1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <http://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * A per-thread cache has been added to malloc. Access to the cache requires
13 no locks and therefore significantly accelerates the fast path to allocate
14 and free small amounts of memory. Refilling an empty cache requires locking
15 the underlying arena. Performance measurements show significant gains in a
16 wide variety of user workloads. Workloads were captured using a special
17 instrumented malloc and analyzed with a malloc simulator. Contributed by
18 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
20 * Unicode 10.0.0 Support: Character encoding, character type info, and
21 transliteration tables are all updated to Unicode 10.0.0, using
22 generator scripts contributed by Mike FABIAN (Red Hat).
23 These updates cause user visible changes, especially the changes in
24 wcwidth for many emoji characters cause problems when emoji sequences
25 are rendered with pango, see for example:
26 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
28 * Collation of Hungarian has been overhauled and is now consistent with "The
29 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
32 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
34 - The GNU C Library will now detect when /etc/resolv.conf has been
35 modified and reload the changed configuration. The new resolver option
36 “no-reload” (RES_NORELOAD) disables this behavior.
38 - The GNU C Library now supports an arbitrary number of search domains
39 (configured using the “search” directive in /etc/resolv.conf);
40 previously, there was a hard limit of six domains. For backward
41 compatibility, applications that directly modify the ‘_res’ global
42 object are still limited to six search domains.
44 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
45 Library will now randomly pick a name server from the configuration as a
46 starting point. (Previously, the second name server was always used.)
48 * The tunables feature is now enabled by default. This allows users to tweak
49 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
51 * New function reallocarray, which resizes an allocated block (like realloc)
52 to the product of two sizes, with a guaranteed clean failure upon integer
53 overflow in the multiplication. Originally from OpenBSD, contributed by
54 Dennis Wölfing and Rüdiger Sonderfeld.
56 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
57 These are extended versions of preadv and pwritev, respectively, taking an
58 additional flags argument. The set of supported flags depends on the
59 running kernel; full support currently requires kernel 4.7 or later.
61 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
62 create a new session ID for the spawned process. This feature is
63 scheduled to be added to the next major revision of POSIX; for the time
64 being, it is available under _GNU_SOURCE.
66 * errno.h is now safe to use from C-preprocessed assembly language on all
67 supported operating systems. In this context, it will only define the
68 Exxxx constants, as preprocessor macros expanding to integer literals.
70 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
71 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
72 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
73 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
75 To compile programs that use this feature, the compiler must support
76 128-bit floating point with the type name _Float128 (as defined by TS
77 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
78 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
79 must be defined to make the new interfaces visible.
81 The new functions and macros correspond to those present for other
82 floating-point types (except for a few obsolescent interfaces not
83 supported for the new type), with F128 or f128 suffixes; for example,
84 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
85 printf or scanf formats for the new type; the strfromf128 and strtof128
86 interfaces should be used instead.
88 Deprecated and removed features, and other changes affecting compatibility:
90 * The synchronization that pthread_spin_unlock performs has been changed to
91 now be equivalent to a C11 atomic store with release memory order to the
92 spin lock's memory location. Previously, several (but not all)
93 architectures used stronger synchronization (e.g., containing what is
94 often called a full barrier). This change can improve performance, but
95 may affect odd fringe uses of spin locks that depend on the previous
96 behavior (e.g., using spin locks as atomic variables to try to implement
97 Dekker's mutual exclusion algorithm).
99 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
102 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
103 will only be built and installed when the GNU C Library is configured with
104 --enable-obsolete-rpc. This allows alternative RPC implementations, such
105 as TIRPC or rpcsvc-proto, to be used.
107 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
108 libnss_compat, are deprecated, and will not be built or installed by
111 The NIS(+) support library, libnsl, is also deprecated. By default, a
112 compatibility shared library will be built and installed, but not headers
113 or development libraries. Only a few NIS-related programs require this
114 library. (In particular, the GNU C Library has never required programs
115 that use 'gethostbyname' to be linked with libnsl.)
117 Replacement implementations based on TIRPC, which additionally support
118 IPv6, are available from <https://github.com/thkukuk/>. The configure
119 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
120 name service modules, to be built and installed.
122 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
123 support is enabled, the configured recursive resolver must support EDNS.
124 (Responding to EDNS-enabled queries with responses which are not
125 EDNS-enabled is fine, but FORMERR responses are not.)
127 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
128 servers have not supported this opcode for a long time.
130 * The _res_opcodes variable has been removed from libresolv. It had been
131 exported by accident.
133 * <string.h> no longer includes inline versions of any string functions,
134 as this kind of optimization is better done by the compiler. The macros
135 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
137 * The nonstandard header <xlocale.h> has been removed. Most programs should
138 use <locale.h> instead. If you have a specific need for the definition of
139 locale_t with no other declarations, please contact
140 libc-alpha@sourceware.org and explain.
142 * The obsolete header <sys/ultrasound.h> has been removed.
144 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
146 * The obsolete function cfree has been removed. Applications should use
149 * The stack_t type no longer has the name struct sigaltstack. This changes
150 the C++ name mangling for interfaces involving this type.
152 * The ucontext_t type no longer has the name struct ucontext. This changes
153 the C++ name mangling for interfaces involving this type.
155 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
156 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
157 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
158 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
159 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
160 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
161 fpregset_t type no longer has the name struct fpu. This changes the C++
162 name mangling for interfaces involving those types.
164 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
165 synced with the kernel:
167 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
168 are not supported on this architecture and have been removed.
170 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
171 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
172 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
174 Programs that assume the GET/SETREGS ptrace requests are universally
175 available will now fail to build, instead of malfunctioning at runtime.
177 Changes to build and runtime requirements:
179 * Linux kernel 3.2 or later is required at runtime, on all architectures
180 supported by that kernel. (This is a change from version 2.25 only for
183 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
185 * On most architectures, GCC 4.9 or later is required to build the GNU C
186 Library. On powerpc64le, GCC 6.2 or later is required.
188 Older GCC versions and non-GNU compilers are still supported when
189 compiling programs that use the GNU C Library. (We do not know exactly
190 how old, and some GNU extensions to C may be _de facto_ required. If you
191 are interested in helping us make this statement less vague, please
192 contact libc-alpha@sourceware.org.)
194 Security related changes:
196 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
197 to avoid fragmentation-based spoofing attacks.
199 The following bugs are resolved with this release:
201 [The release manager will add the list generated by
202 scripts/list-fixed-bugs.py just before the release.]
207 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
208 24731-2:2010, is supported to enable declarations of functions from that
209 TR. Note that not all functions from that TR are supported by the GNU C
212 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
213 18661-1:2014, is supported to enable declarations of functions and macros
214 from that TS. Note that not all features from that TS are supported by
217 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
218 18661-4:2015, is supported to enable declarations of functions and macros
219 from that TS. Note that most features from that TS are not supported by
222 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
223 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
224 Since the GNU C Library defaults to a much newer revision of POSIX, this
225 will only affect programs that specifically request an old conformance
226 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
227 see a change in the visible declarations, but a program compiled with
228 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
231 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
232 defined by all multithreaded code, but glibc has not required this for
235 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
236 means that in a future release, the macros “major”, “minor”, and “makedev”
237 will only be available from <sys/sysmacros.h>.
239 These macros are not part of POSIX nor XSI, and their names frequently
240 collide with user code; see for instance glibc bug 19239 and Red Hat bug
241 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
242 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
245 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
246 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
247 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
249 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
250 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
251 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
252 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
253 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
254 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
255 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
256 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
257 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
258 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
259 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
262 * New <math.h> features are added from TS 18661-1:2014:
264 - Signaling NaN macros: SNANF, SNAN, SNANL.
266 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
267 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
268 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
270 - llogb functions: the llogb, llogbf and llogbl functions, and the
271 FP_LLOGB0 and FP_LLOGBNAN macros.
273 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
276 - Comparison macros: iseqsig.
278 - Classification macros: iscanonical, issubnormal, iszero.
280 - Total order functions: totalorder, totalorderf, totalorderl,
281 totalordermag, totalordermagf, totalordermagl.
283 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
285 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
286 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
288 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
289 are added to libc. They convert a floating-point number into string.
291 * Most of glibc can now be built with the stack smashing protector enabled.
292 It is recommended to build glibc with --enable-stack-protector=strong.
293 Implemented by Nick Alcock (Oracle).
295 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
296 intended to be used instead of memset() to erase sensitive data after use;
297 the compiler will not optimize out calls to explicit_bzero even if they
298 are "unnecessary" (in the sense that no _correct_ program can observe the
299 effects of the memory clear).
301 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
302 to float instead of double. This does not affect the ABI of any libraries
303 that are part of the GNU C Library, but may affect the ABI of other
304 libraries that use this type in their interfaces.
306 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
307 float_t and double_t types are now defined to long double instead of float
308 and double. These options are not the default, and this does not affect
309 the ABI of any libraries that are part of the GNU C Library, but it may
310 affect the ABI of other libraries that use this type in their interfaces,
311 if they are compiled or used with those options.
313 * The getentropy and getrandom functions, and the <sys/random.h> header file
316 * The buffer size for byte-oriented stdio streams is now limited to 8192
317 bytes by default. Previously, on Linux, the default buffer size on most
318 file systems was 4096 bytes (and thus remains unchanged), except on
319 network file systems, where the buffer size was unpredictable and could be
320 as large as several megabytes.
322 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
323 for the Linux quota interface which predates kernel version 2.4.22 has
326 * The malloc_get_state and malloc_set_state functions have been removed.
327 Already-existing binaries that dynamically link to these functions will
328 get a hidden implementation in which malloc_get_state is a stub. As far
329 as we know, these functions are used only by GNU Emacs and this change
330 will not adversely affect already-built Emacs executables. Any undumped
331 Emacs executables, which normally exist only during an Emacs build, should
332 be rebuilt by re-running “./configure; make” in the Emacs build tree.
334 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
335 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
336 “no-ip6-dotint” had already been the default, and support for the
337 “ip6-dotint” option was removed from the Internet in 2006.
339 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
340 flag from <resolv.h> have been removed. The option relied on a
341 backwards-incompatible DNS extension which was never deployed on the
344 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
345 RES_BLAST defined in the <resolv.h> header file have been deprecated.
346 They were already unimplemented.
348 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
349 _res.flags are deprecated. The flag was standardized in RFC 2133, but
350 removed again from the IETF name lookup interface specification in RFC
351 2553. Applications should use getaddrinfo instead.
353 * DNSSEC-related declarations and definitions have been removed from the
354 <arpa/nameser.h> header file, and libresolv will no longer attempt to
355 decode the data part of DNSSEC record types. Previous versions of glibc
356 only implemented minimal support for the previous version of DNSSEC, which
357 is incompatible with the currently deployed version.
359 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
360 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
361 <arpa/nameser.h> header file because the distinction between RR types and
362 meta-RR types is not officially standardized, subject to revision, and
363 thus not suitable for encoding in a macro.
365 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
366 and rhook members of the res_state type in <resolv.h> have been removed.
367 The glibc stub resolver did not support these hooks, but the header file
368 did not reflect that.
370 * For multi-arch support it is recommended to use a GCC which has
371 been built with support for GNU indirect functions. This ensures
372 that correct debugging information is generated for functions
373 selected by IFUNC resolvers. This support can either be enabled by
374 configuring GCC with '--enable-gnu-indirect-function', or by
375 enabling it by default by setting 'default_gnu_indirect_function'
376 variable for a particular architecture in the GCC source file
379 * GDB pretty printers have been added for mutex and condition variable
380 structures in POSIX Threads. When installed and loaded in gdb these pretty
381 printers show various pthread variables in human-readable form when read
382 using the 'print' or 'display' commands in gdb.
384 * Tunables feature added to allow tweaking of the runtime for an application
385 program. This feature can be enabled with the '--enable-tunables' configure
386 flag. The GNU C Library manual has details on usage and README.tunables has
387 instructions on adding new tunables to the library.
389 * A new version of condition variables functions have been implemented in
390 the NPTL implementation of POSIX Threads to provide stronger ordering
393 * A new version of pthread_rwlock functions have been implemented to use a more
394 scalable algorithm primarily through not using a critical section anymore to
397 Security related changes:
399 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
400 have been created with makecontext could fail to terminate due to a
401 missing .cantunwind annotation. This has been observed to lead to a hang
402 (denial of service) in some Go applications compiled with gccgo. Reported
403 by Andreas Schwab. (CVE-2016-6323)
405 * The DNS stub resolver functions would crash due to a NULL pointer
406 dereference when processing a query with a valid DNS question type which
407 was used internally in the implementation. The stub resolver now uses a
408 question type which is outside the range of valid question type values.
411 The following bugs are resolved with this release:
413 [4099] stdio: Overly agressive caching by stream i/o functions.
414 [7065] build: Support building glibc with -fstack-protector or -fstack-
416 [9842] localedata: en_CA: incorrect date format
417 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
418 before it started waiting
419 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
420 [15765] nptl: sem_open is wrongly a cancellation point
421 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
422 [16458] libc: endian.h and netinet/in.h byte order macros return values of
424 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
425 library linked with pthread
426 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
428 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
429 [17252] libc: getrandom and getentropy syscall
430 [17730] malloc: thread-local storage is sometimes improperly free()'d
431 after being __libc_memalign()'d
432 [18241] stdio: failed fseek on memstream does not set errno and can fail
434 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
436 [18463] nptl: pthread_cond_broadcast issue when surrounded by
437 PTHREAD_PRIO_INHERIT mutex on ARM
438 [18784] network: res_query and related function crash for special record
439 type queries (CVE-2015-5180)
440 [19380] math: strtod does not raise "inexact"
441 [19387] string: Integer overflow in memchr
442 [19390] string: Integer overflow in strncat
443 [19398] build: linknamespace tests fail with massively parallel build
444 [19402] nptl: Deadlock with robust shared mutex and asynchronous
446 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
447 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
448 [19514] libc: [PATCH] Fix spelling errors in spelling
449 "implement"/"implementation" in several places
450 [19582] network: Deprecate RES_USE_INET6
451 [19673] manual: clog10 docs appear to be erroneous
452 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
453 causes a segmentation fault
454 [19826] libc: invalid pointer returned from __tls_get_addr with static
456 [20016] network: resolv: Remove hooks support from the API
457 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
459 [20033] math: [x86_64] vectorized math function don't call the __finite
461 [20116] nptl: use after free in pthread_create
462 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
463 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
465 [20311] nptl: please install proc_service.h
466 [20366] libc: Compilation errors in installed headers in strict-compliance
468 [20370] malloc: malloc: Arena free list management is still racy
469 (incorrect fix in bug 19243)
470 [20386] libc: assert (X = 0) does not result in GCC warning
471 [20432] malloc: malloc: Minimize interface required for interposition
472 [20435] libc: Missing unwind info in __startcontext causes infinite loop
473 in _Unwind_Backtrace (CVE-2016-6323)
474 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
475 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
477 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
478 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
479 [20477] network: resolv: incorrect double-checked locking related to
481 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
483 [20495] math: x86_64 performance degradation due to AVX/SSE transition
485 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
486 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
487 cause transition penalty
488 [20517] math: sparcv9 missing fdiml compat symbol
489 [20524] manual: strverscmp is inconsistent
490 [20525] libc: <sys/quota.h> should be based on kernel headers
491 [20539] math: GCC 7 -static -lm fails to link at -O3
492 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
494 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
496 [20591] network: Remove obsolete DNSSEC support
497 [20592] network: DNS resource record type classification macros in
498 <arpa/nameser.h> are incorrect
499 [20593] network: Update DNS RR type definitions
500 [20611] network: getaddrinfo accepts invalid numeric scope IDs
501 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
503 [20629] network: libresolv: Remove support for bitstring labels
505 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
506 assertion: "maxidx >= 2"
507 [20660] math: [arm] Use VSQRT
508 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
509 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
510 [20689] libc: Test for FMA should also check for AVX.
511 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
512 [20715] math: iszero macro breaks existing code
513 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
514 [20728] libc: powerpc: Missing TOC stub in clone
515 [20729] build: glibc-2.24 fails to build for i486 with -Os
516 [20750] build: Build fails with default PIE enabled gcc-6.x
517 [20768] math: [alpha] sqrt fegetenv namespace
518 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
520 [20787] math: float_t is defined as float incorrectly on x86_64 even if
521 __FLT_EVAL_METHOD__ is 2
522 [20790] network: rpcgen buffer overrun in get_prog_declaration
523 [20822] nptl: race condition in __lll_unlock_elision on powerpc
524 [20829] libc: crypt snprintf namespace
525 [20847] libc: tst-vfork3 failure
526 [20855] math: Default bits/mathdef.h has inappropriate float_t
527 [20859] math: [sh4] FP_ILOGB0 invalid
528 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
529 U+20AC), not same as GBK
530 [20915] dynamic-link: global-dynamic TLS broken on aarch64
531 [20916] math: pow handling of sNaN arguments
532 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
533 [20919] math: Bad pow (qNaN, 0) result with -lieee
534 [20940] math: hypot sNaN handling
535 [20947] math: fmax, fmin sNaN handling
536 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
537 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
539 [20971] string: powerpc64/power7 memchr overflows internal pointer check
540 [20973] nptl: robust mutexes: Lost wake-ups
541 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
542 [20978] nis: strlen on null pointer in nss_nisplus
543 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
545 [21014] string: i686 memchr overflows internal pointer check
546 [21019] libc: [mips] n32 lseek incorrect on overflow
547 [21022] libc: [microblaze] __backtrace get_frame_size namespace
548 [21026] libc: [MIPS] readahead syscall is broken on n64
549 [21028] math: Fallback fesetexceptflag should always succeed
550 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
551 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
553 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
554 [21061] librt: [microblaze] librt lost clock_* exports
555 [21073] libc: tunables: insecure environment variables passed to
556 subprocesses with AT_SECURE
557 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
561 * The minimum Linux kernel version that this version of the GNU C Library
562 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
563 kernel version 2.6.32 or later suffices (on architectures that already
564 required kernel versions more recent than 3.2, those requirements remain
565 unchanged). Linux 3.2 or later kernel headers are required on all
568 * The pap_AN locale has been deleted. This has been deprecated for a long
569 time. It has been replaced by pap_AW & pap_CW, both of which have long
570 been included in previous releases.
572 * The readdir_r and readdir64_r functions have been deprecated. It is
573 recommended to use readdir and readdir64 instead.
575 * The type “union wait” has been removed. It was deprecated in the early
576 1990s and never part of POSIX. Application code should use the int type
577 instead of “union wait”.
579 * A new NSS action is added to facilitate large distributed system
580 administration. The action, MERGE, allows remote user stores like LDAP
581 to be merged into local user stores like /etc/groups in order to provide
582 easy to use, updated, and managed sets of merged credentials. The new
583 action can be used by configuring it in /etc/nsswitch.conf:
584 group: files [SUCCESS=merge] nis
585 Implemented by Stephen Gallagher (Red Hat).
587 * The deprecated __malloc_initialize_hook variable has been removed from the
590 * The long unused localedef --old-style option has been removed. It hasn't
591 done anything in over 16 years. Scripts using this option can safely
594 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
595 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
596 return the next representable value in the direction of positive infinity
597 and the nextdown functions return the next representable value in the
598 direction of negative infinity. These are currently enabled as GNU
601 Security related changes:
603 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
604 could result in a stack overflow when getnetbyname was called with an
605 overly long name. (CVE-2016-3075)
607 * Previously, getaddrinfo copied large amounts of address data to the stack,
608 even after the fix for CVE-2013-4458 has been applied, potentially
609 resulting in a stack overflow. getaddrinfo now uses a heap allocation
610 instead. Reported by Michael Petlan. (CVE-2016-3706)
612 * The glob function suffered from a stack-based buffer overflow when it was
613 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
614 Reported by Alexander Cherepanov. (CVE-2016-1234)
616 * The Sun RPC UDP client could exhaust all available stack space when
617 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
618 alloca plugin for GCC. (CVE-2016-4429)
620 * The IPv6 name server management code in libresolv could result in a memory
621 leak for each thread which is created, performs a failing naming lookup,
622 and exits. Over time, this could result in a denial of service due to
623 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
625 The following bugs are resolved with this release:
627 [1170] localedata: ne_NP: update Nepali locale definition file
628 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
630 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
631 [6796] math: fdim() does not set errno on overflow
632 [10354] libc: posix_spawn should use vfork() in more cases than presently
633 [11213] localedata: localedata: add copyright disclaimer to locale files
634 [12143] localedata: chr_US: new Cherokee locale
635 [12450] localedata: sgs_LT: new locale
636 [12676] localedata: ln_CD: new locale
637 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
639 [13304] math: fma, fmaf, fmal produce wrong results
640 [14259] build: --localedir arg to configure is ignored
641 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
642 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
643 [14934] localedata: es_CL: wrong first weekday chilean locale
644 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
646 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
648 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
649 [15368] nptl: raise() is not async-signal-safe
650 [15479] math: ceil, floor, round and trunc raise inexact exception
651 [15578] localedata: kk_KZ: various updates
652 [16003] localedata: pap_AN: punt old locale
653 [16137] localedata: iw_IL: punt old locale
654 [16190] localedata: eo: new esperanto locale
655 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
656 [16742] malloc: race condition: pthread_atfork() called before first
657 malloc() results in unexpected locking behaviour/deadlocks
658 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
660 [16983] localedata: postal_fmt does not allow %l and %n modifiers
661 [17565] localedata: pt_PT: wrong (work-)week start
662 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
663 [17950] build: Build fails with -msse
664 [18205] localedata: be_BY*: wrong first_weekday and first_workday
665 [18433] libc: posix_spawn does not return correctly upon failure to
667 [18453] localedata: charmaps/IBM875: incorrect codes
668 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
670 [18896] localedata: he_IL: improvements for currency
671 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
673 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
675 [19056] libc: Deprecate readdir_r
676 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
678 [19198] localedata: nl_NL: small improvements for Dutch locales
679 [19257] network: Per-thread memory leak in __res_vinit with IPv6
680 nameservers (CVE-2016-5417)
681 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
683 [19400] locale: Language missing in "iso-639.def", trivial fix in
685 [19431] malloc: Deadlock between fflush, getdelim, and fork
686 [19505] libc: Incorrect file descriptor validity checks in
687 posix_spawn_file_actions_add{open,close,dup2}
688 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
690 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
691 `intl/{gettextP,loadinfo}.h'
692 [19534] libc: execle, execlp may use malloc
693 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
694 [19573] network: res_nclose and __res_maybe_init disagree about name
695 server initialization, breaking Hesiod
696 [19575] localedata: Status of GB18030 tables
697 [19581] localedata: sr_* date_fmt string contains additional newline
698 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
700 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
701 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
702 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
703 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
705 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
707 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
709 [19610] dynamic-link: ldconfig -X removes stale symbolic links
710 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
711 [19633] locale: strfmon_l applies global locale to number formatting
712 [19642] network: Memory leak in getnameinfo
713 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
714 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
716 [19654] math: [x86_64] Need testcase for BZ #19590 fix
717 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
719 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
720 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
722 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
724 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
726 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
727 pointers and lengths in error-case.
728 [19727] locale: Converting from/to UTF-xx with iconv() does not always
729 report errors on UTF-16 surrogates values.
730 [19755] nscd: nscd assertion failure in gc
731 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
732 [19759] libc: mempcpy shouldn't be inlined
733 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
734 [19765] libc: s390 needs an optimized mempcpy
735 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
736 NAME_MAX limit assumption (CVE-2016-1234)
737 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
738 [19787] network: Missing and incorrect truncation checks in getnameinfo
739 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
741 [19791] network: Assertion failure in res_query.c with un-connectable name
743 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
744 [19822] math: libm.so install clobbers old version
745 [19825] network: resolv: send_vc can return uninitialized data in second
746 response to getaddrinfo
747 [19830] network: nss_dns: should check RDATA length against buffer length
748 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
749 confronted with A/AAAA records of invalid size
750 [19837] nss: nss_db: No retries for some long lines with a larger buffer
751 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
752 [19853] stdio: Printing IBM long double in decimal with high precision is
754 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
755 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
756 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
758 [19865] network: Assertion failure or memory leak in
759 _nss_dns_getcanonname_r
760 [19868] network: nss_dns: netent code does not skip over non-PTR records
761 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
763 [19881] string: Improve x86-64 memset
764 [19907] string: Incorrect memcpy tests
765 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
767 [19925] libc: termios.h XCASE namespace
768 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
769 [19929] libc: limits.h NL_NMAX namespace
770 [19931] stdio: Memory leak in vfprintf
771 [19957] libc: clone(CLONE_VM) access invalid parent memory
772 [19963] localedata: en_IL: New locale
773 [19989] stdio: stdio.h cuserid namespace
774 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
776 [19996] locale: langinfo.h nl_langinfo_l namespace
777 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
778 [20010] network: getaddrinfo: Stack overflow in hostent translation
780 [20012] stdio: libio: fmemopen append mode failure
781 [20014] stdio: stdio.h namespace for pre-threads POSIX
782 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
783 [20023] libc: fcntl.h timespec namespace
784 [20024] math: [x86_64] vectorized sincos trashes the stack
785 [20031] network: nss_hesiod: Heap overflow in get_txt_records
786 [20041] time: sys/time.h timespec namespace
787 [20043] libc: unistd.h missing cuserid for UNIX98 and before
788 [20044] libc: unistd.h missing pthread_atfork for UNIX98
789 [20051] libc: ttyslot in wrong header under wrong conditions
790 [20054] libc: gethostname not declared for XPG4
791 [20055] libc: termios.h missing tcgetsid for XPG4
792 [20072] dynamic-link: x86 init_cpu_features is called twice in static
794 [20073] libc: sys/stat.h fchmod namespace
795 [20074] libc: stdlib.h rand_r namespace
796 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
797 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
799 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
800 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
802 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
803 [20119] libc: Wrong mask for processors level type from CPUID
804 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
805 AS not supporting AVX512
806 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
808 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
809 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
811 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
812 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
813 [20178] libc: posix_spawn{p} should not call exit
814 [20191] stdio: libio: vtables hardening
815 [20195] string: FMA4 detection requires CPUID execution with register
817 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
818 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
820 [20212] math: acos (sNaN) returns sNaN
821 [20213] math: asin (sNaN) returns sNaN
822 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
823 [20218] math: [i386] asinhl (sNaN) returns sNaN
824 [20219] math: [i386] atanhl (sNaN) returns sNaN
825 [20222] stdio: fopencookie: Mangle function pointers
826 [20224] math: [i386] cbrtl (sNaN) returns sNaN
827 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
828 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
830 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
831 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
832 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
833 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
834 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
835 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
836 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
837 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
838 [20240] math: modf (sNaN) returns sNaN
839 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
840 [20250] math: frexp (sNaN) returns sNaN
841 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
842 [20255] math: [i386] fdim, fdimf return with excess range and precision /
844 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
845 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
846 subscript is below array bounds [-Werror=array-bounds]
847 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
848 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
850 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
851 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
853 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
855 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
856 [20316] localedata: id_ID: Februari instead of Pebruari
857 [20327] string: POWER8 strcasecmp returns incorrect result
858 [20347] math: Failure: Test: j0_downward (0xap+0)
859 [20348] libc: FAIL: misc/tst-preadvwritev64
860 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
861 [20350] libc: There is no test for p{read,write}64
862 [20357] math: Incorrect cos result for 1.5174239687223976
863 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
867 * Unicode 8.0.0 Support: Character encoding, character type info, and
868 transliteration tables are all updated to Unicode 8.0.0, using new
869 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
870 These updates cause user visible changes, such as the fixes for bugs
871 89, 16061, and 18568.
873 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
874 kernel-internal CPU set size. This means that requests that change the
875 CPU affinity which failed before (for example, an all-ones CPU mask) will
876 now succeed. Applications that need to determine the effective CPU
877 affinities need to call sched_getaffinity or pthread_getaffinity_np after
878 setting it because the kernel can adjust it (and the previous size check
879 would not detect this in the majority of cases).
881 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
882 the following new symbols are used: fts64_children, fts64_close,
883 fts64_open, fts64_read and fts64_set.
885 * getaddrinfo now detects certain invalid responses on an internal netlink
886 socket. If such responses are received, an affected process will
887 terminate with an error message of "Unexpected error <number> on netlink
888 descriptor <number>" or "Unexpected netlink response of size <number> on
889 descriptor <number>". The most likely cause for these errors is a
890 multi-threaded application which erroneously closes and reuses the netlink
891 file descriptor while it is used by getaddrinfo.
893 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
894 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
895 unnecessary serialization of memory allocation requests across threads.
896 The defect is now corrected. Users should see a substantial increase in
897 the concurent throughput of allocation requests for applications which
898 trigger this bug. Affected applications typically create and
899 destroy threads frequently. (Bug 19048 was reported and analyzed by
902 * There is now a --disable-timezone-tools configure option for disabling the
903 building and installing of the timezone related utilities (zic, zdump, and
904 tzselect). This is useful for people who build the timezone data and code
905 independent of the GNU C Library.
907 * The obsolete header <regexp.h> has been removed. Programs that require
908 this header must be updated to use <regex.h> instead.
910 * The obsolete functions bdflush, create_module, get_kernel_syms,
911 query_module and uselib are no longer available to newly linked binaries;
912 the header <sys/kdaemon.h> has been removed. These functions and header
913 were specific to systems using the Linux kernel and could not usefully be
914 used with the GNU C Library on systems with version 2.6 or later of the
917 * Optimized string, wcsmbs and memory functions for IBM z13.
918 Implemented by Stefan Liebler.
920 * Newly linked programs that define a variable called signgam will no longer
921 have it set by the lgamma, lgammaf and lgammal functions. Programs that
922 require signgam to be set by those functions must ensure that they use the
923 variable provided by the GNU C Library and declared in <math.h>, without
924 defining their own copy.
926 * The minimum GCC version that can be used to build this version of the GNU
927 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
928 still be used to compile programs using the GNU C Library.
930 Security related changes:
932 * An out-of-bounds value in a broken-out struct tm argument to strftime no
933 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
935 * The LD_POINTER_GUARD environment variable can no longer be used to disable
936 the pointer guard feature. It is always enabled. Previously,
937 LD_POINTER_GUARD could be used to disable security hardening in binaries
938 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
941 * An integer overflow in hcreate and hcreate_r could lead to an
942 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
944 * The catopen function no longer has unbounded stack usage. Reported by
947 * The nan, nanf and nanl functions no longer have unbounded stack usage
948 depending on the length of the string passed as an argument to the
949 functions. Reported by Joseph Myers. (CVE-2014-9761)
951 * A stack-based buffer overflow was found in libresolv when invoked from
952 libnss_dns, allowing specially crafted DNS responses to seize control
953 of execution flow in the DNS client. The buffer overflow occurs in
954 the functions send_dg (send datagram) and send_vc (send TCP) for the
955 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
956 family. The use of AF_UNSPEC triggers the low-level resolver code to
957 send out two parallel queries for A and AAAA. A mismanagement of the
958 buffers used for those queries could result in the response of a query
959 writing beyond the alloca allocated buffer created by
960 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
961 the overflow. Thanks to the Google Security Team and Red Hat for
962 reporting the security impact of this issue, and Robert Holiday of
963 Ciena for reporting the related bug 18665. (CVE-2015-7547)
965 The following bugs are resolved with this release:
967 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
968 [887] math: Math library function "logb" and "nextafter" inconsistent
969 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
970 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
971 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
972 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
974 [4404] localedata: German translation of "Alarm clock" is misleading
975 [6799] math: nextafter() and nexttoward() doen't set errno on
976 overflow/underflow errors
977 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
979 [10432] nis: _nss_nis_setnetgrent assertion failure
980 [11460] libc: fts has no LFS support
981 [12926] network: getaddrinfo()/make_request() may spin forever
982 [13065] nptl: Race condition in pthread barriers
983 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
984 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
986 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
987 [14912] libc: Rename non-installed bits/*.h headers
988 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
989 [15367] math: Let gcc use __builtin_isinf
990 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
991 [15421] math: lgamma wrongly sets signgam for ISO C
992 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
993 FE_INVALID with argument out of range
994 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
996 [15786] dynamic-link: ifunc resolver functions can smash function
998 [15918] math: Unnecessary check for equality in hypotf()
999 [16061] localedata: Review / update transliteration data
1000 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1002 [16141] time: strptime %z offset restriction
1003 [16171] math: drem should be alias of remainder
1004 [16296] math: fegetround is pure?
1005 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1006 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1008 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1009 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1011 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1012 "Exception "Inexact" set" on ppc32
1013 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1014 [16517] math: Missing underflow exception from tanf/tan/tanl
1015 [16519] math: Missing underflow exception from sinhf
1016 [16520] math: Missing underflow exception from tanhf
1017 [16521] math: Missing underflow exception from exp2
1018 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1020 [16734] stdio: fopen calls mmap to allocate its buffer
1021 [16961] math: nan function incorrect handling of bad sequences
1022 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1023 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1024 [16985] locale: localedef: confusing error message when opening output
1026 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1027 [17197] locale: Redundant shift character in iconv conversion output at
1029 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1030 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1031 [17250] dynamic-link: static linking breaks nss loading
1032 (getaddrinfo/getpwnam/etc...)
1033 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1035 [17441] math: isnan() should use __builtin_isnan() in GCC
1036 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1037 (related to lock elision)
1038 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1039 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1040 [17887] time: strptime should be able to parse "+01:00" style timezones
1041 [17905] libc: catopen() Multiple unbounded stack allocations
1043 [18084] libc: backtrace (..., 0) dumps core on x86
1044 [18086] libc: nice() sets errno to 0 on success
1045 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1046 is too large (CVE-2015-8778)
1047 [18251] dynamic-link: SONAME missing when audit modules provides path
1048 [18265] libc: add attributes for wchar string and memory functions
1049 [18370] math: csqrt missing underflows
1050 [18421] libc: [hppa] read-only segment has dynamic relocations
1051 [18472] libc: Obsolete syscall wrappers should be compat symbols
1052 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1053 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1054 [18525] localedata: Remove locale timezone information
1055 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1056 [18568] localedata: Update locale data to Unicode 8.0
1057 [18589] locale: sort-test.sh fails at random
1058 [18595] math: ctan, ctanh missing underflows
1059 [18604] libc: assert macro-expands its argument
1060 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1061 contains a vector instruction exception.
1062 [18611] math: j1, jn missing errno setting on underflow
1063 [18618] localedata: sync Chechen locale definitions with other *_RU
1065 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1066 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1067 [18665] network: In send_dg, the recvfrom function is NOT always using the
1068 buffer size of a newly created buffer (CVE-2015-7547)
1069 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1071 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1073 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1074 [18699] math: tilegx cproj() for various complex infinities does not yield
1076 [18724] libc: Harden put*ent functions against data injection
1077 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1078 [18755] build: build errors with -DNDEBUG
1079 [18757] stdio: fmemopen fails to set errno on failure
1080 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1082 [18781] libc: openat64 lacks O_LARGEFILE
1083 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1084 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1085 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1086 [18790] math: [ldbl-128ibm] tanhl inaccurate
1087 [18795] libc: stpncpy fortification misses buffer lengths that are
1088 statically too large
1089 [18796] build: build fails for --disable-mathvec
1090 [18803] math: hypot missing underflows
1091 [18820] stdio: fmemopen may leak memory on failure
1092 [18823] math: csqrt spurious underflows
1093 [18824] math: fma spurious underflows
1094 [18825] math: pow missing underflows
1095 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1096 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1097 [18870] build: sem_open.c fails to compile with missing symbol
1099 [18872] stdio: Fix memory leak in printf_positional
1100 [18873] libc: posix_fallocate overflow check ineffective
1101 [18875] math: Excess precision leads incorrect libm
1102 [18877] libc: arm: mmap offset regression
1103 [18887] libc: memory corruption when using getmntent on blank lines
1104 [18918] localedata: hu_HU: change time to HH:MM:SS format
1105 [18921] libc: Regression: extraneous stat() and fstat() performed by
1107 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1108 binaries (CVE-2015-8777)
1109 [18951] math: tgamma missing underflows
1110 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1112 [18953] localedata: lt_LT: change currency symbol to the euro
1113 [18956] math: powf inaccuracy
1114 [18961] math: [i386] exp missing underflows
1115 [18966] math: [i386] exp10 missing underflows
1116 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1117 [18969] build: multiple string test failures due to missing locale
1119 [18970] libc: Reference of pthread_setcancelstate in libc.a
1120 [18977] math: float / long double Bessel functions not in XSI POSIX
1121 [18980] math: i386 libm functions return with excess range and precision
1122 [18981] math: i386 scalb*, ldexp return with excess range and precision
1123 [18982] stdio: va_list and vprintf
1124 [18985] time: Passing out of range data to strftime() causes a segfault
1126 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1127 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1128 [19012] locale: iconv_open leaks memory on error path
1129 [19016] math: clog, clog10 inaccuracy
1130 [19018] nptl: Mangle function pointers in tls_dtor_list
1131 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1132 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1133 [19048] malloc: malloc: arena free list can become cyclic, increasing
1135 [19049] math: [powerpc] erfc incorrect zero sign
1136 [19050] math: [powerpc] log* incorrect zero sign
1137 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1138 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1139 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1140 [19074] network: Data race in _res_hconf_reorder_addrs
1141 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1142 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1143 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1145 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1147 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1148 [19086] manual: posix_fallocate64 documented argument order is wrong.
1149 [19088] math: lround, llround missing exceptions close to overflow
1151 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1152 [19095] math: dbl-64 lrint incorrect for 64-bit long
1153 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1154 [19124] dynamic-link: ld.so failed to build with older assmebler
1155 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1156 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1157 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1158 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1159 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1160 pthread_setaffinity_np
1161 [19156] math: [ldbl-128] j0l spurious underflows
1162 [19164] nptl: tst-getcpu fails with many possible CPUs
1163 [19168] math: math/test-ildoubl and math/test-ldouble failure
1164 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1165 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1167 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1168 (FE_NOMASK_ENV) do not clear SSE exceptions
1169 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1170 _int_new_arena/reused_arena
1171 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1172 [19201] math: dbl-64 remainder incorrect sign of zero result
1173 [19205] math: bits/math-finite.h conditions do not match math.h and
1175 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1176 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1177 for C99-based standards
1178 [19212] libc: features.h not -Wundef clean
1179 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1181 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1182 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1183 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1185 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1186 "inexact" exceptions
1187 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1189 [19242] libc: strtol incorrect in Turkish locales
1190 [19243] malloc: reused_arena can pick an arena on the free list, leading
1191 to an assertion failure and reference count corruption
1192 [19253] time: tzset() ineffective when temporary TZ did not include DST
1194 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1195 [19270] math: [hppa] Shared libm missing __isnanl
1196 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1197 MAP_HUGETLB and MAP_STACK defines
1198 [19313] nptl: Wrong __cpu_mask for x32
1199 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1200 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1201 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1202 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1203 [19363] time: x32: times() return value wrongly truncates/sign extends
1205 [19367] dynamic-link: Improve branch prediction on Silvermont
1206 [19369] network: Default domain name not reset by res_ninit when "search"
1207 / "domain" entry is removed from resolv.conf
1208 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1210 [19385] localedata: bg_BG: time separator should be colon, not comma
1211 [19408] libc: linux personality syscall wrapper may erroneously return an
1212 error on 32-bit architectures
1213 [19415] libc: dladdr returns wrong names on hppa
1214 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1215 IBM905, IBM907, and IBM909
1216 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1217 [19443] build: build failures with -DDEBUG
1218 [19451] build: Make check fails on test-double-vlen2
1219 [19462] libc: Glibc failed to build with -Os
1220 [19465] math: Wrong code with -Os
1221 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1223 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1225 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1226 [19486] math: S390: Math tests fail with "Exception Inexact set".
1227 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1228 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1230 [19590] math: Fail to build shared objects that use libmvec.so functions.
1234 * The following bugs are resolved with this release:
1236 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1237 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1238 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1239 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1240 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1241 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1242 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1243 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1244 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1245 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1246 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1247 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1248 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1249 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1250 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1251 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1252 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1253 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1254 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1255 18657, 18676, 18694, 18696, 18887.
1257 * Cache information can be queried via sysconf() function on s390 e.g. with
1258 _SC_LEVEL1_ICACHE_SIZE as argument.
1260 * A buffer overflow in gethostbyname_r and related functions performing DNS
1261 requests has been fixed. If the NSS functions were called with a
1262 misaligned buffer, the buffer length change due to pointer alignment was
1263 not taken into account. This could result in application crashes or,
1264 potentially arbitrary code execution, using crafted, but syntactically
1265 valid DNS responses. (CVE-2015-1781)
1267 * The time zone file parser has been made more robust against crafted time
1268 zone files, avoiding heap buffer overflows related to the processing of
1269 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1270 large time zone data files. Overly long time zone specifiers in the TZ
1271 variable no longer result in stack overflows and crashes.
1273 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1274 for LD and GD on x86 and x86-64, has been implemented. You will need
1275 binutils-2.24 or later to enable this optimization.
1277 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1278 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1279 Hat). These updates cause user visible changes, such as the fix for bug
1282 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1283 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1284 condition in some applications.
1286 * Added vector math library named libmvec with the following vectorized x86_64
1287 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1289 The library can be disabled with --disable-mathvec. Use of the functions is
1290 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1291 Shared library libmvec.so is linked in as needed when using -lm (no need to
1292 specify -lmvec explicitly for not static builds).
1293 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1295 * A new fmemopen implementation has been added with the goal of POSIX
1296 compliance. The new implementation fixes the following long-standing
1297 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1298 old implementation is still present for use be by existing binaries.
1300 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1301 releases. It has been fixed to match 2.19 and older, but binaries built
1302 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1304 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1305 Contributed by Roland McGrath (Google).
1307 * The header <regexp.h> is deprecated, and will be removed in a future
1308 release. Use of this header will trigger a deprecation warning.
1309 Application developers should update their code to use <regex.h> instead.
1311 This header was formerly part of SUS, but was deprecated in 1994 and
1312 removed from the standard in 2001. Also, the glibc implementation
1313 leaks memory. See BZ#18681 for more details.
1317 * The following bugs are resolved with this release:
1319 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1320 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1321 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1322 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1323 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1324 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1325 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1326 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1327 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1328 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1331 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1332 too little memory for the to-be-scanned arguments and overflow the
1333 allocated buffer. The implementation now correctly computes the required
1334 buffer size when using malloc, and switches to malloc from alloca as
1337 * A new semaphore algorithm has been implemented in generic C code for all
1338 machines. Previous custom assembly implementations of semaphore were
1339 difficult to reason about or ensure that they were safe. The new version
1340 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1341 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1342 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1344 * Port to Altera Nios II has been contributed by Mentor Graphics.
1346 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1347 implementations for powerpc64/powerpc64le.
1348 Implemented by Adhemerval Zanella (IBM).
1350 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1351 and powerpc64le. This may improve lock scaling of existing programs on
1352 HTM capable systems. The lock elision code is only enabled with
1353 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1354 powerpc will issue a transaction abort on every syscall to avoid side
1355 effects being visible outside transactions.
1357 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1358 AArch64. Contributed by ARM Ltd.
1360 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1362 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1363 under certain input conditions resulting in the execution of a shell for
1364 command substitution when the application did not request it. The
1365 implementation now checks WRDE_NOCMD immediately before executing the
1366 shell and returns the error WRDE_CMDSUB as expected.
1368 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1369 processing format strings with a large number of format specifiers.
1371 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1372 infinite loop if the DNS response contained a PTR record of an unexpected
1375 * The minimum GCC version that can be used to build this version of the GNU
1376 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1377 still be used to compile programs using the GNU C Library.
1379 * The GNU C Library is now built with -Werror by default. This can be
1380 disabled by configuring with --disable-werror.
1382 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1384 * The obsolete sigvec function has been removed. This was the original
1385 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1386 programs have been using instead for about 25 years. Of course, ABI
1387 compatibility for old binaries using sigvec remains intact.
1389 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1390 with newer versions of bison.
1392 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1393 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1394 registers overlay two consecutive single-precision registers. MIPS32R2
1395 introduced a new FPU mode (FR=1) where double-precision registers extend the
1396 corresponding single-precision registers which is incompatible with the
1397 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1398 require the use of FR=1 making a transition necessary. New o32 ABI
1399 extensions enable users to migrate over time from the original o32 ABI
1400 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1401 tracks the ABI of any loaded object and verifies that new objects are
1402 compatible. Mode transitions will also be requested as required and
1403 unsupportable objects will be rejected. The ABI checks include both soft and
1404 hard float ABIs for o32, n32 and n64.
1406 GCC 5 with GNU binutils 2.25 onwards:
1407 It is strongly recommended that all o32 system libraries are built using the
1408 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1409 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1410 compiler using --with-fp-32=xx to set this by default.
1414 * The following bugs are resolved with this release:
1416 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1417 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1418 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1419 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1420 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1421 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1422 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1423 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1424 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1425 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1426 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1427 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1428 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1429 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1430 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1432 * Reverted change of ABI data structures for s390 and s390x:
1433 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1434 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1435 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1436 are preserved pointing straight to the same implementation as the old ones.
1437 Given that, new callers will simply provide a too-big buffer to these
1438 functions. Any applications/libraries out there that embed jmp_buf or
1439 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1440 against 2.19 headers will have to rebuilt again. This is necessary in any
1441 case to revert the breakage in their ABI caused by the glibc change.
1443 * Support for file description locks is added to systems running the
1444 Linux kernel. The standard file locking interfaces are extended to
1445 operate on file descriptions, not file descriptors, via the use of
1446 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1447 are associated with an open file instead of a process.
1449 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1451 * The minimum Linux kernel version that this version of the GNU C Library
1452 can be used with is 2.6.32.
1454 * Running the testsuite no longer terminates as soon as a test fails.
1455 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1456 with PASS or FAIL lines for individual tests. A summary of the results is
1457 printed, including a list of failing lists, and "make check" exits with
1458 error status if there were any unexpected failures. "make check
1459 stop-on-test-failure=y" may be used to keep the old behavior.
1461 * The am33 port, which had not worked for several years, has been removed
1464 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1465 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1466 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1467 conflicted with POSIX (support for which was removed in 2.19), the
1468 interfaces those macros enabled remain available when compiling with
1469 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1470 test macros defined.
1472 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1474 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1475 This may improve lock scaling of existing programs on TX capable systems.
1476 The lock elision code is only built with --enable-lock-elision=yes and
1477 then requires a GCC version supporting the TX builtins. With lock elision
1478 default mutexes are elided via __builtin_tbegin, if the cpu supports
1479 transactions. By default lock elision is not enabled and the elision code
1482 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1483 copy the path argument. This allowed programs to cause posix_spawn to
1484 deference a dangling pointer, or use an unexpected pathname argument if
1485 the string was modified after the posix_spawn_file_actions_addopen
1488 * All supported architectures now use the main glibc sysdeps directory
1489 instead of some being in a separate "ports" directory (which was
1490 distributed separately before glibc 2.17).
1492 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1493 On configurations that support it (all Linux configurations), it's now
1494 used regardless of the --enable-add-ons switch to configure. It is no
1495 longer possible to build such configurations without pthreads support.
1497 * Locale names, including those obtained from environment variables (LANG
1498 and the LC_* variables), are more tightly checked for proper syntax.
1499 setlocale will now fail (with EINVAL) for locale names that are overly
1500 long, contain slashes without starting with a slash, or contain ".." path
1501 components. (CVE-2014-0475) Previously, some valid locale names were
1502 silently replaced with the "C" locale when running in AT_SECURE mode
1503 (e.g., in a SUID program). This is no longer necessary because of the
1506 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1507 with application code using Intel MPX instructions. (With all previous
1508 versions, the MPX register state could be clobbered when making calls
1509 into or out of a shared library.) Note that while the new dynamic
1510 linker is compatible with all known x86 hardware whether or not it
1511 supports Intel MPX, some x86 instruction-set emulators might fail to
1512 handle the new instruction encodings. This is known to affect Valgrind
1513 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1514 release), and might affect other tools that do instruction emulation.
1516 * Support for loadable gconv transliteration modules has been removed.
1517 The support for transliteration modules has been non-functional for
1518 over a decade, and the removal is prompted by security defects. The
1519 normal gconv conversion modules are still supported. Transliteration
1520 with //TRANSLIT is still possible, and the //IGNORE specifier
1521 continues to be supported. (CVE-2014-5119)
1523 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1524 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1525 resulting a denial-of-service security vulnerability in applications which
1526 use functions related to iconv. (CVE-2014-6040)
1530 * The following bugs are resolved with this release:
1532 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1533 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1534 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1535 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1536 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1537 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1538 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
1539 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
1540 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
1541 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
1542 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
1543 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
1544 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
1545 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
1546 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
1547 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
1548 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
1549 16431, 16453, 16474, 16506, 16510, 16529
1551 * Slovenian translations for glibc messages have been contributed by the
1552 Translation Project's Slovenian team of translators.
1554 * The public headers no longer use __unused nor __block. This change is to
1555 support compiling programs that are derived from BSD sources and use
1556 __unused internally, and to support compiling with Clang's -fblock
1557 extension which uses __block.
1559 * CVE-2012-4412 The strcoll implementation caches indices and rules for
1560 large collation sequences to optimize multiple passes. This cache
1561 computation may overflow for large collation sequences and may cause a
1562 stack or buffer overflow. This is now fixed to use a slower algorithm
1563 which does not use a cache if there is an integer overflow.
1565 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
1566 rules for large collation sequences to optimize multiple passes and falls
1567 back to alloca if malloc fails, resulting in a possible stack overflow.
1568 The implementation now falls back to an uncached collation sequence lookup
1571 * CVE-2013-4788 The pointer guard used for pointer mangling was not
1572 initialized for static applications resulting in the security feature
1573 being disabled. The pointer guard is now correctly initialized to a
1574 random value for static applications. Existing static applications need
1575 to be recompiled to take advantage of the fix (bug 15754).
1577 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
1578 to the d_name member of struct dirent, or omit the terminating NUL
1579 character. (Bugzilla #14699).
1581 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
1582 aligned_alloc functions could allocate too few bytes or corrupt the
1583 heap when passed very large allocation size values (Bugzilla #15855,
1586 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
1587 for AF_INET6 has been fixed (Bugzilla #16072).
1589 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
1590 pap_CW, quz_PE, the_NP.
1592 * Substantially revised locales: gd_GB, ht_HT
1594 * The LC_ADDRESS field was updated to support country_car for almost all
1597 * ISO 1427 definitions were updated.
1599 * ISO 3166 definitions were updated.
1601 * The localedef utility now supports --big-endian and --little-endian
1602 command-line options to generate locales for a different system from that
1603 for which the C library was built.
1605 * Binary locale files now only depend on the endianness of the system for
1606 which they are generated and not on other properties of that system. As a
1607 consequence, binary files generated with new localedef may be incompatible
1608 with old versions of the GNU C Library, and binary files generated with
1609 old localedef may be incompatible with this version of the GNU C Library,
1610 in the following circumstances:
1612 + Locale files may be incompatible on m68k systems.
1614 + Locale archive files (but not separate files for individual locales) may
1615 be incompatible on systems where plain "char" is signed.
1617 * The configure option --disable-versioning has been removed. Builds with
1618 --disable-versioning had not worked for several years.
1620 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
1621 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
1623 * SystemTap probes for malloc have been introduced.
1625 * SystemTap probes for slow multiple precision fallback paths of
1626 transcendental functions have been introduced.
1628 * Support for powerpc64le has been added.
1630 * The soft-float powerpc port now supports e500 processors.
1632 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
1634 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
1635 set of header declarations that are enabled by default, even when other
1636 feature test macros or compiler options such as -std=c99 would otherwise
1637 disable some of those declarations.
1639 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
1640 conflict with POSIX. The libbsd-compat library (which was a dummy library
1641 that did nothing) has also been removed.
1643 * Preliminary documentation about Multi-Thread, Async-Signal and
1644 Async-Cancel Safety has been added.
1646 * Change of ABI data structures for s390 and s390x:
1647 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
1648 allow for future hardware extensions. All library functions that accept or
1649 return these structures were versioned in order to provide backwards
1650 compatibility with existing code. However, not all cases can be handled
1651 with interface versioning. If an application embeds either structure into
1652 another structure and that structure is passed to another compilation unit
1653 which expects the newer larger-sized object then any access to the new
1654 fields will result in undefined behaviour. Similarly any access to fields
1655 that were moved by the enlarging of the structures will cause undefined
1656 behaviour. To fix the undefined behaviour all objects that are part of the
1657 final in-memory image must agree on the size of structures, and this may
1658 require recompilation.
1662 * The following bugs are resolved with this release:
1664 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
1665 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
1666 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
1667 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
1668 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
1669 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
1670 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
1671 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
1672 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
1673 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
1674 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
1675 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
1676 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
1679 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
1680 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
1681 Distributions can re-enable building and using pt_chown via the new configure
1682 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
1683 considerable security risks and should only be used if the distribution
1684 understands and accepts the risks.
1686 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
1689 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
1690 fixed (Bugzilla #15330).
1692 * Add support for calling C++11 thread_local object destructors on thread
1693 and program exit. This needs compiler support for offloading C++11
1694 destructor calls to glibc.
1696 * Improved worst case performance of libm functions with double inputs and
1699 * Support for priority inherited mutexes in pthread condition variables on
1700 non-x86 architectures.
1702 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
1704 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
1706 * Optimized string functions for ARM. Implemented by Will Newton and
1709 * Optimized libm functions for SPARC. Implemented by David S. Miller.
1711 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
1714 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
1715 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
1717 * Added a benchmark framework to track performance of functions in glibc.
1719 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
1720 It is based on draft TS 18661 and currently enabled as a GNU extension.
1722 * On Linux, the clock function now uses the clock_gettime system call
1723 for improved precision, rather than old times system call.
1725 * Added support for version-3 tz format files. This is needed when using
1726 the tz database release 2013e or later, and affects a few unusual cases --
1727 currently only TZ='America/Godthab' for time stamps after 2037.
1729 * Added new API functions pthread_getattr_default_np and
1730 pthread_setattr_default_np to get and set the default pthread
1731 attributes of a process.
1733 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
1734 This may improve lock scaling of existing programs on TSX capable systems.
1735 When the --enable-lock-elision=yes parameter is specified at configure
1736 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
1739 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
1740 availability). Implemented by Ryan S. Arnold.
1742 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
1746 * The following bugs are resolved with this release:
1748 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
1749 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
1750 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
1751 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
1752 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
1753 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
1754 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
1755 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
1756 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
1757 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
1758 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
1759 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
1760 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
1761 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
1762 14879, 14889, 14893, 14898, 14914.
1764 * Optimization of memcpy for MIPS.
1766 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
1767 EMFILE has been fixed (Bugzilla #14889).
1769 * The add-on ports collection is now distributed in the "ports" subdirectory
1770 of the main GNU C Library distribution, rather than separately.
1772 * Port to ARM AArch64 contributed by Linaro.
1774 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
1775 Optimized versions of memcpy, memset, and memcmp added for System z10 and
1777 Implemented by Andreas Krebbel.
1779 * The new function secure_getenv allows secure access to the environment,
1780 returning NULL if running in a SUID/SGID process. This function replaces
1781 the internal function __secure_getenv.
1783 * SystemTap static probes have been added into the dynamic linker.
1784 Implemented by Gary Benson.
1786 * Optimizations of string functions strstr, strcasestr and memmem.
1787 Implemented by Maxim Kuvyrkov.
1789 * The minimum Linux kernel version that this version of the GNU C Library
1790 can be used with is 2.6.16.
1792 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
1793 powerpc POWER7. Implemented by Will Schmidt.
1795 * New configure option --disable-nscd builds the C library such that it
1796 never attempts to contact the Name Service Caching Daemon (nscd).
1797 New configure option --disable-build-nscd avoids building nscd itself;
1798 this is the default if --disable-nscd is used.
1800 * Improved support for cross-compilation, including cross-testing and
1801 bootstrap builds without a previously built glibc.
1803 * Several testsuite tests are now able to test multiple IFUNC variants of an
1804 interface, rather than just testing the one that would be chooen by
1807 * New configure options --with-bugurl and --with-pkgversion, for
1808 distributors to use to embed their bug-reporting and package version
1809 information in --help and --version output.
1811 * The ttyname and ttyname_r functions on Linux now fall back to searching for
1812 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
1813 allows creation of chroots without the procfs mounted on /proc.
1815 * The `crypt' function now fails if passed salt bytes that violate the
1816 specification for those values. On Linux, the `crypt' function will
1817 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
1818 enabled, and fail on encrypted strings using the MD5 or DES algorithm
1819 when the mode is enabled.
1821 * The `clock_*' suite of functions (declared in <time.h>) is now available
1822 directly in the main C library. Previously it was necessary to link with
1823 -lrt to use these functions. This change has the effect that a
1824 single-threaded program that uses a function such as `clock_gettime' (and
1825 is not linked with -lrt) will no longer implicitly load the pthreads
1826 library at runtime and so will not suffer the overheads associated with
1827 multi-thread support in other code such as the C++ runtime library.
1829 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
1834 * The following bugs are resolved with this release:
1836 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
1837 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
1838 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
1839 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
1840 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
1841 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
1842 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
1843 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
1844 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
1845 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
1846 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
1847 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
1848 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
1849 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
1850 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
1851 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
1852 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
1853 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
1854 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
1855 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
1856 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
1857 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
1860 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
1861 configuring glibc with:
1862 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
1863 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
1864 Implemented by H.J. Lu.
1868 + define static_assert
1870 + do not declare gets
1872 + declare at_quick_exit and quick_exit also for ISO C11
1874 + aligned_alloc. NB: The code is deliberately allows the size parameter
1875 to not be a multiple of the alignment. This is a moronic requirement
1876 in the standard but it is only a requirement on the caller, not the
1879 + timespec_get added
1881 + uchar.h support added
1883 + CMPLX, CMPLXF, CMPLXL added
1885 Implemented by Ulrich Drepper.
1887 * Support for the IA-64 has been moved to ports.
1889 * Remove support for anything but ELF binary format
1891 * Checking versions of poll, ppoll added.
1892 Implemented by Ulrich Drepper.
1894 * More generic and 64-bit performance optimizations to math functions.
1895 Implemented by Ulrich Drepper.
1897 * New configure option --enable-obsolete-rpc makes the deprecated RPC
1898 headers and functions available at compile time as they were before
1899 version 2.14. This option will be removed at some time in the future
1900 after the TI-RPC library becomes fully sufficient for the needs of
1901 existing applications.
1903 * Compatibility code for Linux kernel versions before 2.4 has been removed.
1904 Note that glibc is not expected to work with any Linux kernel version
1907 * New header <sys/auxv.h> and function getauxval allowing easy access to
1908 the AT_* key-value pairs passed from the Linux kernel. The header also
1909 defines the HWCAP_* bits associated with the AT_HWCAP key.
1911 * New locales: mag_IN
1913 * New configure option --enable-systemtap builds SystemTap static probes
1914 into libc for setjmp and longjmp and into libpthread for various operations.
1915 So far the setjmp/longjmp probes and some of the libpthread probes are
1916 provided only for i*86 and x86_64.
1917 Implemented by Roland McGrath and Rayson Ho.
1919 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
1921 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
1924 * More optimized functions for SPARC. Implemented by David S. Miller.
1926 * Improved support for cross-compilation, especially bootstrap builds
1927 without a previously built glibc.
1929 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
1930 Chris Metcalf from Tilera.
1932 * Support for the old ARM ABI has been removed from ports. Only the EABI is
1933 now supported for ARM processors.
1935 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
1936 as the name of the dynamic linker, to distinguish it from the
1937 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
1939 * Support for CRIS has been removed from ports.
1941 * A new class of installed header has been documented for low-level
1942 platform-specific functionality. PowerPC added the first instance with a
1943 function to provide time base register access. Contributed by Tulio
1944 Magno Quites Machado Filho.
1946 * ix86 configurations now install header files that are consistent with
1947 what x86-64 configurations install. These same header files can be used
1948 for -m32, -m64, or -mx32 builds.
1949 Contributed by H.J. Lu.
1951 * Math library bug fixes. A thorough audit of all open math library bugs was
1952 conducted by Joseph Myers. Significant progress was made on many math
1953 library bugs resulting in more accurate exceptions and function results.
1954 Many thanks to all those that contributed including Andreas Jaeger for his
1955 patch review and work on the x87 trigonometric instruction issues.
1957 * Timezone data is no longer installed. Timezone-related binaries and scripts
1958 will continue to be installed. Users should obtain their timezone data from
1959 their distribution provider or from the tzdata package at
1960 <ftp://munnari.oz.au/pub/>.
1964 * The following bugs are resolved with this release:
1966 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
1967 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
1968 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
1969 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
1970 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
1971 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
1972 13484, 13506, 13515, 13523, 13524, 13538, 13540
1974 * New program pldd to list loaded object of a process
1975 Implemented by Ulrich Drepper.
1977 * Add nss_db support back to glibc. No more dependency on Berkeley db
1978 and support for initgroups lookups.
1979 Implemented by Ulrich Drepper.
1981 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
1982 Contributed by HJ Lu.
1984 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
1985 Contributed by HJ Lu.
1987 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
1988 on x86-32 and x86-64.
1989 Contributed by Liubov Dmitrieva.
1991 * Optimized strchr and strrchr for SSE on x86-32.
1992 Contributed by Liubov Dmitrieva.
1994 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
1995 for x86-64 and x86-32.
1996 Contributed by Liubov Dmitrieva.
1998 * New interfaces: scandirat, scandirat64
1999 Implemented by Ulrich Drepper.
2001 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2002 Implemented by Ulrich Drepper.
2004 * nscd now also caches the netgroup database.
2005 Implemented by Ulrich Drepper.
2007 * Integrate libm with gcc's -ffinite-math-only option.
2008 Implemented by Ulrich Drepper.
2010 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2011 to math functions. Implemented by Ulrich Drepper.
2013 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2014 Implemented by Ulrich Drepper.
2016 * New Linux interfaces: process_vm_readv, process_vm_writev
2018 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2019 Implemented by Ulrich Drepper.
2021 * Optimized nearbyint and strcasecmp for PPC.
2022 Implemented by Adhemerval Zanella.
2024 * New locales: bho_IN, unm_US, es_CU, ta_LK
2028 * The following bugs are resolved with this release:
2030 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2031 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2032 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2033 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2034 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2035 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2036 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2037 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2038 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2039 12795, 12811, 12813, 12814, 12841
2041 * The RPC implementation in libc is obsoleted. Old programs keep working
2042 but new programs cannot be linked with the routines in libc anymore.
2043 Programs in need of RPC functionality must be linked against TI-RPC.
2044 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2046 Visible changes of this change include (obviously) the inability to link
2047 programs using RPC functions without referencing the TI-RPC library and the
2048 removal of the RPC headers from the glibc headers.
2049 Implemented by Ulrich Drepper.
2051 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2052 syncfs, setns, sendmmsg
2054 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2055 yue_HK, lij_IT, mhr_RU
2057 * New encodings: CP770, CP771, CP772, CP773, CP774
2059 * New program sotruss to trace calls through PLTs
2060 Implemented by Ulrich Drepper.
2062 * The malloc hook implementation is marked deprecated and will be removed
2063 from the default implementation in the next version. The design never
2064 worked ever since the introduction of threads. Even programs which do
2065 not create threads themselves can use multiple threads created internally.
2069 * The following bugs are resolved with this release:
2071 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2072 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2073 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2074 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2077 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2079 * POWER7 optimizations: memset, memcmp, strncmp
2081 * New optimized string functions for x86-64: strnlen (SSE2),
2082 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2083 Implemented by Ulrich Drepper.
2087 * The following bugs are resolved with this release:
2089 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2090 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2091 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2092 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2093 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2094 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2095 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2096 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2098 * New interfaces: pthread_getname_np, pthread_setname_np
2100 * New Linux interface: recvmmsg
2102 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2104 * The dynamic linker now recognizes supported ABI versions from the
2105 EI_ABIVERSION field in the ELF header.
2106 Implemented by Ulrich Drepper.
2108 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2109 will not be used to fill in password fields in the passwd.byname replies.
2110 Instead it is used to synthesize the shadow.byname table, should it be
2111 missing. This is a useful mode in some installations involving Solaris.
2112 Implemented by Ulrich Drepper.
2114 * New locales: kok_IN, sq_MK, cv_RU
2118 * The following bugs are resolved with this release:
2120 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2121 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2122 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2123 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2124 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2125 10789, 10817, 10823, 10840, 10847
2127 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2129 Implemented by Ulrich Drepper.
2131 * Checking version of longjmp added that fails if an uninitialized stack
2132 frame would be created. Implemented by Ulrich Drepper.
2134 * STT_GNU_IFUNC is now supported in static executables.
2135 Implemented by H.J. Lu.
2137 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2139 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2140 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2141 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2142 Contributed by H.J. Lu.
2144 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2145 Implemented by Ulrich Drepper.
2147 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2149 Contributed by H.J. Lu.
2151 * Support for fma instruction in AVX on x86-64.
2152 Implemented by H.J. Lu and Ulrich Drepper.
2154 * AVX support in x86-64 auditing support in ld.so.
2155 Implemented by H.J. Lu.
2157 * STB_GNU_UNIQUE support added.
2158 Implemented by Ulrich Drepper.
2160 * Implement second fallback mode for optimized DNS lookup for even more
2161 broken environments. If two requests from the same port are not
2162 handled correctly close the socket and open a new one before sending
2163 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2164 can be used to select this mode right away, instead of rediscovering the
2165 necessity is every process again.
2166 Implemented by Ulrich Drepper.
2168 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2169 Implemented by Adam Tkac.
2171 * Optimized iconv conversions for S390x.
2172 Implemented by Andreas Krebbel.
2174 * Using condvars with PI mutexes is now more efficient due to kernel
2175 support for requeueing to PI futexes. NPTL support added for x86-64.
2176 Implemented by Ulrich Drepper.
2178 * New locales: ps_AF, my_MM
2182 * The following bugs are resolved with this release:
2184 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2185 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2186 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2187 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2188 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2190 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2191 Implemented by Ulrich Drepper.
2193 * New Linux interfaces: accept4, fallocate, fallocate64.
2194 Implemented by Ulrich Drepper.
2196 * Correct declarations of string function when used in C++ code. This
2197 could lead to compile errors for invalid C++ code.
2199 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2202 * New POSIX 2008 interface: psiginfo
2203 Implemented by Ulrich Drepper.
2205 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2206 Implemented by Ulrich Drepper.
2208 * Support for selecting between multiple function definitions at runtime
2209 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2211 * The libcrypt library can now use the hash function implementations in
2212 NSS. Implemented by Ulrich Drepper.
2214 * The malloc implementation can be compiled to be less memory efficient
2215 but higher performing in multi-threaded programs.
2216 Implemented by Ulrich Drepper.
2218 * New locales: nan_TW@latin, ks_IN
2220 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2221 Implemented by Ulrich Drepper.
2223 * Extended printf hook support. It is possible to use user-defined types
2224 and extend existing format specifiers.
2225 Implemented by Ulrich Drepper.
2227 * Handling for group shadow files has been added.
2228 Implemented by Ulrich Drepper.
2230 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2231 servers (the case, e.g., for some people using the built-in DNS
2232 server in ADSL modems/routers). There is a once-per-process timeout
2233 in case of a broken server. To avoid it, users can run nscd or put
2234 'options single-request' in /etc/resolv.conf.
2235 Implemented by Ulrich Drepper.
2239 * The following bugs are resolved with this release:
2241 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2242 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2243 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2244 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2245 6968, 6974, 6980, 6995, 7008, 7009, 7029
2247 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2248 up at the same time. Implemented by Ulrich Drepper.
2250 * TLS descriptors for LD and GD on x86 and x86-64.
2251 Implemented by Alexandre Oliva.
2253 * getaddrinfo now handles DCCP and UDPlite.
2254 Implemented by Ulrich Drepper.
2256 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2257 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2258 Implemented by Ulrich Drepper.
2260 * New implementation of memmem, strstr, and strcasestr which is O(n).
2261 Implemented by Eric Blake.
2263 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2265 * Implement "e" option for popen to open file descriptor with the
2266 close-on-exec flag set. Implemented by Ulrich Drepper.
2268 * Implement "b" mode for fmemopen. In this mode writes writes don't
2269 implicitly add a NUL byte and seeks from the end of the buffer really
2270 use the buffer end, not the string length as the basis.
2271 Implemented by Ulrich Drepper.
2273 * Many functions, exported and internal, now atomically set the close-on-exec
2274 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2276 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2278 Implemented by Pravin Satpute.
2280 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2284 * The following bugs are resolved with this release:
2286 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2287 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2288 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2289 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2290 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2291 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2292 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2294 * New locales: bo_CN, bo_IN, shs_CA.
2296 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2298 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2299 Implemented by Pravin Satpute.
2301 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2303 * nscd caches now all timeouts for DNS entries
2304 Implemented by Ulrich Drepper.
2306 * nscd is more efficient and wakes up less often.
2307 Implemented by Ulrich Drepper.
2309 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2310 vdprintf, and obstack_vprintf.
2311 Implemented by Jakub Jelinek.
2313 * Faster memset for x86-64.
2314 Implemented by Harsha Jagasia and H.J. Lu.
2316 * Faster memcpy on x86.
2317 Implemented by Ulrich Drepper.
2319 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2320 Implemented by Ulrich Drepper.
2322 * Faster sqrt and sqrtf implemention for some PPC variants.
2323 Implemented by Steven Munroe.
2327 * The following bugs are resolved with this release:
2329 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2330 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2331 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2332 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2333 5071, 5103, 5104, 5112, 5113, 5184, 5186
2335 * More checking functions: fread, fread_unlocked, open*, mq_open.
2336 Implemented by Jakub Jelinek and Ulrich Drepper.
2338 * Extend fortification to C++. Implemented by Jakub Jelinek.
2340 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2341 by not recognizing 'a' as a modifier when those specs are requested.
2342 Implemented by Jakub Jelinek.
2344 * PPC optimizations to math and string functions.
2345 Implemented by Steven Munroe.
2347 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2348 options to be passed. Implemented by Ulrich Drepper.
2350 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2352 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2353 Implemented by Ulrich Drepper.
2355 * Handle private futexes in the NPTL implementation.
2356 Implemented by Jakub Jelinek and Ulrich Drepper.
2358 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2359 Implemented by Roland McGrath and Ulrich Drepper.
2361 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2363 * SHA-256 and SHA-512 based password encryption.
2364 Implemented by Ulrich Drepper.
2366 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2367 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2370 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2371 Implemented by Ulrich Drepper.
2375 * The following bugs are resolved with this release:
2377 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2378 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2379 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2380 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2381 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2382 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2383 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2386 * New Linux interfaces: epoll_pwait, sched_getcpu.
2388 * New generic interfaces: strerror_l.
2390 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2394 * The following bugs are resolved with this release:
2396 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2397 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2398 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2399 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2400 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2401 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2402 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2403 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2404 3225, 3251, 3252, 3253, 3273
2406 * For Linux, the sorting of addresses returned by getaddrinfo now also
2407 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2408 Implemented by Ulrich Drepper.
2410 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2411 Implemented by Ulrich Drepper.
2413 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2415 * New iconv module for MIK. Contributed by Alexander Shopov.
2417 * For sites with broken group and/or passwd database, the auto-propagate
2418 option of nscd can prevent creating ID lookup entries from the results
2419 of a name lookup and vice versa. This usually is no problem but some
2420 site might have problems with the default behavior.
2421 Implemented by Ulrich Drepper.
2423 * Iterating over entire database in NIS can be slow. With the
2424 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2425 to trade time for memory. The entire database will be read at once.
2426 Implemented by Ulrich Drepper.
2428 * The interfaces introduced in RFC 3542 have been implemented by
2431 * Support for the new ELF hash table format was added by Ulrich Drepper.
2433 * Support for priority inheritance mutexes added by Jakub Jelinek and
2436 * Support for priority protected mutexes added by Jakub Jelinek.
2438 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2439 or_IN, csb_PL, fy_NL, sr_ME.
2443 * More overflow detection functions.
2445 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2446 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2448 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2449 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2450 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2451 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2452 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2453 by Masahide Washizawa.
2455 * It is now possible to install audit libraries for the dynamic linker using
2456 LD_AUDIT. Implemented by Ulrich Drepper.
2458 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2459 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2460 For a libc and libpthread that works well on Linux 2.4 kernels, we
2461 recommend using the stable 2.3 branch.
2463 * The new function open_wmemstream acts like open_memstream,
2464 but uses a wchar_t wide character buffer.
2466 * The new function ppoll is an improved version of poll, similar to pselect.
2468 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2469 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2470 renameat, unlinkat, linkat, symlinkat, readlinkat.
2472 * New Linux kernel system calls: unshare,
2473 inotify_init, inotify_add_watch, inotify_rm_watch.
2475 * The euidaccess function is now also known by the alias eaccess,
2476 for compatibility with some other systems.
2478 * Timezone data updated to 2006b version.
2482 * The following bugs are resolved with this release:
2484 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2485 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2486 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2487 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2488 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2489 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2491 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2493 * As of this release, GCC 4 can be used to compile the C Library.
2495 * Timezone data updated to 2005m version.
2499 * The following bugs are resolved with this release:
2501 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2502 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2503 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2504 777, 787, 821, 822, 823, 825
2506 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2510 * Support for RFC 3678. Real implementations exist only for Linux so far.
2511 Implemented by Ulrich Drepper.
2513 * nscd can now cache entries persistently. Expiring entries are reloaded.
2514 For speedups the cache can be shared in memory with client processes.
2515 Implemented by Ulrich Drepper.
2517 * nscd can now perform SELinux checks.
2518 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2520 * getaddrinfo queries are now cached. Canonical name lookup is performed
2522 Implemented by Ulrich Drepper.
2524 * The nothrow function attribute is used when headers are used by gcc when
2525 compiling C code. This can avoid significant amounts of exception
2528 * The malloc functions perform more error checking and are stricter when
2529 it comes to reacting on errors. The default action is to terminate
2530 the process after showing an error message. Implemented by Ulrich Drepper.
2532 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2533 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2535 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2536 namespaces using the new function dlmopen(). This feature is of course,
2537 like most other dynamic loading functionality, not available in statically
2538 linked applications. Implemented by Ulrich Drepper.
2540 * Low-overhead boundary checking variants of string and some stdio functions
2541 were added. These are to be used in conjunction with a gcc patch by
2542 Jakub Jelinek which adds calls to these functions if possible.
2543 Implemented by Jakub Jelinek and Ulrich Drepper.
2545 * Old code for several operating systems and machine architectures that
2546 have not been in working condition in a long time have been removed from
2547 the main source tree maintained by the GNU C Library's maintainers.
2548 These files are now reside in the separate `ports' source module
2549 that is usable as an add-on when building the library.
2553 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
2554 interrogate the dynamic linker, compatible with the Solaris interface.
2556 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
2557 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
2559 * getifaddrs now uses the netlink interface on Linux to get its information.
2560 Implemented by Thorsten Kukuk.
2562 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
2563 Implemented by Ulrich Drepper.
2565 * support for non-executable stacks on x86 has been added. Changes mostly
2568 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
2571 * getaddrinfo now performs destination address selection according to
2576 * Thread-safe interfaces for many functions that access locale data
2577 were added in version 2.3, but these features were omitted from NEWS.
2578 Many functions have variants with an `_l' suffix that take a `locale_t'
2579 object as a parameter rather than consulting the current locale.
2580 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
2581 create and maintain `locale_t' objects. Additionally, the new function
2582 `uselocale' sets "the current locale" (as used by functions not so
2583 parameterized) set for an individual thread. These features were added
2584 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
2586 * The functions getresuid, getresgid, setresuid, and setresgid, which
2587 have long been available on Linux, are now declared in <unistd.h>
2588 and are now also available on the Hurd.
2590 * ELF thread-local storage support (TLS) now works on x86-64.
2592 * The new dynamic string token $LIB is expanded in shared library names.
2593 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
2595 * Aldy Hernandez contributed complete software floating point support for
2596 PowerPC machines with no FPU.
2598 * fexecve is implemented on Linux.
2600 * The `btowc' function should work at least twice as fast due to
2601 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
2603 * With approriate thread add-ons cancelable functions are now implemented
2604 in libc.so as well. No need to call the function in libpthread. This
2605 change allowed to finally disable the incorrect and expensive handling
2606 of weak definition in ld.so.
2608 * Yet more PLT entries in libc.so have been removed. We finally arrived
2609 at the bare minimum. Startup times improved appropriately.
2611 * Support for the new Linux/x86 system call interface was added. The
2612 AT_SYSINFO auxiliary vector entry is recognized and handled.
2616 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
2619 * iconv (the program and the interface) now accepts empty names (excluding
2620 options like //TRANSLIT) to mean "use charset of current locale".
2622 * localedef can now transliterate characters in strings which are not in
2623 the provided charmap. The information from the input locale is used.
2625 * Prelinking support was added for ELF targets. This requires additional
2626 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
2628 * Read-only stdio streams now use mmap to speed up operation by eliminating
2629 copying and buffer underflows. To use add 'm' to the mode string of
2630 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
2632 * The malloc functions were completely rewritten by Wolfram Gloger based
2633 on Doug Lea's malloc-2.7.0.c.
2635 * Isamu Hasegawa contributed a completely new and POSIX-conformant
2636 implementation of regex.
2638 * Bruno Haible upgraded the iconv and locale implementation to support
2641 * Contents of the LC_* and LANG environment variables in the CEN style are
2642 not recognized anymore. It never was used. Change by Ulrich Drepper.
2644 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
2645 thread-local storage (TLS) ABI on some platforms.
2646 Changes by Ulrich Drepper. SH support by Kaz Kojima.
2648 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
2649 EUC-JISX0213, and TSCII.
2651 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
2652 BSD-compatible interface for getting all network interface addresses.
2653 Implementation for IPv4 by Roland McGrath.
2655 * Loading of locale data is faster due to the introduction of a locale
2656 archive. Implemented by Roland McGrath and Ulrich Drepper.
2658 * Startup times are significantly reduced by not using exported functions
2659 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
2662 * Steven Munroe contributed a port to PowerPC64/Linux.
2666 * The Hurd now uses the GNU libio implementation of stdio rather than the
2667 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
2669 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
2670 with the same behavior as the Linux system call of the same name.
2674 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
2675 128-bit long double format.
2677 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
2678 IBM1161, and IBM1162 charsets.
2680 * Andreas Jaeger contributed a port to x86-64/Linux.
2682 * Peter Bruin contributed a port to PowerPC/Hurd.
2684 * libc functions using I/O streams now can handle wide-oriented streams
2687 * optimizations in the dynamic linker. Binaries created by recent binutils
2688 versions start up quicker due to reduced time spent on relocations.
2690 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
2694 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
2695 asinh, atanh, j0 for the 128-bit long double format.
2697 * Bruno Haible updated all the code handling Unicode in some form to
2698 support Unicode 3.1.
2700 * Speed of regex for single-byte locales is back to previous levels.
2701 Patch by Isamu Hasegawa.
2703 * Alpha, SPARC, and IA-64 now also using floating stacks.
2705 * Startup time of internationalized applications greatly improved through
2706 iconv cache. Use iconvconfig to generate the cache file.
2707 Contributed by Ulrich Drepper.
2709 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
2710 severe performance problems. Patch by David Mosberger.
2712 * The Hurd port got a lot more functionality like AIO, various stdio
2713 extensions, etc. Mainly done by Roland McGrath.
2715 * mtrace can now lookup symbols in shared libraries.
2719 * Intel's IA-64 math library is largely integrated. It provides fast and
2720 accurate implementatations for most basic and standard math functions
2721 in float, double, and long double format.
2723 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
2724 and asin for the 96-bit long double format and asin, log, tan for the
2725 128-bit long double format.
2727 * The beginning of a last-bit accurate math library by IBM Haifa were added.
2728 The basic double functions exist today. Contributed by Abraham Ziv
2729 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
2730 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
2732 * An asynchronous name lookup library was added. The interface is designed
2733 after POSIX AIO. The proposal was circulated beforehand to get comments.
2734 No negative ones came in. Implemented by Ulrich Drepper.
2736 * Port to S390/64bit contributed by Martin Schwidefsky
2737 <schwidefsky@de.ibm.com>.
2739 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
2740 of functions for Linux/IA-64.
2742 * The RPC code is now thread safe. Threads can now use the same service
2743 of different services at the same time. Patch by Eric Norum
2744 <eric.norum@usask.ca> with some help by Ulrich Drepper.
2746 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
2747 family of functions for Linux/S390.
2749 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
2750 of functions for Linux/x86.
2752 * Port to Linux/CRIS contributed by Axis Communications.
2756 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
2757 we can now check for namespace violations and missing declarations. The
2758 result is that almost all headers are now Unix-compliant (as defined in
2759 the upcoming XPG6). The negative side is that some programs might need
2760 corrections, too, if they depend on the incorrect form of the headers in
2761 previous versions which defined too many symbols and included too many
2764 * regex now handles multibyte character sets correctly.
2765 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
2767 * iconv (the program) does now conform to the upcoming XPG6 and handles
2768 charmaps. Instead of the charset names the path of charmaps can be
2769 provided and the conversion happens based on this data.
2770 Contributed by Ulrich Drepper.
2772 * The locale program now provides detailed information about the installed
2777 only lists the names of the supported locales
2781 provides details such as country, language, and codeset name.
2782 Contributed by Ulrich Drepper.
2786 * The gencat program now parses the input file according to the charset
2787 selected by the LC_CTYPE category. This is important for stateful
2788 character sets. To make generating catalogs easier there is a way
2789 to overwrite the charset selected by the locale: before the first
2790 message or $ quote line the catalog can contain a line like
2792 $ codeset=ISO-8859-2
2794 to select the charset (ISO-8859-2 in this case).
2796 Implemented by Shinya Hanataka and Ulrich Drepper.
2798 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
2799 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
2800 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
2802 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
2803 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
2805 * The LANGUAGE environment variable is now ignored unless the locale is
2806 changed from the default "C" locale.
2808 * The usual bug fixes.
2812 * Greg McGary added runtime support for bounds checking using gcc's
2813 new -fbounded-pointers option. ix86 target is complete. PowerPC
2816 * Thorsten Kukuk added secure mode support to nscd.
2818 * The Berkeley db libraries have been removed.
2820 Related, the nss_db module is now in a separate package since it
2821 obviously requires a database library being available.
2823 * Wide character I/O streams implemented by Ulrich Drepper.
2825 * Functions from the extended socket API added by Ulrich Drepper.
2827 * Functions feenableexcept and fedisableexcept to control the
2828 behaviour of individual exceptions have been added by Andreas Jaeger.
2830 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
2832 * The resolver code has been updated from bind 8.2.3-T5B which supports
2833 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
2836 This change could in some situations effect backward compatibility. Since
2837 now `_res' is a thread-local instead of a global variable, modifying it
2838 in one thread does not have any effect in other threads.
2840 The resolver library was also extended to allow IPv6 as the transport
2841 protocol for the requests. This work was done by Stig Venaas.
2843 * Compatibility code for K&R C compilers has been removed from the
2844 header files. A ISO C compiler is needed to use the library
2845 (conforming to either C89 or C99 standard).
2847 * Complete rewrite of the localedef program to support multibyte character
2848 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
2849 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
2850 Implemented by Ulrich Drepper.
2852 Bruno Haible significantly improved the generation and use of the data
2853 structures for the wide character tables.
2855 * Plural handling in gettext implemented by Ulrich Drepper.
2857 * The utmp daemon has been removed.
2859 * The port to MIPS-Linux has been finished by Andreas Jaeger.
2861 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
2864 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
2866 * POSIX barriers implemented by Kaz Kylheku.
2868 * POSIX spawn function family implemented by Ulrich Drepper.
2870 * POSIX spinlocks are now available.
2872 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
2874 * the configure option --enable-kernel=X.Y.Z allows to strip out
2875 compatibility for kernel versions before X.Y.Z. This is currently only
2876 implemented for Linux.
2878 * the sockaddr_in6 structure changed. The IPv6 working group added a new
2879 field sin6_scope_id. This means that all programs using IPv6 should be
2880 recompiled. Don't expect binary compatibility with previous glibc
2883 * various conversion modules for IBM character sets contributed by
2886 * IA-64 port by Jes Sorensen and HJ Lu.
2900 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
2902 * New xdr functions are added; some rpc functions are now 64bit clean.
2904 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
2906 * Fixed known incompatibilities with glibc 2.0.
2908 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
2910 * Optimized a number of functions (especially the ELF dynamic loader).
2912 * Update timezone data files.
2914 * lots of charmaps corrections
2916 * some new locale definitions and charmaps
2921 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
2922 conform to POSIX member type requirements. Symbol versions have been
2923 adjusted as needed within the library, and for direct use by applications,
2924 but there is potential for problems if third-party libraries use
2925 struct timeval as part of their interface. This does not present
2926 a problem for X and other "essential" system libraries.
2928 * An additional locale model to support C++ Standard Library locale
2929 model and probably more was implemented by Ulrich Drepper.
2931 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
2934 * Miles Bader provided the `argp' function family to support hierachical
2935 command line argument parsing, layered on top of getopt.
2937 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
2939 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
2942 * scanf recognizes the %a and %A format for scanning floating point numbers.
2944 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
2945 information and interfaces for the available integer types.
2947 * about 130 new math functions were added to implement the ISO C9x math
2950 * the new header <complex.h> contains definitions of the complex math
2951 functions from ISO C 9X.
2953 * the new header <tgmath.h> defines generic macros to use complex or
2954 real valued functions.
2956 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
2958 * Andreas Jaeger provided a test suite for the math library.
2960 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
2962 * Ulrich Drepper added character set conversion functions (iconv).
2964 * Optimized string functions have been added.
2966 * The localedata addon is now part of glibc.
2968 * An implementation of profiling shared libraries was added by Ulrich Drepper.
2970 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
2971 daemon for NSS (nscd).
2973 Missing a better place here are some numbers on improvements. Under
2974 Linux 2.1.125 un-tar-ing the kernel sources takes
2978 using local files 12.19s 6.88s 22.91s
2980 using NIS 13.92s 8.91s 26.34s
2982 using NIS & nscd 10.37s 7.34s 25.30s
2984 using NIS+ 27.57s 30.37s 640.46s
2986 using NIS+ & nscd 10.25s 7.83s 26.51s
2988 using NIS & old nscd [1] 13.83s 8.32s 29.60s
2990 Keep in mind that non-namelookup related operations dominate above times.
2991 It was just a common complain that using NIS+ unpacking the kernel is
2994 [1] The old nscd implementation is not available anymore since it was
2995 distributed with glibc up to version 2.0.98 and thus is now replaced.
2997 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
2999 * Mark Kettenis provided a Hesiod NSS module.
3001 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3002 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3004 * rcmd can now handle netgroups (Dick Streefland).
3006 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3009 * Support for the IPv6 protocol has been added to the socket API, as per the
3010 latest draft standards.
3012 * Support for Linux 2.2 has been added.
3014 * Interface changes relative to the latest 2.0.x release:
3015 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3016 addseverity NEW: Unix98
3017 alphasort64 NEW: LFS
3018 argp_err_exit_status NEW: argp, GNU ext
3019 argp_error NEW: argp, GNU ext
3020 argp_failure NEW: argp, GNU ext
3021 argp_help NEW: argp, GNU ext
3022 argp_parse NEW: argp, GNU ext
3023 argp_program_bug_address NEW: argp, GNU ext
3024 argp_program_version NEW: argp, GNU ext
3025 argp_program_version_hook NEW: argp, GNU ext
3026 argp_state_help NEW: argp, GNU ext
3027 argp_usage NEW: argp, GNU ext
3028 authdes_create NEW: Secure RPC
3029 authdes_getucred NEW: Secure RPC
3030 authdes_pk_create NEW: Secure RPC
3031 backtrace NEW: GNU ext.
3032 backtrace_symbols NEW: GNU ext.
3033 backtrace_symbols_fd NEW: GNU ext.
3035 cacosf NEW: ISO C 9x
3036 cacosh NEW: ISO C 9x
3037 cacoshf NEW: ISO C 9x
3038 cacoshl NEW: ISO C 9x
3039 cacosl NEW: ISO C 9x
3046 casinf NEW: ISO C 9x
3047 casinh NEW: ISO C 9x
3048 casinhf NEW: ISO C 9x
3049 casinhl NEW: ISO C 9x
3050 casinl NEW: ISO C 9x
3052 catanf NEW: ISO C 9x
3053 catanh NEW: ISO C 9x
3054 catanhf NEW: ISO C 9x
3055 catanhl NEW: ISO C 9x
3056 catanl NEW: ISO C 9x
3057 cbc_crypt NEW: Secure RPC
3061 ccoshf NEW: ISO C 9x
3062 ccoshl NEW: ISO C 9x
3068 cimagf NEW: ISO C 9x
3069 cimagl NEW: ISO C 9x
3070 clearerr_locked REMOVED
3071 clntunix_create NEW: sunrpc ext
3073 clog10 NEW: ISO C 9x
3074 clog10f NEW: ISO C 9x
3075 clog10l NEW: ISO C 9x
3085 cprojf NEW: ISO C 9x
3086 cprojl NEW: ISO C 9x
3088 crealf NEW: ISO C 9x
3089 creall NEW: ISO C 9x
3094 csinhf NEW: ISO C 9x
3095 csinhl NEW: ISO C 9x
3098 csqrtf NEW: ISO C 9x
3099 csqrtl NEW: ISO C 9x
3103 ctanhf NEW: ISO C 9x
3104 ctanhl NEW: ISO C 9x
3106 des_setparity NEW: Secure RPC
3107 ecb_crypt NEW: Secure RPC
3108 endutxent NEW: Unix98
3110 exp10f NEW: ISO C 9x
3111 exp10l NEW: ISO C 9x
3115 fattach NEW: STREAMS
3116 fdetach NEW: STREAMS
3120 feclearexcept NEW: ISO C 9x
3121 fegetenv NEW: ISO C 9x
3122 fegetexceptflag NEW: ISO C 9x
3123 fegetround NEW: ISO C 9x
3124 feholdexcept NEW: ISO C 9x
3126 feraiseexcept NEW: ISO C 9x
3127 ferror_locked REMOVED
3128 fesetenv NEW: ISO C 9x
3129 fesetexceptflag NEW: ISO C 9x
3130 fesetround NEW: ISO C 9x
3131 fetestexcept NEW: ISO C 9x
3132 feupdateenv NEW: ISO C 9x
3133 fflush_locked REMOVED
3137 fgets_unlocked NEW: GNU ext.
3138 fileno_locked REMOVED
3150 fputc_locked REMOVED
3151 fputs_unlocked NEW: GNU ext.
3152 fread_unlocked NEW: GNU ext.
3157 fstatvfs NEW: Unix98
3161 ftruncate64 NEW: LFS
3163 fwrite_unlocked NEW: GNU ext.
3164 gai_strerror NEW: IPv6
3168 getchar_locked REMOVED
3170 getdate_err NEW: Unix98
3171 getdate_r NEW: GNU ext.
3173 getnameinfo NEW: IPv6
3174 getnetname NEW: Secure RPC
3175 getpmsg NEW: STREAMS
3176 getpt NEW: Unix98 PTY
3177 getrlimit64 NEW: LFS
3178 getutxent NEW: Unix98
3179 getutxid NEW: Unix98
3180 getutxline NEW: Unix98
3181 glob64 NEW: GNU ext.
3182 globfree64 NEW: GNU ext.
3183 gnu_get_libc_release NEW: GNU ext.
3184 gnu_get_libc_version NEW: GNU ext.
3185 grantpt NEW: Unix98 PTY
3186 host2netname NEW: Secure RPC
3188 iconv_close NEW: iconv
3189 iconv_open NEW: iconv
3190 if_freenameindex NEW: IPv6
3191 if_indextoname NEW: IPv6
3192 if_nameindex NEW: IPv6
3193 if_nametoindex NEW: IPv6
3194 in6addr_any NEW: IPv6
3195 in6addr_loopback NEW: IPv6
3196 inet6_isipv4mapped NEW: IPv6
3197 isastream NEW: STREAMS
3198 iswblank NEW: GNU ext.
3199 key_decryptsession NEW: Secure RPC
3200 key_decryptsession_pk NEW: Secure RPC
3201 key_encryptsession NEW: Secure RPC
3202 key_encryptsession_pk NEW: Secure RPC
3203 key_gendes NEW: Secure RPC
3204 key_get_conv NEW: Secure RPC
3205 key_secretkey_is_set NEW: Secure RPC
3206 key_setnet NEW: Secure RPC
3207 key_setsecret NEW: Secure RPC
3208 llrint NEW: ISO C 9x
3209 llrintf NEW: ISO C 9x
3210 llrintl NEW: ISO C 9x
3211 llround NEW: ISO C 9x
3212 llroundf NEW: ISO C 9x
3213 llroundl NEW: ISO C 9x
3218 lrintf NEW: ISO C 9x
3219 lrintl NEW: ISO C 9x
3220 lround NEW: ISO C 9x
3221 lroundf NEW: ISO C 9x
3222 lroundl NEW: ISO C 9x
3224 makecontext NEW: Unix98
3225 mempcpy NEW: GNU ext.
3228 modify_ldt NEW: kernel
3232 nearbyint NEW: ISO C 9x
3233 nearbyintf NEW: ISO C 9x
3234 nearbyintl NEW: ISO C 9x
3235 netname2host NEW: Secure RPC
3236 netname2user NEW: Secure RPC
3237 nexttoward NEW: ISO C 9x
3238 nexttowardf NEW: ISO C 9x
3239 nexttowardl NEW: ISO C 9x
3243 passwd2des NEW: Secure RPC
3245 pow10f NEW: GNU ext.
3246 pow10l NEW: GNU ext.
3249 printf_size NEW: GNU ext.
3250 printf_size_info NEW: GNU ext.
3251 profil_counter REMOVED
3252 pthread_mutexattr_getkind_np REPLACED
3253 pthread_mutexattr_setkind_np REPLACED
3254 ptsname NEW: Unix98 PTY
3255 ptsname_r NEW: Unix98 PTY
3257 putchar_locked REMOVED
3258 putgrent NEW: GNU ext.
3260 putpmsg NEW: STREAMS
3261 pututxline NEW: Unix98
3265 readdir64_r NEW: LFS
3266 remquo NEW: ISO C 9x
3267 remquof NEW: ISO C 9x
3268 remquol NEW: ISO C 9x
3270 roundf NEW: ISO C 9x
3271 roundl NEW: ISO C 9x
3273 scalbln NEW: ISO C 9x
3274 scalblnf NEW: ISO C 9x
3275 scalblnl NEW: ISO C 9x
3277 sendfile NEW: kernel
3278 setcontext NEW: Unix98
3279 setrlimit64 NEW: LFS
3280 setutxent NEW: Unix98
3282 sigignore NEW: Unix98
3283 sigqueue NEW: POSIX.1b
3284 sigrelse NEW: Unix98
3285 sigset NEW: POSIX.1b
3286 sigtimedwait NEW: POSIX.1b
3287 sigwaitinfo NEW: POSIX.1b
3288 sincos NEW: GNU ext.
3289 sincosf NEW: GNU ext.
3290 sincosl NEW: GNU ext.
3294 strcasestr NEW: GNU ext.
3295 strtoimax NEW: ISO C 9x
3296 strtoumax NEW: ISO C 9x
3297 strverscmp NEW: GNU ext.
3298 svcauthdes_stats NEW: Secure RPC
3299 svcunix_create NEW: sunrpc etx
3300 svcunixfd_create NEW: sunrpc ext
3301 swapcontext NEW: Unix98
3302 tcgetsid NEW: Unix98 PTY
3303 tdestroy NEW: GNU ext.
3304 tgamma NEW: ISO C 9x
3305 tgammaf NEW: ISO C 9x
3306 tgammal NEW: ISO C 9x
3310 truncf NEW: ISO C 9x
3311 truncl NEW: ISO C 9x
3313 unlockpt NEW: Unix98 PTY
3314 updwtmpx NEW: Unix98
3315 user2netname NEW: Secure RPC
3316 utmpxname NEW: Unix98
3317 versionsort NEW: GNU ext.
3318 versionsort64 NEW: GNU ext.
3320 wcscasecmp NEW: GNU ext.
3321 wcsncasecmp NEW: GNU ext.
3322 wcsnlen NEW: GNU ext.
3323 wcstoimax NEW: ISO C 9x
3324 wcstoll NEW: ISO C 9x
3325 wcstoull NEW: ISO C 9x
3326 wcstoumax NEW: ISO C 9x
3328 wordexp NEW: POSIX.2
3329 wordfree NEW: POSIX.2
3330 write_profiling REMOVED
3331 xdecrypt NEW: Secure RPC
3332 xdr_authdes_cred NEW: Secure RPC
3333 xdr_authdes_verf NEW: Secure RPC
3334 xdr_cryptkeyarg NEW: Secure RPC
3335 xdr_cryptkeyarg2 NEW: Secure RPC
3336 xdr_cryptkeyres NEW: Secure RPC
3337 xdr_getcredres NEW: Secure RPC
3338 xdr_key_netstarg NEW: Secure RPC
3339 xdr_key_netstres NEW: Secure RPC
3340 xdr_keybuf NEW: Secure RPC
3341 xdr_keystatus NEW: Secure RPC
3342 xdr_netnamestr NEW: Secure RPC
3343 xdr_sizeof NEW: Secure RPC
3344 xdr_unixcred NEW: sunrpc ext
3345 xencrypt NEW: Secure RPC
3346 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3357 * inet_ntoa is thread-safe
3359 * updwtmp is moved from libutil to libc
3361 * rewrite of cbrt function
3363 * update of timezone data
3377 * add atoll function
3379 * fix complex problems in Berkeley DB code
3381 * fix math functions
3385 * fixed lots of header problems (especially Linux/GNU specific)
3387 * dynamic loader preserves all registers
3389 * Roland McGrath provided support for handling of auxiliary objects in
3390 the ELF dynamic loader.
3392 * support for parallel builds is improved
3396 * GNU extensions are no longer declared by default. To enable them you
3397 must define the macro `_GNU_SOURCE' in your program or compile with
3400 * The library has changed from using GNU ld symbol aliases to using weak
3401 symbols where available. The ELF object file format supports weak
3402 symbols; GNU ld also supports weak symbols in the a.out format. (There
3403 is also now support for other GNU ld extensions in ELF. Use the
3404 `--with-elf' option to configure to indicate you have ELF, and
3405 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3406 of many files which contained only symbol aliases, reducing the size of
3407 the source and the compiled library; many other files were renamed to
3408 less cryptic names previously occupied by the symbol alias files.
3409 There is a new header file <elf.h> for programs which operate on
3410 files in the ELF format.
3412 * Converted to Autoconf version 2, so `configure' has more options.
3413 Run `configure --help' to see the details.
3415 * The library can now be configured to build profiling, highly-optimized
3416 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3417 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3418 `configure' enable building these extra libraries. The shared library is
3419 built by default when using both ELF and GNU ld. When shared libraries
3420 are enabled, the new library `-ldl' is available for arbitrary run-time
3421 loading of shared objects; its interface is defined in <dlfcn.h>. The
3422 new header file <link.h> gives access to the internals of the run-time
3423 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3424 application of same name on other systems and it provides information
3425 about dynamically linked binaries.
3427 * The C library now provides the run-time support code for profiling
3428 executables compiled with `-pg'. Programs can control the profiling code
3429 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3430 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3431 the support for this file format was contributed by David Mosberger-Tang.
3433 * The math code has been replaced with a math library based on fdlibm from
3434 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3435 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3436 functions. The math functions now reside in a separate library, so
3437 programs using them will need to use `-lm' their linking commands.
3439 * John C. Bowman contributed optimized ix87 assembler inline functions.
3441 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3442 mechanism similar to that found in Solaris 2. This is now used for the
3443 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3444 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3445 file controls what services are used for each individual database. This
3446 works by loading shared libraries with names specified in `nsswitch.conf',
3447 so service modules can be changed or added at any time without even
3448 relinking any program. Currently there are the file, db, and NIS based
3449 NSS services available.
3451 * The new functions `strtoq' and `strtouq' parse integer values from
3452 strings, like `strtol' and `strtoul', but they return `long long int' and
3453 `unsigned long long int' values, respectively (64-bit quantities).
3455 * The new functions `strtof' and `strtold' parse floating-point values from
3456 strings, like `strtod', but they return `float' and `long double' values,
3457 respectively (on some machines `double' and `long double' are the same).
3459 * Ulrich Drepper has contributed new implementations of the floating-point
3460 printing and reading code used in the `printf' family of functions and
3461 `strtod', `strtof', and `strtold'. These new functions are perfectly
3462 accurate, and much faster than the old ones.
3464 * The implementation of the POSIX locale model was completely rewritten by
3465 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3466 to compile the POSIX locale definition.
3468 * The former dummy implementations of the strcoll and strxfrm function are
3469 now replaced by fully functional code contributed by Ulrich Drepper. The
3470 collation information comes from the POSIX locale definitions.
3472 * The new header <langinfo.h> defines an interface for accessing
3473 various locale-dependent data (using the locale chosen with `setlocale').
3475 * Ulrich Drepper has contributed a new suite of functions for operation on
3476 wide-character and multibyte-character strings, in <wchar.h>;
3477 and classification and case conversion of wide characters, in <wctype.h>.
3478 These new functions are conforming to the ISO C, Amendement 1 specification.
3480 * There is now a second implementation of the standard I/O library available.
3481 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3482 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3484 * You can now use positional parameter specifications in format strings
3485 for the `printf' and `scanf' families of functions. For example,
3486 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3487 ``Number 6, Mr Jones''. This is mainly useful when providing different
3488 format strings for different languages, whose grammars may dictate
3489 different orderings of the values being printed. To support this
3490 feature, the interface for `register_printf_handler' has changed; see
3491 the header file <printf.h> for details.
3493 * The `printf' and `scanf' families of functions now understand a new
3494 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3495 to group numbers as indicated by the locale; for `scanf' and friends, this
3496 says to accept as valid only a number with all the proper grouping
3497 separators in the right places. In the default "C" locale, numbers are
3498 not grouped; but locales for specific countries will define the usual
3499 conventions (i.e. separate thousands with `,' in the US locale).
3501 * The pgrp functions have been regularized, slightly incompatibly but much
3502 less confusingly. The core functions are now `getpgid' and `setpgid',
3503 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3504 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3505 provided for compatibility. There is no longer an incompatible `getpgrp'
3506 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3508 * The new header file <fts.h> and suite of functions simplify programs that
3509 operate on directory trees. This code comes from 4.4 BSD.
3511 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3512 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3513 NSS scheme used in glibc.
3515 * There is a new malloc debugging hook `__memalign_hook'.
3517 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3518 `unsigned int' in <sys/types.h>. These are for compatibility only and
3519 their use is discouraged.
3521 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3522 done differently, so that it works even without GNU ld.
3524 * New function `euidaccess' checks allowed access to a file like `access',
3525 but using the effective IDs instead of the real IDs.
3527 * The time zone data files have been updated for the latest and greatest
3528 local time conventions of the countries of the world.
3530 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3533 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3534 for formatting floating-point numbers. They are provided only for
3535 compatibility; new programs should use `sprintf' instead. There are
3536 also equivalent function for the `long double' floating-point type and
3537 all functions also exist in a reentrant form.
3539 * The new auxiliary library `-lutil' from 4.4 BSD contains various
3540 functions for maintaining the login-record files (primarily of use to
3541 system programs such as `login'), and convenient functions for
3542 allocating and initializing a pseudo-terminal (pty) device.
3544 * Ulrich Drepper has contributed new support for System V style
3545 shared memory and IPC on systems that support it.
3547 * Ulrich Drepper has contributed several miscellaneous new functions found
3548 in System V: The `hsearch' family of functions provide an effective
3549 implementation of hash tables; `a64l' and `l64a' provide a very simple
3550 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
3553 * Ulrich Drepper has contributed new reentrant counterparts for the
3554 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
3556 * Ulrich Drepper has contributed new, highly-optimized versions of several
3557 string functions for the i486/Pentium family of processors.
3559 * Ulrich Drepper has updated the Linux-specific code, based largely
3560 on work done in Hongjiu Lu's version of GNU libc for Linux.
3561 The GNU library now supports Linux versions 2.0.10 and later,
3562 using the ELF object file format (i[3456]86-*-linux).
3564 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
3566 * David Mosberger-Tang and Richard Henderson have ported the C library
3567 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
3568 dynamic linking support for ELF/Alpha.
3570 * Richard Henderson contributed several Alpha optimized assembler function
3571 for arithmetic and string handling.
3573 * Ulrich Drepper has contributed a new set of message catalog functions to
3574 support multiple languages using the <libintl.h> interface, for use with
3575 his new package GNU gettext. Translation volunteers have contributed
3576 catalogs of the library's messages in Spanish, German, and Korean.
3578 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
3579 program and the `catgets' function for reading the catalog files it
3580 creates. (The <libintl.h> interface is preferred; we include the
3581 <nl_types.h> interface using `catgets' only for source compatibility with
3582 programs already written to use it.)
3584 * New header file <values.h> gives SVID-compatible names for <limits.h>
3587 * Various new macros, declarations, and small header files for compatibility
3590 * New function `group_member' is a convenient way to check if a process has
3591 a given effective group ID.
3593 * When using GCC 2.7 and later, the socket functions are now declared in a
3594 special way so that passing an argument of type `struct sockaddr_in *',
3595 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
3596 `struct sockaddr *' type, does not generate a type-clash warning.
3598 * New function `error' declared in header file <error.h> is a convenient
3599 function for printing error messages and optionally exiting; this is the
3600 canonical function used in GNU programs. The new functions `err', `warn',
3601 and friends in header file <err.h> are the canonical 4.4 BSD interface for
3602 doing the same thing.
3604 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
3605 POSIX.2 `glob' function to do ~ and {...} expansion.
3607 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
3608 `clearenv' which is used in POSIX.9 is also available.
3610 * New function `getsid' returns session ID number on systems that support it.
3612 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
3613 files <db.h> and <mpool.h> provide a rich set of functions for several
3614 types of simple databases stored in memory and in files, and <ndbm.h> is
3615 an old `ndbm'-compatible interface using the `db' functions. Link with
3616 `-ldb' to get these functions.
3618 * New macro `strdupa' copies a string like `strdup', but uses local stack
3619 space from `alloca' instead of dynamic heap space from `malloc'.
3621 * New function `strnlen' is like `strlen' but searches only a given maximum
3622 number of characters for the null terminator. `stpncpy', `strndup' and
3623 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
3626 * New function `statfs' in header <sys/statfs.h>.
3628 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
3629 provide convenient functions for operating on blocks of null-terminated
3632 * A new suite of functions in <utmp.h> handle all the details of reading
3633 and writing the utmp file.
3635 * An implementation of the NIS/YP(tm) based NSS service was contributed by
3638 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
3639 completely POSIX compliant and also implemented the extended functionality
3640 to handle alternate digit representation and alternate era date formats.
3642 * Ulrich Drepper provided an implementation of the `strptime' function
3643 defined in XPG4.2 which transforms a string into a `struct tm' value.
3645 * Paul Eggert provided the tzselect shell script as part of the timezone
3646 code. The shell script makes it easy to select the correct timezone
3649 * The implementation of the malloc family of functions is completely replaced
3650 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
3651 The implementation uses the mmap function (if available) and it is
3652 optimized for the use in multi threaded programs.
3654 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
3655 functions. This new functionality is usable by specifying a special
3656 salt string and it is compatible with implementation on *BSD systems.
3658 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
3659 `getsubopt' to handle second level command line options, `bsd_signal'
3660 to access BSD style `signal' functionality, the obsolete `regexp' style
3663 * the `lchown' function is available on system which support this
3666 * The implementation of the shadow password handling function was contributed
3669 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
3671 * POSIX.1g support was added. The <sys/select.h> header is available,
3672 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
3673 implementation of `getaddrinfo'.
3677 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
3679 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
3680 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
3682 * The new function `daemon' from 4.4 BSD is useful for server programs that
3683 want to put themselves in the background.
3685 * Joel Sherrill has contributed support for several standalone boards that
3686 run without an operating system.
3688 * `printf', `scanf' and friends now accept a `q' type modifier for long
3689 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
3691 * All of the code taken from BSD (notably most of the math and networking
3692 routines) has been updated from the BSD 4.4-Lite release.
3694 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
3696 * The new functions `getdomainname' and `setdomainname' fetch or change the
3697 YP/NIS domain name. These are system calls which exist on systems which
3700 * The time zone data files have been updated for the latest international
3703 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
3704 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
3708 * The C library now includes support for Sun RPC, from Sun's free
3709 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
3710 are included. (There is still no support for YP.)
3712 * Tom Quinn has contributed a port of the C library to SGI machines running
3713 Irix 4 (mips-sgi-irix4).
3715 * The new `lockf' function is a simplified interface to the locking
3716 facilities of `fcntl', included for compatibility.
3718 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
3720 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
3722 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
3725 * The error message printed by `assert' for a failed assertion now includes
3726 the name of the program (if using GNU ld) and the name of the calling
3727 function (with versions of GCC that support this).
3729 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
3731 * The library now includes the <sys/mman.h> header file and memory
3732 management functions `mmap', `munmap', `mprotect', `msync', and
3733 `madvise', on systems that support those facilities.
3735 * The interface for `mcheck' has changed slightly: the function called to
3736 abort the program when an allocation inconsistency is detected now takes
3737 an argument that indicates the type of failure. The new function
3738 `mprobe' lets you request a consistency check for a particular block at
3739 any time (checks are normally done only when you call `free' or `realloc'
3742 * It is now possible to easily cross-compile the C library, building on one
3743 system a library to run on another machine and/or operating system. All
3744 you need to do is set the variable `HOST_CC' in `configparms' to the
3745 native compiler for programs to run on the machine you are building on (a
3746 few generator programs are used on Unix systems); set `CC' to the
3749 * The new function `fexecve' (only implemented on the GNU system) executes
3750 a program file given a file descriptor already open on the file.
3754 * Brendan Kehoe has contributed most of a port to the DEC Alpha
3755 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
3757 * You can set the variable `libprefix' in `configparms' to specify a prefix
3758 to be prepended to installed library files; this makes it easy to install
3759 the GNU C library to be linked as `-lgnuc' or whatever.
3761 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
3762 copies a limited number of characters from a string, and returns the
3763 address of the last character written.
3765 * You no longer need to check for whether the installed `stddef.h' is
3766 compatible with the GNU C library. configure now checks for you.
3768 * You can now define a per-stream `fileno' function to convert the
3769 stream's cookie into an integral file descriptor.
3771 * ``malloc (0)'' no longer returns a null pointer. Instead, it
3772 allocates zero bytes of storage, and returns a unique pointer which
3773 you can pass to `realloc' or `free'. The behavior is undefined if
3774 you dereference this pointer.
3776 * The C library now runs on Sony NEWS m68k machines running either
3777 NewsOS 3 or NewsOS 4.
3779 * The new `syscall' function is a system-dependent primitive function
3780 for invoking system calls. It has the canonical behavior on Unix
3781 systems, including unreliable return values for some calls (such as
3782 `pipe', `fork' and `getppid').
3784 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
3785 to `EAGAIN', which is the preferred name. On systems whose kernels
3786 use two distinct codes, the C library now translates EWOULDBLOCK to
3787 EAGAIN in every system call function.
3791 * The GNU C Library Reference Manual is now distributed with the library.
3792 `make dvi' will produce a DVI file of the printed manual.
3793 `make info' will produce Info files that you can read on line using C-h i
3794 in Emacs or the `info' program.
3795 Please send comments on the manual to bug-glibc-manual@gnu.org.
3797 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
3799 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
3801 * Jason Merrill has contributed a port to the Sequent Symmetry running
3802 Dynix version 3 (i386-sequent-dynix).
3804 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
3805 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
3807 * New function `memory_warnings' lets you arrange to get warnings when
3808 malloc is running out of memory to allocate, like Emacs gives you.
3810 * The C library now contains the relocating allocator used in Emacs 19 for
3811 its editing buffers. This allocator (ralloc) minimizes allocation
3812 overhead and fragmentation by moving allocated regions around whenever it
3813 needs to. You always refer to a ralloc'd region with a "handle" (a
3814 pointer to a pointer--an object of type `void **').
3816 * There is a new `printf' format: `%m' gives you the string corresponding
3817 to the error code in `errno'.
3819 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
3820 or `%[' conversion, but instead of filling in a fixed-sized buffer you
3821 pass, the `a' modifier says to fill in a `char **' you pass with a
3824 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
3825 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
3826 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
3828 * `mkstemp' is a traditional Unix function to atomically create and open a
3829 uniquely-named temporary file.
3833 * The standard location for the file that says what the local timezone is
3834 has changed again. It is now `/usr/local/etc/localtime' (or more
3835 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
3837 * The distribution no longer contains any files with names longer than 14
3840 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
3841 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
3843 * New functions `ttyslot' and `syslog' from 4.4 BSD.
3847 * The configuration process has changed quite a bit. The `configure'
3848 script is now used just like the configuration scripts for other GNU
3849 packages. The `sysdeps' directory hierarchy is much rearranged.
3850 The file `INSTALL' explains the new scheme in detail.
3852 * The header files no longer need to be processed into ANSI C and
3853 traditional C versions. There is just one set of files to install, and
3854 it will work with ANSI or old C compilers (including `gcc -traditional').
3856 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
3857 MIPS DECStation running Ultrix 4.
3859 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
3860 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
3861 made itself into a shared library.
3863 * Yet further improved support for the i386, running 4.3 BSD-like systems
3864 (such as Mach 3 with the Unix single-server), or System V.
3866 * New function `strncasecmp' to do case-insensitive string comparison
3867 with limited length.
3869 * New function `strsep' is a reentrant alternative to `strtok'.
3871 * New functions `scandir' and `alphasort' for searching directories.
3873 * New function `setenv' is a better interface to `putenv'.
3875 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
3876 function for traversing a directory tree.
3878 * The GNU obstack package is now also part of the C library.
3879 The new function `open_obstack_stream' creates a stdio stream that
3880 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
3881 formatted output directly to an obstack.
3883 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
3884 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
3886 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
3888 * The new functions `strfry' and `memfrob' do mysterious and wonderful
3889 things to your strings.
3891 * There are some new test programs: test-fseek, testmb, and testrand.
3893 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
3894 These ports are not finished, but are a good starting place for really
3895 supporting those systems.
3897 * `/etc/localtime' is now the standard location for the file that says what
3898 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
3899 This follows the general principle that `/etc' is the place for all local
3900 configuration files.
3902 * The C library header files now use `extern "C"' when used by the C++
3903 compiler, so the C library should now work with C++ code.
3905 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
3906 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
3909 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
3910 function which uses the `merge sort' algorithm, and is said to be
3911 significantly faster than the old GNU `qsort' function. Merge sort is
3912 now the standard `qsort' function. The new algorithm can require a lot
3913 of temporary storage; so, the old sorting function is called when the
3914 required storage is not available.
3916 * The C library now includes Michael Glad's Ultra Fast Crypt, which
3917 provides the Unix `crypt' function, plus some other entry points.
3919 * The code and header files taken from 4.4 BSD have been updated with the
3920 latest files released from Berkeley.
3922 ----------------------------------------------------------------------
3923 Copyright information:
3925 Copyright (C) 1992-2017 Free Software Foundation, Inc.
3927 Permission is granted to anyone to make or distribute verbatim copies
3928 of this document as received, in any medium, provided that the
3929 copyright notice and this permission notice are preserved,
3930 thus giving the recipient permission to redistribute in turn.
3932 Permission is granted to distribute modified versions
3933 of this document, or of portions of it,
3934 under the above conditions, provided also that they
3935 carry prominent notices stating who last changed them.
3938 version-control: never