1 GNU C Library NEWS -- history of user-visible changes.
2 Copyright (C) 1992-2018 Free Software Foundation, Inc.
3 See the end for copying conditions.
5 Please send GNU C library bug reports via <https://sourceware.org/bugzilla/>
6 using `glibc' in the "product" field.
12 * The GNU C Library can now be compiled with support for building static
13 PIE executables (See --enable-static-pie in INSTALL). These static PIE
14 exectuables are like static executables but can be loaded at any address
15 and provide additional security hardening benefits at the cost of some
16 memory and performance. When the library is built with --enable-static-pie
17 the resulting libc.a is usable with GCC 8 and above to create static PIE
18 executables using the GCC option '-static-pie'. This feature is currently
19 supported on i386, x86_64, x32 and aarch64.
21 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
22 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
25 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
27 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
29 * In order to support faster and safer process termination the malloc API
30 family of functions will no longer print a failure address and stack
31 backtrace after detecting heap corruption. The goal is to minimize the
32 amount of work done after corruption is detected and to avoid potential
33 security issues in continued process execution. Reducing shutdown time
34 leads to lower overall process restart latency, so there is benefit both
35 from a security and performance perspective.
37 * The abort function terminates the process immediately, without flushing
38 stdio streams. Previous glibc versions used to flush streams, resulting
39 in deadlocks and further data corruption. This change also affects
40 process aborts as the result of assertion failures.
42 * On platforms where long double has the IEEE binary128 format (aarch64,
43 alpha, mips64, s390 and sparc), the math library now implements _Float128
44 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015. These
45 are the same interfaces added in version 2.26 for some platforms where
46 this format is supported but is not the format of long double.
48 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
49 mips64, powerpc64le, s390, sparc and x86_64), the math library now
50 implements interfaces for that type, as defined by ISO/IEC TS
51 18661-3:2015. These are corresponding interfaces to those supported for
54 * The math library now implements interfaces for the _Float32, _Float64 and
55 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
56 corresponding interfaces to those supported for _Float128.
58 * glibc now implements the memfd_create and mlock2 functions on Linux.
60 * Support for memory protection keys was added. The <sys/mman.h> header now
61 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
64 * The copy_file_range function was added.
66 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
68 * The ldconfig utility now processes `include' directives using the C/POSIX
69 collation ordering. Previous glibc versions used locale-specific
70 ordering, the change might break systems that relied on that.
72 * Support for two grammatical forms of month names has been added.
73 In a call to strftime, the "%B" and "%b" format specifiers will now
74 produce the grammatical form required when the month is used as part
75 of a complete date. New "%OB" and "%Ob" specifiers produce the form
76 required when the month is named by itself. For instance, in Greek
77 and in many Slavic and Baltic languages, "%B" will produce the month
78 in genitive case, and "%OB" will produce the month in nominative case.
80 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
81 are all valid and will all accept any known form of month
82 name---standalone or complete, abbreviated or full. In a call to
83 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
84 the strings used by "%B" and "%b", respectively. New query
85 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
86 used by "%OB" and "%Ob", respectively.
88 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
89 define the strings for %OB and %Ob, respectively; these have the
90 same syntax as "mon" and "ab_mon".
92 This feature is currently a GNU extension, but it is expected to
93 be added to the next revision of POSIX, and it is also already
94 available on some BSD-derived operating systems.
96 Deprecated and removed features, and other changes affecting compatibility:
98 * Support for statically linked applications which call dlopen is deprecated
99 and will be removed in a future version of glibc. Applications which call
100 dlopen need to be linked dynamically instead.
102 * Support for old programs which use internal stdio data structures and
103 functions is deprecated. This includes programs which use the C++ streams
104 provided by libstdc++ in GCC 2.95. Programs which use the internal
105 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
106 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
107 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
108 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
109 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
110 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
111 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
112 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
113 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
114 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
115 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
116 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
117 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
118 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
119 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
120 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
121 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
122 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
123 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
124 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
125 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
126 future version of glibc. Unlike other symbol removals, these old
127 applications will not be supported using compatibility symbols.
129 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
130 defined by <sys/ptrace.h>.
132 * libm no longer supports SVID error handling (calling a user-provided
133 matherr function on error) or the _LIB_VERSION variable to control error
134 handling. (SVID error handling and the _LIB_VERSION variable still work
135 for binaries linked against older versions of the GNU C Library.) The
136 libieee.a library is no longer provided. math.h no longer defines struct
137 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
138 TLOSS, PLOSS and HUGE.
140 * The libm functions pow10, pow10f and pow10l are no longer supported for
141 new programs. Programs should use the standard names exp10, exp10f and
142 exp10l for these functions instead.
144 * The mcontext_t type is no longer the same as struct sigcontext. On
145 platforms where it was previously the same, this changes the C++ name
146 mangling for interfaces involving this type.
148 * The add-ons mechanism for building additional packages at the same time as
149 glibc has been removed. The --enable-add-ons configure option is now
152 * The --without-fp configure option is now ignored. Whether hardware
153 floating-point instructions are used is now configured based on whether
154 the compiler used at configure time (without any options implied by a
155 --with-cpu= configure option) uses such instructions.
157 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
158 the specified string can be parsed as a domain name.
160 * In the malloc_info output, the <heap> element may contain another <aspace>
161 element, "subheaps", which contains the number of sub-heaps.
163 * The libresolv function p_secstodate is no longer supported for new
166 * The tilepro-*-linux-gnu configuration is no longer supported.
168 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
169 and will be removed in a future release. Software that is still using
170 either header should be updated to use standard <stdio.h> interfaces
173 libio.h was originally the header for a set of supported GNU extensions,
174 but they have not been maintained as such in many years, they are now
175 standing in the way of improvements to stdio, and we don't think there are
176 any remaining external users. _G_config.h was never intended for public
177 use, but predates the bits convention.
179 Changes to build and runtime requirements:
181 * bison version 2.7 or later is required to generate code in the 'intl'
184 Security related changes:
186 CVE-2009-5064: The ldd script would sometimes run the program under
187 examination directly, without preventing code execution through the
188 dynamic linker. (The glibc project disputes that this is a security
189 vulnerability; only trusted binaries must be examined using the ldd
192 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
193 suffered from a one-byte overflow during ~ operator processing (either
194 on the stack or the heap, depending on the length of the user name).
195 Reported by Tim Rühsen.
197 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
198 would sometimes fail to free memory allocated during ~ operator
199 processing, leading to a memory leak and, potentially, to a denial
202 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
203 without GLOB_NOESCAPE, could write past the end of a buffer while
204 unescaping user names. Reported by Tim Rühsen.
206 CVE-2017-17426: The malloc function, when called with an object size near
207 the value SIZE_MAX, would return a pointer to a buffer which is too small,
208 instead of NULL. This was a regression introduced with the new malloc
209 thread cache in glibc 2.26. Reported by Iain Buclaw.
211 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
212 to the allocation of too much memory. (This is not a security bug per se,
213 it is mentioned here only because of the CVE assignment.) Reported by
216 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
217 of the number of search path components. (This is not a security
218 vulnerability per se because no trust boundary is crossed if the fix for
219 CVE-2017-1000366 has been applied, but it is mentioned here only because
220 of the CVE assignment.) Reported by Qualys.
222 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
223 for AT_SECURE or SUID binaries could be used to load libraries from the
226 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
227 succeeds without returning an absolute path due to unexpected behaviour
228 of the Linux kernel getcwd syscall. Reported by halfdog.
230 The following bugs are resolved with this release:
232 [The release manager will add the list generated by
233 scripts/list-fixed-bugs.py just before the release.]
240 * A per-thread cache has been added to malloc. Access to the cache requires
241 no locks and therefore significantly accelerates the fast path to allocate
242 and free small amounts of memory. Refilling an empty cache requires locking
243 the underlying arena. Performance measurements show significant gains in a
244 wide variety of user workloads. Workloads were captured using a special
245 instrumented malloc and analyzed with a malloc simulator. Contributed by
246 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
248 * Unicode 10.0.0 Support: Character encoding, character type info, and
249 transliteration tables are all updated to Unicode 10.0.0, using
250 generator scripts contributed by Mike FABIAN (Red Hat).
251 These updates cause user visible changes, especially the changes in
252 wcwidth for many emoji characters cause problems when emoji sequences
253 are rendered with pango, see for example:
254 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
256 * Collation of Hungarian has been overhauled and is now consistent with "The
257 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
260 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
262 - The GNU C Library will now detect when /etc/resolv.conf has been
263 modified and reload the changed configuration. The new resolver option
264 “no-reload” (RES_NORELOAD) disables this behavior.
266 - The GNU C Library now supports an arbitrary number of search domains
267 (configured using the “search” directive in /etc/resolv.conf);
268 previously, there was a hard limit of six domains. For backward
269 compatibility, applications that directly modify the ‘_res’ global
270 object are still limited to six search domains.
272 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
273 Library will now randomly pick a name server from the configuration as a
274 starting point. (Previously, the second name server was always used.)
276 * The tunables feature is now enabled by default. This allows users to tweak
277 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
279 * New function reallocarray, which resizes an allocated block (like realloc)
280 to the product of two sizes, with a guaranteed clean failure upon integer
281 overflow in the multiplication. Originally from OpenBSD, contributed by
282 Dennis Wölfing and Rüdiger Sonderfeld.
284 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
285 These are extended versions of preadv and pwritev, respectively, taking an
286 additional flags argument. The set of supported flags depends on the
287 running kernel; full support currently requires kernel 4.7 or later.
289 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
290 create a new session ID for the spawned process. This feature is
291 scheduled to be added to the next major revision of POSIX; for the time
292 being, it is available under _GNU_SOURCE.
294 * errno.h is now safe to use from C-preprocessed assembly language on all
295 supported operating systems. In this context, it will only define the
296 Exxxx constants, as preprocessor macros expanding to integer literals.
298 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
299 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
300 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
301 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
303 To compile programs that use this feature, the compiler must support
304 128-bit floating point with the type name _Float128 (as defined by TS
305 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
306 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
307 must be defined to make the new interfaces visible.
309 The new functions and macros correspond to those present for other
310 floating-point types (except for a few obsolescent interfaces not
311 supported for the new type), with F128 or f128 suffixes; for example,
312 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
313 printf or scanf formats for the new type; the strfromf128 and strtof128
314 interfaces should be used instead.
316 Deprecated and removed features, and other changes affecting compatibility:
318 * The synchronization that pthread_spin_unlock performs has been changed to
319 now be equivalent to a C11 atomic store with release memory order to the
320 spin lock's memory location. Previously, several (but not all)
321 architectures used stronger synchronization (e.g., containing what is
322 often called a full barrier). This change can improve performance, but
323 may affect odd fringe uses of spin locks that depend on the previous
324 behavior (e.g., using spin locks as atomic variables to try to implement
325 Dekker's mutual exclusion algorithm).
327 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
330 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
331 will only be built and installed when the GNU C Library is configured with
332 --enable-obsolete-rpc. This allows alternative RPC implementations, such
333 as TIRPC or rpcsvc-proto, to be used.
335 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
336 libnss_compat, are deprecated, and will not be built or installed by
339 The NIS(+) support library, libnsl, is also deprecated. By default, a
340 compatibility shared library will be built and installed, but not headers
341 or development libraries. Only a few NIS-related programs require this
342 library. (In particular, the GNU C Library has never required programs
343 that use 'gethostbyname' to be linked with libnsl.)
345 Replacement implementations based on TIRPC, which additionally support
346 IPv6, are available from <https://github.com/thkukuk/>. The configure
347 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
348 name service modules, to be built and installed.
350 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
351 support is enabled, the configured recursive resolver must support EDNS.
352 (Responding to EDNS-enabled queries with responses which are not
353 EDNS-enabled is fine, but FORMERR responses are not.)
355 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
356 servers have not supported this opcode for a long time.
358 * The _res_opcodes variable has been removed from libresolv. It had been
359 exported by accident.
361 * <string.h> no longer includes inline versions of any string functions,
362 as this kind of optimization is better done by the compiler. The macros
363 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
365 * The nonstandard header <xlocale.h> has been removed. Most programs should
366 use <locale.h> instead. If you have a specific need for the definition of
367 locale_t with no other declarations, please contact
368 libc-alpha@sourceware.org and explain.
370 * The obsolete header <sys/ultrasound.h> has been removed.
372 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
374 * The obsolete function cfree has been removed. Applications should use
377 * The stack_t type no longer has the name struct sigaltstack. This changes
378 the C++ name mangling for interfaces involving this type.
380 * The ucontext_t type no longer has the name struct ucontext. This changes
381 the C++ name mangling for interfaces involving this type.
383 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
384 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
385 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
386 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
387 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
388 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
389 fpregset_t type no longer has the name struct fpu. This changes the C++
390 name mangling for interfaces involving those types.
392 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
393 synced with the kernel:
395 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
396 are not supported on this architecture and have been removed.
398 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
399 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
400 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
402 Programs that assume the GET/SETREGS ptrace requests are universally
403 available will now fail to build, instead of malfunctioning at runtime.
405 Changes to build and runtime requirements:
407 * Linux kernel 3.2 or later is required at runtime, on all architectures
408 supported by that kernel. (This is a change from version 2.25 only for
411 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
413 * On most architectures, GCC 4.9 or later is required to build the GNU C
414 Library. On powerpc64le, GCC 6.2 or later is required.
416 Older GCC versions and non-GNU compilers are still supported when
417 compiling programs that use the GNU C Library. (We do not know exactly
418 how old, and some GNU extensions to C may be _de facto_ required. If you
419 are interested in helping us make this statement less vague, please
420 contact libc-alpha@sourceware.org.)
422 Security related changes:
424 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
425 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
427 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
428 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
430 * Avoid printing a backtrace from the __stack_chk_fail function since it is
431 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
434 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
435 fixed (CVE-2017-12133).
437 The following bugs are resolved with this release:
439 [984] network: Respond to changed resolv.conf in gethostbyname
440 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
441 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
443 [12189] libc: __stack_chk_fail should not attempt a backtrace
445 [14096] time: Race condition on timezone/tst-timezone.out
446 [14172] localedata: az_IR: new locale
447 [14995] build: glibc fails to build if gold is the default linker, even if
449 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
450 [16637] network: inet_pton function is accepting IPv6 with bad format
451 [16640] string: string/strtok.c: undefined behaviour inconsistent between
452 x86 and other generic code
453 [16875] localedata: ko_KR: fix lang_name
454 [17225] localedata: ar_SY: localized month names for May and June are
456 [17297] localedata: da_DK: wrong date_fmt string
457 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
458 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
459 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
460 [19066] localedata: ar_SA abbreviated day and month names are in English
461 [19569] network: resolv: Support an arbitrary number of search domains
462 [19570] network: Implement random DNS server selection in the stub
464 [19838] locale: localedef fails on PA-RISC
465 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
466 order of 0D36 and 0D37
467 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
469 [20098] libc: FAIL: debug/backtrace-tst on hppa
470 [20257] network: sunrpc: clntudp_call does not enforce timeout when
472 [20275] localedata: locale day/abday/mon/abmon should not have trailing
474 [20313] localedata: Update locale data to Unicode 9.0
475 [20424] manual: Document how to provide a malloc replacement
476 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
478 [20686] locale: Add el_GR@euro to SUPPORTED.
479 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
480 failures consistently
481 [21015] dynamic-link: Document and fix --enable-bind-now
482 [21016] nptl: pthread_cond support is broken on hppa
483 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
484 frame-pointer on i386
485 [21049] libc: segfault in longjmp_chk() due to clobbered processor
487 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
488 [21088] libc: Build fails with --enable-static-nss
489 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
490 generic c code is used
491 [21109] libc: Tunables broken on big-endian
492 [21112] math: powf has large ulp errors with base close to 1 and exponent
494 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
496 [21120] malloc: glibc malloc is incompatible with GCC 7
497 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
499 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
500 (-0.0) when linking with -lieee
501 [21171] math: log10, log2 and lgamma return incorrect results
502 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
503 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
504 [21207] localedata: ce_RU: update weekdays from CLDR
505 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
506 [21217] localedata: Update months from CLDR-31
507 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
508 [21243] libc: support_delete_temp_file should issue warning for failed
510 [21244] libc: support resolv_test_start() socket fd close should be
512 [21253] libc: localedef randomly segfaults when using -fstack-check due to
513 new posix_spawn implementation
514 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
515 leads to lower CPU frequency
516 [21259] libc: [alpha] termios.h missing IXANY for POSIX
517 [21261] libc: [sparc64] bits/setjmp.h namespace
518 [21267] network: [mips] bits/socket.h IOC* namespace
519 [21268] libc: [alpha] termios.h NL2, NL3 namespace
520 [21270] libc: mmap64 silently truncates large offset values
521 [21275] libc: posix_spawn always crashes on ia64 now
522 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
523 [21280] math: [powerpc] logbl for POWER7 return incorrect results
524 [21289] libc: Incorrect declaration for 32-bit platforms with
525 _FILE_OFFSET_BITS=64 causes build error
526 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
527 not support gethostbyname4_r
528 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
530 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
532 [21340] libc: Support POSIX_SPAWN_SETSID
533 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
534 [21359] network: ns_name_pack needs additional byte in destination buffer
535 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
536 against fragmentation attacks (CVE-2017-12132)
537 [21369] network: resolv: Remove EDNS fallback
538 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
540 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
541 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
542 [21393] stdio: Missing dup3 error check in freopen, freopen64
543 [21396] libc: Use AVX2 memcpy/memset on Skylake server
544 [21399] localedata: Bad description for U00EC in
545 localedata/charmaps/CP1254
546 [21411] malloc: realloc documentation error
547 [21426] network: sys/socket.h uio.h namespace
548 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
549 [21445] libc: signal.h bsd_signal namespace
550 [21455] network: Network headers stdint.h namespace
551 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
553 [21475] network: resolv: Overlong search path is truncated mid-label
554 [21511] libc: sigstack namespace
555 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
556 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
557 bits/syscall.h: No such file or directory
558 [21517] libc: struct sigaltstack namespace
559 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
560 [21533] localedata: Update locale data to Unicode 10.0
562 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
563 junk at end of line, first unrecognized character is `@'
564 [21538] libc: SIG_HOLD missing for XPG4
565 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
566 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
567 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
568 [21543] libc: sigevent namespace
569 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
570 size for general purpose registers in mcontext_t structure
571 [21550] libc: sigwait namespace
572 [21552] libc: XPG4 bsd_signal namespace
573 [21554] libc: sigpause namespace
574 [21560] libc: sys/wait.h signal.h namespace
575 [21561] libc: waitid namespace
576 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
577 '/usr/include/stdlib.h': Permission denied
578 [21575] libc: sys/wait.h missing struct rusage definition
579 [21584] libc: sigaltstack etc namespace
580 [21597] libc: siginterrupt namespace
581 [21607] math: hppa: FAIL: math/test-tgmath
582 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
584 [21622] libc: [tile] missing SA_* for POSIX.1:2008
585 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
586 stack and heap (CVE-2017-1000366)
587 [21625] libc: wait3 namespace
588 [21654] nss: Incorrect pointer alignment in NSS group merge result
590 [21657] network: Parse interface zone id for node-local multicast
591 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
593 [21666] libc: .symver is used on common symbol
594 [21668] network: resolv: res_init cross-thread broadcast introduces race
596 [21687] math: tgmath.h totalorder, totalordermag return type
597 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
599 [21696] libc: Incorrect assumption of of __cpu_mask in
600 posix/sched_cpucount.c
601 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
602 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
604 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
605 leading to relocation crash
606 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
607 `resolv_conf_matches (resp, conf)' failed.
608 [21710] localedata: Added Samoan language locale for Samoa
609 [21711] localedata: Pashto yesstr/nostr locale are missing
610 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
611 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
612 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
613 [21723] localedata: yesstr/nostr missing for Chinese language locale
614 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
616 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
618 [21728] localedata: New Locale for Tongan language
619 [21729] localedata: incorrect LC_NAME fields for hi_IN
620 [21733] localedata: yesstr and nostr are missing for zh_HK
621 [21734] localedata: Missing yesstr and nostr are for kw_GB
622 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
623 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
624 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
625 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
627 [21744] libc: Tests failing on --enable-tunables --enable-stack-
629 [21749] localedata: Wrong abbreviated day name (“abday”) for
631 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
632 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
633 [21759] localedata: missing yesstr and nostr for Tigrinya
634 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
635 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
636 [21767] localedata: Missing Bislama locales
637 [21768] localedata: Missing yesstr and nostr for aa_ET
638 [21770] localedata: Missing Field in li_NL
639 [21778] nptl: Robust mutex may deadlock
640 [21779] libc: MicroBlaze segfaults when loading libpthread
641 [21783] localedata: Fix int_select international_call_prefixes
642 [21784] localedata: Inconsistency in country_isbn
643 [21788] localedata: Missing Country Postal Abbreviations
644 [21794] localedata: Added-country_isbn-for-Italy
645 [21795] localedata: Add/Fix country_isbn for France
646 [21796] localedata: Added country_isbn for Republic of Korea
647 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
648 [21799] localedata: Added int_select international_call_prefixes
649 [21801] localedata: Added int_select international_call_prefixes
650 [21804] nptl: Double semicolon in thread-shared-types.h
651 [21807] localedata: LC_ADDRESS fix for pap_CW
652 [21808] localedata: Fix LC_ADDRESS for pap_AW
653 [21821] localedata: Added country_name in mai_IN
654 [21822] localedata: Fix LC_TIME for mai_IN
655 [21823] localedata: missing yesstr, nostr for sa_IN
656 [21825] localedata: Fix name_mrs for mag_IN
657 [21828] localedata: 2.26 changelog should mention user visible changes
659 [21835] localedata: Added Maithili language locale for Nepal
660 [21838] localedata: Removed redundant data for the_NP
661 [21839] localedata: Fix LC_MONETARY for ta_LK
662 [21844] localedata: Fix Latin characters and Months Sequence.
663 [21848] localedata: Fix mai_NP Title Name
668 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
669 24731-2:2010, is supported to enable declarations of functions from that
670 TR. Note that not all functions from that TR are supported by the GNU C
673 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
674 18661-1:2014, is supported to enable declarations of functions and macros
675 from that TS. Note that not all features from that TS are supported by
678 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
679 18661-4:2015, is supported to enable declarations of functions and macros
680 from that TS. Note that most features from that TS are not supported by
683 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
684 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
685 Since the GNU C Library defaults to a much newer revision of POSIX, this
686 will only affect programs that specifically request an old conformance
687 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
688 see a change in the visible declarations, but a program compiled with
689 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
692 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
693 defined by all multithreaded code, but glibc has not required this for
696 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
697 means that in a future release, the macros “major”, “minor”, and “makedev”
698 will only be available from <sys/sysmacros.h>.
700 These macros are not part of POSIX nor XSI, and their names frequently
701 collide with user code; see for instance glibc bug 19239 and Red Hat bug
702 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
703 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
706 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
707 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
708 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
710 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
711 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
712 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
713 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
714 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
715 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
716 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
717 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
718 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
719 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
720 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
723 * New <math.h> features are added from TS 18661-1:2014:
725 - Signaling NaN macros: SNANF, SNAN, SNANL.
727 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
728 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
729 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
731 - llogb functions: the llogb, llogbf and llogbl functions, and the
732 FP_LLOGB0 and FP_LLOGBNAN macros.
734 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
737 - Comparison macros: iseqsig.
739 - Classification macros: iscanonical, issubnormal, iszero.
741 - Total order functions: totalorder, totalorderf, totalorderl,
742 totalordermag, totalordermagf, totalordermagl.
744 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
746 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
747 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
749 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
750 are added to libc. They convert a floating-point number into string.
752 * Most of glibc can now be built with the stack smashing protector enabled.
753 It is recommended to build glibc with --enable-stack-protector=strong.
754 Implemented by Nick Alcock (Oracle).
756 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
757 intended to be used instead of memset() to erase sensitive data after use;
758 the compiler will not optimize out calls to explicit_bzero even if they
759 are "unnecessary" (in the sense that no _correct_ program can observe the
760 effects of the memory clear).
762 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
763 to float instead of double. This does not affect the ABI of any libraries
764 that are part of the GNU C Library, but may affect the ABI of other
765 libraries that use this type in their interfaces.
767 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
768 float_t and double_t types are now defined to long double instead of float
769 and double. These options are not the default, and this does not affect
770 the ABI of any libraries that are part of the GNU C Library, but it may
771 affect the ABI of other libraries that use this type in their interfaces,
772 if they are compiled or used with those options.
774 * The getentropy and getrandom functions, and the <sys/random.h> header file
777 * The buffer size for byte-oriented stdio streams is now limited to 8192
778 bytes by default. Previously, on Linux, the default buffer size on most
779 file systems was 4096 bytes (and thus remains unchanged), except on
780 network file systems, where the buffer size was unpredictable and could be
781 as large as several megabytes.
783 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
784 for the Linux quota interface which predates kernel version 2.4.22 has
787 * The malloc_get_state and malloc_set_state functions have been removed.
788 Already-existing binaries that dynamically link to these functions will
789 get a hidden implementation in which malloc_get_state is a stub. As far
790 as we know, these functions are used only by GNU Emacs and this change
791 will not adversely affect already-built Emacs executables. Any undumped
792 Emacs executables, which normally exist only during an Emacs build, should
793 be rebuilt by re-running “./configure; make” in the Emacs build tree.
795 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
796 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
797 “no-ip6-dotint” had already been the default, and support for the
798 “ip6-dotint” option was removed from the Internet in 2006.
800 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
801 flag from <resolv.h> have been removed. The option relied on a
802 backwards-incompatible DNS extension which was never deployed on the
805 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
806 RES_BLAST defined in the <resolv.h> header file have been deprecated.
807 They were already unimplemented.
809 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
810 _res.flags are deprecated. The flag was standardized in RFC 2133, but
811 removed again from the IETF name lookup interface specification in RFC
812 2553. Applications should use getaddrinfo instead.
814 * DNSSEC-related declarations and definitions have been removed from the
815 <arpa/nameser.h> header file, and libresolv will no longer attempt to
816 decode the data part of DNSSEC record types. Previous versions of glibc
817 only implemented minimal support for the previous version of DNSSEC, which
818 is incompatible with the currently deployed version.
820 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
821 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
822 <arpa/nameser.h> header file because the distinction between RR types and
823 meta-RR types is not officially standardized, subject to revision, and
824 thus not suitable for encoding in a macro.
826 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
827 and rhook members of the res_state type in <resolv.h> have been removed.
828 The glibc stub resolver did not support these hooks, but the header file
829 did not reflect that.
831 * For multi-arch support it is recommended to use a GCC which has
832 been built with support for GNU indirect functions. This ensures
833 that correct debugging information is generated for functions
834 selected by IFUNC resolvers. This support can either be enabled by
835 configuring GCC with '--enable-gnu-indirect-function', or by
836 enabling it by default by setting 'default_gnu_indirect_function'
837 variable for a particular architecture in the GCC source file
840 * GDB pretty printers have been added for mutex and condition variable
841 structures in POSIX Threads. When installed and loaded in gdb these pretty
842 printers show various pthread variables in human-readable form when read
843 using the 'print' or 'display' commands in gdb.
845 * Tunables feature added to allow tweaking of the runtime for an application
846 program. This feature can be enabled with the '--enable-tunables' configure
847 flag. The GNU C Library manual has details on usage and README.tunables has
848 instructions on adding new tunables to the library.
850 * A new version of condition variables functions have been implemented in
851 the NPTL implementation of POSIX Threads to provide stronger ordering
854 * A new version of pthread_rwlock functions have been implemented to use a more
855 scalable algorithm primarily through not using a critical section anymore to
858 Security related changes:
860 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
861 have been created with makecontext could fail to terminate due to a
862 missing .cantunwind annotation. This has been observed to lead to a hang
863 (denial of service) in some Go applications compiled with gccgo. Reported
864 by Andreas Schwab. (CVE-2016-6323)
866 * The DNS stub resolver functions would crash due to a NULL pointer
867 dereference when processing a query with a valid DNS question type which
868 was used internally in the implementation. The stub resolver now uses a
869 question type which is outside the range of valid question type values.
872 The following bugs are resolved with this release:
874 [4099] stdio: Overly agressive caching by stream i/o functions.
875 [7065] build: Support building glibc with -fstack-protector or -fstack-
877 [9842] localedata: en_CA: incorrect date format
878 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
879 before it started waiting
880 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
881 [15765] nptl: sem_open is wrongly a cancellation point
882 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
883 [16458] libc: endian.h and netinet/in.h byte order macros return values of
885 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
886 library linked with pthread
887 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
889 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
890 [17252] libc: getrandom and getentropy syscall
891 [17730] malloc: thread-local storage is sometimes improperly free()'d
892 after being __libc_memalign()'d
893 [18241] stdio: failed fseek on memstream does not set errno and can fail
895 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
897 [18463] nptl: pthread_cond_broadcast issue when surrounded by
898 PTHREAD_PRIO_INHERIT mutex on ARM
899 [18784] network: res_query and related function crash for special record
900 type queries (CVE-2015-5180)
901 [19380] math: strtod does not raise "inexact"
902 [19387] string: Integer overflow in memchr
903 [19390] string: Integer overflow in strncat
904 [19398] build: linknamespace tests fail with massively parallel build
905 [19402] nptl: Deadlock with robust shared mutex and asynchronous
907 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
908 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
909 [19514] libc: [PATCH] Fix spelling errors in spelling
910 "implement"/"implementation" in several places
911 [19582] network: Deprecate RES_USE_INET6
912 [19673] manual: clog10 docs appear to be erroneous
913 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
914 causes a segmentation fault
915 [19826] libc: invalid pointer returned from __tls_get_addr with static
917 [20016] network: resolv: Remove hooks support from the API
918 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
920 [20033] math: [x86_64] vectorized math function don't call the __finite
922 [20116] nptl: use after free in pthread_create
923 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
924 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
926 [20311] nptl: please install proc_service.h
927 [20366] libc: Compilation errors in installed headers in strict-compliance
929 [20370] malloc: malloc: Arena free list management is still racy
930 (incorrect fix in bug 19243)
931 [20386] libc: assert (X = 0) does not result in GCC warning
932 [20432] malloc: malloc: Minimize interface required for interposition
933 [20435] libc: Missing unwind info in __startcontext causes infinite loop
934 in _Unwind_Backtrace (CVE-2016-6323)
935 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
936 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
938 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
939 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
940 [20477] network: resolv: incorrect double-checked locking related to
942 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
944 [20495] math: x86_64 performance degradation due to AVX/SSE transition
946 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
947 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
948 cause transition penalty
949 [20517] math: sparcv9 missing fdiml compat symbol
950 [20524] manual: strverscmp is inconsistent
951 [20525] libc: <sys/quota.h> should be based on kernel headers
952 [20539] math: GCC 7 -static -lm fails to link at -O3
953 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
955 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
957 [20591] network: Remove obsolete DNSSEC support
958 [20592] network: DNS resource record type classification macros in
959 <arpa/nameser.h> are incorrect
960 [20593] network: Update DNS RR type definitions
961 [20611] network: getaddrinfo accepts invalid numeric scope IDs
962 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
964 [20629] network: libresolv: Remove support for bitstring labels
966 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
967 assertion: "maxidx >= 2"
968 [20660] math: [arm] Use VSQRT
969 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
970 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
971 [20689] libc: Test for FMA should also check for AVX.
972 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
973 [20715] math: iszero macro breaks existing code
974 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
975 [20728] libc: powerpc: Missing TOC stub in clone
976 [20729] build: glibc-2.24 fails to build for i486 with -Os
977 [20750] build: Build fails with default PIE enabled gcc-6.x
978 [20768] math: [alpha] sqrt fegetenv namespace
979 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
981 [20787] math: float_t is defined as float incorrectly on x86_64 even if
982 __FLT_EVAL_METHOD__ is 2
983 [20790] network: rpcgen buffer overrun in get_prog_declaration
984 [20822] nptl: race condition in __lll_unlock_elision on powerpc
985 [20829] libc: crypt snprintf namespace
986 [20847] libc: tst-vfork3 failure
987 [20855] math: Default bits/mathdef.h has inappropriate float_t
988 [20859] math: [sh4] FP_ILOGB0 invalid
989 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
990 U+20AC), not same as GBK
991 [20915] dynamic-link: global-dynamic TLS broken on aarch64
992 [20916] math: pow handling of sNaN arguments
993 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
994 [20919] math: Bad pow (qNaN, 0) result with -lieee
995 [20940] math: hypot sNaN handling
996 [20947] math: fmax, fmin sNaN handling
997 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
998 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1000 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1001 [20973] nptl: robust mutexes: Lost wake-ups
1002 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1003 [20978] nis: strlen on null pointer in nss_nisplus
1004 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1006 [21014] string: i686 memchr overflows internal pointer check
1007 [21019] libc: [mips] n32 lseek incorrect on overflow
1008 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1009 [21026] libc: [MIPS] readahead syscall is broken on n64
1010 [21028] math: Fallback fesetexceptflag should always succeed
1011 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1012 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1014 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1015 [21061] librt: [microblaze] librt lost clock_* exports
1016 [21073] libc: tunables: insecure environment variables passed to
1017 subprocesses with AT_SECURE
1018 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1022 * The minimum Linux kernel version that this version of the GNU C Library
1023 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1024 kernel version 2.6.32 or later suffices (on architectures that already
1025 required kernel versions more recent than 3.2, those requirements remain
1026 unchanged). Linux 3.2 or later kernel headers are required on all
1029 * The pap_AN locale has been deleted. This has been deprecated for a long
1030 time. It has been replaced by pap_AW & pap_CW, both of which have long
1031 been included in previous releases.
1033 * The readdir_r and readdir64_r functions have been deprecated. It is
1034 recommended to use readdir and readdir64 instead.
1036 * The type “union wait” has been removed. It was deprecated in the early
1037 1990s and never part of POSIX. Application code should use the int type
1038 instead of “union wait”.
1040 * A new NSS action is added to facilitate large distributed system
1041 administration. The action, MERGE, allows remote user stores like LDAP
1042 to be merged into local user stores like /etc/groups in order to provide
1043 easy to use, updated, and managed sets of merged credentials. The new
1044 action can be used by configuring it in /etc/nsswitch.conf:
1045 group: files [SUCCESS=merge] nis
1046 Implemented by Stephen Gallagher (Red Hat).
1048 * The deprecated __malloc_initialize_hook variable has been removed from the
1051 * The long unused localedef --old-style option has been removed. It hasn't
1052 done anything in over 16 years. Scripts using this option can safely
1055 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1056 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1057 return the next representable value in the direction of positive infinity
1058 and the nextdown functions return the next representable value in the
1059 direction of negative infinity. These are currently enabled as GNU
1062 Security related changes:
1064 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1065 could result in a stack overflow when getnetbyname was called with an
1066 overly long name. (CVE-2016-3075)
1068 * Previously, getaddrinfo copied large amounts of address data to the stack,
1069 even after the fix for CVE-2013-4458 has been applied, potentially
1070 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1071 instead. Reported by Michael Petlan. (CVE-2016-3706)
1073 * The glob function suffered from a stack-based buffer overflow when it was
1074 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1075 Reported by Alexander Cherepanov. (CVE-2016-1234)
1077 * The Sun RPC UDP client could exhaust all available stack space when
1078 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1079 alloca plugin for GCC. (CVE-2016-4429)
1081 * The IPv6 name server management code in libresolv could result in a memory
1082 leak for each thread which is created, performs a failing naming lookup,
1083 and exits. Over time, this could result in a denial of service due to
1084 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1086 The following bugs are resolved with this release:
1088 [1170] localedata: ne_NP: update Nepali locale definition file
1089 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1091 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1092 [6796] math: fdim() does not set errno on overflow
1093 [10354] libc: posix_spawn should use vfork() in more cases than presently
1094 [11213] localedata: localedata: add copyright disclaimer to locale files
1095 [12143] localedata: chr_US: new Cherokee locale
1096 [12450] localedata: sgs_LT: new locale
1097 [12676] localedata: ln_CD: new locale
1098 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1100 [13304] math: fma, fmaf, fmal produce wrong results
1101 [14259] build: --localedir arg to configure is ignored
1102 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1103 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1104 [14934] localedata: es_CL: wrong first weekday chilean locale
1105 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1107 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1109 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1110 [15368] nptl: raise() is not async-signal-safe
1111 [15479] math: ceil, floor, round and trunc raise inexact exception
1112 [15578] localedata: kk_KZ: various updates
1113 [16003] localedata: pap_AN: punt old locale
1114 [16137] localedata: iw_IL: punt old locale
1115 [16190] localedata: eo: new esperanto locale
1116 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1117 [16742] malloc: race condition: pthread_atfork() called before first
1118 malloc() results in unexpected locking behaviour/deadlocks
1119 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1121 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1122 [17565] localedata: pt_PT: wrong (work-)week start
1123 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1124 [17950] build: Build fails with -msse
1125 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1126 [18433] libc: posix_spawn does not return correctly upon failure to
1128 [18453] localedata: charmaps/IBM875: incorrect codes
1129 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1131 [18896] localedata: he_IL: improvements for currency
1132 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1133 Romanian locale data
1134 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1136 [19056] libc: Deprecate readdir_r
1137 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1139 [19198] localedata: nl_NL: small improvements for Dutch locales
1140 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1141 nameservers (CVE-2016-5417)
1142 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1144 [19400] locale: Language missing in "iso-639.def", trivial fix in
1146 [19431] malloc: Deadlock between fflush, getdelim, and fork
1147 [19505] libc: Incorrect file descriptor validity checks in
1148 posix_spawn_file_actions_add{open,close,dup2}
1149 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1150 when using RTLD_NEXT
1151 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1152 `intl/{gettextP,loadinfo}.h'
1153 [19534] libc: execle, execlp may use malloc
1154 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1155 [19573] network: res_nclose and __res_maybe_init disagree about name
1156 server initialization, breaking Hesiod
1157 [19575] localedata: Status of GB18030 tables
1158 [19581] localedata: sr_* date_fmt string contains additional newline
1159 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1161 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1162 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1163 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1164 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1166 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1168 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1170 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1171 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1172 [19633] locale: strfmon_l applies global locale to number formatting
1173 [19642] network: Memory leak in getnameinfo
1174 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1175 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1177 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1178 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1180 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1181 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1183 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1185 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1187 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1188 pointers and lengths in error-case.
1189 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1190 report errors on UTF-16 surrogates values.
1191 [19755] nscd: nscd assertion failure in gc
1192 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1193 [19759] libc: mempcpy shouldn't be inlined
1194 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1195 [19765] libc: s390 needs an optimized mempcpy
1196 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1197 NAME_MAX limit assumption (CVE-2016-1234)
1198 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1199 [19787] network: Missing and incorrect truncation checks in getnameinfo
1200 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1202 [19791] network: Assertion failure in res_query.c with un-connectable name
1204 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1205 [19822] math: libm.so install clobbers old version
1206 [19825] network: resolv: send_vc can return uninitialized data in second
1207 response to getaddrinfo
1208 [19830] network: nss_dns: should check RDATA length against buffer length
1209 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1210 confronted with A/AAAA records of invalid size
1211 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1212 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1213 [19853] stdio: Printing IBM long double in decimal with high precision is
1215 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1216 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1217 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1219 [19865] network: Assertion failure or memory leak in
1220 _nss_dns_getcanonname_r
1221 [19868] network: nss_dns: netent code does not skip over non-PTR records
1222 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1224 [19881] string: Improve x86-64 memset
1225 [19907] string: Incorrect memcpy tests
1226 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1228 [19925] libc: termios.h XCASE namespace
1229 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1230 [19929] libc: limits.h NL_NMAX namespace
1231 [19931] stdio: Memory leak in vfprintf
1232 [19957] libc: clone(CLONE_VM) access invalid parent memory
1233 [19963] localedata: en_IL: New locale
1234 [19989] stdio: stdio.h cuserid namespace
1235 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1237 [19996] locale: langinfo.h nl_langinfo_l namespace
1238 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1239 [20010] network: getaddrinfo: Stack overflow in hostent translation
1241 [20012] stdio: libio: fmemopen append mode failure
1242 [20014] stdio: stdio.h namespace for pre-threads POSIX
1243 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1244 [20023] libc: fcntl.h timespec namespace
1245 [20024] math: [x86_64] vectorized sincos trashes the stack
1246 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1247 [20041] time: sys/time.h timespec namespace
1248 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1249 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1250 [20051] libc: ttyslot in wrong header under wrong conditions
1251 [20054] libc: gethostname not declared for XPG4
1252 [20055] libc: termios.h missing tcgetsid for XPG4
1253 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1255 [20073] libc: sys/stat.h fchmod namespace
1256 [20074] libc: stdlib.h rand_r namespace
1257 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1258 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1260 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1261 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1263 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1264 [20119] libc: Wrong mask for processors level type from CPUID
1265 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1266 AS not supporting AVX512
1267 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1269 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1270 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1272 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1273 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1274 [20178] libc: posix_spawn{p} should not call exit
1275 [20191] stdio: libio: vtables hardening
1276 [20195] string: FMA4 detection requires CPUID execution with register
1278 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1279 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1281 [20212] math: acos (sNaN) returns sNaN
1282 [20213] math: asin (sNaN) returns sNaN
1283 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1284 [20218] math: [i386] asinhl (sNaN) returns sNaN
1285 [20219] math: [i386] atanhl (sNaN) returns sNaN
1286 [20222] stdio: fopencookie: Mangle function pointers
1287 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1288 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1289 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1291 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1292 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1293 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1294 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1295 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1296 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1297 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1298 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1299 [20240] math: modf (sNaN) returns sNaN
1300 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1301 [20250] math: frexp (sNaN) returns sNaN
1302 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1303 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1305 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1306 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1307 subscript is below array bounds [-Werror=array-bounds]
1308 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1309 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1311 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1312 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1314 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1315 "invalid" exceptions
1316 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1317 [20316] localedata: id_ID: Februari instead of Pebruari
1318 [20327] string: POWER8 strcasecmp returns incorrect result
1319 [20347] math: Failure: Test: j0_downward (0xap+0)
1320 [20348] libc: FAIL: misc/tst-preadvwritev64
1321 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1322 [20350] libc: There is no test for p{read,write}64
1323 [20357] math: Incorrect cos result for 1.5174239687223976
1324 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1328 * Unicode 8.0.0 Support: Character encoding, character type info, and
1329 transliteration tables are all updated to Unicode 8.0.0, using new
1330 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1331 These updates cause user visible changes, such as the fixes for bugs
1332 89, 16061, and 18568.
1334 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1335 kernel-internal CPU set size. This means that requests that change the
1336 CPU affinity which failed before (for example, an all-ones CPU mask) will
1337 now succeed. Applications that need to determine the effective CPU
1338 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1339 setting it because the kernel can adjust it (and the previous size check
1340 would not detect this in the majority of cases).
1342 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1343 the following new symbols are used: fts64_children, fts64_close,
1344 fts64_open, fts64_read and fts64_set.
1346 * getaddrinfo now detects certain invalid responses on an internal netlink
1347 socket. If such responses are received, an affected process will
1348 terminate with an error message of "Unexpected error <number> on netlink
1349 descriptor <number>" or "Unexpected netlink response of size <number> on
1350 descriptor <number>". The most likely cause for these errors is a
1351 multi-threaded application which erroneously closes and reuses the netlink
1352 file descriptor while it is used by getaddrinfo.
1354 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1355 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1356 unnecessary serialization of memory allocation requests across threads.
1357 The defect is now corrected. Users should see a substantial increase in
1358 the concurent throughput of allocation requests for applications which
1359 trigger this bug. Affected applications typically create and
1360 destroy threads frequently. (Bug 19048 was reported and analyzed by
1363 * There is now a --disable-timezone-tools configure option for disabling the
1364 building and installing of the timezone related utilities (zic, zdump, and
1365 tzselect). This is useful for people who build the timezone data and code
1366 independent of the GNU C Library.
1368 * The obsolete header <regexp.h> has been removed. Programs that require
1369 this header must be updated to use <regex.h> instead.
1371 * The obsolete functions bdflush, create_module, get_kernel_syms,
1372 query_module and uselib are no longer available to newly linked binaries;
1373 the header <sys/kdaemon.h> has been removed. These functions and header
1374 were specific to systems using the Linux kernel and could not usefully be
1375 used with the GNU C Library on systems with version 2.6 or later of the
1378 * Optimized string, wcsmbs and memory functions for IBM z13.
1379 Implemented by Stefan Liebler.
1381 * Newly linked programs that define a variable called signgam will no longer
1382 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1383 require signgam to be set by those functions must ensure that they use the
1384 variable provided by the GNU C Library and declared in <math.h>, without
1385 defining their own copy.
1387 * The minimum GCC version that can be used to build this version of the GNU
1388 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1389 still be used to compile programs using the GNU C Library.
1391 Security related changes:
1393 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1394 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1396 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1397 the pointer guard feature. It is always enabled. Previously,
1398 LD_POINTER_GUARD could be used to disable security hardening in binaries
1399 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1402 * An integer overflow in hcreate and hcreate_r could lead to an
1403 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1405 * The catopen function no longer has unbounded stack usage. Reported by
1406 Max. (CVE-2015-8779)
1408 * The nan, nanf and nanl functions no longer have unbounded stack usage
1409 depending on the length of the string passed as an argument to the
1410 functions. Reported by Joseph Myers. (CVE-2014-9761)
1412 * A stack-based buffer overflow was found in libresolv when invoked from
1413 libnss_dns, allowing specially crafted DNS responses to seize control
1414 of execution flow in the DNS client. The buffer overflow occurs in
1415 the functions send_dg (send datagram) and send_vc (send TCP) for the
1416 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1417 family. The use of AF_UNSPEC triggers the low-level resolver code to
1418 send out two parallel queries for A and AAAA. A mismanagement of the
1419 buffers used for those queries could result in the response of a query
1420 writing beyond the alloca allocated buffer created by
1421 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1422 the overflow. Thanks to the Google Security Team and Red Hat for
1423 reporting the security impact of this issue, and Robert Holiday of
1424 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1426 The following bugs are resolved with this release:
1428 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1429 [887] math: Math library function "logb" and "nextafter" inconsistent
1430 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1431 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1432 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1433 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1435 [4404] localedata: German translation of "Alarm clock" is misleading
1436 [6799] math: nextafter() and nexttoward() doen't set errno on
1437 overflow/underflow errors
1438 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1440 [10432] nis: _nss_nis_setnetgrent assertion failure
1441 [11460] libc: fts has no LFS support
1442 [12926] network: getaddrinfo()/make_request() may spin forever
1443 [13065] nptl: Race condition in pthread barriers
1444 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1445 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1447 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1448 [14912] libc: Rename non-installed bits/*.h headers
1449 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1450 [15367] math: Let gcc use __builtin_isinf
1451 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1452 [15421] math: lgamma wrongly sets signgam for ISO C
1453 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1454 FE_INVALID with argument out of range
1455 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1457 [15786] dynamic-link: ifunc resolver functions can smash function
1459 [15918] math: Unnecessary check for equality in hypotf()
1460 [16061] localedata: Review / update transliteration data
1461 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1463 [16141] time: strptime %z offset restriction
1464 [16171] math: drem should be alias of remainder
1465 [16296] math: fegetround is pure?
1466 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1467 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1469 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1470 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1472 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1473 "Exception "Inexact" set" on ppc32
1474 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1475 [16517] math: Missing underflow exception from tanf/tan/tanl
1476 [16519] math: Missing underflow exception from sinhf
1477 [16520] math: Missing underflow exception from tanhf
1478 [16521] math: Missing underflow exception from exp2
1479 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1481 [16734] stdio: fopen calls mmap to allocate its buffer
1482 [16961] math: nan function incorrect handling of bad sequences
1483 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1484 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1485 [16985] locale: localedef: confusing error message when opening output
1487 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1488 [17197] locale: Redundant shift character in iconv conversion output at
1490 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1491 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1492 [17250] dynamic-link: static linking breaks nss loading
1493 (getaddrinfo/getpwnam/etc...)
1494 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1496 [17441] math: isnan() should use __builtin_isnan() in GCC
1497 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1498 (related to lock elision)
1499 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1500 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1501 [17887] time: strptime should be able to parse "+01:00" style timezones
1502 [17905] libc: catopen() Multiple unbounded stack allocations
1504 [18084] libc: backtrace (..., 0) dumps core on x86
1505 [18086] libc: nice() sets errno to 0 on success
1506 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1507 is too large (CVE-2015-8778)
1508 [18251] dynamic-link: SONAME missing when audit modules provides path
1509 [18265] libc: add attributes for wchar string and memory functions
1510 [18370] math: csqrt missing underflows
1511 [18421] libc: [hppa] read-only segment has dynamic relocations
1512 [18472] libc: Obsolete syscall wrappers should be compat symbols
1513 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1514 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1515 [18525] localedata: Remove locale timezone information
1516 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1517 [18568] localedata: Update locale data to Unicode 8.0
1518 [18589] locale: sort-test.sh fails at random
1519 [18595] math: ctan, ctanh missing underflows
1520 [18604] libc: assert macro-expands its argument
1521 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1522 contains a vector instruction exception.
1523 [18611] math: j1, jn missing errno setting on underflow
1524 [18618] localedata: sync Chechen locale definitions with other *_RU
1526 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1527 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1528 [18665] network: In send_dg, the recvfrom function is NOT always using the
1529 buffer size of a newly created buffer (CVE-2015-7547)
1530 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1532 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1534 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1535 [18699] math: tilegx cproj() for various complex infinities does not yield
1537 [18724] libc: Harden put*ent functions against data injection
1538 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1539 [18755] build: build errors with -DNDEBUG
1540 [18757] stdio: fmemopen fails to set errno on failure
1541 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1543 [18781] libc: openat64 lacks O_LARGEFILE
1544 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1545 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1546 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1547 [18790] math: [ldbl-128ibm] tanhl inaccurate
1548 [18795] libc: stpncpy fortification misses buffer lengths that are
1549 statically too large
1550 [18796] build: build fails for --disable-mathvec
1551 [18803] math: hypot missing underflows
1552 [18820] stdio: fmemopen may leak memory on failure
1553 [18823] math: csqrt spurious underflows
1554 [18824] math: fma spurious underflows
1555 [18825] math: pow missing underflows
1556 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1557 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1558 [18870] build: sem_open.c fails to compile with missing symbol
1560 [18872] stdio: Fix memory leak in printf_positional
1561 [18873] libc: posix_fallocate overflow check ineffective
1562 [18875] math: Excess precision leads incorrect libm
1563 [18877] libc: arm: mmap offset regression
1564 [18887] libc: memory corruption when using getmntent on blank lines
1565 [18918] localedata: hu_HU: change time to HH:MM:SS format
1566 [18921] libc: Regression: extraneous stat() and fstat() performed by
1568 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1569 binaries (CVE-2015-8777)
1570 [18951] math: tgamma missing underflows
1571 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1573 [18953] localedata: lt_LT: change currency symbol to the euro
1574 [18956] math: powf inaccuracy
1575 [18961] math: [i386] exp missing underflows
1576 [18966] math: [i386] exp10 missing underflows
1577 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1578 [18969] build: multiple string test failures due to missing locale
1580 [18970] libc: Reference of pthread_setcancelstate in libc.a
1581 [18977] math: float / long double Bessel functions not in XSI POSIX
1582 [18980] math: i386 libm functions return with excess range and precision
1583 [18981] math: i386 scalb*, ldexp return with excess range and precision
1584 [18982] stdio: va_list and vprintf
1585 [18985] time: Passing out of range data to strftime() causes a segfault
1587 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1588 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1589 [19012] locale: iconv_open leaks memory on error path
1590 [19016] math: clog, clog10 inaccuracy
1591 [19018] nptl: Mangle function pointers in tls_dtor_list
1592 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1593 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1594 [19048] malloc: malloc: arena free list can become cyclic, increasing
1596 [19049] math: [powerpc] erfc incorrect zero sign
1597 [19050] math: [powerpc] log* incorrect zero sign
1598 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1599 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1600 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1601 [19074] network: Data race in _res_hconf_reorder_addrs
1602 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1603 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1604 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1606 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1608 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1609 [19086] manual: posix_fallocate64 documented argument order is wrong.
1610 [19088] math: lround, llround missing exceptions close to overflow
1612 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1613 [19095] math: dbl-64 lrint incorrect for 64-bit long
1614 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1615 [19124] dynamic-link: ld.so failed to build with older assmebler
1616 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1617 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1618 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1619 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1620 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1621 pthread_setaffinity_np
1622 [19156] math: [ldbl-128] j0l spurious underflows
1623 [19164] nptl: tst-getcpu fails with many possible CPUs
1624 [19168] math: math/test-ildoubl and math/test-ldouble failure
1625 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1626 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1628 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1629 (FE_NOMASK_ENV) do not clear SSE exceptions
1630 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1631 _int_new_arena/reused_arena
1632 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1633 [19201] math: dbl-64 remainder incorrect sign of zero result
1634 [19205] math: bits/math-finite.h conditions do not match math.h and
1636 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1637 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1638 for C99-based standards
1639 [19212] libc: features.h not -Wundef clean
1640 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1642 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1643 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1644 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1646 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1647 "inexact" exceptions
1648 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1650 [19242] libc: strtol incorrect in Turkish locales
1651 [19243] malloc: reused_arena can pick an arena on the free list, leading
1652 to an assertion failure and reference count corruption
1653 [19253] time: tzset() ineffective when temporary TZ did not include DST
1655 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1656 [19270] math: [hppa] Shared libm missing __isnanl
1657 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1658 MAP_HUGETLB and MAP_STACK defines
1659 [19313] nptl: Wrong __cpu_mask for x32
1660 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1661 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1662 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1663 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1664 [19363] time: x32: times() return value wrongly truncates/sign extends
1666 [19367] dynamic-link: Improve branch prediction on Silvermont
1667 [19369] network: Default domain name not reset by res_ninit when "search"
1668 / "domain" entry is removed from resolv.conf
1669 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1671 [19385] localedata: bg_BG: time separator should be colon, not comma
1672 [19408] libc: linux personality syscall wrapper may erroneously return an
1673 error on 32-bit architectures
1674 [19415] libc: dladdr returns wrong names on hppa
1675 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1676 IBM905, IBM907, and IBM909
1677 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1678 [19443] build: build failures with -DDEBUG
1679 [19451] build: Make check fails on test-double-vlen2
1680 [19462] libc: Glibc failed to build with -Os
1681 [19465] math: Wrong code with -Os
1682 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1684 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1686 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1687 [19486] math: S390: Math tests fail with "Exception Inexact set".
1688 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1689 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1691 [19590] math: Fail to build shared objects that use libmvec.so functions.
1695 * The following bugs are resolved with this release:
1697 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1698 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1699 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1700 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1701 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1702 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1703 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1704 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1705 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1706 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1707 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1708 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1709 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1710 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1711 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1712 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1713 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1714 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1715 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1716 18657, 18676, 18694, 18696, 18887.
1718 * Cache information can be queried via sysconf() function on s390 e.g. with
1719 _SC_LEVEL1_ICACHE_SIZE as argument.
1721 * A buffer overflow in gethostbyname_r and related functions performing DNS
1722 requests has been fixed. If the NSS functions were called with a
1723 misaligned buffer, the buffer length change due to pointer alignment was
1724 not taken into account. This could result in application crashes or,
1725 potentially arbitrary code execution, using crafted, but syntactically
1726 valid DNS responses. (CVE-2015-1781)
1728 * The time zone file parser has been made more robust against crafted time
1729 zone files, avoiding heap buffer overflows related to the processing of
1730 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1731 large time zone data files. Overly long time zone specifiers in the TZ
1732 variable no longer result in stack overflows and crashes.
1734 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1735 for LD and GD on x86 and x86-64, has been implemented. You will need
1736 binutils-2.24 or later to enable this optimization.
1738 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1739 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1740 Hat). These updates cause user visible changes, such as the fix for bug
1743 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1744 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1745 condition in some applications.
1747 * Added vector math library named libmvec with the following vectorized x86_64
1748 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1750 The library can be disabled with --disable-mathvec. Use of the functions is
1751 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1752 Shared library libmvec.so is linked in as needed when using -lm (no need to
1753 specify -lmvec explicitly for not static builds).
1754 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1756 * A new fmemopen implementation has been added with the goal of POSIX
1757 compliance. The new implementation fixes the following long-standing
1758 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1759 old implementation is still present for use be by existing binaries.
1761 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1762 releases. It has been fixed to match 2.19 and older, but binaries built
1763 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1765 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1766 Contributed by Roland McGrath (Google).
1768 * The header <regexp.h> is deprecated, and will be removed in a future
1769 release. Use of this header will trigger a deprecation warning.
1770 Application developers should update their code to use <regex.h> instead.
1772 This header was formerly part of SUS, but was deprecated in 1994 and
1773 removed from the standard in 2001. Also, the glibc implementation
1774 leaks memory. See BZ#18681 for more details.
1778 * The following bugs are resolved with this release:
1780 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1781 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1782 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1783 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1784 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1785 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1786 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1787 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1788 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1789 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1792 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1793 too little memory for the to-be-scanned arguments and overflow the
1794 allocated buffer. The implementation now correctly computes the required
1795 buffer size when using malloc, and switches to malloc from alloca as
1798 * A new semaphore algorithm has been implemented in generic C code for all
1799 machines. Previous custom assembly implementations of semaphore were
1800 difficult to reason about or ensure that they were safe. The new version
1801 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1802 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1803 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1805 * Port to Altera Nios II has been contributed by Mentor Graphics.
1807 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1808 implementations for powerpc64/powerpc64le.
1809 Implemented by Adhemerval Zanella (IBM).
1811 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1812 and powerpc64le. This may improve lock scaling of existing programs on
1813 HTM capable systems. The lock elision code is only enabled with
1814 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1815 powerpc will issue a transaction abort on every syscall to avoid side
1816 effects being visible outside transactions.
1818 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1819 AArch64. Contributed by ARM Ltd.
1821 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1823 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1824 under certain input conditions resulting in the execution of a shell for
1825 command substitution when the application did not request it. The
1826 implementation now checks WRDE_NOCMD immediately before executing the
1827 shell and returns the error WRDE_CMDSUB as expected.
1829 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1830 processing format strings with a large number of format specifiers.
1832 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1833 infinite loop if the DNS response contained a PTR record of an unexpected
1836 * The minimum GCC version that can be used to build this version of the GNU
1837 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1838 still be used to compile programs using the GNU C Library.
1840 * The GNU C Library is now built with -Werror by default. This can be
1841 disabled by configuring with --disable-werror.
1843 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1845 * The obsolete sigvec function has been removed. This was the original
1846 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1847 programs have been using instead for about 25 years. Of course, ABI
1848 compatibility for old binaries using sigvec remains intact.
1850 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1851 with newer versions of bison.
1853 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1854 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1855 registers overlay two consecutive single-precision registers. MIPS32R2
1856 introduced a new FPU mode (FR=1) where double-precision registers extend the
1857 corresponding single-precision registers which is incompatible with the
1858 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1859 require the use of FR=1 making a transition necessary. New o32 ABI
1860 extensions enable users to migrate over time from the original o32 ABI
1861 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1862 tracks the ABI of any loaded object and verifies that new objects are
1863 compatible. Mode transitions will also be requested as required and
1864 unsupportable objects will be rejected. The ABI checks include both soft and
1865 hard float ABIs for o32, n32 and n64.
1867 GCC 5 with GNU binutils 2.25 onwards:
1868 It is strongly recommended that all o32 system libraries are built using the
1869 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1870 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1871 compiler using --with-fp-32=xx to set this by default.
1875 * The following bugs are resolved with this release:
1877 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1878 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1879 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1880 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1881 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1882 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1883 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1884 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1885 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1886 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1887 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1888 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1889 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1890 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1891 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1893 * Reverted change of ABI data structures for s390 and s390x:
1894 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1895 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1896 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1897 are preserved pointing straight to the same implementation as the old ones.
1898 Given that, new callers will simply provide a too-big buffer to these
1899 functions. Any applications/libraries out there that embed jmp_buf or
1900 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1901 against 2.19 headers will have to rebuilt again. This is necessary in any
1902 case to revert the breakage in their ABI caused by the glibc change.
1904 * Support for file description locks is added to systems running the
1905 Linux kernel. The standard file locking interfaces are extended to
1906 operate on file descriptions, not file descriptors, via the use of
1907 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1908 are associated with an open file instead of a process.
1910 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1912 * The minimum Linux kernel version that this version of the GNU C Library
1913 can be used with is 2.6.32.
1915 * Running the testsuite no longer terminates as soon as a test fails.
1916 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1917 with PASS or FAIL lines for individual tests. A summary of the results is
1918 printed, including a list of failing lists, and "make check" exits with
1919 error status if there were any unexpected failures. "make check
1920 stop-on-test-failure=y" may be used to keep the old behavior.
1922 * The am33 port, which had not worked for several years, has been removed
1925 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1926 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1927 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1928 conflicted with POSIX (support for which was removed in 2.19), the
1929 interfaces those macros enabled remain available when compiling with
1930 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1931 test macros defined.
1933 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1935 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1936 This may improve lock scaling of existing programs on TX capable systems.
1937 The lock elision code is only built with --enable-lock-elision=yes and
1938 then requires a GCC version supporting the TX builtins. With lock elision
1939 default mutexes are elided via __builtin_tbegin, if the cpu supports
1940 transactions. By default lock elision is not enabled and the elision code
1943 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1944 copy the path argument. This allowed programs to cause posix_spawn to
1945 deference a dangling pointer, or use an unexpected pathname argument if
1946 the string was modified after the posix_spawn_file_actions_addopen
1949 * All supported architectures now use the main glibc sysdeps directory
1950 instead of some being in a separate "ports" directory (which was
1951 distributed separately before glibc 2.17).
1953 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1954 On configurations that support it (all Linux configurations), it's now
1955 used regardless of the --enable-add-ons switch to configure. It is no
1956 longer possible to build such configurations without pthreads support.
1958 * Locale names, including those obtained from environment variables (LANG
1959 and the LC_* variables), are more tightly checked for proper syntax.
1960 setlocale will now fail (with EINVAL) for locale names that are overly
1961 long, contain slashes without starting with a slash, or contain ".." path
1962 components. (CVE-2014-0475) Previously, some valid locale names were
1963 silently replaced with the "C" locale when running in AT_SECURE mode
1964 (e.g., in a SUID program). This is no longer necessary because of the
1967 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1968 with application code using Intel MPX instructions. (With all previous
1969 versions, the MPX register state could be clobbered when making calls
1970 into or out of a shared library.) Note that while the new dynamic
1971 linker is compatible with all known x86 hardware whether or not it
1972 supports Intel MPX, some x86 instruction-set emulators might fail to
1973 handle the new instruction encodings. This is known to affect Valgrind
1974 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1975 release), and might affect other tools that do instruction emulation.
1977 * Support for loadable gconv transliteration modules has been removed.
1978 The support for transliteration modules has been non-functional for
1979 over a decade, and the removal is prompted by security defects. The
1980 normal gconv conversion modules are still supported. Transliteration
1981 with //TRANSLIT is still possible, and the //IGNORE specifier
1982 continues to be supported. (CVE-2014-5119)
1984 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
1985 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
1986 resulting a denial-of-service security vulnerability in applications which
1987 use functions related to iconv. (CVE-2014-6040)
1991 * The following bugs are resolved with this release:
1993 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
1994 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
1995 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
1996 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
1997 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
1998 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
1999 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2000 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2001 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2002 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2003 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2004 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2005 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2006 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2007 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2008 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2009 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2010 16431, 16453, 16474, 16506, 16510, 16529
2012 * Slovenian translations for glibc messages have been contributed by the
2013 Translation Project's Slovenian team of translators.
2015 * The public headers no longer use __unused nor __block. This change is to
2016 support compiling programs that are derived from BSD sources and use
2017 __unused internally, and to support compiling with Clang's -fblock
2018 extension which uses __block.
2020 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2021 large collation sequences to optimize multiple passes. This cache
2022 computation may overflow for large collation sequences and may cause a
2023 stack or buffer overflow. This is now fixed to use a slower algorithm
2024 which does not use a cache if there is an integer overflow.
2026 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2027 rules for large collation sequences to optimize multiple passes and falls
2028 back to alloca if malloc fails, resulting in a possible stack overflow.
2029 The implementation now falls back to an uncached collation sequence lookup
2032 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2033 initialized for static applications resulting in the security feature
2034 being disabled. The pointer guard is now correctly initialized to a
2035 random value for static applications. Existing static applications need
2036 to be recompiled to take advantage of the fix (bug 15754).
2038 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2039 to the d_name member of struct dirent, or omit the terminating NUL
2040 character. (Bugzilla #14699).
2042 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2043 aligned_alloc functions could allocate too few bytes or corrupt the
2044 heap when passed very large allocation size values (Bugzilla #15855,
2047 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2048 for AF_INET6 has been fixed (Bugzilla #16072).
2050 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2051 pap_CW, quz_PE, the_NP.
2053 * Substantially revised locales: gd_GB, ht_HT
2055 * The LC_ADDRESS field was updated to support country_car for almost all
2058 * ISO 1427 definitions were updated.
2060 * ISO 3166 definitions were updated.
2062 * The localedef utility now supports --big-endian and --little-endian
2063 command-line options to generate locales for a different system from that
2064 for which the C library was built.
2066 * Binary locale files now only depend on the endianness of the system for
2067 which they are generated and not on other properties of that system. As a
2068 consequence, binary files generated with new localedef may be incompatible
2069 with old versions of the GNU C Library, and binary files generated with
2070 old localedef may be incompatible with this version of the GNU C Library,
2071 in the following circumstances:
2073 + Locale files may be incompatible on m68k systems.
2075 + Locale archive files (but not separate files for individual locales) may
2076 be incompatible on systems where plain "char" is signed.
2078 * The configure option --disable-versioning has been removed. Builds with
2079 --disable-versioning had not worked for several years.
2081 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2082 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2084 * SystemTap probes for malloc have been introduced.
2086 * SystemTap probes for slow multiple precision fallback paths of
2087 transcendental functions have been introduced.
2089 * Support for powerpc64le has been added.
2091 * The soft-float powerpc port now supports e500 processors.
2093 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2095 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2096 set of header declarations that are enabled by default, even when other
2097 feature test macros or compiler options such as -std=c99 would otherwise
2098 disable some of those declarations.
2100 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2101 conflict with POSIX. The libbsd-compat library (which was a dummy library
2102 that did nothing) has also been removed.
2104 * Preliminary documentation about Multi-Thread, Async-Signal and
2105 Async-Cancel Safety has been added.
2107 * Change of ABI data structures for s390 and s390x:
2108 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2109 allow for future hardware extensions. All library functions that accept or
2110 return these structures were versioned in order to provide backwards
2111 compatibility with existing code. However, not all cases can be handled
2112 with interface versioning. If an application embeds either structure into
2113 another structure and that structure is passed to another compilation unit
2114 which expects the newer larger-sized object then any access to the new
2115 fields will result in undefined behaviour. Similarly any access to fields
2116 that were moved by the enlarging of the structures will cause undefined
2117 behaviour. To fix the undefined behaviour all objects that are part of the
2118 final in-memory image must agree on the size of structures, and this may
2119 require recompilation.
2123 * The following bugs are resolved with this release:
2125 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2126 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2127 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2128 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2129 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2130 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2131 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2132 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2133 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2134 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2135 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2136 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2137 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2140 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2141 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2142 Distributions can re-enable building and using pt_chown via the new configure
2143 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2144 considerable security risks and should only be used if the distribution
2145 understands and accepts the risks.
2147 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2150 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2151 fixed (Bugzilla #15330).
2153 * Add support for calling C++11 thread_local object destructors on thread
2154 and program exit. This needs compiler support for offloading C++11
2155 destructor calls to glibc.
2157 * Improved worst case performance of libm functions with double inputs and
2160 * Support for priority inherited mutexes in pthread condition variables on
2161 non-x86 architectures.
2163 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2165 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2167 * Optimized string functions for ARM. Implemented by Will Newton and
2170 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2172 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2175 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2176 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2178 * Added a benchmark framework to track performance of functions in glibc.
2180 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2181 It is based on draft TS 18661 and currently enabled as a GNU extension.
2183 * On Linux, the clock function now uses the clock_gettime system call
2184 for improved precision, rather than old times system call.
2186 * Added support for version-3 tz format files. This is needed when using
2187 the tz database release 2013e or later, and affects a few unusual cases --
2188 currently only TZ='America/Godthab' for time stamps after 2037.
2190 * Added new API functions pthread_getattr_default_np and
2191 pthread_setattr_default_np to get and set the default pthread
2192 attributes of a process.
2194 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2195 This may improve lock scaling of existing programs on TSX capable systems.
2196 When the --enable-lock-elision=yes parameter is specified at configure
2197 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2200 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2201 availability). Implemented by Ryan S. Arnold.
2203 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2207 * The following bugs are resolved with this release:
2209 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2210 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2211 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2212 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2213 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2214 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2215 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2216 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2217 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2218 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2219 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2220 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2221 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2222 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2223 14879, 14889, 14893, 14898, 14914.
2225 * Optimization of memcpy for MIPS.
2227 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2228 EMFILE has been fixed (Bugzilla #14889).
2230 * The add-on ports collection is now distributed in the "ports" subdirectory
2231 of the main GNU C Library distribution, rather than separately.
2233 * Port to ARM AArch64 contributed by Linaro.
2235 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2236 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2238 Implemented by Andreas Krebbel.
2240 * The new function secure_getenv allows secure access to the environment,
2241 returning NULL if running in a SUID/SGID process. This function replaces
2242 the internal function __secure_getenv.
2244 * SystemTap static probes have been added into the dynamic linker.
2245 Implemented by Gary Benson.
2247 * Optimizations of string functions strstr, strcasestr and memmem.
2248 Implemented by Maxim Kuvyrkov.
2250 * The minimum Linux kernel version that this version of the GNU C Library
2251 can be used with is 2.6.16.
2253 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2254 powerpc POWER7. Implemented by Will Schmidt.
2256 * New configure option --disable-nscd builds the C library such that it
2257 never attempts to contact the Name Service Caching Daemon (nscd).
2258 New configure option --disable-build-nscd avoids building nscd itself;
2259 this is the default if --disable-nscd is used.
2261 * Improved support for cross-compilation, including cross-testing and
2262 bootstrap builds without a previously built glibc.
2264 * Several testsuite tests are now able to test multiple IFUNC variants of an
2265 interface, rather than just testing the one that would be chooen by
2268 * New configure options --with-bugurl and --with-pkgversion, for
2269 distributors to use to embed their bug-reporting and package version
2270 information in --help and --version output.
2272 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2273 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2274 allows creation of chroots without the procfs mounted on /proc.
2276 * The `crypt' function now fails if passed salt bytes that violate the
2277 specification for those values. On Linux, the `crypt' function will
2278 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2279 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2280 when the mode is enabled.
2282 * The `clock_*' suite of functions (declared in <time.h>) is now available
2283 directly in the main C library. Previously it was necessary to link with
2284 -lrt to use these functions. This change has the effect that a
2285 single-threaded program that uses a function such as `clock_gettime' (and
2286 is not linked with -lrt) will no longer implicitly load the pthreads
2287 library at runtime and so will not suffer the overheads associated with
2288 multi-thread support in other code such as the C++ runtime library.
2290 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2295 * The following bugs are resolved with this release:
2297 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2298 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2299 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2300 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2301 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2302 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2303 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2304 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2305 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2306 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2307 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2308 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2309 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2310 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2311 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2312 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2313 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2314 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2315 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2316 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2317 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2318 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2321 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2322 configuring glibc with:
2323 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2324 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2325 Implemented by H.J. Lu.
2329 + define static_assert
2331 + do not declare gets
2333 + declare at_quick_exit and quick_exit also for ISO C11
2335 + aligned_alloc. NB: The code is deliberately allows the size parameter
2336 to not be a multiple of the alignment. This is a moronic requirement
2337 in the standard but it is only a requirement on the caller, not the
2340 + timespec_get added
2342 + uchar.h support added
2344 + CMPLX, CMPLXF, CMPLXL added
2346 Implemented by Ulrich Drepper.
2348 * Support for the IA-64 has been moved to ports.
2350 * Remove support for anything but ELF binary format
2352 * Checking versions of poll, ppoll added.
2353 Implemented by Ulrich Drepper.
2355 * More generic and 64-bit performance optimizations to math functions.
2356 Implemented by Ulrich Drepper.
2358 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2359 headers and functions available at compile time as they were before
2360 version 2.14. This option will be removed at some time in the future
2361 after the TI-RPC library becomes fully sufficient for the needs of
2362 existing applications.
2364 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2365 Note that glibc is not expected to work with any Linux kernel version
2368 * New header <sys/auxv.h> and function getauxval allowing easy access to
2369 the AT_* key-value pairs passed from the Linux kernel. The header also
2370 defines the HWCAP_* bits associated with the AT_HWCAP key.
2372 * New locales: mag_IN
2374 * New configure option --enable-systemtap builds SystemTap static probes
2375 into libc for setjmp and longjmp and into libpthread for various operations.
2376 So far the setjmp/longjmp probes and some of the libpthread probes are
2377 provided only for i*86 and x86_64.
2378 Implemented by Roland McGrath and Rayson Ho.
2380 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2382 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2385 * More optimized functions for SPARC. Implemented by David S. Miller.
2387 * Improved support for cross-compilation, especially bootstrap builds
2388 without a previously built glibc.
2390 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2391 Chris Metcalf from Tilera.
2393 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2394 now supported for ARM processors.
2396 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2397 as the name of the dynamic linker, to distinguish it from the
2398 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2400 * Support for CRIS has been removed from ports.
2402 * A new class of installed header has been documented for low-level
2403 platform-specific functionality. PowerPC added the first instance with a
2404 function to provide time base register access. Contributed by Tulio
2405 Magno Quites Machado Filho.
2407 * ix86 configurations now install header files that are consistent with
2408 what x86-64 configurations install. These same header files can be used
2409 for -m32, -m64, or -mx32 builds.
2410 Contributed by H.J. Lu.
2412 * Math library bug fixes. A thorough audit of all open math library bugs was
2413 conducted by Joseph Myers. Significant progress was made on many math
2414 library bugs resulting in more accurate exceptions and function results.
2415 Many thanks to all those that contributed including Andreas Jaeger for his
2416 patch review and work on the x87 trigonometric instruction issues.
2418 * Timezone data is no longer installed. Timezone-related binaries and scripts
2419 will continue to be installed. Users should obtain their timezone data from
2420 their distribution provider or from the tzdata package at
2421 <ftp://munnari.oz.au/pub/>.
2425 * The following bugs are resolved with this release:
2427 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2428 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2429 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2430 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2431 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2432 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2433 13484, 13506, 13515, 13523, 13524, 13538, 13540
2435 * New program pldd to list loaded object of a process
2436 Implemented by Ulrich Drepper.
2438 * Add nss_db support back to glibc. No more dependency on Berkeley db
2439 and support for initgroups lookups.
2440 Implemented by Ulrich Drepper.
2442 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2443 Contributed by HJ Lu.
2445 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2446 Contributed by HJ Lu.
2448 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2449 on x86-32 and x86-64.
2450 Contributed by Liubov Dmitrieva.
2452 * Optimized strchr and strrchr for SSE on x86-32.
2453 Contributed by Liubov Dmitrieva.
2455 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2456 for x86-64 and x86-32.
2457 Contributed by Liubov Dmitrieva.
2459 * New interfaces: scandirat, scandirat64
2460 Implemented by Ulrich Drepper.
2462 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2463 Implemented by Ulrich Drepper.
2465 * nscd now also caches the netgroup database.
2466 Implemented by Ulrich Drepper.
2468 * Integrate libm with gcc's -ffinite-math-only option.
2469 Implemented by Ulrich Drepper.
2471 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2472 to math functions. Implemented by Ulrich Drepper.
2474 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2475 Implemented by Ulrich Drepper.
2477 * New Linux interfaces: process_vm_readv, process_vm_writev
2479 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2480 Implemented by Ulrich Drepper.
2482 * Optimized nearbyint and strcasecmp for PPC.
2483 Implemented by Adhemerval Zanella.
2485 * New locales: bho_IN, unm_US, es_CU, ta_LK
2489 * The following bugs are resolved with this release:
2491 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2492 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2493 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2494 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2495 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2496 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2497 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2498 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2499 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2500 12795, 12811, 12813, 12814, 12841
2502 * The RPC implementation in libc is obsoleted. Old programs keep working
2503 but new programs cannot be linked with the routines in libc anymore.
2504 Programs in need of RPC functionality must be linked against TI-RPC.
2505 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2507 Visible changes of this change include (obviously) the inability to link
2508 programs using RPC functions without referencing the TI-RPC library and the
2509 removal of the RPC headers from the glibc headers.
2510 Implemented by Ulrich Drepper.
2512 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2513 syncfs, setns, sendmmsg
2515 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2516 yue_HK, lij_IT, mhr_RU
2518 * New encodings: CP770, CP771, CP772, CP773, CP774
2520 * New program sotruss to trace calls through PLTs
2521 Implemented by Ulrich Drepper.
2523 * The malloc hook implementation is marked deprecated and will be removed
2524 from the default implementation in the next version. The design never
2525 worked ever since the introduction of threads. Even programs which do
2526 not create threads themselves can use multiple threads created internally.
2530 * The following bugs are resolved with this release:
2532 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2533 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2534 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2535 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2538 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2540 * POWER7 optimizations: memset, memcmp, strncmp
2542 * New optimized string functions for x86-64: strnlen (SSE2),
2543 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2544 Implemented by Ulrich Drepper.
2548 * The following bugs are resolved with this release:
2550 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2551 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2552 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2553 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2554 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2555 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2556 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2557 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2559 * New interfaces: pthread_getname_np, pthread_setname_np
2561 * New Linux interface: recvmmsg
2563 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2565 * The dynamic linker now recognizes supported ABI versions from the
2566 EI_ABIVERSION field in the ELF header.
2567 Implemented by Ulrich Drepper.
2569 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2570 will not be used to fill in password fields in the passwd.byname replies.
2571 Instead it is used to synthesize the shadow.byname table, should it be
2572 missing. This is a useful mode in some installations involving Solaris.
2573 Implemented by Ulrich Drepper.
2575 * New locales: kok_IN, sq_MK, cv_RU
2579 * The following bugs are resolved with this release:
2581 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2582 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2583 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2584 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2585 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2586 10789, 10817, 10823, 10840, 10847
2588 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2590 Implemented by Ulrich Drepper.
2592 * Checking version of longjmp added that fails if an uninitialized stack
2593 frame would be created. Implemented by Ulrich Drepper.
2595 * STT_GNU_IFUNC is now supported in static executables.
2596 Implemented by H.J. Lu.
2598 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2600 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2601 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2602 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2603 Contributed by H.J. Lu.
2605 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2606 Implemented by Ulrich Drepper.
2608 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2610 Contributed by H.J. Lu.
2612 * Support for fma instruction in AVX on x86-64.
2613 Implemented by H.J. Lu and Ulrich Drepper.
2615 * AVX support in x86-64 auditing support in ld.so.
2616 Implemented by H.J. Lu.
2618 * STB_GNU_UNIQUE support added.
2619 Implemented by Ulrich Drepper.
2621 * Implement second fallback mode for optimized DNS lookup for even more
2622 broken environments. If two requests from the same port are not
2623 handled correctly close the socket and open a new one before sending
2624 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2625 can be used to select this mode right away, instead of rediscovering the
2626 necessity is every process again.
2627 Implemented by Ulrich Drepper.
2629 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2630 Implemented by Adam Tkac.
2632 * Optimized iconv conversions for S390x.
2633 Implemented by Andreas Krebbel.
2635 * Using condvars with PI mutexes is now more efficient due to kernel
2636 support for requeueing to PI futexes. NPTL support added for x86-64.
2637 Implemented by Ulrich Drepper.
2639 * New locales: ps_AF, my_MM
2643 * The following bugs are resolved with this release:
2645 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2646 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2647 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2648 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2649 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2651 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2652 Implemented by Ulrich Drepper.
2654 * New Linux interfaces: accept4, fallocate, fallocate64.
2655 Implemented by Ulrich Drepper.
2657 * Correct declarations of string function when used in C++ code. This
2658 could lead to compile errors for invalid C++ code.
2660 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2663 * New POSIX 2008 interface: psiginfo
2664 Implemented by Ulrich Drepper.
2666 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2667 Implemented by Ulrich Drepper.
2669 * Support for selecting between multiple function definitions at runtime
2670 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2672 * The libcrypt library can now use the hash function implementations in
2673 NSS. Implemented by Ulrich Drepper.
2675 * The malloc implementation can be compiled to be less memory efficient
2676 but higher performing in multi-threaded programs.
2677 Implemented by Ulrich Drepper.
2679 * New locales: nan_TW@latin, ks_IN
2681 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2682 Implemented by Ulrich Drepper.
2684 * Extended printf hook support. It is possible to use user-defined types
2685 and extend existing format specifiers.
2686 Implemented by Ulrich Drepper.
2688 * Handling for group shadow files has been added.
2689 Implemented by Ulrich Drepper.
2691 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2692 servers (the case, e.g., for some people using the built-in DNS
2693 server in ADSL modems/routers). There is a once-per-process timeout
2694 in case of a broken server. To avoid it, users can run nscd or put
2695 'options single-request' in /etc/resolv.conf.
2696 Implemented by Ulrich Drepper.
2700 * The following bugs are resolved with this release:
2702 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2703 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2704 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2705 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2706 6968, 6974, 6980, 6995, 7008, 7009, 7029
2708 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2709 up at the same time. Implemented by Ulrich Drepper.
2711 * TLS descriptors for LD and GD on x86 and x86-64.
2712 Implemented by Alexandre Oliva.
2714 * getaddrinfo now handles DCCP and UDPlite.
2715 Implemented by Ulrich Drepper.
2717 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2718 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2719 Implemented by Ulrich Drepper.
2721 * New implementation of memmem, strstr, and strcasestr which is O(n).
2722 Implemented by Eric Blake.
2724 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2726 * Implement "e" option for popen to open file descriptor with the
2727 close-on-exec flag set. Implemented by Ulrich Drepper.
2729 * Implement "b" mode for fmemopen. In this mode writes writes don't
2730 implicitly add a NUL byte and seeks from the end of the buffer really
2731 use the buffer end, not the string length as the basis.
2732 Implemented by Ulrich Drepper.
2734 * Many functions, exported and internal, now atomically set the close-on-exec
2735 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2737 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2739 Implemented by Pravin Satpute.
2741 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2745 * The following bugs are resolved with this release:
2747 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2748 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2749 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2750 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2751 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2752 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2753 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2755 * New locales: bo_CN, bo_IN, shs_CA.
2757 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2759 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2760 Implemented by Pravin Satpute.
2762 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2764 * nscd caches now all timeouts for DNS entries
2765 Implemented by Ulrich Drepper.
2767 * nscd is more efficient and wakes up less often.
2768 Implemented by Ulrich Drepper.
2770 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2771 vdprintf, and obstack_vprintf.
2772 Implemented by Jakub Jelinek.
2774 * Faster memset for x86-64.
2775 Implemented by Harsha Jagasia and H.J. Lu.
2777 * Faster memcpy on x86.
2778 Implemented by Ulrich Drepper.
2780 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2781 Implemented by Ulrich Drepper.
2783 * Faster sqrt and sqrtf implemention for some PPC variants.
2784 Implemented by Steven Munroe.
2788 * The following bugs are resolved with this release:
2790 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2791 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2792 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2793 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2794 5071, 5103, 5104, 5112, 5113, 5184, 5186
2796 * More checking functions: fread, fread_unlocked, open*, mq_open.
2797 Implemented by Jakub Jelinek and Ulrich Drepper.
2799 * Extend fortification to C++. Implemented by Jakub Jelinek.
2801 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2802 by not recognizing 'a' as a modifier when those specs are requested.
2803 Implemented by Jakub Jelinek.
2805 * PPC optimizations to math and string functions.
2806 Implemented by Steven Munroe.
2808 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2809 options to be passed. Implemented by Ulrich Drepper.
2811 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2813 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2814 Implemented by Ulrich Drepper.
2816 * Handle private futexes in the NPTL implementation.
2817 Implemented by Jakub Jelinek and Ulrich Drepper.
2819 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2820 Implemented by Roland McGrath and Ulrich Drepper.
2822 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2824 * SHA-256 and SHA-512 based password encryption.
2825 Implemented by Ulrich Drepper.
2827 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2828 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2831 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2832 Implemented by Ulrich Drepper.
2836 * The following bugs are resolved with this release:
2838 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2839 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2840 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2841 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2842 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2843 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2844 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2847 * New Linux interfaces: epoll_pwait, sched_getcpu.
2849 * New generic interfaces: strerror_l.
2851 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2855 * The following bugs are resolved with this release:
2857 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2858 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2859 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2860 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2861 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2862 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2863 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2864 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2865 3225, 3251, 3252, 3253, 3273
2867 * For Linux, the sorting of addresses returned by getaddrinfo now also
2868 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2869 Implemented by Ulrich Drepper.
2871 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2872 Implemented by Ulrich Drepper.
2874 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2876 * New iconv module for MIK. Contributed by Alexander Shopov.
2878 * For sites with broken group and/or passwd database, the auto-propagate
2879 option of nscd can prevent creating ID lookup entries from the results
2880 of a name lookup and vice versa. This usually is no problem but some
2881 site might have problems with the default behavior.
2882 Implemented by Ulrich Drepper.
2884 * Iterating over entire database in NIS can be slow. With the
2885 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2886 to trade time for memory. The entire database will be read at once.
2887 Implemented by Ulrich Drepper.
2889 * The interfaces introduced in RFC 3542 have been implemented by
2892 * Support for the new ELF hash table format was added by Ulrich Drepper.
2894 * Support for priority inheritance mutexes added by Jakub Jelinek and
2897 * Support for priority protected mutexes added by Jakub Jelinek.
2899 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2900 or_IN, csb_PL, fy_NL, sr_ME.
2904 * More overflow detection functions.
2906 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2907 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2909 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2910 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2911 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2912 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2913 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2914 by Masahide Washizawa.
2916 * It is now possible to install audit libraries for the dynamic linker using
2917 LD_AUDIT. Implemented by Ulrich Drepper.
2919 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2920 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2921 For a libc and libpthread that works well on Linux 2.4 kernels, we
2922 recommend using the stable 2.3 branch.
2924 * The new function open_wmemstream acts like open_memstream,
2925 but uses a wchar_t wide character buffer.
2927 * The new function ppoll is an improved version of poll, similar to pselect.
2929 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2930 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2931 renameat, unlinkat, linkat, symlinkat, readlinkat.
2933 * New Linux kernel system calls: unshare,
2934 inotify_init, inotify_add_watch, inotify_rm_watch.
2936 * The euidaccess function is now also known by the alias eaccess,
2937 for compatibility with some other systems.
2939 * Timezone data updated to 2006b version.
2943 * The following bugs are resolved with this release:
2945 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2946 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2947 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2948 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2949 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2950 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2952 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2954 * As of this release, GCC 4 can be used to compile the C Library.
2956 * Timezone data updated to 2005m version.
2960 * The following bugs are resolved with this release:
2962 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2963 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2964 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2965 777, 787, 821, 822, 823, 825
2967 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2971 * Support for RFC 3678. Real implementations exist only for Linux so far.
2972 Implemented by Ulrich Drepper.
2974 * nscd can now cache entries persistently. Expiring entries are reloaded.
2975 For speedups the cache can be shared in memory with client processes.
2976 Implemented by Ulrich Drepper.
2978 * nscd can now perform SELinux checks.
2979 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2981 * getaddrinfo queries are now cached. Canonical name lookup is performed
2983 Implemented by Ulrich Drepper.
2985 * The nothrow function attribute is used when headers are used by gcc when
2986 compiling C code. This can avoid significant amounts of exception
2989 * The malloc functions perform more error checking and are stricter when
2990 it comes to reacting on errors. The default action is to terminate
2991 the process after showing an error message. Implemented by Ulrich Drepper.
2993 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
2994 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
2996 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
2997 namespaces using the new function dlmopen(). This feature is of course,
2998 like most other dynamic loading functionality, not available in statically
2999 linked applications. Implemented by Ulrich Drepper.
3001 * Low-overhead boundary checking variants of string and some stdio functions
3002 were added. These are to be used in conjunction with a gcc patch by
3003 Jakub Jelinek which adds calls to these functions if possible.
3004 Implemented by Jakub Jelinek and Ulrich Drepper.
3006 * Old code for several operating systems and machine architectures that
3007 have not been in working condition in a long time have been removed from
3008 the main source tree maintained by the GNU C Library's maintainers.
3009 These files are now reside in the separate `ports' source module
3010 that is usable as an add-on when building the library.
3014 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3015 interrogate the dynamic linker, compatible with the Solaris interface.
3017 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3018 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3020 * getifaddrs now uses the netlink interface on Linux to get its information.
3021 Implemented by Thorsten Kukuk.
3023 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3024 Implemented by Ulrich Drepper.
3026 * support for non-executable stacks on x86 has been added. Changes mostly
3029 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3032 * getaddrinfo now performs destination address selection according to
3037 * Thread-safe interfaces for many functions that access locale data
3038 were added in version 2.3, but these features were omitted from NEWS.
3039 Many functions have variants with an `_l' suffix that take a `locale_t'
3040 object as a parameter rather than consulting the current locale.
3041 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3042 create and maintain `locale_t' objects. Additionally, the new function
3043 `uselocale' sets "the current locale" (as used by functions not so
3044 parameterized) set for an individual thread. These features were added
3045 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3047 * The functions getresuid, getresgid, setresuid, and setresgid, which
3048 have long been available on Linux, are now declared in <unistd.h>
3049 and are now also available on the Hurd.
3051 * ELF thread-local storage support (TLS) now works on x86-64.
3053 * The new dynamic string token $LIB is expanded in shared library names.
3054 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3056 * Aldy Hernandez contributed complete software floating point support for
3057 PowerPC machines with no FPU.
3059 * fexecve is implemented on Linux.
3061 * The `btowc' function should work at least twice as fast due to
3062 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3064 * With approriate thread add-ons cancelable functions are now implemented
3065 in libc.so as well. No need to call the function in libpthread. This
3066 change allowed to finally disable the incorrect and expensive handling
3067 of weak definition in ld.so.
3069 * Yet more PLT entries in libc.so have been removed. We finally arrived
3070 at the bare minimum. Startup times improved appropriately.
3072 * Support for the new Linux/x86 system call interface was added. The
3073 AT_SYSINFO auxiliary vector entry is recognized and handled.
3077 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3080 * iconv (the program and the interface) now accepts empty names (excluding
3081 options like //TRANSLIT) to mean "use charset of current locale".
3083 * localedef can now transliterate characters in strings which are not in
3084 the provided charmap. The information from the input locale is used.
3086 * Prelinking support was added for ELF targets. This requires additional
3087 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3089 * Read-only stdio streams now use mmap to speed up operation by eliminating
3090 copying and buffer underflows. To use add 'm' to the mode string of
3091 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3093 * The malloc functions were completely rewritten by Wolfram Gloger based
3094 on Doug Lea's malloc-2.7.0.c.
3096 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3097 implementation of regex.
3099 * Bruno Haible upgraded the iconv and locale implementation to support
3102 * Contents of the LC_* and LANG environment variables in the CEN style are
3103 not recognized anymore. It never was used. Change by Ulrich Drepper.
3105 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3106 thread-local storage (TLS) ABI on some platforms.
3107 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3109 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3110 EUC-JISX0213, and TSCII.
3112 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3113 BSD-compatible interface for getting all network interface addresses.
3114 Implementation for IPv4 by Roland McGrath.
3116 * Loading of locale data is faster due to the introduction of a locale
3117 archive. Implemented by Roland McGrath and Ulrich Drepper.
3119 * Startup times are significantly reduced by not using exported functions
3120 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3123 * Steven Munroe contributed a port to PowerPC64/Linux.
3127 * The Hurd now uses the GNU libio implementation of stdio rather than the
3128 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3130 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3131 with the same behavior as the Linux system call of the same name.
3135 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3136 128-bit long double format.
3138 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3139 IBM1161, and IBM1162 charsets.
3141 * Andreas Jaeger contributed a port to x86-64/Linux.
3143 * Peter Bruin contributed a port to PowerPC/Hurd.
3145 * libc functions using I/O streams now can handle wide-oriented streams
3148 * optimizations in the dynamic linker. Binaries created by recent binutils
3149 versions start up quicker due to reduced time spent on relocations.
3151 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3155 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3156 asinh, atanh, j0 for the 128-bit long double format.
3158 * Bruno Haible updated all the code handling Unicode in some form to
3159 support Unicode 3.1.
3161 * Speed of regex for single-byte locales is back to previous levels.
3162 Patch by Isamu Hasegawa.
3164 * Alpha, SPARC, and IA-64 now also using floating stacks.
3166 * Startup time of internationalized applications greatly improved through
3167 iconv cache. Use iconvconfig to generate the cache file.
3168 Contributed by Ulrich Drepper.
3170 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3171 severe performance problems. Patch by David Mosberger.
3173 * The Hurd port got a lot more functionality like AIO, various stdio
3174 extensions, etc. Mainly done by Roland McGrath.
3176 * mtrace can now lookup symbols in shared libraries.
3180 * Intel's IA-64 math library is largely integrated. It provides fast and
3181 accurate implementatations for most basic and standard math functions
3182 in float, double, and long double format.
3184 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3185 and asin for the 96-bit long double format and asin, log, tan for the
3186 128-bit long double format.
3188 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3189 The basic double functions exist today. Contributed by Abraham Ziv
3190 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3191 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3193 * An asynchronous name lookup library was added. The interface is designed
3194 after POSIX AIO. The proposal was circulated beforehand to get comments.
3195 No negative ones came in. Implemented by Ulrich Drepper.
3197 * Port to S390/64bit contributed by Martin Schwidefsky
3198 <schwidefsky@de.ibm.com>.
3200 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3201 of functions for Linux/IA-64.
3203 * The RPC code is now thread safe. Threads can now use the same service
3204 of different services at the same time. Patch by Eric Norum
3205 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3207 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3208 family of functions for Linux/S390.
3210 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3211 of functions for Linux/x86.
3213 * Port to Linux/CRIS contributed by Axis Communications.
3217 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3218 we can now check for namespace violations and missing declarations. The
3219 result is that almost all headers are now Unix-compliant (as defined in
3220 the upcoming XPG6). The negative side is that some programs might need
3221 corrections, too, if they depend on the incorrect form of the headers in
3222 previous versions which defined too many symbols and included too many
3225 * regex now handles multibyte character sets correctly.
3226 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3228 * iconv (the program) does now conform to the upcoming XPG6 and handles
3229 charmaps. Instead of the charset names the path of charmaps can be
3230 provided and the conversion happens based on this data.
3231 Contributed by Ulrich Drepper.
3233 * The locale program now provides detailed information about the installed
3238 only lists the names of the supported locales
3242 provides details such as country, language, and codeset name.
3243 Contributed by Ulrich Drepper.
3247 * The gencat program now parses the input file according to the charset
3248 selected by the LC_CTYPE category. This is important for stateful
3249 character sets. To make generating catalogs easier there is a way
3250 to overwrite the charset selected by the locale: before the first
3251 message or $ quote line the catalog can contain a line like
3253 $ codeset=ISO-8859-2
3255 to select the charset (ISO-8859-2 in this case).
3257 Implemented by Shinya Hanataka and Ulrich Drepper.
3259 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3260 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3261 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3263 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3264 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3266 * The LANGUAGE environment variable is now ignored unless the locale is
3267 changed from the default "C" locale.
3269 * The usual bug fixes.
3273 * Greg McGary added runtime support for bounds checking using gcc's
3274 new -fbounded-pointers option. ix86 target is complete. PowerPC
3277 * Thorsten Kukuk added secure mode support to nscd.
3279 * The Berkeley db libraries have been removed.
3281 Related, the nss_db module is now in a separate package since it
3282 obviously requires a database library being available.
3284 * Wide character I/O streams implemented by Ulrich Drepper.
3286 * Functions from the extended socket API added by Ulrich Drepper.
3288 * Functions feenableexcept and fedisableexcept to control the
3289 behaviour of individual exceptions have been added by Andreas Jaeger.
3291 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3293 * The resolver code has been updated from bind 8.2.3-T5B which supports
3294 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3297 This change could in some situations effect backward compatibility. Since
3298 now `_res' is a thread-local instead of a global variable, modifying it
3299 in one thread does not have any effect in other threads.
3301 The resolver library was also extended to allow IPv6 as the transport
3302 protocol for the requests. This work was done by Stig Venaas.
3304 * Compatibility code for K&R C compilers has been removed from the
3305 header files. A ISO C compiler is needed to use the library
3306 (conforming to either C89 or C99 standard).
3308 * Complete rewrite of the localedef program to support multibyte character
3309 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3310 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3311 Implemented by Ulrich Drepper.
3313 Bruno Haible significantly improved the generation and use of the data
3314 structures for the wide character tables.
3316 * Plural handling in gettext implemented by Ulrich Drepper.
3318 * The utmp daemon has been removed.
3320 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3322 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3325 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3327 * POSIX barriers implemented by Kaz Kylheku.
3329 * POSIX spawn function family implemented by Ulrich Drepper.
3331 * POSIX spinlocks are now available.
3333 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3335 * the configure option --enable-kernel=X.Y.Z allows to strip out
3336 compatibility for kernel versions before X.Y.Z. This is currently only
3337 implemented for Linux.
3339 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3340 field sin6_scope_id. This means that all programs using IPv6 should be
3341 recompiled. Don't expect binary compatibility with previous glibc
3344 * various conversion modules for IBM character sets contributed by
3347 * IA-64 port by Jes Sorensen and HJ Lu.
3361 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3363 * New xdr functions are added; some rpc functions are now 64bit clean.
3365 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3367 * Fixed known incompatibilities with glibc 2.0.
3369 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3371 * Optimized a number of functions (especially the ELF dynamic loader).
3373 * Update timezone data files.
3375 * lots of charmaps corrections
3377 * some new locale definitions and charmaps
3382 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3383 conform to POSIX member type requirements. Symbol versions have been
3384 adjusted as needed within the library, and for direct use by applications,
3385 but there is potential for problems if third-party libraries use
3386 struct timeval as part of their interface. This does not present
3387 a problem for X and other "essential" system libraries.
3389 * An additional locale model to support C++ Standard Library locale
3390 model and probably more was implemented by Ulrich Drepper.
3392 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3395 * Miles Bader provided the `argp' function family to support hierachical
3396 command line argument parsing, layered on top of getopt.
3398 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3400 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3403 * scanf recognizes the %a and %A format for scanning floating point numbers.
3405 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3406 information and interfaces for the available integer types.
3408 * about 130 new math functions were added to implement the ISO C9x math
3411 * the new header <complex.h> contains definitions of the complex math
3412 functions from ISO C 9X.
3414 * the new header <tgmath.h> defines generic macros to use complex or
3415 real valued functions.
3417 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3419 * Andreas Jaeger provided a test suite for the math library.
3421 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3423 * Ulrich Drepper added character set conversion functions (iconv).
3425 * Optimized string functions have been added.
3427 * The localedata addon is now part of glibc.
3429 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3431 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3432 daemon for NSS (nscd).
3434 Missing a better place here are some numbers on improvements. Under
3435 Linux 2.1.125 un-tar-ing the kernel sources takes
3439 using local files 12.19s 6.88s 22.91s
3441 using NIS 13.92s 8.91s 26.34s
3443 using NIS & nscd 10.37s 7.34s 25.30s
3445 using NIS+ 27.57s 30.37s 640.46s
3447 using NIS+ & nscd 10.25s 7.83s 26.51s
3449 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3451 Keep in mind that non-namelookup related operations dominate above times.
3452 It was just a common complain that using NIS+ unpacking the kernel is
3455 [1] The old nscd implementation is not available anymore since it was
3456 distributed with glibc up to version 2.0.98 and thus is now replaced.
3458 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3460 * Mark Kettenis provided a Hesiod NSS module.
3462 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3463 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3465 * rcmd can now handle netgroups (Dick Streefland).
3467 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3470 * Support for the IPv6 protocol has been added to the socket API, as per the
3471 latest draft standards.
3473 * Support for Linux 2.2 has been added.
3475 * Interface changes relative to the latest 2.0.x release:
3476 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3477 addseverity NEW: Unix98
3478 alphasort64 NEW: LFS
3479 argp_err_exit_status NEW: argp, GNU ext
3480 argp_error NEW: argp, GNU ext
3481 argp_failure NEW: argp, GNU ext
3482 argp_help NEW: argp, GNU ext
3483 argp_parse NEW: argp, GNU ext
3484 argp_program_bug_address NEW: argp, GNU ext
3485 argp_program_version NEW: argp, GNU ext
3486 argp_program_version_hook NEW: argp, GNU ext
3487 argp_state_help NEW: argp, GNU ext
3488 argp_usage NEW: argp, GNU ext
3489 authdes_create NEW: Secure RPC
3490 authdes_getucred NEW: Secure RPC
3491 authdes_pk_create NEW: Secure RPC
3492 backtrace NEW: GNU ext.
3493 backtrace_symbols NEW: GNU ext.
3494 backtrace_symbols_fd NEW: GNU ext.
3496 cacosf NEW: ISO C 9x
3497 cacosh NEW: ISO C 9x
3498 cacoshf NEW: ISO C 9x
3499 cacoshl NEW: ISO C 9x
3500 cacosl NEW: ISO C 9x
3507 casinf NEW: ISO C 9x
3508 casinh NEW: ISO C 9x
3509 casinhf NEW: ISO C 9x
3510 casinhl NEW: ISO C 9x
3511 casinl NEW: ISO C 9x
3513 catanf NEW: ISO C 9x
3514 catanh NEW: ISO C 9x
3515 catanhf NEW: ISO C 9x
3516 catanhl NEW: ISO C 9x
3517 catanl NEW: ISO C 9x
3518 cbc_crypt NEW: Secure RPC
3522 ccoshf NEW: ISO C 9x
3523 ccoshl NEW: ISO C 9x
3529 cimagf NEW: ISO C 9x
3530 cimagl NEW: ISO C 9x
3531 clearerr_locked REMOVED
3532 clntunix_create NEW: sunrpc ext
3534 clog10 NEW: ISO C 9x
3535 clog10f NEW: ISO C 9x
3536 clog10l NEW: ISO C 9x
3546 cprojf NEW: ISO C 9x
3547 cprojl NEW: ISO C 9x
3549 crealf NEW: ISO C 9x
3550 creall NEW: ISO C 9x
3555 csinhf NEW: ISO C 9x
3556 csinhl NEW: ISO C 9x
3559 csqrtf NEW: ISO C 9x
3560 csqrtl NEW: ISO C 9x
3564 ctanhf NEW: ISO C 9x
3565 ctanhl NEW: ISO C 9x
3567 des_setparity NEW: Secure RPC
3568 ecb_crypt NEW: Secure RPC
3569 endutxent NEW: Unix98
3571 exp10f NEW: ISO C 9x
3572 exp10l NEW: ISO C 9x
3576 fattach NEW: STREAMS
3577 fdetach NEW: STREAMS
3581 feclearexcept NEW: ISO C 9x
3582 fegetenv NEW: ISO C 9x
3583 fegetexceptflag NEW: ISO C 9x
3584 fegetround NEW: ISO C 9x
3585 feholdexcept NEW: ISO C 9x
3587 feraiseexcept NEW: ISO C 9x
3588 ferror_locked REMOVED
3589 fesetenv NEW: ISO C 9x
3590 fesetexceptflag NEW: ISO C 9x
3591 fesetround NEW: ISO C 9x
3592 fetestexcept NEW: ISO C 9x
3593 feupdateenv NEW: ISO C 9x
3594 fflush_locked REMOVED
3598 fgets_unlocked NEW: GNU ext.
3599 fileno_locked REMOVED
3611 fputc_locked REMOVED
3612 fputs_unlocked NEW: GNU ext.
3613 fread_unlocked NEW: GNU ext.
3618 fstatvfs NEW: Unix98
3622 ftruncate64 NEW: LFS
3624 fwrite_unlocked NEW: GNU ext.
3625 gai_strerror NEW: IPv6
3629 getchar_locked REMOVED
3631 getdate_err NEW: Unix98
3632 getdate_r NEW: GNU ext.
3634 getnameinfo NEW: IPv6
3635 getnetname NEW: Secure RPC
3636 getpmsg NEW: STREAMS
3637 getpt NEW: Unix98 PTY
3638 getrlimit64 NEW: LFS
3639 getutxent NEW: Unix98
3640 getutxid NEW: Unix98
3641 getutxline NEW: Unix98
3642 glob64 NEW: GNU ext.
3643 globfree64 NEW: GNU ext.
3644 gnu_get_libc_release NEW: GNU ext.
3645 gnu_get_libc_version NEW: GNU ext.
3646 grantpt NEW: Unix98 PTY
3647 host2netname NEW: Secure RPC
3649 iconv_close NEW: iconv
3650 iconv_open NEW: iconv
3651 if_freenameindex NEW: IPv6
3652 if_indextoname NEW: IPv6
3653 if_nameindex NEW: IPv6
3654 if_nametoindex NEW: IPv6
3655 in6addr_any NEW: IPv6
3656 in6addr_loopback NEW: IPv6
3657 inet6_isipv4mapped NEW: IPv6
3658 isastream NEW: STREAMS
3659 iswblank NEW: GNU ext.
3660 key_decryptsession NEW: Secure RPC
3661 key_decryptsession_pk NEW: Secure RPC
3662 key_encryptsession NEW: Secure RPC
3663 key_encryptsession_pk NEW: Secure RPC
3664 key_gendes NEW: Secure RPC
3665 key_get_conv NEW: Secure RPC
3666 key_secretkey_is_set NEW: Secure RPC
3667 key_setnet NEW: Secure RPC
3668 key_setsecret NEW: Secure RPC
3669 llrint NEW: ISO C 9x
3670 llrintf NEW: ISO C 9x
3671 llrintl NEW: ISO C 9x
3672 llround NEW: ISO C 9x
3673 llroundf NEW: ISO C 9x
3674 llroundl NEW: ISO C 9x
3679 lrintf NEW: ISO C 9x
3680 lrintl NEW: ISO C 9x
3681 lround NEW: ISO C 9x
3682 lroundf NEW: ISO C 9x
3683 lroundl NEW: ISO C 9x
3685 makecontext NEW: Unix98
3686 mempcpy NEW: GNU ext.
3689 modify_ldt NEW: kernel
3693 nearbyint NEW: ISO C 9x
3694 nearbyintf NEW: ISO C 9x
3695 nearbyintl NEW: ISO C 9x
3696 netname2host NEW: Secure RPC
3697 netname2user NEW: Secure RPC
3698 nexttoward NEW: ISO C 9x
3699 nexttowardf NEW: ISO C 9x
3700 nexttowardl NEW: ISO C 9x
3704 passwd2des NEW: Secure RPC
3706 pow10f NEW: GNU ext.
3707 pow10l NEW: GNU ext.
3710 printf_size NEW: GNU ext.
3711 printf_size_info NEW: GNU ext.
3712 profil_counter REMOVED
3713 pthread_mutexattr_getkind_np REPLACED
3714 pthread_mutexattr_setkind_np REPLACED
3715 ptsname NEW: Unix98 PTY
3716 ptsname_r NEW: Unix98 PTY
3718 putchar_locked REMOVED
3719 putgrent NEW: GNU ext.
3721 putpmsg NEW: STREAMS
3722 pututxline NEW: Unix98
3726 readdir64_r NEW: LFS
3727 remquo NEW: ISO C 9x
3728 remquof NEW: ISO C 9x
3729 remquol NEW: ISO C 9x
3731 roundf NEW: ISO C 9x
3732 roundl NEW: ISO C 9x
3734 scalbln NEW: ISO C 9x
3735 scalblnf NEW: ISO C 9x
3736 scalblnl NEW: ISO C 9x
3738 sendfile NEW: kernel
3739 setcontext NEW: Unix98
3740 setrlimit64 NEW: LFS
3741 setutxent NEW: Unix98
3743 sigignore NEW: Unix98
3744 sigqueue NEW: POSIX.1b
3745 sigrelse NEW: Unix98
3746 sigset NEW: POSIX.1b
3747 sigtimedwait NEW: POSIX.1b
3748 sigwaitinfo NEW: POSIX.1b
3749 sincos NEW: GNU ext.
3750 sincosf NEW: GNU ext.
3751 sincosl NEW: GNU ext.
3755 strcasestr NEW: GNU ext.
3756 strtoimax NEW: ISO C 9x
3757 strtoumax NEW: ISO C 9x
3758 strverscmp NEW: GNU ext.
3759 svcauthdes_stats NEW: Secure RPC
3760 svcunix_create NEW: sunrpc etx
3761 svcunixfd_create NEW: sunrpc ext
3762 swapcontext NEW: Unix98
3763 tcgetsid NEW: Unix98 PTY
3764 tdestroy NEW: GNU ext.
3765 tgamma NEW: ISO C 9x
3766 tgammaf NEW: ISO C 9x
3767 tgammal NEW: ISO C 9x
3771 truncf NEW: ISO C 9x
3772 truncl NEW: ISO C 9x
3774 unlockpt NEW: Unix98 PTY
3775 updwtmpx NEW: Unix98
3776 user2netname NEW: Secure RPC
3777 utmpxname NEW: Unix98
3778 versionsort NEW: GNU ext.
3779 versionsort64 NEW: GNU ext.
3781 wcscasecmp NEW: GNU ext.
3782 wcsncasecmp NEW: GNU ext.
3783 wcsnlen NEW: GNU ext.
3784 wcstoimax NEW: ISO C 9x
3785 wcstoll NEW: ISO C 9x
3786 wcstoull NEW: ISO C 9x
3787 wcstoumax NEW: ISO C 9x
3789 wordexp NEW: POSIX.2
3790 wordfree NEW: POSIX.2
3791 write_profiling REMOVED
3792 xdecrypt NEW: Secure RPC
3793 xdr_authdes_cred NEW: Secure RPC
3794 xdr_authdes_verf NEW: Secure RPC
3795 xdr_cryptkeyarg NEW: Secure RPC
3796 xdr_cryptkeyarg2 NEW: Secure RPC
3797 xdr_cryptkeyres NEW: Secure RPC
3798 xdr_getcredres NEW: Secure RPC
3799 xdr_key_netstarg NEW: Secure RPC
3800 xdr_key_netstres NEW: Secure RPC
3801 xdr_keybuf NEW: Secure RPC
3802 xdr_keystatus NEW: Secure RPC
3803 xdr_netnamestr NEW: Secure RPC
3804 xdr_sizeof NEW: Secure RPC
3805 xdr_unixcred NEW: sunrpc ext
3806 xencrypt NEW: Secure RPC
3807 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3818 * inet_ntoa is thread-safe
3820 * updwtmp is moved from libutil to libc
3822 * rewrite of cbrt function
3824 * update of timezone data
3838 * add atoll function
3840 * fix complex problems in Berkeley DB code
3842 * fix math functions
3846 * fixed lots of header problems (especially Linux/GNU specific)
3848 * dynamic loader preserves all registers
3850 * Roland McGrath provided support for handling of auxiliary objects in
3851 the ELF dynamic loader.
3853 * support for parallel builds is improved
3857 * GNU extensions are no longer declared by default. To enable them you
3858 must define the macro `_GNU_SOURCE' in your program or compile with
3861 * The library has changed from using GNU ld symbol aliases to using weak
3862 symbols where available. The ELF object file format supports weak
3863 symbols; GNU ld also supports weak symbols in the a.out format. (There
3864 is also now support for other GNU ld extensions in ELF. Use the
3865 `--with-elf' option to configure to indicate you have ELF, and
3866 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3867 of many files which contained only symbol aliases, reducing the size of
3868 the source and the compiled library; many other files were renamed to
3869 less cryptic names previously occupied by the symbol alias files.
3870 There is a new header file <elf.h> for programs which operate on
3871 files in the ELF format.
3873 * Converted to Autoconf version 2, so `configure' has more options.
3874 Run `configure --help' to see the details.
3876 * The library can now be configured to build profiling, highly-optimized
3877 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3878 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3879 `configure' enable building these extra libraries. The shared library is
3880 built by default when using both ELF and GNU ld. When shared libraries
3881 are enabled, the new library `-ldl' is available for arbitrary run-time
3882 loading of shared objects; its interface is defined in <dlfcn.h>. The
3883 new header file <link.h> gives access to the internals of the run-time
3884 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3885 application of same name on other systems and it provides information
3886 about dynamically linked binaries.
3888 * The C library now provides the run-time support code for profiling
3889 executables compiled with `-pg'. Programs can control the profiling code
3890 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3891 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3892 the support for this file format was contributed by David Mosberger-Tang.
3894 * The math code has been replaced with a math library based on fdlibm from
3895 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3896 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3897 functions. The math functions now reside in a separate library, so
3898 programs using them will need to use `-lm' their linking commands.
3900 * John C. Bowman contributed optimized ix87 assembler inline functions.
3902 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3903 mechanism similar to that found in Solaris 2. This is now used for the
3904 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3905 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3906 file controls what services are used for each individual database. This
3907 works by loading shared libraries with names specified in `nsswitch.conf',
3908 so service modules can be changed or added at any time without even
3909 relinking any program. Currently there are the file, db, and NIS based
3910 NSS services available.
3912 * The new functions `strtoq' and `strtouq' parse integer values from
3913 strings, like `strtol' and `strtoul', but they return `long long int' and
3914 `unsigned long long int' values, respectively (64-bit quantities).
3916 * The new functions `strtof' and `strtold' parse floating-point values from
3917 strings, like `strtod', but they return `float' and `long double' values,
3918 respectively (on some machines `double' and `long double' are the same).
3920 * Ulrich Drepper has contributed new implementations of the floating-point
3921 printing and reading code used in the `printf' family of functions and
3922 `strtod', `strtof', and `strtold'. These new functions are perfectly
3923 accurate, and much faster than the old ones.
3925 * The implementation of the POSIX locale model was completely rewritten by
3926 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3927 to compile the POSIX locale definition.
3929 * The former dummy implementations of the strcoll and strxfrm function are
3930 now replaced by fully functional code contributed by Ulrich Drepper. The
3931 collation information comes from the POSIX locale definitions.
3933 * The new header <langinfo.h> defines an interface for accessing
3934 various locale-dependent data (using the locale chosen with `setlocale').
3936 * Ulrich Drepper has contributed a new suite of functions for operation on
3937 wide-character and multibyte-character strings, in <wchar.h>;
3938 and classification and case conversion of wide characters, in <wctype.h>.
3939 These new functions are conforming to the ISO C, Amendement 1 specification.
3941 * There is now a second implementation of the standard I/O library available.
3942 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3943 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3945 * You can now use positional parameter specifications in format strings
3946 for the `printf' and `scanf' families of functions. For example,
3947 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3948 ``Number 6, Mr Jones''. This is mainly useful when providing different
3949 format strings for different languages, whose grammars may dictate
3950 different orderings of the values being printed. To support this
3951 feature, the interface for `register_printf_handler' has changed; see
3952 the header file <printf.h> for details.
3954 * The `printf' and `scanf' families of functions now understand a new
3955 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3956 to group numbers as indicated by the locale; for `scanf' and friends, this
3957 says to accept as valid only a number with all the proper grouping
3958 separators in the right places. In the default "C" locale, numbers are
3959 not grouped; but locales for specific countries will define the usual
3960 conventions (i.e. separate thousands with `,' in the US locale).
3962 * The pgrp functions have been regularized, slightly incompatibly but much
3963 less confusingly. The core functions are now `getpgid' and `setpgid',
3964 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3965 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3966 provided for compatibility. There is no longer an incompatible `getpgrp'
3967 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3969 * The new header file <fts.h> and suite of functions simplify programs that
3970 operate on directory trees. This code comes from 4.4 BSD.
3972 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3973 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3974 NSS scheme used in glibc.
3976 * There is a new malloc debugging hook `__memalign_hook'.
3978 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3979 `unsigned int' in <sys/types.h>. These are for compatibility only and
3980 their use is discouraged.
3982 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3983 done differently, so that it works even without GNU ld.
3985 * New function `euidaccess' checks allowed access to a file like `access',
3986 but using the effective IDs instead of the real IDs.
3988 * The time zone data files have been updated for the latest and greatest
3989 local time conventions of the countries of the world.
3991 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
3994 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
3995 for formatting floating-point numbers. They are provided only for
3996 compatibility; new programs should use `sprintf' instead. There are
3997 also equivalent function for the `long double' floating-point type and
3998 all functions also exist in a reentrant form.
4000 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4001 functions for maintaining the login-record files (primarily of use to
4002 system programs such as `login'), and convenient functions for
4003 allocating and initializing a pseudo-terminal (pty) device.
4005 * Ulrich Drepper has contributed new support for System V style
4006 shared memory and IPC on systems that support it.
4008 * Ulrich Drepper has contributed several miscellaneous new functions found
4009 in System V: The `hsearch' family of functions provide an effective
4010 implementation of hash tables; `a64l' and `l64a' provide a very simple
4011 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4014 * Ulrich Drepper has contributed new reentrant counterparts for the
4015 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4017 * Ulrich Drepper has contributed new, highly-optimized versions of several
4018 string functions for the i486/Pentium family of processors.
4020 * Ulrich Drepper has updated the Linux-specific code, based largely
4021 on work done in Hongjiu Lu's version of GNU libc for Linux.
4022 The GNU library now supports Linux versions 2.0.10 and later,
4023 using the ELF object file format (i[3456]86-*-linux).
4025 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4027 * David Mosberger-Tang and Richard Henderson have ported the C library
4028 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4029 dynamic linking support for ELF/Alpha.
4031 * Richard Henderson contributed several Alpha optimized assembler function
4032 for arithmetic and string handling.
4034 * Ulrich Drepper has contributed a new set of message catalog functions to
4035 support multiple languages using the <libintl.h> interface, for use with
4036 his new package GNU gettext. Translation volunteers have contributed
4037 catalogs of the library's messages in Spanish, German, and Korean.
4039 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4040 program and the `catgets' function for reading the catalog files it
4041 creates. (The <libintl.h> interface is preferred; we include the
4042 <nl_types.h> interface using `catgets' only for source compatibility with
4043 programs already written to use it.)
4045 * New header file <values.h> gives SVID-compatible names for <limits.h>
4048 * Various new macros, declarations, and small header files for compatibility
4051 * New function `group_member' is a convenient way to check if a process has
4052 a given effective group ID.
4054 * When using GCC 2.7 and later, the socket functions are now declared in a
4055 special way so that passing an argument of type `struct sockaddr_in *',
4056 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4057 `struct sockaddr *' type, does not generate a type-clash warning.
4059 * New function `error' declared in header file <error.h> is a convenient
4060 function for printing error messages and optionally exiting; this is the
4061 canonical function used in GNU programs. The new functions `err', `warn',
4062 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4063 doing the same thing.
4065 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4066 POSIX.2 `glob' function to do ~ and {...} expansion.
4068 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4069 `clearenv' which is used in POSIX.9 is also available.
4071 * New function `getsid' returns session ID number on systems that support it.
4073 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4074 files <db.h> and <mpool.h> provide a rich set of functions for several
4075 types of simple databases stored in memory and in files, and <ndbm.h> is
4076 an old `ndbm'-compatible interface using the `db' functions. Link with
4077 `-ldb' to get these functions.
4079 * New macro `strdupa' copies a string like `strdup', but uses local stack
4080 space from `alloca' instead of dynamic heap space from `malloc'.
4082 * New function `strnlen' is like `strlen' but searches only a given maximum
4083 number of characters for the null terminator. `stpncpy', `strndup' and
4084 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4087 * New function `statfs' in header <sys/statfs.h>.
4089 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4090 provide convenient functions for operating on blocks of null-terminated
4093 * A new suite of functions in <utmp.h> handle all the details of reading
4094 and writing the utmp file.
4096 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4099 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4100 completely POSIX compliant and also implemented the extended functionality
4101 to handle alternate digit representation and alternate era date formats.
4103 * Ulrich Drepper provided an implementation of the `strptime' function
4104 defined in XPG4.2 which transforms a string into a `struct tm' value.
4106 * Paul Eggert provided the tzselect shell script as part of the timezone
4107 code. The shell script makes it easy to select the correct timezone
4110 * The implementation of the malloc family of functions is completely replaced
4111 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4112 The implementation uses the mmap function (if available) and it is
4113 optimized for the use in multi threaded programs.
4115 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4116 functions. This new functionality is usable by specifying a special
4117 salt string and it is compatible with implementation on *BSD systems.
4119 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4120 `getsubopt' to handle second level command line options, `bsd_signal'
4121 to access BSD style `signal' functionality, the obsolete `regexp' style
4124 * the `lchown' function is available on system which support this
4127 * The implementation of the shadow password handling function was contributed
4130 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4132 * POSIX.1g support was added. The <sys/select.h> header is available,
4133 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4134 implementation of `getaddrinfo'.
4138 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4140 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4141 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4143 * The new function `daemon' from 4.4 BSD is useful for server programs that
4144 want to put themselves in the background.
4146 * Joel Sherrill has contributed support for several standalone boards that
4147 run without an operating system.
4149 * `printf', `scanf' and friends now accept a `q' type modifier for long
4150 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4152 * All of the code taken from BSD (notably most of the math and networking
4153 routines) has been updated from the BSD 4.4-Lite release.
4155 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4157 * The new functions `getdomainname' and `setdomainname' fetch or change the
4158 YP/NIS domain name. These are system calls which exist on systems which
4161 * The time zone data files have been updated for the latest international
4164 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4165 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4169 * The C library now includes support for Sun RPC, from Sun's free
4170 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4171 are included. (There is still no support for YP.)
4173 * Tom Quinn has contributed a port of the C library to SGI machines running
4174 Irix 4 (mips-sgi-irix4).
4176 * The new `lockf' function is a simplified interface to the locking
4177 facilities of `fcntl', included for compatibility.
4179 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4181 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4183 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4186 * The error message printed by `assert' for a failed assertion now includes
4187 the name of the program (if using GNU ld) and the name of the calling
4188 function (with versions of GCC that support this).
4190 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4192 * The library now includes the <sys/mman.h> header file and memory
4193 management functions `mmap', `munmap', `mprotect', `msync', and
4194 `madvise', on systems that support those facilities.
4196 * The interface for `mcheck' has changed slightly: the function called to
4197 abort the program when an allocation inconsistency is detected now takes
4198 an argument that indicates the type of failure. The new function
4199 `mprobe' lets you request a consistency check for a particular block at
4200 any time (checks are normally done only when you call `free' or `realloc'
4203 * It is now possible to easily cross-compile the C library, building on one
4204 system a library to run on another machine and/or operating system. All
4205 you need to do is set the variable `HOST_CC' in `configparms' to the
4206 native compiler for programs to run on the machine you are building on (a
4207 few generator programs are used on Unix systems); set `CC' to the
4210 * The new function `fexecve' (only implemented on the GNU system) executes
4211 a program file given a file descriptor already open on the file.
4215 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4216 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4218 * You can set the variable `libprefix' in `configparms' to specify a prefix
4219 to be prepended to installed library files; this makes it easy to install
4220 the GNU C library to be linked as `-lgnuc' or whatever.
4222 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4223 copies a limited number of characters from a string, and returns the
4224 address of the last character written.
4226 * You no longer need to check for whether the installed `stddef.h' is
4227 compatible with the GNU C library. configure now checks for you.
4229 * You can now define a per-stream `fileno' function to convert the
4230 stream's cookie into an integral file descriptor.
4232 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4233 allocates zero bytes of storage, and returns a unique pointer which
4234 you can pass to `realloc' or `free'. The behavior is undefined if
4235 you dereference this pointer.
4237 * The C library now runs on Sony NEWS m68k machines running either
4238 NewsOS 3 or NewsOS 4.
4240 * The new `syscall' function is a system-dependent primitive function
4241 for invoking system calls. It has the canonical behavior on Unix
4242 systems, including unreliable return values for some calls (such as
4243 `pipe', `fork' and `getppid').
4245 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4246 to `EAGAIN', which is the preferred name. On systems whose kernels
4247 use two distinct codes, the C library now translates EWOULDBLOCK to
4248 EAGAIN in every system call function.
4252 * The GNU C Library Reference Manual is now distributed with the library.
4253 `make dvi' will produce a DVI file of the printed manual.
4254 `make info' will produce Info files that you can read on line using C-h i
4255 in Emacs or the `info' program.
4256 Please send comments on the manual to bug-glibc-manual@gnu.org.
4258 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4260 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4262 * Jason Merrill has contributed a port to the Sequent Symmetry running
4263 Dynix version 3 (i386-sequent-dynix).
4265 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4266 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4268 * New function `memory_warnings' lets you arrange to get warnings when
4269 malloc is running out of memory to allocate, like Emacs gives you.
4271 * The C library now contains the relocating allocator used in Emacs 19 for
4272 its editing buffers. This allocator (ralloc) minimizes allocation
4273 overhead and fragmentation by moving allocated regions around whenever it
4274 needs to. You always refer to a ralloc'd region with a "handle" (a
4275 pointer to a pointer--an object of type `void **').
4277 * There is a new `printf' format: `%m' gives you the string corresponding
4278 to the error code in `errno'.
4280 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4281 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4282 pass, the `a' modifier says to fill in a `char **' you pass with a
4285 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4286 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4287 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4289 * `mkstemp' is a traditional Unix function to atomically create and open a
4290 uniquely-named temporary file.
4294 * The standard location for the file that says what the local timezone is
4295 has changed again. It is now `/usr/local/etc/localtime' (or more
4296 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4298 * The distribution no longer contains any files with names longer than 14
4301 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4302 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4304 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4308 * The configuration process has changed quite a bit. The `configure'
4309 script is now used just like the configuration scripts for other GNU
4310 packages. The `sysdeps' directory hierarchy is much rearranged.
4311 The file `INSTALL' explains the new scheme in detail.
4313 * The header files no longer need to be processed into ANSI C and
4314 traditional C versions. There is just one set of files to install, and
4315 it will work with ANSI or old C compilers (including `gcc -traditional').
4317 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4318 MIPS DECStation running Ultrix 4.
4320 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4321 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4322 made itself into a shared library.
4324 * Yet further improved support for the i386, running 4.3 BSD-like systems
4325 (such as Mach 3 with the Unix single-server), or System V.
4327 * New function `strncasecmp' to do case-insensitive string comparison
4328 with limited length.
4330 * New function `strsep' is a reentrant alternative to `strtok'.
4332 * New functions `scandir' and `alphasort' for searching directories.
4334 * New function `setenv' is a better interface to `putenv'.
4336 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4337 function for traversing a directory tree.
4339 * The GNU obstack package is now also part of the C library.
4340 The new function `open_obstack_stream' creates a stdio stream that
4341 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4342 formatted output directly to an obstack.
4344 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4345 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4347 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4349 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4350 things to your strings.
4352 * There are some new test programs: test-fseek, testmb, and testrand.
4354 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4355 These ports are not finished, but are a good starting place for really
4356 supporting those systems.
4358 * `/etc/localtime' is now the standard location for the file that says what
4359 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4360 This follows the general principle that `/etc' is the place for all local
4361 configuration files.
4363 * The C library header files now use `extern "C"' when used by the C++
4364 compiler, so the C library should now work with C++ code.
4366 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4367 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4370 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4371 function which uses the `merge sort' algorithm, and is said to be
4372 significantly faster than the old GNU `qsort' function. Merge sort is
4373 now the standard `qsort' function. The new algorithm can require a lot
4374 of temporary storage; so, the old sorting function is called when the
4375 required storage is not available.
4377 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4378 provides the Unix `crypt' function, plus some other entry points.
4380 * The code and header files taken from 4.4 BSD have been updated with the
4381 latest files released from Berkeley.
4383 ----------------------------------------------------------------------
4384 Copyright information:
4386 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4388 Permission is granted to anyone to make or distribute verbatim copies
4389 of this document as received, in any medium, provided that the
4390 copyright notice and this permission notice are preserved,
4391 thus giving the recipient permission to redistribute in turn.
4393 Permission is granted to distribute modified versions
4394 of this document, or of portions of it,
4395 under the above conditions, provided also that they
4396 carry prominent notices stating who last changed them.
4399 version-control: never