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
511 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
513 [21728] localedata: New Locale for Tongan language
514 [21729] localedata: incorrect LC_NAME fields for hi_IN
515 [21733] localedata: yesstr and nostr are missing for zh_HK
516 [21734] localedata: Missing yesstr and nostr are for kw_GB
517 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
518 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
519 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
520 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
522 [21744] libc: Tests failing on --enable-tunables --enable-stack-
524 [21749] localedata: Wrong abbreviated day name (“abday”) for
526 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
527 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
528 [21759] localedata: missing yesstr and nostr for Tigrinya
529 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
530 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
531 [21767] localedata: Missing Bislama locales
532 [21768] localedata: Missing yesstr and nostr for aa_ET
533 [21770] localedata: Missing Field in li_NL
534 [21778] nptl: Robust mutex may deadlock
535 [21779] libc: MicroBlaze segfaults when loading libpthread
536 [21783] localedata: Fix int_select international_call_prefixes
537 [21784] localedata: Inconsistency in country_isbn
538 [21788] localedata: Missing Country Postal Abbreviations
539 [21794] localedata: Added-country_isbn-for-Italy
540 [21795] localedata: Add/Fix country_isbn for France
541 [21796] localedata: Added country_isbn for Republic of Korea
542 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
543 [21799] localedata: Added int_select international_call_prefixes
544 [21801] localedata: Added int_select international_call_prefixes
545 [21804] nptl: Double semicolon in thread-shared-types.h
546 [21807] localedata: LC_ADDRESS fix for pap_CW
547 [21808] localedata: Fix LC_ADDRESS for pap_AW
548 [21821] localedata: Added country_name in mai_IN
549 [21822] localedata: Fix LC_TIME for mai_IN
550 [21823] localedata: missing yesstr, nostr for sa_IN
551 [21825] localedata: Fix name_mrs for mag_IN
552 [21828] localedata: 2.26 changelog should mention user visible changes
554 [21835] localedata: Added Maithili language locale for Nepal
555 [21838] localedata: Removed redundant data for the_NP
556 [21839] localedata: Fix LC_MONETARY for ta_LK
557 [21844] localedata: Fix Latin characters and Months Sequence.
558 [21848] localedata: Fix mai_NP Title Name
563 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
564 24731-2:2010, is supported to enable declarations of functions from that
565 TR. Note that not all functions from that TR are supported by the GNU C
568 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
569 18661-1:2014, is supported to enable declarations of functions and macros
570 from that TS. Note that not all features from that TS are supported by
573 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
574 18661-4:2015, is supported to enable declarations of functions and macros
575 from that TS. Note that most features from that TS are not supported by
578 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
579 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
580 Since the GNU C Library defaults to a much newer revision of POSIX, this
581 will only affect programs that specifically request an old conformance
582 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
583 see a change in the visible declarations, but a program compiled with
584 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
587 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
588 defined by all multithreaded code, but glibc has not required this for
591 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
592 means that in a future release, the macros “major”, “minor”, and “makedev”
593 will only be available from <sys/sysmacros.h>.
595 These macros are not part of POSIX nor XSI, and their names frequently
596 collide with user code; see for instance glibc bug 19239 and Red Hat bug
597 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
598 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
601 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
602 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
603 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
605 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
606 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
607 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
608 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
609 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
610 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
611 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
612 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
613 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
614 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
615 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
618 * New <math.h> features are added from TS 18661-1:2014:
620 - Signaling NaN macros: SNANF, SNAN, SNANL.
622 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
623 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
624 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
626 - llogb functions: the llogb, llogbf and llogbl functions, and the
627 FP_LLOGB0 and FP_LLOGBNAN macros.
629 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
632 - Comparison macros: iseqsig.
634 - Classification macros: iscanonical, issubnormal, iszero.
636 - Total order functions: totalorder, totalorderf, totalorderl,
637 totalordermag, totalordermagf, totalordermagl.
639 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
641 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
642 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
644 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
645 are added to libc. They convert a floating-point number into string.
647 * Most of glibc can now be built with the stack smashing protector enabled.
648 It is recommended to build glibc with --enable-stack-protector=strong.
649 Implemented by Nick Alcock (Oracle).
651 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
652 intended to be used instead of memset() to erase sensitive data after use;
653 the compiler will not optimize out calls to explicit_bzero even if they
654 are "unnecessary" (in the sense that no _correct_ program can observe the
655 effects of the memory clear).
657 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
658 to float instead of double. This does not affect the ABI of any libraries
659 that are part of the GNU C Library, but may affect the ABI of other
660 libraries that use this type in their interfaces.
662 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
663 float_t and double_t types are now defined to long double instead of float
664 and double. These options are not the default, and this does not affect
665 the ABI of any libraries that are part of the GNU C Library, but it may
666 affect the ABI of other libraries that use this type in their interfaces,
667 if they are compiled or used with those options.
669 * The getentropy and getrandom functions, and the <sys/random.h> header file
672 * The buffer size for byte-oriented stdio streams is now limited to 8192
673 bytes by default. Previously, on Linux, the default buffer size on most
674 file systems was 4096 bytes (and thus remains unchanged), except on
675 network file systems, where the buffer size was unpredictable and could be
676 as large as several megabytes.
678 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
679 for the Linux quota interface which predates kernel version 2.4.22 has
682 * The malloc_get_state and malloc_set_state functions have been removed.
683 Already-existing binaries that dynamically link to these functions will
684 get a hidden implementation in which malloc_get_state is a stub. As far
685 as we know, these functions are used only by GNU Emacs and this change
686 will not adversely affect already-built Emacs executables. Any undumped
687 Emacs executables, which normally exist only during an Emacs build, should
688 be rebuilt by re-running “./configure; make” in the Emacs build tree.
690 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
691 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
692 “no-ip6-dotint” had already been the default, and support for the
693 “ip6-dotint” option was removed from the Internet in 2006.
695 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
696 flag from <resolv.h> have been removed. The option relied on a
697 backwards-incompatible DNS extension which was never deployed on the
700 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
701 RES_BLAST defined in the <resolv.h> header file have been deprecated.
702 They were already unimplemented.
704 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
705 _res.flags are deprecated. The flag was standardized in RFC 2133, but
706 removed again from the IETF name lookup interface specification in RFC
707 2553. Applications should use getaddrinfo instead.
709 * DNSSEC-related declarations and definitions have been removed from the
710 <arpa/nameser.h> header file, and libresolv will no longer attempt to
711 decode the data part of DNSSEC record types. Previous versions of glibc
712 only implemented minimal support for the previous version of DNSSEC, which
713 is incompatible with the currently deployed version.
715 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
716 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
717 <arpa/nameser.h> header file because the distinction between RR types and
718 meta-RR types is not officially standardized, subject to revision, and
719 thus not suitable for encoding in a macro.
721 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
722 and rhook members of the res_state type in <resolv.h> have been removed.
723 The glibc stub resolver did not support these hooks, but the header file
724 did not reflect that.
726 * For multi-arch support it is recommended to use a GCC which has
727 been built with support for GNU indirect functions. This ensures
728 that correct debugging information is generated for functions
729 selected by IFUNC resolvers. This support can either be enabled by
730 configuring GCC with '--enable-gnu-indirect-function', or by
731 enabling it by default by setting 'default_gnu_indirect_function'
732 variable for a particular architecture in the GCC source file
735 * GDB pretty printers have been added for mutex and condition variable
736 structures in POSIX Threads. When installed and loaded in gdb these pretty
737 printers show various pthread variables in human-readable form when read
738 using the 'print' or 'display' commands in gdb.
740 * Tunables feature added to allow tweaking of the runtime for an application
741 program. This feature can be enabled with the '--enable-tunables' configure
742 flag. The GNU C Library manual has details on usage and README.tunables has
743 instructions on adding new tunables to the library.
745 * A new version of condition variables functions have been implemented in
746 the NPTL implementation of POSIX Threads to provide stronger ordering
749 * A new version of pthread_rwlock functions have been implemented to use a more
750 scalable algorithm primarily through not using a critical section anymore to
753 Security related changes:
755 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
756 have been created with makecontext could fail to terminate due to a
757 missing .cantunwind annotation. This has been observed to lead to a hang
758 (denial of service) in some Go applications compiled with gccgo. Reported
759 by Andreas Schwab. (CVE-2016-6323)
761 * The DNS stub resolver functions would crash due to a NULL pointer
762 dereference when processing a query with a valid DNS question type which
763 was used internally in the implementation. The stub resolver now uses a
764 question type which is outside the range of valid question type values.
767 The following bugs are resolved with this release:
769 [4099] stdio: Overly agressive caching by stream i/o functions.
770 [7065] build: Support building glibc with -fstack-protector or -fstack-
772 [9842] localedata: en_CA: incorrect date format
773 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
774 before it started waiting
775 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
776 [15765] nptl: sem_open is wrongly a cancellation point
777 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
778 [16458] libc: endian.h and netinet/in.h byte order macros return values of
780 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
781 library linked with pthread
782 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
784 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
785 [17252] libc: getrandom and getentropy syscall
786 [17730] malloc: thread-local storage is sometimes improperly free()'d
787 after being __libc_memalign()'d
788 [18241] stdio: failed fseek on memstream does not set errno and can fail
790 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
792 [18463] nptl: pthread_cond_broadcast issue when surrounded by
793 PTHREAD_PRIO_INHERIT mutex on ARM
794 [18784] network: res_query and related function crash for special record
795 type queries (CVE-2015-5180)
796 [19380] math: strtod does not raise "inexact"
797 [19387] string: Integer overflow in memchr
798 [19390] string: Integer overflow in strncat
799 [19398] build: linknamespace tests fail with massively parallel build
800 [19402] nptl: Deadlock with robust shared mutex and asynchronous
802 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
803 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
804 [19514] libc: [PATCH] Fix spelling errors in spelling
805 "implement"/"implementation" in several places
806 [19582] network: Deprecate RES_USE_INET6
807 [19673] manual: clog10 docs appear to be erroneous
808 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
809 causes a segmentation fault
810 [19826] libc: invalid pointer returned from __tls_get_addr with static
812 [20016] network: resolv: Remove hooks support from the API
813 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
815 [20033] math: [x86_64] vectorized math function don't call the __finite
817 [20116] nptl: use after free in pthread_create
818 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
819 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
821 [20311] nptl: please install proc_service.h
822 [20366] libc: Compilation errors in installed headers in strict-compliance
824 [20370] malloc: malloc: Arena free list management is still racy
825 (incorrect fix in bug 19243)
826 [20386] libc: assert (X = 0) does not result in GCC warning
827 [20432] malloc: malloc: Minimize interface required for interposition
828 [20435] libc: Missing unwind info in __startcontext causes infinite loop
829 in _Unwind_Backtrace (CVE-2016-6323)
830 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
831 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
833 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
834 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
835 [20477] network: resolv: incorrect double-checked locking related to
837 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
839 [20495] math: x86_64 performance degradation due to AVX/SSE transition
841 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
842 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
843 cause transition penalty
844 [20517] math: sparcv9 missing fdiml compat symbol
845 [20524] manual: strverscmp is inconsistent
846 [20525] libc: <sys/quota.h> should be based on kernel headers
847 [20539] math: GCC 7 -static -lm fails to link at -O3
848 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
850 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
852 [20591] network: Remove obsolete DNSSEC support
853 [20592] network: DNS resource record type classification macros in
854 <arpa/nameser.h> are incorrect
855 [20593] network: Update DNS RR type definitions
856 [20611] network: getaddrinfo accepts invalid numeric scope IDs
857 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
859 [20629] network: libresolv: Remove support for bitstring labels
861 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
862 assertion: "maxidx >= 2"
863 [20660] math: [arm] Use VSQRT
864 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
865 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
866 [20689] libc: Test for FMA should also check for AVX.
867 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
868 [20715] math: iszero macro breaks existing code
869 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
870 [20728] libc: powerpc: Missing TOC stub in clone
871 [20729] build: glibc-2.24 fails to build for i486 with -Os
872 [20750] build: Build fails with default PIE enabled gcc-6.x
873 [20768] math: [alpha] sqrt fegetenv namespace
874 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
876 [20787] math: float_t is defined as float incorrectly on x86_64 even if
877 __FLT_EVAL_METHOD__ is 2
878 [20790] network: rpcgen buffer overrun in get_prog_declaration
879 [20822] nptl: race condition in __lll_unlock_elision on powerpc
880 [20829] libc: crypt snprintf namespace
881 [20847] libc: tst-vfork3 failure
882 [20855] math: Default bits/mathdef.h has inappropriate float_t
883 [20859] math: [sh4] FP_ILOGB0 invalid
884 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
885 U+20AC), not same as GBK
886 [20915] dynamic-link: global-dynamic TLS broken on aarch64
887 [20916] math: pow handling of sNaN arguments
888 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
889 [20919] math: Bad pow (qNaN, 0) result with -lieee
890 [20940] math: hypot sNaN handling
891 [20947] math: fmax, fmin sNaN handling
892 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
893 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
895 [20971] string: powerpc64/power7 memchr overflows internal pointer check
896 [20973] nptl: robust mutexes: Lost wake-ups
897 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
898 [20978] nis: strlen on null pointer in nss_nisplus
899 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
901 [21014] string: i686 memchr overflows internal pointer check
902 [21019] libc: [mips] n32 lseek incorrect on overflow
903 [21022] libc: [microblaze] __backtrace get_frame_size namespace
904 [21026] libc: [MIPS] readahead syscall is broken on n64
905 [21028] math: Fallback fesetexceptflag should always succeed
906 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
907 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
909 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
910 [21061] librt: [microblaze] librt lost clock_* exports
911 [21073] libc: tunables: insecure environment variables passed to
912 subprocesses with AT_SECURE
913 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
917 * The minimum Linux kernel version that this version of the GNU C Library
918 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
919 kernel version 2.6.32 or later suffices (on architectures that already
920 required kernel versions more recent than 3.2, those requirements remain
921 unchanged). Linux 3.2 or later kernel headers are required on all
924 * The pap_AN locale has been deleted. This has been deprecated for a long
925 time. It has been replaced by pap_AW & pap_CW, both of which have long
926 been included in previous releases.
928 * The readdir_r and readdir64_r functions have been deprecated. It is
929 recommended to use readdir and readdir64 instead.
931 * The type “union wait” has been removed. It was deprecated in the early
932 1990s and never part of POSIX. Application code should use the int type
933 instead of “union wait”.
935 * A new NSS action is added to facilitate large distributed system
936 administration. The action, MERGE, allows remote user stores like LDAP
937 to be merged into local user stores like /etc/groups in order to provide
938 easy to use, updated, and managed sets of merged credentials. The new
939 action can be used by configuring it in /etc/nsswitch.conf:
940 group: files [SUCCESS=merge] nis
941 Implemented by Stephen Gallagher (Red Hat).
943 * The deprecated __malloc_initialize_hook variable has been removed from the
946 * The long unused localedef --old-style option has been removed. It hasn't
947 done anything in over 16 years. Scripts using this option can safely
950 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
951 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
952 return the next representable value in the direction of positive infinity
953 and the nextdown functions return the next representable value in the
954 direction of negative infinity. These are currently enabled as GNU
957 Security related changes:
959 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
960 could result in a stack overflow when getnetbyname was called with an
961 overly long name. (CVE-2016-3075)
963 * Previously, getaddrinfo copied large amounts of address data to the stack,
964 even after the fix for CVE-2013-4458 has been applied, potentially
965 resulting in a stack overflow. getaddrinfo now uses a heap allocation
966 instead. Reported by Michael Petlan. (CVE-2016-3706)
968 * The glob function suffered from a stack-based buffer overflow when it was
969 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
970 Reported by Alexander Cherepanov. (CVE-2016-1234)
972 * The Sun RPC UDP client could exhaust all available stack space when
973 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
974 alloca plugin for GCC. (CVE-2016-4429)
976 * The IPv6 name server management code in libresolv could result in a memory
977 leak for each thread which is created, performs a failing naming lookup,
978 and exits. Over time, this could result in a denial of service due to
979 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
981 The following bugs are resolved with this release:
983 [1170] localedata: ne_NP: update Nepali locale definition file
984 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
986 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
987 [6796] math: fdim() does not set errno on overflow
988 [10354] libc: posix_spawn should use vfork() in more cases than presently
989 [11213] localedata: localedata: add copyright disclaimer to locale files
990 [12143] localedata: chr_US: new Cherokee locale
991 [12450] localedata: sgs_LT: new locale
992 [12676] localedata: ln_CD: new locale
993 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
995 [13304] math: fma, fmaf, fmal produce wrong results
996 [14259] build: --localedir arg to configure is ignored
997 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
998 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
999 [14934] localedata: es_CL: wrong first weekday chilean locale
1000 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1002 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1004 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1005 [15368] nptl: raise() is not async-signal-safe
1006 [15479] math: ceil, floor, round and trunc raise inexact exception
1007 [15578] localedata: kk_KZ: various updates
1008 [16003] localedata: pap_AN: punt old locale
1009 [16137] localedata: iw_IL: punt old locale
1010 [16190] localedata: eo: new esperanto locale
1011 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1012 [16742] malloc: race condition: pthread_atfork() called before first
1013 malloc() results in unexpected locking behaviour/deadlocks
1014 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1016 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1017 [17565] localedata: pt_PT: wrong (work-)week start
1018 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1019 [17950] build: Build fails with -msse
1020 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1021 [18433] libc: posix_spawn does not return correctly upon failure to
1023 [18453] localedata: charmaps/IBM875: incorrect codes
1024 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1026 [18896] localedata: he_IL: improvements for currency
1027 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1028 Romanian locale data
1029 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1031 [19056] libc: Deprecate readdir_r
1032 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1034 [19198] localedata: nl_NL: small improvements for Dutch locales
1035 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1036 nameservers (CVE-2016-5417)
1037 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1039 [19400] locale: Language missing in "iso-639.def", trivial fix in
1041 [19431] malloc: Deadlock between fflush, getdelim, and fork
1042 [19505] libc: Incorrect file descriptor validity checks in
1043 posix_spawn_file_actions_add{open,close,dup2}
1044 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1045 when using RTLD_NEXT
1046 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1047 `intl/{gettextP,loadinfo}.h'
1048 [19534] libc: execle, execlp may use malloc
1049 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1050 [19573] network: res_nclose and __res_maybe_init disagree about name
1051 server initialization, breaking Hesiod
1052 [19575] localedata: Status of GB18030 tables
1053 [19581] localedata: sr_* date_fmt string contains additional newline
1054 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1056 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1057 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1058 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1059 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1061 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1063 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1065 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1066 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1067 [19633] locale: strfmon_l applies global locale to number formatting
1068 [19642] network: Memory leak in getnameinfo
1069 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1070 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1072 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1073 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1075 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1076 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1078 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1080 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1082 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1083 pointers and lengths in error-case.
1084 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1085 report errors on UTF-16 surrogates values.
1086 [19755] nscd: nscd assertion failure in gc
1087 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1088 [19759] libc: mempcpy shouldn't be inlined
1089 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1090 [19765] libc: s390 needs an optimized mempcpy
1091 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1092 NAME_MAX limit assumption (CVE-2016-1234)
1093 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1094 [19787] network: Missing and incorrect truncation checks in getnameinfo
1095 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1097 [19791] network: Assertion failure in res_query.c with un-connectable name
1099 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1100 [19822] math: libm.so install clobbers old version
1101 [19825] network: resolv: send_vc can return uninitialized data in second
1102 response to getaddrinfo
1103 [19830] network: nss_dns: should check RDATA length against buffer length
1104 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1105 confronted with A/AAAA records of invalid size
1106 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1107 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1108 [19853] stdio: Printing IBM long double in decimal with high precision is
1110 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1111 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1112 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1114 [19865] network: Assertion failure or memory leak in
1115 _nss_dns_getcanonname_r
1116 [19868] network: nss_dns: netent code does not skip over non-PTR records
1117 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1119 [19881] string: Improve x86-64 memset
1120 [19907] string: Incorrect memcpy tests
1121 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1123 [19925] libc: termios.h XCASE namespace
1124 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1125 [19929] libc: limits.h NL_NMAX namespace
1126 [19931] stdio: Memory leak in vfprintf
1127 [19957] libc: clone(CLONE_VM) access invalid parent memory
1128 [19963] localedata: en_IL: New locale
1129 [19989] stdio: stdio.h cuserid namespace
1130 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1132 [19996] locale: langinfo.h nl_langinfo_l namespace
1133 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1134 [20010] network: getaddrinfo: Stack overflow in hostent translation
1136 [20012] stdio: libio: fmemopen append mode failure
1137 [20014] stdio: stdio.h namespace for pre-threads POSIX
1138 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1139 [20023] libc: fcntl.h timespec namespace
1140 [20024] math: [x86_64] vectorized sincos trashes the stack
1141 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1142 [20041] time: sys/time.h timespec namespace
1143 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1144 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1145 [20051] libc: ttyslot in wrong header under wrong conditions
1146 [20054] libc: gethostname not declared for XPG4
1147 [20055] libc: termios.h missing tcgetsid for XPG4
1148 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1150 [20073] libc: sys/stat.h fchmod namespace
1151 [20074] libc: stdlib.h rand_r namespace
1152 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1153 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1155 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1156 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1158 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1159 [20119] libc: Wrong mask for processors level type from CPUID
1160 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1161 AS not supporting AVX512
1162 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1164 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1165 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1167 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1168 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1169 [20178] libc: posix_spawn{p} should not call exit
1170 [20191] stdio: libio: vtables hardening
1171 [20195] string: FMA4 detection requires CPUID execution with register
1173 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1174 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1176 [20212] math: acos (sNaN) returns sNaN
1177 [20213] math: asin (sNaN) returns sNaN
1178 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1179 [20218] math: [i386] asinhl (sNaN) returns sNaN
1180 [20219] math: [i386] atanhl (sNaN) returns sNaN
1181 [20222] stdio: fopencookie: Mangle function pointers
1182 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1183 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1184 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1186 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1187 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1188 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1189 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1190 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1191 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1192 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1193 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1194 [20240] math: modf (sNaN) returns sNaN
1195 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1196 [20250] math: frexp (sNaN) returns sNaN
1197 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1198 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1200 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1201 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1202 subscript is below array bounds [-Werror=array-bounds]
1203 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1204 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1206 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1207 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1209 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1210 "invalid" exceptions
1211 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1212 [20316] localedata: id_ID: Februari instead of Pebruari
1213 [20327] string: POWER8 strcasecmp returns incorrect result
1214 [20347] math: Failure: Test: j0_downward (0xap+0)
1215 [20348] libc: FAIL: misc/tst-preadvwritev64
1216 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1217 [20350] libc: There is no test for p{read,write}64
1218 [20357] math: Incorrect cos result for 1.5174239687223976
1219 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1223 * Unicode 8.0.0 Support: Character encoding, character type info, and
1224 transliteration tables are all updated to Unicode 8.0.0, using new
1225 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1226 These updates cause user visible changes, such as the fixes for bugs
1227 89, 16061, and 18568.
1229 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1230 kernel-internal CPU set size. This means that requests that change the
1231 CPU affinity which failed before (for example, an all-ones CPU mask) will
1232 now succeed. Applications that need to determine the effective CPU
1233 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1234 setting it because the kernel can adjust it (and the previous size check
1235 would not detect this in the majority of cases).
1237 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1238 the following new symbols are used: fts64_children, fts64_close,
1239 fts64_open, fts64_read and fts64_set.
1241 * getaddrinfo now detects certain invalid responses on an internal netlink
1242 socket. If such responses are received, an affected process will
1243 terminate with an error message of "Unexpected error <number> on netlink
1244 descriptor <number>" or "Unexpected netlink response of size <number> on
1245 descriptor <number>". The most likely cause for these errors is a
1246 multi-threaded application which erroneously closes and reuses the netlink
1247 file descriptor while it is used by getaddrinfo.
1249 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1250 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1251 unnecessary serialization of memory allocation requests across threads.
1252 The defect is now corrected. Users should see a substantial increase in
1253 the concurent throughput of allocation requests for applications which
1254 trigger this bug. Affected applications typically create and
1255 destroy threads frequently. (Bug 19048 was reported and analyzed by
1258 * There is now a --disable-timezone-tools configure option for disabling the
1259 building and installing of the timezone related utilities (zic, zdump, and
1260 tzselect). This is useful for people who build the timezone data and code
1261 independent of the GNU C Library.
1263 * The obsolete header <regexp.h> has been removed. Programs that require
1264 this header must be updated to use <regex.h> instead.
1266 * The obsolete functions bdflush, create_module, get_kernel_syms,
1267 query_module and uselib are no longer available to newly linked binaries;
1268 the header <sys/kdaemon.h> has been removed. These functions and header
1269 were specific to systems using the Linux kernel and could not usefully be
1270 used with the GNU C Library on systems with version 2.6 or later of the
1273 * Optimized string, wcsmbs and memory functions for IBM z13.
1274 Implemented by Stefan Liebler.
1276 * Newly linked programs that define a variable called signgam will no longer
1277 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1278 require signgam to be set by those functions must ensure that they use the
1279 variable provided by the GNU C Library and declared in <math.h>, without
1280 defining their own copy.
1282 * The minimum GCC version that can be used to build this version of the GNU
1283 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1284 still be used to compile programs using the GNU C Library.
1286 Security related changes:
1288 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1289 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1291 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1292 the pointer guard feature. It is always enabled. Previously,
1293 LD_POINTER_GUARD could be used to disable security hardening in binaries
1294 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1297 * An integer overflow in hcreate and hcreate_r could lead to an
1298 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1300 * The catopen function no longer has unbounded stack usage. Reported by
1301 Max. (CVE-2015-8779)
1303 * The nan, nanf and nanl functions no longer have unbounded stack usage
1304 depending on the length of the string passed as an argument to the
1305 functions. Reported by Joseph Myers. (CVE-2014-9761)
1307 * A stack-based buffer overflow was found in libresolv when invoked from
1308 libnss_dns, allowing specially crafted DNS responses to seize control
1309 of execution flow in the DNS client. The buffer overflow occurs in
1310 the functions send_dg (send datagram) and send_vc (send TCP) for the
1311 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1312 family. The use of AF_UNSPEC triggers the low-level resolver code to
1313 send out two parallel queries for A and AAAA. A mismanagement of the
1314 buffers used for those queries could result in the response of a query
1315 writing beyond the alloca allocated buffer created by
1316 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1317 the overflow. Thanks to the Google Security Team and Red Hat for
1318 reporting the security impact of this issue, and Robert Holiday of
1319 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1321 The following bugs are resolved with this release:
1323 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1324 [887] math: Math library function "logb" and "nextafter" inconsistent
1325 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1326 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1327 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1328 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1330 [4404] localedata: German translation of "Alarm clock" is misleading
1331 [6799] math: nextafter() and nexttoward() doen't set errno on
1332 overflow/underflow errors
1333 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1335 [10432] nis: _nss_nis_setnetgrent assertion failure
1336 [11460] libc: fts has no LFS support
1337 [12926] network: getaddrinfo()/make_request() may spin forever
1338 [13065] nptl: Race condition in pthread barriers
1339 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1340 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1342 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1343 [14912] libc: Rename non-installed bits/*.h headers
1344 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1345 [15367] math: Let gcc use __builtin_isinf
1346 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1347 [15421] math: lgamma wrongly sets signgam for ISO C
1348 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1349 FE_INVALID with argument out of range
1350 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1352 [15786] dynamic-link: ifunc resolver functions can smash function
1354 [15918] math: Unnecessary check for equality in hypotf()
1355 [16061] localedata: Review / update transliteration data
1356 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1358 [16141] time: strptime %z offset restriction
1359 [16171] math: drem should be alias of remainder
1360 [16296] math: fegetround is pure?
1361 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1362 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1364 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1365 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1367 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1368 "Exception "Inexact" set" on ppc32
1369 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1370 [16517] math: Missing underflow exception from tanf/tan/tanl
1371 [16519] math: Missing underflow exception from sinhf
1372 [16520] math: Missing underflow exception from tanhf
1373 [16521] math: Missing underflow exception from exp2
1374 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1376 [16734] stdio: fopen calls mmap to allocate its buffer
1377 [16961] math: nan function incorrect handling of bad sequences
1378 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1379 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1380 [16985] locale: localedef: confusing error message when opening output
1382 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1383 [17197] locale: Redundant shift character in iconv conversion output at
1385 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1386 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1387 [17250] dynamic-link: static linking breaks nss loading
1388 (getaddrinfo/getpwnam/etc...)
1389 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1391 [17441] math: isnan() should use __builtin_isnan() in GCC
1392 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1393 (related to lock elision)
1394 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1395 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1396 [17887] time: strptime should be able to parse "+01:00" style timezones
1397 [17905] libc: catopen() Multiple unbounded stack allocations
1399 [18084] libc: backtrace (..., 0) dumps core on x86
1400 [18086] libc: nice() sets errno to 0 on success
1401 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1402 is too large (CVE-2015-8778)
1403 [18251] dynamic-link: SONAME missing when audit modules provides path
1404 [18265] libc: add attributes for wchar string and memory functions
1405 [18370] math: csqrt missing underflows
1406 [18421] libc: [hppa] read-only segment has dynamic relocations
1407 [18472] libc: Obsolete syscall wrappers should be compat symbols
1408 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1409 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1410 [18525] localedata: Remove locale timezone information
1411 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1412 [18568] localedata: Update locale data to Unicode 8.0
1413 [18589] locale: sort-test.sh fails at random
1414 [18595] math: ctan, ctanh missing underflows
1415 [18604] libc: assert macro-expands its argument
1416 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1417 contains a vector instruction exception.
1418 [18611] math: j1, jn missing errno setting on underflow
1419 [18618] localedata: sync Chechen locale definitions with other *_RU
1421 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1422 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1423 [18665] network: In send_dg, the recvfrom function is NOT always using the
1424 buffer size of a newly created buffer (CVE-2015-7547)
1425 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1427 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1429 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1430 [18699] math: tilegx cproj() for various complex infinities does not yield
1432 [18724] libc: Harden put*ent functions against data injection
1433 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1434 [18755] build: build errors with -DNDEBUG
1435 [18757] stdio: fmemopen fails to set errno on failure
1436 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1438 [18781] libc: openat64 lacks O_LARGEFILE
1439 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1440 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1441 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1442 [18790] math: [ldbl-128ibm] tanhl inaccurate
1443 [18795] libc: stpncpy fortification misses buffer lengths that are
1444 statically too large
1445 [18796] build: build fails for --disable-mathvec
1446 [18803] math: hypot missing underflows
1447 [18820] stdio: fmemopen may leak memory on failure
1448 [18823] math: csqrt spurious underflows
1449 [18824] math: fma spurious underflows
1450 [18825] math: pow missing underflows
1451 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1452 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1453 [18870] build: sem_open.c fails to compile with missing symbol
1455 [18872] stdio: Fix memory leak in printf_positional
1456 [18873] libc: posix_fallocate overflow check ineffective
1457 [18875] math: Excess precision leads incorrect libm
1458 [18877] libc: arm: mmap offset regression
1459 [18887] libc: memory corruption when using getmntent on blank lines
1460 [18918] localedata: hu_HU: change time to HH:MM:SS format
1461 [18921] libc: Regression: extraneous stat() and fstat() performed by
1463 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1464 binaries (CVE-2015-8777)
1465 [18951] math: tgamma missing underflows
1466 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1468 [18953] localedata: lt_LT: change currency symbol to the euro
1469 [18956] math: powf inaccuracy
1470 [18961] math: [i386] exp missing underflows
1471 [18966] math: [i386] exp10 missing underflows
1472 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1473 [18969] build: multiple string test failures due to missing locale
1475 [18970] libc: Reference of pthread_setcancelstate in libc.a
1476 [18977] math: float / long double Bessel functions not in XSI POSIX
1477 [18980] math: i386 libm functions return with excess range and precision
1478 [18981] math: i386 scalb*, ldexp return with excess range and precision
1479 [18982] stdio: va_list and vprintf
1480 [18985] time: Passing out of range data to strftime() causes a segfault
1482 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1483 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1484 [19012] locale: iconv_open leaks memory on error path
1485 [19016] math: clog, clog10 inaccuracy
1486 [19018] nptl: Mangle function pointers in tls_dtor_list
1487 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1488 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1489 [19048] malloc: malloc: arena free list can become cyclic, increasing
1491 [19049] math: [powerpc] erfc incorrect zero sign
1492 [19050] math: [powerpc] log* incorrect zero sign
1493 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1494 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1495 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1496 [19074] network: Data race in _res_hconf_reorder_addrs
1497 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1498 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1499 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1501 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1503 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1504 [19086] manual: posix_fallocate64 documented argument order is wrong.
1505 [19088] math: lround, llround missing exceptions close to overflow
1507 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1508 [19095] math: dbl-64 lrint incorrect for 64-bit long
1509 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1510 [19124] dynamic-link: ld.so failed to build with older assmebler
1511 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1512 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1513 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1514 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1515 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1516 pthread_setaffinity_np
1517 [19156] math: [ldbl-128] j0l spurious underflows
1518 [19164] nptl: tst-getcpu fails with many possible CPUs
1519 [19168] math: math/test-ildoubl and math/test-ldouble failure
1520 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1521 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1523 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1524 (FE_NOMASK_ENV) do not clear SSE exceptions
1525 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1526 _int_new_arena/reused_arena
1527 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1528 [19201] math: dbl-64 remainder incorrect sign of zero result
1529 [19205] math: bits/math-finite.h conditions do not match math.h and
1531 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1532 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1533 for C99-based standards
1534 [19212] libc: features.h not -Wundef clean
1535 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1537 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1538 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1539 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1541 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1542 "inexact" exceptions
1543 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1545 [19242] libc: strtol incorrect in Turkish locales
1546 [19243] malloc: reused_arena can pick an arena on the free list, leading
1547 to an assertion failure and reference count corruption
1548 [19253] time: tzset() ineffective when temporary TZ did not include DST
1550 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1551 [19270] math: [hppa] Shared libm missing __isnanl
1552 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1553 MAP_HUGETLB and MAP_STACK defines
1554 [19313] nptl: Wrong __cpu_mask for x32
1555 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1556 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1557 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1558 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1559 [19363] time: x32: times() return value wrongly truncates/sign extends
1561 [19367] dynamic-link: Improve branch prediction on Silvermont
1562 [19369] network: Default domain name not reset by res_ninit when "search"
1563 / "domain" entry is removed from resolv.conf
1564 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1566 [19385] localedata: bg_BG: time separator should be colon, not comma
1567 [19408] libc: linux personality syscall wrapper may erroneously return an
1568 error on 32-bit architectures
1569 [19415] libc: dladdr returns wrong names on hppa
1570 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1571 IBM905, IBM907, and IBM909
1572 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1573 [19443] build: build failures with -DDEBUG
1574 [19451] build: Make check fails on test-double-vlen2
1575 [19462] libc: Glibc failed to build with -Os
1576 [19465] math: Wrong code with -Os
1577 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1579 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1581 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1582 [19486] math: S390: Math tests fail with "Exception Inexact set".
1583 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1584 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1586 [19590] math: Fail to build shared objects that use libmvec.so functions.
1590 * The following bugs are resolved with this release:
1592 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1593 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1594 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1595 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1596 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1597 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1598 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1599 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1600 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1601 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1602 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1603 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1604 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1605 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1606 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1607 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1608 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1609 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1610 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1611 18657, 18676, 18694, 18696, 18887.
1613 * Cache information can be queried via sysconf() function on s390 e.g. with
1614 _SC_LEVEL1_ICACHE_SIZE as argument.
1616 * A buffer overflow in gethostbyname_r and related functions performing DNS
1617 requests has been fixed. If the NSS functions were called with a
1618 misaligned buffer, the buffer length change due to pointer alignment was
1619 not taken into account. This could result in application crashes or,
1620 potentially arbitrary code execution, using crafted, but syntactically
1621 valid DNS responses. (CVE-2015-1781)
1623 * The time zone file parser has been made more robust against crafted time
1624 zone files, avoiding heap buffer overflows related to the processing of
1625 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1626 large time zone data files. Overly long time zone specifiers in the TZ
1627 variable no longer result in stack overflows and crashes.
1629 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1630 for LD and GD on x86 and x86-64, has been implemented. You will need
1631 binutils-2.24 or later to enable this optimization.
1633 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1634 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1635 Hat). These updates cause user visible changes, such as the fix for bug
1638 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1639 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1640 condition in some applications.
1642 * Added vector math library named libmvec with the following vectorized x86_64
1643 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1645 The library can be disabled with --disable-mathvec. Use of the functions is
1646 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1647 Shared library libmvec.so is linked in as needed when using -lm (no need to
1648 specify -lmvec explicitly for not static builds).
1649 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1651 * A new fmemopen implementation has been added with the goal of POSIX
1652 compliance. The new implementation fixes the following long-standing
1653 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1654 old implementation is still present for use be by existing binaries.
1656 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1657 releases. It has been fixed to match 2.19 and older, but binaries built
1658 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1660 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1661 Contributed by Roland McGrath (Google).
1663 * The header <regexp.h> is deprecated, and will be removed in a future
1664 release. Use of this header will trigger a deprecation warning.
1665 Application developers should update their code to use <regex.h> instead.
1667 This header was formerly part of SUS, but was deprecated in 1994 and
1668 removed from the standard in 2001. Also, the glibc implementation
1669 leaks memory. See BZ#18681 for more details.
1673 * The following bugs are resolved with this release:
1675 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1676 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1677 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1678 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1679 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1680 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1681 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1682 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1683 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1684 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1687 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1688 too little memory for the to-be-scanned arguments and overflow the
1689 allocated buffer. The implementation now correctly computes the required
1690 buffer size when using malloc, and switches to malloc from alloca as
1693 * A new semaphore algorithm has been implemented in generic C code for all
1694 machines. Previous custom assembly implementations of semaphore were
1695 difficult to reason about or ensure that they were safe. The new version
1696 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1697 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1698 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1700 * Port to Altera Nios II has been contributed by Mentor Graphics.
1702 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1703 implementations for powerpc64/powerpc64le.
1704 Implemented by Adhemerval Zanella (IBM).
1706 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1707 and powerpc64le. This may improve lock scaling of existing programs on
1708 HTM capable systems. The lock elision code is only enabled with
1709 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1710 powerpc will issue a transaction abort on every syscall to avoid side
1711 effects being visible outside transactions.
1713 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1714 AArch64. Contributed by ARM Ltd.
1716 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1718 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1719 under certain input conditions resulting in the execution of a shell for
1720 command substitution when the application did not request it. The
1721 implementation now checks WRDE_NOCMD immediately before executing the
1722 shell and returns the error WRDE_CMDSUB as expected.
1724 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1725 processing format strings with a large number of format specifiers.
1727 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1728 infinite loop if the DNS response contained a PTR record of an unexpected
1731 * The minimum GCC version that can be used to build this version of the GNU
1732 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1733 still be used to compile programs using the GNU C Library.
1735 * The GNU C Library is now built with -Werror by default. This can be
1736 disabled by configuring with --disable-werror.
1738 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1740 * The obsolete sigvec function has been removed. This was the original
1741 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1742 programs have been using instead for about 25 years. Of course, ABI
1743 compatibility for old binaries using sigvec remains intact.
1745 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1746 with newer versions of bison.
1748 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1749 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1750 registers overlay two consecutive single-precision registers. MIPS32R2
1751 introduced a new FPU mode (FR=1) where double-precision registers extend the
1752 corresponding single-precision registers which is incompatible with the
1753 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1754 require the use of FR=1 making a transition necessary. New o32 ABI
1755 extensions enable users to migrate over time from the original o32 ABI
1756 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1757 tracks the ABI of any loaded object and verifies that new objects are
1758 compatible. Mode transitions will also be requested as required and
1759 unsupportable objects will be rejected. The ABI checks include both soft and
1760 hard float ABIs for o32, n32 and n64.
1762 GCC 5 with GNU binutils 2.25 onwards:
1763 It is strongly recommended that all o32 system libraries are built using the
1764 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1765 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1766 compiler using --with-fp-32=xx to set this by default.
1770 * The following bugs are resolved with this release:
1772 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1773 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1774 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1775 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1776 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1777 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1778 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1779 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1780 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1781 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1782 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1783 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1784 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1785 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1786 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1788 * Reverted change of ABI data structures for s390 and s390x:
1789 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1790 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1791 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1792 are preserved pointing straight to the same implementation as the old ones.
1793 Given that, new callers will simply provide a too-big buffer to these
1794 functions. Any applications/libraries out there that embed jmp_buf or
1795 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1796 against 2.19 headers will have to rebuilt again. This is necessary in any
1797 case to revert the breakage in their ABI caused by the glibc change.
1799 * Support for file description locks is added to systems running the
1800 Linux kernel. The standard file locking interfaces are extended to
1801 operate on file descriptions, not file descriptors, via the use of
1802 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1803 are associated with an open file instead of a process.
1805 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1807 * The minimum Linux kernel version that this version of the GNU C Library
1808 can be used with is 2.6.32.
1810 * Running the testsuite no longer terminates as soon as a test fails.
1811 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1812 with PASS or FAIL lines for individual tests. A summary of the results is
1813 printed, including a list of failing lists, and "make check" exits with
1814 error status if there were any unexpected failures. "make check
1815 stop-on-test-failure=y" may be used to keep the old behavior.
1817 * The am33 port, which had not worked for several years, has been removed
1820 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1821 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1822 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1823 conflicted with POSIX (support for which was removed in 2.19), the
1824 interfaces those macros enabled remain available when compiling with
1825 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1826 test macros defined.
1828 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1830 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1831 This may improve lock scaling of existing programs on TX capable systems.
1832 The lock elision code is only built with --enable-lock-elision=yes and
1833 then requires a GCC version supporting the TX builtins. With lock elision
1834 default mutexes are elided via __builtin_tbegin, if the cpu supports
1835 transactions. By default lock elision is not enabled and the elision code
1838 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1839 copy the path argument. This allowed programs to cause posix_spawn to
1840 deference a dangling pointer, or use an unexpected pathname argument if
1841 the string was modified after the posix_spawn_file_actions_addopen
1844 * All supported architectures now use the main glibc sysdeps directory
1845 instead of some being in a separate "ports" directory (which was
1846 distributed separately before glibc 2.17).
1848 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1849 On configurations that support it (all Linux configurations), it's now
1850 used regardless of the --enable-add-ons switch to configure. It is no
1851 longer possible to build such configurations without pthreads support.
1853 * Locale names, including those obtained from environment variables (LANG
1854 and the LC_* variables), are more tightly checked for proper syntax.
1855 setlocale will now fail (with EINVAL) for locale names that are overly
1856 long, contain slashes without starting with a slash, or contain ".." path
1857 components. (CVE-2014-0475) Previously, some valid locale names were
1858 silently replaced with the "C" locale when running in AT_SECURE mode
1859 (e.g., in a SUID program). This is no longer necessary because of the
1862 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1863 with application code using Intel MPX instructions. (With all previous
1864 versions, the MPX register state could be clobbered when making calls
1865 into or out of a shared library.) Note that while the new dynamic
1866 linker is compatible with all known x86 hardware whether or not it
1867 supports Intel MPX, some x86 instruction-set emulators might fail to
1868 handle the new instruction encodings. This is known to affect Valgrind
1869 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1870 release), and might affect other tools that do instruction emulation.
1872 * Support for loadable gconv transliteration modules has been removed.
1873 The support for transliteration modules has been non-functional for
1874 over a decade, and the removal is prompted by security defects. The
1875 normal gconv conversion modules are still supported. Transliteration
1876 with //TRANSLIT is still possible, and the //IGNORE specifier
1877 continues to be supported. (CVE-2014-5119)
1879 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1880 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1881 resulting a denial-of-service security vulnerability in applications which
1882 use functions related to iconv. (CVE-2014-6040)
1886 * The following bugs are resolved with this release:
1888 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1889 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1890 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1891 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1892 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1893 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1894 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1895 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1896 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1897 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1898 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1899 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1900 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1901 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1902 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1903 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1904 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1905 16431, 16453, 16474, 16506, 16510, 16529
1907 * Slovenian translations for glibc messages have been contributed by the
1908 Translation Project's Slovenian team of translators.
1910 * The public headers no longer use __unused nor __block. This change is to
1911 support compiling programs that are derived from BSD sources and use
1912 __unused internally, and to support compiling with Clang's -fblock
1913 extension which uses __block.
1915 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1916 large collation sequences to optimize multiple passes. This cache
1917 computation may overflow for large collation sequences and may cause a
1918 stack or buffer overflow. This is now fixed to use a slower algorithm
1919 which does not use a cache if there is an integer overflow.
1921 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1922 rules for large collation sequences to optimize multiple passes and falls
1923 back to alloca if malloc fails, resulting in a possible stack overflow.
1924 The implementation now falls back to an uncached collation sequence lookup
1927 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1928 initialized for static applications resulting in the security feature
1929 being disabled. The pointer guard is now correctly initialized to a
1930 random value for static applications. Existing static applications need
1931 to be recompiled to take advantage of the fix (bug 15754).
1933 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1934 to the d_name member of struct dirent, or omit the terminating NUL
1935 character. (Bugzilla #14699).
1937 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1938 aligned_alloc functions could allocate too few bytes or corrupt the
1939 heap when passed very large allocation size values (Bugzilla #15855,
1942 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1943 for AF_INET6 has been fixed (Bugzilla #16072).
1945 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1946 pap_CW, quz_PE, the_NP.
1948 * Substantially revised locales: gd_GB, ht_HT
1950 * The LC_ADDRESS field was updated to support country_car for almost all
1953 * ISO 1427 definitions were updated.
1955 * ISO 3166 definitions were updated.
1957 * The localedef utility now supports --big-endian and --little-endian
1958 command-line options to generate locales for a different system from that
1959 for which the C library was built.
1961 * Binary locale files now only depend on the endianness of the system for
1962 which they are generated and not on other properties of that system. As a
1963 consequence, binary files generated with new localedef may be incompatible
1964 with old versions of the GNU C Library, and binary files generated with
1965 old localedef may be incompatible with this version of the GNU C Library,
1966 in the following circumstances:
1968 + Locale files may be incompatible on m68k systems.
1970 + Locale archive files (but not separate files for individual locales) may
1971 be incompatible on systems where plain "char" is signed.
1973 * The configure option --disable-versioning has been removed. Builds with
1974 --disable-versioning had not worked for several years.
1976 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1977 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1979 * SystemTap probes for malloc have been introduced.
1981 * SystemTap probes for slow multiple precision fallback paths of
1982 transcendental functions have been introduced.
1984 * Support for powerpc64le has been added.
1986 * The soft-float powerpc port now supports e500 processors.
1988 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1990 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1991 set of header declarations that are enabled by default, even when other
1992 feature test macros or compiler options such as -std=c99 would otherwise
1993 disable some of those declarations.
1995 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1996 conflict with POSIX. The libbsd-compat library (which was a dummy library
1997 that did nothing) has also been removed.
1999 * Preliminary documentation about Multi-Thread, Async-Signal and
2000 Async-Cancel Safety has been added.
2002 * Change of ABI data structures for s390 and s390x:
2003 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2004 allow for future hardware extensions. All library functions that accept or
2005 return these structures were versioned in order to provide backwards
2006 compatibility with existing code. However, not all cases can be handled
2007 with interface versioning. If an application embeds either structure into
2008 another structure and that structure is passed to another compilation unit
2009 which expects the newer larger-sized object then any access to the new
2010 fields will result in undefined behaviour. Similarly any access to fields
2011 that were moved by the enlarging of the structures will cause undefined
2012 behaviour. To fix the undefined behaviour all objects that are part of the
2013 final in-memory image must agree on the size of structures, and this may
2014 require recompilation.
2018 * The following bugs are resolved with this release:
2020 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2021 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2022 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2023 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2024 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2025 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2026 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2027 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2028 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2029 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2030 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2031 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2032 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2035 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2036 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2037 Distributions can re-enable building and using pt_chown via the new configure
2038 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2039 considerable security risks and should only be used if the distribution
2040 understands and accepts the risks.
2042 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2045 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2046 fixed (Bugzilla #15330).
2048 * Add support for calling C++11 thread_local object destructors on thread
2049 and program exit. This needs compiler support for offloading C++11
2050 destructor calls to glibc.
2052 * Improved worst case performance of libm functions with double inputs and
2055 * Support for priority inherited mutexes in pthread condition variables on
2056 non-x86 architectures.
2058 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2060 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2062 * Optimized string functions for ARM. Implemented by Will Newton and
2065 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2067 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2070 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2071 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2073 * Added a benchmark framework to track performance of functions in glibc.
2075 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2076 It is based on draft TS 18661 and currently enabled as a GNU extension.
2078 * On Linux, the clock function now uses the clock_gettime system call
2079 for improved precision, rather than old times system call.
2081 * Added support for version-3 tz format files. This is needed when using
2082 the tz database release 2013e or later, and affects a few unusual cases --
2083 currently only TZ='America/Godthab' for time stamps after 2037.
2085 * Added new API functions pthread_getattr_default_np and
2086 pthread_setattr_default_np to get and set the default pthread
2087 attributes of a process.
2089 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2090 This may improve lock scaling of existing programs on TSX capable systems.
2091 When the --enable-lock-elision=yes parameter is specified at configure
2092 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2095 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2096 availability). Implemented by Ryan S. Arnold.
2098 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2102 * The following bugs are resolved with this release:
2104 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2105 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2106 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2107 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2108 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2109 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2110 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2111 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2112 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2113 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2114 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2115 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2116 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2117 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2118 14879, 14889, 14893, 14898, 14914.
2120 * Optimization of memcpy for MIPS.
2122 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2123 EMFILE has been fixed (Bugzilla #14889).
2125 * The add-on ports collection is now distributed in the "ports" subdirectory
2126 of the main GNU C Library distribution, rather than separately.
2128 * Port to ARM AArch64 contributed by Linaro.
2130 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2131 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2133 Implemented by Andreas Krebbel.
2135 * The new function secure_getenv allows secure access to the environment,
2136 returning NULL if running in a SUID/SGID process. This function replaces
2137 the internal function __secure_getenv.
2139 * SystemTap static probes have been added into the dynamic linker.
2140 Implemented by Gary Benson.
2142 * Optimizations of string functions strstr, strcasestr and memmem.
2143 Implemented by Maxim Kuvyrkov.
2145 * The minimum Linux kernel version that this version of the GNU C Library
2146 can be used with is 2.6.16.
2148 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2149 powerpc POWER7. Implemented by Will Schmidt.
2151 * New configure option --disable-nscd builds the C library such that it
2152 never attempts to contact the Name Service Caching Daemon (nscd).
2153 New configure option --disable-build-nscd avoids building nscd itself;
2154 this is the default if --disable-nscd is used.
2156 * Improved support for cross-compilation, including cross-testing and
2157 bootstrap builds without a previously built glibc.
2159 * Several testsuite tests are now able to test multiple IFUNC variants of an
2160 interface, rather than just testing the one that would be chooen by
2163 * New configure options --with-bugurl and --with-pkgversion, for
2164 distributors to use to embed their bug-reporting and package version
2165 information in --help and --version output.
2167 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2168 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2169 allows creation of chroots without the procfs mounted on /proc.
2171 * The `crypt' function now fails if passed salt bytes that violate the
2172 specification for those values. On Linux, the `crypt' function will
2173 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2174 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2175 when the mode is enabled.
2177 * The `clock_*' suite of functions (declared in <time.h>) is now available
2178 directly in the main C library. Previously it was necessary to link with
2179 -lrt to use these functions. This change has the effect that a
2180 single-threaded program that uses a function such as `clock_gettime' (and
2181 is not linked with -lrt) will no longer implicitly load the pthreads
2182 library at runtime and so will not suffer the overheads associated with
2183 multi-thread support in other code such as the C++ runtime library.
2185 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2190 * The following bugs are resolved with this release:
2192 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2193 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2194 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2195 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2196 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2197 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2198 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2199 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2200 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2201 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2202 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2203 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2204 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2205 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2206 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2207 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2208 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2209 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2210 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2211 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2212 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2213 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2216 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2217 configuring glibc with:
2218 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2219 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2220 Implemented by H.J. Lu.
2224 + define static_assert
2226 + do not declare gets
2228 + declare at_quick_exit and quick_exit also for ISO C11
2230 + aligned_alloc. NB: The code is deliberately allows the size parameter
2231 to not be a multiple of the alignment. This is a moronic requirement
2232 in the standard but it is only a requirement on the caller, not the
2235 + timespec_get added
2237 + uchar.h support added
2239 + CMPLX, CMPLXF, CMPLXL added
2241 Implemented by Ulrich Drepper.
2243 * Support for the IA-64 has been moved to ports.
2245 * Remove support for anything but ELF binary format
2247 * Checking versions of poll, ppoll added.
2248 Implemented by Ulrich Drepper.
2250 * More generic and 64-bit performance optimizations to math functions.
2251 Implemented by Ulrich Drepper.
2253 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2254 headers and functions available at compile time as they were before
2255 version 2.14. This option will be removed at some time in the future
2256 after the TI-RPC library becomes fully sufficient for the needs of
2257 existing applications.
2259 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2260 Note that glibc is not expected to work with any Linux kernel version
2263 * New header <sys/auxv.h> and function getauxval allowing easy access to
2264 the AT_* key-value pairs passed from the Linux kernel. The header also
2265 defines the HWCAP_* bits associated with the AT_HWCAP key.
2267 * New locales: mag_IN
2269 * New configure option --enable-systemtap builds SystemTap static probes
2270 into libc for setjmp and longjmp and into libpthread for various operations.
2271 So far the setjmp/longjmp probes and some of the libpthread probes are
2272 provided only for i*86 and x86_64.
2273 Implemented by Roland McGrath and Rayson Ho.
2275 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2277 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2280 * More optimized functions for SPARC. Implemented by David S. Miller.
2282 * Improved support for cross-compilation, especially bootstrap builds
2283 without a previously built glibc.
2285 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2286 Chris Metcalf from Tilera.
2288 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2289 now supported for ARM processors.
2291 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2292 as the name of the dynamic linker, to distinguish it from the
2293 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2295 * Support for CRIS has been removed from ports.
2297 * A new class of installed header has been documented for low-level
2298 platform-specific functionality. PowerPC added the first instance with a
2299 function to provide time base register access. Contributed by Tulio
2300 Magno Quites Machado Filho.
2302 * ix86 configurations now install header files that are consistent with
2303 what x86-64 configurations install. These same header files can be used
2304 for -m32, -m64, or -mx32 builds.
2305 Contributed by H.J. Lu.
2307 * Math library bug fixes. A thorough audit of all open math library bugs was
2308 conducted by Joseph Myers. Significant progress was made on many math
2309 library bugs resulting in more accurate exceptions and function results.
2310 Many thanks to all those that contributed including Andreas Jaeger for his
2311 patch review and work on the x87 trigonometric instruction issues.
2313 * Timezone data is no longer installed. Timezone-related binaries and scripts
2314 will continue to be installed. Users should obtain their timezone data from
2315 their distribution provider or from the tzdata package at
2316 <ftp://munnari.oz.au/pub/>.
2320 * The following bugs are resolved with this release:
2322 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2323 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2324 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2325 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2326 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2327 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2328 13484, 13506, 13515, 13523, 13524, 13538, 13540
2330 * New program pldd to list loaded object of a process
2331 Implemented by Ulrich Drepper.
2333 * Add nss_db support back to glibc. No more dependency on Berkeley db
2334 and support for initgroups lookups.
2335 Implemented by Ulrich Drepper.
2337 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2338 Contributed by HJ Lu.
2340 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2341 Contributed by HJ Lu.
2343 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2344 on x86-32 and x86-64.
2345 Contributed by Liubov Dmitrieva.
2347 * Optimized strchr and strrchr for SSE on x86-32.
2348 Contributed by Liubov Dmitrieva.
2350 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2351 for x86-64 and x86-32.
2352 Contributed by Liubov Dmitrieva.
2354 * New interfaces: scandirat, scandirat64
2355 Implemented by Ulrich Drepper.
2357 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2358 Implemented by Ulrich Drepper.
2360 * nscd now also caches the netgroup database.
2361 Implemented by Ulrich Drepper.
2363 * Integrate libm with gcc's -ffinite-math-only option.
2364 Implemented by Ulrich Drepper.
2366 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2367 to math functions. Implemented by Ulrich Drepper.
2369 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2370 Implemented by Ulrich Drepper.
2372 * New Linux interfaces: process_vm_readv, process_vm_writev
2374 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2375 Implemented by Ulrich Drepper.
2377 * Optimized nearbyint and strcasecmp for PPC.
2378 Implemented by Adhemerval Zanella.
2380 * New locales: bho_IN, unm_US, es_CU, ta_LK
2384 * The following bugs are resolved with this release:
2386 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2387 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2388 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2389 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2390 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2391 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2392 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2393 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2394 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2395 12795, 12811, 12813, 12814, 12841
2397 * The RPC implementation in libc is obsoleted. Old programs keep working
2398 but new programs cannot be linked with the routines in libc anymore.
2399 Programs in need of RPC functionality must be linked against TI-RPC.
2400 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2402 Visible changes of this change include (obviously) the inability to link
2403 programs using RPC functions without referencing the TI-RPC library and the
2404 removal of the RPC headers from the glibc headers.
2405 Implemented by Ulrich Drepper.
2407 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2408 syncfs, setns, sendmmsg
2410 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2411 yue_HK, lij_IT, mhr_RU
2413 * New encodings: CP770, CP771, CP772, CP773, CP774
2415 * New program sotruss to trace calls through PLTs
2416 Implemented by Ulrich Drepper.
2418 * The malloc hook implementation is marked deprecated and will be removed
2419 from the default implementation in the next version. The design never
2420 worked ever since the introduction of threads. Even programs which do
2421 not create threads themselves can use multiple threads created internally.
2425 * The following bugs are resolved with this release:
2427 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2428 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2429 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2430 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2433 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2435 * POWER7 optimizations: memset, memcmp, strncmp
2437 * New optimized string functions for x86-64: strnlen (SSE2),
2438 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2439 Implemented by Ulrich Drepper.
2443 * The following bugs are resolved with this release:
2445 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2446 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2447 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2448 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2449 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2450 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2451 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2452 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2454 * New interfaces: pthread_getname_np, pthread_setname_np
2456 * New Linux interface: recvmmsg
2458 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2460 * The dynamic linker now recognizes supported ABI versions from the
2461 EI_ABIVERSION field in the ELF header.
2462 Implemented by Ulrich Drepper.
2464 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2465 will not be used to fill in password fields in the passwd.byname replies.
2466 Instead it is used to synthesize the shadow.byname table, should it be
2467 missing. This is a useful mode in some installations involving Solaris.
2468 Implemented by Ulrich Drepper.
2470 * New locales: kok_IN, sq_MK, cv_RU
2474 * The following bugs are resolved with this release:
2476 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2477 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2478 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2479 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2480 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2481 10789, 10817, 10823, 10840, 10847
2483 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2485 Implemented by Ulrich Drepper.
2487 * Checking version of longjmp added that fails if an uninitialized stack
2488 frame would be created. Implemented by Ulrich Drepper.
2490 * STT_GNU_IFUNC is now supported in static executables.
2491 Implemented by H.J. Lu.
2493 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2495 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2496 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2497 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2498 Contributed by H.J. Lu.
2500 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2501 Implemented by Ulrich Drepper.
2503 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2505 Contributed by H.J. Lu.
2507 * Support for fma instruction in AVX on x86-64.
2508 Implemented by H.J. Lu and Ulrich Drepper.
2510 * AVX support in x86-64 auditing support in ld.so.
2511 Implemented by H.J. Lu.
2513 * STB_GNU_UNIQUE support added.
2514 Implemented by Ulrich Drepper.
2516 * Implement second fallback mode for optimized DNS lookup for even more
2517 broken environments. If two requests from the same port are not
2518 handled correctly close the socket and open a new one before sending
2519 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2520 can be used to select this mode right away, instead of rediscovering the
2521 necessity is every process again.
2522 Implemented by Ulrich Drepper.
2524 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2525 Implemented by Adam Tkac.
2527 * Optimized iconv conversions for S390x.
2528 Implemented by Andreas Krebbel.
2530 * Using condvars with PI mutexes is now more efficient due to kernel
2531 support for requeueing to PI futexes. NPTL support added for x86-64.
2532 Implemented by Ulrich Drepper.
2534 * New locales: ps_AF, my_MM
2538 * The following bugs are resolved with this release:
2540 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2541 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2542 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2543 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2544 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2546 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2547 Implemented by Ulrich Drepper.
2549 * New Linux interfaces: accept4, fallocate, fallocate64.
2550 Implemented by Ulrich Drepper.
2552 * Correct declarations of string function when used in C++ code. This
2553 could lead to compile errors for invalid C++ code.
2555 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2558 * New POSIX 2008 interface: psiginfo
2559 Implemented by Ulrich Drepper.
2561 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2562 Implemented by Ulrich Drepper.
2564 * Support for selecting between multiple function definitions at runtime
2565 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2567 * The libcrypt library can now use the hash function implementations in
2568 NSS. Implemented by Ulrich Drepper.
2570 * The malloc implementation can be compiled to be less memory efficient
2571 but higher performing in multi-threaded programs.
2572 Implemented by Ulrich Drepper.
2574 * New locales: nan_TW@latin, ks_IN
2576 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2577 Implemented by Ulrich Drepper.
2579 * Extended printf hook support. It is possible to use user-defined types
2580 and extend existing format specifiers.
2581 Implemented by Ulrich Drepper.
2583 * Handling for group shadow files has been added.
2584 Implemented by Ulrich Drepper.
2586 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2587 servers (the case, e.g., for some people using the built-in DNS
2588 server in ADSL modems/routers). There is a once-per-process timeout
2589 in case of a broken server. To avoid it, users can run nscd or put
2590 'options single-request' in /etc/resolv.conf.
2591 Implemented by Ulrich Drepper.
2595 * The following bugs are resolved with this release:
2597 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2598 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2599 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2600 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2601 6968, 6974, 6980, 6995, 7008, 7009, 7029
2603 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2604 up at the same time. Implemented by Ulrich Drepper.
2606 * TLS descriptors for LD and GD on x86 and x86-64.
2607 Implemented by Alexandre Oliva.
2609 * getaddrinfo now handles DCCP and UDPlite.
2610 Implemented by Ulrich Drepper.
2612 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2613 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2614 Implemented by Ulrich Drepper.
2616 * New implementation of memmem, strstr, and strcasestr which is O(n).
2617 Implemented by Eric Blake.
2619 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2621 * Implement "e" option for popen to open file descriptor with the
2622 close-on-exec flag set. Implemented by Ulrich Drepper.
2624 * Implement "b" mode for fmemopen. In this mode writes writes don't
2625 implicitly add a NUL byte and seeks from the end of the buffer really
2626 use the buffer end, not the string length as the basis.
2627 Implemented by Ulrich Drepper.
2629 * Many functions, exported and internal, now atomically set the close-on-exec
2630 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2632 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2634 Implemented by Pravin Satpute.
2636 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2640 * The following bugs are resolved with this release:
2642 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2643 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2644 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2645 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2646 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2647 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2648 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2650 * New locales: bo_CN, bo_IN, shs_CA.
2652 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2654 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2655 Implemented by Pravin Satpute.
2657 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2659 * nscd caches now all timeouts for DNS entries
2660 Implemented by Ulrich Drepper.
2662 * nscd is more efficient and wakes up less often.
2663 Implemented by Ulrich Drepper.
2665 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2666 vdprintf, and obstack_vprintf.
2667 Implemented by Jakub Jelinek.
2669 * Faster memset for x86-64.
2670 Implemented by Harsha Jagasia and H.J. Lu.
2672 * Faster memcpy on x86.
2673 Implemented by Ulrich Drepper.
2675 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2676 Implemented by Ulrich Drepper.
2678 * Faster sqrt and sqrtf implemention for some PPC variants.
2679 Implemented by Steven Munroe.
2683 * The following bugs are resolved with this release:
2685 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2686 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2687 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2688 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2689 5071, 5103, 5104, 5112, 5113, 5184, 5186
2691 * More checking functions: fread, fread_unlocked, open*, mq_open.
2692 Implemented by Jakub Jelinek and Ulrich Drepper.
2694 * Extend fortification to C++. Implemented by Jakub Jelinek.
2696 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2697 by not recognizing 'a' as a modifier when those specs are requested.
2698 Implemented by Jakub Jelinek.
2700 * PPC optimizations to math and string functions.
2701 Implemented by Steven Munroe.
2703 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2704 options to be passed. Implemented by Ulrich Drepper.
2706 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2708 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2709 Implemented by Ulrich Drepper.
2711 * Handle private futexes in the NPTL implementation.
2712 Implemented by Jakub Jelinek and Ulrich Drepper.
2714 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2715 Implemented by Roland McGrath and Ulrich Drepper.
2717 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2719 * SHA-256 and SHA-512 based password encryption.
2720 Implemented by Ulrich Drepper.
2722 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2723 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2726 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2727 Implemented by Ulrich Drepper.
2731 * The following bugs are resolved with this release:
2733 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2734 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2735 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2736 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2737 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2738 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2739 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2742 * New Linux interfaces: epoll_pwait, sched_getcpu.
2744 * New generic interfaces: strerror_l.
2746 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2750 * The following bugs are resolved with this release:
2752 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2753 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2754 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2755 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2756 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2757 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2758 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2759 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2760 3225, 3251, 3252, 3253, 3273
2762 * For Linux, the sorting of addresses returned by getaddrinfo now also
2763 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2764 Implemented by Ulrich Drepper.
2766 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2767 Implemented by Ulrich Drepper.
2769 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2771 * New iconv module for MIK. Contributed by Alexander Shopov.
2773 * For sites with broken group and/or passwd database, the auto-propagate
2774 option of nscd can prevent creating ID lookup entries from the results
2775 of a name lookup and vice versa. This usually is no problem but some
2776 site might have problems with the default behavior.
2777 Implemented by Ulrich Drepper.
2779 * Iterating over entire database in NIS can be slow. With the
2780 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2781 to trade time for memory. The entire database will be read at once.
2782 Implemented by Ulrich Drepper.
2784 * The interfaces introduced in RFC 3542 have been implemented by
2787 * Support for the new ELF hash table format was added by Ulrich Drepper.
2789 * Support for priority inheritance mutexes added by Jakub Jelinek and
2792 * Support for priority protected mutexes added by Jakub Jelinek.
2794 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2795 or_IN, csb_PL, fy_NL, sr_ME.
2799 * More overflow detection functions.
2801 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2802 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2804 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2805 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2806 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2807 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2808 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2809 by Masahide Washizawa.
2811 * It is now possible to install audit libraries for the dynamic linker using
2812 LD_AUDIT. Implemented by Ulrich Drepper.
2814 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2815 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2816 For a libc and libpthread that works well on Linux 2.4 kernels, we
2817 recommend using the stable 2.3 branch.
2819 * The new function open_wmemstream acts like open_memstream,
2820 but uses a wchar_t wide character buffer.
2822 * The new function ppoll is an improved version of poll, similar to pselect.
2824 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2825 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2826 renameat, unlinkat, linkat, symlinkat, readlinkat.
2828 * New Linux kernel system calls: unshare,
2829 inotify_init, inotify_add_watch, inotify_rm_watch.
2831 * The euidaccess function is now also known by the alias eaccess,
2832 for compatibility with some other systems.
2834 * Timezone data updated to 2006b version.
2838 * The following bugs are resolved with this release:
2840 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2841 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2842 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2843 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2844 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2845 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2847 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2849 * As of this release, GCC 4 can be used to compile the C Library.
2851 * Timezone data updated to 2005m version.
2855 * The following bugs are resolved with this release:
2857 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2858 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2859 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2860 777, 787, 821, 822, 823, 825
2862 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2866 * Support for RFC 3678. Real implementations exist only for Linux so far.
2867 Implemented by Ulrich Drepper.
2869 * nscd can now cache entries persistently. Expiring entries are reloaded.
2870 For speedups the cache can be shared in memory with client processes.
2871 Implemented by Ulrich Drepper.
2873 * nscd can now perform SELinux checks.
2874 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2876 * getaddrinfo queries are now cached. Canonical name lookup is performed
2878 Implemented by Ulrich Drepper.
2880 * The nothrow function attribute is used when headers are used by gcc when
2881 compiling C code. This can avoid significant amounts of exception
2884 * The malloc functions perform more error checking and are stricter when
2885 it comes to reacting on errors. The default action is to terminate
2886 the process after showing an error message. Implemented by Ulrich Drepper.
2888 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2889 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2891 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2892 namespaces using the new function dlmopen(). This feature is of course,
2893 like most other dynamic loading functionality, not available in statically
2894 linked applications. Implemented by Ulrich Drepper.
2896 * Low-overhead boundary checking variants of string and some stdio functions
2897 were added. These are to be used in conjunction with a gcc patch by
2898 Jakub Jelinek which adds calls to these functions if possible.
2899 Implemented by Jakub Jelinek and Ulrich Drepper.
2901 * Old code for several operating systems and machine architectures that
2902 have not been in working condition in a long time have been removed from
2903 the main source tree maintained by the GNU C Library's maintainers.
2904 These files are now reside in the separate `ports' source module
2905 that is usable as an add-on when building the library.
2909 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2910 interrogate the dynamic linker, compatible with the Solaris interface.
2912 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2913 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2915 * getifaddrs now uses the netlink interface on Linux to get its information.
2916 Implemented by Thorsten Kukuk.
2918 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2919 Implemented by Ulrich Drepper.
2921 * support for non-executable stacks on x86 has been added. Changes mostly
2924 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2927 * getaddrinfo now performs destination address selection according to
2932 * Thread-safe interfaces for many functions that access locale data
2933 were added in version 2.3, but these features were omitted from NEWS.
2934 Many functions have variants with an `_l' suffix that take a `locale_t'
2935 object as a parameter rather than consulting the current locale.
2936 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2937 create and maintain `locale_t' objects. Additionally, the new function
2938 `uselocale' sets "the current locale" (as used by functions not so
2939 parameterized) set for an individual thread. These features were added
2940 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2942 * The functions getresuid, getresgid, setresuid, and setresgid, which
2943 have long been available on Linux, are now declared in <unistd.h>
2944 and are now also available on the Hurd.
2946 * ELF thread-local storage support (TLS) now works on x86-64.
2948 * The new dynamic string token $LIB is expanded in shared library names.
2949 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2951 * Aldy Hernandez contributed complete software floating point support for
2952 PowerPC machines with no FPU.
2954 * fexecve is implemented on Linux.
2956 * The `btowc' function should work at least twice as fast due to
2957 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2959 * With approriate thread add-ons cancelable functions are now implemented
2960 in libc.so as well. No need to call the function in libpthread. This
2961 change allowed to finally disable the incorrect and expensive handling
2962 of weak definition in ld.so.
2964 * Yet more PLT entries in libc.so have been removed. We finally arrived
2965 at the bare minimum. Startup times improved appropriately.
2967 * Support for the new Linux/x86 system call interface was added. The
2968 AT_SYSINFO auxiliary vector entry is recognized and handled.
2972 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2975 * iconv (the program and the interface) now accepts empty names (excluding
2976 options like //TRANSLIT) to mean "use charset of current locale".
2978 * localedef can now transliterate characters in strings which are not in
2979 the provided charmap. The information from the input locale is used.
2981 * Prelinking support was added for ELF targets. This requires additional
2982 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2984 * Read-only stdio streams now use mmap to speed up operation by eliminating
2985 copying and buffer underflows. To use add 'm' to the mode string of
2986 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2988 * The malloc functions were completely rewritten by Wolfram Gloger based
2989 on Doug Lea's malloc-2.7.0.c.
2991 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2992 implementation of regex.
2994 * Bruno Haible upgraded the iconv and locale implementation to support
2997 * Contents of the LC_* and LANG environment variables in the CEN style are
2998 not recognized anymore. It never was used. Change by Ulrich Drepper.
3000 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3001 thread-local storage (TLS) ABI on some platforms.
3002 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3004 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3005 EUC-JISX0213, and TSCII.
3007 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3008 BSD-compatible interface for getting all network interface addresses.
3009 Implementation for IPv4 by Roland McGrath.
3011 * Loading of locale data is faster due to the introduction of a locale
3012 archive. Implemented by Roland McGrath and Ulrich Drepper.
3014 * Startup times are significantly reduced by not using exported functions
3015 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3018 * Steven Munroe contributed a port to PowerPC64/Linux.
3022 * The Hurd now uses the GNU libio implementation of stdio rather than the
3023 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3025 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3026 with the same behavior as the Linux system call of the same name.
3030 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3031 128-bit long double format.
3033 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3034 IBM1161, and IBM1162 charsets.
3036 * Andreas Jaeger contributed a port to x86-64/Linux.
3038 * Peter Bruin contributed a port to PowerPC/Hurd.
3040 * libc functions using I/O streams now can handle wide-oriented streams
3043 * optimizations in the dynamic linker. Binaries created by recent binutils
3044 versions start up quicker due to reduced time spent on relocations.
3046 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3050 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3051 asinh, atanh, j0 for the 128-bit long double format.
3053 * Bruno Haible updated all the code handling Unicode in some form to
3054 support Unicode 3.1.
3056 * Speed of regex for single-byte locales is back to previous levels.
3057 Patch by Isamu Hasegawa.
3059 * Alpha, SPARC, and IA-64 now also using floating stacks.
3061 * Startup time of internationalized applications greatly improved through
3062 iconv cache. Use iconvconfig to generate the cache file.
3063 Contributed by Ulrich Drepper.
3065 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3066 severe performance problems. Patch by David Mosberger.
3068 * The Hurd port got a lot more functionality like AIO, various stdio
3069 extensions, etc. Mainly done by Roland McGrath.
3071 * mtrace can now lookup symbols in shared libraries.
3075 * Intel's IA-64 math library is largely integrated. It provides fast and
3076 accurate implementatations for most basic and standard math functions
3077 in float, double, and long double format.
3079 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3080 and asin for the 96-bit long double format and asin, log, tan for the
3081 128-bit long double format.
3083 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3084 The basic double functions exist today. Contributed by Abraham Ziv
3085 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3086 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3088 * An asynchronous name lookup library was added. The interface is designed
3089 after POSIX AIO. The proposal was circulated beforehand to get comments.
3090 No negative ones came in. Implemented by Ulrich Drepper.
3092 * Port to S390/64bit contributed by Martin Schwidefsky
3093 <schwidefsky@de.ibm.com>.
3095 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3096 of functions for Linux/IA-64.
3098 * The RPC code is now thread safe. Threads can now use the same service
3099 of different services at the same time. Patch by Eric Norum
3100 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3102 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3103 family of functions for Linux/S390.
3105 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3106 of functions for Linux/x86.
3108 * Port to Linux/CRIS contributed by Axis Communications.
3112 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3113 we can now check for namespace violations and missing declarations. The
3114 result is that almost all headers are now Unix-compliant (as defined in
3115 the upcoming XPG6). The negative side is that some programs might need
3116 corrections, too, if they depend on the incorrect form of the headers in
3117 previous versions which defined too many symbols and included too many
3120 * regex now handles multibyte character sets correctly.
3121 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3123 * iconv (the program) does now conform to the upcoming XPG6 and handles
3124 charmaps. Instead of the charset names the path of charmaps can be
3125 provided and the conversion happens based on this data.
3126 Contributed by Ulrich Drepper.
3128 * The locale program now provides detailed information about the installed
3133 only lists the names of the supported locales
3137 provides details such as country, language, and codeset name.
3138 Contributed by Ulrich Drepper.
3142 * The gencat program now parses the input file according to the charset
3143 selected by the LC_CTYPE category. This is important for stateful
3144 character sets. To make generating catalogs easier there is a way
3145 to overwrite the charset selected by the locale: before the first
3146 message or $ quote line the catalog can contain a line like
3148 $ codeset=ISO-8859-2
3150 to select the charset (ISO-8859-2 in this case).
3152 Implemented by Shinya Hanataka and Ulrich Drepper.
3154 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3155 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3156 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3158 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3159 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3161 * The LANGUAGE environment variable is now ignored unless the locale is
3162 changed from the default "C" locale.
3164 * The usual bug fixes.
3168 * Greg McGary added runtime support for bounds checking using gcc's
3169 new -fbounded-pointers option. ix86 target is complete. PowerPC
3172 * Thorsten Kukuk added secure mode support to nscd.
3174 * The Berkeley db libraries have been removed.
3176 Related, the nss_db module is now in a separate package since it
3177 obviously requires a database library being available.
3179 * Wide character I/O streams implemented by Ulrich Drepper.
3181 * Functions from the extended socket API added by Ulrich Drepper.
3183 * Functions feenableexcept and fedisableexcept to control the
3184 behaviour of individual exceptions have been added by Andreas Jaeger.
3186 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3188 * The resolver code has been updated from bind 8.2.3-T5B which supports
3189 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3192 This change could in some situations effect backward compatibility. Since
3193 now `_res' is a thread-local instead of a global variable, modifying it
3194 in one thread does not have any effect in other threads.
3196 The resolver library was also extended to allow IPv6 as the transport
3197 protocol for the requests. This work was done by Stig Venaas.
3199 * Compatibility code for K&R C compilers has been removed from the
3200 header files. A ISO C compiler is needed to use the library
3201 (conforming to either C89 or C99 standard).
3203 * Complete rewrite of the localedef program to support multibyte character
3204 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3205 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3206 Implemented by Ulrich Drepper.
3208 Bruno Haible significantly improved the generation and use of the data
3209 structures for the wide character tables.
3211 * Plural handling in gettext implemented by Ulrich Drepper.
3213 * The utmp daemon has been removed.
3215 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3217 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3220 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3222 * POSIX barriers implemented by Kaz Kylheku.
3224 * POSIX spawn function family implemented by Ulrich Drepper.
3226 * POSIX spinlocks are now available.
3228 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3230 * the configure option --enable-kernel=X.Y.Z allows to strip out
3231 compatibility for kernel versions before X.Y.Z. This is currently only
3232 implemented for Linux.
3234 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3235 field sin6_scope_id. This means that all programs using IPv6 should be
3236 recompiled. Don't expect binary compatibility with previous glibc
3239 * various conversion modules for IBM character sets contributed by
3242 * IA-64 port by Jes Sorensen and HJ Lu.
3256 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3258 * New xdr functions are added; some rpc functions are now 64bit clean.
3260 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3262 * Fixed known incompatibilities with glibc 2.0.
3264 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3266 * Optimized a number of functions (especially the ELF dynamic loader).
3268 * Update timezone data files.
3270 * lots of charmaps corrections
3272 * some new locale definitions and charmaps
3277 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3278 conform to POSIX member type requirements. Symbol versions have been
3279 adjusted as needed within the library, and for direct use by applications,
3280 but there is potential for problems if third-party libraries use
3281 struct timeval as part of their interface. This does not present
3282 a problem for X and other "essential" system libraries.
3284 * An additional locale model to support C++ Standard Library locale
3285 model and probably more was implemented by Ulrich Drepper.
3287 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3290 * Miles Bader provided the `argp' function family to support hierachical
3291 command line argument parsing, layered on top of getopt.
3293 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3295 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3298 * scanf recognizes the %a and %A format for scanning floating point numbers.
3300 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3301 information and interfaces for the available integer types.
3303 * about 130 new math functions were added to implement the ISO C9x math
3306 * the new header <complex.h> contains definitions of the complex math
3307 functions from ISO C 9X.
3309 * the new header <tgmath.h> defines generic macros to use complex or
3310 real valued functions.
3312 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3314 * Andreas Jaeger provided a test suite for the math library.
3316 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3318 * Ulrich Drepper added character set conversion functions (iconv).
3320 * Optimized string functions have been added.
3322 * The localedata addon is now part of glibc.
3324 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3326 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3327 daemon for NSS (nscd).
3329 Missing a better place here are some numbers on improvements. Under
3330 Linux 2.1.125 un-tar-ing the kernel sources takes
3334 using local files 12.19s 6.88s 22.91s
3336 using NIS 13.92s 8.91s 26.34s
3338 using NIS & nscd 10.37s 7.34s 25.30s
3340 using NIS+ 27.57s 30.37s 640.46s
3342 using NIS+ & nscd 10.25s 7.83s 26.51s
3344 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3346 Keep in mind that non-namelookup related operations dominate above times.
3347 It was just a common complain that using NIS+ unpacking the kernel is
3350 [1] The old nscd implementation is not available anymore since it was
3351 distributed with glibc up to version 2.0.98 and thus is now replaced.
3353 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3355 * Mark Kettenis provided a Hesiod NSS module.
3357 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3358 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3360 * rcmd can now handle netgroups (Dick Streefland).
3362 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3365 * Support for the IPv6 protocol has been added to the socket API, as per the
3366 latest draft standards.
3368 * Support for Linux 2.2 has been added.
3370 * Interface changes relative to the latest 2.0.x release:
3371 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3372 addseverity NEW: Unix98
3373 alphasort64 NEW: LFS
3374 argp_err_exit_status NEW: argp, GNU ext
3375 argp_error NEW: argp, GNU ext
3376 argp_failure NEW: argp, GNU ext
3377 argp_help NEW: argp, GNU ext
3378 argp_parse NEW: argp, GNU ext
3379 argp_program_bug_address NEW: argp, GNU ext
3380 argp_program_version NEW: argp, GNU ext
3381 argp_program_version_hook NEW: argp, GNU ext
3382 argp_state_help NEW: argp, GNU ext
3383 argp_usage NEW: argp, GNU ext
3384 authdes_create NEW: Secure RPC
3385 authdes_getucred NEW: Secure RPC
3386 authdes_pk_create NEW: Secure RPC
3387 backtrace NEW: GNU ext.
3388 backtrace_symbols NEW: GNU ext.
3389 backtrace_symbols_fd NEW: GNU ext.
3391 cacosf NEW: ISO C 9x
3392 cacosh NEW: ISO C 9x
3393 cacoshf NEW: ISO C 9x
3394 cacoshl NEW: ISO C 9x
3395 cacosl NEW: ISO C 9x
3402 casinf NEW: ISO C 9x
3403 casinh NEW: ISO C 9x
3404 casinhf NEW: ISO C 9x
3405 casinhl NEW: ISO C 9x
3406 casinl NEW: ISO C 9x
3408 catanf NEW: ISO C 9x
3409 catanh NEW: ISO C 9x
3410 catanhf NEW: ISO C 9x
3411 catanhl NEW: ISO C 9x
3412 catanl NEW: ISO C 9x
3413 cbc_crypt NEW: Secure RPC
3417 ccoshf NEW: ISO C 9x
3418 ccoshl NEW: ISO C 9x
3424 cimagf NEW: ISO C 9x
3425 cimagl NEW: ISO C 9x
3426 clearerr_locked REMOVED
3427 clntunix_create NEW: sunrpc ext
3429 clog10 NEW: ISO C 9x
3430 clog10f NEW: ISO C 9x
3431 clog10l NEW: ISO C 9x
3441 cprojf NEW: ISO C 9x
3442 cprojl NEW: ISO C 9x
3444 crealf NEW: ISO C 9x
3445 creall NEW: ISO C 9x
3450 csinhf NEW: ISO C 9x
3451 csinhl NEW: ISO C 9x
3454 csqrtf NEW: ISO C 9x
3455 csqrtl NEW: ISO C 9x
3459 ctanhf NEW: ISO C 9x
3460 ctanhl NEW: ISO C 9x
3462 des_setparity NEW: Secure RPC
3463 ecb_crypt NEW: Secure RPC
3464 endutxent NEW: Unix98
3466 exp10f NEW: ISO C 9x
3467 exp10l NEW: ISO C 9x
3471 fattach NEW: STREAMS
3472 fdetach NEW: STREAMS
3476 feclearexcept NEW: ISO C 9x
3477 fegetenv NEW: ISO C 9x
3478 fegetexceptflag NEW: ISO C 9x
3479 fegetround NEW: ISO C 9x
3480 feholdexcept NEW: ISO C 9x
3482 feraiseexcept NEW: ISO C 9x
3483 ferror_locked REMOVED
3484 fesetenv NEW: ISO C 9x
3485 fesetexceptflag NEW: ISO C 9x
3486 fesetround NEW: ISO C 9x
3487 fetestexcept NEW: ISO C 9x
3488 feupdateenv NEW: ISO C 9x
3489 fflush_locked REMOVED
3493 fgets_unlocked NEW: GNU ext.
3494 fileno_locked REMOVED
3506 fputc_locked REMOVED
3507 fputs_unlocked NEW: GNU ext.
3508 fread_unlocked NEW: GNU ext.
3513 fstatvfs NEW: Unix98
3517 ftruncate64 NEW: LFS
3519 fwrite_unlocked NEW: GNU ext.
3520 gai_strerror NEW: IPv6
3524 getchar_locked REMOVED
3526 getdate_err NEW: Unix98
3527 getdate_r NEW: GNU ext.
3529 getnameinfo NEW: IPv6
3530 getnetname NEW: Secure RPC
3531 getpmsg NEW: STREAMS
3532 getpt NEW: Unix98 PTY
3533 getrlimit64 NEW: LFS
3534 getutxent NEW: Unix98
3535 getutxid NEW: Unix98
3536 getutxline NEW: Unix98
3537 glob64 NEW: GNU ext.
3538 globfree64 NEW: GNU ext.
3539 gnu_get_libc_release NEW: GNU ext.
3540 gnu_get_libc_version NEW: GNU ext.
3541 grantpt NEW: Unix98 PTY
3542 host2netname NEW: Secure RPC
3544 iconv_close NEW: iconv
3545 iconv_open NEW: iconv
3546 if_freenameindex NEW: IPv6
3547 if_indextoname NEW: IPv6
3548 if_nameindex NEW: IPv6
3549 if_nametoindex NEW: IPv6
3550 in6addr_any NEW: IPv6
3551 in6addr_loopback NEW: IPv6
3552 inet6_isipv4mapped NEW: IPv6
3553 isastream NEW: STREAMS
3554 iswblank NEW: GNU ext.
3555 key_decryptsession NEW: Secure RPC
3556 key_decryptsession_pk NEW: Secure RPC
3557 key_encryptsession NEW: Secure RPC
3558 key_encryptsession_pk NEW: Secure RPC
3559 key_gendes NEW: Secure RPC
3560 key_get_conv NEW: Secure RPC
3561 key_secretkey_is_set NEW: Secure RPC
3562 key_setnet NEW: Secure RPC
3563 key_setsecret NEW: Secure RPC
3564 llrint NEW: ISO C 9x
3565 llrintf NEW: ISO C 9x
3566 llrintl NEW: ISO C 9x
3567 llround NEW: ISO C 9x
3568 llroundf NEW: ISO C 9x
3569 llroundl NEW: ISO C 9x
3574 lrintf NEW: ISO C 9x
3575 lrintl NEW: ISO C 9x
3576 lround NEW: ISO C 9x
3577 lroundf NEW: ISO C 9x
3578 lroundl NEW: ISO C 9x
3580 makecontext NEW: Unix98
3581 mempcpy NEW: GNU ext.
3584 modify_ldt NEW: kernel
3588 nearbyint NEW: ISO C 9x
3589 nearbyintf NEW: ISO C 9x
3590 nearbyintl NEW: ISO C 9x
3591 netname2host NEW: Secure RPC
3592 netname2user NEW: Secure RPC
3593 nexttoward NEW: ISO C 9x
3594 nexttowardf NEW: ISO C 9x
3595 nexttowardl NEW: ISO C 9x
3599 passwd2des NEW: Secure RPC
3601 pow10f NEW: GNU ext.
3602 pow10l NEW: GNU ext.
3605 printf_size NEW: GNU ext.
3606 printf_size_info NEW: GNU ext.
3607 profil_counter REMOVED
3608 pthread_mutexattr_getkind_np REPLACED
3609 pthread_mutexattr_setkind_np REPLACED
3610 ptsname NEW: Unix98 PTY
3611 ptsname_r NEW: Unix98 PTY
3613 putchar_locked REMOVED
3614 putgrent NEW: GNU ext.
3616 putpmsg NEW: STREAMS
3617 pututxline NEW: Unix98
3621 readdir64_r NEW: LFS
3622 remquo NEW: ISO C 9x
3623 remquof NEW: ISO C 9x
3624 remquol NEW: ISO C 9x
3626 roundf NEW: ISO C 9x
3627 roundl NEW: ISO C 9x
3629 scalbln NEW: ISO C 9x
3630 scalblnf NEW: ISO C 9x
3631 scalblnl NEW: ISO C 9x
3633 sendfile NEW: kernel
3634 setcontext NEW: Unix98
3635 setrlimit64 NEW: LFS
3636 setutxent NEW: Unix98
3638 sigignore NEW: Unix98
3639 sigqueue NEW: POSIX.1b
3640 sigrelse NEW: Unix98
3641 sigset NEW: POSIX.1b
3642 sigtimedwait NEW: POSIX.1b
3643 sigwaitinfo NEW: POSIX.1b
3644 sincos NEW: GNU ext.
3645 sincosf NEW: GNU ext.
3646 sincosl NEW: GNU ext.
3650 strcasestr NEW: GNU ext.
3651 strtoimax NEW: ISO C 9x
3652 strtoumax NEW: ISO C 9x
3653 strverscmp NEW: GNU ext.
3654 svcauthdes_stats NEW: Secure RPC
3655 svcunix_create NEW: sunrpc etx
3656 svcunixfd_create NEW: sunrpc ext
3657 swapcontext NEW: Unix98
3658 tcgetsid NEW: Unix98 PTY
3659 tdestroy NEW: GNU ext.
3660 tgamma NEW: ISO C 9x
3661 tgammaf NEW: ISO C 9x
3662 tgammal NEW: ISO C 9x
3666 truncf NEW: ISO C 9x
3667 truncl NEW: ISO C 9x
3669 unlockpt NEW: Unix98 PTY
3670 updwtmpx NEW: Unix98
3671 user2netname NEW: Secure RPC
3672 utmpxname NEW: Unix98
3673 versionsort NEW: GNU ext.
3674 versionsort64 NEW: GNU ext.
3676 wcscasecmp NEW: GNU ext.
3677 wcsncasecmp NEW: GNU ext.
3678 wcsnlen NEW: GNU ext.
3679 wcstoimax NEW: ISO C 9x
3680 wcstoll NEW: ISO C 9x
3681 wcstoull NEW: ISO C 9x
3682 wcstoumax NEW: ISO C 9x
3684 wordexp NEW: POSIX.2
3685 wordfree NEW: POSIX.2
3686 write_profiling REMOVED
3687 xdecrypt NEW: Secure RPC
3688 xdr_authdes_cred NEW: Secure RPC
3689 xdr_authdes_verf NEW: Secure RPC
3690 xdr_cryptkeyarg NEW: Secure RPC
3691 xdr_cryptkeyarg2 NEW: Secure RPC
3692 xdr_cryptkeyres NEW: Secure RPC
3693 xdr_getcredres NEW: Secure RPC
3694 xdr_key_netstarg NEW: Secure RPC
3695 xdr_key_netstres NEW: Secure RPC
3696 xdr_keybuf NEW: Secure RPC
3697 xdr_keystatus NEW: Secure RPC
3698 xdr_netnamestr NEW: Secure RPC
3699 xdr_sizeof NEW: Secure RPC
3700 xdr_unixcred NEW: sunrpc ext
3701 xencrypt NEW: Secure RPC
3702 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3713 * inet_ntoa is thread-safe
3715 * updwtmp is moved from libutil to libc
3717 * rewrite of cbrt function
3719 * update of timezone data
3733 * add atoll function
3735 * fix complex problems in Berkeley DB code
3737 * fix math functions
3741 * fixed lots of header problems (especially Linux/GNU specific)
3743 * dynamic loader preserves all registers
3745 * Roland McGrath provided support for handling of auxiliary objects in
3746 the ELF dynamic loader.
3748 * support for parallel builds is improved
3752 * GNU extensions are no longer declared by default. To enable them you
3753 must define the macro `_GNU_SOURCE' in your program or compile with
3756 * The library has changed from using GNU ld symbol aliases to using weak
3757 symbols where available. The ELF object file format supports weak
3758 symbols; GNU ld also supports weak symbols in the a.out format. (There
3759 is also now support for other GNU ld extensions in ELF. Use the
3760 `--with-elf' option to configure to indicate you have ELF, and
3761 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3762 of many files which contained only symbol aliases, reducing the size of
3763 the source and the compiled library; many other files were renamed to
3764 less cryptic names previously occupied by the symbol alias files.
3765 There is a new header file <elf.h> for programs which operate on
3766 files in the ELF format.
3768 * Converted to Autoconf version 2, so `configure' has more options.
3769 Run `configure --help' to see the details.
3771 * The library can now be configured to build profiling, highly-optimized
3772 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3773 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3774 `configure' enable building these extra libraries. The shared library is
3775 built by default when using both ELF and GNU ld. When shared libraries
3776 are enabled, the new library `-ldl' is available for arbitrary run-time
3777 loading of shared objects; its interface is defined in <dlfcn.h>. The
3778 new header file <link.h> gives access to the internals of the run-time
3779 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3780 application of same name on other systems and it provides information
3781 about dynamically linked binaries.
3783 * The C library now provides the run-time support code for profiling
3784 executables compiled with `-pg'. Programs can control the profiling code
3785 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3786 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3787 the support for this file format was contributed by David Mosberger-Tang.
3789 * The math code has been replaced with a math library based on fdlibm from
3790 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3791 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3792 functions. The math functions now reside in a separate library, so
3793 programs using them will need to use `-lm' their linking commands.
3795 * John C. Bowman contributed optimized ix87 assembler inline functions.
3797 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3798 mechanism similar to that found in Solaris 2. This is now used for the
3799 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3800 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3801 file controls what services are used for each individual database. This
3802 works by loading shared libraries with names specified in `nsswitch.conf',
3803 so service modules can be changed or added at any time without even
3804 relinking any program. Currently there are the file, db, and NIS based
3805 NSS services available.
3807 * The new functions `strtoq' and `strtouq' parse integer values from
3808 strings, like `strtol' and `strtoul', but they return `long long int' and
3809 `unsigned long long int' values, respectively (64-bit quantities).
3811 * The new functions `strtof' and `strtold' parse floating-point values from
3812 strings, like `strtod', but they return `float' and `long double' values,
3813 respectively (on some machines `double' and `long double' are the same).
3815 * Ulrich Drepper has contributed new implementations of the floating-point
3816 printing and reading code used in the `printf' family of functions and
3817 `strtod', `strtof', and `strtold'. These new functions are perfectly
3818 accurate, and much faster than the old ones.
3820 * The implementation of the POSIX locale model was completely rewritten by
3821 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3822 to compile the POSIX locale definition.
3824 * The former dummy implementations of the strcoll and strxfrm function are
3825 now replaced by fully functional code contributed by Ulrich Drepper. The
3826 collation information comes from the POSIX locale definitions.
3828 * The new header <langinfo.h> defines an interface for accessing
3829 various locale-dependent data (using the locale chosen with `setlocale').
3831 * Ulrich Drepper has contributed a new suite of functions for operation on
3832 wide-character and multibyte-character strings, in <wchar.h>;
3833 and classification and case conversion of wide characters, in <wctype.h>.
3834 These new functions are conforming to the ISO C, Amendement 1 specification.
3836 * There is now a second implementation of the standard I/O library available.
3837 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3838 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3840 * You can now use positional parameter specifications in format strings
3841 for the `printf' and `scanf' families of functions. For example,
3842 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3843 ``Number 6, Mr Jones''. This is mainly useful when providing different
3844 format strings for different languages, whose grammars may dictate
3845 different orderings of the values being printed. To support this
3846 feature, the interface for `register_printf_handler' has changed; see
3847 the header file <printf.h> for details.
3849 * The `printf' and `scanf' families of functions now understand a new
3850 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3851 to group numbers as indicated by the locale; for `scanf' and friends, this
3852 says to accept as valid only a number with all the proper grouping
3853 separators in the right places. In the default "C" locale, numbers are
3854 not grouped; but locales for specific countries will define the usual
3855 conventions (i.e. separate thousands with `,' in the US locale).
3857 * The pgrp functions have been regularized, slightly incompatibly but much
3858 less confusingly. The core functions are now `getpgid' and `setpgid',
3859 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3860 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3861 provided for compatibility. There is no longer an incompatible `getpgrp'
3862 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3864 * The new header file <fts.h> and suite of functions simplify programs that
3865 operate on directory trees. This code comes from 4.4 BSD.
3867 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3868 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3869 NSS scheme used in glibc.
3871 * There is a new malloc debugging hook `__memalign_hook'.
3873 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3874 `unsigned int' in <sys/types.h>. These are for compatibility only and
3875 their use is discouraged.
3877 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3878 done differently, so that it works even without GNU ld.
3880 * New function `euidaccess' checks allowed access to a file like `access',
3881 but using the effective IDs instead of the real IDs.
3883 * The time zone data files have been updated for the latest and greatest
3884 local time conventions of the countries of the world.
3886 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3889 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3890 for formatting floating-point numbers. They are provided only for
3891 compatibility; new programs should use `sprintf' instead. There are
3892 also equivalent function for the `long double' floating-point type and
3893 all functions also exist in a reentrant form.
3895 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3896 functions for maintaining the login-record files (primarily of use to
3897 system programs such as `login'), and convenient functions for
3898 allocating and initializing a pseudo-terminal (pty) device.
3900 * Ulrich Drepper has contributed new support for System V style
3901 shared memory and IPC on systems that support it.
3903 * Ulrich Drepper has contributed several miscellaneous new functions found
3904 in System V: The `hsearch' family of functions provide an effective
3905 implementation of hash tables; `a64l' and `l64a' provide a very simple
3906 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3909 * Ulrich Drepper has contributed new reentrant counterparts for the
3910 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3912 * Ulrich Drepper has contributed new, highly-optimized versions of several
3913 string functions for the i486/Pentium family of processors.
3915 * Ulrich Drepper has updated the Linux-specific code, based largely
3916 on work done in Hongjiu Lu's version of GNU libc for Linux.
3917 The GNU library now supports Linux versions 2.0.10 and later,
3918 using the ELF object file format (i[3456]86-*-linux).
3920 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3922 * David Mosberger-Tang and Richard Henderson have ported the C library
3923 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3924 dynamic linking support for ELF/Alpha.
3926 * Richard Henderson contributed several Alpha optimized assembler function
3927 for arithmetic and string handling.
3929 * Ulrich Drepper has contributed a new set of message catalog functions to
3930 support multiple languages using the <libintl.h> interface, for use with
3931 his new package GNU gettext. Translation volunteers have contributed
3932 catalogs of the library's messages in Spanish, German, and Korean.
3934 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3935 program and the `catgets' function for reading the catalog files it
3936 creates. (The <libintl.h> interface is preferred; we include the
3937 <nl_types.h> interface using `catgets' only for source compatibility with
3938 programs already written to use it.)
3940 * New header file <values.h> gives SVID-compatible names for <limits.h>
3943 * Various new macros, declarations, and small header files for compatibility
3946 * New function `group_member' is a convenient way to check if a process has
3947 a given effective group ID.
3949 * When using GCC 2.7 and later, the socket functions are now declared in a
3950 special way so that passing an argument of type `struct sockaddr_in *',
3951 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3952 `struct sockaddr *' type, does not generate a type-clash warning.
3954 * New function `error' declared in header file <error.h> is a convenient
3955 function for printing error messages and optionally exiting; this is the
3956 canonical function used in GNU programs. The new functions `err', `warn',
3957 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3958 doing the same thing.
3960 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3961 POSIX.2 `glob' function to do ~ and {...} expansion.
3963 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3964 `clearenv' which is used in POSIX.9 is also available.
3966 * New function `getsid' returns session ID number on systems that support it.
3968 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3969 files <db.h> and <mpool.h> provide a rich set of functions for several
3970 types of simple databases stored in memory and in files, and <ndbm.h> is
3971 an old `ndbm'-compatible interface using the `db' functions. Link with
3972 `-ldb' to get these functions.
3974 * New macro `strdupa' copies a string like `strdup', but uses local stack
3975 space from `alloca' instead of dynamic heap space from `malloc'.
3977 * New function `strnlen' is like `strlen' but searches only a given maximum
3978 number of characters for the null terminator. `stpncpy', `strndup' and
3979 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3982 * New function `statfs' in header <sys/statfs.h>.
3984 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3985 provide convenient functions for operating on blocks of null-terminated
3988 * A new suite of functions in <utmp.h> handle all the details of reading
3989 and writing the utmp file.
3991 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3994 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3995 completely POSIX compliant and also implemented the extended functionality
3996 to handle alternate digit representation and alternate era date formats.
3998 * Ulrich Drepper provided an implementation of the `strptime' function
3999 defined in XPG4.2 which transforms a string into a `struct tm' value.
4001 * Paul Eggert provided the tzselect shell script as part of the timezone
4002 code. The shell script makes it easy to select the correct timezone
4005 * The implementation of the malloc family of functions is completely replaced
4006 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4007 The implementation uses the mmap function (if available) and it is
4008 optimized for the use in multi threaded programs.
4010 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4011 functions. This new functionality is usable by specifying a special
4012 salt string and it is compatible with implementation on *BSD systems.
4014 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4015 `getsubopt' to handle second level command line options, `bsd_signal'
4016 to access BSD style `signal' functionality, the obsolete `regexp' style
4019 * the `lchown' function is available on system which support this
4022 * The implementation of the shadow password handling function was contributed
4025 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4027 * POSIX.1g support was added. The <sys/select.h> header is available,
4028 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4029 implementation of `getaddrinfo'.
4033 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4035 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4036 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4038 * The new function `daemon' from 4.4 BSD is useful for server programs that
4039 want to put themselves in the background.
4041 * Joel Sherrill has contributed support for several standalone boards that
4042 run without an operating system.
4044 * `printf', `scanf' and friends now accept a `q' type modifier for long
4045 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4047 * All of the code taken from BSD (notably most of the math and networking
4048 routines) has been updated from the BSD 4.4-Lite release.
4050 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4052 * The new functions `getdomainname' and `setdomainname' fetch or change the
4053 YP/NIS domain name. These are system calls which exist on systems which
4056 * The time zone data files have been updated for the latest international
4059 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4060 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4064 * The C library now includes support for Sun RPC, from Sun's free
4065 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4066 are included. (There is still no support for YP.)
4068 * Tom Quinn has contributed a port of the C library to SGI machines running
4069 Irix 4 (mips-sgi-irix4).
4071 * The new `lockf' function is a simplified interface to the locking
4072 facilities of `fcntl', included for compatibility.
4074 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4076 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4078 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4081 * The error message printed by `assert' for a failed assertion now includes
4082 the name of the program (if using GNU ld) and the name of the calling
4083 function (with versions of GCC that support this).
4085 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4087 * The library now includes the <sys/mman.h> header file and memory
4088 management functions `mmap', `munmap', `mprotect', `msync', and
4089 `madvise', on systems that support those facilities.
4091 * The interface for `mcheck' has changed slightly: the function called to
4092 abort the program when an allocation inconsistency is detected now takes
4093 an argument that indicates the type of failure. The new function
4094 `mprobe' lets you request a consistency check for a particular block at
4095 any time (checks are normally done only when you call `free' or `realloc'
4098 * It is now possible to easily cross-compile the C library, building on one
4099 system a library to run on another machine and/or operating system. All
4100 you need to do is set the variable `HOST_CC' in `configparms' to the
4101 native compiler for programs to run on the machine you are building on (a
4102 few generator programs are used on Unix systems); set `CC' to the
4105 * The new function `fexecve' (only implemented on the GNU system) executes
4106 a program file given a file descriptor already open on the file.
4110 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4111 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4113 * You can set the variable `libprefix' in `configparms' to specify a prefix
4114 to be prepended to installed library files; this makes it easy to install
4115 the GNU C library to be linked as `-lgnuc' or whatever.
4117 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4118 copies a limited number of characters from a string, and returns the
4119 address of the last character written.
4121 * You no longer need to check for whether the installed `stddef.h' is
4122 compatible with the GNU C library. configure now checks for you.
4124 * You can now define a per-stream `fileno' function to convert the
4125 stream's cookie into an integral file descriptor.
4127 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4128 allocates zero bytes of storage, and returns a unique pointer which
4129 you can pass to `realloc' or `free'. The behavior is undefined if
4130 you dereference this pointer.
4132 * The C library now runs on Sony NEWS m68k machines running either
4133 NewsOS 3 or NewsOS 4.
4135 * The new `syscall' function is a system-dependent primitive function
4136 for invoking system calls. It has the canonical behavior on Unix
4137 systems, including unreliable return values for some calls (such as
4138 `pipe', `fork' and `getppid').
4140 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4141 to `EAGAIN', which is the preferred name. On systems whose kernels
4142 use two distinct codes, the C library now translates EWOULDBLOCK to
4143 EAGAIN in every system call function.
4147 * The GNU C Library Reference Manual is now distributed with the library.
4148 `make dvi' will produce a DVI file of the printed manual.
4149 `make info' will produce Info files that you can read on line using C-h i
4150 in Emacs or the `info' program.
4151 Please send comments on the manual to bug-glibc-manual@gnu.org.
4153 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4155 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4157 * Jason Merrill has contributed a port to the Sequent Symmetry running
4158 Dynix version 3 (i386-sequent-dynix).
4160 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4161 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4163 * New function `memory_warnings' lets you arrange to get warnings when
4164 malloc is running out of memory to allocate, like Emacs gives you.
4166 * The C library now contains the relocating allocator used in Emacs 19 for
4167 its editing buffers. This allocator (ralloc) minimizes allocation
4168 overhead and fragmentation by moving allocated regions around whenever it
4169 needs to. You always refer to a ralloc'd region with a "handle" (a
4170 pointer to a pointer--an object of type `void **').
4172 * There is a new `printf' format: `%m' gives you the string corresponding
4173 to the error code in `errno'.
4175 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4176 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4177 pass, the `a' modifier says to fill in a `char **' you pass with a
4180 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4181 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4182 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4184 * `mkstemp' is a traditional Unix function to atomically create and open a
4185 uniquely-named temporary file.
4189 * The standard location for the file that says what the local timezone is
4190 has changed again. It is now `/usr/local/etc/localtime' (or more
4191 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4193 * The distribution no longer contains any files with names longer than 14
4196 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4197 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4199 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4203 * The configuration process has changed quite a bit. The `configure'
4204 script is now used just like the configuration scripts for other GNU
4205 packages. The `sysdeps' directory hierarchy is much rearranged.
4206 The file `INSTALL' explains the new scheme in detail.
4208 * The header files no longer need to be processed into ANSI C and
4209 traditional C versions. There is just one set of files to install, and
4210 it will work with ANSI or old C compilers (including `gcc -traditional').
4212 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4213 MIPS DECStation running Ultrix 4.
4215 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4216 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4217 made itself into a shared library.
4219 * Yet further improved support for the i386, running 4.3 BSD-like systems
4220 (such as Mach 3 with the Unix single-server), or System V.
4222 * New function `strncasecmp' to do case-insensitive string comparison
4223 with limited length.
4225 * New function `strsep' is a reentrant alternative to `strtok'.
4227 * New functions `scandir' and `alphasort' for searching directories.
4229 * New function `setenv' is a better interface to `putenv'.
4231 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4232 function for traversing a directory tree.
4234 * The GNU obstack package is now also part of the C library.
4235 The new function `open_obstack_stream' creates a stdio stream that
4236 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4237 formatted output directly to an obstack.
4239 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4240 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4242 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4244 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4245 things to your strings.
4247 * There are some new test programs: test-fseek, testmb, and testrand.
4249 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4250 These ports are not finished, but are a good starting place for really
4251 supporting those systems.
4253 * `/etc/localtime' is now the standard location for the file that says what
4254 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4255 This follows the general principle that `/etc' is the place for all local
4256 configuration files.
4258 * The C library header files now use `extern "C"' when used by the C++
4259 compiler, so the C library should now work with C++ code.
4261 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4262 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4265 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4266 function which uses the `merge sort' algorithm, and is said to be
4267 significantly faster than the old GNU `qsort' function. Merge sort is
4268 now the standard `qsort' function. The new algorithm can require a lot
4269 of temporary storage; so, the old sorting function is called when the
4270 required storage is not available.
4272 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4273 provides the Unix `crypt' function, plus some other entry points.
4275 * The code and header files taken from 4.4 BSD have been updated with the
4276 latest files released from Berkeley.
4278 ----------------------------------------------------------------------
4279 Copyright information:
4281 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4283 Permission is granted to anyone to make or distribute verbatim copies
4284 of this document as received, in any medium, provided that the
4285 copyright notice and this permission notice are preserved,
4286 thus giving the recipient permission to redistribute in turn.
4288 Permission is granted to distribute modified versions
4289 of this document, or of portions of it,
4290 under the above conditions, provided also that they
4291 carry prominent notices stating who last changed them.
4294 version-control: never