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, cosf,
13 sinf and tan with FMA, contributed by Arjan van de Ven and H.J. Lu
16 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
18 * Optimized generic expf, exp2f, logf, log2f, powf and sinf.
20 * In order to support faster and safer process termination the malloc API
21 family of functions will no longer print a failure address and stack
22 backtrace after detecting heap corruption. The goal is to minimize the
23 amount of work done after corruption is detected and to avoid potential
24 security issues in continued process execution. Reducing shutdown time
25 leads to lower overall process restart latency, so there is benefit both
26 from a security and performance perspective.
28 * The abort function terminates the process immediately, without flushing
29 stdio streams. Previous glibc versions used to flush streams, resulting
30 in deadlocks and further data corruption. This change also affects
31 process aborts as the result of assertion failures.
33 * On platforms where long double has the IEEE binary128 format (aarch64,
34 alpha, mips64, s390 and sparc), the math library now implements _Float128
35 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015. These
36 are the same interfaces added in version 2.26 for some platforms where
37 this format is supported but is not the format of long double.
39 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
40 mips64, powerpc64le, s390, sparc and x86_64), the math library now
41 implements interfaces for that type, as defined by ISO/IEC TS
42 18661-3:2015. These are corresponding interfaces to those supported for
45 * The math library now implements interfaces for the _Float32, _Float64 and
46 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
47 corresponding interfaces to those supported for _Float128.
49 * glibc now implements the memfd_create and mlock2 functions on Linux.
51 * Support for memory protection keys was added. The <sys/mman.h> header now
52 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
55 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
57 Deprecated and removed features, and other changes affecting compatibility:
59 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
60 defined by <sys/ptrace.h>.
62 * libm no longer supports SVID error handling (calling a user-provided
63 matherr function on error) or the _LIB_VERSION variable to control error
64 handling. (SVID error handling and the _LIB_VERSION variable still work
65 for binaries linked against older versions of the GNU C Library.) The
66 libieee.a library is no longer provided. math.h no longer defines struct
67 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
68 TLOSS, PLOSS and HUGE.
70 * The libm functions pow10, pow10f and pow10l are no longer supported for
71 new programs. Programs should use the standard names exp10, exp10f and
72 exp10l for these functions instead.
74 * The mcontext_t type is no longer the same as struct sigcontext. On
75 platforms where it was previously the same, this changes the C++ name
76 mangling for interfaces involving this type.
78 * The add-ons mechanism for building additional packages at the same time as
79 glibc has been removed. The --enable-add-ons configure option is now
82 * The --without-fp configure option is now ignored. Whether hardware
83 floating-point instructions are used is now configured based on whether
84 the compiler used at configure time (without any options implied by a
85 --with-cpu= configure option) uses such instructions.
87 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
88 the specified string can be parsed as a domain name.
90 * In the malloc_info output, the <heap> element may contain another <aspace>
91 element, "subheaps", which contains the number of sub-heaps.
93 * The libresolv function p_secstodate is no longer supported for new
96 Changes to build and runtime requirements:
98 * bison version 2.7 or later is required to generate code in the 'intl'
101 Security related changes:
103 CVE-2009-5064: The ldd script would sometimes run the program under
104 examination directly, without preventing code execution through the
105 dynamic linker. (The glibc project disputes that this is a security
106 vulnerability; only trusted binaries must be examined using the ldd
109 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
110 suffered from a one-byte overflow during ~ operator processing (either
111 on the stack or the heap, depending on the length of the user name).
112 Reported by Tim Rühsen.
114 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
115 would sometimes fail to free memory allocated during ~ operator
116 processing, leading to a memory leak and, potentially, to a denial
119 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
120 without GLOB_NOESCAPE, could write past the end of a buffer while
121 unescaping user names. Reported by Tim Rühsen.
123 CVE-2017-17426: The malloc function, when called with an object size near
124 the value SIZE_MAX, would return a pointer to a buffer which is too small,
125 instead of NULL. This was a regression introduced with the new malloc
126 thread cache in glibc 2.26. Reported by Iain Buclaw.
128 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
129 to the allocation of too much memory. (This is not a security bug per se,
130 it is mentioned here only because of the CVE assignment.) Reported by
133 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
134 of the number of search path components. (This is not a security
135 vulnerability per se because no trust boundary is crossed if the fix for
136 CVE-2017-1000366 has been applied, but it is mentioned here only because
137 of the CVE assignment.) Reported by Qualys.
139 The following bugs are resolved with this release:
141 [The release manager will add the list generated by
142 scripts/list-fixed-bugs.py just before the release.]
149 * A per-thread cache has been added to malloc. Access to the cache requires
150 no locks and therefore significantly accelerates the fast path to allocate
151 and free small amounts of memory. Refilling an empty cache requires locking
152 the underlying arena. Performance measurements show significant gains in a
153 wide variety of user workloads. Workloads were captured using a special
154 instrumented malloc and analyzed with a malloc simulator. Contributed by
155 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
157 * Unicode 10.0.0 Support: Character encoding, character type info, and
158 transliteration tables are all updated to Unicode 10.0.0, using
159 generator scripts contributed by Mike FABIAN (Red Hat).
160 These updates cause user visible changes, especially the changes in
161 wcwidth for many emoji characters cause problems when emoji sequences
162 are rendered with pango, see for example:
163 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
165 * Collation of Hungarian has been overhauled and is now consistent with "The
166 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
169 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
171 - The GNU C Library will now detect when /etc/resolv.conf has been
172 modified and reload the changed configuration. The new resolver option
173 “no-reload” (RES_NORELOAD) disables this behavior.
175 - The GNU C Library now supports an arbitrary number of search domains
176 (configured using the “search” directive in /etc/resolv.conf);
177 previously, there was a hard limit of six domains. For backward
178 compatibility, applications that directly modify the ‘_res’ global
179 object are still limited to six search domains.
181 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
182 Library will now randomly pick a name server from the configuration as a
183 starting point. (Previously, the second name server was always used.)
185 * The tunables feature is now enabled by default. This allows users to tweak
186 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
188 * New function reallocarray, which resizes an allocated block (like realloc)
189 to the product of two sizes, with a guaranteed clean failure upon integer
190 overflow in the multiplication. Originally from OpenBSD, contributed by
191 Dennis Wölfing and Rüdiger Sonderfeld.
193 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
194 These are extended versions of preadv and pwritev, respectively, taking an
195 additional flags argument. The set of supported flags depends on the
196 running kernel; full support currently requires kernel 4.7 or later.
198 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
199 create a new session ID for the spawned process. This feature is
200 scheduled to be added to the next major revision of POSIX; for the time
201 being, it is available under _GNU_SOURCE.
203 * errno.h is now safe to use from C-preprocessed assembly language on all
204 supported operating systems. In this context, it will only define the
205 Exxxx constants, as preprocessor macros expanding to integer literals.
207 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
208 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
209 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
210 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
212 To compile programs that use this feature, the compiler must support
213 128-bit floating point with the type name _Float128 (as defined by TS
214 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
215 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
216 must be defined to make the new interfaces visible.
218 The new functions and macros correspond to those present for other
219 floating-point types (except for a few obsolescent interfaces not
220 supported for the new type), with F128 or f128 suffixes; for example,
221 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
222 printf or scanf formats for the new type; the strfromf128 and strtof128
223 interfaces should be used instead.
225 Deprecated and removed features, and other changes affecting compatibility:
227 * The synchronization that pthread_spin_unlock performs has been changed to
228 now be equivalent to a C11 atomic store with release memory order to the
229 spin lock's memory location. Previously, several (but not all)
230 architectures used stronger synchronization (e.g., containing what is
231 often called a full barrier). This change can improve performance, but
232 may affect odd fringe uses of spin locks that depend on the previous
233 behavior (e.g., using spin locks as atomic variables to try to implement
234 Dekker's mutual exclusion algorithm).
236 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
239 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
240 will only be built and installed when the GNU C Library is configured with
241 --enable-obsolete-rpc. This allows alternative RPC implementations, such
242 as TIRPC or rpcsvc-proto, to be used.
244 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
245 libnss_compat, are deprecated, and will not be built or installed by
248 The NIS(+) support library, libnsl, is also deprecated. By default, a
249 compatibility shared library will be built and installed, but not headers
250 or development libraries. Only a few NIS-related programs require this
251 library. (In particular, the GNU C Library has never required programs
252 that use 'gethostbyname' to be linked with libnsl.)
254 Replacement implementations based on TIRPC, which additionally support
255 IPv6, are available from <https://github.com/thkukuk/>. The configure
256 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
257 name service modules, to be built and installed.
259 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
260 support is enabled, the configured recursive resolver must support EDNS.
261 (Responding to EDNS-enabled queries with responses which are not
262 EDNS-enabled is fine, but FORMERR responses are not.)
264 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
265 servers have not supported this opcode for a long time.
267 * The _res_opcodes variable has been removed from libresolv. It had been
268 exported by accident.
270 * <string.h> no longer includes inline versions of any string functions,
271 as this kind of optimization is better done by the compiler. The macros
272 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
274 * The nonstandard header <xlocale.h> has been removed. Most programs should
275 use <locale.h> instead. If you have a specific need for the definition of
276 locale_t with no other declarations, please contact
277 libc-alpha@sourceware.org and explain.
279 * The obsolete header <sys/ultrasound.h> has been removed.
281 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
283 * The obsolete function cfree has been removed. Applications should use
286 * The stack_t type no longer has the name struct sigaltstack. This changes
287 the C++ name mangling for interfaces involving this type.
289 * The ucontext_t type no longer has the name struct ucontext. This changes
290 the C++ name mangling for interfaces involving this type.
292 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
293 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
294 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
295 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
296 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
297 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
298 fpregset_t type no longer has the name struct fpu. This changes the C++
299 name mangling for interfaces involving those types.
301 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
302 synced with the kernel:
304 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
305 are not supported on this architecture and have been removed.
307 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
308 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
309 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
311 Programs that assume the GET/SETREGS ptrace requests are universally
312 available will now fail to build, instead of malfunctioning at runtime.
314 Changes to build and runtime requirements:
316 * Linux kernel 3.2 or later is required at runtime, on all architectures
317 supported by that kernel. (This is a change from version 2.25 only for
320 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
322 * On most architectures, GCC 4.9 or later is required to build the GNU C
323 Library. On powerpc64le, GCC 6.2 or later is required.
325 Older GCC versions and non-GNU compilers are still supported when
326 compiling programs that use the GNU C Library. (We do not know exactly
327 how old, and some GNU extensions to C may be _de facto_ required. If you
328 are interested in helping us make this statement less vague, please
329 contact libc-alpha@sourceware.org.)
331 Security related changes:
333 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
334 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
336 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
337 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
339 * Avoid printing a backtrace from the __stack_chk_fail function since it is
340 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
343 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
344 fixed (CVE-2017-12133).
346 The following bugs are resolved with this release:
348 [984] network: Respond to changed resolv.conf in gethostbyname
349 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
350 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
352 [12189] libc: __stack_chk_fail should not attempt a backtrace
354 [14096] time: Race condition on timezone/tst-timezone.out
355 [14172] localedata: az_IR: new locale
356 [14995] build: glibc fails to build if gold is the default linker, even if
358 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
359 [16637] network: inet_pton function is accepting IPv6 with bad format
360 [16640] string: string/strtok.c: undefined behaviour inconsistent between
361 x86 and other generic code
362 [16875] localedata: ko_KR: fix lang_name
363 [17225] localedata: ar_SY: localized month names for May and June are
365 [17297] localedata: da_DK: wrong date_fmt string
366 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
367 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
368 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
369 [19066] localedata: ar_SA abbreviated day and month names are in English
370 [19569] network: resolv: Support an arbitrary number of search domains
371 [19570] network: Implement random DNS server selection in the stub
373 [19838] locale: localedef fails on PA-RISC
374 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
375 order of 0D36 and 0D37
376 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
378 [20098] libc: FAIL: debug/backtrace-tst on hppa
379 [20257] network: sunrpc: clntudp_call does not enforce timeout when
381 [20275] localedata: locale day/abday/mon/abmon should not have trailing
383 [20313] localedata: Update locale data to Unicode 9.0
384 [20424] manual: Document how to provide a malloc replacement
385 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
387 [20686] locale: Add el_GR@euro to SUPPORTED.
388 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
389 failures consistently
390 [21015] dynamic-link: Document and fix --enable-bind-now
391 [21016] nptl: pthread_cond support is broken on hppa
392 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
393 frame-pointer on i386
394 [21049] libc: segfault in longjmp_chk() due to clobbered processor
396 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
397 [21088] libc: Build fails with --enable-static-nss
398 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
399 generic c code is used
400 [21109] libc: Tunables broken on big-endian
401 [21112] math: powf has large ulp errors with base close to 1 and exponent
403 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
405 [21120] malloc: glibc malloc is incompatible with GCC 7
406 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
408 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
409 (-0.0) when linking with -lieee
410 [21171] math: log10, log2 and lgamma return incorrect results
411 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
412 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
413 [21207] localedata: ce_RU: update weekdays from CLDR
414 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
415 [21217] localedata: Update months from CLDR-31
416 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
417 [21243] libc: support_delete_temp_file should issue warning for failed
419 [21244] libc: support resolv_test_start() socket fd close should be
421 [21253] libc: localedef randomly segfaults when using -fstack-check due to
422 new posix_spawn implementation
423 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
424 leads to lower CPU frequency
425 [21259] libc: [alpha] termios.h missing IXANY for POSIX
426 [21261] libc: [sparc64] bits/setjmp.h namespace
427 [21267] network: [mips] bits/socket.h IOC* namespace
428 [21268] libc: [alpha] termios.h NL2, NL3 namespace
429 [21270] libc: mmap64 silently truncates large offset values
430 [21275] libc: posix_spawn always crashes on ia64 now
431 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
432 [21280] math: [powerpc] logbl for POWER7 return incorrect results
433 [21289] libc: Incorrect declaration for 32-bit platforms with
434 _FILE_OFFSET_BITS=64 causes build error
435 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
436 not support gethostbyname4_r
437 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
439 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
441 [21340] libc: Support POSIX_SPAWN_SETSID
442 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
443 [21359] network: ns_name_pack needs additional byte in destination buffer
444 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
445 against fragmentation attacks (CVE-2017-12132)
446 [21369] network: resolv: Remove EDNS fallback
447 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
449 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
450 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
451 [21393] stdio: Missing dup3 error check in freopen, freopen64
452 [21396] libc: Use AVX2 memcpy/memset on Skylake server
453 [21399] localedata: Bad description for U00EC in
454 localedata/charmaps/CP1254
455 [21411] malloc: realloc documentation error
456 [21426] network: sys/socket.h uio.h namespace
457 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
458 [21445] libc: signal.h bsd_signal namespace
459 [21455] network: Network headers stdint.h namespace
460 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
462 [21475] network: resolv: Overlong search path is truncated mid-label
463 [21511] libc: sigstack namespace
464 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
465 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
466 bits/syscall.h: No such file or directory
467 [21517] libc: struct sigaltstack namespace
468 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
469 [21533] localedata: Update locale data to Unicode 10.0
471 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
472 junk at end of line, first unrecognized character is `@'
473 [21538] libc: SIG_HOLD missing for XPG4
474 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
475 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
476 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
477 [21543] libc: sigevent namespace
478 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
479 size for general purpose registers in mcontext_t structure
480 [21550] libc: sigwait namespace
481 [21552] libc: XPG4 bsd_signal namespace
482 [21554] libc: sigpause namespace
483 [21560] libc: sys/wait.h signal.h namespace
484 [21561] libc: waitid namespace
485 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
486 '/usr/include/stdlib.h': Permission denied
487 [21575] libc: sys/wait.h missing struct rusage definition
488 [21584] libc: sigaltstack etc namespace
489 [21597] libc: siginterrupt namespace
490 [21607] math: hppa: FAIL: math/test-tgmath
491 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
493 [21622] libc: [tile] missing SA_* for POSIX.1:2008
494 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
495 stack and heap (CVE-2017-1000366)
496 [21625] libc: wait3 namespace
497 [21654] nss: Incorrect pointer alignment in NSS group merge result
499 [21657] network: Parse interface zone id for node-local multicast
500 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
502 [21666] libc: .symver is used on common symbol
503 [21668] network: resolv: res_init cross-thread broadcast introduces race
505 [21687] math: tgmath.h totalorder, totalordermag return type
506 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
508 [21696] libc: Incorrect assumption of of __cpu_mask in
509 posix/sched_cpucount.c
510 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
511 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
513 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
514 leading to relocation crash
515 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
516 `resolv_conf_matches (resp, conf)' failed.
517 [21710] localedata: Added Samoan language locale for Samoa
518 [21711] localedata: Pashto yesstr/nostr locale are missing
519 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
520 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
521 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
522 [21723] localedata: yesstr/nostr missing for Chinese language locale
523 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
525 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
527 [21728] localedata: New Locale for Tongan language
528 [21729] localedata: incorrect LC_NAME fields for hi_IN
529 [21733] localedata: yesstr and nostr are missing for zh_HK
530 [21734] localedata: Missing yesstr and nostr are for kw_GB
531 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
532 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
533 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
534 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
536 [21744] libc: Tests failing on --enable-tunables --enable-stack-
538 [21749] localedata: Wrong abbreviated day name (“abday”) for
540 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
541 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
542 [21759] localedata: missing yesstr and nostr for Tigrinya
543 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
544 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
545 [21767] localedata: Missing Bislama locales
546 [21768] localedata: Missing yesstr and nostr for aa_ET
547 [21770] localedata: Missing Field in li_NL
548 [21778] nptl: Robust mutex may deadlock
549 [21779] libc: MicroBlaze segfaults when loading libpthread
550 [21783] localedata: Fix int_select international_call_prefixes
551 [21784] localedata: Inconsistency in country_isbn
552 [21788] localedata: Missing Country Postal Abbreviations
553 [21794] localedata: Added-country_isbn-for-Italy
554 [21795] localedata: Add/Fix country_isbn for France
555 [21796] localedata: Added country_isbn for Republic of Korea
556 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
557 [21799] localedata: Added int_select international_call_prefixes
558 [21801] localedata: Added int_select international_call_prefixes
559 [21804] nptl: Double semicolon in thread-shared-types.h
560 [21807] localedata: LC_ADDRESS fix for pap_CW
561 [21808] localedata: Fix LC_ADDRESS for pap_AW
562 [21821] localedata: Added country_name in mai_IN
563 [21822] localedata: Fix LC_TIME for mai_IN
564 [21823] localedata: missing yesstr, nostr for sa_IN
565 [21825] localedata: Fix name_mrs for mag_IN
566 [21828] localedata: 2.26 changelog should mention user visible changes
568 [21835] localedata: Added Maithili language locale for Nepal
569 [21838] localedata: Removed redundant data for the_NP
570 [21839] localedata: Fix LC_MONETARY for ta_LK
571 [21844] localedata: Fix Latin characters and Months Sequence.
572 [21848] localedata: Fix mai_NP Title Name
577 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
578 24731-2:2010, is supported to enable declarations of functions from that
579 TR. Note that not all functions from that TR are supported by the GNU C
582 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
583 18661-1:2014, is supported to enable declarations of functions and macros
584 from that TS. Note that not all features from that TS are supported by
587 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
588 18661-4:2015, is supported to enable declarations of functions and macros
589 from that TS. Note that most features from that TS are not supported by
592 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
593 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
594 Since the GNU C Library defaults to a much newer revision of POSIX, this
595 will only affect programs that specifically request an old conformance
596 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
597 see a change in the visible declarations, but a program compiled with
598 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
601 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
602 defined by all multithreaded code, but glibc has not required this for
605 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
606 means that in a future release, the macros “major”, “minor”, and “makedev”
607 will only be available from <sys/sysmacros.h>.
609 These macros are not part of POSIX nor XSI, and their names frequently
610 collide with user code; see for instance glibc bug 19239 and Red Hat bug
611 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
612 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
615 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
616 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
617 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
619 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
620 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
621 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
622 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
623 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
624 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
625 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
626 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
627 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
628 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
629 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
632 * New <math.h> features are added from TS 18661-1:2014:
634 - Signaling NaN macros: SNANF, SNAN, SNANL.
636 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
637 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
638 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
640 - llogb functions: the llogb, llogbf and llogbl functions, and the
641 FP_LLOGB0 and FP_LLOGBNAN macros.
643 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
646 - Comparison macros: iseqsig.
648 - Classification macros: iscanonical, issubnormal, iszero.
650 - Total order functions: totalorder, totalorderf, totalorderl,
651 totalordermag, totalordermagf, totalordermagl.
653 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
655 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
656 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
658 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
659 are added to libc. They convert a floating-point number into string.
661 * Most of glibc can now be built with the stack smashing protector enabled.
662 It is recommended to build glibc with --enable-stack-protector=strong.
663 Implemented by Nick Alcock (Oracle).
665 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
666 intended to be used instead of memset() to erase sensitive data after use;
667 the compiler will not optimize out calls to explicit_bzero even if they
668 are "unnecessary" (in the sense that no _correct_ program can observe the
669 effects of the memory clear).
671 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
672 to float instead of double. This does not affect the ABI of any libraries
673 that are part of the GNU C Library, but may affect the ABI of other
674 libraries that use this type in their interfaces.
676 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
677 float_t and double_t types are now defined to long double instead of float
678 and double. These options are not the default, and this does not affect
679 the ABI of any libraries that are part of the GNU C Library, but it may
680 affect the ABI of other libraries that use this type in their interfaces,
681 if they are compiled or used with those options.
683 * The getentropy and getrandom functions, and the <sys/random.h> header file
686 * The buffer size for byte-oriented stdio streams is now limited to 8192
687 bytes by default. Previously, on Linux, the default buffer size on most
688 file systems was 4096 bytes (and thus remains unchanged), except on
689 network file systems, where the buffer size was unpredictable and could be
690 as large as several megabytes.
692 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
693 for the Linux quota interface which predates kernel version 2.4.22 has
696 * The malloc_get_state and malloc_set_state functions have been removed.
697 Already-existing binaries that dynamically link to these functions will
698 get a hidden implementation in which malloc_get_state is a stub. As far
699 as we know, these functions are used only by GNU Emacs and this change
700 will not adversely affect already-built Emacs executables. Any undumped
701 Emacs executables, which normally exist only during an Emacs build, should
702 be rebuilt by re-running “./configure; make” in the Emacs build tree.
704 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
705 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
706 “no-ip6-dotint” had already been the default, and support for the
707 “ip6-dotint” option was removed from the Internet in 2006.
709 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
710 flag from <resolv.h> have been removed. The option relied on a
711 backwards-incompatible DNS extension which was never deployed on the
714 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
715 RES_BLAST defined in the <resolv.h> header file have been deprecated.
716 They were already unimplemented.
718 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
719 _res.flags are deprecated. The flag was standardized in RFC 2133, but
720 removed again from the IETF name lookup interface specification in RFC
721 2553. Applications should use getaddrinfo instead.
723 * DNSSEC-related declarations and definitions have been removed from the
724 <arpa/nameser.h> header file, and libresolv will no longer attempt to
725 decode the data part of DNSSEC record types. Previous versions of glibc
726 only implemented minimal support for the previous version of DNSSEC, which
727 is incompatible with the currently deployed version.
729 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
730 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
731 <arpa/nameser.h> header file because the distinction between RR types and
732 meta-RR types is not officially standardized, subject to revision, and
733 thus not suitable for encoding in a macro.
735 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
736 and rhook members of the res_state type in <resolv.h> have been removed.
737 The glibc stub resolver did not support these hooks, but the header file
738 did not reflect that.
740 * For multi-arch support it is recommended to use a GCC which has
741 been built with support for GNU indirect functions. This ensures
742 that correct debugging information is generated for functions
743 selected by IFUNC resolvers. This support can either be enabled by
744 configuring GCC with '--enable-gnu-indirect-function', or by
745 enabling it by default by setting 'default_gnu_indirect_function'
746 variable for a particular architecture in the GCC source file
749 * GDB pretty printers have been added for mutex and condition variable
750 structures in POSIX Threads. When installed and loaded in gdb these pretty
751 printers show various pthread variables in human-readable form when read
752 using the 'print' or 'display' commands in gdb.
754 * Tunables feature added to allow tweaking of the runtime for an application
755 program. This feature can be enabled with the '--enable-tunables' configure
756 flag. The GNU C Library manual has details on usage and README.tunables has
757 instructions on adding new tunables to the library.
759 * A new version of condition variables functions have been implemented in
760 the NPTL implementation of POSIX Threads to provide stronger ordering
763 * A new version of pthread_rwlock functions have been implemented to use a more
764 scalable algorithm primarily through not using a critical section anymore to
767 Security related changes:
769 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
770 have been created with makecontext could fail to terminate due to a
771 missing .cantunwind annotation. This has been observed to lead to a hang
772 (denial of service) in some Go applications compiled with gccgo. Reported
773 by Andreas Schwab. (CVE-2016-6323)
775 * The DNS stub resolver functions would crash due to a NULL pointer
776 dereference when processing a query with a valid DNS question type which
777 was used internally in the implementation. The stub resolver now uses a
778 question type which is outside the range of valid question type values.
781 The following bugs are resolved with this release:
783 [4099] stdio: Overly agressive caching by stream i/o functions.
784 [7065] build: Support building glibc with -fstack-protector or -fstack-
786 [9842] localedata: en_CA: incorrect date format
787 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
788 before it started waiting
789 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
790 [15765] nptl: sem_open is wrongly a cancellation point
791 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
792 [16458] libc: endian.h and netinet/in.h byte order macros return values of
794 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
795 library linked with pthread
796 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
798 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
799 [17252] libc: getrandom and getentropy syscall
800 [17730] malloc: thread-local storage is sometimes improperly free()'d
801 after being __libc_memalign()'d
802 [18241] stdio: failed fseek on memstream does not set errno and can fail
804 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
806 [18463] nptl: pthread_cond_broadcast issue when surrounded by
807 PTHREAD_PRIO_INHERIT mutex on ARM
808 [18784] network: res_query and related function crash for special record
809 type queries (CVE-2015-5180)
810 [19380] math: strtod does not raise "inexact"
811 [19387] string: Integer overflow in memchr
812 [19390] string: Integer overflow in strncat
813 [19398] build: linknamespace tests fail with massively parallel build
814 [19402] nptl: Deadlock with robust shared mutex and asynchronous
816 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
817 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
818 [19514] libc: [PATCH] Fix spelling errors in spelling
819 "implement"/"implementation" in several places
820 [19582] network: Deprecate RES_USE_INET6
821 [19673] manual: clog10 docs appear to be erroneous
822 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
823 causes a segmentation fault
824 [19826] libc: invalid pointer returned from __tls_get_addr with static
826 [20016] network: resolv: Remove hooks support from the API
827 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
829 [20033] math: [x86_64] vectorized math function don't call the __finite
831 [20116] nptl: use after free in pthread_create
832 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
833 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
835 [20311] nptl: please install proc_service.h
836 [20366] libc: Compilation errors in installed headers in strict-compliance
838 [20370] malloc: malloc: Arena free list management is still racy
839 (incorrect fix in bug 19243)
840 [20386] libc: assert (X = 0) does not result in GCC warning
841 [20432] malloc: malloc: Minimize interface required for interposition
842 [20435] libc: Missing unwind info in __startcontext causes infinite loop
843 in _Unwind_Backtrace (CVE-2016-6323)
844 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
845 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
847 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
848 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
849 [20477] network: resolv: incorrect double-checked locking related to
851 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
853 [20495] math: x86_64 performance degradation due to AVX/SSE transition
855 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
856 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
857 cause transition penalty
858 [20517] math: sparcv9 missing fdiml compat symbol
859 [20524] manual: strverscmp is inconsistent
860 [20525] libc: <sys/quota.h> should be based on kernel headers
861 [20539] math: GCC 7 -static -lm fails to link at -O3
862 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
864 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
866 [20591] network: Remove obsolete DNSSEC support
867 [20592] network: DNS resource record type classification macros in
868 <arpa/nameser.h> are incorrect
869 [20593] network: Update DNS RR type definitions
870 [20611] network: getaddrinfo accepts invalid numeric scope IDs
871 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
873 [20629] network: libresolv: Remove support for bitstring labels
875 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
876 assertion: "maxidx >= 2"
877 [20660] math: [arm] Use VSQRT
878 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
879 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
880 [20689] libc: Test for FMA should also check for AVX.
881 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
882 [20715] math: iszero macro breaks existing code
883 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
884 [20728] libc: powerpc: Missing TOC stub in clone
885 [20729] build: glibc-2.24 fails to build for i486 with -Os
886 [20750] build: Build fails with default PIE enabled gcc-6.x
887 [20768] math: [alpha] sqrt fegetenv namespace
888 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
890 [20787] math: float_t is defined as float incorrectly on x86_64 even if
891 __FLT_EVAL_METHOD__ is 2
892 [20790] network: rpcgen buffer overrun in get_prog_declaration
893 [20822] nptl: race condition in __lll_unlock_elision on powerpc
894 [20829] libc: crypt snprintf namespace
895 [20847] libc: tst-vfork3 failure
896 [20855] math: Default bits/mathdef.h has inappropriate float_t
897 [20859] math: [sh4] FP_ILOGB0 invalid
898 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
899 U+20AC), not same as GBK
900 [20915] dynamic-link: global-dynamic TLS broken on aarch64
901 [20916] math: pow handling of sNaN arguments
902 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
903 [20919] math: Bad pow (qNaN, 0) result with -lieee
904 [20940] math: hypot sNaN handling
905 [20947] math: fmax, fmin sNaN handling
906 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
907 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
909 [20971] string: powerpc64/power7 memchr overflows internal pointer check
910 [20973] nptl: robust mutexes: Lost wake-ups
911 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
912 [20978] nis: strlen on null pointer in nss_nisplus
913 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
915 [21014] string: i686 memchr overflows internal pointer check
916 [21019] libc: [mips] n32 lseek incorrect on overflow
917 [21022] libc: [microblaze] __backtrace get_frame_size namespace
918 [21026] libc: [MIPS] readahead syscall is broken on n64
919 [21028] math: Fallback fesetexceptflag should always succeed
920 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
921 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
923 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
924 [21061] librt: [microblaze] librt lost clock_* exports
925 [21073] libc: tunables: insecure environment variables passed to
926 subprocesses with AT_SECURE
927 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
931 * The minimum Linux kernel version that this version of the GNU C Library
932 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
933 kernel version 2.6.32 or later suffices (on architectures that already
934 required kernel versions more recent than 3.2, those requirements remain
935 unchanged). Linux 3.2 or later kernel headers are required on all
938 * The pap_AN locale has been deleted. This has been deprecated for a long
939 time. It has been replaced by pap_AW & pap_CW, both of which have long
940 been included in previous releases.
942 * The readdir_r and readdir64_r functions have been deprecated. It is
943 recommended to use readdir and readdir64 instead.
945 * The type “union wait” has been removed. It was deprecated in the early
946 1990s and never part of POSIX. Application code should use the int type
947 instead of “union wait”.
949 * A new NSS action is added to facilitate large distributed system
950 administration. The action, MERGE, allows remote user stores like LDAP
951 to be merged into local user stores like /etc/groups in order to provide
952 easy to use, updated, and managed sets of merged credentials. The new
953 action can be used by configuring it in /etc/nsswitch.conf:
954 group: files [SUCCESS=merge] nis
955 Implemented by Stephen Gallagher (Red Hat).
957 * The deprecated __malloc_initialize_hook variable has been removed from the
960 * The long unused localedef --old-style option has been removed. It hasn't
961 done anything in over 16 years. Scripts using this option can safely
964 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
965 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
966 return the next representable value in the direction of positive infinity
967 and the nextdown functions return the next representable value in the
968 direction of negative infinity. These are currently enabled as GNU
971 Security related changes:
973 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
974 could result in a stack overflow when getnetbyname was called with an
975 overly long name. (CVE-2016-3075)
977 * Previously, getaddrinfo copied large amounts of address data to the stack,
978 even after the fix for CVE-2013-4458 has been applied, potentially
979 resulting in a stack overflow. getaddrinfo now uses a heap allocation
980 instead. Reported by Michael Petlan. (CVE-2016-3706)
982 * The glob function suffered from a stack-based buffer overflow when it was
983 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
984 Reported by Alexander Cherepanov. (CVE-2016-1234)
986 * The Sun RPC UDP client could exhaust all available stack space when
987 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
988 alloca plugin for GCC. (CVE-2016-4429)
990 * The IPv6 name server management code in libresolv could result in a memory
991 leak for each thread which is created, performs a failing naming lookup,
992 and exits. Over time, this could result in a denial of service due to
993 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
995 The following bugs are resolved with this release:
997 [1170] localedata: ne_NP: update Nepali locale definition file
998 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1000 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1001 [6796] math: fdim() does not set errno on overflow
1002 [10354] libc: posix_spawn should use vfork() in more cases than presently
1003 [11213] localedata: localedata: add copyright disclaimer to locale files
1004 [12143] localedata: chr_US: new Cherokee locale
1005 [12450] localedata: sgs_LT: new locale
1006 [12676] localedata: ln_CD: new locale
1007 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1009 [13304] math: fma, fmaf, fmal produce wrong results
1010 [14259] build: --localedir arg to configure is ignored
1011 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1012 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1013 [14934] localedata: es_CL: wrong first weekday chilean locale
1014 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1016 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1018 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1019 [15368] nptl: raise() is not async-signal-safe
1020 [15479] math: ceil, floor, round and trunc raise inexact exception
1021 [15578] localedata: kk_KZ: various updates
1022 [16003] localedata: pap_AN: punt old locale
1023 [16137] localedata: iw_IL: punt old locale
1024 [16190] localedata: eo: new esperanto locale
1025 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1026 [16742] malloc: race condition: pthread_atfork() called before first
1027 malloc() results in unexpected locking behaviour/deadlocks
1028 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1030 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1031 [17565] localedata: pt_PT: wrong (work-)week start
1032 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1033 [17950] build: Build fails with -msse
1034 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1035 [18433] libc: posix_spawn does not return correctly upon failure to
1037 [18453] localedata: charmaps/IBM875: incorrect codes
1038 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1040 [18896] localedata: he_IL: improvements for currency
1041 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1042 Romanian locale data
1043 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1045 [19056] libc: Deprecate readdir_r
1046 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1048 [19198] localedata: nl_NL: small improvements for Dutch locales
1049 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1050 nameservers (CVE-2016-5417)
1051 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1053 [19400] locale: Language missing in "iso-639.def", trivial fix in
1055 [19431] malloc: Deadlock between fflush, getdelim, and fork
1056 [19505] libc: Incorrect file descriptor validity checks in
1057 posix_spawn_file_actions_add{open,close,dup2}
1058 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1059 when using RTLD_NEXT
1060 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1061 `intl/{gettextP,loadinfo}.h'
1062 [19534] libc: execle, execlp may use malloc
1063 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1064 [19573] network: res_nclose and __res_maybe_init disagree about name
1065 server initialization, breaking Hesiod
1066 [19575] localedata: Status of GB18030 tables
1067 [19581] localedata: sr_* date_fmt string contains additional newline
1068 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1070 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1071 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1072 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1073 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1075 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1077 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1079 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1080 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1081 [19633] locale: strfmon_l applies global locale to number formatting
1082 [19642] network: Memory leak in getnameinfo
1083 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1084 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1086 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1087 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1089 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1090 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1092 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1094 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1096 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1097 pointers and lengths in error-case.
1098 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1099 report errors on UTF-16 surrogates values.
1100 [19755] nscd: nscd assertion failure in gc
1101 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1102 [19759] libc: mempcpy shouldn't be inlined
1103 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1104 [19765] libc: s390 needs an optimized mempcpy
1105 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1106 NAME_MAX limit assumption (CVE-2016-1234)
1107 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1108 [19787] network: Missing and incorrect truncation checks in getnameinfo
1109 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1111 [19791] network: Assertion failure in res_query.c with un-connectable name
1113 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1114 [19822] math: libm.so install clobbers old version
1115 [19825] network: resolv: send_vc can return uninitialized data in second
1116 response to getaddrinfo
1117 [19830] network: nss_dns: should check RDATA length against buffer length
1118 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1119 confronted with A/AAAA records of invalid size
1120 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1121 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1122 [19853] stdio: Printing IBM long double in decimal with high precision is
1124 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1125 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1126 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1128 [19865] network: Assertion failure or memory leak in
1129 _nss_dns_getcanonname_r
1130 [19868] network: nss_dns: netent code does not skip over non-PTR records
1131 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1133 [19881] string: Improve x86-64 memset
1134 [19907] string: Incorrect memcpy tests
1135 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1137 [19925] libc: termios.h XCASE namespace
1138 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1139 [19929] libc: limits.h NL_NMAX namespace
1140 [19931] stdio: Memory leak in vfprintf
1141 [19957] libc: clone(CLONE_VM) access invalid parent memory
1142 [19963] localedata: en_IL: New locale
1143 [19989] stdio: stdio.h cuserid namespace
1144 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1146 [19996] locale: langinfo.h nl_langinfo_l namespace
1147 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1148 [20010] network: getaddrinfo: Stack overflow in hostent translation
1150 [20012] stdio: libio: fmemopen append mode failure
1151 [20014] stdio: stdio.h namespace for pre-threads POSIX
1152 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1153 [20023] libc: fcntl.h timespec namespace
1154 [20024] math: [x86_64] vectorized sincos trashes the stack
1155 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1156 [20041] time: sys/time.h timespec namespace
1157 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1158 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1159 [20051] libc: ttyslot in wrong header under wrong conditions
1160 [20054] libc: gethostname not declared for XPG4
1161 [20055] libc: termios.h missing tcgetsid for XPG4
1162 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1164 [20073] libc: sys/stat.h fchmod namespace
1165 [20074] libc: stdlib.h rand_r namespace
1166 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1167 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1169 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1170 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1172 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1173 [20119] libc: Wrong mask for processors level type from CPUID
1174 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1175 AS not supporting AVX512
1176 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1178 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1179 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1181 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1182 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1183 [20178] libc: posix_spawn{p} should not call exit
1184 [20191] stdio: libio: vtables hardening
1185 [20195] string: FMA4 detection requires CPUID execution with register
1187 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1188 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1190 [20212] math: acos (sNaN) returns sNaN
1191 [20213] math: asin (sNaN) returns sNaN
1192 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1193 [20218] math: [i386] asinhl (sNaN) returns sNaN
1194 [20219] math: [i386] atanhl (sNaN) returns sNaN
1195 [20222] stdio: fopencookie: Mangle function pointers
1196 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1197 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1198 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1200 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1201 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1202 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1203 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1204 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1205 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1206 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1207 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1208 [20240] math: modf (sNaN) returns sNaN
1209 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1210 [20250] math: frexp (sNaN) returns sNaN
1211 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1212 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1214 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1215 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1216 subscript is below array bounds [-Werror=array-bounds]
1217 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1218 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1220 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1221 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1223 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1224 "invalid" exceptions
1225 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1226 [20316] localedata: id_ID: Februari instead of Pebruari
1227 [20327] string: POWER8 strcasecmp returns incorrect result
1228 [20347] math: Failure: Test: j0_downward (0xap+0)
1229 [20348] libc: FAIL: misc/tst-preadvwritev64
1230 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1231 [20350] libc: There is no test for p{read,write}64
1232 [20357] math: Incorrect cos result for 1.5174239687223976
1233 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1237 * Unicode 8.0.0 Support: Character encoding, character type info, and
1238 transliteration tables are all updated to Unicode 8.0.0, using new
1239 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1240 These updates cause user visible changes, such as the fixes for bugs
1241 89, 16061, and 18568.
1243 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1244 kernel-internal CPU set size. This means that requests that change the
1245 CPU affinity which failed before (for example, an all-ones CPU mask) will
1246 now succeed. Applications that need to determine the effective CPU
1247 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1248 setting it because the kernel can adjust it (and the previous size check
1249 would not detect this in the majority of cases).
1251 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1252 the following new symbols are used: fts64_children, fts64_close,
1253 fts64_open, fts64_read and fts64_set.
1255 * getaddrinfo now detects certain invalid responses on an internal netlink
1256 socket. If such responses are received, an affected process will
1257 terminate with an error message of "Unexpected error <number> on netlink
1258 descriptor <number>" or "Unexpected netlink response of size <number> on
1259 descriptor <number>". The most likely cause for these errors is a
1260 multi-threaded application which erroneously closes and reuses the netlink
1261 file descriptor while it is used by getaddrinfo.
1263 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1264 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1265 unnecessary serialization of memory allocation requests across threads.
1266 The defect is now corrected. Users should see a substantial increase in
1267 the concurent throughput of allocation requests for applications which
1268 trigger this bug. Affected applications typically create and
1269 destroy threads frequently. (Bug 19048 was reported and analyzed by
1272 * There is now a --disable-timezone-tools configure option for disabling the
1273 building and installing of the timezone related utilities (zic, zdump, and
1274 tzselect). This is useful for people who build the timezone data and code
1275 independent of the GNU C Library.
1277 * The obsolete header <regexp.h> has been removed. Programs that require
1278 this header must be updated to use <regex.h> instead.
1280 * The obsolete functions bdflush, create_module, get_kernel_syms,
1281 query_module and uselib are no longer available to newly linked binaries;
1282 the header <sys/kdaemon.h> has been removed. These functions and header
1283 were specific to systems using the Linux kernel and could not usefully be
1284 used with the GNU C Library on systems with version 2.6 or later of the
1287 * Optimized string, wcsmbs and memory functions for IBM z13.
1288 Implemented by Stefan Liebler.
1290 * Newly linked programs that define a variable called signgam will no longer
1291 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1292 require signgam to be set by those functions must ensure that they use the
1293 variable provided by the GNU C Library and declared in <math.h>, without
1294 defining their own copy.
1296 * The minimum GCC version that can be used to build this version of the GNU
1297 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1298 still be used to compile programs using the GNU C Library.
1300 Security related changes:
1302 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1303 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1305 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1306 the pointer guard feature. It is always enabled. Previously,
1307 LD_POINTER_GUARD could be used to disable security hardening in binaries
1308 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1311 * An integer overflow in hcreate and hcreate_r could lead to an
1312 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1314 * The catopen function no longer has unbounded stack usage. Reported by
1315 Max. (CVE-2015-8779)
1317 * The nan, nanf and nanl functions no longer have unbounded stack usage
1318 depending on the length of the string passed as an argument to the
1319 functions. Reported by Joseph Myers. (CVE-2014-9761)
1321 * A stack-based buffer overflow was found in libresolv when invoked from
1322 libnss_dns, allowing specially crafted DNS responses to seize control
1323 of execution flow in the DNS client. The buffer overflow occurs in
1324 the functions send_dg (send datagram) and send_vc (send TCP) for the
1325 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1326 family. The use of AF_UNSPEC triggers the low-level resolver code to
1327 send out two parallel queries for A and AAAA. A mismanagement of the
1328 buffers used for those queries could result in the response of a query
1329 writing beyond the alloca allocated buffer created by
1330 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1331 the overflow. Thanks to the Google Security Team and Red Hat for
1332 reporting the security impact of this issue, and Robert Holiday of
1333 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1335 The following bugs are resolved with this release:
1337 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1338 [887] math: Math library function "logb" and "nextafter" inconsistent
1339 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1340 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1341 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1342 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1344 [4404] localedata: German translation of "Alarm clock" is misleading
1345 [6799] math: nextafter() and nexttoward() doen't set errno on
1346 overflow/underflow errors
1347 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1349 [10432] nis: _nss_nis_setnetgrent assertion failure
1350 [11460] libc: fts has no LFS support
1351 [12926] network: getaddrinfo()/make_request() may spin forever
1352 [13065] nptl: Race condition in pthread barriers
1353 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1354 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1356 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1357 [14912] libc: Rename non-installed bits/*.h headers
1358 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1359 [15367] math: Let gcc use __builtin_isinf
1360 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1361 [15421] math: lgamma wrongly sets signgam for ISO C
1362 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1363 FE_INVALID with argument out of range
1364 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1366 [15786] dynamic-link: ifunc resolver functions can smash function
1368 [15918] math: Unnecessary check for equality in hypotf()
1369 [16061] localedata: Review / update transliteration data
1370 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1372 [16141] time: strptime %z offset restriction
1373 [16171] math: drem should be alias of remainder
1374 [16296] math: fegetround is pure?
1375 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1376 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1378 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1379 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1381 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1382 "Exception "Inexact" set" on ppc32
1383 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1384 [16517] math: Missing underflow exception from tanf/tan/tanl
1385 [16519] math: Missing underflow exception from sinhf
1386 [16520] math: Missing underflow exception from tanhf
1387 [16521] math: Missing underflow exception from exp2
1388 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1390 [16734] stdio: fopen calls mmap to allocate its buffer
1391 [16961] math: nan function incorrect handling of bad sequences
1392 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1393 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1394 [16985] locale: localedef: confusing error message when opening output
1396 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1397 [17197] locale: Redundant shift character in iconv conversion output at
1399 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1400 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1401 [17250] dynamic-link: static linking breaks nss loading
1402 (getaddrinfo/getpwnam/etc...)
1403 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1405 [17441] math: isnan() should use __builtin_isnan() in GCC
1406 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1407 (related to lock elision)
1408 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1409 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1410 [17887] time: strptime should be able to parse "+01:00" style timezones
1411 [17905] libc: catopen() Multiple unbounded stack allocations
1413 [18084] libc: backtrace (..., 0) dumps core on x86
1414 [18086] libc: nice() sets errno to 0 on success
1415 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1416 is too large (CVE-2015-8778)
1417 [18251] dynamic-link: SONAME missing when audit modules provides path
1418 [18265] libc: add attributes for wchar string and memory functions
1419 [18370] math: csqrt missing underflows
1420 [18421] libc: [hppa] read-only segment has dynamic relocations
1421 [18472] libc: Obsolete syscall wrappers should be compat symbols
1422 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1423 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1424 [18525] localedata: Remove locale timezone information
1425 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1426 [18568] localedata: Update locale data to Unicode 8.0
1427 [18589] locale: sort-test.sh fails at random
1428 [18595] math: ctan, ctanh missing underflows
1429 [18604] libc: assert macro-expands its argument
1430 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1431 contains a vector instruction exception.
1432 [18611] math: j1, jn missing errno setting on underflow
1433 [18618] localedata: sync Chechen locale definitions with other *_RU
1435 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1436 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1437 [18665] network: In send_dg, the recvfrom function is NOT always using the
1438 buffer size of a newly created buffer (CVE-2015-7547)
1439 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1441 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1443 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1444 [18699] math: tilegx cproj() for various complex infinities does not yield
1446 [18724] libc: Harden put*ent functions against data injection
1447 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1448 [18755] build: build errors with -DNDEBUG
1449 [18757] stdio: fmemopen fails to set errno on failure
1450 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1452 [18781] libc: openat64 lacks O_LARGEFILE
1453 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1454 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1455 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1456 [18790] math: [ldbl-128ibm] tanhl inaccurate
1457 [18795] libc: stpncpy fortification misses buffer lengths that are
1458 statically too large
1459 [18796] build: build fails for --disable-mathvec
1460 [18803] math: hypot missing underflows
1461 [18820] stdio: fmemopen may leak memory on failure
1462 [18823] math: csqrt spurious underflows
1463 [18824] math: fma spurious underflows
1464 [18825] math: pow missing underflows
1465 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1466 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1467 [18870] build: sem_open.c fails to compile with missing symbol
1469 [18872] stdio: Fix memory leak in printf_positional
1470 [18873] libc: posix_fallocate overflow check ineffective
1471 [18875] math: Excess precision leads incorrect libm
1472 [18877] libc: arm: mmap offset regression
1473 [18887] libc: memory corruption when using getmntent on blank lines
1474 [18918] localedata: hu_HU: change time to HH:MM:SS format
1475 [18921] libc: Regression: extraneous stat() and fstat() performed by
1477 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1478 binaries (CVE-2015-8777)
1479 [18951] math: tgamma missing underflows
1480 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1482 [18953] localedata: lt_LT: change currency symbol to the euro
1483 [18956] math: powf inaccuracy
1484 [18961] math: [i386] exp missing underflows
1485 [18966] math: [i386] exp10 missing underflows
1486 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1487 [18969] build: multiple string test failures due to missing locale
1489 [18970] libc: Reference of pthread_setcancelstate in libc.a
1490 [18977] math: float / long double Bessel functions not in XSI POSIX
1491 [18980] math: i386 libm functions return with excess range and precision
1492 [18981] math: i386 scalb*, ldexp return with excess range and precision
1493 [18982] stdio: va_list and vprintf
1494 [18985] time: Passing out of range data to strftime() causes a segfault
1496 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1497 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1498 [19012] locale: iconv_open leaks memory on error path
1499 [19016] math: clog, clog10 inaccuracy
1500 [19018] nptl: Mangle function pointers in tls_dtor_list
1501 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1502 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1503 [19048] malloc: malloc: arena free list can become cyclic, increasing
1505 [19049] math: [powerpc] erfc incorrect zero sign
1506 [19050] math: [powerpc] log* incorrect zero sign
1507 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1508 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1509 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1510 [19074] network: Data race in _res_hconf_reorder_addrs
1511 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1512 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1513 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1515 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1517 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1518 [19086] manual: posix_fallocate64 documented argument order is wrong.
1519 [19088] math: lround, llround missing exceptions close to overflow
1521 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1522 [19095] math: dbl-64 lrint incorrect for 64-bit long
1523 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1524 [19124] dynamic-link: ld.so failed to build with older assmebler
1525 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1526 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1527 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1528 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1529 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1530 pthread_setaffinity_np
1531 [19156] math: [ldbl-128] j0l spurious underflows
1532 [19164] nptl: tst-getcpu fails with many possible CPUs
1533 [19168] math: math/test-ildoubl and math/test-ldouble failure
1534 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1535 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1537 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1538 (FE_NOMASK_ENV) do not clear SSE exceptions
1539 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1540 _int_new_arena/reused_arena
1541 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1542 [19201] math: dbl-64 remainder incorrect sign of zero result
1543 [19205] math: bits/math-finite.h conditions do not match math.h and
1545 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1546 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1547 for C99-based standards
1548 [19212] libc: features.h not -Wundef clean
1549 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1551 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1552 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1553 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1555 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1556 "inexact" exceptions
1557 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1559 [19242] libc: strtol incorrect in Turkish locales
1560 [19243] malloc: reused_arena can pick an arena on the free list, leading
1561 to an assertion failure and reference count corruption
1562 [19253] time: tzset() ineffective when temporary TZ did not include DST
1564 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1565 [19270] math: [hppa] Shared libm missing __isnanl
1566 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1567 MAP_HUGETLB and MAP_STACK defines
1568 [19313] nptl: Wrong __cpu_mask for x32
1569 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1570 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1571 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1572 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1573 [19363] time: x32: times() return value wrongly truncates/sign extends
1575 [19367] dynamic-link: Improve branch prediction on Silvermont
1576 [19369] network: Default domain name not reset by res_ninit when "search"
1577 / "domain" entry is removed from resolv.conf
1578 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1580 [19385] localedata: bg_BG: time separator should be colon, not comma
1581 [19408] libc: linux personality syscall wrapper may erroneously return an
1582 error on 32-bit architectures
1583 [19415] libc: dladdr returns wrong names on hppa
1584 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1585 IBM905, IBM907, and IBM909
1586 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1587 [19443] build: build failures with -DDEBUG
1588 [19451] build: Make check fails on test-double-vlen2
1589 [19462] libc: Glibc failed to build with -Os
1590 [19465] math: Wrong code with -Os
1591 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1593 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1595 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1596 [19486] math: S390: Math tests fail with "Exception Inexact set".
1597 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1598 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1600 [19590] math: Fail to build shared objects that use libmvec.so functions.
1604 * The following bugs are resolved with this release:
1606 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1607 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1608 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1609 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1610 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1611 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1612 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1613 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1614 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1615 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1616 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1617 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1618 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1619 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1620 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1621 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1622 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1623 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1624 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1625 18657, 18676, 18694, 18696, 18887.
1627 * Cache information can be queried via sysconf() function on s390 e.g. with
1628 _SC_LEVEL1_ICACHE_SIZE as argument.
1630 * A buffer overflow in gethostbyname_r and related functions performing DNS
1631 requests has been fixed. If the NSS functions were called with a
1632 misaligned buffer, the buffer length change due to pointer alignment was
1633 not taken into account. This could result in application crashes or,
1634 potentially arbitrary code execution, using crafted, but syntactically
1635 valid DNS responses. (CVE-2015-1781)
1637 * The time zone file parser has been made more robust against crafted time
1638 zone files, avoiding heap buffer overflows related to the processing of
1639 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1640 large time zone data files. Overly long time zone specifiers in the TZ
1641 variable no longer result in stack overflows and crashes.
1643 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1644 for LD and GD on x86 and x86-64, has been implemented. You will need
1645 binutils-2.24 or later to enable this optimization.
1647 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1648 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1649 Hat). These updates cause user visible changes, such as the fix for bug
1652 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1653 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1654 condition in some applications.
1656 * Added vector math library named libmvec with the following vectorized x86_64
1657 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1659 The library can be disabled with --disable-mathvec. Use of the functions is
1660 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1661 Shared library libmvec.so is linked in as needed when using -lm (no need to
1662 specify -lmvec explicitly for not static builds).
1663 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1665 * A new fmemopen implementation has been added with the goal of POSIX
1666 compliance. The new implementation fixes the following long-standing
1667 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1668 old implementation is still present for use be by existing binaries.
1670 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1671 releases. It has been fixed to match 2.19 and older, but binaries built
1672 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1674 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1675 Contributed by Roland McGrath (Google).
1677 * The header <regexp.h> is deprecated, and will be removed in a future
1678 release. Use of this header will trigger a deprecation warning.
1679 Application developers should update their code to use <regex.h> instead.
1681 This header was formerly part of SUS, but was deprecated in 1994 and
1682 removed from the standard in 2001. Also, the glibc implementation
1683 leaks memory. See BZ#18681 for more details.
1687 * The following bugs are resolved with this release:
1689 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1690 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1691 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1692 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1693 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1694 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1695 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1696 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1697 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1698 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1701 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1702 too little memory for the to-be-scanned arguments and overflow the
1703 allocated buffer. The implementation now correctly computes the required
1704 buffer size when using malloc, and switches to malloc from alloca as
1707 * A new semaphore algorithm has been implemented in generic C code for all
1708 machines. Previous custom assembly implementations of semaphore were
1709 difficult to reason about or ensure that they were safe. The new version
1710 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1711 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1712 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1714 * Port to Altera Nios II has been contributed by Mentor Graphics.
1716 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1717 implementations for powerpc64/powerpc64le.
1718 Implemented by Adhemerval Zanella (IBM).
1720 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1721 and powerpc64le. This may improve lock scaling of existing programs on
1722 HTM capable systems. The lock elision code is only enabled with
1723 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1724 powerpc will issue a transaction abort on every syscall to avoid side
1725 effects being visible outside transactions.
1727 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1728 AArch64. Contributed by ARM Ltd.
1730 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1732 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1733 under certain input conditions resulting in the execution of a shell for
1734 command substitution when the application did not request it. The
1735 implementation now checks WRDE_NOCMD immediately before executing the
1736 shell and returns the error WRDE_CMDSUB as expected.
1738 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1739 processing format strings with a large number of format specifiers.
1741 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1742 infinite loop if the DNS response contained a PTR record of an unexpected
1745 * The minimum GCC version that can be used to build this version of the GNU
1746 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1747 still be used to compile programs using the GNU C Library.
1749 * The GNU C Library is now built with -Werror by default. This can be
1750 disabled by configuring with --disable-werror.
1752 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1754 * The obsolete sigvec function has been removed. This was the original
1755 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1756 programs have been using instead for about 25 years. Of course, ABI
1757 compatibility for old binaries using sigvec remains intact.
1759 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1760 with newer versions of bison.
1762 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1763 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1764 registers overlay two consecutive single-precision registers. MIPS32R2
1765 introduced a new FPU mode (FR=1) where double-precision registers extend the
1766 corresponding single-precision registers which is incompatible with the
1767 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1768 require the use of FR=1 making a transition necessary. New o32 ABI
1769 extensions enable users to migrate over time from the original o32 ABI
1770 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1771 tracks the ABI of any loaded object and verifies that new objects are
1772 compatible. Mode transitions will also be requested as required and
1773 unsupportable objects will be rejected. The ABI checks include both soft and
1774 hard float ABIs for o32, n32 and n64.
1776 GCC 5 with GNU binutils 2.25 onwards:
1777 It is strongly recommended that all o32 system libraries are built using the
1778 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1779 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1780 compiler using --with-fp-32=xx to set this by default.
1784 * The following bugs are resolved with this release:
1786 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1787 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1788 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1789 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1790 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1791 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1792 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1793 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1794 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1795 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1796 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1797 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1798 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1799 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1800 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1802 * Reverted change of ABI data structures for s390 and s390x:
1803 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1804 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1805 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1806 are preserved pointing straight to the same implementation as the old ones.
1807 Given that, new callers will simply provide a too-big buffer to these
1808 functions. Any applications/libraries out there that embed jmp_buf or
1809 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1810 against 2.19 headers will have to rebuilt again. This is necessary in any
1811 case to revert the breakage in their ABI caused by the glibc change.
1813 * Support for file description locks is added to systems running the
1814 Linux kernel. The standard file locking interfaces are extended to
1815 operate on file descriptions, not file descriptors, via the use of
1816 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1817 are associated with an open file instead of a process.
1819 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1821 * The minimum Linux kernel version that this version of the GNU C Library
1822 can be used with is 2.6.32.
1824 * Running the testsuite no longer terminates as soon as a test fails.
1825 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1826 with PASS or FAIL lines for individual tests. A summary of the results is
1827 printed, including a list of failing lists, and "make check" exits with
1828 error status if there were any unexpected failures. "make check
1829 stop-on-test-failure=y" may be used to keep the old behavior.
1831 * The am33 port, which had not worked for several years, has been removed
1834 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1835 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1836 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1837 conflicted with POSIX (support for which was removed in 2.19), the
1838 interfaces those macros enabled remain available when compiling with
1839 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1840 test macros defined.
1842 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1844 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1845 This may improve lock scaling of existing programs on TX capable systems.
1846 The lock elision code is only built with --enable-lock-elision=yes and
1847 then requires a GCC version supporting the TX builtins. With lock elision
1848 default mutexes are elided via __builtin_tbegin, if the cpu supports
1849 transactions. By default lock elision is not enabled and the elision code
1852 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1853 copy the path argument. This allowed programs to cause posix_spawn to
1854 deference a dangling pointer, or use an unexpected pathname argument if
1855 the string was modified after the posix_spawn_file_actions_addopen
1858 * All supported architectures now use the main glibc sysdeps directory
1859 instead of some being in a separate "ports" directory (which was
1860 distributed separately before glibc 2.17).
1862 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1863 On configurations that support it (all Linux configurations), it's now
1864 used regardless of the --enable-add-ons switch to configure. It is no
1865 longer possible to build such configurations without pthreads support.
1867 * Locale names, including those obtained from environment variables (LANG
1868 and the LC_* variables), are more tightly checked for proper syntax.
1869 setlocale will now fail (with EINVAL) for locale names that are overly
1870 long, contain slashes without starting with a slash, or contain ".." path
1871 components. (CVE-2014-0475) Previously, some valid locale names were
1872 silently replaced with the "C" locale when running in AT_SECURE mode
1873 (e.g., in a SUID program). This is no longer necessary because of the
1876 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1877 with application code using Intel MPX instructions. (With all previous
1878 versions, the MPX register state could be clobbered when making calls
1879 into or out of a shared library.) Note that while the new dynamic
1880 linker is compatible with all known x86 hardware whether or not it
1881 supports Intel MPX, some x86 instruction-set emulators might fail to
1882 handle the new instruction encodings. This is known to affect Valgrind
1883 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1884 release), and might affect other tools that do instruction emulation.
1886 * Support for loadable gconv transliteration modules has been removed.
1887 The support for transliteration modules has been non-functional for
1888 over a decade, and the removal is prompted by security defects. The
1889 normal gconv conversion modules are still supported. Transliteration
1890 with //TRANSLIT is still possible, and the //IGNORE specifier
1891 continues to be supported. (CVE-2014-5119)
1893 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1894 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1895 resulting a denial-of-service security vulnerability in applications which
1896 use functions related to iconv. (CVE-2014-6040)
1900 * The following bugs are resolved with this release:
1902 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1903 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1904 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1905 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1906 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1907 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1908 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1909 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1910 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1911 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1912 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1913 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1914 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1915 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1916 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1917 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1918 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1919 16431, 16453, 16474, 16506, 16510, 16529
1921 * Slovenian translations for glibc messages have been contributed by the
1922 Translation Project's Slovenian team of translators.
1924 * The public headers no longer use __unused nor __block. This change is to
1925 support compiling programs that are derived from BSD sources and use
1926 __unused internally, and to support compiling with Clang's -fblock
1927 extension which uses __block.
1929 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1930 large collation sequences to optimize multiple passes. This cache
1931 computation may overflow for large collation sequences and may cause a
1932 stack or buffer overflow. This is now fixed to use a slower algorithm
1933 which does not use a cache if there is an integer overflow.
1935 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1936 rules for large collation sequences to optimize multiple passes and falls
1937 back to alloca if malloc fails, resulting in a possible stack overflow.
1938 The implementation now falls back to an uncached collation sequence lookup
1941 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1942 initialized for static applications resulting in the security feature
1943 being disabled. The pointer guard is now correctly initialized to a
1944 random value for static applications. Existing static applications need
1945 to be recompiled to take advantage of the fix (bug 15754).
1947 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1948 to the d_name member of struct dirent, or omit the terminating NUL
1949 character. (Bugzilla #14699).
1951 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1952 aligned_alloc functions could allocate too few bytes or corrupt the
1953 heap when passed very large allocation size values (Bugzilla #15855,
1956 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1957 for AF_INET6 has been fixed (Bugzilla #16072).
1959 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1960 pap_CW, quz_PE, the_NP.
1962 * Substantially revised locales: gd_GB, ht_HT
1964 * The LC_ADDRESS field was updated to support country_car for almost all
1967 * ISO 1427 definitions were updated.
1969 * ISO 3166 definitions were updated.
1971 * The localedef utility now supports --big-endian and --little-endian
1972 command-line options to generate locales for a different system from that
1973 for which the C library was built.
1975 * Binary locale files now only depend on the endianness of the system for
1976 which they are generated and not on other properties of that system. As a
1977 consequence, binary files generated with new localedef may be incompatible
1978 with old versions of the GNU C Library, and binary files generated with
1979 old localedef may be incompatible with this version of the GNU C Library,
1980 in the following circumstances:
1982 + Locale files may be incompatible on m68k systems.
1984 + Locale archive files (but not separate files for individual locales) may
1985 be incompatible on systems where plain "char" is signed.
1987 * The configure option --disable-versioning has been removed. Builds with
1988 --disable-versioning had not worked for several years.
1990 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1991 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1993 * SystemTap probes for malloc have been introduced.
1995 * SystemTap probes for slow multiple precision fallback paths of
1996 transcendental functions have been introduced.
1998 * Support for powerpc64le has been added.
2000 * The soft-float powerpc port now supports e500 processors.
2002 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2004 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2005 set of header declarations that are enabled by default, even when other
2006 feature test macros or compiler options such as -std=c99 would otherwise
2007 disable some of those declarations.
2009 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2010 conflict with POSIX. The libbsd-compat library (which was a dummy library
2011 that did nothing) has also been removed.
2013 * Preliminary documentation about Multi-Thread, Async-Signal and
2014 Async-Cancel Safety has been added.
2016 * Change of ABI data structures for s390 and s390x:
2017 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2018 allow for future hardware extensions. All library functions that accept or
2019 return these structures were versioned in order to provide backwards
2020 compatibility with existing code. However, not all cases can be handled
2021 with interface versioning. If an application embeds either structure into
2022 another structure and that structure is passed to another compilation unit
2023 which expects the newer larger-sized object then any access to the new
2024 fields will result in undefined behaviour. Similarly any access to fields
2025 that were moved by the enlarging of the structures will cause undefined
2026 behaviour. To fix the undefined behaviour all objects that are part of the
2027 final in-memory image must agree on the size of structures, and this may
2028 require recompilation.
2032 * The following bugs are resolved with this release:
2034 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2035 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2036 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2037 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2038 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2039 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2040 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2041 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2042 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2043 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2044 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2045 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2046 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2049 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2050 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2051 Distributions can re-enable building and using pt_chown via the new configure
2052 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2053 considerable security risks and should only be used if the distribution
2054 understands and accepts the risks.
2056 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2059 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2060 fixed (Bugzilla #15330).
2062 * Add support for calling C++11 thread_local object destructors on thread
2063 and program exit. This needs compiler support for offloading C++11
2064 destructor calls to glibc.
2066 * Improved worst case performance of libm functions with double inputs and
2069 * Support for priority inherited mutexes in pthread condition variables on
2070 non-x86 architectures.
2072 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2074 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2076 * Optimized string functions for ARM. Implemented by Will Newton and
2079 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2081 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2084 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2085 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2087 * Added a benchmark framework to track performance of functions in glibc.
2089 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2090 It is based on draft TS 18661 and currently enabled as a GNU extension.
2092 * On Linux, the clock function now uses the clock_gettime system call
2093 for improved precision, rather than old times system call.
2095 * Added support for version-3 tz format files. This is needed when using
2096 the tz database release 2013e or later, and affects a few unusual cases --
2097 currently only TZ='America/Godthab' for time stamps after 2037.
2099 * Added new API functions pthread_getattr_default_np and
2100 pthread_setattr_default_np to get and set the default pthread
2101 attributes of a process.
2103 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2104 This may improve lock scaling of existing programs on TSX capable systems.
2105 When the --enable-lock-elision=yes parameter is specified at configure
2106 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2109 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2110 availability). Implemented by Ryan S. Arnold.
2112 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2116 * The following bugs are resolved with this release:
2118 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2119 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2120 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2121 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2122 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2123 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2124 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2125 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2126 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2127 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2128 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2129 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2130 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2131 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2132 14879, 14889, 14893, 14898, 14914.
2134 * Optimization of memcpy for MIPS.
2136 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2137 EMFILE has been fixed (Bugzilla #14889).
2139 * The add-on ports collection is now distributed in the "ports" subdirectory
2140 of the main GNU C Library distribution, rather than separately.
2142 * Port to ARM AArch64 contributed by Linaro.
2144 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2145 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2147 Implemented by Andreas Krebbel.
2149 * The new function secure_getenv allows secure access to the environment,
2150 returning NULL if running in a SUID/SGID process. This function replaces
2151 the internal function __secure_getenv.
2153 * SystemTap static probes have been added into the dynamic linker.
2154 Implemented by Gary Benson.
2156 * Optimizations of string functions strstr, strcasestr and memmem.
2157 Implemented by Maxim Kuvyrkov.
2159 * The minimum Linux kernel version that this version of the GNU C Library
2160 can be used with is 2.6.16.
2162 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2163 powerpc POWER7. Implemented by Will Schmidt.
2165 * New configure option --disable-nscd builds the C library such that it
2166 never attempts to contact the Name Service Caching Daemon (nscd).
2167 New configure option --disable-build-nscd avoids building nscd itself;
2168 this is the default if --disable-nscd is used.
2170 * Improved support for cross-compilation, including cross-testing and
2171 bootstrap builds without a previously built glibc.
2173 * Several testsuite tests are now able to test multiple IFUNC variants of an
2174 interface, rather than just testing the one that would be chooen by
2177 * New configure options --with-bugurl and --with-pkgversion, for
2178 distributors to use to embed their bug-reporting and package version
2179 information in --help and --version output.
2181 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2182 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2183 allows creation of chroots without the procfs mounted on /proc.
2185 * The `crypt' function now fails if passed salt bytes that violate the
2186 specification for those values. On Linux, the `crypt' function will
2187 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2188 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2189 when the mode is enabled.
2191 * The `clock_*' suite of functions (declared in <time.h>) is now available
2192 directly in the main C library. Previously it was necessary to link with
2193 -lrt to use these functions. This change has the effect that a
2194 single-threaded program that uses a function such as `clock_gettime' (and
2195 is not linked with -lrt) will no longer implicitly load the pthreads
2196 library at runtime and so will not suffer the overheads associated with
2197 multi-thread support in other code such as the C++ runtime library.
2199 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2204 * The following bugs are resolved with this release:
2206 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2207 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2208 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2209 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2210 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2211 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2212 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2213 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2214 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2215 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2216 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2217 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2218 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2219 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2220 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2221 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2222 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2223 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2224 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2225 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2226 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2227 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2230 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2231 configuring glibc with:
2232 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2233 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2234 Implemented by H.J. Lu.
2238 + define static_assert
2240 + do not declare gets
2242 + declare at_quick_exit and quick_exit also for ISO C11
2244 + aligned_alloc. NB: The code is deliberately allows the size parameter
2245 to not be a multiple of the alignment. This is a moronic requirement
2246 in the standard but it is only a requirement on the caller, not the
2249 + timespec_get added
2251 + uchar.h support added
2253 + CMPLX, CMPLXF, CMPLXL added
2255 Implemented by Ulrich Drepper.
2257 * Support for the IA-64 has been moved to ports.
2259 * Remove support for anything but ELF binary format
2261 * Checking versions of poll, ppoll added.
2262 Implemented by Ulrich Drepper.
2264 * More generic and 64-bit performance optimizations to math functions.
2265 Implemented by Ulrich Drepper.
2267 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2268 headers and functions available at compile time as they were before
2269 version 2.14. This option will be removed at some time in the future
2270 after the TI-RPC library becomes fully sufficient for the needs of
2271 existing applications.
2273 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2274 Note that glibc is not expected to work with any Linux kernel version
2277 * New header <sys/auxv.h> and function getauxval allowing easy access to
2278 the AT_* key-value pairs passed from the Linux kernel. The header also
2279 defines the HWCAP_* bits associated with the AT_HWCAP key.
2281 * New locales: mag_IN
2283 * New configure option --enable-systemtap builds SystemTap static probes
2284 into libc for setjmp and longjmp and into libpthread for various operations.
2285 So far the setjmp/longjmp probes and some of the libpthread probes are
2286 provided only for i*86 and x86_64.
2287 Implemented by Roland McGrath and Rayson Ho.
2289 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2291 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2294 * More optimized functions for SPARC. Implemented by David S. Miller.
2296 * Improved support for cross-compilation, especially bootstrap builds
2297 without a previously built glibc.
2299 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2300 Chris Metcalf from Tilera.
2302 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2303 now supported for ARM processors.
2305 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2306 as the name of the dynamic linker, to distinguish it from the
2307 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2309 * Support for CRIS has been removed from ports.
2311 * A new class of installed header has been documented for low-level
2312 platform-specific functionality. PowerPC added the first instance with a
2313 function to provide time base register access. Contributed by Tulio
2314 Magno Quites Machado Filho.
2316 * ix86 configurations now install header files that are consistent with
2317 what x86-64 configurations install. These same header files can be used
2318 for -m32, -m64, or -mx32 builds.
2319 Contributed by H.J. Lu.
2321 * Math library bug fixes. A thorough audit of all open math library bugs was
2322 conducted by Joseph Myers. Significant progress was made on many math
2323 library bugs resulting in more accurate exceptions and function results.
2324 Many thanks to all those that contributed including Andreas Jaeger for his
2325 patch review and work on the x87 trigonometric instruction issues.
2327 * Timezone data is no longer installed. Timezone-related binaries and scripts
2328 will continue to be installed. Users should obtain their timezone data from
2329 their distribution provider or from the tzdata package at
2330 <ftp://munnari.oz.au/pub/>.
2334 * The following bugs are resolved with this release:
2336 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2337 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2338 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2339 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2340 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2341 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2342 13484, 13506, 13515, 13523, 13524, 13538, 13540
2344 * New program pldd to list loaded object of a process
2345 Implemented by Ulrich Drepper.
2347 * Add nss_db support back to glibc. No more dependency on Berkeley db
2348 and support for initgroups lookups.
2349 Implemented by Ulrich Drepper.
2351 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2352 Contributed by HJ Lu.
2354 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2355 Contributed by HJ Lu.
2357 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2358 on x86-32 and x86-64.
2359 Contributed by Liubov Dmitrieva.
2361 * Optimized strchr and strrchr for SSE on x86-32.
2362 Contributed by Liubov Dmitrieva.
2364 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2365 for x86-64 and x86-32.
2366 Contributed by Liubov Dmitrieva.
2368 * New interfaces: scandirat, scandirat64
2369 Implemented by Ulrich Drepper.
2371 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2372 Implemented by Ulrich Drepper.
2374 * nscd now also caches the netgroup database.
2375 Implemented by Ulrich Drepper.
2377 * Integrate libm with gcc's -ffinite-math-only option.
2378 Implemented by Ulrich Drepper.
2380 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2381 to math functions. Implemented by Ulrich Drepper.
2383 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2384 Implemented by Ulrich Drepper.
2386 * New Linux interfaces: process_vm_readv, process_vm_writev
2388 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2389 Implemented by Ulrich Drepper.
2391 * Optimized nearbyint and strcasecmp for PPC.
2392 Implemented by Adhemerval Zanella.
2394 * New locales: bho_IN, unm_US, es_CU, ta_LK
2398 * The following bugs are resolved with this release:
2400 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2401 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2402 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2403 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2404 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2405 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2406 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2407 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2408 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2409 12795, 12811, 12813, 12814, 12841
2411 * The RPC implementation in libc is obsoleted. Old programs keep working
2412 but new programs cannot be linked with the routines in libc anymore.
2413 Programs in need of RPC functionality must be linked against TI-RPC.
2414 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2416 Visible changes of this change include (obviously) the inability to link
2417 programs using RPC functions without referencing the TI-RPC library and the
2418 removal of the RPC headers from the glibc headers.
2419 Implemented by Ulrich Drepper.
2421 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2422 syncfs, setns, sendmmsg
2424 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2425 yue_HK, lij_IT, mhr_RU
2427 * New encodings: CP770, CP771, CP772, CP773, CP774
2429 * New program sotruss to trace calls through PLTs
2430 Implemented by Ulrich Drepper.
2432 * The malloc hook implementation is marked deprecated and will be removed
2433 from the default implementation in the next version. The design never
2434 worked ever since the introduction of threads. Even programs which do
2435 not create threads themselves can use multiple threads created internally.
2439 * The following bugs are resolved with this release:
2441 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2442 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2443 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2444 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2447 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2449 * POWER7 optimizations: memset, memcmp, strncmp
2451 * New optimized string functions for x86-64: strnlen (SSE2),
2452 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2453 Implemented by Ulrich Drepper.
2457 * The following bugs are resolved with this release:
2459 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2460 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2461 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2462 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2463 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2464 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2465 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2466 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2468 * New interfaces: pthread_getname_np, pthread_setname_np
2470 * New Linux interface: recvmmsg
2472 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2474 * The dynamic linker now recognizes supported ABI versions from the
2475 EI_ABIVERSION field in the ELF header.
2476 Implemented by Ulrich Drepper.
2478 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2479 will not be used to fill in password fields in the passwd.byname replies.
2480 Instead it is used to synthesize the shadow.byname table, should it be
2481 missing. This is a useful mode in some installations involving Solaris.
2482 Implemented by Ulrich Drepper.
2484 * New locales: kok_IN, sq_MK, cv_RU
2488 * The following bugs are resolved with this release:
2490 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2491 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2492 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2493 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2494 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2495 10789, 10817, 10823, 10840, 10847
2497 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2499 Implemented by Ulrich Drepper.
2501 * Checking version of longjmp added that fails if an uninitialized stack
2502 frame would be created. Implemented by Ulrich Drepper.
2504 * STT_GNU_IFUNC is now supported in static executables.
2505 Implemented by H.J. Lu.
2507 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2509 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2510 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2511 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2512 Contributed by H.J. Lu.
2514 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2515 Implemented by Ulrich Drepper.
2517 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2519 Contributed by H.J. Lu.
2521 * Support for fma instruction in AVX on x86-64.
2522 Implemented by H.J. Lu and Ulrich Drepper.
2524 * AVX support in x86-64 auditing support in ld.so.
2525 Implemented by H.J. Lu.
2527 * STB_GNU_UNIQUE support added.
2528 Implemented by Ulrich Drepper.
2530 * Implement second fallback mode for optimized DNS lookup for even more
2531 broken environments. If two requests from the same port are not
2532 handled correctly close the socket and open a new one before sending
2533 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2534 can be used to select this mode right away, instead of rediscovering the
2535 necessity is every process again.
2536 Implemented by Ulrich Drepper.
2538 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2539 Implemented by Adam Tkac.
2541 * Optimized iconv conversions for S390x.
2542 Implemented by Andreas Krebbel.
2544 * Using condvars with PI mutexes is now more efficient due to kernel
2545 support for requeueing to PI futexes. NPTL support added for x86-64.
2546 Implemented by Ulrich Drepper.
2548 * New locales: ps_AF, my_MM
2552 * The following bugs are resolved with this release:
2554 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2555 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2556 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2557 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2558 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2560 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2561 Implemented by Ulrich Drepper.
2563 * New Linux interfaces: accept4, fallocate, fallocate64.
2564 Implemented by Ulrich Drepper.
2566 * Correct declarations of string function when used in C++ code. This
2567 could lead to compile errors for invalid C++ code.
2569 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2572 * New POSIX 2008 interface: psiginfo
2573 Implemented by Ulrich Drepper.
2575 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2576 Implemented by Ulrich Drepper.
2578 * Support for selecting between multiple function definitions at runtime
2579 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2581 * The libcrypt library can now use the hash function implementations in
2582 NSS. Implemented by Ulrich Drepper.
2584 * The malloc implementation can be compiled to be less memory efficient
2585 but higher performing in multi-threaded programs.
2586 Implemented by Ulrich Drepper.
2588 * New locales: nan_TW@latin, ks_IN
2590 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2591 Implemented by Ulrich Drepper.
2593 * Extended printf hook support. It is possible to use user-defined types
2594 and extend existing format specifiers.
2595 Implemented by Ulrich Drepper.
2597 * Handling for group shadow files has been added.
2598 Implemented by Ulrich Drepper.
2600 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2601 servers (the case, e.g., for some people using the built-in DNS
2602 server in ADSL modems/routers). There is a once-per-process timeout
2603 in case of a broken server. To avoid it, users can run nscd or put
2604 'options single-request' in /etc/resolv.conf.
2605 Implemented by Ulrich Drepper.
2609 * The following bugs are resolved with this release:
2611 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2612 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2613 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2614 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2615 6968, 6974, 6980, 6995, 7008, 7009, 7029
2617 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2618 up at the same time. Implemented by Ulrich Drepper.
2620 * TLS descriptors for LD and GD on x86 and x86-64.
2621 Implemented by Alexandre Oliva.
2623 * getaddrinfo now handles DCCP and UDPlite.
2624 Implemented by Ulrich Drepper.
2626 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2627 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2628 Implemented by Ulrich Drepper.
2630 * New implementation of memmem, strstr, and strcasestr which is O(n).
2631 Implemented by Eric Blake.
2633 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2635 * Implement "e" option for popen to open file descriptor with the
2636 close-on-exec flag set. Implemented by Ulrich Drepper.
2638 * Implement "b" mode for fmemopen. In this mode writes writes don't
2639 implicitly add a NUL byte and seeks from the end of the buffer really
2640 use the buffer end, not the string length as the basis.
2641 Implemented by Ulrich Drepper.
2643 * Many functions, exported and internal, now atomically set the close-on-exec
2644 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2646 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2648 Implemented by Pravin Satpute.
2650 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2654 * The following bugs are resolved with this release:
2656 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2657 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2658 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2659 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2660 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2661 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2662 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2664 * New locales: bo_CN, bo_IN, shs_CA.
2666 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2668 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2669 Implemented by Pravin Satpute.
2671 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2673 * nscd caches now all timeouts for DNS entries
2674 Implemented by Ulrich Drepper.
2676 * nscd is more efficient and wakes up less often.
2677 Implemented by Ulrich Drepper.
2679 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2680 vdprintf, and obstack_vprintf.
2681 Implemented by Jakub Jelinek.
2683 * Faster memset for x86-64.
2684 Implemented by Harsha Jagasia and H.J. Lu.
2686 * Faster memcpy on x86.
2687 Implemented by Ulrich Drepper.
2689 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2690 Implemented by Ulrich Drepper.
2692 * Faster sqrt and sqrtf implemention for some PPC variants.
2693 Implemented by Steven Munroe.
2697 * The following bugs are resolved with this release:
2699 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2700 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2701 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2702 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2703 5071, 5103, 5104, 5112, 5113, 5184, 5186
2705 * More checking functions: fread, fread_unlocked, open*, mq_open.
2706 Implemented by Jakub Jelinek and Ulrich Drepper.
2708 * Extend fortification to C++. Implemented by Jakub Jelinek.
2710 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2711 by not recognizing 'a' as a modifier when those specs are requested.
2712 Implemented by Jakub Jelinek.
2714 * PPC optimizations to math and string functions.
2715 Implemented by Steven Munroe.
2717 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2718 options to be passed. Implemented by Ulrich Drepper.
2720 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2722 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2723 Implemented by Ulrich Drepper.
2725 * Handle private futexes in the NPTL implementation.
2726 Implemented by Jakub Jelinek and Ulrich Drepper.
2728 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2729 Implemented by Roland McGrath and Ulrich Drepper.
2731 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2733 * SHA-256 and SHA-512 based password encryption.
2734 Implemented by Ulrich Drepper.
2736 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2737 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2740 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2741 Implemented by Ulrich Drepper.
2745 * The following bugs are resolved with this release:
2747 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2748 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2749 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2750 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2751 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2752 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2753 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2756 * New Linux interfaces: epoll_pwait, sched_getcpu.
2758 * New generic interfaces: strerror_l.
2760 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2764 * The following bugs are resolved with this release:
2766 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2767 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2768 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2769 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2770 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2771 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2772 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2773 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2774 3225, 3251, 3252, 3253, 3273
2776 * For Linux, the sorting of addresses returned by getaddrinfo now also
2777 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2778 Implemented by Ulrich Drepper.
2780 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2781 Implemented by Ulrich Drepper.
2783 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2785 * New iconv module for MIK. Contributed by Alexander Shopov.
2787 * For sites with broken group and/or passwd database, the auto-propagate
2788 option of nscd can prevent creating ID lookup entries from the results
2789 of a name lookup and vice versa. This usually is no problem but some
2790 site might have problems with the default behavior.
2791 Implemented by Ulrich Drepper.
2793 * Iterating over entire database in NIS can be slow. With the
2794 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2795 to trade time for memory. The entire database will be read at once.
2796 Implemented by Ulrich Drepper.
2798 * The interfaces introduced in RFC 3542 have been implemented by
2801 * Support for the new ELF hash table format was added by Ulrich Drepper.
2803 * Support for priority inheritance mutexes added by Jakub Jelinek and
2806 * Support for priority protected mutexes added by Jakub Jelinek.
2808 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2809 or_IN, csb_PL, fy_NL, sr_ME.
2813 * More overflow detection functions.
2815 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2816 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2818 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2819 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2820 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2821 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2822 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2823 by Masahide Washizawa.
2825 * It is now possible to install audit libraries for the dynamic linker using
2826 LD_AUDIT. Implemented by Ulrich Drepper.
2828 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2829 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2830 For a libc and libpthread that works well on Linux 2.4 kernels, we
2831 recommend using the stable 2.3 branch.
2833 * The new function open_wmemstream acts like open_memstream,
2834 but uses a wchar_t wide character buffer.
2836 * The new function ppoll is an improved version of poll, similar to pselect.
2838 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2839 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2840 renameat, unlinkat, linkat, symlinkat, readlinkat.
2842 * New Linux kernel system calls: unshare,
2843 inotify_init, inotify_add_watch, inotify_rm_watch.
2845 * The euidaccess function is now also known by the alias eaccess,
2846 for compatibility with some other systems.
2848 * Timezone data updated to 2006b version.
2852 * The following bugs are resolved with this release:
2854 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2855 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2856 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2857 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2858 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2859 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2861 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2863 * As of this release, GCC 4 can be used to compile the C Library.
2865 * Timezone data updated to 2005m version.
2869 * The following bugs are resolved with this release:
2871 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2872 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2873 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2874 777, 787, 821, 822, 823, 825
2876 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2880 * Support for RFC 3678. Real implementations exist only for Linux so far.
2881 Implemented by Ulrich Drepper.
2883 * nscd can now cache entries persistently. Expiring entries are reloaded.
2884 For speedups the cache can be shared in memory with client processes.
2885 Implemented by Ulrich Drepper.
2887 * nscd can now perform SELinux checks.
2888 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2890 * getaddrinfo queries are now cached. Canonical name lookup is performed
2892 Implemented by Ulrich Drepper.
2894 * The nothrow function attribute is used when headers are used by gcc when
2895 compiling C code. This can avoid significant amounts of exception
2898 * The malloc functions perform more error checking and are stricter when
2899 it comes to reacting on errors. The default action is to terminate
2900 the process after showing an error message. Implemented by Ulrich Drepper.
2902 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2903 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2905 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2906 namespaces using the new function dlmopen(). This feature is of course,
2907 like most other dynamic loading functionality, not available in statically
2908 linked applications. Implemented by Ulrich Drepper.
2910 * Low-overhead boundary checking variants of string and some stdio functions
2911 were added. These are to be used in conjunction with a gcc patch by
2912 Jakub Jelinek which adds calls to these functions if possible.
2913 Implemented by Jakub Jelinek and Ulrich Drepper.
2915 * Old code for several operating systems and machine architectures that
2916 have not been in working condition in a long time have been removed from
2917 the main source tree maintained by the GNU C Library's maintainers.
2918 These files are now reside in the separate `ports' source module
2919 that is usable as an add-on when building the library.
2923 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2924 interrogate the dynamic linker, compatible with the Solaris interface.
2926 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2927 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2929 * getifaddrs now uses the netlink interface on Linux to get its information.
2930 Implemented by Thorsten Kukuk.
2932 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2933 Implemented by Ulrich Drepper.
2935 * support for non-executable stacks on x86 has been added. Changes mostly
2938 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2941 * getaddrinfo now performs destination address selection according to
2946 * Thread-safe interfaces for many functions that access locale data
2947 were added in version 2.3, but these features were omitted from NEWS.
2948 Many functions have variants with an `_l' suffix that take a `locale_t'
2949 object as a parameter rather than consulting the current locale.
2950 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2951 create and maintain `locale_t' objects. Additionally, the new function
2952 `uselocale' sets "the current locale" (as used by functions not so
2953 parameterized) set for an individual thread. These features were added
2954 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2956 * The functions getresuid, getresgid, setresuid, and setresgid, which
2957 have long been available on Linux, are now declared in <unistd.h>
2958 and are now also available on the Hurd.
2960 * ELF thread-local storage support (TLS) now works on x86-64.
2962 * The new dynamic string token $LIB is expanded in shared library names.
2963 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2965 * Aldy Hernandez contributed complete software floating point support for
2966 PowerPC machines with no FPU.
2968 * fexecve is implemented on Linux.
2970 * The `btowc' function should work at least twice as fast due to
2971 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2973 * With approriate thread add-ons cancelable functions are now implemented
2974 in libc.so as well. No need to call the function in libpthread. This
2975 change allowed to finally disable the incorrect and expensive handling
2976 of weak definition in ld.so.
2978 * Yet more PLT entries in libc.so have been removed. We finally arrived
2979 at the bare minimum. Startup times improved appropriately.
2981 * Support for the new Linux/x86 system call interface was added. The
2982 AT_SYSINFO auxiliary vector entry is recognized and handled.
2986 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2989 * iconv (the program and the interface) now accepts empty names (excluding
2990 options like //TRANSLIT) to mean "use charset of current locale".
2992 * localedef can now transliterate characters in strings which are not in
2993 the provided charmap. The information from the input locale is used.
2995 * Prelinking support was added for ELF targets. This requires additional
2996 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2998 * Read-only stdio streams now use mmap to speed up operation by eliminating
2999 copying and buffer underflows. To use add 'm' to the mode string of
3000 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3002 * The malloc functions were completely rewritten by Wolfram Gloger based
3003 on Doug Lea's malloc-2.7.0.c.
3005 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3006 implementation of regex.
3008 * Bruno Haible upgraded the iconv and locale implementation to support
3011 * Contents of the LC_* and LANG environment variables in the CEN style are
3012 not recognized anymore. It never was used. Change by Ulrich Drepper.
3014 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3015 thread-local storage (TLS) ABI on some platforms.
3016 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3018 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3019 EUC-JISX0213, and TSCII.
3021 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3022 BSD-compatible interface for getting all network interface addresses.
3023 Implementation for IPv4 by Roland McGrath.
3025 * Loading of locale data is faster due to the introduction of a locale
3026 archive. Implemented by Roland McGrath and Ulrich Drepper.
3028 * Startup times are significantly reduced by not using exported functions
3029 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3032 * Steven Munroe contributed a port to PowerPC64/Linux.
3036 * The Hurd now uses the GNU libio implementation of stdio rather than the
3037 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3039 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3040 with the same behavior as the Linux system call of the same name.
3044 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3045 128-bit long double format.
3047 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3048 IBM1161, and IBM1162 charsets.
3050 * Andreas Jaeger contributed a port to x86-64/Linux.
3052 * Peter Bruin contributed a port to PowerPC/Hurd.
3054 * libc functions using I/O streams now can handle wide-oriented streams
3057 * optimizations in the dynamic linker. Binaries created by recent binutils
3058 versions start up quicker due to reduced time spent on relocations.
3060 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3064 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3065 asinh, atanh, j0 for the 128-bit long double format.
3067 * Bruno Haible updated all the code handling Unicode in some form to
3068 support Unicode 3.1.
3070 * Speed of regex for single-byte locales is back to previous levels.
3071 Patch by Isamu Hasegawa.
3073 * Alpha, SPARC, and IA-64 now also using floating stacks.
3075 * Startup time of internationalized applications greatly improved through
3076 iconv cache. Use iconvconfig to generate the cache file.
3077 Contributed by Ulrich Drepper.
3079 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3080 severe performance problems. Patch by David Mosberger.
3082 * The Hurd port got a lot more functionality like AIO, various stdio
3083 extensions, etc. Mainly done by Roland McGrath.
3085 * mtrace can now lookup symbols in shared libraries.
3089 * Intel's IA-64 math library is largely integrated. It provides fast and
3090 accurate implementatations for most basic and standard math functions
3091 in float, double, and long double format.
3093 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3094 and asin for the 96-bit long double format and asin, log, tan for the
3095 128-bit long double format.
3097 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3098 The basic double functions exist today. Contributed by Abraham Ziv
3099 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3100 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3102 * An asynchronous name lookup library was added. The interface is designed
3103 after POSIX AIO. The proposal was circulated beforehand to get comments.
3104 No negative ones came in. Implemented by Ulrich Drepper.
3106 * Port to S390/64bit contributed by Martin Schwidefsky
3107 <schwidefsky@de.ibm.com>.
3109 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3110 of functions for Linux/IA-64.
3112 * The RPC code is now thread safe. Threads can now use the same service
3113 of different services at the same time. Patch by Eric Norum
3114 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3116 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3117 family of functions for Linux/S390.
3119 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3120 of functions for Linux/x86.
3122 * Port to Linux/CRIS contributed by Axis Communications.
3126 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3127 we can now check for namespace violations and missing declarations. The
3128 result is that almost all headers are now Unix-compliant (as defined in
3129 the upcoming XPG6). The negative side is that some programs might need
3130 corrections, too, if they depend on the incorrect form of the headers in
3131 previous versions which defined too many symbols and included too many
3134 * regex now handles multibyte character sets correctly.
3135 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3137 * iconv (the program) does now conform to the upcoming XPG6 and handles
3138 charmaps. Instead of the charset names the path of charmaps can be
3139 provided and the conversion happens based on this data.
3140 Contributed by Ulrich Drepper.
3142 * The locale program now provides detailed information about the installed
3147 only lists the names of the supported locales
3151 provides details such as country, language, and codeset name.
3152 Contributed by Ulrich Drepper.
3156 * The gencat program now parses the input file according to the charset
3157 selected by the LC_CTYPE category. This is important for stateful
3158 character sets. To make generating catalogs easier there is a way
3159 to overwrite the charset selected by the locale: before the first
3160 message or $ quote line the catalog can contain a line like
3162 $ codeset=ISO-8859-2
3164 to select the charset (ISO-8859-2 in this case).
3166 Implemented by Shinya Hanataka and Ulrich Drepper.
3168 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3169 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3170 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3172 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3173 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3175 * The LANGUAGE environment variable is now ignored unless the locale is
3176 changed from the default "C" locale.
3178 * The usual bug fixes.
3182 * Greg McGary added runtime support for bounds checking using gcc's
3183 new -fbounded-pointers option. ix86 target is complete. PowerPC
3186 * Thorsten Kukuk added secure mode support to nscd.
3188 * The Berkeley db libraries have been removed.
3190 Related, the nss_db module is now in a separate package since it
3191 obviously requires a database library being available.
3193 * Wide character I/O streams implemented by Ulrich Drepper.
3195 * Functions from the extended socket API added by Ulrich Drepper.
3197 * Functions feenableexcept and fedisableexcept to control the
3198 behaviour of individual exceptions have been added by Andreas Jaeger.
3200 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3202 * The resolver code has been updated from bind 8.2.3-T5B which supports
3203 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3206 This change could in some situations effect backward compatibility. Since
3207 now `_res' is a thread-local instead of a global variable, modifying it
3208 in one thread does not have any effect in other threads.
3210 The resolver library was also extended to allow IPv6 as the transport
3211 protocol for the requests. This work was done by Stig Venaas.
3213 * Compatibility code for K&R C compilers has been removed from the
3214 header files. A ISO C compiler is needed to use the library
3215 (conforming to either C89 or C99 standard).
3217 * Complete rewrite of the localedef program to support multibyte character
3218 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3219 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3220 Implemented by Ulrich Drepper.
3222 Bruno Haible significantly improved the generation and use of the data
3223 structures for the wide character tables.
3225 * Plural handling in gettext implemented by Ulrich Drepper.
3227 * The utmp daemon has been removed.
3229 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3231 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3234 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3236 * POSIX barriers implemented by Kaz Kylheku.
3238 * POSIX spawn function family implemented by Ulrich Drepper.
3240 * POSIX spinlocks are now available.
3242 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3244 * the configure option --enable-kernel=X.Y.Z allows to strip out
3245 compatibility for kernel versions before X.Y.Z. This is currently only
3246 implemented for Linux.
3248 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3249 field sin6_scope_id. This means that all programs using IPv6 should be
3250 recompiled. Don't expect binary compatibility with previous glibc
3253 * various conversion modules for IBM character sets contributed by
3256 * IA-64 port by Jes Sorensen and HJ Lu.
3270 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3272 * New xdr functions are added; some rpc functions are now 64bit clean.
3274 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3276 * Fixed known incompatibilities with glibc 2.0.
3278 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3280 * Optimized a number of functions (especially the ELF dynamic loader).
3282 * Update timezone data files.
3284 * lots of charmaps corrections
3286 * some new locale definitions and charmaps
3291 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3292 conform to POSIX member type requirements. Symbol versions have been
3293 adjusted as needed within the library, and for direct use by applications,
3294 but there is potential for problems if third-party libraries use
3295 struct timeval as part of their interface. This does not present
3296 a problem for X and other "essential" system libraries.
3298 * An additional locale model to support C++ Standard Library locale
3299 model and probably more was implemented by Ulrich Drepper.
3301 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3304 * Miles Bader provided the `argp' function family to support hierachical
3305 command line argument parsing, layered on top of getopt.
3307 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3309 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3312 * scanf recognizes the %a and %A format for scanning floating point numbers.
3314 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3315 information and interfaces for the available integer types.
3317 * about 130 new math functions were added to implement the ISO C9x math
3320 * the new header <complex.h> contains definitions of the complex math
3321 functions from ISO C 9X.
3323 * the new header <tgmath.h> defines generic macros to use complex or
3324 real valued functions.
3326 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3328 * Andreas Jaeger provided a test suite for the math library.
3330 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3332 * Ulrich Drepper added character set conversion functions (iconv).
3334 * Optimized string functions have been added.
3336 * The localedata addon is now part of glibc.
3338 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3340 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3341 daemon for NSS (nscd).
3343 Missing a better place here are some numbers on improvements. Under
3344 Linux 2.1.125 un-tar-ing the kernel sources takes
3348 using local files 12.19s 6.88s 22.91s
3350 using NIS 13.92s 8.91s 26.34s
3352 using NIS & nscd 10.37s 7.34s 25.30s
3354 using NIS+ 27.57s 30.37s 640.46s
3356 using NIS+ & nscd 10.25s 7.83s 26.51s
3358 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3360 Keep in mind that non-namelookup related operations dominate above times.
3361 It was just a common complain that using NIS+ unpacking the kernel is
3364 [1] The old nscd implementation is not available anymore since it was
3365 distributed with glibc up to version 2.0.98 and thus is now replaced.
3367 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3369 * Mark Kettenis provided a Hesiod NSS module.
3371 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3372 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3374 * rcmd can now handle netgroups (Dick Streefland).
3376 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3379 * Support for the IPv6 protocol has been added to the socket API, as per the
3380 latest draft standards.
3382 * Support for Linux 2.2 has been added.
3384 * Interface changes relative to the latest 2.0.x release:
3385 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3386 addseverity NEW: Unix98
3387 alphasort64 NEW: LFS
3388 argp_err_exit_status NEW: argp, GNU ext
3389 argp_error NEW: argp, GNU ext
3390 argp_failure NEW: argp, GNU ext
3391 argp_help NEW: argp, GNU ext
3392 argp_parse NEW: argp, GNU ext
3393 argp_program_bug_address NEW: argp, GNU ext
3394 argp_program_version NEW: argp, GNU ext
3395 argp_program_version_hook NEW: argp, GNU ext
3396 argp_state_help NEW: argp, GNU ext
3397 argp_usage NEW: argp, GNU ext
3398 authdes_create NEW: Secure RPC
3399 authdes_getucred NEW: Secure RPC
3400 authdes_pk_create NEW: Secure RPC
3401 backtrace NEW: GNU ext.
3402 backtrace_symbols NEW: GNU ext.
3403 backtrace_symbols_fd NEW: GNU ext.
3405 cacosf NEW: ISO C 9x
3406 cacosh NEW: ISO C 9x
3407 cacoshf NEW: ISO C 9x
3408 cacoshl NEW: ISO C 9x
3409 cacosl NEW: ISO C 9x
3416 casinf NEW: ISO C 9x
3417 casinh NEW: ISO C 9x
3418 casinhf NEW: ISO C 9x
3419 casinhl NEW: ISO C 9x
3420 casinl NEW: ISO C 9x
3422 catanf NEW: ISO C 9x
3423 catanh NEW: ISO C 9x
3424 catanhf NEW: ISO C 9x
3425 catanhl NEW: ISO C 9x
3426 catanl NEW: ISO C 9x
3427 cbc_crypt NEW: Secure RPC
3431 ccoshf NEW: ISO C 9x
3432 ccoshl NEW: ISO C 9x
3438 cimagf NEW: ISO C 9x
3439 cimagl NEW: ISO C 9x
3440 clearerr_locked REMOVED
3441 clntunix_create NEW: sunrpc ext
3443 clog10 NEW: ISO C 9x
3444 clog10f NEW: ISO C 9x
3445 clog10l NEW: ISO C 9x
3455 cprojf NEW: ISO C 9x
3456 cprojl NEW: ISO C 9x
3458 crealf NEW: ISO C 9x
3459 creall NEW: ISO C 9x
3464 csinhf NEW: ISO C 9x
3465 csinhl NEW: ISO C 9x
3468 csqrtf NEW: ISO C 9x
3469 csqrtl NEW: ISO C 9x
3473 ctanhf NEW: ISO C 9x
3474 ctanhl NEW: ISO C 9x
3476 des_setparity NEW: Secure RPC
3477 ecb_crypt NEW: Secure RPC
3478 endutxent NEW: Unix98
3480 exp10f NEW: ISO C 9x
3481 exp10l NEW: ISO C 9x
3485 fattach NEW: STREAMS
3486 fdetach NEW: STREAMS
3490 feclearexcept NEW: ISO C 9x
3491 fegetenv NEW: ISO C 9x
3492 fegetexceptflag NEW: ISO C 9x
3493 fegetround NEW: ISO C 9x
3494 feholdexcept NEW: ISO C 9x
3496 feraiseexcept NEW: ISO C 9x
3497 ferror_locked REMOVED
3498 fesetenv NEW: ISO C 9x
3499 fesetexceptflag NEW: ISO C 9x
3500 fesetround NEW: ISO C 9x
3501 fetestexcept NEW: ISO C 9x
3502 feupdateenv NEW: ISO C 9x
3503 fflush_locked REMOVED
3507 fgets_unlocked NEW: GNU ext.
3508 fileno_locked REMOVED
3520 fputc_locked REMOVED
3521 fputs_unlocked NEW: GNU ext.
3522 fread_unlocked NEW: GNU ext.
3527 fstatvfs NEW: Unix98
3531 ftruncate64 NEW: LFS
3533 fwrite_unlocked NEW: GNU ext.
3534 gai_strerror NEW: IPv6
3538 getchar_locked REMOVED
3540 getdate_err NEW: Unix98
3541 getdate_r NEW: GNU ext.
3543 getnameinfo NEW: IPv6
3544 getnetname NEW: Secure RPC
3545 getpmsg NEW: STREAMS
3546 getpt NEW: Unix98 PTY
3547 getrlimit64 NEW: LFS
3548 getutxent NEW: Unix98
3549 getutxid NEW: Unix98
3550 getutxline NEW: Unix98
3551 glob64 NEW: GNU ext.
3552 globfree64 NEW: GNU ext.
3553 gnu_get_libc_release NEW: GNU ext.
3554 gnu_get_libc_version NEW: GNU ext.
3555 grantpt NEW: Unix98 PTY
3556 host2netname NEW: Secure RPC
3558 iconv_close NEW: iconv
3559 iconv_open NEW: iconv
3560 if_freenameindex NEW: IPv6
3561 if_indextoname NEW: IPv6
3562 if_nameindex NEW: IPv6
3563 if_nametoindex NEW: IPv6
3564 in6addr_any NEW: IPv6
3565 in6addr_loopback NEW: IPv6
3566 inet6_isipv4mapped NEW: IPv6
3567 isastream NEW: STREAMS
3568 iswblank NEW: GNU ext.
3569 key_decryptsession NEW: Secure RPC
3570 key_decryptsession_pk NEW: Secure RPC
3571 key_encryptsession NEW: Secure RPC
3572 key_encryptsession_pk NEW: Secure RPC
3573 key_gendes NEW: Secure RPC
3574 key_get_conv NEW: Secure RPC
3575 key_secretkey_is_set NEW: Secure RPC
3576 key_setnet NEW: Secure RPC
3577 key_setsecret NEW: Secure RPC
3578 llrint NEW: ISO C 9x
3579 llrintf NEW: ISO C 9x
3580 llrintl NEW: ISO C 9x
3581 llround NEW: ISO C 9x
3582 llroundf NEW: ISO C 9x
3583 llroundl NEW: ISO C 9x
3588 lrintf NEW: ISO C 9x
3589 lrintl NEW: ISO C 9x
3590 lround NEW: ISO C 9x
3591 lroundf NEW: ISO C 9x
3592 lroundl NEW: ISO C 9x
3594 makecontext NEW: Unix98
3595 mempcpy NEW: GNU ext.
3598 modify_ldt NEW: kernel
3602 nearbyint NEW: ISO C 9x
3603 nearbyintf NEW: ISO C 9x
3604 nearbyintl NEW: ISO C 9x
3605 netname2host NEW: Secure RPC
3606 netname2user NEW: Secure RPC
3607 nexttoward NEW: ISO C 9x
3608 nexttowardf NEW: ISO C 9x
3609 nexttowardl NEW: ISO C 9x
3613 passwd2des NEW: Secure RPC
3615 pow10f NEW: GNU ext.
3616 pow10l NEW: GNU ext.
3619 printf_size NEW: GNU ext.
3620 printf_size_info NEW: GNU ext.
3621 profil_counter REMOVED
3622 pthread_mutexattr_getkind_np REPLACED
3623 pthread_mutexattr_setkind_np REPLACED
3624 ptsname NEW: Unix98 PTY
3625 ptsname_r NEW: Unix98 PTY
3627 putchar_locked REMOVED
3628 putgrent NEW: GNU ext.
3630 putpmsg NEW: STREAMS
3631 pututxline NEW: Unix98
3635 readdir64_r NEW: LFS
3636 remquo NEW: ISO C 9x
3637 remquof NEW: ISO C 9x
3638 remquol NEW: ISO C 9x
3640 roundf NEW: ISO C 9x
3641 roundl NEW: ISO C 9x
3643 scalbln NEW: ISO C 9x
3644 scalblnf NEW: ISO C 9x
3645 scalblnl NEW: ISO C 9x
3647 sendfile NEW: kernel
3648 setcontext NEW: Unix98
3649 setrlimit64 NEW: LFS
3650 setutxent NEW: Unix98
3652 sigignore NEW: Unix98
3653 sigqueue NEW: POSIX.1b
3654 sigrelse NEW: Unix98
3655 sigset NEW: POSIX.1b
3656 sigtimedwait NEW: POSIX.1b
3657 sigwaitinfo NEW: POSIX.1b
3658 sincos NEW: GNU ext.
3659 sincosf NEW: GNU ext.
3660 sincosl NEW: GNU ext.
3664 strcasestr NEW: GNU ext.
3665 strtoimax NEW: ISO C 9x
3666 strtoumax NEW: ISO C 9x
3667 strverscmp NEW: GNU ext.
3668 svcauthdes_stats NEW: Secure RPC
3669 svcunix_create NEW: sunrpc etx
3670 svcunixfd_create NEW: sunrpc ext
3671 swapcontext NEW: Unix98
3672 tcgetsid NEW: Unix98 PTY
3673 tdestroy NEW: GNU ext.
3674 tgamma NEW: ISO C 9x
3675 tgammaf NEW: ISO C 9x
3676 tgammal NEW: ISO C 9x
3680 truncf NEW: ISO C 9x
3681 truncl NEW: ISO C 9x
3683 unlockpt NEW: Unix98 PTY
3684 updwtmpx NEW: Unix98
3685 user2netname NEW: Secure RPC
3686 utmpxname NEW: Unix98
3687 versionsort NEW: GNU ext.
3688 versionsort64 NEW: GNU ext.
3690 wcscasecmp NEW: GNU ext.
3691 wcsncasecmp NEW: GNU ext.
3692 wcsnlen NEW: GNU ext.
3693 wcstoimax NEW: ISO C 9x
3694 wcstoll NEW: ISO C 9x
3695 wcstoull NEW: ISO C 9x
3696 wcstoumax NEW: ISO C 9x
3698 wordexp NEW: POSIX.2
3699 wordfree NEW: POSIX.2
3700 write_profiling REMOVED
3701 xdecrypt NEW: Secure RPC
3702 xdr_authdes_cred NEW: Secure RPC
3703 xdr_authdes_verf NEW: Secure RPC
3704 xdr_cryptkeyarg NEW: Secure RPC
3705 xdr_cryptkeyarg2 NEW: Secure RPC
3706 xdr_cryptkeyres NEW: Secure RPC
3707 xdr_getcredres NEW: Secure RPC
3708 xdr_key_netstarg NEW: Secure RPC
3709 xdr_key_netstres NEW: Secure RPC
3710 xdr_keybuf NEW: Secure RPC
3711 xdr_keystatus NEW: Secure RPC
3712 xdr_netnamestr NEW: Secure RPC
3713 xdr_sizeof NEW: Secure RPC
3714 xdr_unixcred NEW: sunrpc ext
3715 xencrypt NEW: Secure RPC
3716 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3727 * inet_ntoa is thread-safe
3729 * updwtmp is moved from libutil to libc
3731 * rewrite of cbrt function
3733 * update of timezone data
3747 * add atoll function
3749 * fix complex problems in Berkeley DB code
3751 * fix math functions
3755 * fixed lots of header problems (especially Linux/GNU specific)
3757 * dynamic loader preserves all registers
3759 * Roland McGrath provided support for handling of auxiliary objects in
3760 the ELF dynamic loader.
3762 * support for parallel builds is improved
3766 * GNU extensions are no longer declared by default. To enable them you
3767 must define the macro `_GNU_SOURCE' in your program or compile with
3770 * The library has changed from using GNU ld symbol aliases to using weak
3771 symbols where available. The ELF object file format supports weak
3772 symbols; GNU ld also supports weak symbols in the a.out format. (There
3773 is also now support for other GNU ld extensions in ELF. Use the
3774 `--with-elf' option to configure to indicate you have ELF, and
3775 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3776 of many files which contained only symbol aliases, reducing the size of
3777 the source and the compiled library; many other files were renamed to
3778 less cryptic names previously occupied by the symbol alias files.
3779 There is a new header file <elf.h> for programs which operate on
3780 files in the ELF format.
3782 * Converted to Autoconf version 2, so `configure' has more options.
3783 Run `configure --help' to see the details.
3785 * The library can now be configured to build profiling, highly-optimized
3786 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3787 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3788 `configure' enable building these extra libraries. The shared library is
3789 built by default when using both ELF and GNU ld. When shared libraries
3790 are enabled, the new library `-ldl' is available for arbitrary run-time
3791 loading of shared objects; its interface is defined in <dlfcn.h>. The
3792 new header file <link.h> gives access to the internals of the run-time
3793 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3794 application of same name on other systems and it provides information
3795 about dynamically linked binaries.
3797 * The C library now provides the run-time support code for profiling
3798 executables compiled with `-pg'. Programs can control the profiling code
3799 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3800 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3801 the support for this file format was contributed by David Mosberger-Tang.
3803 * The math code has been replaced with a math library based on fdlibm from
3804 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3805 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3806 functions. The math functions now reside in a separate library, so
3807 programs using them will need to use `-lm' their linking commands.
3809 * John C. Bowman contributed optimized ix87 assembler inline functions.
3811 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3812 mechanism similar to that found in Solaris 2. This is now used for the
3813 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3814 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3815 file controls what services are used for each individual database. This
3816 works by loading shared libraries with names specified in `nsswitch.conf',
3817 so service modules can be changed or added at any time without even
3818 relinking any program. Currently there are the file, db, and NIS based
3819 NSS services available.
3821 * The new functions `strtoq' and `strtouq' parse integer values from
3822 strings, like `strtol' and `strtoul', but they return `long long int' and
3823 `unsigned long long int' values, respectively (64-bit quantities).
3825 * The new functions `strtof' and `strtold' parse floating-point values from
3826 strings, like `strtod', but they return `float' and `long double' values,
3827 respectively (on some machines `double' and `long double' are the same).
3829 * Ulrich Drepper has contributed new implementations of the floating-point
3830 printing and reading code used in the `printf' family of functions and
3831 `strtod', `strtof', and `strtold'. These new functions are perfectly
3832 accurate, and much faster than the old ones.
3834 * The implementation of the POSIX locale model was completely rewritten by
3835 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3836 to compile the POSIX locale definition.
3838 * The former dummy implementations of the strcoll and strxfrm function are
3839 now replaced by fully functional code contributed by Ulrich Drepper. The
3840 collation information comes from the POSIX locale definitions.
3842 * The new header <langinfo.h> defines an interface for accessing
3843 various locale-dependent data (using the locale chosen with `setlocale').
3845 * Ulrich Drepper has contributed a new suite of functions for operation on
3846 wide-character and multibyte-character strings, in <wchar.h>;
3847 and classification and case conversion of wide characters, in <wctype.h>.
3848 These new functions are conforming to the ISO C, Amendement 1 specification.
3850 * There is now a second implementation of the standard I/O library available.
3851 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3852 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3854 * You can now use positional parameter specifications in format strings
3855 for the `printf' and `scanf' families of functions. For example,
3856 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3857 ``Number 6, Mr Jones''. This is mainly useful when providing different
3858 format strings for different languages, whose grammars may dictate
3859 different orderings of the values being printed. To support this
3860 feature, the interface for `register_printf_handler' has changed; see
3861 the header file <printf.h> for details.
3863 * The `printf' and `scanf' families of functions now understand a new
3864 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3865 to group numbers as indicated by the locale; for `scanf' and friends, this
3866 says to accept as valid only a number with all the proper grouping
3867 separators in the right places. In the default "C" locale, numbers are
3868 not grouped; but locales for specific countries will define the usual
3869 conventions (i.e. separate thousands with `,' in the US locale).
3871 * The pgrp functions have been regularized, slightly incompatibly but much
3872 less confusingly. The core functions are now `getpgid' and `setpgid',
3873 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3874 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3875 provided for compatibility. There is no longer an incompatible `getpgrp'
3876 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3878 * The new header file <fts.h> and suite of functions simplify programs that
3879 operate on directory trees. This code comes from 4.4 BSD.
3881 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3882 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3883 NSS scheme used in glibc.
3885 * There is a new malloc debugging hook `__memalign_hook'.
3887 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3888 `unsigned int' in <sys/types.h>. These are for compatibility only and
3889 their use is discouraged.
3891 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3892 done differently, so that it works even without GNU ld.
3894 * New function `euidaccess' checks allowed access to a file like `access',
3895 but using the effective IDs instead of the real IDs.
3897 * The time zone data files have been updated for the latest and greatest
3898 local time conventions of the countries of the world.
3900 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3903 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3904 for formatting floating-point numbers. They are provided only for
3905 compatibility; new programs should use `sprintf' instead. There are
3906 also equivalent function for the `long double' floating-point type and
3907 all functions also exist in a reentrant form.
3909 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3910 functions for maintaining the login-record files (primarily of use to
3911 system programs such as `login'), and convenient functions for
3912 allocating and initializing a pseudo-terminal (pty) device.
3914 * Ulrich Drepper has contributed new support for System V style
3915 shared memory and IPC on systems that support it.
3917 * Ulrich Drepper has contributed several miscellaneous new functions found
3918 in System V: The `hsearch' family of functions provide an effective
3919 implementation of hash tables; `a64l' and `l64a' provide a very simple
3920 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3923 * Ulrich Drepper has contributed new reentrant counterparts for the
3924 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3926 * Ulrich Drepper has contributed new, highly-optimized versions of several
3927 string functions for the i486/Pentium family of processors.
3929 * Ulrich Drepper has updated the Linux-specific code, based largely
3930 on work done in Hongjiu Lu's version of GNU libc for Linux.
3931 The GNU library now supports Linux versions 2.0.10 and later,
3932 using the ELF object file format (i[3456]86-*-linux).
3934 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3936 * David Mosberger-Tang and Richard Henderson have ported the C library
3937 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3938 dynamic linking support for ELF/Alpha.
3940 * Richard Henderson contributed several Alpha optimized assembler function
3941 for arithmetic and string handling.
3943 * Ulrich Drepper has contributed a new set of message catalog functions to
3944 support multiple languages using the <libintl.h> interface, for use with
3945 his new package GNU gettext. Translation volunteers have contributed
3946 catalogs of the library's messages in Spanish, German, and Korean.
3948 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3949 program and the `catgets' function for reading the catalog files it
3950 creates. (The <libintl.h> interface is preferred; we include the
3951 <nl_types.h> interface using `catgets' only for source compatibility with
3952 programs already written to use it.)
3954 * New header file <values.h> gives SVID-compatible names for <limits.h>
3957 * Various new macros, declarations, and small header files for compatibility
3960 * New function `group_member' is a convenient way to check if a process has
3961 a given effective group ID.
3963 * When using GCC 2.7 and later, the socket functions are now declared in a
3964 special way so that passing an argument of type `struct sockaddr_in *',
3965 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3966 `struct sockaddr *' type, does not generate a type-clash warning.
3968 * New function `error' declared in header file <error.h> is a convenient
3969 function for printing error messages and optionally exiting; this is the
3970 canonical function used in GNU programs. The new functions `err', `warn',
3971 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3972 doing the same thing.
3974 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3975 POSIX.2 `glob' function to do ~ and {...} expansion.
3977 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3978 `clearenv' which is used in POSIX.9 is also available.
3980 * New function `getsid' returns session ID number on systems that support it.
3982 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3983 files <db.h> and <mpool.h> provide a rich set of functions for several
3984 types of simple databases stored in memory and in files, and <ndbm.h> is
3985 an old `ndbm'-compatible interface using the `db' functions. Link with
3986 `-ldb' to get these functions.
3988 * New macro `strdupa' copies a string like `strdup', but uses local stack
3989 space from `alloca' instead of dynamic heap space from `malloc'.
3991 * New function `strnlen' is like `strlen' but searches only a given maximum
3992 number of characters for the null terminator. `stpncpy', `strndup' and
3993 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3996 * New function `statfs' in header <sys/statfs.h>.
3998 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3999 provide convenient functions for operating on blocks of null-terminated
4002 * A new suite of functions in <utmp.h> handle all the details of reading
4003 and writing the utmp file.
4005 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4008 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4009 completely POSIX compliant and also implemented the extended functionality
4010 to handle alternate digit representation and alternate era date formats.
4012 * Ulrich Drepper provided an implementation of the `strptime' function
4013 defined in XPG4.2 which transforms a string into a `struct tm' value.
4015 * Paul Eggert provided the tzselect shell script as part of the timezone
4016 code. The shell script makes it easy to select the correct timezone
4019 * The implementation of the malloc family of functions is completely replaced
4020 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4021 The implementation uses the mmap function (if available) and it is
4022 optimized for the use in multi threaded programs.
4024 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4025 functions. This new functionality is usable by specifying a special
4026 salt string and it is compatible with implementation on *BSD systems.
4028 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4029 `getsubopt' to handle second level command line options, `bsd_signal'
4030 to access BSD style `signal' functionality, the obsolete `regexp' style
4033 * the `lchown' function is available on system which support this
4036 * The implementation of the shadow password handling function was contributed
4039 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4041 * POSIX.1g support was added. The <sys/select.h> header is available,
4042 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4043 implementation of `getaddrinfo'.
4047 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4049 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4050 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4052 * The new function `daemon' from 4.4 BSD is useful for server programs that
4053 want to put themselves in the background.
4055 * Joel Sherrill has contributed support for several standalone boards that
4056 run without an operating system.
4058 * `printf', `scanf' and friends now accept a `q' type modifier for long
4059 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4061 * All of the code taken from BSD (notably most of the math and networking
4062 routines) has been updated from the BSD 4.4-Lite release.
4064 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4066 * The new functions `getdomainname' and `setdomainname' fetch or change the
4067 YP/NIS domain name. These are system calls which exist on systems which
4070 * The time zone data files have been updated for the latest international
4073 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4074 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4078 * The C library now includes support for Sun RPC, from Sun's free
4079 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4080 are included. (There is still no support for YP.)
4082 * Tom Quinn has contributed a port of the C library to SGI machines running
4083 Irix 4 (mips-sgi-irix4).
4085 * The new `lockf' function is a simplified interface to the locking
4086 facilities of `fcntl', included for compatibility.
4088 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4090 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4092 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4095 * The error message printed by `assert' for a failed assertion now includes
4096 the name of the program (if using GNU ld) and the name of the calling
4097 function (with versions of GCC that support this).
4099 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4101 * The library now includes the <sys/mman.h> header file and memory
4102 management functions `mmap', `munmap', `mprotect', `msync', and
4103 `madvise', on systems that support those facilities.
4105 * The interface for `mcheck' has changed slightly: the function called to
4106 abort the program when an allocation inconsistency is detected now takes
4107 an argument that indicates the type of failure. The new function
4108 `mprobe' lets you request a consistency check for a particular block at
4109 any time (checks are normally done only when you call `free' or `realloc'
4112 * It is now possible to easily cross-compile the C library, building on one
4113 system a library to run on another machine and/or operating system. All
4114 you need to do is set the variable `HOST_CC' in `configparms' to the
4115 native compiler for programs to run on the machine you are building on (a
4116 few generator programs are used on Unix systems); set `CC' to the
4119 * The new function `fexecve' (only implemented on the GNU system) executes
4120 a program file given a file descriptor already open on the file.
4124 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4125 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4127 * You can set the variable `libprefix' in `configparms' to specify a prefix
4128 to be prepended to installed library files; this makes it easy to install
4129 the GNU C library to be linked as `-lgnuc' or whatever.
4131 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4132 copies a limited number of characters from a string, and returns the
4133 address of the last character written.
4135 * You no longer need to check for whether the installed `stddef.h' is
4136 compatible with the GNU C library. configure now checks for you.
4138 * You can now define a per-stream `fileno' function to convert the
4139 stream's cookie into an integral file descriptor.
4141 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4142 allocates zero bytes of storage, and returns a unique pointer which
4143 you can pass to `realloc' or `free'. The behavior is undefined if
4144 you dereference this pointer.
4146 * The C library now runs on Sony NEWS m68k machines running either
4147 NewsOS 3 or NewsOS 4.
4149 * The new `syscall' function is a system-dependent primitive function
4150 for invoking system calls. It has the canonical behavior on Unix
4151 systems, including unreliable return values for some calls (such as
4152 `pipe', `fork' and `getppid').
4154 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4155 to `EAGAIN', which is the preferred name. On systems whose kernels
4156 use two distinct codes, the C library now translates EWOULDBLOCK to
4157 EAGAIN in every system call function.
4161 * The GNU C Library Reference Manual is now distributed with the library.
4162 `make dvi' will produce a DVI file of the printed manual.
4163 `make info' will produce Info files that you can read on line using C-h i
4164 in Emacs or the `info' program.
4165 Please send comments on the manual to bug-glibc-manual@gnu.org.
4167 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4169 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4171 * Jason Merrill has contributed a port to the Sequent Symmetry running
4172 Dynix version 3 (i386-sequent-dynix).
4174 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4175 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4177 * New function `memory_warnings' lets you arrange to get warnings when
4178 malloc is running out of memory to allocate, like Emacs gives you.
4180 * The C library now contains the relocating allocator used in Emacs 19 for
4181 its editing buffers. This allocator (ralloc) minimizes allocation
4182 overhead and fragmentation by moving allocated regions around whenever it
4183 needs to. You always refer to a ralloc'd region with a "handle" (a
4184 pointer to a pointer--an object of type `void **').
4186 * There is a new `printf' format: `%m' gives you the string corresponding
4187 to the error code in `errno'.
4189 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4190 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4191 pass, the `a' modifier says to fill in a `char **' you pass with a
4194 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4195 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4196 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4198 * `mkstemp' is a traditional Unix function to atomically create and open a
4199 uniquely-named temporary file.
4203 * The standard location for the file that says what the local timezone is
4204 has changed again. It is now `/usr/local/etc/localtime' (or more
4205 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4207 * The distribution no longer contains any files with names longer than 14
4210 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4211 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4213 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4217 * The configuration process has changed quite a bit. The `configure'
4218 script is now used just like the configuration scripts for other GNU
4219 packages. The `sysdeps' directory hierarchy is much rearranged.
4220 The file `INSTALL' explains the new scheme in detail.
4222 * The header files no longer need to be processed into ANSI C and
4223 traditional C versions. There is just one set of files to install, and
4224 it will work with ANSI or old C compilers (including `gcc -traditional').
4226 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4227 MIPS DECStation running Ultrix 4.
4229 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4230 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4231 made itself into a shared library.
4233 * Yet further improved support for the i386, running 4.3 BSD-like systems
4234 (such as Mach 3 with the Unix single-server), or System V.
4236 * New function `strncasecmp' to do case-insensitive string comparison
4237 with limited length.
4239 * New function `strsep' is a reentrant alternative to `strtok'.
4241 * New functions `scandir' and `alphasort' for searching directories.
4243 * New function `setenv' is a better interface to `putenv'.
4245 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4246 function for traversing a directory tree.
4248 * The GNU obstack package is now also part of the C library.
4249 The new function `open_obstack_stream' creates a stdio stream that
4250 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4251 formatted output directly to an obstack.
4253 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4254 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4256 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4258 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4259 things to your strings.
4261 * There are some new test programs: test-fseek, testmb, and testrand.
4263 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4264 These ports are not finished, but are a good starting place for really
4265 supporting those systems.
4267 * `/etc/localtime' is now the standard location for the file that says what
4268 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4269 This follows the general principle that `/etc' is the place for all local
4270 configuration files.
4272 * The C library header files now use `extern "C"' when used by the C++
4273 compiler, so the C library should now work with C++ code.
4275 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4276 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4279 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4280 function which uses the `merge sort' algorithm, and is said to be
4281 significantly faster than the old GNU `qsort' function. Merge sort is
4282 now the standard `qsort' function. The new algorithm can require a lot
4283 of temporary storage; so, the old sorting function is called when the
4284 required storage is not available.
4286 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4287 provides the Unix `crypt' function, plus some other entry points.
4289 * The code and header files taken from 4.4 BSD have been updated with the
4290 latest files released from Berkeley.
4292 ----------------------------------------------------------------------
4293 Copyright information:
4295 Copyright (C) 1992-2017 Free Software Foundation, Inc.
4297 Permission is granted to anyone to make or distribute verbatim copies
4298 of this document as received, in any medium, provided that the
4299 copyright notice and this permission notice are preserved,
4300 thus giving the recipient permission to redistribute in turn.
4302 Permission is granted to distribute modified versions
4303 of this document, or of portions of it,
4304 under the above conditions, provided also that they
4305 carry prominent notices stating who last changed them.
4308 version-control: never