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 and tan
13 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 and powf.
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 Deprecated and removed features, and other changes affecting compatibility:
40 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
41 defined by <sys/ptrace.h>.
43 * libm no longer supports SVID error handling (calling a user-provided
44 matherr function on error) or the _LIB_VERSION variable to control error
45 handling. (SVID error handling and the _LIB_VERSION variable still work
46 for binaries linked against older versions of the GNU C Library.) The
47 libieee.a library is no longer provided. math.h no longer defines struct
48 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
49 TLOSS, PLOSS and HUGE.
51 * The libm functions pow10, pow10f and pow10l are no longer supported for
52 new programs. Programs should use the standard names exp10, exp10f and
53 exp10l for these functions instead.
55 * The mcontext_t type is no longer the same as struct sigcontext. On
56 platforms where it was previously the same, this changes the C++ name
57 mangling for interfaces involving this type.
59 * The add-ons mechanism for building additional packages at the same time as
60 glibc has been removed. The --enable-add-ons configure option is now
63 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
64 the specified string can be parsed as a domain name.
66 * In the malloc_info output, the <heap> element may contain another <aspace>
67 element, "subheaps", which contains the number of sub-heaps.
69 Changes to build and runtime requirements:
71 [Add changes to build and runtime requirements here]
73 Security related changes:
75 CVE-2009-5064: The ldd script would sometimes run the program under
76 examination directly, without preventing code execution through the
77 dynamic linker. (The glibc project disputes that this is a security
78 vulnerability; only trusted binaries must be examined using the ldd
81 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
82 suffered from a one-byte overflow during ~ operator processing (either
83 on the stack or the heap, depending on the length of the user name).
84 Reported by Tim Rühsen.
86 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
87 would sometimes fail to free memory allocated during ~ operator
88 processing, leading to a memory leak and, potentially, to a denial
91 The glob function, when invoked with GLOB_TILDE and without
92 GLOB_NOESCAPE, could write past the end of a buffer while
93 unescaping user names. Reported by Tim Rühsen.
95 The following bugs are resolved with this release:
97 [The release manager will add the list generated by
98 scripts/list-fixed-bugs.py just before the release.]
105 * A per-thread cache has been added to malloc. Access to the cache requires
106 no locks and therefore significantly accelerates the fast path to allocate
107 and free small amounts of memory. Refilling an empty cache requires locking
108 the underlying arena. Performance measurements show significant gains in a
109 wide variety of user workloads. Workloads were captured using a special
110 instrumented malloc and analyzed with a malloc simulator. Contributed by
111 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
113 * Unicode 10.0.0 Support: Character encoding, character type info, and
114 transliteration tables are all updated to Unicode 10.0.0, using
115 generator scripts contributed by Mike FABIAN (Red Hat).
116 These updates cause user visible changes, especially the changes in
117 wcwidth for many emoji characters cause problems when emoji sequences
118 are rendered with pango, see for example:
119 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
121 * Collation of Hungarian has been overhauled and is now consistent with "The
122 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
125 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
127 - The GNU C Library will now detect when /etc/resolv.conf has been
128 modified and reload the changed configuration. The new resolver option
129 “no-reload” (RES_NORELOAD) disables this behavior.
131 - The GNU C Library now supports an arbitrary number of search domains
132 (configured using the “search” directive in /etc/resolv.conf);
133 previously, there was a hard limit of six domains. For backward
134 compatibility, applications that directly modify the ‘_res’ global
135 object are still limited to six search domains.
137 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
138 Library will now randomly pick a name server from the configuration as a
139 starting point. (Previously, the second name server was always used.)
141 * The tunables feature is now enabled by default. This allows users to tweak
142 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
144 * New function reallocarray, which resizes an allocated block (like realloc)
145 to the product of two sizes, with a guaranteed clean failure upon integer
146 overflow in the multiplication. Originally from OpenBSD, contributed by
147 Dennis Wölfing and Rüdiger Sonderfeld.
149 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
150 These are extended versions of preadv and pwritev, respectively, taking an
151 additional flags argument. The set of supported flags depends on the
152 running kernel; full support currently requires kernel 4.7 or later.
154 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
155 create a new session ID for the spawned process. This feature is
156 scheduled to be added to the next major revision of POSIX; for the time
157 being, it is available under _GNU_SOURCE.
159 * errno.h is now safe to use from C-preprocessed assembly language on all
160 supported operating systems. In this context, it will only define the
161 Exxxx constants, as preprocessor macros expanding to integer literals.
163 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
164 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
165 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
166 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
168 To compile programs that use this feature, the compiler must support
169 128-bit floating point with the type name _Float128 (as defined by TS
170 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
171 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
172 must be defined to make the new interfaces visible.
174 The new functions and macros correspond to those present for other
175 floating-point types (except for a few obsolescent interfaces not
176 supported for the new type), with F128 or f128 suffixes; for example,
177 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
178 printf or scanf formats for the new type; the strfromf128 and strtof128
179 interfaces should be used instead.
181 Deprecated and removed features, and other changes affecting compatibility:
183 * The synchronization that pthread_spin_unlock performs has been changed to
184 now be equivalent to a C11 atomic store with release memory order to the
185 spin lock's memory location. Previously, several (but not all)
186 architectures used stronger synchronization (e.g., containing what is
187 often called a full barrier). This change can improve performance, but
188 may affect odd fringe uses of spin locks that depend on the previous
189 behavior (e.g., using spin locks as atomic variables to try to implement
190 Dekker's mutual exclusion algorithm).
192 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
195 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
196 will only be built and installed when the GNU C Library is configured with
197 --enable-obsolete-rpc. This allows alternative RPC implementations, such
198 as TIRPC or rpcsvc-proto, to be used.
200 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
201 libnss_compat, are deprecated, and will not be built or installed by
204 The NIS(+) support library, libnsl, is also deprecated. By default, a
205 compatibility shared library will be built and installed, but not headers
206 or development libraries. Only a few NIS-related programs require this
207 library. (In particular, the GNU C Library has never required programs
208 that use 'gethostbyname' to be linked with libnsl.)
210 Replacement implementations based on TIRPC, which additionally support
211 IPv6, are available from <https://github.com/thkukuk/>. The configure
212 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
213 name service modules, to be built and installed.
215 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
216 support is enabled, the configured recursive resolver must support EDNS.
217 (Responding to EDNS-enabled queries with responses which are not
218 EDNS-enabled is fine, but FORMERR responses are not.)
220 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
221 servers have not supported this opcode for a long time.
223 * The _res_opcodes variable has been removed from libresolv. It had been
224 exported by accident.
226 * <string.h> no longer includes inline versions of any string functions,
227 as this kind of optimization is better done by the compiler. The macros
228 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
230 * The nonstandard header <xlocale.h> has been removed. Most programs should
231 use <locale.h> instead. If you have a specific need for the definition of
232 locale_t with no other declarations, please contact
233 libc-alpha@sourceware.org and explain.
235 * The obsolete header <sys/ultrasound.h> has been removed.
237 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
239 * The obsolete function cfree has been removed. Applications should use
242 * The stack_t type no longer has the name struct sigaltstack. This changes
243 the C++ name mangling for interfaces involving this type.
245 * The ucontext_t type no longer has the name struct ucontext. This changes
246 the C++ name mangling for interfaces involving this type.
248 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
249 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
250 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
251 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
252 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
253 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
254 fpregset_t type no longer has the name struct fpu. This changes the C++
255 name mangling for interfaces involving those types.
257 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
258 synced with the kernel:
260 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
261 are not supported on this architecture and have been removed.
263 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
264 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
265 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
267 Programs that assume the GET/SETREGS ptrace requests are universally
268 available will now fail to build, instead of malfunctioning at runtime.
270 Changes to build and runtime requirements:
272 * Linux kernel 3.2 or later is required at runtime, on all architectures
273 supported by that kernel. (This is a change from version 2.25 only for
276 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
278 * On most architectures, GCC 4.9 or later is required to build the GNU C
279 Library. On powerpc64le, GCC 6.2 or later is required.
281 Older GCC versions and non-GNU compilers are still supported when
282 compiling programs that use the GNU C Library. (We do not know exactly
283 how old, and some GNU extensions to C may be _de facto_ required. If you
284 are interested in helping us make this statement less vague, please
285 contact libc-alpha@sourceware.org.)
287 Security related changes:
289 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
290 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
292 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
293 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
295 * Avoid printing a backtrace from the __stack_chk_fail function since it is
296 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
299 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
300 fixed (CVE-2017-12133).
302 The following bugs are resolved with this release:
304 [984] network: Respond to changed resolv.conf in gethostbyname
305 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
306 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
308 [12189] libc: __stack_chk_fail should not attempt a backtrace
310 [14096] time: Race condition on timezone/tst-timezone.out
311 [14172] localedata: az_IR: new locale
312 [14995] build: glibc fails to build if gold is the default linker, even if
314 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
315 [16637] network: inet_pton function is accepting IPv6 with bad format
316 [16640] string: string/strtok.c: undefined behaviour inconsistent between
317 x86 and other generic code
318 [16875] localedata: ko_KR: fix lang_name
319 [17225] localedata: ar_SY: localized month names for May and June are
321 [17297] localedata: da_DK: wrong date_fmt string
322 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
323 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
324 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
325 [19066] localedata: ar_SA abbreviated day and month names are in English
326 [19569] network: resolv: Support an arbitrary number of search domains
327 [19570] network: Implement random DNS server selection in the stub
329 [19838] locale: localedef fails on PA-RISC
330 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
331 order of 0D36 and 0D37
332 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
334 [20098] libc: FAIL: debug/backtrace-tst on hppa
335 [20257] network: sunrpc: clntudp_call does not enforce timeout when
337 [20275] localedata: locale day/abday/mon/abmon should not have trailing
339 [20313] localedata: Update locale data to Unicode 9.0
340 [20424] manual: Document how to provide a malloc replacement
341 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
343 [20686] locale: Add el_GR@euro to SUPPORTED.
344 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
345 failures consistently
346 [21015] dynamic-link: Document and fix --enable-bind-now
347 [21016] nptl: pthread_cond support is broken on hppa
348 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
349 frame-pointer on i386
350 [21049] libc: segfault in longjmp_chk() due to clobbered processor
352 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
353 [21088] libc: Build fails with --enable-static-nss
354 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
355 generic c code is used
356 [21109] libc: Tunables broken on big-endian
357 [21112] math: powf has large ulp errors with base close to 1 and exponent
359 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
361 [21120] malloc: glibc malloc is incompatible with GCC 7
362 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
364 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
365 (-0.0) when linking with -lieee
366 [21171] math: log10, log2 and lgamma return incorrect results
367 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
368 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
369 [21207] localedata: ce_RU: update weekdays from CLDR
370 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
371 [21217] localedata: Update months from CLDR-31
372 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
373 [21243] libc: support_delete_temp_file should issue warning for failed
375 [21244] libc: support resolv_test_start() socket fd close should be
377 [21253] libc: localedef randomly segfaults when using -fstack-check due to
378 new posix_spawn implementation
379 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
380 leads to lower CPU frequency
381 [21259] libc: [alpha] termios.h missing IXANY for POSIX
382 [21261] libc: [sparc64] bits/setjmp.h namespace
383 [21267] network: [mips] bits/socket.h IOC* namespace
384 [21268] libc: [alpha] termios.h NL2, NL3 namespace
385 [21270] libc: mmap64 silently truncates large offset values
386 [21275] libc: posix_spawn always crashes on ia64 now
387 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
388 [21280] math: [powerpc] logbl for POWER7 return incorrect results
389 [21289] libc: Incorrect declaration for 32-bit platforms with
390 _FILE_OFFSET_BITS=64 causes build error
391 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
392 not support gethostbyname4_r
393 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
395 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
397 [21340] libc: Support POSIX_SPAWN_SETSID
398 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
399 [21359] network: ns_name_pack needs additional byte in destination buffer
400 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
401 against fragmentation attacks (CVE-2017-12132)
402 [21369] network: resolv: Remove EDNS fallback
403 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
405 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
406 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
407 [21393] stdio: Missing dup3 error check in freopen, freopen64
408 [21396] libc: Use AVX2 memcpy/memset on Skylake server
409 [21399] localedata: Bad description for U00EC in
410 localedata/charmaps/CP1254
411 [21411] malloc: realloc documentation error
412 [21426] network: sys/socket.h uio.h namespace
413 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
414 [21445] libc: signal.h bsd_signal namespace
415 [21455] network: Network headers stdint.h namespace
416 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
418 [21475] network: resolv: Overlong search path is truncated mid-label
419 [21511] libc: sigstack namespace
420 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
421 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
422 bits/syscall.h: No such file or directory
423 [21517] libc: struct sigaltstack namespace
424 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
425 [21533] localedata: Update locale data to Unicode 10.0
427 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
428 junk at end of line, first unrecognized character is `@'
429 [21538] libc: SIG_HOLD missing for XPG4
430 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
431 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
432 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
433 [21543] libc: sigevent namespace
434 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
435 size for general purpose registers in mcontext_t structure
436 [21550] libc: sigwait namespace
437 [21552] libc: XPG4 bsd_signal namespace
438 [21554] libc: sigpause namespace
439 [21560] libc: sys/wait.h signal.h namespace
440 [21561] libc: waitid namespace
441 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
442 '/usr/include/stdlib.h': Permission denied
443 [21575] libc: sys/wait.h missing struct rusage definition
444 [21584] libc: sigaltstack etc namespace
445 [21597] libc: siginterrupt namespace
446 [21607] math: hppa: FAIL: math/test-tgmath
447 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
449 [21622] libc: [tile] missing SA_* for POSIX.1:2008
450 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
451 stack and heap (CVE-2017-1000366)
452 [21625] libc: wait3 namespace
453 [21654] nss: Incorrect pointer alignment in NSS group merge result
455 [21657] network: Parse interface zone id for node-local multicast
456 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
458 [21666] libc: .symver is used on common symbol
459 [21668] network: resolv: res_init cross-thread broadcast introduces race
461 [21687] math: tgmath.h totalorder, totalordermag return type
462 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
464 [21696] libc: Incorrect assumption of of __cpu_mask in
465 posix/sched_cpucount.c
466 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
467 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
469 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
470 leading to relocation crash
471 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
472 `resolv_conf_matches (resp, conf)' failed.
473 [21710] localedata: Added Samoan language locale for Samoa
474 [21711] localedata: Pashto yesstr/nostr locale are missing
475 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
476 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
477 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
478 [21723] localedata: yesstr/nostr missing for Chinese language locale
479 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
481 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
483 [21728] localedata: New Locale for Tongan language
484 [21729] localedata: incorrect LC_NAME fields for hi_IN
485 [21733] localedata: yesstr and nostr are missing for zh_HK
486 [21734] localedata: Missing yesstr and nostr are for kw_GB
487 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
488 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
489 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
490 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
492 [21744] libc: Tests failing on --enable-tunables --enable-stack-
494 [21749] localedata: Wrong abbreviated day name (“abday”) for
496 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
497 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
498 [21759] localedata: missing yesstr and nostr for Tigrinya
499 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
500 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
501 [21767] localedata: Missing Bislama locales
502 [21768] localedata: Missing yesstr and nostr for aa_ET
503 [21770] localedata: Missing Field in li_NL
504 [21778] nptl: Robust mutex may deadlock
505 [21779] libc: MicroBlaze segfaults when loading libpthread
506 [21783] localedata: Fix int_select international_call_prefixes
507 [21784] localedata: Inconsistency in country_isbn
508 [21788] localedata: Missing Country Postal Abbreviations
509 [21794] localedata: Added-country_isbn-for-Italy
510 [21795] localedata: Add/Fix country_isbn for France
511 [21796] localedata: Added country_isbn for Republic of Korea
512 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
513 [21799] localedata: Added int_select international_call_prefixes
514 [21801] localedata: Added int_select international_call_prefixes
515 [21804] nptl: Double semicolon in thread-shared-types.h
516 [21807] localedata: LC_ADDRESS fix for pap_CW
517 [21808] localedata: Fix LC_ADDRESS for pap_AW
518 [21821] localedata: Added country_name in mai_IN
519 [21822] localedata: Fix LC_TIME for mai_IN
520 [21823] localedata: missing yesstr, nostr for sa_IN
521 [21825] localedata: Fix name_mrs for mag_IN
522 [21828] localedata: 2.26 changelog should mention user visible changes
524 [21835] localedata: Added Maithili language locale for Nepal
525 [21838] localedata: Removed redundant data for the_NP
526 [21839] localedata: Fix LC_MONETARY for ta_LK
527 [21844] localedata: Fix Latin characters and Months Sequence.
528 [21848] localedata: Fix mai_NP Title Name
533 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
534 24731-2:2010, is supported to enable declarations of functions from that
535 TR. Note that not all functions from that TR are supported by the GNU C
538 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
539 18661-1:2014, is supported to enable declarations of functions and macros
540 from that TS. Note that not all features from that TS are supported by
543 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
544 18661-4:2015, is supported to enable declarations of functions and macros
545 from that TS. Note that most features from that TS are not supported by
548 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
549 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
550 Since the GNU C Library defaults to a much newer revision of POSIX, this
551 will only affect programs that specifically request an old conformance
552 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
553 see a change in the visible declarations, but a program compiled with
554 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
557 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
558 defined by all multithreaded code, but glibc has not required this for
561 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
562 means that in a future release, the macros “major”, “minor”, and “makedev”
563 will only be available from <sys/sysmacros.h>.
565 These macros are not part of POSIX nor XSI, and their names frequently
566 collide with user code; see for instance glibc bug 19239 and Red Hat bug
567 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
568 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
571 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
572 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
573 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
575 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
576 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
577 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
578 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
579 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
580 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
581 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
582 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
583 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
584 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
585 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
588 * New <math.h> features are added from TS 18661-1:2014:
590 - Signaling NaN macros: SNANF, SNAN, SNANL.
592 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
593 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
594 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
596 - llogb functions: the llogb, llogbf and llogbl functions, and the
597 FP_LLOGB0 and FP_LLOGBNAN macros.
599 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
602 - Comparison macros: iseqsig.
604 - Classification macros: iscanonical, issubnormal, iszero.
606 - Total order functions: totalorder, totalorderf, totalorderl,
607 totalordermag, totalordermagf, totalordermagl.
609 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
611 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
612 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
614 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
615 are added to libc. They convert a floating-point number into string.
617 * Most of glibc can now be built with the stack smashing protector enabled.
618 It is recommended to build glibc with --enable-stack-protector=strong.
619 Implemented by Nick Alcock (Oracle).
621 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
622 intended to be used instead of memset() to erase sensitive data after use;
623 the compiler will not optimize out calls to explicit_bzero even if they
624 are "unnecessary" (in the sense that no _correct_ program can observe the
625 effects of the memory clear).
627 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
628 to float instead of double. This does not affect the ABI of any libraries
629 that are part of the GNU C Library, but may affect the ABI of other
630 libraries that use this type in their interfaces.
632 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
633 float_t and double_t types are now defined to long double instead of float
634 and double. These options are not the default, and this does not affect
635 the ABI of any libraries that are part of the GNU C Library, but it may
636 affect the ABI of other libraries that use this type in their interfaces,
637 if they are compiled or used with those options.
639 * The getentropy and getrandom functions, and the <sys/random.h> header file
642 * The buffer size for byte-oriented stdio streams is now limited to 8192
643 bytes by default. Previously, on Linux, the default buffer size on most
644 file systems was 4096 bytes (and thus remains unchanged), except on
645 network file systems, where the buffer size was unpredictable and could be
646 as large as several megabytes.
648 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
649 for the Linux quota interface which predates kernel version 2.4.22 has
652 * The malloc_get_state and malloc_set_state functions have been removed.
653 Already-existing binaries that dynamically link to these functions will
654 get a hidden implementation in which malloc_get_state is a stub. As far
655 as we know, these functions are used only by GNU Emacs and this change
656 will not adversely affect already-built Emacs executables. Any undumped
657 Emacs executables, which normally exist only during an Emacs build, should
658 be rebuilt by re-running “./configure; make” in the Emacs build tree.
660 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
661 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
662 “no-ip6-dotint” had already been the default, and support for the
663 “ip6-dotint” option was removed from the Internet in 2006.
665 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
666 flag from <resolv.h> have been removed. The option relied on a
667 backwards-incompatible DNS extension which was never deployed on the
670 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
671 RES_BLAST defined in the <resolv.h> header file have been deprecated.
672 They were already unimplemented.
674 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
675 _res.flags are deprecated. The flag was standardized in RFC 2133, but
676 removed again from the IETF name lookup interface specification in RFC
677 2553. Applications should use getaddrinfo instead.
679 * DNSSEC-related declarations and definitions have been removed from the
680 <arpa/nameser.h> header file, and libresolv will no longer attempt to
681 decode the data part of DNSSEC record types. Previous versions of glibc
682 only implemented minimal support for the previous version of DNSSEC, which
683 is incompatible with the currently deployed version.
685 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
686 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
687 <arpa/nameser.h> header file because the distinction between RR types and
688 meta-RR types is not officially standardized, subject to revision, and
689 thus not suitable for encoding in a macro.
691 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
692 and rhook members of the res_state type in <resolv.h> have been removed.
693 The glibc stub resolver did not support these hooks, but the header file
694 did not reflect that.
696 * For multi-arch support it is recommended to use a GCC which has
697 been built with support for GNU indirect functions. This ensures
698 that correct debugging information is generated for functions
699 selected by IFUNC resolvers. This support can either be enabled by
700 configuring GCC with '--enable-gnu-indirect-function', or by
701 enabling it by default by setting 'default_gnu_indirect_function'
702 variable for a particular architecture in the GCC source file
705 * GDB pretty printers have been added for mutex and condition variable
706 structures in POSIX Threads. When installed and loaded in gdb these pretty
707 printers show various pthread variables in human-readable form when read
708 using the 'print' or 'display' commands in gdb.
710 * Tunables feature added to allow tweaking of the runtime for an application
711 program. This feature can be enabled with the '--enable-tunables' configure
712 flag. The GNU C Library manual has details on usage and README.tunables has
713 instructions on adding new tunables to the library.
715 * A new version of condition variables functions have been implemented in
716 the NPTL implementation of POSIX Threads to provide stronger ordering
719 * A new version of pthread_rwlock functions have been implemented to use a more
720 scalable algorithm primarily through not using a critical section anymore to
723 Security related changes:
725 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
726 have been created with makecontext could fail to terminate due to a
727 missing .cantunwind annotation. This has been observed to lead to a hang
728 (denial of service) in some Go applications compiled with gccgo. Reported
729 by Andreas Schwab. (CVE-2016-6323)
731 * The DNS stub resolver functions would crash due to a NULL pointer
732 dereference when processing a query with a valid DNS question type which
733 was used internally in the implementation. The stub resolver now uses a
734 question type which is outside the range of valid question type values.
737 The following bugs are resolved with this release:
739 [4099] stdio: Overly agressive caching by stream i/o functions.
740 [7065] build: Support building glibc with -fstack-protector or -fstack-
742 [9842] localedata: en_CA: incorrect date format
743 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
744 before it started waiting
745 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
746 [15765] nptl: sem_open is wrongly a cancellation point
747 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
748 [16458] libc: endian.h and netinet/in.h byte order macros return values of
750 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
751 library linked with pthread
752 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
754 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
755 [17252] libc: getrandom and getentropy syscall
756 [17730] malloc: thread-local storage is sometimes improperly free()'d
757 after being __libc_memalign()'d
758 [18241] stdio: failed fseek on memstream does not set errno and can fail
760 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
762 [18463] nptl: pthread_cond_broadcast issue when surrounded by
763 PTHREAD_PRIO_INHERIT mutex on ARM
764 [18784] network: res_query and related function crash for special record
765 type queries (CVE-2015-5180)
766 [19380] math: strtod does not raise "inexact"
767 [19387] string: Integer overflow in memchr
768 [19390] string: Integer overflow in strncat
769 [19398] build: linknamespace tests fail with massively parallel build
770 [19402] nptl: Deadlock with robust shared mutex and asynchronous
772 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
773 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
774 [19514] libc: [PATCH] Fix spelling errors in spelling
775 "implement"/"implementation" in several places
776 [19582] network: Deprecate RES_USE_INET6
777 [19673] manual: clog10 docs appear to be erroneous
778 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
779 causes a segmentation fault
780 [19826] libc: invalid pointer returned from __tls_get_addr with static
782 [20016] network: resolv: Remove hooks support from the API
783 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
785 [20033] math: [x86_64] vectorized math function don't call the __finite
787 [20116] nptl: use after free in pthread_create
788 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
789 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
791 [20311] nptl: please install proc_service.h
792 [20366] libc: Compilation errors in installed headers in strict-compliance
794 [20370] malloc: malloc: Arena free list management is still racy
795 (incorrect fix in bug 19243)
796 [20386] libc: assert (X = 0) does not result in GCC warning
797 [20432] malloc: malloc: Minimize interface required for interposition
798 [20435] libc: Missing unwind info in __startcontext causes infinite loop
799 in _Unwind_Backtrace (CVE-2016-6323)
800 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
801 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
803 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
804 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
805 [20477] network: resolv: incorrect double-checked locking related to
807 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
809 [20495] math: x86_64 performance degradation due to AVX/SSE transition
811 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
812 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
813 cause transition penalty
814 [20517] math: sparcv9 missing fdiml compat symbol
815 [20524] manual: strverscmp is inconsistent
816 [20525] libc: <sys/quota.h> should be based on kernel headers
817 [20539] math: GCC 7 -static -lm fails to link at -O3
818 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
820 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
822 [20591] network: Remove obsolete DNSSEC support
823 [20592] network: DNS resource record type classification macros in
824 <arpa/nameser.h> are incorrect
825 [20593] network: Update DNS RR type definitions
826 [20611] network: getaddrinfo accepts invalid numeric scope IDs
827 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
829 [20629] network: libresolv: Remove support for bitstring labels
831 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
832 assertion: "maxidx >= 2"
833 [20660] math: [arm] Use VSQRT
834 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
835 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
836 [20689] libc: Test for FMA should also check for AVX.
837 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
838 [20715] math: iszero macro breaks existing code
839 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
840 [20728] libc: powerpc: Missing TOC stub in clone
841 [20729] build: glibc-2.24 fails to build for i486 with -Os
842 [20750] build: Build fails with default PIE enabled gcc-6.x
843 [20768] math: [alpha] sqrt fegetenv namespace
844 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
846 [20787] math: float_t is defined as float incorrectly on x86_64 even if
847 __FLT_EVAL_METHOD__ is 2
848 [20790] network: rpcgen buffer overrun in get_prog_declaration
849 [20822] nptl: race condition in __lll_unlock_elision on powerpc
850 [20829] libc: crypt snprintf namespace
851 [20847] libc: tst-vfork3 failure
852 [20855] math: Default bits/mathdef.h has inappropriate float_t
853 [20859] math: [sh4] FP_ILOGB0 invalid
854 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
855 U+20AC), not same as GBK
856 [20915] dynamic-link: global-dynamic TLS broken on aarch64
857 [20916] math: pow handling of sNaN arguments
858 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
859 [20919] math: Bad pow (qNaN, 0) result with -lieee
860 [20940] math: hypot sNaN handling
861 [20947] math: fmax, fmin sNaN handling
862 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
863 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
865 [20971] string: powerpc64/power7 memchr overflows internal pointer check
866 [20973] nptl: robust mutexes: Lost wake-ups
867 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
868 [20978] nis: strlen on null pointer in nss_nisplus
869 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
871 [21014] string: i686 memchr overflows internal pointer check
872 [21019] libc: [mips] n32 lseek incorrect on overflow
873 [21022] libc: [microblaze] __backtrace get_frame_size namespace
874 [21026] libc: [MIPS] readahead syscall is broken on n64
875 [21028] math: Fallback fesetexceptflag should always succeed
876 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
877 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
879 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
880 [21061] librt: [microblaze] librt lost clock_* exports
881 [21073] libc: tunables: insecure environment variables passed to
882 subprocesses with AT_SECURE
883 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
887 * The minimum Linux kernel version that this version of the GNU C Library
888 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
889 kernel version 2.6.32 or later suffices (on architectures that already
890 required kernel versions more recent than 3.2, those requirements remain
891 unchanged). Linux 3.2 or later kernel headers are required on all
894 * The pap_AN locale has been deleted. This has been deprecated for a long
895 time. It has been replaced by pap_AW & pap_CW, both of which have long
896 been included in previous releases.
898 * The readdir_r and readdir64_r functions have been deprecated. It is
899 recommended to use readdir and readdir64 instead.
901 * The type “union wait” has been removed. It was deprecated in the early
902 1990s and never part of POSIX. Application code should use the int type
903 instead of “union wait”.
905 * A new NSS action is added to facilitate large distributed system
906 administration. The action, MERGE, allows remote user stores like LDAP
907 to be merged into local user stores like /etc/groups in order to provide
908 easy to use, updated, and managed sets of merged credentials. The new
909 action can be used by configuring it in /etc/nsswitch.conf:
910 group: files [SUCCESS=merge] nis
911 Implemented by Stephen Gallagher (Red Hat).
913 * The deprecated __malloc_initialize_hook variable has been removed from the
916 * The long unused localedef --old-style option has been removed. It hasn't
917 done anything in over 16 years. Scripts using this option can safely
920 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
921 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
922 return the next representable value in the direction of positive infinity
923 and the nextdown functions return the next representable value in the
924 direction of negative infinity. These are currently enabled as GNU
927 Security related changes:
929 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
930 could result in a stack overflow when getnetbyname was called with an
931 overly long name. (CVE-2016-3075)
933 * Previously, getaddrinfo copied large amounts of address data to the stack,
934 even after the fix for CVE-2013-4458 has been applied, potentially
935 resulting in a stack overflow. getaddrinfo now uses a heap allocation
936 instead. Reported by Michael Petlan. (CVE-2016-3706)
938 * The glob function suffered from a stack-based buffer overflow when it was
939 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
940 Reported by Alexander Cherepanov. (CVE-2016-1234)
942 * The Sun RPC UDP client could exhaust all available stack space when
943 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
944 alloca plugin for GCC. (CVE-2016-4429)
946 * The IPv6 name server management code in libresolv could result in a memory
947 leak for each thread which is created, performs a failing naming lookup,
948 and exits. Over time, this could result in a denial of service due to
949 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
951 The following bugs are resolved with this release:
953 [1170] localedata: ne_NP: update Nepali locale definition file
954 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
956 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
957 [6796] math: fdim() does not set errno on overflow
958 [10354] libc: posix_spawn should use vfork() in more cases than presently
959 [11213] localedata: localedata: add copyright disclaimer to locale files
960 [12143] localedata: chr_US: new Cherokee locale
961 [12450] localedata: sgs_LT: new locale
962 [12676] localedata: ln_CD: new locale
963 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
965 [13304] math: fma, fmaf, fmal produce wrong results
966 [14259] build: --localedir arg to configure is ignored
967 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
968 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
969 [14934] localedata: es_CL: wrong first weekday chilean locale
970 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
972 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
974 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
975 [15368] nptl: raise() is not async-signal-safe
976 [15479] math: ceil, floor, round and trunc raise inexact exception
977 [15578] localedata: kk_KZ: various updates
978 [16003] localedata: pap_AN: punt old locale
979 [16137] localedata: iw_IL: punt old locale
980 [16190] localedata: eo: new esperanto locale
981 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
982 [16742] malloc: race condition: pthread_atfork() called before first
983 malloc() results in unexpected locking behaviour/deadlocks
984 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
986 [16983] localedata: postal_fmt does not allow %l and %n modifiers
987 [17565] localedata: pt_PT: wrong (work-)week start
988 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
989 [17950] build: Build fails with -msse
990 [18205] localedata: be_BY*: wrong first_weekday and first_workday
991 [18433] libc: posix_spawn does not return correctly upon failure to
993 [18453] localedata: charmaps/IBM875: incorrect codes
994 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
996 [18896] localedata: he_IL: improvements for currency
997 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
999 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1001 [19056] libc: Deprecate readdir_r
1002 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1004 [19198] localedata: nl_NL: small improvements for Dutch locales
1005 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1006 nameservers (CVE-2016-5417)
1007 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1009 [19400] locale: Language missing in "iso-639.def", trivial fix in
1011 [19431] malloc: Deadlock between fflush, getdelim, and fork
1012 [19505] libc: Incorrect file descriptor validity checks in
1013 posix_spawn_file_actions_add{open,close,dup2}
1014 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1015 when using RTLD_NEXT
1016 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1017 `intl/{gettextP,loadinfo}.h'
1018 [19534] libc: execle, execlp may use malloc
1019 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1020 [19573] network: res_nclose and __res_maybe_init disagree about name
1021 server initialization, breaking Hesiod
1022 [19575] localedata: Status of GB18030 tables
1023 [19581] localedata: sr_* date_fmt string contains additional newline
1024 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1026 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1027 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1028 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1029 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1031 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1033 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1035 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1036 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1037 [19633] locale: strfmon_l applies global locale to number formatting
1038 [19642] network: Memory leak in getnameinfo
1039 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1040 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1042 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1043 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1045 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1046 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1048 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1050 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1052 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1053 pointers and lengths in error-case.
1054 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1055 report errors on UTF-16 surrogates values.
1056 [19755] nscd: nscd assertion failure in gc
1057 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1058 [19759] libc: mempcpy shouldn't be inlined
1059 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1060 [19765] libc: s390 needs an optimized mempcpy
1061 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1062 NAME_MAX limit assumption (CVE-2016-1234)
1063 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1064 [19787] network: Missing and incorrect truncation checks in getnameinfo
1065 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1067 [19791] network: Assertion failure in res_query.c with un-connectable name
1069 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1070 [19822] math: libm.so install clobbers old version
1071 [19825] network: resolv: send_vc can return uninitialized data in second
1072 response to getaddrinfo
1073 [19830] network: nss_dns: should check RDATA length against buffer length
1074 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1075 confronted with A/AAAA records of invalid size
1076 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1077 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1078 [19853] stdio: Printing IBM long double in decimal with high precision is
1080 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1081 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1082 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1084 [19865] network: Assertion failure or memory leak in
1085 _nss_dns_getcanonname_r
1086 [19868] network: nss_dns: netent code does not skip over non-PTR records
1087 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1089 [19881] string: Improve x86-64 memset
1090 [19907] string: Incorrect memcpy tests
1091 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1093 [19925] libc: termios.h XCASE namespace
1094 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1095 [19929] libc: limits.h NL_NMAX namespace
1096 [19931] stdio: Memory leak in vfprintf
1097 [19957] libc: clone(CLONE_VM) access invalid parent memory
1098 [19963] localedata: en_IL: New locale
1099 [19989] stdio: stdio.h cuserid namespace
1100 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1102 [19996] locale: langinfo.h nl_langinfo_l namespace
1103 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1104 [20010] network: getaddrinfo: Stack overflow in hostent translation
1106 [20012] stdio: libio: fmemopen append mode failure
1107 [20014] stdio: stdio.h namespace for pre-threads POSIX
1108 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1109 [20023] libc: fcntl.h timespec namespace
1110 [20024] math: [x86_64] vectorized sincos trashes the stack
1111 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1112 [20041] time: sys/time.h timespec namespace
1113 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1114 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1115 [20051] libc: ttyslot in wrong header under wrong conditions
1116 [20054] libc: gethostname not declared for XPG4
1117 [20055] libc: termios.h missing tcgetsid for XPG4
1118 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1120 [20073] libc: sys/stat.h fchmod namespace
1121 [20074] libc: stdlib.h rand_r namespace
1122 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1123 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1125 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1126 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1128 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1129 [20119] libc: Wrong mask for processors level type from CPUID
1130 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1131 AS not supporting AVX512
1132 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1134 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1135 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1137 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1138 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1139 [20178] libc: posix_spawn{p} should not call exit
1140 [20191] stdio: libio: vtables hardening
1141 [20195] string: FMA4 detection requires CPUID execution with register
1143 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1144 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1146 [20212] math: acos (sNaN) returns sNaN
1147 [20213] math: asin (sNaN) returns sNaN
1148 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1149 [20218] math: [i386] asinhl (sNaN) returns sNaN
1150 [20219] math: [i386] atanhl (sNaN) returns sNaN
1151 [20222] stdio: fopencookie: Mangle function pointers
1152 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1153 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1154 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1156 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1157 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1158 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1159 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1160 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1161 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1162 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1163 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1164 [20240] math: modf (sNaN) returns sNaN
1165 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1166 [20250] math: frexp (sNaN) returns sNaN
1167 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1168 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1170 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1171 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1172 subscript is below array bounds [-Werror=array-bounds]
1173 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1174 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1176 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1177 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1179 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1180 "invalid" exceptions
1181 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1182 [20316] localedata: id_ID: Februari instead of Pebruari
1183 [20327] string: POWER8 strcasecmp returns incorrect result
1184 [20347] math: Failure: Test: j0_downward (0xap+0)
1185 [20348] libc: FAIL: misc/tst-preadvwritev64
1186 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1187 [20350] libc: There is no test for p{read,write}64
1188 [20357] math: Incorrect cos result for 1.5174239687223976
1189 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1193 * Unicode 8.0.0 Support: Character encoding, character type info, and
1194 transliteration tables are all updated to Unicode 8.0.0, using new
1195 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1196 These updates cause user visible changes, such as the fixes for bugs
1197 89, 16061, and 18568.
1199 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1200 kernel-internal CPU set size. This means that requests that change the
1201 CPU affinity which failed before (for example, an all-ones CPU mask) will
1202 now succeed. Applications that need to determine the effective CPU
1203 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1204 setting it because the kernel can adjust it (and the previous size check
1205 would not detect this in the majority of cases).
1207 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1208 the following new symbols are used: fts64_children, fts64_close,
1209 fts64_open, fts64_read and fts64_set.
1211 * getaddrinfo now detects certain invalid responses on an internal netlink
1212 socket. If such responses are received, an affected process will
1213 terminate with an error message of "Unexpected error <number> on netlink
1214 descriptor <number>" or "Unexpected netlink response of size <number> on
1215 descriptor <number>". The most likely cause for these errors is a
1216 multi-threaded application which erroneously closes and reuses the netlink
1217 file descriptor while it is used by getaddrinfo.
1219 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1220 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1221 unnecessary serialization of memory allocation requests across threads.
1222 The defect is now corrected. Users should see a substantial increase in
1223 the concurent throughput of allocation requests for applications which
1224 trigger this bug. Affected applications typically create and
1225 destroy threads frequently. (Bug 19048 was reported and analyzed by
1228 * There is now a --disable-timezone-tools configure option for disabling the
1229 building and installing of the timezone related utilities (zic, zdump, and
1230 tzselect). This is useful for people who build the timezone data and code
1231 independent of the GNU C Library.
1233 * The obsolete header <regexp.h> has been removed. Programs that require
1234 this header must be updated to use <regex.h> instead.
1236 * The obsolete functions bdflush, create_module, get_kernel_syms,
1237 query_module and uselib are no longer available to newly linked binaries;
1238 the header <sys/kdaemon.h> has been removed. These functions and header
1239 were specific to systems using the Linux kernel and could not usefully be
1240 used with the GNU C Library on systems with version 2.6 or later of the
1243 * Optimized string, wcsmbs and memory functions for IBM z13.
1244 Implemented by Stefan Liebler.
1246 * Newly linked programs that define a variable called signgam will no longer
1247 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1248 require signgam to be set by those functions must ensure that they use the
1249 variable provided by the GNU C Library and declared in <math.h>, without
1250 defining their own copy.
1252 * The minimum GCC version that can be used to build this version of the GNU
1253 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1254 still be used to compile programs using the GNU C Library.
1256 Security related changes:
1258 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1259 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1261 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1262 the pointer guard feature. It is always enabled. Previously,
1263 LD_POINTER_GUARD could be used to disable security hardening in binaries
1264 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1267 * An integer overflow in hcreate and hcreate_r could lead to an
1268 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1270 * The catopen function no longer has unbounded stack usage. Reported by
1271 Max. (CVE-2015-8779)
1273 * The nan, nanf and nanl functions no longer have unbounded stack usage
1274 depending on the length of the string passed as an argument to the
1275 functions. Reported by Joseph Myers. (CVE-2014-9761)
1277 * A stack-based buffer overflow was found in libresolv when invoked from
1278 libnss_dns, allowing specially crafted DNS responses to seize control
1279 of execution flow in the DNS client. The buffer overflow occurs in
1280 the functions send_dg (send datagram) and send_vc (send TCP) for the
1281 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1282 family. The use of AF_UNSPEC triggers the low-level resolver code to
1283 send out two parallel queries for A and AAAA. A mismanagement of the
1284 buffers used for those queries could result in the response of a query
1285 writing beyond the alloca allocated buffer created by
1286 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1287 the overflow. Thanks to the Google Security Team and Red Hat for
1288 reporting the security impact of this issue, and Robert Holiday of
1289 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1291 The following bugs are resolved with this release:
1293 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1294 [887] math: Math library function "logb" and "nextafter" inconsistent
1295 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1296 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1297 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1298 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1300 [4404] localedata: German translation of "Alarm clock" is misleading
1301 [6799] math: nextafter() and nexttoward() doen't set errno on
1302 overflow/underflow errors
1303 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1305 [10432] nis: _nss_nis_setnetgrent assertion failure
1306 [11460] libc: fts has no LFS support
1307 [12926] network: getaddrinfo()/make_request() may spin forever
1308 [13065] nptl: Race condition in pthread barriers
1309 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1310 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1312 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1313 [14912] libc: Rename non-installed bits/*.h headers
1314 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1315 [15367] math: Let gcc use __builtin_isinf
1316 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1317 [15421] math: lgamma wrongly sets signgam for ISO C
1318 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1319 FE_INVALID with argument out of range
1320 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1322 [15786] dynamic-link: ifunc resolver functions can smash function
1324 [15918] math: Unnecessary check for equality in hypotf()
1325 [16061] localedata: Review / update transliteration data
1326 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1328 [16141] time: strptime %z offset restriction
1329 [16171] math: drem should be alias of remainder
1330 [16296] math: fegetround is pure?
1331 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1332 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1334 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1335 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1337 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1338 "Exception "Inexact" set" on ppc32
1339 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1340 [16517] math: Missing underflow exception from tanf/tan/tanl
1341 [16519] math: Missing underflow exception from sinhf
1342 [16520] math: Missing underflow exception from tanhf
1343 [16521] math: Missing underflow exception from exp2
1344 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1346 [16734] stdio: fopen calls mmap to allocate its buffer
1347 [16961] math: nan function incorrect handling of bad sequences
1348 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1349 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1350 [16985] locale: localedef: confusing error message when opening output
1352 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1353 [17197] locale: Redundant shift character in iconv conversion output at
1355 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1356 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1357 [17250] dynamic-link: static linking breaks nss loading
1358 (getaddrinfo/getpwnam/etc...)
1359 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1361 [17441] math: isnan() should use __builtin_isnan() in GCC
1362 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1363 (related to lock elision)
1364 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1365 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1366 [17887] time: strptime should be able to parse "+01:00" style timezones
1367 [17905] libc: catopen() Multiple unbounded stack allocations
1369 [18084] libc: backtrace (..., 0) dumps core on x86
1370 [18086] libc: nice() sets errno to 0 on success
1371 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1372 is too large (CVE-2015-8778)
1373 [18251] dynamic-link: SONAME missing when audit modules provides path
1374 [18265] libc: add attributes for wchar string and memory functions
1375 [18370] math: csqrt missing underflows
1376 [18421] libc: [hppa] read-only segment has dynamic relocations
1377 [18472] libc: Obsolete syscall wrappers should be compat symbols
1378 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1379 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1380 [18525] localedata: Remove locale timezone information
1381 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1382 [18568] localedata: Update locale data to Unicode 8.0
1383 [18589] locale: sort-test.sh fails at random
1384 [18595] math: ctan, ctanh missing underflows
1385 [18604] libc: assert macro-expands its argument
1386 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1387 contains a vector instruction exception.
1388 [18611] math: j1, jn missing errno setting on underflow
1389 [18618] localedata: sync Chechen locale definitions with other *_RU
1391 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1392 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1393 [18665] network: In send_dg, the recvfrom function is NOT always using the
1394 buffer size of a newly created buffer (CVE-2015-7547)
1395 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1397 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1399 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1400 [18699] math: tilegx cproj() for various complex infinities does not yield
1402 [18724] libc: Harden put*ent functions against data injection
1403 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1404 [18755] build: build errors with -DNDEBUG
1405 [18757] stdio: fmemopen fails to set errno on failure
1406 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1408 [18781] libc: openat64 lacks O_LARGEFILE
1409 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1410 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1411 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1412 [18790] math: [ldbl-128ibm] tanhl inaccurate
1413 [18795] libc: stpncpy fortification misses buffer lengths that are
1414 statically too large
1415 [18796] build: build fails for --disable-mathvec
1416 [18803] math: hypot missing underflows
1417 [18820] stdio: fmemopen may leak memory on failure
1418 [18823] math: csqrt spurious underflows
1419 [18824] math: fma spurious underflows
1420 [18825] math: pow missing underflows
1421 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1422 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1423 [18870] build: sem_open.c fails to compile with missing symbol
1425 [18872] stdio: Fix memory leak in printf_positional
1426 [18873] libc: posix_fallocate overflow check ineffective
1427 [18875] math: Excess precision leads incorrect libm
1428 [18877] libc: arm: mmap offset regression
1429 [18887] libc: memory corruption when using getmntent on blank lines
1430 [18918] localedata: hu_HU: change time to HH:MM:SS format
1431 [18921] libc: Regression: extraneous stat() and fstat() performed by
1433 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1434 binaries (CVE-2015-8777)
1435 [18951] math: tgamma missing underflows
1436 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1438 [18953] localedata: lt_LT: change currency symbol to the euro
1439 [18956] math: powf inaccuracy
1440 [18961] math: [i386] exp missing underflows
1441 [18966] math: [i386] exp10 missing underflows
1442 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1443 [18969] build: multiple string test failures due to missing locale
1445 [18970] libc: Reference of pthread_setcancelstate in libc.a
1446 [18977] math: float / long double Bessel functions not in XSI POSIX
1447 [18980] math: i386 libm functions return with excess range and precision
1448 [18981] math: i386 scalb*, ldexp return with excess range and precision
1449 [18982] stdio: va_list and vprintf
1450 [18985] time: Passing out of range data to strftime() causes a segfault
1452 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1453 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1454 [19012] locale: iconv_open leaks memory on error path
1455 [19016] math: clog, clog10 inaccuracy
1456 [19018] nptl: Mangle function pointers in tls_dtor_list
1457 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1458 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1459 [19048] malloc: malloc: arena free list can become cyclic, increasing
1461 [19049] math: [powerpc] erfc incorrect zero sign
1462 [19050] math: [powerpc] log* incorrect zero sign
1463 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1464 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1465 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1466 [19074] network: Data race in _res_hconf_reorder_addrs
1467 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1468 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1469 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1471 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1473 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1474 [19086] manual: posix_fallocate64 documented argument order is wrong.
1475 [19088] math: lround, llround missing exceptions close to overflow
1477 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1478 [19095] math: dbl-64 lrint incorrect for 64-bit long
1479 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1480 [19124] dynamic-link: ld.so failed to build with older assmebler
1481 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1482 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1483 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1484 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1485 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1486 pthread_setaffinity_np
1487 [19156] math: [ldbl-128] j0l spurious underflows
1488 [19164] nptl: tst-getcpu fails with many possible CPUs
1489 [19168] math: math/test-ildoubl and math/test-ldouble failure
1490 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1491 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1493 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1494 (FE_NOMASK_ENV) do not clear SSE exceptions
1495 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1496 _int_new_arena/reused_arena
1497 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1498 [19201] math: dbl-64 remainder incorrect sign of zero result
1499 [19205] math: bits/math-finite.h conditions do not match math.h and
1501 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1502 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1503 for C99-based standards
1504 [19212] libc: features.h not -Wundef clean
1505 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1507 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1508 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1509 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1511 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1512 "inexact" exceptions
1513 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1515 [19242] libc: strtol incorrect in Turkish locales
1516 [19243] malloc: reused_arena can pick an arena on the free list, leading
1517 to an assertion failure and reference count corruption
1518 [19253] time: tzset() ineffective when temporary TZ did not include DST
1520 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1521 [19270] math: [hppa] Shared libm missing __isnanl
1522 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1523 MAP_HUGETLB and MAP_STACK defines
1524 [19313] nptl: Wrong __cpu_mask for x32
1525 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1526 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1527 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1528 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1529 [19363] time: x32: times() return value wrongly truncates/sign extends
1531 [19367] dynamic-link: Improve branch prediction on Silvermont
1532 [19369] network: Default domain name not reset by res_ninit when "search"
1533 / "domain" entry is removed from resolv.conf
1534 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1536 [19385] localedata: bg_BG: time separator should be colon, not comma
1537 [19408] libc: linux personality syscall wrapper may erroneously return an
1538 error on 32-bit architectures
1539 [19415] libc: dladdr returns wrong names on hppa
1540 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1541 IBM905, IBM907, and IBM909
1542 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1543 [19443] build: build failures with -DDEBUG
1544 [19451] build: Make check fails on test-double-vlen2
1545 [19462] libc: Glibc failed to build with -Os
1546 [19465] math: Wrong code with -Os
1547 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1549 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1551 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1552 [19486] math: S390: Math tests fail with "Exception Inexact set".
1553 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1554 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1556 [19590] math: Fail to build shared objects that use libmvec.so functions.
1560 * The following bugs are resolved with this release:
1562 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1563 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1564 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1565 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1566 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1567 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1568 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1569 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1570 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1571 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1572 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1573 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1574 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1575 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1576 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1577 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1578 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1579 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1580 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1581 18657, 18676, 18694, 18696, 18887.
1583 * Cache information can be queried via sysconf() function on s390 e.g. with
1584 _SC_LEVEL1_ICACHE_SIZE as argument.
1586 * A buffer overflow in gethostbyname_r and related functions performing DNS
1587 requests has been fixed. If the NSS functions were called with a
1588 misaligned buffer, the buffer length change due to pointer alignment was
1589 not taken into account. This could result in application crashes or,
1590 potentially arbitrary code execution, using crafted, but syntactically
1591 valid DNS responses. (CVE-2015-1781)
1593 * The time zone file parser has been made more robust against crafted time
1594 zone files, avoiding heap buffer overflows related to the processing of
1595 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1596 large time zone data files. Overly long time zone specifiers in the TZ
1597 variable no longer result in stack overflows and crashes.
1599 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1600 for LD and GD on x86 and x86-64, has been implemented. You will need
1601 binutils-2.24 or later to enable this optimization.
1603 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1604 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1605 Hat). These updates cause user visible changes, such as the fix for bug
1608 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1609 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1610 condition in some applications.
1612 * Added vector math library named libmvec with the following vectorized x86_64
1613 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1615 The library can be disabled with --disable-mathvec. Use of the functions is
1616 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1617 Shared library libmvec.so is linked in as needed when using -lm (no need to
1618 specify -lmvec explicitly for not static builds).
1619 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1621 * A new fmemopen implementation has been added with the goal of POSIX
1622 compliance. The new implementation fixes the following long-standing
1623 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1624 old implementation is still present for use be by existing binaries.
1626 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1627 releases. It has been fixed to match 2.19 and older, but binaries built
1628 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1630 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1631 Contributed by Roland McGrath (Google).
1633 * The header <regexp.h> is deprecated, and will be removed in a future
1634 release. Use of this header will trigger a deprecation warning.
1635 Application developers should update their code to use <regex.h> instead.
1637 This header was formerly part of SUS, but was deprecated in 1994 and
1638 removed from the standard in 2001. Also, the glibc implementation
1639 leaks memory. See BZ#18681 for more details.
1643 * The following bugs are resolved with this release:
1645 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1646 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1647 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1648 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1649 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1650 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1651 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1652 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1653 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1654 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1657 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1658 too little memory for the to-be-scanned arguments and overflow the
1659 allocated buffer. The implementation now correctly computes the required
1660 buffer size when using malloc, and switches to malloc from alloca as
1663 * A new semaphore algorithm has been implemented in generic C code for all
1664 machines. Previous custom assembly implementations of semaphore were
1665 difficult to reason about or ensure that they were safe. The new version
1666 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1667 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1668 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1670 * Port to Altera Nios II has been contributed by Mentor Graphics.
1672 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1673 implementations for powerpc64/powerpc64le.
1674 Implemented by Adhemerval Zanella (IBM).
1676 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1677 and powerpc64le. This may improve lock scaling of existing programs on
1678 HTM capable systems. The lock elision code is only enabled with
1679 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1680 powerpc will issue a transaction abort on every syscall to avoid side
1681 effects being visible outside transactions.
1683 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1684 AArch64. Contributed by ARM Ltd.
1686 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1688 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1689 under certain input conditions resulting in the execution of a shell for
1690 command substitution when the application did not request it. The
1691 implementation now checks WRDE_NOCMD immediately before executing the
1692 shell and returns the error WRDE_CMDSUB as expected.
1694 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1695 processing format strings with a large number of format specifiers.
1697 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1698 infinite loop if the DNS response contained a PTR record of an unexpected
1701 * The minimum GCC version that can be used to build this version of the GNU
1702 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1703 still be used to compile programs using the GNU C Library.
1705 * The GNU C Library is now built with -Werror by default. This can be
1706 disabled by configuring with --disable-werror.
1708 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1710 * The obsolete sigvec function has been removed. This was the original
1711 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1712 programs have been using instead for about 25 years. Of course, ABI
1713 compatibility for old binaries using sigvec remains intact.
1715 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1716 with newer versions of bison.
1718 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1719 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1720 registers overlay two consecutive single-precision registers. MIPS32R2
1721 introduced a new FPU mode (FR=1) where double-precision registers extend the
1722 corresponding single-precision registers which is incompatible with the
1723 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1724 require the use of FR=1 making a transition necessary. New o32 ABI
1725 extensions enable users to migrate over time from the original o32 ABI
1726 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1727 tracks the ABI of any loaded object and verifies that new objects are
1728 compatible. Mode transitions will also be requested as required and
1729 unsupportable objects will be rejected. The ABI checks include both soft and
1730 hard float ABIs for o32, n32 and n64.
1732 GCC 5 with GNU binutils 2.25 onwards:
1733 It is strongly recommended that all o32 system libraries are built using the
1734 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1735 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1736 compiler using --with-fp-32=xx to set this by default.
1740 * The following bugs are resolved with this release:
1742 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1743 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1744 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1745 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1746 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1747 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1748 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1749 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1750 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1751 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1752 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1753 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1754 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1755 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1756 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1758 * Reverted change of ABI data structures for s390 and s390x:
1759 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1760 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1761 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1762 are preserved pointing straight to the same implementation as the old ones.
1763 Given that, new callers will simply provide a too-big buffer to these
1764 functions. Any applications/libraries out there that embed jmp_buf or
1765 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1766 against 2.19 headers will have to rebuilt again. This is necessary in any
1767 case to revert the breakage in their ABI caused by the glibc change.
1769 * Support for file description locks is added to systems running the
1770 Linux kernel. The standard file locking interfaces are extended to
1771 operate on file descriptions, not file descriptors, via the use of
1772 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1773 are associated with an open file instead of a process.
1775 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1777 * The minimum Linux kernel version that this version of the GNU C Library
1778 can be used with is 2.6.32.
1780 * Running the testsuite no longer terminates as soon as a test fails.
1781 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1782 with PASS or FAIL lines for individual tests. A summary of the results is
1783 printed, including a list of failing lists, and "make check" exits with
1784 error status if there were any unexpected failures. "make check
1785 stop-on-test-failure=y" may be used to keep the old behavior.
1787 * The am33 port, which had not worked for several years, has been removed
1790 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1791 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1792 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1793 conflicted with POSIX (support for which was removed in 2.19), the
1794 interfaces those macros enabled remain available when compiling with
1795 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1796 test macros defined.
1798 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1800 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1801 This may improve lock scaling of existing programs on TX capable systems.
1802 The lock elision code is only built with --enable-lock-elision=yes and
1803 then requires a GCC version supporting the TX builtins. With lock elision
1804 default mutexes are elided via __builtin_tbegin, if the cpu supports
1805 transactions. By default lock elision is not enabled and the elision code
1808 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1809 copy the path argument. This allowed programs to cause posix_spawn to
1810 deference a dangling pointer, or use an unexpected pathname argument if
1811 the string was modified after the posix_spawn_file_actions_addopen
1814 * All supported architectures now use the main glibc sysdeps directory
1815 instead of some being in a separate "ports" directory (which was
1816 distributed separately before glibc 2.17).
1818 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1819 On configurations that support it (all Linux configurations), it's now
1820 used regardless of the --enable-add-ons switch to configure. It is no
1821 longer possible to build such configurations without pthreads support.
1823 * Locale names, including those obtained from environment variables (LANG
1824 and the LC_* variables), are more tightly checked for proper syntax.
1825 setlocale will now fail (with EINVAL) for locale names that are overly
1826 long, contain slashes without starting with a slash, or contain ".." path
1827 components. (CVE-2014-0475) Previously, some valid locale names were
1828 silently replaced with the "C" locale when running in AT_SECURE mode
1829 (e.g., in a SUID program). This is no longer necessary because of the
1832 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1833 with application code using Intel MPX instructions. (With all previous
1834 versions, the MPX register state could be clobbered when making calls
1835 into or out of a shared library.) Note that while the new dynamic
1836 linker is compatible with all known x86 hardware whether or not it
1837 supports Intel MPX, some x86 instruction-set emulators might fail to
1838 handle the new instruction encodings. This is known to affect Valgrind
1839 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1840 release), and might affect other tools that do instruction emulation.
1842 * Support for loadable gconv transliteration modules has been removed.
1843 The support for transliteration modules has been non-functional for
1844 over a decade, and the removal is prompted by security defects. The
1845 normal gconv conversion modules are still supported. Transliteration
1846 with //TRANSLIT is still possible, and the //IGNORE specifier
1847 continues to be supported. (CVE-2014-5119)
1849 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1850 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1851 resulting a denial-of-service security vulnerability in applications which
1852 use functions related to iconv. (CVE-2014-6040)
1856 * The following bugs are resolved with this release:
1858 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1859 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1860 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1861 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1862 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1863 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1864 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1865 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1866 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1867 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1868 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1869 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1870 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1871 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1872 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1873 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1874 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1875 16431, 16453, 16474, 16506, 16510, 16529
1877 * Slovenian translations for glibc messages have been contributed by the
1878 Translation Project's Slovenian team of translators.
1880 * The public headers no longer use __unused nor __block. This change is to
1881 support compiling programs that are derived from BSD sources and use
1882 __unused internally, and to support compiling with Clang's -fblock
1883 extension which uses __block.
1885 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1886 large collation sequences to optimize multiple passes. This cache
1887 computation may overflow for large collation sequences and may cause a
1888 stack or buffer overflow. This is now fixed to use a slower algorithm
1889 which does not use a cache if there is an integer overflow.
1891 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1892 rules for large collation sequences to optimize multiple passes and falls
1893 back to alloca if malloc fails, resulting in a possible stack overflow.
1894 The implementation now falls back to an uncached collation sequence lookup
1897 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1898 initialized for static applications resulting in the security feature
1899 being disabled. The pointer guard is now correctly initialized to a
1900 random value for static applications. Existing static applications need
1901 to be recompiled to take advantage of the fix (bug 15754).
1903 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1904 to the d_name member of struct dirent, or omit the terminating NUL
1905 character. (Bugzilla #14699).
1907 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1908 aligned_alloc functions could allocate too few bytes or corrupt the
1909 heap when passed very large allocation size values (Bugzilla #15855,
1912 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1913 for AF_INET6 has been fixed (Bugzilla #16072).
1915 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1916 pap_CW, quz_PE, the_NP.
1918 * Substantially revised locales: gd_GB, ht_HT
1920 * The LC_ADDRESS field was updated to support country_car for almost all
1923 * ISO 1427 definitions were updated.
1925 * ISO 3166 definitions were updated.
1927 * The localedef utility now supports --big-endian and --little-endian
1928 command-line options to generate locales for a different system from that
1929 for which the C library was built.
1931 * Binary locale files now only depend on the endianness of the system for
1932 which they are generated and not on other properties of that system. As a
1933 consequence, binary files generated with new localedef may be incompatible
1934 with old versions of the GNU C Library, and binary files generated with
1935 old localedef may be incompatible with this version of the GNU C Library,
1936 in the following circumstances:
1938 + Locale files may be incompatible on m68k systems.
1940 + Locale archive files (but not separate files for individual locales) may
1941 be incompatible on systems where plain "char" is signed.
1943 * The configure option --disable-versioning has been removed. Builds with
1944 --disable-versioning had not worked for several years.
1946 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1947 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1949 * SystemTap probes for malloc have been introduced.
1951 * SystemTap probes for slow multiple precision fallback paths of
1952 transcendental functions have been introduced.
1954 * Support for powerpc64le has been added.
1956 * The soft-float powerpc port now supports e500 processors.
1958 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1960 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1961 set of header declarations that are enabled by default, even when other
1962 feature test macros or compiler options such as -std=c99 would otherwise
1963 disable some of those declarations.
1965 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1966 conflict with POSIX. The libbsd-compat library (which was a dummy library
1967 that did nothing) has also been removed.
1969 * Preliminary documentation about Multi-Thread, Async-Signal and
1970 Async-Cancel Safety has been added.
1972 * Change of ABI data structures for s390 and s390x:
1973 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1974 allow for future hardware extensions. All library functions that accept or
1975 return these structures were versioned in order to provide backwards
1976 compatibility with existing code. However, not all cases can be handled
1977 with interface versioning. If an application embeds either structure into
1978 another structure and that structure is passed to another compilation unit
1979 which expects the newer larger-sized object then any access to the new
1980 fields will result in undefined behaviour. Similarly any access to fields
1981 that were moved by the enlarging of the structures will cause undefined
1982 behaviour. To fix the undefined behaviour all objects that are part of the
1983 final in-memory image must agree on the size of structures, and this may
1984 require recompilation.
1988 * The following bugs are resolved with this release:
1990 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1991 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1992 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1993 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1994 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1995 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1996 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1997 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1998 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1999 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2000 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2001 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2002 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2005 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2006 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2007 Distributions can re-enable building and using pt_chown via the new configure
2008 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2009 considerable security risks and should only be used if the distribution
2010 understands and accepts the risks.
2012 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2015 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2016 fixed (Bugzilla #15330).
2018 * Add support for calling C++11 thread_local object destructors on thread
2019 and program exit. This needs compiler support for offloading C++11
2020 destructor calls to glibc.
2022 * Improved worst case performance of libm functions with double inputs and
2025 * Support for priority inherited mutexes in pthread condition variables on
2026 non-x86 architectures.
2028 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2030 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2032 * Optimized string functions for ARM. Implemented by Will Newton and
2035 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2037 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2040 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2041 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2043 * Added a benchmark framework to track performance of functions in glibc.
2045 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2046 It is based on draft TS 18661 and currently enabled as a GNU extension.
2048 * On Linux, the clock function now uses the clock_gettime system call
2049 for improved precision, rather than old times system call.
2051 * Added support for version-3 tz format files. This is needed when using
2052 the tz database release 2013e or later, and affects a few unusual cases --
2053 currently only TZ='America/Godthab' for time stamps after 2037.
2055 * Added new API functions pthread_getattr_default_np and
2056 pthread_setattr_default_np to get and set the default pthread
2057 attributes of a process.
2059 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2060 This may improve lock scaling of existing programs on TSX capable systems.
2061 When the --enable-lock-elision=yes parameter is specified at configure
2062 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2065 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2066 availability). Implemented by Ryan S. Arnold.
2068 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2072 * The following bugs are resolved with this release:
2074 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2075 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2076 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2077 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2078 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2079 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2080 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2081 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2082 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2083 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2084 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2085 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2086 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2087 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2088 14879, 14889, 14893, 14898, 14914.
2090 * Optimization of memcpy for MIPS.
2092 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2093 EMFILE has been fixed (Bugzilla #14889).
2095 * The add-on ports collection is now distributed in the "ports" subdirectory
2096 of the main GNU C Library distribution, rather than separately.
2098 * Port to ARM AArch64 contributed by Linaro.
2100 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2101 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2103 Implemented by Andreas Krebbel.
2105 * The new function secure_getenv allows secure access to the environment,
2106 returning NULL if running in a SUID/SGID process. This function replaces
2107 the internal function __secure_getenv.
2109 * SystemTap static probes have been added into the dynamic linker.
2110 Implemented by Gary Benson.
2112 * Optimizations of string functions strstr, strcasestr and memmem.
2113 Implemented by Maxim Kuvyrkov.
2115 * The minimum Linux kernel version that this version of the GNU C Library
2116 can be used with is 2.6.16.
2118 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2119 powerpc POWER7. Implemented by Will Schmidt.
2121 * New configure option --disable-nscd builds the C library such that it
2122 never attempts to contact the Name Service Caching Daemon (nscd).
2123 New configure option --disable-build-nscd avoids building nscd itself;
2124 this is the default if --disable-nscd is used.
2126 * Improved support for cross-compilation, including cross-testing and
2127 bootstrap builds without a previously built glibc.
2129 * Several testsuite tests are now able to test multiple IFUNC variants of an
2130 interface, rather than just testing the one that would be chooen by
2133 * New configure options --with-bugurl and --with-pkgversion, for
2134 distributors to use to embed their bug-reporting and package version
2135 information in --help and --version output.
2137 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2138 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2139 allows creation of chroots without the procfs mounted on /proc.
2141 * The `crypt' function now fails if passed salt bytes that violate the
2142 specification for those values. On Linux, the `crypt' function will
2143 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2144 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2145 when the mode is enabled.
2147 * The `clock_*' suite of functions (declared in <time.h>) is now available
2148 directly in the main C library. Previously it was necessary to link with
2149 -lrt to use these functions. This change has the effect that a
2150 single-threaded program that uses a function such as `clock_gettime' (and
2151 is not linked with -lrt) will no longer implicitly load the pthreads
2152 library at runtime and so will not suffer the overheads associated with
2153 multi-thread support in other code such as the C++ runtime library.
2155 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2160 * The following bugs are resolved with this release:
2162 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2163 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2164 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2165 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2166 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2167 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2168 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2169 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2170 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2171 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2172 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2173 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2174 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2175 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2176 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2177 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2178 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2179 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2180 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2181 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2182 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2183 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2186 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2187 configuring glibc with:
2188 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2189 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2190 Implemented by H.J. Lu.
2194 + define static_assert
2196 + do not declare gets
2198 + declare at_quick_exit and quick_exit also for ISO C11
2200 + aligned_alloc. NB: The code is deliberately allows the size parameter
2201 to not be a multiple of the alignment. This is a moronic requirement
2202 in the standard but it is only a requirement on the caller, not the
2205 + timespec_get added
2207 + uchar.h support added
2209 + CMPLX, CMPLXF, CMPLXL added
2211 Implemented by Ulrich Drepper.
2213 * Support for the IA-64 has been moved to ports.
2215 * Remove support for anything but ELF binary format
2217 * Checking versions of poll, ppoll added.
2218 Implemented by Ulrich Drepper.
2220 * More generic and 64-bit performance optimizations to math functions.
2221 Implemented by Ulrich Drepper.
2223 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2224 headers and functions available at compile time as they were before
2225 version 2.14. This option will be removed at some time in the future
2226 after the TI-RPC library becomes fully sufficient for the needs of
2227 existing applications.
2229 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2230 Note that glibc is not expected to work with any Linux kernel version
2233 * New header <sys/auxv.h> and function getauxval allowing easy access to
2234 the AT_* key-value pairs passed from the Linux kernel. The header also
2235 defines the HWCAP_* bits associated with the AT_HWCAP key.
2237 * New locales: mag_IN
2239 * New configure option --enable-systemtap builds SystemTap static probes
2240 into libc for setjmp and longjmp and into libpthread for various operations.
2241 So far the setjmp/longjmp probes and some of the libpthread probes are
2242 provided only for i*86 and x86_64.
2243 Implemented by Roland McGrath and Rayson Ho.
2245 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2247 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2250 * More optimized functions for SPARC. Implemented by David S. Miller.
2252 * Improved support for cross-compilation, especially bootstrap builds
2253 without a previously built glibc.
2255 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2256 Chris Metcalf from Tilera.
2258 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2259 now supported for ARM processors.
2261 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2262 as the name of the dynamic linker, to distinguish it from the
2263 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2265 * Support for CRIS has been removed from ports.
2267 * A new class of installed header has been documented for low-level
2268 platform-specific functionality. PowerPC added the first instance with a
2269 function to provide time base register access. Contributed by Tulio
2270 Magno Quites Machado Filho.
2272 * ix86 configurations now install header files that are consistent with
2273 what x86-64 configurations install. These same header files can be used
2274 for -m32, -m64, or -mx32 builds.
2275 Contributed by H.J. Lu.
2277 * Math library bug fixes. A thorough audit of all open math library bugs was
2278 conducted by Joseph Myers. Significant progress was made on many math
2279 library bugs resulting in more accurate exceptions and function results.
2280 Many thanks to all those that contributed including Andreas Jaeger for his
2281 patch review and work on the x87 trigonometric instruction issues.
2283 * Timezone data is no longer installed. Timezone-related binaries and scripts
2284 will continue to be installed. Users should obtain their timezone data from
2285 their distribution provider or from the tzdata package at
2286 <ftp://munnari.oz.au/pub/>.
2290 * The following bugs are resolved with this release:
2292 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2293 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2294 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2295 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2296 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2297 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2298 13484, 13506, 13515, 13523, 13524, 13538, 13540
2300 * New program pldd to list loaded object of a process
2301 Implemented by Ulrich Drepper.
2303 * Add nss_db support back to glibc. No more dependency on Berkeley db
2304 and support for initgroups lookups.
2305 Implemented by Ulrich Drepper.
2307 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2308 Contributed by HJ Lu.
2310 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2311 Contributed by HJ Lu.
2313 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2314 on x86-32 and x86-64.
2315 Contributed by Liubov Dmitrieva.
2317 * Optimized strchr and strrchr for SSE on x86-32.
2318 Contributed by Liubov Dmitrieva.
2320 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2321 for x86-64 and x86-32.
2322 Contributed by Liubov Dmitrieva.
2324 * New interfaces: scandirat, scandirat64
2325 Implemented by Ulrich Drepper.
2327 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2328 Implemented by Ulrich Drepper.
2330 * nscd now also caches the netgroup database.
2331 Implemented by Ulrich Drepper.
2333 * Integrate libm with gcc's -ffinite-math-only option.
2334 Implemented by Ulrich Drepper.
2336 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2337 to math functions. Implemented by Ulrich Drepper.
2339 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2340 Implemented by Ulrich Drepper.
2342 * New Linux interfaces: process_vm_readv, process_vm_writev
2344 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2345 Implemented by Ulrich Drepper.
2347 * Optimized nearbyint and strcasecmp for PPC.
2348 Implemented by Adhemerval Zanella.
2350 * New locales: bho_IN, unm_US, es_CU, ta_LK
2354 * The following bugs are resolved with this release:
2356 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2357 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2358 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2359 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2360 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2361 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2362 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2363 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2364 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2365 12795, 12811, 12813, 12814, 12841
2367 * The RPC implementation in libc is obsoleted. Old programs keep working
2368 but new programs cannot be linked with the routines in libc anymore.
2369 Programs in need of RPC functionality must be linked against TI-RPC.
2370 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2372 Visible changes of this change include (obviously) the inability to link
2373 programs using RPC functions without referencing the TI-RPC library and the
2374 removal of the RPC headers from the glibc headers.
2375 Implemented by Ulrich Drepper.
2377 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2378 syncfs, setns, sendmmsg
2380 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2381 yue_HK, lij_IT, mhr_RU
2383 * New encodings: CP770, CP771, CP772, CP773, CP774
2385 * New program sotruss to trace calls through PLTs
2386 Implemented by Ulrich Drepper.
2388 * The malloc hook implementation is marked deprecated and will be removed
2389 from the default implementation in the next version. The design never
2390 worked ever since the introduction of threads. Even programs which do
2391 not create threads themselves can use multiple threads created internally.
2395 * The following bugs are resolved with this release:
2397 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2398 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2399 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2400 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2403 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2405 * POWER7 optimizations: memset, memcmp, strncmp
2407 * New optimized string functions for x86-64: strnlen (SSE2),
2408 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2409 Implemented by Ulrich Drepper.
2413 * The following bugs are resolved with this release:
2415 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2416 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2417 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2418 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2419 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2420 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2421 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2422 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2424 * New interfaces: pthread_getname_np, pthread_setname_np
2426 * New Linux interface: recvmmsg
2428 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2430 * The dynamic linker now recognizes supported ABI versions from the
2431 EI_ABIVERSION field in the ELF header.
2432 Implemented by Ulrich Drepper.
2434 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2435 will not be used to fill in password fields in the passwd.byname replies.
2436 Instead it is used to synthesize the shadow.byname table, should it be
2437 missing. This is a useful mode in some installations involving Solaris.
2438 Implemented by Ulrich Drepper.
2440 * New locales: kok_IN, sq_MK, cv_RU
2444 * The following bugs are resolved with this release:
2446 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2447 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2448 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2449 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2450 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2451 10789, 10817, 10823, 10840, 10847
2453 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2455 Implemented by Ulrich Drepper.
2457 * Checking version of longjmp added that fails if an uninitialized stack
2458 frame would be created. Implemented by Ulrich Drepper.
2460 * STT_GNU_IFUNC is now supported in static executables.
2461 Implemented by H.J. Lu.
2463 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2465 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2466 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2467 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2468 Contributed by H.J. Lu.
2470 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2471 Implemented by Ulrich Drepper.
2473 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2475 Contributed by H.J. Lu.
2477 * Support for fma instruction in AVX on x86-64.
2478 Implemented by H.J. Lu and Ulrich Drepper.
2480 * AVX support in x86-64 auditing support in ld.so.
2481 Implemented by H.J. Lu.
2483 * STB_GNU_UNIQUE support added.
2484 Implemented by Ulrich Drepper.
2486 * Implement second fallback mode for optimized DNS lookup for even more
2487 broken environments. If two requests from the same port are not
2488 handled correctly close the socket and open a new one before sending
2489 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2490 can be used to select this mode right away, instead of rediscovering the
2491 necessity is every process again.
2492 Implemented by Ulrich Drepper.
2494 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2495 Implemented by Adam Tkac.
2497 * Optimized iconv conversions for S390x.
2498 Implemented by Andreas Krebbel.
2500 * Using condvars with PI mutexes is now more efficient due to kernel
2501 support for requeueing to PI futexes. NPTL support added for x86-64.
2502 Implemented by Ulrich Drepper.
2504 * New locales: ps_AF, my_MM
2508 * The following bugs are resolved with this release:
2510 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2511 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2512 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2513 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2514 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2516 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2517 Implemented by Ulrich Drepper.
2519 * New Linux interfaces: accept4, fallocate, fallocate64.
2520 Implemented by Ulrich Drepper.
2522 * Correct declarations of string function when used in C++ code. This
2523 could lead to compile errors for invalid C++ code.
2525 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2528 * New POSIX 2008 interface: psiginfo
2529 Implemented by Ulrich Drepper.
2531 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2532 Implemented by Ulrich Drepper.
2534 * Support for selecting between multiple function definitions at runtime
2535 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2537 * The libcrypt library can now use the hash function implementations in
2538 NSS. Implemented by Ulrich Drepper.
2540 * The malloc implementation can be compiled to be less memory efficient
2541 but higher performing in multi-threaded programs.
2542 Implemented by Ulrich Drepper.
2544 * New locales: nan_TW@latin, ks_IN
2546 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2547 Implemented by Ulrich Drepper.
2549 * Extended printf hook support. It is possible to use user-defined types
2550 and extend existing format specifiers.
2551 Implemented by Ulrich Drepper.
2553 * Handling for group shadow files has been added.
2554 Implemented by Ulrich Drepper.
2556 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2557 servers (the case, e.g., for some people using the built-in DNS
2558 server in ADSL modems/routers). There is a once-per-process timeout
2559 in case of a broken server. To avoid it, users can run nscd or put
2560 'options single-request' in /etc/resolv.conf.
2561 Implemented by Ulrich Drepper.
2565 * The following bugs are resolved with this release:
2567 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2568 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2569 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2570 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2571 6968, 6974, 6980, 6995, 7008, 7009, 7029
2573 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2574 up at the same time. Implemented by Ulrich Drepper.
2576 * TLS descriptors for LD and GD on x86 and x86-64.
2577 Implemented by Alexandre Oliva.
2579 * getaddrinfo now handles DCCP and UDPlite.
2580 Implemented by Ulrich Drepper.
2582 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2583 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2584 Implemented by Ulrich Drepper.
2586 * New implementation of memmem, strstr, and strcasestr which is O(n).
2587 Implemented by Eric Blake.
2589 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2591 * Implement "e" option for popen to open file descriptor with the
2592 close-on-exec flag set. Implemented by Ulrich Drepper.
2594 * Implement "b" mode for fmemopen. In this mode writes writes don't
2595 implicitly add a NUL byte and seeks from the end of the buffer really
2596 use the buffer end, not the string length as the basis.
2597 Implemented by Ulrich Drepper.
2599 * Many functions, exported and internal, now atomically set the close-on-exec
2600 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2602 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2604 Implemented by Pravin Satpute.
2606 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2610 * The following bugs are resolved with this release:
2612 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2613 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2614 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2615 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2616 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2617 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2618 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2620 * New locales: bo_CN, bo_IN, shs_CA.
2622 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2624 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2625 Implemented by Pravin Satpute.
2627 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2629 * nscd caches now all timeouts for DNS entries
2630 Implemented by Ulrich Drepper.
2632 * nscd is more efficient and wakes up less often.
2633 Implemented by Ulrich Drepper.
2635 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2636 vdprintf, and obstack_vprintf.
2637 Implemented by Jakub Jelinek.
2639 * Faster memset for x86-64.
2640 Implemented by Harsha Jagasia and H.J. Lu.
2642 * Faster memcpy on x86.
2643 Implemented by Ulrich Drepper.
2645 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2646 Implemented by Ulrich Drepper.
2648 * Faster sqrt and sqrtf implemention for some PPC variants.
2649 Implemented by Steven Munroe.
2653 * The following bugs are resolved with this release:
2655 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2656 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2657 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2658 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2659 5071, 5103, 5104, 5112, 5113, 5184, 5186
2661 * More checking functions: fread, fread_unlocked, open*, mq_open.
2662 Implemented by Jakub Jelinek and Ulrich Drepper.
2664 * Extend fortification to C++. Implemented by Jakub Jelinek.
2666 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2667 by not recognizing 'a' as a modifier when those specs are requested.
2668 Implemented by Jakub Jelinek.
2670 * PPC optimizations to math and string functions.
2671 Implemented by Steven Munroe.
2673 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2674 options to be passed. Implemented by Ulrich Drepper.
2676 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2678 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2679 Implemented by Ulrich Drepper.
2681 * Handle private futexes in the NPTL implementation.
2682 Implemented by Jakub Jelinek and Ulrich Drepper.
2684 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2685 Implemented by Roland McGrath and Ulrich Drepper.
2687 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2689 * SHA-256 and SHA-512 based password encryption.
2690 Implemented by Ulrich Drepper.
2692 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2693 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2696 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2697 Implemented by Ulrich Drepper.
2701 * The following bugs are resolved with this release:
2703 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2704 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2705 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2706 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2707 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2708 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2709 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2712 * New Linux interfaces: epoll_pwait, sched_getcpu.
2714 * New generic interfaces: strerror_l.
2716 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2720 * The following bugs are resolved with this release:
2722 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2723 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2724 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2725 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2726 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2727 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2728 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2729 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2730 3225, 3251, 3252, 3253, 3273
2732 * For Linux, the sorting of addresses returned by getaddrinfo now also
2733 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2734 Implemented by Ulrich Drepper.
2736 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2737 Implemented by Ulrich Drepper.
2739 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2741 * New iconv module for MIK. Contributed by Alexander Shopov.
2743 * For sites with broken group and/or passwd database, the auto-propagate
2744 option of nscd can prevent creating ID lookup entries from the results
2745 of a name lookup and vice versa. This usually is no problem but some
2746 site might have problems with the default behavior.
2747 Implemented by Ulrich Drepper.
2749 * Iterating over entire database in NIS can be slow. With the
2750 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2751 to trade time for memory. The entire database will be read at once.
2752 Implemented by Ulrich Drepper.
2754 * The interfaces introduced in RFC 3542 have been implemented by
2757 * Support for the new ELF hash table format was added by Ulrich Drepper.
2759 * Support for priority inheritance mutexes added by Jakub Jelinek and
2762 * Support for priority protected mutexes added by Jakub Jelinek.
2764 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2765 or_IN, csb_PL, fy_NL, sr_ME.
2769 * More overflow detection functions.
2771 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2772 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2774 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2775 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2776 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2777 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2778 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2779 by Masahide Washizawa.
2781 * It is now possible to install audit libraries for the dynamic linker using
2782 LD_AUDIT. Implemented by Ulrich Drepper.
2784 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2785 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2786 For a libc and libpthread that works well on Linux 2.4 kernels, we
2787 recommend using the stable 2.3 branch.
2789 * The new function open_wmemstream acts like open_memstream,
2790 but uses a wchar_t wide character buffer.
2792 * The new function ppoll is an improved version of poll, similar to pselect.
2794 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2795 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2796 renameat, unlinkat, linkat, symlinkat, readlinkat.
2798 * New Linux kernel system calls: unshare,
2799 inotify_init, inotify_add_watch, inotify_rm_watch.
2801 * The euidaccess function is now also known by the alias eaccess,
2802 for compatibility with some other systems.
2804 * Timezone data updated to 2006b version.
2808 * The following bugs are resolved with this release:
2810 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2811 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2812 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2813 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2814 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2815 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2817 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2819 * As of this release, GCC 4 can be used to compile the C Library.
2821 * Timezone data updated to 2005m version.
2825 * The following bugs are resolved with this release:
2827 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2828 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2829 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2830 777, 787, 821, 822, 823, 825
2832 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2836 * Support for RFC 3678. Real implementations exist only for Linux so far.
2837 Implemented by Ulrich Drepper.
2839 * nscd can now cache entries persistently. Expiring entries are reloaded.
2840 For speedups the cache can be shared in memory with client processes.
2841 Implemented by Ulrich Drepper.
2843 * nscd can now perform SELinux checks.
2844 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2846 * getaddrinfo queries are now cached. Canonical name lookup is performed
2848 Implemented by Ulrich Drepper.
2850 * The nothrow function attribute is used when headers are used by gcc when
2851 compiling C code. This can avoid significant amounts of exception
2854 * The malloc functions perform more error checking and are stricter when
2855 it comes to reacting on errors. The default action is to terminate
2856 the process after showing an error message. Implemented by Ulrich Drepper.
2858 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2859 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2861 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2862 namespaces using the new function dlmopen(). This feature is of course,
2863 like most other dynamic loading functionality, not available in statically
2864 linked applications. Implemented by Ulrich Drepper.
2866 * Low-overhead boundary checking variants of string and some stdio functions
2867 were added. These are to be used in conjunction with a gcc patch by
2868 Jakub Jelinek which adds calls to these functions if possible.
2869 Implemented by Jakub Jelinek and Ulrich Drepper.
2871 * Old code for several operating systems and machine architectures that
2872 have not been in working condition in a long time have been removed from
2873 the main source tree maintained by the GNU C Library's maintainers.
2874 These files are now reside in the separate `ports' source module
2875 that is usable as an add-on when building the library.
2879 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2880 interrogate the dynamic linker, compatible with the Solaris interface.
2882 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2883 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2885 * getifaddrs now uses the netlink interface on Linux to get its information.
2886 Implemented by Thorsten Kukuk.
2888 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2889 Implemented by Ulrich Drepper.
2891 * support for non-executable stacks on x86 has been added. Changes mostly
2894 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2897 * getaddrinfo now performs destination address selection according to
2902 * Thread-safe interfaces for many functions that access locale data
2903 were added in version 2.3, but these features were omitted from NEWS.
2904 Many functions have variants with an `_l' suffix that take a `locale_t'
2905 object as a parameter rather than consulting the current locale.
2906 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2907 create and maintain `locale_t' objects. Additionally, the new function
2908 `uselocale' sets "the current locale" (as used by functions not so
2909 parameterized) set for an individual thread. These features were added
2910 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2912 * The functions getresuid, getresgid, setresuid, and setresgid, which
2913 have long been available on Linux, are now declared in <unistd.h>
2914 and are now also available on the Hurd.
2916 * ELF thread-local storage support (TLS) now works on x86-64.
2918 * The new dynamic string token $LIB is expanded in shared library names.
2919 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2921 * Aldy Hernandez contributed complete software floating point support for
2922 PowerPC machines with no FPU.
2924 * fexecve is implemented on Linux.
2926 * The `btowc' function should work at least twice as fast due to
2927 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2929 * With approriate thread add-ons cancelable functions are now implemented
2930 in libc.so as well. No need to call the function in libpthread. This
2931 change allowed to finally disable the incorrect and expensive handling
2932 of weak definition in ld.so.
2934 * Yet more PLT entries in libc.so have been removed. We finally arrived
2935 at the bare minimum. Startup times improved appropriately.
2937 * Support for the new Linux/x86 system call interface was added. The
2938 AT_SYSINFO auxiliary vector entry is recognized and handled.
2942 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2945 * iconv (the program and the interface) now accepts empty names (excluding
2946 options like //TRANSLIT) to mean "use charset of current locale".
2948 * localedef can now transliterate characters in strings which are not in
2949 the provided charmap. The information from the input locale is used.
2951 * Prelinking support was added for ELF targets. This requires additional
2952 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2954 * Read-only stdio streams now use mmap to speed up operation by eliminating
2955 copying and buffer underflows. To use add 'm' to the mode string of
2956 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2958 * The malloc functions were completely rewritten by Wolfram Gloger based
2959 on Doug Lea's malloc-2.7.0.c.
2961 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2962 implementation of regex.
2964 * Bruno Haible upgraded the iconv and locale implementation to support
2967 * Contents of the LC_* and LANG environment variables in the CEN style are
2968 not recognized anymore. It never was used. Change by Ulrich Drepper.
2970 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2971 thread-local storage (TLS) ABI on some platforms.
2972 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2974 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2975 EUC-JISX0213, and TSCII.
2977 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2978 BSD-compatible interface for getting all network interface addresses.
2979 Implementation for IPv4 by Roland McGrath.
2981 * Loading of locale data is faster due to the introduction of a locale
2982 archive. Implemented by Roland McGrath and Ulrich Drepper.
2984 * Startup times are significantly reduced by not using exported functions
2985 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2988 * Steven Munroe contributed a port to PowerPC64/Linux.
2992 * The Hurd now uses the GNU libio implementation of stdio rather than the
2993 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2995 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2996 with the same behavior as the Linux system call of the same name.
3000 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3001 128-bit long double format.
3003 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3004 IBM1161, and IBM1162 charsets.
3006 * Andreas Jaeger contributed a port to x86-64/Linux.
3008 * Peter Bruin contributed a port to PowerPC/Hurd.
3010 * libc functions using I/O streams now can handle wide-oriented streams
3013 * optimizations in the dynamic linker. Binaries created by recent binutils
3014 versions start up quicker due to reduced time spent on relocations.
3016 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3020 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3021 asinh, atanh, j0 for the 128-bit long double format.
3023 * Bruno Haible updated all the code handling Unicode in some form to
3024 support Unicode 3.1.
3026 * Speed of regex for single-byte locales is back to previous levels.
3027 Patch by Isamu Hasegawa.
3029 * Alpha, SPARC, and IA-64 now also using floating stacks.
3031 * Startup time of internationalized applications greatly improved through
3032 iconv cache. Use iconvconfig to generate the cache file.
3033 Contributed by Ulrich Drepper.
3035 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3036 severe performance problems. Patch by David Mosberger.
3038 * The Hurd port got a lot more functionality like AIO, various stdio
3039 extensions, etc. Mainly done by Roland McGrath.
3041 * mtrace can now lookup symbols in shared libraries.
3045 * Intel's IA-64 math library is largely integrated. It provides fast and
3046 accurate implementatations for most basic and standard math functions
3047 in float, double, and long double format.
3049 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3050 and asin for the 96-bit long double format and asin, log, tan for the
3051 128-bit long double format.
3053 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3054 The basic double functions exist today. Contributed by Abraham Ziv
3055 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3056 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3058 * An asynchronous name lookup library was added. The interface is designed
3059 after POSIX AIO. The proposal was circulated beforehand to get comments.
3060 No negative ones came in. Implemented by Ulrich Drepper.
3062 * Port to S390/64bit contributed by Martin Schwidefsky
3063 <schwidefsky@de.ibm.com>.
3065 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3066 of functions for Linux/IA-64.
3068 * The RPC code is now thread safe. Threads can now use the same service
3069 of different services at the same time. Patch by Eric Norum
3070 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3072 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3073 family of functions for Linux/S390.
3075 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3076 of functions for Linux/x86.
3078 * Port to Linux/CRIS contributed by Axis Communications.
3082 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3083 we can now check for namespace violations and missing declarations. The
3084 result is that almost all headers are now Unix-compliant (as defined in
3085 the upcoming XPG6). The negative side is that some programs might need
3086 corrections, too, if they depend on the incorrect form of the headers in
3087 previous versions which defined too many symbols and included too many
3090 * regex now handles multibyte character sets correctly.
3091 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3093 * iconv (the program) does now conform to the upcoming XPG6 and handles
3094 charmaps. Instead of the charset names the path of charmaps can be
3095 provided and the conversion happens based on this data.
3096 Contributed by Ulrich Drepper.
3098 * The locale program now provides detailed information about the installed
3103 only lists the names of the supported locales
3107 provides details such as country, language, and codeset name.
3108 Contributed by Ulrich Drepper.
3112 * The gencat program now parses the input file according to the charset
3113 selected by the LC_CTYPE category. This is important for stateful
3114 character sets. To make generating catalogs easier there is a way
3115 to overwrite the charset selected by the locale: before the first
3116 message or $ quote line the catalog can contain a line like
3118 $ codeset=ISO-8859-2
3120 to select the charset (ISO-8859-2 in this case).
3122 Implemented by Shinya Hanataka and Ulrich Drepper.
3124 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3125 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3126 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3128 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3129 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3131 * The LANGUAGE environment variable is now ignored unless the locale is
3132 changed from the default "C" locale.
3134 * The usual bug fixes.
3138 * Greg McGary added runtime support for bounds checking using gcc's
3139 new -fbounded-pointers option. ix86 target is complete. PowerPC
3142 * Thorsten Kukuk added secure mode support to nscd.
3144 * The Berkeley db libraries have been removed.
3146 Related, the nss_db module is now in a separate package since it
3147 obviously requires a database library being available.
3149 * Wide character I/O streams implemented by Ulrich Drepper.
3151 * Functions from the extended socket API added by Ulrich Drepper.
3153 * Functions feenableexcept and fedisableexcept to control the
3154 behaviour of individual exceptions have been added by Andreas Jaeger.
3156 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3158 * The resolver code has been updated from bind 8.2.3-T5B which supports
3159 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3162 This change could in some situations effect backward compatibility. Since
3163 now `_res' is a thread-local instead of a global variable, modifying it
3164 in one thread does not have any effect in other threads.
3166 The resolver library was also extended to allow IPv6 as the transport
3167 protocol for the requests. This work was done by Stig Venaas.
3169 * Compatibility code for K&R C compilers has been removed from the
3170 header files. A ISO C compiler is needed to use the library
3171 (conforming to either C89 or C99 standard).
3173 * Complete rewrite of the localedef program to support multibyte character
3174 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3175 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3176 Implemented by Ulrich Drepper.
3178 Bruno Haible significantly improved the generation and use of the data
3179 structures for the wide character tables.
3181 * Plural handling in gettext implemented by Ulrich Drepper.
3183 * The utmp daemon has been removed.
3185 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3187 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3190 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3192 * POSIX barriers implemented by Kaz Kylheku.
3194 * POSIX spawn function family implemented by Ulrich Drepper.
3196 * POSIX spinlocks are now available.
3198 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3200 * the configure option --enable-kernel=X.Y.Z allows to strip out
3201 compatibility for kernel versions before X.Y.Z. This is currently only
3202 implemented for Linux.
3204 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3205 field sin6_scope_id. This means that all programs using IPv6 should be
3206 recompiled. Don't expect binary compatibility with previous glibc
3209 * various conversion modules for IBM character sets contributed by
3212 * IA-64 port by Jes Sorensen and HJ Lu.
3226 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3228 * New xdr functions are added; some rpc functions are now 64bit clean.
3230 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3232 * Fixed known incompatibilities with glibc 2.0.
3234 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3236 * Optimized a number of functions (especially the ELF dynamic loader).
3238 * Update timezone data files.
3240 * lots of charmaps corrections
3242 * some new locale definitions and charmaps
3247 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3248 conform to POSIX member type requirements. Symbol versions have been
3249 adjusted as needed within the library, and for direct use by applications,
3250 but there is potential for problems if third-party libraries use
3251 struct timeval as part of their interface. This does not present
3252 a problem for X and other "essential" system libraries.
3254 * An additional locale model to support C++ Standard Library locale
3255 model and probably more was implemented by Ulrich Drepper.
3257 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3260 * Miles Bader provided the `argp' function family to support hierachical
3261 command line argument parsing, layered on top of getopt.
3263 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3265 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3268 * scanf recognizes the %a and %A format for scanning floating point numbers.
3270 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3271 information and interfaces for the available integer types.
3273 * about 130 new math functions were added to implement the ISO C9x math
3276 * the new header <complex.h> contains definitions of the complex math
3277 functions from ISO C 9X.
3279 * the new header <tgmath.h> defines generic macros to use complex or
3280 real valued functions.
3282 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3284 * Andreas Jaeger provided a test suite for the math library.
3286 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3288 * Ulrich Drepper added character set conversion functions (iconv).
3290 * Optimized string functions have been added.
3292 * The localedata addon is now part of glibc.
3294 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3296 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3297 daemon for NSS (nscd).
3299 Missing a better place here are some numbers on improvements. Under
3300 Linux 2.1.125 un-tar-ing the kernel sources takes
3304 using local files 12.19s 6.88s 22.91s
3306 using NIS 13.92s 8.91s 26.34s
3308 using NIS & nscd 10.37s 7.34s 25.30s
3310 using NIS+ 27.57s 30.37s 640.46s
3312 using NIS+ & nscd 10.25s 7.83s 26.51s
3314 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3316 Keep in mind that non-namelookup related operations dominate above times.
3317 It was just a common complain that using NIS+ unpacking the kernel is
3320 [1] The old nscd implementation is not available anymore since it was
3321 distributed with glibc up to version 2.0.98 and thus is now replaced.
3323 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3325 * Mark Kettenis provided a Hesiod NSS module.
3327 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3328 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3330 * rcmd can now handle netgroups (Dick Streefland).
3332 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3335 * Support for the IPv6 protocol has been added to the socket API, as per the
3336 latest draft standards.
3338 * Support for Linux 2.2 has been added.
3340 * Interface changes relative to the latest 2.0.x release:
3341 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3342 addseverity NEW: Unix98
3343 alphasort64 NEW: LFS
3344 argp_err_exit_status NEW: argp, GNU ext
3345 argp_error NEW: argp, GNU ext
3346 argp_failure NEW: argp, GNU ext
3347 argp_help NEW: argp, GNU ext
3348 argp_parse NEW: argp, GNU ext
3349 argp_program_bug_address NEW: argp, GNU ext
3350 argp_program_version NEW: argp, GNU ext
3351 argp_program_version_hook NEW: argp, GNU ext
3352 argp_state_help NEW: argp, GNU ext
3353 argp_usage NEW: argp, GNU ext
3354 authdes_create NEW: Secure RPC
3355 authdes_getucred NEW: Secure RPC
3356 authdes_pk_create NEW: Secure RPC
3357 backtrace NEW: GNU ext.
3358 backtrace_symbols NEW: GNU ext.
3359 backtrace_symbols_fd NEW: GNU ext.
3361 cacosf NEW: ISO C 9x
3362 cacosh NEW: ISO C 9x
3363 cacoshf NEW: ISO C 9x
3364 cacoshl NEW: ISO C 9x
3365 cacosl NEW: ISO C 9x
3372 casinf NEW: ISO C 9x
3373 casinh NEW: ISO C 9x
3374 casinhf NEW: ISO C 9x
3375 casinhl NEW: ISO C 9x
3376 casinl NEW: ISO C 9x
3378 catanf NEW: ISO C 9x
3379 catanh NEW: ISO C 9x
3380 catanhf NEW: ISO C 9x
3381 catanhl NEW: ISO C 9x
3382 catanl NEW: ISO C 9x
3383 cbc_crypt NEW: Secure RPC
3387 ccoshf NEW: ISO C 9x
3388 ccoshl NEW: ISO C 9x
3394 cimagf NEW: ISO C 9x
3395 cimagl NEW: ISO C 9x
3396 clearerr_locked REMOVED
3397 clntunix_create NEW: sunrpc ext
3399 clog10 NEW: ISO C 9x
3400 clog10f NEW: ISO C 9x
3401 clog10l NEW: ISO C 9x
3411 cprojf NEW: ISO C 9x
3412 cprojl NEW: ISO C 9x
3414 crealf NEW: ISO C 9x
3415 creall NEW: ISO C 9x
3420 csinhf NEW: ISO C 9x
3421 csinhl NEW: ISO C 9x
3424 csqrtf NEW: ISO C 9x
3425 csqrtl NEW: ISO C 9x
3429 ctanhf NEW: ISO C 9x
3430 ctanhl NEW: ISO C 9x
3432 des_setparity NEW: Secure RPC
3433 ecb_crypt NEW: Secure RPC
3434 endutxent NEW: Unix98
3436 exp10f NEW: ISO C 9x
3437 exp10l NEW: ISO C 9x
3441 fattach NEW: STREAMS
3442 fdetach NEW: STREAMS
3446 feclearexcept NEW: ISO C 9x
3447 fegetenv NEW: ISO C 9x
3448 fegetexceptflag NEW: ISO C 9x
3449 fegetround NEW: ISO C 9x
3450 feholdexcept NEW: ISO C 9x
3452 feraiseexcept NEW: ISO C 9x
3453 ferror_locked REMOVED
3454 fesetenv NEW: ISO C 9x
3455 fesetexceptflag NEW: ISO C 9x
3456 fesetround NEW: ISO C 9x
3457 fetestexcept NEW: ISO C 9x
3458 feupdateenv NEW: ISO C 9x
3459 fflush_locked REMOVED
3463 fgets_unlocked NEW: GNU ext.
3464 fileno_locked REMOVED
3476 fputc_locked REMOVED
3477 fputs_unlocked NEW: GNU ext.
3478 fread_unlocked NEW: GNU ext.
3483 fstatvfs NEW: Unix98
3487 ftruncate64 NEW: LFS
3489 fwrite_unlocked NEW: GNU ext.
3490 gai_strerror NEW: IPv6
3494 getchar_locked REMOVED
3496 getdate_err NEW: Unix98
3497 getdate_r NEW: GNU ext.
3499 getnameinfo NEW: IPv6
3500 getnetname NEW: Secure RPC
3501 getpmsg NEW: STREAMS
3502 getpt NEW: Unix98 PTY
3503 getrlimit64 NEW: LFS
3504 getutxent NEW: Unix98
3505 getutxid NEW: Unix98
3506 getutxline NEW: Unix98
3507 glob64 NEW: GNU ext.
3508 globfree64 NEW: GNU ext.
3509 gnu_get_libc_release NEW: GNU ext.
3510 gnu_get_libc_version NEW: GNU ext.
3511 grantpt NEW: Unix98 PTY
3512 host2netname NEW: Secure RPC
3514 iconv_close NEW: iconv
3515 iconv_open NEW: iconv
3516 if_freenameindex NEW: IPv6
3517 if_indextoname NEW: IPv6
3518 if_nameindex NEW: IPv6
3519 if_nametoindex NEW: IPv6
3520 in6addr_any NEW: IPv6
3521 in6addr_loopback NEW: IPv6
3522 inet6_isipv4mapped NEW: IPv6
3523 isastream NEW: STREAMS
3524 iswblank NEW: GNU ext.
3525 key_decryptsession NEW: Secure RPC
3526 key_decryptsession_pk NEW: Secure RPC
3527 key_encryptsession NEW: Secure RPC
3528 key_encryptsession_pk NEW: Secure RPC
3529 key_gendes NEW: Secure RPC
3530 key_get_conv NEW: Secure RPC
3531 key_secretkey_is_set NEW: Secure RPC
3532 key_setnet NEW: Secure RPC
3533 key_setsecret NEW: Secure RPC
3534 llrint NEW: ISO C 9x
3535 llrintf NEW: ISO C 9x
3536 llrintl NEW: ISO C 9x
3537 llround NEW: ISO C 9x
3538 llroundf NEW: ISO C 9x
3539 llroundl NEW: ISO C 9x
3544 lrintf NEW: ISO C 9x
3545 lrintl NEW: ISO C 9x
3546 lround NEW: ISO C 9x
3547 lroundf NEW: ISO C 9x
3548 lroundl NEW: ISO C 9x
3550 makecontext NEW: Unix98
3551 mempcpy NEW: GNU ext.
3554 modify_ldt NEW: kernel
3558 nearbyint NEW: ISO C 9x
3559 nearbyintf NEW: ISO C 9x
3560 nearbyintl NEW: ISO C 9x
3561 netname2host NEW: Secure RPC
3562 netname2user NEW: Secure RPC
3563 nexttoward NEW: ISO C 9x
3564 nexttowardf NEW: ISO C 9x
3565 nexttowardl NEW: ISO C 9x
3569 passwd2des NEW: Secure RPC
3571 pow10f NEW: GNU ext.
3572 pow10l NEW: GNU ext.
3575 printf_size NEW: GNU ext.
3576 printf_size_info NEW: GNU ext.
3577 profil_counter REMOVED
3578 pthread_mutexattr_getkind_np REPLACED
3579 pthread_mutexattr_setkind_np REPLACED
3580 ptsname NEW: Unix98 PTY
3581 ptsname_r NEW: Unix98 PTY
3583 putchar_locked REMOVED
3584 putgrent NEW: GNU ext.
3586 putpmsg NEW: STREAMS
3587 pututxline NEW: Unix98
3591 readdir64_r NEW: LFS
3592 remquo NEW: ISO C 9x
3593 remquof NEW: ISO C 9x
3594 remquol NEW: ISO C 9x
3596 roundf NEW: ISO C 9x
3597 roundl NEW: ISO C 9x
3599 scalbln NEW: ISO C 9x
3600 scalblnf NEW: ISO C 9x
3601 scalblnl NEW: ISO C 9x
3603 sendfile NEW: kernel
3604 setcontext NEW: Unix98
3605 setrlimit64 NEW: LFS
3606 setutxent NEW: Unix98
3608 sigignore NEW: Unix98
3609 sigqueue NEW: POSIX.1b
3610 sigrelse NEW: Unix98
3611 sigset NEW: POSIX.1b
3612 sigtimedwait NEW: POSIX.1b
3613 sigwaitinfo NEW: POSIX.1b
3614 sincos NEW: GNU ext.
3615 sincosf NEW: GNU ext.
3616 sincosl NEW: GNU ext.
3620 strcasestr NEW: GNU ext.
3621 strtoimax NEW: ISO C 9x
3622 strtoumax NEW: ISO C 9x
3623 strverscmp NEW: GNU ext.
3624 svcauthdes_stats NEW: Secure RPC
3625 svcunix_create NEW: sunrpc etx
3626 svcunixfd_create NEW: sunrpc ext
3627 swapcontext NEW: Unix98
3628 tcgetsid NEW: Unix98 PTY
3629 tdestroy NEW: GNU ext.
3630 tgamma NEW: ISO C 9x
3631 tgammaf NEW: ISO C 9x
3632 tgammal NEW: ISO C 9x
3636 truncf NEW: ISO C 9x
3637 truncl NEW: ISO C 9x
3639 unlockpt NEW: Unix98 PTY
3640 updwtmpx NEW: Unix98
3641 user2netname NEW: Secure RPC
3642 utmpxname NEW: Unix98
3643 versionsort NEW: GNU ext.
3644 versionsort64 NEW: GNU ext.
3646 wcscasecmp NEW: GNU ext.
3647 wcsncasecmp NEW: GNU ext.
3648 wcsnlen NEW: GNU ext.
3649 wcstoimax NEW: ISO C 9x
3650 wcstoll NEW: ISO C 9x
3651 wcstoull NEW: ISO C 9x
3652 wcstoumax NEW: ISO C 9x
3654 wordexp NEW: POSIX.2
3655 wordfree NEW: POSIX.2
3656 write_profiling REMOVED
3657 xdecrypt NEW: Secure RPC
3658 xdr_authdes_cred NEW: Secure RPC
3659 xdr_authdes_verf NEW: Secure RPC
3660 xdr_cryptkeyarg NEW: Secure RPC
3661 xdr_cryptkeyarg2 NEW: Secure RPC
3662 xdr_cryptkeyres NEW: Secure RPC
3663 xdr_getcredres NEW: Secure RPC
3664 xdr_key_netstarg NEW: Secure RPC
3665 xdr_key_netstres NEW: Secure RPC
3666 xdr_keybuf NEW: Secure RPC
3667 xdr_keystatus NEW: Secure RPC
3668 xdr_netnamestr NEW: Secure RPC
3669 xdr_sizeof NEW: Secure RPC
3670 xdr_unixcred NEW: sunrpc ext
3671 xencrypt NEW: Secure RPC
3672 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3683 * inet_ntoa is thread-safe
3685 * updwtmp is moved from libutil to libc
3687 * rewrite of cbrt function
3689 * update of timezone data
3703 * add atoll function
3705 * fix complex problems in Berkeley DB code
3707 * fix math functions
3711 * fixed lots of header problems (especially Linux/GNU specific)
3713 * dynamic loader preserves all registers
3715 * Roland McGrath provided support for handling of auxiliary objects in
3716 the ELF dynamic loader.
3718 * support for parallel builds is improved
3722 * GNU extensions are no longer declared by default. To enable them you
3723 must define the macro `_GNU_SOURCE' in your program or compile with
3726 * The library has changed from using GNU ld symbol aliases to using weak
3727 symbols where available. The ELF object file format supports weak
3728 symbols; GNU ld also supports weak symbols in the a.out format. (There
3729 is also now support for other GNU ld extensions in ELF. Use the
3730 `--with-elf' option to configure to indicate you have ELF, and
3731 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3732 of many files which contained only symbol aliases, reducing the size of
3733 the source and the compiled library; many other files were renamed to
3734 less cryptic names previously occupied by the symbol alias files.
3735 There is a new header file <elf.h> for programs which operate on
3736 files in the ELF format.
3738 * Converted to Autoconf version 2, so `configure' has more options.
3739 Run `configure --help' to see the details.
3741 * The library can now be configured to build profiling, highly-optimized
3742 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3743 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3744 `configure' enable building these extra libraries. The shared library is
3745 built by default when using both ELF and GNU ld. When shared libraries
3746 are enabled, the new library `-ldl' is available for arbitrary run-time
3747 loading of shared objects; its interface is defined in <dlfcn.h>. The
3748 new header file <link.h> gives access to the internals of the run-time
3749 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3750 application of same name on other systems and it provides information
3751 about dynamically linked binaries.
3753 * The C library now provides the run-time support code for profiling
3754 executables compiled with `-pg'. Programs can control the profiling code
3755 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3756 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3757 the support for this file format was contributed by David Mosberger-Tang.
3759 * The math code has been replaced with a math library based on fdlibm from
3760 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3761 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3762 functions. The math functions now reside in a separate library, so
3763 programs using them will need to use `-lm' their linking commands.
3765 * John C. Bowman contributed optimized ix87 assembler inline functions.
3767 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3768 mechanism similar to that found in Solaris 2. This is now used for the
3769 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3770 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3771 file controls what services are used for each individual database. This
3772 works by loading shared libraries with names specified in `nsswitch.conf',
3773 so service modules can be changed or added at any time without even
3774 relinking any program. Currently there are the file, db, and NIS based
3775 NSS services available.
3777 * The new functions `strtoq' and `strtouq' parse integer values from
3778 strings, like `strtol' and `strtoul', but they return `long long int' and
3779 `unsigned long long int' values, respectively (64-bit quantities).
3781 * The new functions `strtof' and `strtold' parse floating-point values from
3782 strings, like `strtod', but they return `float' and `long double' values,
3783 respectively (on some machines `double' and `long double' are the same).
3785 * Ulrich Drepper has contributed new implementations of the floating-point
3786 printing and reading code used in the `printf' family of functions and
3787 `strtod', `strtof', and `strtold'. These new functions are perfectly
3788 accurate, and much faster than the old ones.
3790 * The implementation of the POSIX locale model was completely rewritten by
3791 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3792 to compile the POSIX locale definition.
3794 * The former dummy implementations of the strcoll and strxfrm function are
3795 now replaced by fully functional code contributed by Ulrich Drepper. The
3796 collation information comes from the POSIX locale definitions.
3798 * The new header <langinfo.h> defines an interface for accessing
3799 various locale-dependent data (using the locale chosen with `setlocale').
3801 * Ulrich Drepper has contributed a new suite of functions for operation on
3802 wide-character and multibyte-character strings, in <wchar.h>;
3803 and classification and case conversion of wide characters, in <wctype.h>.
3804 These new functions are conforming to the ISO C, Amendement 1 specification.
3806 * There is now a second implementation of the standard I/O library available.
3807 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3808 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3810 * You can now use positional parameter specifications in format strings
3811 for the `printf' and `scanf' families of functions. For example,
3812 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3813 ``Number 6, Mr Jones''. This is mainly useful when providing different
3814 format strings for different languages, whose grammars may dictate
3815 different orderings of the values being printed. To support this
3816 feature, the interface for `register_printf_handler' has changed; see
3817 the header file <printf.h> for details.
3819 * The `printf' and `scanf' families of functions now understand a new
3820 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3821 to group numbers as indicated by the locale; for `scanf' and friends, this
3822 says to accept as valid only a number with all the proper grouping
3823 separators in the right places. In the default "C" locale, numbers are
3824 not grouped; but locales for specific countries will define the usual
3825 conventions (i.e. separate thousands with `,' in the US locale).
3827 * The pgrp functions have been regularized, slightly incompatibly but much
3828 less confusingly. The core functions are now `getpgid' and `setpgid',
3829 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3830 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3831 provided for compatibility. There is no longer an incompatible `getpgrp'
3832 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3834 * The new header file <fts.h> and suite of functions simplify programs that
3835 operate on directory trees. This code comes from 4.4 BSD.
3837 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3838 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3839 NSS scheme used in glibc.
3841 * There is a new malloc debugging hook `__memalign_hook'.
3843 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3844 `unsigned int' in <sys/types.h>. These are for compatibility only and
3845 their use is discouraged.
3847 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3848 done differently, so that it works even without GNU ld.
3850 * New function `euidaccess' checks allowed access to a file like `access',
3851 but using the effective IDs instead of the real IDs.
3853 * The time zone data files have been updated for the latest and greatest
3854 local time conventions of the countries of the world.
3856 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3859 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3860 for formatting floating-point numbers. They are provided only for
3861 compatibility; new programs should use `sprintf' instead. There are
3862 also equivalent function for the `long double' floating-point type and
3863 all functions also exist in a reentrant form.
3865 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3866 functions for maintaining the login-record files (primarily of use to
3867 system programs such as `login'), and convenient functions for
3868 allocating and initializing a pseudo-terminal (pty) device.
3870 * Ulrich Drepper has contributed new support for System V style
3871 shared memory and IPC on systems that support it.
3873 * Ulrich Drepper has contributed several miscellaneous new functions found
3874 in System V: The `hsearch' family of functions provide an effective
3875 implementation of hash tables; `a64l' and `l64a' provide a very simple
3876 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3879 * Ulrich Drepper has contributed new reentrant counterparts for the
3880 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3882 * Ulrich Drepper has contributed new, highly-optimized versions of several
3883 string functions for the i486/Pentium family of processors.
3885 * Ulrich Drepper has updated the Linux-specific code, based largely
3886 on work done in Hongjiu Lu's version of GNU libc for Linux.
3887 The GNU library now supports Linux versions 2.0.10 and later,
3888 using the ELF object file format (i[3456]86-*-linux).
3890 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3892 * David Mosberger-Tang and Richard Henderson have ported the C library
3893 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3894 dynamic linking support for ELF/Alpha.
3896 * Richard Henderson contributed several Alpha optimized assembler function
3897 for arithmetic and string handling.
3899 * Ulrich Drepper has contributed a new set of message catalog functions to
3900 support multiple languages using the <libintl.h> interface, for use with
3901 his new package GNU gettext. Translation volunteers have contributed
3902 catalogs of the library's messages in Spanish, German, and Korean.
3904 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3905 program and the `catgets' function for reading the catalog files it
3906 creates. (The <libintl.h> interface is preferred; we include the
3907 <nl_types.h> interface using `catgets' only for source compatibility with
3908 programs already written to use it.)
3910 * New header file <values.h> gives SVID-compatible names for <limits.h>
3913 * Various new macros, declarations, and small header files for compatibility
3916 * New function `group_member' is a convenient way to check if a process has
3917 a given effective group ID.
3919 * When using GCC 2.7 and later, the socket functions are now declared in a
3920 special way so that passing an argument of type `struct sockaddr_in *',
3921 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3922 `struct sockaddr *' type, does not generate a type-clash warning.
3924 * New function `error' declared in header file <error.h> is a convenient
3925 function for printing error messages and optionally exiting; this is the
3926 canonical function used in GNU programs. The new functions `err', `warn',
3927 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3928 doing the same thing.
3930 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3931 POSIX.2 `glob' function to do ~ and {...} expansion.
3933 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3934 `clearenv' which is used in POSIX.9 is also available.
3936 * New function `getsid' returns session ID number on systems that support it.
3938 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3939 files <db.h> and <mpool.h> provide a rich set of functions for several
3940 types of simple databases stored in memory and in files, and <ndbm.h> is
3941 an old `ndbm'-compatible interface using the `db' functions. Link with
3942 `-ldb' to get these functions.
3944 * New macro `strdupa' copies a string like `strdup', but uses local stack
3945 space from `alloca' instead of dynamic heap space from `malloc'.
3947 * New function `strnlen' is like `strlen' but searches only a given maximum
3948 number of characters for the null terminator. `stpncpy', `strndup' and
3949 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3952 * New function `statfs' in header <sys/statfs.h>.
3954 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3955 provide convenient functions for operating on blocks of null-terminated
3958 * A new suite of functions in <utmp.h> handle all the details of reading
3959 and writing the utmp file.
3961 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3964 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3965 completely POSIX compliant and also implemented the extended functionality
3966 to handle alternate digit representation and alternate era date formats.
3968 * Ulrich Drepper provided an implementation of the `strptime' function
3969 defined in XPG4.2 which transforms a string into a `struct tm' value.
3971 * Paul Eggert provided the tzselect shell script as part of the timezone
3972 code. The shell script makes it easy to select the correct timezone
3975 * The implementation of the malloc family of functions is completely replaced
3976 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3977 The implementation uses the mmap function (if available) and it is
3978 optimized for the use in multi threaded programs.
3980 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3981 functions. This new functionality is usable by specifying a special
3982 salt string and it is compatible with implementation on *BSD systems.
3984 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3985 `getsubopt' to handle second level command line options, `bsd_signal'
3986 to access BSD style `signal' functionality, the obsolete `regexp' style
3989 * the `lchown' function is available on system which support this
3992 * The implementation of the shadow password handling function was contributed
3995 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3997 * POSIX.1g support was added. The <sys/select.h> header is available,
3998 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3999 implementation of `getaddrinfo'.
4003 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4005 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4006 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4008 * The new function `daemon' from 4.4 BSD is useful for server programs that
4009 want to put themselves in the background.
4011 * Joel Sherrill has contributed support for several standalone boards that
4012 run without an operating system.
4014 * `printf', `scanf' and friends now accept a `q' type modifier for long
4015 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4017 * All of the code taken from BSD (notably most of the math and networking
4018 routines) has been updated from the BSD 4.4-Lite release.
4020 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4022 * The new functions `getdomainname' and `setdomainname' fetch or change the
4023 YP/NIS domain name. These are system calls which exist on systems which
4026 * The time zone data files have been updated for the latest international
4029 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4030 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4034 * The C library now includes support for Sun RPC, from Sun's free
4035 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4036 are included. (There is still no support for YP.)
4038 * Tom Quinn has contributed a port of the C library to SGI machines running
4039 Irix 4 (mips-sgi-irix4).
4041 * The new `lockf' function is a simplified interface to the locking
4042 facilities of `fcntl', included for compatibility.
4044 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4046 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4048 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4051 * The error message printed by `assert' for a failed assertion now includes
4052 the name of the program (if using GNU ld) and the name of the calling
4053 function (with versions of GCC that support this).
4055 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4057 * The library now includes the <sys/mman.h> header file and memory
4058 management functions `mmap', `munmap', `mprotect', `msync', and
4059 `madvise', on systems that support those facilities.
4061 * The interface for `mcheck' has changed slightly: the function called to
4062 abort the program when an allocation inconsistency is detected now takes
4063 an argument that indicates the type of failure. The new function
4064 `mprobe' lets you request a consistency check for a particular block at
4065 any time (checks are normally done only when you call `free' or `realloc'
4068 * It is now possible to easily cross-compile the C library, building on one
4069 system a library to run on another machine and/or operating system. All
4070 you need to do is set the variable `HOST_CC' in `configparms' to the
4071 native compiler for programs to run on the machine you are building on (a
4072 few generator programs are used on Unix systems); set `CC' to the
4075 * The new function `fexecve' (only implemented on the GNU system) executes
4076 a program file given a file descriptor already open on the file.
4080 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4081 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4083 * You can set the variable `libprefix' in `configparms' to specify a prefix
4084 to be prepended to installed library files; this makes it easy to install
4085 the GNU C library to be linked as `-lgnuc' or whatever.
4087 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4088 copies a limited number of characters from a string, and returns the
4089 address of the last character written.
4091 * You no longer need to check for whether the installed `stddef.h' is
4092 compatible with the GNU C library. configure now checks for you.
4094 * You can now define a per-stream `fileno' function to convert the
4095 stream's cookie into an integral file descriptor.
4097 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4098 allocates zero bytes of storage, and returns a unique pointer which
4099 you can pass to `realloc' or `free'. The behavior is undefined if
4100 you dereference this pointer.
4102 * The C library now runs on Sony NEWS m68k machines running either
4103 NewsOS 3 or NewsOS 4.
4105 * The new `syscall' function is a system-dependent primitive function
4106 for invoking system calls. It has the canonical behavior on Unix
4107 systems, including unreliable return values for some calls (such as
4108 `pipe', `fork' and `getppid').
4110 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4111 to `EAGAIN', which is the preferred name. On systems whose kernels
4112 use two distinct codes, the C library now translates EWOULDBLOCK to
4113 EAGAIN in every system call function.
4117 * The GNU C Library Reference Manual is now distributed with the library.
4118 `make dvi' will produce a DVI file of the printed manual.
4119 `make info' will produce Info files that you can read on line using C-h i
4120 in Emacs or the `info' program.
4121 Please send comments on the manual to bug-glibc-manual@gnu.org.
4123 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4125 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4127 * Jason Merrill has contributed a port to the Sequent Symmetry running
4128 Dynix version 3 (i386-sequent-dynix).
4130 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4131 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4133 * New function `memory_warnings' lets you arrange to get warnings when
4134 malloc is running out of memory to allocate, like Emacs gives you.
4136 * The C library now contains the relocating allocator used in Emacs 19 for
4137 its editing buffers. This allocator (ralloc) minimizes allocation
4138 overhead and fragmentation by moving allocated regions around whenever it
4139 needs to. You always refer to a ralloc'd region with a "handle" (a
4140 pointer to a pointer--an object of type `void **').
4142 * There is a new `printf' format: `%m' gives you the string corresponding
4143 to the error code in `errno'.
4145 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4146 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4147 pass, the `a' modifier says to fill in a `char **' you pass with a
4150 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4151 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4152 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4154 * `mkstemp' is a traditional Unix function to atomically create and open a
4155 uniquely-named temporary file.
4159 * The standard location for the file that says what the local timezone is
4160 has changed again. It is now `/usr/local/etc/localtime' (or more
4161 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4163 * The distribution no longer contains any files with names longer than 14
4166 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4167 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4169 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4173 * The configuration process has changed quite a bit. The `configure'
4174 script is now used just like the configuration scripts for other GNU
4175 packages. The `sysdeps' directory hierarchy is much rearranged.
4176 The file `INSTALL' explains the new scheme in detail.
4178 * The header files no longer need to be processed into ANSI C and
4179 traditional C versions. There is just one set of files to install, and
4180 it will work with ANSI or old C compilers (including `gcc -traditional').
4182 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4183 MIPS DECStation running Ultrix 4.
4185 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4186 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4187 made itself into a shared library.
4189 * Yet further improved support for the i386, running 4.3 BSD-like systems
4190 (such as Mach 3 with the Unix single-server), or System V.
4192 * New function `strncasecmp' to do case-insensitive string comparison
4193 with limited length.
4195 * New function `strsep' is a reentrant alternative to `strtok'.
4197 * New functions `scandir' and `alphasort' for searching directories.
4199 * New function `setenv' is a better interface to `putenv'.
4201 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4202 function for traversing a directory tree.
4204 * The GNU obstack package is now also part of the C library.
4205 The new function `open_obstack_stream' creates a stdio stream that
4206 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4207 formatted output directly to an obstack.
4209 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4210 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4212 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4214 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4215 things to your strings.
4217 * There are some new test programs: test-fseek, testmb, and testrand.
4219 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4220 These ports are not finished, but are a good starting place for really
4221 supporting those systems.
4223 * `/etc/localtime' is now the standard location for the file that says what
4224 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4225 This follows the general principle that `/etc' is the place for all local
4226 configuration files.
4228 * The C library header files now use `extern "C"' when used by the C++
4229 compiler, so the C library should now work with C++ code.
4231 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4232 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4235 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4236 function which uses the `merge sort' algorithm, and is said to be
4237 significantly faster than the old GNU `qsort' function. Merge sort is
4238 now the standard `qsort' function. The new algorithm can require a lot
4239 of temporary storage; so, the old sorting function is called when the
4240 required storage is not available.
4242 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4243 provides the Unix `crypt' function, plus some other entry points.
4245 * The code and header files taken from 4.4 BSD have been updated with the
4246 latest files released from Berkeley.
4248 ----------------------------------------------------------------------
4249 Copyright information:
4251 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4253 Permission is granted to anyone to make or distribute verbatim copies
4254 of this document as received, in any medium, provided that the
4255 copyright notice and this permission notice are preserved,
4256 thus giving the recipient permission to redistribute in turn.
4258 Permission is granted to distribute modified versions
4259 of this document, or of portions of it,
4260 under the above conditions, provided also that they
4261 carry prominent notices stating who last changed them.
4264 version-control: never