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 executables 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 and x32 with binutils 2.29 or later, and on
20 aarch64 with binutils 2.30 or later.
22 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
23 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
26 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
28 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
30 * In order to support faster and safer process termination the malloc API
31 family of functions will no longer print a failure address and stack
32 backtrace after detecting heap corruption. The goal is to minimize the
33 amount of work done after corruption is detected and to avoid potential
34 security issues in continued process execution. Reducing shutdown time
35 leads to lower overall process restart latency, so there is benefit both
36 from a security and performance perspective.
38 * The abort function terminates the process immediately, without flushing
39 stdio streams. Previous glibc versions used to flush streams, resulting
40 in deadlocks and further data corruption. This change also affects
41 process aborts as the result of assertion failures.
43 * On platforms where long double has the IEEE binary128 format (aarch64,
44 alpha, mips64, s390 and sparc), the math library now implements _Float128
45 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015. These
46 are the same interfaces added in version 2.26 for some platforms where
47 this format is supported but is not the format of long double.
49 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
50 mips64, powerpc64le, s390, sparc and x86_64), the math library now
51 implements interfaces for that type, as defined by ISO/IEC TS
52 18661-3:2015. These are corresponding interfaces to those supported for
55 * The math library now implements interfaces for the _Float32, _Float64 and
56 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
57 corresponding interfaces to those supported for _Float128.
59 * glibc now implements the memfd_create and mlock2 functions on Linux.
61 * Support for memory protection keys was added. The <sys/mman.h> header now
62 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
65 * The copy_file_range function was added.
67 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
69 * The ldconfig utility now processes `include' directives using the C/POSIX
70 collation ordering. Previous glibc versions used locale-specific
71 ordering, the change might break systems that relied on that.
73 * Support for two grammatical forms of month names has been added.
74 In a call to strftime, the "%B" and "%b" format specifiers will now
75 produce the grammatical form required when the month is used as part
76 of a complete date. New "%OB" and "%Ob" specifiers produce the form
77 required when the month is named by itself. For instance, in Greek
78 and in many Slavic and Baltic languages, "%B" will produce the month
79 in genitive case, and "%OB" will produce the month in nominative case.
81 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
82 are all valid and will all accept any known form of month
83 name---standalone or complete, abbreviated or full. In a call to
84 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
85 the strings used by "%B" and "%b", respectively. New query
86 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
87 used by "%OB" and "%Ob", respectively.
89 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
90 define the strings for %OB and %Ob, respectively; these have the
91 same syntax as "mon" and "ab_mon".
93 This feature is currently a GNU extension, but it is expected to
94 be added to the next revision of POSIX, and it is also already
95 available on some BSD-derived operating systems.
97 This feature will cause existing statically compiled applications
98 to fail to load locales and fall back to the builtin C/POSIX locales.
99 See notes below for other changes affecting compatibility.
101 Deprecated and removed features, and other changes affecting compatibility:
103 * Statically compiled applications attempting to load locales compiled for the
104 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
105 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
106 support for two grammatical forms of the month names, also extends the locale
107 data binary format. Static applications needing locale support must be
108 recompiled to match the runtime and data they are deployed with. In some
109 distributions there is an upgrade window where dynamically linked applications
110 may use a new library but the old locale data and also fall back to the
111 builtin C/POSIX locales; restarting the application process is sufficient to
114 * Support for statically linked applications which call dlopen is deprecated
115 and will be removed in a future version of glibc. Applications which call
116 dlopen need to be linked dynamically instead.
118 * Support for old programs which use internal stdio data structures and
119 functions is deprecated. This includes programs which use the C++ streams
120 provided by libstdc++ in GCC 2.95. Programs which use the internal
121 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
122 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
123 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
124 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
125 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
126 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
127 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
128 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
129 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
130 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
131 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
132 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
133 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
134 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
135 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
136 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
137 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
138 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
139 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
140 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
141 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
142 future version of glibc. Unlike other symbol removals, these old
143 applications will not be supported using compatibility symbols.
145 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
146 defined by <sys/ptrace.h>.
148 * libm no longer supports SVID error handling (calling a user-provided
149 matherr function on error) or the _LIB_VERSION variable to control error
150 handling. (SVID error handling and the _LIB_VERSION variable still work
151 for binaries linked against older versions of the GNU C Library.) The
152 libieee.a library is no longer provided. math.h no longer defines struct
153 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
154 TLOSS, PLOSS and HUGE.
156 * The libm functions pow10, pow10f and pow10l are no longer supported for
157 new programs. Programs should use the standard names exp10, exp10f and
158 exp10l for these functions instead.
160 * The mcontext_t type is no longer the same as struct sigcontext. On
161 platforms where it was previously the same, this changes the C++ name
162 mangling for interfaces involving this type.
164 * The add-ons mechanism for building additional packages at the same time as
165 glibc has been removed. The --enable-add-ons configure option is now
168 * The --without-fp configure option is now ignored. Whether hardware
169 floating-point instructions are used is now configured based on whether
170 the compiler used at configure time (without any options implied by a
171 --with-cpu= configure option) uses such instructions.
173 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
174 the specified string can be parsed as a domain name.
176 * In the malloc_info output, the <heap> element may contain another <aspace>
177 element, "subheaps", which contains the number of sub-heaps.
179 * The libresolv function p_secstodate is no longer supported for new
182 * The tilepro-*-linux-gnu configuration is no longer supported.
184 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
185 and will be removed in a future release. Software that is still using
186 either header should be updated to use standard <stdio.h> interfaces
189 libio.h was originally the header for a set of supported GNU extensions,
190 but they have not been maintained as such in many years, they are now
191 standing in the way of improvements to stdio, and we don't think there are
192 any remaining external users. _G_config.h was never intended for public
193 use, but predates the bits convention.
195 Changes to build and runtime requirements:
197 * bison version 2.7 or later is required to generate code in the 'intl'
200 Security related changes:
202 CVE-2009-5064: The ldd script would sometimes run the program under
203 examination directly, without preventing code execution through the
204 dynamic linker. (The glibc project disputes that this is a security
205 vulnerability; only trusted binaries must be examined using the ldd
208 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
209 suffered from a one-byte overflow during ~ operator processing (either
210 on the stack or the heap, depending on the length of the user name).
211 Reported by Tim Rühsen.
213 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
214 would sometimes fail to free memory allocated during ~ operator
215 processing, leading to a memory leak and, potentially, to a denial
218 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
219 without GLOB_NOESCAPE, could write past the end of a buffer while
220 unescaping user names. Reported by Tim Rühsen.
222 CVE-2017-17426: The malloc function, when called with an object size near
223 the value SIZE_MAX, would return a pointer to a buffer which is too small,
224 instead of NULL. This was a regression introduced with the new malloc
225 thread cache in glibc 2.26. Reported by Iain Buclaw.
227 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
228 to the allocation of too much memory. (This is not a security bug per se,
229 it is mentioned here only because of the CVE assignment.) Reported by
232 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
233 of the number of search path components. (This is not a security
234 vulnerability per se because no trust boundary is crossed if the fix for
235 CVE-2017-1000366 has been applied, but it is mentioned here only because
236 of the CVE assignment.) Reported by Qualys.
238 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
239 for AT_SECURE or SUID binaries could be used to load libraries from the
242 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
243 succeeds without returning an absolute path due to unexpected behaviour
244 of the Linux kernel getcwd syscall. Reported by halfdog.
246 The following bugs are resolved with this release:
248 [The release manager will add the list generated by
249 scripts/list-fixed-bugs.py just before the release.]
256 * A per-thread cache has been added to malloc. Access to the cache requires
257 no locks and therefore significantly accelerates the fast path to allocate
258 and free small amounts of memory. Refilling an empty cache requires locking
259 the underlying arena. Performance measurements show significant gains in a
260 wide variety of user workloads. Workloads were captured using a special
261 instrumented malloc and analyzed with a malloc simulator. Contributed by
262 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
264 * Unicode 10.0.0 Support: Character encoding, character type info, and
265 transliteration tables are all updated to Unicode 10.0.0, using
266 generator scripts contributed by Mike FABIAN (Red Hat).
267 These updates cause user visible changes, especially the changes in
268 wcwidth for many emoji characters cause problems when emoji sequences
269 are rendered with pango, see for example:
270 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
272 * Collation of Hungarian has been overhauled and is now consistent with "The
273 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
276 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
278 - The GNU C Library will now detect when /etc/resolv.conf has been
279 modified and reload the changed configuration. The new resolver option
280 “no-reload” (RES_NORELOAD) disables this behavior.
282 - The GNU C Library now supports an arbitrary number of search domains
283 (configured using the “search” directive in /etc/resolv.conf);
284 previously, there was a hard limit of six domains. For backward
285 compatibility, applications that directly modify the ‘_res’ global
286 object are still limited to six search domains.
288 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
289 Library will now randomly pick a name server from the configuration as a
290 starting point. (Previously, the second name server was always used.)
292 * The tunables feature is now enabled by default. This allows users to tweak
293 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
295 * New function reallocarray, which resizes an allocated block (like realloc)
296 to the product of two sizes, with a guaranteed clean failure upon integer
297 overflow in the multiplication. Originally from OpenBSD, contributed by
298 Dennis Wölfing and Rüdiger Sonderfeld.
300 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
301 These are extended versions of preadv and pwritev, respectively, taking an
302 additional flags argument. The set of supported flags depends on the
303 running kernel; full support currently requires kernel 4.7 or later.
305 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
306 create a new session ID for the spawned process. This feature is
307 scheduled to be added to the next major revision of POSIX; for the time
308 being, it is available under _GNU_SOURCE.
310 * errno.h is now safe to use from C-preprocessed assembly language on all
311 supported operating systems. In this context, it will only define the
312 Exxxx constants, as preprocessor macros expanding to integer literals.
314 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
315 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
316 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
317 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
319 To compile programs that use this feature, the compiler must support
320 128-bit floating point with the type name _Float128 (as defined by TS
321 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
322 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
323 must be defined to make the new interfaces visible.
325 The new functions and macros correspond to those present for other
326 floating-point types (except for a few obsolescent interfaces not
327 supported for the new type), with F128 or f128 suffixes; for example,
328 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
329 printf or scanf formats for the new type; the strfromf128 and strtof128
330 interfaces should be used instead.
332 Deprecated and removed features, and other changes affecting compatibility:
334 * The synchronization that pthread_spin_unlock performs has been changed to
335 now be equivalent to a C11 atomic store with release memory order to the
336 spin lock's memory location. Previously, several (but not all)
337 architectures used stronger synchronization (e.g., containing what is
338 often called a full barrier). This change can improve performance, but
339 may affect odd fringe uses of spin locks that depend on the previous
340 behavior (e.g., using spin locks as atomic variables to try to implement
341 Dekker's mutual exclusion algorithm).
343 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
346 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
347 will only be built and installed when the GNU C Library is configured with
348 --enable-obsolete-rpc. This allows alternative RPC implementations, such
349 as TIRPC or rpcsvc-proto, to be used.
351 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
352 libnss_compat, are deprecated, and will not be built or installed by
355 The NIS(+) support library, libnsl, is also deprecated. By default, a
356 compatibility shared library will be built and installed, but not headers
357 or development libraries. Only a few NIS-related programs require this
358 library. (In particular, the GNU C Library has never required programs
359 that use 'gethostbyname' to be linked with libnsl.)
361 Replacement implementations based on TIRPC, which additionally support
362 IPv6, are available from <https://github.com/thkukuk/>. The configure
363 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
364 name service modules, to be built and installed.
366 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
367 support is enabled, the configured recursive resolver must support EDNS.
368 (Responding to EDNS-enabled queries with responses which are not
369 EDNS-enabled is fine, but FORMERR responses are not.)
371 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
372 servers have not supported this opcode for a long time.
374 * The _res_opcodes variable has been removed from libresolv. It had been
375 exported by accident.
377 * <string.h> no longer includes inline versions of any string functions,
378 as this kind of optimization is better done by the compiler. The macros
379 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
381 * The nonstandard header <xlocale.h> has been removed. Most programs should
382 use <locale.h> instead. If you have a specific need for the definition of
383 locale_t with no other declarations, please contact
384 libc-alpha@sourceware.org and explain.
386 * The obsolete header <sys/ultrasound.h> has been removed.
388 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
390 * The obsolete function cfree has been removed. Applications should use
393 * The stack_t type no longer has the name struct sigaltstack. This changes
394 the C++ name mangling for interfaces involving this type.
396 * The ucontext_t type no longer has the name struct ucontext. This changes
397 the C++ name mangling for interfaces involving this type.
399 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
400 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
401 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
402 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
403 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
404 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
405 fpregset_t type no longer has the name struct fpu. This changes the C++
406 name mangling for interfaces involving those types.
408 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
409 synced with the kernel:
411 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
412 are not supported on this architecture and have been removed.
414 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
415 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
416 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
418 Programs that assume the GET/SETREGS ptrace requests are universally
419 available will now fail to build, instead of malfunctioning at runtime.
421 Changes to build and runtime requirements:
423 * Linux kernel 3.2 or later is required at runtime, on all architectures
424 supported by that kernel. (This is a change from version 2.25 only for
427 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
429 * On most architectures, GCC 4.9 or later is required to build the GNU C
430 Library. On powerpc64le, GCC 6.2 or later is required.
432 Older GCC versions and non-GNU compilers are still supported when
433 compiling programs that use the GNU C Library. (We do not know exactly
434 how old, and some GNU extensions to C may be _de facto_ required. If you
435 are interested in helping us make this statement less vague, please
436 contact libc-alpha@sourceware.org.)
438 Security related changes:
440 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
441 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
443 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
444 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
446 * Avoid printing a backtrace from the __stack_chk_fail function since it is
447 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
450 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
451 fixed (CVE-2017-12133).
453 The following bugs are resolved with this release:
455 [984] network: Respond to changed resolv.conf in gethostbyname
456 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
457 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
459 [12189] libc: __stack_chk_fail should not attempt a backtrace
461 [14096] time: Race condition on timezone/tst-timezone.out
462 [14172] localedata: az_IR: new locale
463 [14995] build: glibc fails to build if gold is the default linker, even if
465 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
466 [16637] network: inet_pton function is accepting IPv6 with bad format
467 [16640] string: string/strtok.c: undefined behaviour inconsistent between
468 x86 and other generic code
469 [16875] localedata: ko_KR: fix lang_name
470 [17225] localedata: ar_SY: localized month names for May and June are
472 [17297] localedata: da_DK: wrong date_fmt string
473 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
474 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
475 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
476 [19066] localedata: ar_SA abbreviated day and month names are in English
477 [19569] network: resolv: Support an arbitrary number of search domains
478 [19570] network: Implement random DNS server selection in the stub
480 [19838] locale: localedef fails on PA-RISC
481 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
482 order of 0D36 and 0D37
483 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
485 [20098] libc: FAIL: debug/backtrace-tst on hppa
486 [20257] network: sunrpc: clntudp_call does not enforce timeout when
488 [20275] localedata: locale day/abday/mon/abmon should not have trailing
490 [20313] localedata: Update locale data to Unicode 9.0
491 [20424] manual: Document how to provide a malloc replacement
492 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
494 [20686] locale: Add el_GR@euro to SUPPORTED.
495 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
496 failures consistently
497 [21015] dynamic-link: Document and fix --enable-bind-now
498 [21016] nptl: pthread_cond support is broken on hppa
499 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
500 frame-pointer on i386
501 [21049] libc: segfault in longjmp_chk() due to clobbered processor
503 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
504 [21088] libc: Build fails with --enable-static-nss
505 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
506 generic c code is used
507 [21109] libc: Tunables broken on big-endian
508 [21112] math: powf has large ulp errors with base close to 1 and exponent
510 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
512 [21120] malloc: glibc malloc is incompatible with GCC 7
513 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
515 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
516 (-0.0) when linking with -lieee
517 [21171] math: log10, log2 and lgamma return incorrect results
518 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
519 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
520 [21207] localedata: ce_RU: update weekdays from CLDR
521 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
522 [21217] localedata: Update months from CLDR-31
523 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
524 [21243] libc: support_delete_temp_file should issue warning for failed
526 [21244] libc: support resolv_test_start() socket fd close should be
528 [21253] libc: localedef randomly segfaults when using -fstack-check due to
529 new posix_spawn implementation
530 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
531 leads to lower CPU frequency
532 [21259] libc: [alpha] termios.h missing IXANY for POSIX
533 [21261] libc: [sparc64] bits/setjmp.h namespace
534 [21267] network: [mips] bits/socket.h IOC* namespace
535 [21268] libc: [alpha] termios.h NL2, NL3 namespace
536 [21270] libc: mmap64 silently truncates large offset values
537 [21275] libc: posix_spawn always crashes on ia64 now
538 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
539 [21280] math: [powerpc] logbl for POWER7 return incorrect results
540 [21289] libc: Incorrect declaration for 32-bit platforms with
541 _FILE_OFFSET_BITS=64 causes build error
542 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
543 not support gethostbyname4_r
544 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
546 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
548 [21340] libc: Support POSIX_SPAWN_SETSID
549 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
550 [21359] network: ns_name_pack needs additional byte in destination buffer
551 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
552 against fragmentation attacks (CVE-2017-12132)
553 [21369] network: resolv: Remove EDNS fallback
554 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
556 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
557 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
558 [21393] stdio: Missing dup3 error check in freopen, freopen64
559 [21396] libc: Use AVX2 memcpy/memset on Skylake server
560 [21399] localedata: Bad description for U00EC in
561 localedata/charmaps/CP1254
562 [21411] malloc: realloc documentation error
563 [21426] network: sys/socket.h uio.h namespace
564 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
565 [21445] libc: signal.h bsd_signal namespace
566 [21455] network: Network headers stdint.h namespace
567 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
569 [21475] network: resolv: Overlong search path is truncated mid-label
570 [21511] libc: sigstack namespace
571 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
572 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
573 bits/syscall.h: No such file or directory
574 [21517] libc: struct sigaltstack namespace
575 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
576 [21533] localedata: Update locale data to Unicode 10.0
578 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
579 junk at end of line, first unrecognized character is `@'
580 [21538] libc: SIG_HOLD missing for XPG4
581 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
582 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
583 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
584 [21543] libc: sigevent namespace
585 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
586 size for general purpose registers in mcontext_t structure
587 [21550] libc: sigwait namespace
588 [21552] libc: XPG4 bsd_signal namespace
589 [21554] libc: sigpause namespace
590 [21560] libc: sys/wait.h signal.h namespace
591 [21561] libc: waitid namespace
592 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
593 '/usr/include/stdlib.h': Permission denied
594 [21575] libc: sys/wait.h missing struct rusage definition
595 [21584] libc: sigaltstack etc namespace
596 [21597] libc: siginterrupt namespace
597 [21607] math: hppa: FAIL: math/test-tgmath
598 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
600 [21622] libc: [tile] missing SA_* for POSIX.1:2008
601 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
602 stack and heap (CVE-2017-1000366)
603 [21625] libc: wait3 namespace
604 [21654] nss: Incorrect pointer alignment in NSS group merge result
606 [21657] network: Parse interface zone id for node-local multicast
607 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
609 [21666] libc: .symver is used on common symbol
610 [21668] network: resolv: res_init cross-thread broadcast introduces race
612 [21687] math: tgmath.h totalorder, totalordermag return type
613 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
615 [21696] libc: Incorrect assumption of of __cpu_mask in
616 posix/sched_cpucount.c
617 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
618 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
620 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
621 leading to relocation crash
622 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
623 `resolv_conf_matches (resp, conf)' failed.
624 [21710] localedata: Added Samoan language locale for Samoa
625 [21711] localedata: Pashto yesstr/nostr locale are missing
626 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
627 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
628 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
629 [21723] localedata: yesstr/nostr missing for Chinese language locale
630 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
632 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
634 [21728] localedata: New Locale for Tongan language
635 [21729] localedata: incorrect LC_NAME fields for hi_IN
636 [21733] localedata: yesstr and nostr are missing for zh_HK
637 [21734] localedata: Missing yesstr and nostr are for kw_GB
638 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
639 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
640 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
641 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
643 [21744] libc: Tests failing on --enable-tunables --enable-stack-
645 [21749] localedata: Wrong abbreviated day name (“abday”) for
647 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
648 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
649 [21759] localedata: missing yesstr and nostr for Tigrinya
650 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
651 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
652 [21767] localedata: Missing Bislama locales
653 [21768] localedata: Missing yesstr and nostr for aa_ET
654 [21770] localedata: Missing Field in li_NL
655 [21778] nptl: Robust mutex may deadlock
656 [21779] libc: MicroBlaze segfaults when loading libpthread
657 [21783] localedata: Fix int_select international_call_prefixes
658 [21784] localedata: Inconsistency in country_isbn
659 [21788] localedata: Missing Country Postal Abbreviations
660 [21794] localedata: Added-country_isbn-for-Italy
661 [21795] localedata: Add/Fix country_isbn for France
662 [21796] localedata: Added country_isbn for Republic of Korea
663 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
664 [21799] localedata: Added int_select international_call_prefixes
665 [21801] localedata: Added int_select international_call_prefixes
666 [21804] nptl: Double semicolon in thread-shared-types.h
667 [21807] localedata: LC_ADDRESS fix for pap_CW
668 [21808] localedata: Fix LC_ADDRESS for pap_AW
669 [21821] localedata: Added country_name in mai_IN
670 [21822] localedata: Fix LC_TIME for mai_IN
671 [21823] localedata: missing yesstr, nostr for sa_IN
672 [21825] localedata: Fix name_mrs for mag_IN
673 [21828] localedata: 2.26 changelog should mention user visible changes
675 [21835] localedata: Added Maithili language locale for Nepal
676 [21838] localedata: Removed redundant data for the_NP
677 [21839] localedata: Fix LC_MONETARY for ta_LK
678 [21844] localedata: Fix Latin characters and Months Sequence.
679 [21848] localedata: Fix mai_NP Title Name
684 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
685 24731-2:2010, is supported to enable declarations of functions from that
686 TR. Note that not all functions from that TR are supported by the GNU C
689 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
690 18661-1:2014, is supported to enable declarations of functions and macros
691 from that TS. Note that not all features from that TS are supported by
694 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
695 18661-4:2015, is supported to enable declarations of functions and macros
696 from that TS. Note that most features from that TS are not supported by
699 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
700 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
701 Since the GNU C Library defaults to a much newer revision of POSIX, this
702 will only affect programs that specifically request an old conformance
703 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
704 see a change in the visible declarations, but a program compiled with
705 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
708 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
709 defined by all multithreaded code, but glibc has not required this for
712 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
713 means that in a future release, the macros “major”, “minor”, and “makedev”
714 will only be available from <sys/sysmacros.h>.
716 These macros are not part of POSIX nor XSI, and their names frequently
717 collide with user code; see for instance glibc bug 19239 and Red Hat bug
718 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
719 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
722 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
723 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
724 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
726 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
727 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
728 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
729 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
730 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
731 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
732 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
733 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
734 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
735 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
736 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
739 * New <math.h> features are added from TS 18661-1:2014:
741 - Signaling NaN macros: SNANF, SNAN, SNANL.
743 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
744 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
745 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
747 - llogb functions: the llogb, llogbf and llogbl functions, and the
748 FP_LLOGB0 and FP_LLOGBNAN macros.
750 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
753 - Comparison macros: iseqsig.
755 - Classification macros: iscanonical, issubnormal, iszero.
757 - Total order functions: totalorder, totalorderf, totalorderl,
758 totalordermag, totalordermagf, totalordermagl.
760 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
762 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
763 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
765 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
766 are added to libc. They convert a floating-point number into string.
768 * Most of glibc can now be built with the stack smashing protector enabled.
769 It is recommended to build glibc with --enable-stack-protector=strong.
770 Implemented by Nick Alcock (Oracle).
772 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
773 intended to be used instead of memset() to erase sensitive data after use;
774 the compiler will not optimize out calls to explicit_bzero even if they
775 are "unnecessary" (in the sense that no _correct_ program can observe the
776 effects of the memory clear).
778 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
779 to float instead of double. This does not affect the ABI of any libraries
780 that are part of the GNU C Library, but may affect the ABI of other
781 libraries that use this type in their interfaces.
783 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
784 float_t and double_t types are now defined to long double instead of float
785 and double. These options are not the default, and this does not affect
786 the ABI of any libraries that are part of the GNU C Library, but it may
787 affect the ABI of other libraries that use this type in their interfaces,
788 if they are compiled or used with those options.
790 * The getentropy and getrandom functions, and the <sys/random.h> header file
793 * The buffer size for byte-oriented stdio streams is now limited to 8192
794 bytes by default. Previously, on Linux, the default buffer size on most
795 file systems was 4096 bytes (and thus remains unchanged), except on
796 network file systems, where the buffer size was unpredictable and could be
797 as large as several megabytes.
799 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
800 for the Linux quota interface which predates kernel version 2.4.22 has
803 * The malloc_get_state and malloc_set_state functions have been removed.
804 Already-existing binaries that dynamically link to these functions will
805 get a hidden implementation in which malloc_get_state is a stub. As far
806 as we know, these functions are used only by GNU Emacs and this change
807 will not adversely affect already-built Emacs executables. Any undumped
808 Emacs executables, which normally exist only during an Emacs build, should
809 be rebuilt by re-running “./configure; make” in the Emacs build tree.
811 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
812 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
813 “no-ip6-dotint” had already been the default, and support for the
814 “ip6-dotint” option was removed from the Internet in 2006.
816 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
817 flag from <resolv.h> have been removed. The option relied on a
818 backwards-incompatible DNS extension which was never deployed on the
821 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
822 RES_BLAST defined in the <resolv.h> header file have been deprecated.
823 They were already unimplemented.
825 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
826 _res.flags are deprecated. The flag was standardized in RFC 2133, but
827 removed again from the IETF name lookup interface specification in RFC
828 2553. Applications should use getaddrinfo instead.
830 * DNSSEC-related declarations and definitions have been removed from the
831 <arpa/nameser.h> header file, and libresolv will no longer attempt to
832 decode the data part of DNSSEC record types. Previous versions of glibc
833 only implemented minimal support for the previous version of DNSSEC, which
834 is incompatible with the currently deployed version.
836 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
837 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
838 <arpa/nameser.h> header file because the distinction between RR types and
839 meta-RR types is not officially standardized, subject to revision, and
840 thus not suitable for encoding in a macro.
842 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
843 and rhook members of the res_state type in <resolv.h> have been removed.
844 The glibc stub resolver did not support these hooks, but the header file
845 did not reflect that.
847 * For multi-arch support it is recommended to use a GCC which has
848 been built with support for GNU indirect functions. This ensures
849 that correct debugging information is generated for functions
850 selected by IFUNC resolvers. This support can either be enabled by
851 configuring GCC with '--enable-gnu-indirect-function', or by
852 enabling it by default by setting 'default_gnu_indirect_function'
853 variable for a particular architecture in the GCC source file
856 * GDB pretty printers have been added for mutex and condition variable
857 structures in POSIX Threads. When installed and loaded in gdb these pretty
858 printers show various pthread variables in human-readable form when read
859 using the 'print' or 'display' commands in gdb.
861 * Tunables feature added to allow tweaking of the runtime for an application
862 program. This feature can be enabled with the '--enable-tunables' configure
863 flag. The GNU C Library manual has details on usage and README.tunables has
864 instructions on adding new tunables to the library.
866 * A new version of condition variables functions have been implemented in
867 the NPTL implementation of POSIX Threads to provide stronger ordering
870 * A new version of pthread_rwlock functions have been implemented to use a more
871 scalable algorithm primarily through not using a critical section anymore to
874 Security related changes:
876 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
877 have been created with makecontext could fail to terminate due to a
878 missing .cantunwind annotation. This has been observed to lead to a hang
879 (denial of service) in some Go applications compiled with gccgo. Reported
880 by Andreas Schwab. (CVE-2016-6323)
882 * The DNS stub resolver functions would crash due to a NULL pointer
883 dereference when processing a query with a valid DNS question type which
884 was used internally in the implementation. The stub resolver now uses a
885 question type which is outside the range of valid question type values.
888 The following bugs are resolved with this release:
890 [4099] stdio: Overly agressive caching by stream i/o functions.
891 [7065] build: Support building glibc with -fstack-protector or -fstack-
893 [9842] localedata: en_CA: incorrect date format
894 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
895 before it started waiting
896 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
897 [15765] nptl: sem_open is wrongly a cancellation point
898 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
899 [16458] libc: endian.h and netinet/in.h byte order macros return values of
901 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
902 library linked with pthread
903 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
905 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
906 [17252] libc: getrandom and getentropy syscall
907 [17730] malloc: thread-local storage is sometimes improperly free()'d
908 after being __libc_memalign()'d
909 [18241] stdio: failed fseek on memstream does not set errno and can fail
911 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
913 [18463] nptl: pthread_cond_broadcast issue when surrounded by
914 PTHREAD_PRIO_INHERIT mutex on ARM
915 [18784] network: res_query and related function crash for special record
916 type queries (CVE-2015-5180)
917 [19380] math: strtod does not raise "inexact"
918 [19387] string: Integer overflow in memchr
919 [19390] string: Integer overflow in strncat
920 [19398] build: linknamespace tests fail with massively parallel build
921 [19402] nptl: Deadlock with robust shared mutex and asynchronous
923 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
924 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
925 [19514] libc: [PATCH] Fix spelling errors in spelling
926 "implement"/"implementation" in several places
927 [19582] network: Deprecate RES_USE_INET6
928 [19673] manual: clog10 docs appear to be erroneous
929 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
930 causes a segmentation fault
931 [19826] libc: invalid pointer returned from __tls_get_addr with static
933 [20016] network: resolv: Remove hooks support from the API
934 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
936 [20033] math: [x86_64] vectorized math function don't call the __finite
938 [20116] nptl: use after free in pthread_create
939 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
940 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
942 [20311] nptl: please install proc_service.h
943 [20366] libc: Compilation errors in installed headers in strict-compliance
945 [20370] malloc: malloc: Arena free list management is still racy
946 (incorrect fix in bug 19243)
947 [20386] libc: assert (X = 0) does not result in GCC warning
948 [20432] malloc: malloc: Minimize interface required for interposition
949 [20435] libc: Missing unwind info in __startcontext causes infinite loop
950 in _Unwind_Backtrace (CVE-2016-6323)
951 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
952 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
954 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
955 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
956 [20477] network: resolv: incorrect double-checked locking related to
958 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
960 [20495] math: x86_64 performance degradation due to AVX/SSE transition
962 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
963 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
964 cause transition penalty
965 [20517] math: sparcv9 missing fdiml compat symbol
966 [20524] manual: strverscmp is inconsistent
967 [20525] libc: <sys/quota.h> should be based on kernel headers
968 [20539] math: GCC 7 -static -lm fails to link at -O3
969 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
971 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
973 [20591] network: Remove obsolete DNSSEC support
974 [20592] network: DNS resource record type classification macros in
975 <arpa/nameser.h> are incorrect
976 [20593] network: Update DNS RR type definitions
977 [20611] network: getaddrinfo accepts invalid numeric scope IDs
978 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
980 [20629] network: libresolv: Remove support for bitstring labels
982 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
983 assertion: "maxidx >= 2"
984 [20660] math: [arm] Use VSQRT
985 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
986 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
987 [20689] libc: Test for FMA should also check for AVX.
988 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
989 [20715] math: iszero macro breaks existing code
990 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
991 [20728] libc: powerpc: Missing TOC stub in clone
992 [20729] build: glibc-2.24 fails to build for i486 with -Os
993 [20750] build: Build fails with default PIE enabled gcc-6.x
994 [20768] math: [alpha] sqrt fegetenv namespace
995 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
997 [20787] math: float_t is defined as float incorrectly on x86_64 even if
998 __FLT_EVAL_METHOD__ is 2
999 [20790] network: rpcgen buffer overrun in get_prog_declaration
1000 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1001 [20829] libc: crypt snprintf namespace
1002 [20847] libc: tst-vfork3 failure
1003 [20855] math: Default bits/mathdef.h has inappropriate float_t
1004 [20859] math: [sh4] FP_ILOGB0 invalid
1005 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1006 U+20AC), not same as GBK
1007 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1008 [20916] math: pow handling of sNaN arguments
1009 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1010 [20919] math: Bad pow (qNaN, 0) result with -lieee
1011 [20940] math: hypot sNaN handling
1012 [20947] math: fmax, fmin sNaN handling
1013 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1014 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1016 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1017 [20973] nptl: robust mutexes: Lost wake-ups
1018 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1019 [20978] nis: strlen on null pointer in nss_nisplus
1020 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1022 [21014] string: i686 memchr overflows internal pointer check
1023 [21019] libc: [mips] n32 lseek incorrect on overflow
1024 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1025 [21026] libc: [MIPS] readahead syscall is broken on n64
1026 [21028] math: Fallback fesetexceptflag should always succeed
1027 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1028 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1030 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1031 [21061] librt: [microblaze] librt lost clock_* exports
1032 [21073] libc: tunables: insecure environment variables passed to
1033 subprocesses with AT_SECURE
1034 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1038 * The minimum Linux kernel version that this version of the GNU C Library
1039 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1040 kernel version 2.6.32 or later suffices (on architectures that already
1041 required kernel versions more recent than 3.2, those requirements remain
1042 unchanged). Linux 3.2 or later kernel headers are required on all
1045 * The pap_AN locale has been deleted. This has been deprecated for a long
1046 time. It has been replaced by pap_AW & pap_CW, both of which have long
1047 been included in previous releases.
1049 * The readdir_r and readdir64_r functions have been deprecated. It is
1050 recommended to use readdir and readdir64 instead.
1052 * The type “union wait” has been removed. It was deprecated in the early
1053 1990s and never part of POSIX. Application code should use the int type
1054 instead of “union wait”.
1056 * A new NSS action is added to facilitate large distributed system
1057 administration. The action, MERGE, allows remote user stores like LDAP
1058 to be merged into local user stores like /etc/groups in order to provide
1059 easy to use, updated, and managed sets of merged credentials. The new
1060 action can be used by configuring it in /etc/nsswitch.conf:
1061 group: files [SUCCESS=merge] nis
1062 Implemented by Stephen Gallagher (Red Hat).
1064 * The deprecated __malloc_initialize_hook variable has been removed from the
1067 * The long unused localedef --old-style option has been removed. It hasn't
1068 done anything in over 16 years. Scripts using this option can safely
1071 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1072 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1073 return the next representable value in the direction of positive infinity
1074 and the nextdown functions return the next representable value in the
1075 direction of negative infinity. These are currently enabled as GNU
1078 Security related changes:
1080 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1081 could result in a stack overflow when getnetbyname was called with an
1082 overly long name. (CVE-2016-3075)
1084 * Previously, getaddrinfo copied large amounts of address data to the stack,
1085 even after the fix for CVE-2013-4458 has been applied, potentially
1086 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1087 instead. Reported by Michael Petlan. (CVE-2016-3706)
1089 * The glob function suffered from a stack-based buffer overflow when it was
1090 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1091 Reported by Alexander Cherepanov. (CVE-2016-1234)
1093 * The Sun RPC UDP client could exhaust all available stack space when
1094 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1095 alloca plugin for GCC. (CVE-2016-4429)
1097 * The IPv6 name server management code in libresolv could result in a memory
1098 leak for each thread which is created, performs a failing naming lookup,
1099 and exits. Over time, this could result in a denial of service due to
1100 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1102 The following bugs are resolved with this release:
1104 [1170] localedata: ne_NP: update Nepali locale definition file
1105 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1107 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1108 [6796] math: fdim() does not set errno on overflow
1109 [10354] libc: posix_spawn should use vfork() in more cases than presently
1110 [11213] localedata: localedata: add copyright disclaimer to locale files
1111 [12143] localedata: chr_US: new Cherokee locale
1112 [12450] localedata: sgs_LT: new locale
1113 [12676] localedata: ln_CD: new locale
1114 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1116 [13304] math: fma, fmaf, fmal produce wrong results
1117 [14259] build: --localedir arg to configure is ignored
1118 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1119 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1120 [14934] localedata: es_CL: wrong first weekday chilean locale
1121 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1123 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1125 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1126 [15368] nptl: raise() is not async-signal-safe
1127 [15479] math: ceil, floor, round and trunc raise inexact exception
1128 [15578] localedata: kk_KZ: various updates
1129 [16003] localedata: pap_AN: punt old locale
1130 [16137] localedata: iw_IL: punt old locale
1131 [16190] localedata: eo: new esperanto locale
1132 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1133 [16742] malloc: race condition: pthread_atfork() called before first
1134 malloc() results in unexpected locking behaviour/deadlocks
1135 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1137 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1138 [17565] localedata: pt_PT: wrong (work-)week start
1139 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1140 [17950] build: Build fails with -msse
1141 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1142 [18433] libc: posix_spawn does not return correctly upon failure to
1144 [18453] localedata: charmaps/IBM875: incorrect codes
1145 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1147 [18896] localedata: he_IL: improvements for currency
1148 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1149 Romanian locale data
1150 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1152 [19056] libc: Deprecate readdir_r
1153 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1155 [19198] localedata: nl_NL: small improvements for Dutch locales
1156 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1157 nameservers (CVE-2016-5417)
1158 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1160 [19400] locale: Language missing in "iso-639.def", trivial fix in
1162 [19431] malloc: Deadlock between fflush, getdelim, and fork
1163 [19505] libc: Incorrect file descriptor validity checks in
1164 posix_spawn_file_actions_add{open,close,dup2}
1165 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1166 when using RTLD_NEXT
1167 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1168 `intl/{gettextP,loadinfo}.h'
1169 [19534] libc: execle, execlp may use malloc
1170 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1171 [19573] network: res_nclose and __res_maybe_init disagree about name
1172 server initialization, breaking Hesiod
1173 [19575] localedata: Status of GB18030 tables
1174 [19581] localedata: sr_* date_fmt string contains additional newline
1175 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1177 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1178 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1179 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1180 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1182 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1184 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1186 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1187 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1188 [19633] locale: strfmon_l applies global locale to number formatting
1189 [19642] network: Memory leak in getnameinfo
1190 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1191 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1193 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1194 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1196 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1197 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1199 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1201 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1203 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1204 pointers and lengths in error-case.
1205 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1206 report errors on UTF-16 surrogates values.
1207 [19755] nscd: nscd assertion failure in gc
1208 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1209 [19759] libc: mempcpy shouldn't be inlined
1210 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1211 [19765] libc: s390 needs an optimized mempcpy
1212 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1213 NAME_MAX limit assumption (CVE-2016-1234)
1214 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1215 [19787] network: Missing and incorrect truncation checks in getnameinfo
1216 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1218 [19791] network: Assertion failure in res_query.c with un-connectable name
1220 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1221 [19822] math: libm.so install clobbers old version
1222 [19825] network: resolv: send_vc can return uninitialized data in second
1223 response to getaddrinfo
1224 [19830] network: nss_dns: should check RDATA length against buffer length
1225 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1226 confronted with A/AAAA records of invalid size
1227 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1228 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1229 [19853] stdio: Printing IBM long double in decimal with high precision is
1231 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1232 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1233 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1235 [19865] network: Assertion failure or memory leak in
1236 _nss_dns_getcanonname_r
1237 [19868] network: nss_dns: netent code does not skip over non-PTR records
1238 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1240 [19881] string: Improve x86-64 memset
1241 [19907] string: Incorrect memcpy tests
1242 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1244 [19925] libc: termios.h XCASE namespace
1245 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1246 [19929] libc: limits.h NL_NMAX namespace
1247 [19931] stdio: Memory leak in vfprintf
1248 [19957] libc: clone(CLONE_VM) access invalid parent memory
1249 [19963] localedata: en_IL: New locale
1250 [19989] stdio: stdio.h cuserid namespace
1251 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1253 [19996] locale: langinfo.h nl_langinfo_l namespace
1254 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1255 [20010] network: getaddrinfo: Stack overflow in hostent translation
1257 [20012] stdio: libio: fmemopen append mode failure
1258 [20014] stdio: stdio.h namespace for pre-threads POSIX
1259 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1260 [20023] libc: fcntl.h timespec namespace
1261 [20024] math: [x86_64] vectorized sincos trashes the stack
1262 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1263 [20041] time: sys/time.h timespec namespace
1264 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1265 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1266 [20051] libc: ttyslot in wrong header under wrong conditions
1267 [20054] libc: gethostname not declared for XPG4
1268 [20055] libc: termios.h missing tcgetsid for XPG4
1269 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1271 [20073] libc: sys/stat.h fchmod namespace
1272 [20074] libc: stdlib.h rand_r namespace
1273 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1274 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1276 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1277 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1279 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1280 [20119] libc: Wrong mask for processors level type from CPUID
1281 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1282 AS not supporting AVX512
1283 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1285 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1286 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1288 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1289 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1290 [20178] libc: posix_spawn{p} should not call exit
1291 [20191] stdio: libio: vtables hardening
1292 [20195] string: FMA4 detection requires CPUID execution with register
1294 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1295 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1297 [20212] math: acos (sNaN) returns sNaN
1298 [20213] math: asin (sNaN) returns sNaN
1299 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1300 [20218] math: [i386] asinhl (sNaN) returns sNaN
1301 [20219] math: [i386] atanhl (sNaN) returns sNaN
1302 [20222] stdio: fopencookie: Mangle function pointers
1303 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1304 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1305 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1307 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1308 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1309 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1310 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1311 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1312 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1313 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1314 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1315 [20240] math: modf (sNaN) returns sNaN
1316 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1317 [20250] math: frexp (sNaN) returns sNaN
1318 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1319 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1321 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1322 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1323 subscript is below array bounds [-Werror=array-bounds]
1324 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1325 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1327 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1328 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1330 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1331 "invalid" exceptions
1332 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1333 [20316] localedata: id_ID: Februari instead of Pebruari
1334 [20327] string: POWER8 strcasecmp returns incorrect result
1335 [20347] math: Failure: Test: j0_downward (0xap+0)
1336 [20348] libc: FAIL: misc/tst-preadvwritev64
1337 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1338 [20350] libc: There is no test for p{read,write}64
1339 [20357] math: Incorrect cos result for 1.5174239687223976
1340 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1344 * Unicode 8.0.0 Support: Character encoding, character type info, and
1345 transliteration tables are all updated to Unicode 8.0.0, using new
1346 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1347 These updates cause user visible changes, such as the fixes for bugs
1348 89, 16061, and 18568.
1350 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1351 kernel-internal CPU set size. This means that requests that change the
1352 CPU affinity which failed before (for example, an all-ones CPU mask) will
1353 now succeed. Applications that need to determine the effective CPU
1354 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1355 setting it because the kernel can adjust it (and the previous size check
1356 would not detect this in the majority of cases).
1358 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1359 the following new symbols are used: fts64_children, fts64_close,
1360 fts64_open, fts64_read and fts64_set.
1362 * getaddrinfo now detects certain invalid responses on an internal netlink
1363 socket. If such responses are received, an affected process will
1364 terminate with an error message of "Unexpected error <number> on netlink
1365 descriptor <number>" or "Unexpected netlink response of size <number> on
1366 descriptor <number>". The most likely cause for these errors is a
1367 multi-threaded application which erroneously closes and reuses the netlink
1368 file descriptor while it is used by getaddrinfo.
1370 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1371 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1372 unnecessary serialization of memory allocation requests across threads.
1373 The defect is now corrected. Users should see a substantial increase in
1374 the concurent throughput of allocation requests for applications which
1375 trigger this bug. Affected applications typically create and
1376 destroy threads frequently. (Bug 19048 was reported and analyzed by
1379 * There is now a --disable-timezone-tools configure option for disabling the
1380 building and installing of the timezone related utilities (zic, zdump, and
1381 tzselect). This is useful for people who build the timezone data and code
1382 independent of the GNU C Library.
1384 * The obsolete header <regexp.h> has been removed. Programs that require
1385 this header must be updated to use <regex.h> instead.
1387 * The obsolete functions bdflush, create_module, get_kernel_syms,
1388 query_module and uselib are no longer available to newly linked binaries;
1389 the header <sys/kdaemon.h> has been removed. These functions and header
1390 were specific to systems using the Linux kernel and could not usefully be
1391 used with the GNU C Library on systems with version 2.6 or later of the
1394 * Optimized string, wcsmbs and memory functions for IBM z13.
1395 Implemented by Stefan Liebler.
1397 * Newly linked programs that define a variable called signgam will no longer
1398 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1399 require signgam to be set by those functions must ensure that they use the
1400 variable provided by the GNU C Library and declared in <math.h>, without
1401 defining their own copy.
1403 * The minimum GCC version that can be used to build this version of the GNU
1404 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1405 still be used to compile programs using the GNU C Library.
1407 Security related changes:
1409 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1410 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1412 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1413 the pointer guard feature. It is always enabled. Previously,
1414 LD_POINTER_GUARD could be used to disable security hardening in binaries
1415 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1418 * An integer overflow in hcreate and hcreate_r could lead to an
1419 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1421 * The catopen function no longer has unbounded stack usage. Reported by
1422 Max. (CVE-2015-8779)
1424 * The nan, nanf and nanl functions no longer have unbounded stack usage
1425 depending on the length of the string passed as an argument to the
1426 functions. Reported by Joseph Myers. (CVE-2014-9761)
1428 * A stack-based buffer overflow was found in libresolv when invoked from
1429 libnss_dns, allowing specially crafted DNS responses to seize control
1430 of execution flow in the DNS client. The buffer overflow occurs in
1431 the functions send_dg (send datagram) and send_vc (send TCP) for the
1432 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1433 family. The use of AF_UNSPEC triggers the low-level resolver code to
1434 send out two parallel queries for A and AAAA. A mismanagement of the
1435 buffers used for those queries could result in the response of a query
1436 writing beyond the alloca allocated buffer created by
1437 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1438 the overflow. Thanks to the Google Security Team and Red Hat for
1439 reporting the security impact of this issue, and Robert Holiday of
1440 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1442 The following bugs are resolved with this release:
1444 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1445 [887] math: Math library function "logb" and "nextafter" inconsistent
1446 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1447 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1448 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1449 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1451 [4404] localedata: German translation of "Alarm clock" is misleading
1452 [6799] math: nextafter() and nexttoward() doen't set errno on
1453 overflow/underflow errors
1454 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1456 [10432] nis: _nss_nis_setnetgrent assertion failure
1457 [11460] libc: fts has no LFS support
1458 [12926] network: getaddrinfo()/make_request() may spin forever
1459 [13065] nptl: Race condition in pthread barriers
1460 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1461 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1463 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1464 [14912] libc: Rename non-installed bits/*.h headers
1465 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1466 [15367] math: Let gcc use __builtin_isinf
1467 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1468 [15421] math: lgamma wrongly sets signgam for ISO C
1469 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1470 FE_INVALID with argument out of range
1471 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1473 [15786] dynamic-link: ifunc resolver functions can smash function
1475 [15918] math: Unnecessary check for equality in hypotf()
1476 [16061] localedata: Review / update transliteration data
1477 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1479 [16141] time: strptime %z offset restriction
1480 [16171] math: drem should be alias of remainder
1481 [16296] math: fegetround is pure?
1482 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1483 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1485 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1486 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1488 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1489 "Exception "Inexact" set" on ppc32
1490 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1491 [16517] math: Missing underflow exception from tanf/tan/tanl
1492 [16519] math: Missing underflow exception from sinhf
1493 [16520] math: Missing underflow exception from tanhf
1494 [16521] math: Missing underflow exception from exp2
1495 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1497 [16734] stdio: fopen calls mmap to allocate its buffer
1498 [16961] math: nan function incorrect handling of bad sequences
1499 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1500 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1501 [16985] locale: localedef: confusing error message when opening output
1503 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1504 [17197] locale: Redundant shift character in iconv conversion output at
1506 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1507 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1508 [17250] dynamic-link: static linking breaks nss loading
1509 (getaddrinfo/getpwnam/etc...)
1510 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1512 [17441] math: isnan() should use __builtin_isnan() in GCC
1513 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1514 (related to lock elision)
1515 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1516 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1517 [17887] time: strptime should be able to parse "+01:00" style timezones
1518 [17905] libc: catopen() Multiple unbounded stack allocations
1520 [18084] libc: backtrace (..., 0) dumps core on x86
1521 [18086] libc: nice() sets errno to 0 on success
1522 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1523 is too large (CVE-2015-8778)
1524 [18251] dynamic-link: SONAME missing when audit modules provides path
1525 [18265] libc: add attributes for wchar string and memory functions
1526 [18370] math: csqrt missing underflows
1527 [18421] libc: [hppa] read-only segment has dynamic relocations
1528 [18472] libc: Obsolete syscall wrappers should be compat symbols
1529 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1530 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1531 [18525] localedata: Remove locale timezone information
1532 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1533 [18568] localedata: Update locale data to Unicode 8.0
1534 [18589] locale: sort-test.sh fails at random
1535 [18595] math: ctan, ctanh missing underflows
1536 [18604] libc: assert macro-expands its argument
1537 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1538 contains a vector instruction exception.
1539 [18611] math: j1, jn missing errno setting on underflow
1540 [18618] localedata: sync Chechen locale definitions with other *_RU
1542 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1543 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1544 [18665] network: In send_dg, the recvfrom function is NOT always using the
1545 buffer size of a newly created buffer (CVE-2015-7547)
1546 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1548 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1550 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1551 [18699] math: tilegx cproj() for various complex infinities does not yield
1553 [18724] libc: Harden put*ent functions against data injection
1554 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1555 [18755] build: build errors with -DNDEBUG
1556 [18757] stdio: fmemopen fails to set errno on failure
1557 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1559 [18781] libc: openat64 lacks O_LARGEFILE
1560 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1561 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1562 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1563 [18790] math: [ldbl-128ibm] tanhl inaccurate
1564 [18795] libc: stpncpy fortification misses buffer lengths that are
1565 statically too large
1566 [18796] build: build fails for --disable-mathvec
1567 [18803] math: hypot missing underflows
1568 [18820] stdio: fmemopen may leak memory on failure
1569 [18823] math: csqrt spurious underflows
1570 [18824] math: fma spurious underflows
1571 [18825] math: pow missing underflows
1572 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1573 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1574 [18870] build: sem_open.c fails to compile with missing symbol
1576 [18872] stdio: Fix memory leak in printf_positional
1577 [18873] libc: posix_fallocate overflow check ineffective
1578 [18875] math: Excess precision leads incorrect libm
1579 [18877] libc: arm: mmap offset regression
1580 [18887] libc: memory corruption when using getmntent on blank lines
1581 [18918] localedata: hu_HU: change time to HH:MM:SS format
1582 [18921] libc: Regression: extraneous stat() and fstat() performed by
1584 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1585 binaries (CVE-2015-8777)
1586 [18951] math: tgamma missing underflows
1587 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1589 [18953] localedata: lt_LT: change currency symbol to the euro
1590 [18956] math: powf inaccuracy
1591 [18961] math: [i386] exp missing underflows
1592 [18966] math: [i386] exp10 missing underflows
1593 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1594 [18969] build: multiple string test failures due to missing locale
1596 [18970] libc: Reference of pthread_setcancelstate in libc.a
1597 [18977] math: float / long double Bessel functions not in XSI POSIX
1598 [18980] math: i386 libm functions return with excess range and precision
1599 [18981] math: i386 scalb*, ldexp return with excess range and precision
1600 [18982] stdio: va_list and vprintf
1601 [18985] time: Passing out of range data to strftime() causes a segfault
1603 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1604 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1605 [19012] locale: iconv_open leaks memory on error path
1606 [19016] math: clog, clog10 inaccuracy
1607 [19018] nptl: Mangle function pointers in tls_dtor_list
1608 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1609 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1610 [19048] malloc: malloc: arena free list can become cyclic, increasing
1612 [19049] math: [powerpc] erfc incorrect zero sign
1613 [19050] math: [powerpc] log* incorrect zero sign
1614 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1615 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1616 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1617 [19074] network: Data race in _res_hconf_reorder_addrs
1618 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1619 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1620 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1622 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1624 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1625 [19086] manual: posix_fallocate64 documented argument order is wrong.
1626 [19088] math: lround, llround missing exceptions close to overflow
1628 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1629 [19095] math: dbl-64 lrint incorrect for 64-bit long
1630 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1631 [19124] dynamic-link: ld.so failed to build with older assmebler
1632 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1633 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1634 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1635 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1636 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1637 pthread_setaffinity_np
1638 [19156] math: [ldbl-128] j0l spurious underflows
1639 [19164] nptl: tst-getcpu fails with many possible CPUs
1640 [19168] math: math/test-ildoubl and math/test-ldouble failure
1641 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
1642 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
1644 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
1645 (FE_NOMASK_ENV) do not clear SSE exceptions
1646 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
1647 _int_new_arena/reused_arena
1648 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
1649 [19201] math: dbl-64 remainder incorrect sign of zero result
1650 [19205] math: bits/math-finite.h conditions do not match math.h and
1652 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
1653 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
1654 for C99-based standards
1655 [19212] libc: features.h not -Wundef clean
1656 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
1658 [19214] libc: Family and model identification for AMD CPU's are incorrect.
1659 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
1660 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
1662 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
1663 "inexact" exceptions
1664 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
1666 [19242] libc: strtol incorrect in Turkish locales
1667 [19243] malloc: reused_arena can pick an arena on the free list, leading
1668 to an assertion failure and reference count corruption
1669 [19253] time: tzset() ineffective when temporary TZ did not include DST
1671 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
1672 [19270] math: [hppa] Shared libm missing __isnanl
1673 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
1674 MAP_HUGETLB and MAP_STACK defines
1675 [19313] nptl: Wrong __cpu_mask for x32
1676 [19347] libc: grantpt: try to force a specific gid even without pt_chown
1677 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
1678 [19350] math: [ldbl-128ibm] sinhl spurious overflows
1679 [19351] math: [ldbl-128ibm] logl inaccurate near 1
1680 [19363] time: x32: times() return value wrongly truncates/sign extends
1682 [19367] dynamic-link: Improve branch prediction on Silvermont
1683 [19369] network: Default domain name not reset by res_ninit when "search"
1684 / "domain" entry is removed from resolv.conf
1685 [19375] math: powerpc: incorrect results for POWER7 logb with negative
1687 [19385] localedata: bg_BG: time separator should be colon, not comma
1688 [19408] libc: linux personality syscall wrapper may erroneously return an
1689 error on 32-bit architectures
1690 [19415] libc: dladdr returns wrong names on hppa
1691 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
1692 IBM905, IBM907, and IBM909
1693 [19439] math: Unix98 isinf and isnan functions conflict with C++11
1694 [19443] build: build failures with -DDEBUG
1695 [19451] build: Make check fails on test-double-vlen2
1696 [19462] libc: Glibc failed to build with -Os
1697 [19465] math: Wrong code with -Os
1698 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
1700 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
1702 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
1703 [19486] math: S390: Math tests fail with "Exception Inexact set".
1704 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
1705 [19550] libc: [mips] mmap negative offset handling inconsistent with other
1707 [19590] math: Fail to build shared objects that use libmvec.so functions.
1711 * The following bugs are resolved with this release:
1713 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
1714 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
1715 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
1716 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
1717 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
1718 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
1719 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
1720 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
1721 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
1722 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
1723 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
1724 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
1725 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
1726 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
1727 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
1728 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
1729 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
1730 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
1731 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
1732 18657, 18676, 18694, 18696, 18887.
1734 * Cache information can be queried via sysconf() function on s390 e.g. with
1735 _SC_LEVEL1_ICACHE_SIZE as argument.
1737 * A buffer overflow in gethostbyname_r and related functions performing DNS
1738 requests has been fixed. If the NSS functions were called with a
1739 misaligned buffer, the buffer length change due to pointer alignment was
1740 not taken into account. This could result in application crashes or,
1741 potentially arbitrary code execution, using crafted, but syntactically
1742 valid DNS responses. (CVE-2015-1781)
1744 * The time zone file parser has been made more robust against crafted time
1745 zone files, avoiding heap buffer overflows related to the processing of
1746 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
1747 large time zone data files. Overly long time zone specifiers in the TZ
1748 variable no longer result in stack overflows and crashes.
1750 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
1751 for LD and GD on x86 and x86-64, has been implemented. You will need
1752 binutils-2.24 or later to enable this optimization.
1754 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
1755 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
1756 Hat). These updates cause user visible changes, such as the fix for bug
1759 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
1760 and getXXbyYY NSS calls for the same database, causing a denial-of-service
1761 condition in some applications.
1763 * Added vector math library named libmvec with the following vectorized x86_64
1764 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
1766 The library can be disabled with --disable-mathvec. Use of the functions is
1767 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
1768 Shared library libmvec.so is linked in as needed when using -lm (no need to
1769 specify -lmvec explicitly for not static builds).
1770 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
1772 * A new fmemopen implementation has been added with the goal of POSIX
1773 compliance. The new implementation fixes the following long-standing
1774 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
1775 old implementation is still present for use be by existing binaries.
1777 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
1778 releases. It has been fixed to match 2.19 and older, but binaries built
1779 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
1781 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
1782 Contributed by Roland McGrath (Google).
1784 * The header <regexp.h> is deprecated, and will be removed in a future
1785 release. Use of this header will trigger a deprecation warning.
1786 Application developers should update their code to use <regex.h> instead.
1788 This header was formerly part of SUS, but was deprecated in 1994 and
1789 removed from the standard in 2001. Also, the glibc implementation
1790 leaks memory. See BZ#18681 for more details.
1794 * The following bugs are resolved with this release:
1796 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
1797 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
1798 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
1799 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
1800 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
1801 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
1802 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
1803 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
1804 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
1805 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
1808 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
1809 too little memory for the to-be-scanned arguments and overflow the
1810 allocated buffer. The implementation now correctly computes the required
1811 buffer size when using malloc, and switches to malloc from alloca as
1814 * A new semaphore algorithm has been implemented in generic C code for all
1815 machines. Previous custom assembly implementations of semaphore were
1816 difficult to reason about or ensure that they were safe. The new version
1817 of semaphore supports machines with 64-bit or 32-bit atomic operations.
1818 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
1819 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
1821 * Port to Altera Nios II has been contributed by Mentor Graphics.
1823 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
1824 implementations for powerpc64/powerpc64le.
1825 Implemented by Adhemerval Zanella (IBM).
1827 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
1828 and powerpc64le. This may improve lock scaling of existing programs on
1829 HTM capable systems. The lock elision code is only enabled with
1830 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
1831 powerpc will issue a transaction abort on every syscall to avoid side
1832 effects being visible outside transactions.
1834 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
1835 AArch64. Contributed by ARM Ltd.
1837 * i386 memcpy functions optimized with SSE2 unaligned load/store.
1839 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
1840 under certain input conditions resulting in the execution of a shell for
1841 command substitution when the application did not request it. The
1842 implementation now checks WRDE_NOCMD immediately before executing the
1843 shell and returns the error WRDE_CMDSUB as expected.
1845 * CVE-2012-3406 printf-style functions could run into a stack overflow when
1846 processing format strings with a large number of format specifiers.
1848 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
1849 infinite loop if the DNS response contained a PTR record of an unexpected
1852 * The minimum GCC version that can be used to build this version of the GNU
1853 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
1854 still be used to compile programs using the GNU C Library.
1856 * The GNU C Library is now built with -Werror by default. This can be
1857 disabled by configuring with --disable-werror.
1859 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
1861 * The obsolete sigvec function has been removed. This was the original
1862 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
1863 programs have been using instead for about 25 years. Of course, ABI
1864 compatibility for old binaries using sigvec remains intact.
1866 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
1867 with newer versions of bison.
1869 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
1870 The original MIPS o32 hard-float ABI requires an FPU where double-precision
1871 registers overlay two consecutive single-precision registers. MIPS32R2
1872 introduced a new FPU mode (FR=1) where double-precision registers extend the
1873 corresponding single-precision registers which is incompatible with the
1874 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
1875 require the use of FR=1 making a transition necessary. New o32 ABI
1876 extensions enable users to migrate over time from the original o32 ABI
1877 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
1878 tracks the ABI of any loaded object and verifies that new objects are
1879 compatible. Mode transitions will also be requested as required and
1880 unsupportable objects will be rejected. The ABI checks include both soft and
1881 hard float ABIs for o32, n32 and n64.
1883 GCC 5 with GNU binutils 2.25 onwards:
1884 It is strongly recommended that all o32 system libraries are built using the
1885 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
1886 with the original and all new o32 ABI extensions. Configure a MIPS GCC
1887 compiler using --with-fp-32=xx to set this by default.
1891 * The following bugs are resolved with this release:
1893 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
1894 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
1895 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
1896 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
1897 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
1898 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
1899 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
1900 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
1901 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
1902 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
1903 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
1904 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
1905 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
1906 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
1907 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
1909 * Reverted change of ABI data structures for s390 and s390x:
1910 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
1911 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
1912 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
1913 are preserved pointing straight to the same implementation as the old ones.
1914 Given that, new callers will simply provide a too-big buffer to these
1915 functions. Any applications/libraries out there that embed jmp_buf or
1916 ucontext_t in an ABI-relevant data structure that have already been rebuilt
1917 against 2.19 headers will have to rebuilt again. This is necessary in any
1918 case to revert the breakage in their ABI caused by the glibc change.
1920 * Support for file description locks is added to systems running the
1921 Linux kernel. The standard file locking interfaces are extended to
1922 operate on file descriptions, not file descriptors, via the use of
1923 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
1924 are associated with an open file instead of a process.
1926 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
1928 * The minimum Linux kernel version that this version of the GNU C Library
1929 can be used with is 2.6.32.
1931 * Running the testsuite no longer terminates as soon as a test fails.
1932 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
1933 with PASS or FAIL lines for individual tests. A summary of the results is
1934 printed, including a list of failing lists, and "make check" exits with
1935 error status if there were any unexpected failures. "make check
1936 stop-on-test-failure=y" may be used to keep the old behavior.
1938 * The am33 port, which had not worked for several years, has been removed
1941 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
1942 supported; they now act the same as _DEFAULT_SOURCE (but generate a
1943 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
1944 conflicted with POSIX (support for which was removed in 2.19), the
1945 interfaces those macros enabled remain available when compiling with
1946 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
1947 test macros defined.
1949 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
1951 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
1952 This may improve lock scaling of existing programs on TX capable systems.
1953 The lock elision code is only built with --enable-lock-elision=yes and
1954 then requires a GCC version supporting the TX builtins. With lock elision
1955 default mutexes are elided via __builtin_tbegin, if the cpu supports
1956 transactions. By default lock elision is not enabled and the elision code
1959 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
1960 copy the path argument. This allowed programs to cause posix_spawn to
1961 deference a dangling pointer, or use an unexpected pathname argument if
1962 the string was modified after the posix_spawn_file_actions_addopen
1965 * All supported architectures now use the main glibc sysdeps directory
1966 instead of some being in a separate "ports" directory (which was
1967 distributed separately before glibc 2.17).
1969 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
1970 On configurations that support it (all Linux configurations), it's now
1971 used regardless of the --enable-add-ons switch to configure. It is no
1972 longer possible to build such configurations without pthreads support.
1974 * Locale names, including those obtained from environment variables (LANG
1975 and the LC_* variables), are more tightly checked for proper syntax.
1976 setlocale will now fail (with EINVAL) for locale names that are overly
1977 long, contain slashes without starting with a slash, or contain ".." path
1978 components. (CVE-2014-0475) Previously, some valid locale names were
1979 silently replaced with the "C" locale when running in AT_SECURE mode
1980 (e.g., in a SUID program). This is no longer necessary because of the
1983 * On x86-64, the dynamic linker's lazy-binding support is now compatible
1984 with application code using Intel MPX instructions. (With all previous
1985 versions, the MPX register state could be clobbered when making calls
1986 into or out of a shared library.) Note that while the new dynamic
1987 linker is compatible with all known x86 hardware whether or not it
1988 supports Intel MPX, some x86 instruction-set emulators might fail to
1989 handle the new instruction encodings. This is known to affect Valgrind
1990 versions up through 3.9 (but will be fixed in the forthcoming 3.10
1991 release), and might affect other tools that do instruction emulation.
1993 * Support for loadable gconv transliteration modules has been removed.
1994 The support for transliteration modules has been non-functional for
1995 over a decade, and the removal is prompted by security defects. The
1996 normal gconv conversion modules are still supported. Transliteration
1997 with //TRANSLIT is still possible, and the //IGNORE specifier
1998 continues to be supported. (CVE-2014-5119)
2000 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2001 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2002 resulting a denial-of-service security vulnerability in applications which
2003 use functions related to iconv. (CVE-2014-6040)
2007 * The following bugs are resolved with this release:
2009 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2010 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2011 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2012 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2013 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2014 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2015 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2016 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2017 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2018 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2019 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2020 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2021 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2022 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2023 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2024 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2025 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2026 16431, 16453, 16474, 16506, 16510, 16529
2028 * Slovenian translations for glibc messages have been contributed by the
2029 Translation Project's Slovenian team of translators.
2031 * The public headers no longer use __unused nor __block. This change is to
2032 support compiling programs that are derived from BSD sources and use
2033 __unused internally, and to support compiling with Clang's -fblock
2034 extension which uses __block.
2036 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2037 large collation sequences to optimize multiple passes. This cache
2038 computation may overflow for large collation sequences and may cause a
2039 stack or buffer overflow. This is now fixed to use a slower algorithm
2040 which does not use a cache if there is an integer overflow.
2042 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2043 rules for large collation sequences to optimize multiple passes and falls
2044 back to alloca if malloc fails, resulting in a possible stack overflow.
2045 The implementation now falls back to an uncached collation sequence lookup
2048 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2049 initialized for static applications resulting in the security feature
2050 being disabled. The pointer guard is now correctly initialized to a
2051 random value for static applications. Existing static applications need
2052 to be recompiled to take advantage of the fix (bug 15754).
2054 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2055 to the d_name member of struct dirent, or omit the terminating NUL
2056 character. (Bugzilla #14699).
2058 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2059 aligned_alloc functions could allocate too few bytes or corrupt the
2060 heap when passed very large allocation size values (Bugzilla #15855,
2063 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2064 for AF_INET6 has been fixed (Bugzilla #16072).
2066 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2067 pap_CW, quz_PE, the_NP.
2069 * Substantially revised locales: gd_GB, ht_HT
2071 * The LC_ADDRESS field was updated to support country_car for almost all
2074 * ISO 1427 definitions were updated.
2076 * ISO 3166 definitions were updated.
2078 * The localedef utility now supports --big-endian and --little-endian
2079 command-line options to generate locales for a different system from that
2080 for which the C library was built.
2082 * Binary locale files now only depend on the endianness of the system for
2083 which they are generated and not on other properties of that system. As a
2084 consequence, binary files generated with new localedef may be incompatible
2085 with old versions of the GNU C Library, and binary files generated with
2086 old localedef may be incompatible with this version of the GNU C Library,
2087 in the following circumstances:
2089 + Locale files may be incompatible on m68k systems.
2091 + Locale archive files (but not separate files for individual locales) may
2092 be incompatible on systems where plain "char" is signed.
2094 * The configure option --disable-versioning has been removed. Builds with
2095 --disable-versioning had not worked for several years.
2097 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2098 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2100 * SystemTap probes for malloc have been introduced.
2102 * SystemTap probes for slow multiple precision fallback paths of
2103 transcendental functions have been introduced.
2105 * Support for powerpc64le has been added.
2107 * The soft-float powerpc port now supports e500 processors.
2109 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2111 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2112 set of header declarations that are enabled by default, even when other
2113 feature test macros or compiler options such as -std=c99 would otherwise
2114 disable some of those declarations.
2116 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2117 conflict with POSIX. The libbsd-compat library (which was a dummy library
2118 that did nothing) has also been removed.
2120 * Preliminary documentation about Multi-Thread, Async-Signal and
2121 Async-Cancel Safety has been added.
2123 * Change of ABI data structures for s390 and s390x:
2124 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2125 allow for future hardware extensions. All library functions that accept or
2126 return these structures were versioned in order to provide backwards
2127 compatibility with existing code. However, not all cases can be handled
2128 with interface versioning. If an application embeds either structure into
2129 another structure and that structure is passed to another compilation unit
2130 which expects the newer larger-sized object then any access to the new
2131 fields will result in undefined behaviour. Similarly any access to fields
2132 that were moved by the enlarging of the structures will cause undefined
2133 behaviour. To fix the undefined behaviour all objects that are part of the
2134 final in-memory image must agree on the size of structures, and this may
2135 require recompilation.
2139 * The following bugs are resolved with this release:
2141 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2142 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2143 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2144 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2145 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2146 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2147 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2148 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2149 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2150 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2151 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2152 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2153 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2156 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2157 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2158 Distributions can re-enable building and using pt_chown via the new configure
2159 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2160 considerable security risks and should only be used if the distribution
2161 understands and accepts the risks.
2163 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2166 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2167 fixed (Bugzilla #15330).
2169 * Add support for calling C++11 thread_local object destructors on thread
2170 and program exit. This needs compiler support for offloading C++11
2171 destructor calls to glibc.
2173 * Improved worst case performance of libm functions with double inputs and
2176 * Support for priority inherited mutexes in pthread condition variables on
2177 non-x86 architectures.
2179 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2181 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2183 * Optimized string functions for ARM. Implemented by Will Newton and
2186 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2188 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2191 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2192 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2194 * Added a benchmark framework to track performance of functions in glibc.
2196 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2197 It is based on draft TS 18661 and currently enabled as a GNU extension.
2199 * On Linux, the clock function now uses the clock_gettime system call
2200 for improved precision, rather than old times system call.
2202 * Added support for version-3 tz format files. This is needed when using
2203 the tz database release 2013e or later, and affects a few unusual cases --
2204 currently only TZ='America/Godthab' for time stamps after 2037.
2206 * Added new API functions pthread_getattr_default_np and
2207 pthread_setattr_default_np to get and set the default pthread
2208 attributes of a process.
2210 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2211 This may improve lock scaling of existing programs on TSX capable systems.
2212 When the --enable-lock-elision=yes parameter is specified at configure
2213 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2216 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2217 availability). Implemented by Ryan S. Arnold.
2219 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2223 * The following bugs are resolved with this release:
2225 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2226 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2227 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2228 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2229 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2230 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2231 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2232 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2233 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2234 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2235 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2236 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2237 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2238 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2239 14879, 14889, 14893, 14898, 14914.
2241 * Optimization of memcpy for MIPS.
2243 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2244 EMFILE has been fixed (Bugzilla #14889).
2246 * The add-on ports collection is now distributed in the "ports" subdirectory
2247 of the main GNU C Library distribution, rather than separately.
2249 * Port to ARM AArch64 contributed by Linaro.
2251 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2252 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2254 Implemented by Andreas Krebbel.
2256 * The new function secure_getenv allows secure access to the environment,
2257 returning NULL if running in a SUID/SGID process. This function replaces
2258 the internal function __secure_getenv.
2260 * SystemTap static probes have been added into the dynamic linker.
2261 Implemented by Gary Benson.
2263 * Optimizations of string functions strstr, strcasestr and memmem.
2264 Implemented by Maxim Kuvyrkov.
2266 * The minimum Linux kernel version that this version of the GNU C Library
2267 can be used with is 2.6.16.
2269 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2270 powerpc POWER7. Implemented by Will Schmidt.
2272 * New configure option --disable-nscd builds the C library such that it
2273 never attempts to contact the Name Service Caching Daemon (nscd).
2274 New configure option --disable-build-nscd avoids building nscd itself;
2275 this is the default if --disable-nscd is used.
2277 * Improved support for cross-compilation, including cross-testing and
2278 bootstrap builds without a previously built glibc.
2280 * Several testsuite tests are now able to test multiple IFUNC variants of an
2281 interface, rather than just testing the one that would be chooen by
2284 * New configure options --with-bugurl and --with-pkgversion, for
2285 distributors to use to embed their bug-reporting and package version
2286 information in --help and --version output.
2288 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2289 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2290 allows creation of chroots without the procfs mounted on /proc.
2292 * The `crypt' function now fails if passed salt bytes that violate the
2293 specification for those values. On Linux, the `crypt' function will
2294 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2295 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2296 when the mode is enabled.
2298 * The `clock_*' suite of functions (declared in <time.h>) is now available
2299 directly in the main C library. Previously it was necessary to link with
2300 -lrt to use these functions. This change has the effect that a
2301 single-threaded program that uses a function such as `clock_gettime' (and
2302 is not linked with -lrt) will no longer implicitly load the pthreads
2303 library at runtime and so will not suffer the overheads associated with
2304 multi-thread support in other code such as the C++ runtime library.
2306 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2311 * The following bugs are resolved with this release:
2313 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2314 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2315 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2316 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2317 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2318 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2319 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2320 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2321 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2322 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2323 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2324 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2325 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2326 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2327 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2328 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2329 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2330 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2331 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2332 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2333 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2334 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2337 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2338 configuring glibc with:
2339 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2340 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2341 Implemented by H.J. Lu.
2345 + define static_assert
2347 + do not declare gets
2349 + declare at_quick_exit and quick_exit also for ISO C11
2351 + aligned_alloc. NB: The code is deliberately allows the size parameter
2352 to not be a multiple of the alignment. This is a moronic requirement
2353 in the standard but it is only a requirement on the caller, not the
2356 + timespec_get added
2358 + uchar.h support added
2360 + CMPLX, CMPLXF, CMPLXL added
2362 Implemented by Ulrich Drepper.
2364 * Support for the IA-64 has been moved to ports.
2366 * Remove support for anything but ELF binary format
2368 * Checking versions of poll, ppoll added.
2369 Implemented by Ulrich Drepper.
2371 * More generic and 64-bit performance optimizations to math functions.
2372 Implemented by Ulrich Drepper.
2374 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2375 headers and functions available at compile time as they were before
2376 version 2.14. This option will be removed at some time in the future
2377 after the TI-RPC library becomes fully sufficient for the needs of
2378 existing applications.
2380 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2381 Note that glibc is not expected to work with any Linux kernel version
2384 * New header <sys/auxv.h> and function getauxval allowing easy access to
2385 the AT_* key-value pairs passed from the Linux kernel. The header also
2386 defines the HWCAP_* bits associated with the AT_HWCAP key.
2388 * New locales: mag_IN
2390 * New configure option --enable-systemtap builds SystemTap static probes
2391 into libc for setjmp and longjmp and into libpthread for various operations.
2392 So far the setjmp/longjmp probes and some of the libpthread probes are
2393 provided only for i*86 and x86_64.
2394 Implemented by Roland McGrath and Rayson Ho.
2396 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2398 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2401 * More optimized functions for SPARC. Implemented by David S. Miller.
2403 * Improved support for cross-compilation, especially bootstrap builds
2404 without a previously built glibc.
2406 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2407 Chris Metcalf from Tilera.
2409 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2410 now supported for ARM processors.
2412 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2413 as the name of the dynamic linker, to distinguish it from the
2414 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2416 * Support for CRIS has been removed from ports.
2418 * A new class of installed header has been documented for low-level
2419 platform-specific functionality. PowerPC added the first instance with a
2420 function to provide time base register access. Contributed by Tulio
2421 Magno Quites Machado Filho.
2423 * ix86 configurations now install header files that are consistent with
2424 what x86-64 configurations install. These same header files can be used
2425 for -m32, -m64, or -mx32 builds.
2426 Contributed by H.J. Lu.
2428 * Math library bug fixes. A thorough audit of all open math library bugs was
2429 conducted by Joseph Myers. Significant progress was made on many math
2430 library bugs resulting in more accurate exceptions and function results.
2431 Many thanks to all those that contributed including Andreas Jaeger for his
2432 patch review and work on the x87 trigonometric instruction issues.
2434 * Timezone data is no longer installed. Timezone-related binaries and scripts
2435 will continue to be installed. Users should obtain their timezone data from
2436 their distribution provider or from the tzdata package at
2437 <ftp://munnari.oz.au/pub/>.
2441 * The following bugs are resolved with this release:
2443 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2444 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2445 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2446 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2447 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2448 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2449 13484, 13506, 13515, 13523, 13524, 13538, 13540
2451 * New program pldd to list loaded object of a process
2452 Implemented by Ulrich Drepper.
2454 * Add nss_db support back to glibc. No more dependency on Berkeley db
2455 and support for initgroups lookups.
2456 Implemented by Ulrich Drepper.
2458 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2459 Contributed by HJ Lu.
2461 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2462 Contributed by HJ Lu.
2464 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2465 on x86-32 and x86-64.
2466 Contributed by Liubov Dmitrieva.
2468 * Optimized strchr and strrchr for SSE on x86-32.
2469 Contributed by Liubov Dmitrieva.
2471 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2472 for x86-64 and x86-32.
2473 Contributed by Liubov Dmitrieva.
2475 * New interfaces: scandirat, scandirat64
2476 Implemented by Ulrich Drepper.
2478 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2479 Implemented by Ulrich Drepper.
2481 * nscd now also caches the netgroup database.
2482 Implemented by Ulrich Drepper.
2484 * Integrate libm with gcc's -ffinite-math-only option.
2485 Implemented by Ulrich Drepper.
2487 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2488 to math functions. Implemented by Ulrich Drepper.
2490 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2491 Implemented by Ulrich Drepper.
2493 * New Linux interfaces: process_vm_readv, process_vm_writev
2495 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2496 Implemented by Ulrich Drepper.
2498 * Optimized nearbyint and strcasecmp for PPC.
2499 Implemented by Adhemerval Zanella.
2501 * New locales: bho_IN, unm_US, es_CU, ta_LK
2505 * The following bugs are resolved with this release:
2507 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2508 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2509 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2510 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2511 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2512 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2513 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2514 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2515 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2516 12795, 12811, 12813, 12814, 12841
2518 * The RPC implementation in libc is obsoleted. Old programs keep working
2519 but new programs cannot be linked with the routines in libc anymore.
2520 Programs in need of RPC functionality must be linked against TI-RPC.
2521 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2523 Visible changes of this change include (obviously) the inability to link
2524 programs using RPC functions without referencing the TI-RPC library and the
2525 removal of the RPC headers from the glibc headers.
2526 Implemented by Ulrich Drepper.
2528 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2529 syncfs, setns, sendmmsg
2531 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2532 yue_HK, lij_IT, mhr_RU
2534 * New encodings: CP770, CP771, CP772, CP773, CP774
2536 * New program sotruss to trace calls through PLTs
2537 Implemented by Ulrich Drepper.
2539 * The malloc hook implementation is marked deprecated and will be removed
2540 from the default implementation in the next version. The design never
2541 worked ever since the introduction of threads. Even programs which do
2542 not create threads themselves can use multiple threads created internally.
2546 * The following bugs are resolved with this release:
2548 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2549 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2550 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2551 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2554 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2556 * POWER7 optimizations: memset, memcmp, strncmp
2558 * New optimized string functions for x86-64: strnlen (SSE2),
2559 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2560 Implemented by Ulrich Drepper.
2564 * The following bugs are resolved with this release:
2566 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2567 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2568 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2569 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2570 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2571 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2572 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2573 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2575 * New interfaces: pthread_getname_np, pthread_setname_np
2577 * New Linux interface: recvmmsg
2579 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2581 * The dynamic linker now recognizes supported ABI versions from the
2582 EI_ABIVERSION field in the ELF header.
2583 Implemented by Ulrich Drepper.
2585 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2586 will not be used to fill in password fields in the passwd.byname replies.
2587 Instead it is used to synthesize the shadow.byname table, should it be
2588 missing. This is a useful mode in some installations involving Solaris.
2589 Implemented by Ulrich Drepper.
2591 * New locales: kok_IN, sq_MK, cv_RU
2595 * The following bugs are resolved with this release:
2597 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2598 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2599 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2600 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2601 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2602 10789, 10817, 10823, 10840, 10847
2604 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2606 Implemented by Ulrich Drepper.
2608 * Checking version of longjmp added that fails if an uninitialized stack
2609 frame would be created. Implemented by Ulrich Drepper.
2611 * STT_GNU_IFUNC is now supported in static executables.
2612 Implemented by H.J. Lu.
2614 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2616 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2617 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2618 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2619 Contributed by H.J. Lu.
2621 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2622 Implemented by Ulrich Drepper.
2624 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2626 Contributed by H.J. Lu.
2628 * Support for fma instruction in AVX on x86-64.
2629 Implemented by H.J. Lu and Ulrich Drepper.
2631 * AVX support in x86-64 auditing support in ld.so.
2632 Implemented by H.J. Lu.
2634 * STB_GNU_UNIQUE support added.
2635 Implemented by Ulrich Drepper.
2637 * Implement second fallback mode for optimized DNS lookup for even more
2638 broken environments. If two requests from the same port are not
2639 handled correctly close the socket and open a new one before sending
2640 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2641 can be used to select this mode right away, instead of rediscovering the
2642 necessity is every process again.
2643 Implemented by Ulrich Drepper.
2645 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
2646 Implemented by Adam Tkac.
2648 * Optimized iconv conversions for S390x.
2649 Implemented by Andreas Krebbel.
2651 * Using condvars with PI mutexes is now more efficient due to kernel
2652 support for requeueing to PI futexes. NPTL support added for x86-64.
2653 Implemented by Ulrich Drepper.
2655 * New locales: ps_AF, my_MM
2659 * The following bugs are resolved with this release:
2661 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
2662 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
2663 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
2664 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
2665 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
2667 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
2668 Implemented by Ulrich Drepper.
2670 * New Linux interfaces: accept4, fallocate, fallocate64.
2671 Implemented by Ulrich Drepper.
2673 * Correct declarations of string function when used in C++ code. This
2674 could lead to compile errors for invalid C++ code.
2676 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
2679 * New POSIX 2008 interface: psiginfo
2680 Implemented by Ulrich Drepper.
2682 * New ISO C++1x interfaces: quick_exit, at_quick_exit
2683 Implemented by Ulrich Drepper.
2685 * Support for selecting between multiple function definitions at runtime
2686 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
2688 * The libcrypt library can now use the hash function implementations in
2689 NSS. Implemented by Ulrich Drepper.
2691 * The malloc implementation can be compiled to be less memory efficient
2692 but higher performing in multi-threaded programs.
2693 Implemented by Ulrich Drepper.
2695 * New locales: nan_TW@latin, ks_IN
2697 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
2698 Implemented by Ulrich Drepper.
2700 * Extended printf hook support. It is possible to use user-defined types
2701 and extend existing format specifiers.
2702 Implemented by Ulrich Drepper.
2704 * Handling for group shadow files has been added.
2705 Implemented by Ulrich Drepper.
2707 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
2708 servers (the case, e.g., for some people using the built-in DNS
2709 server in ADSL modems/routers). There is a once-per-process timeout
2710 in case of a broken server. To avoid it, users can run nscd or put
2711 'options single-request' in /etc/resolv.conf.
2712 Implemented by Ulrich Drepper.
2716 * The following bugs are resolved with this release:
2718 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
2719 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
2720 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
2721 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
2722 6968, 6974, 6980, 6995, 7008, 7009, 7029
2724 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
2725 up at the same time. Implemented by Ulrich Drepper.
2727 * TLS descriptors for LD and GD on x86 and x86-64.
2728 Implemented by Alexandre Oliva.
2730 * getaddrinfo now handles DCCP and UDPlite.
2731 Implemented by Ulrich Drepper.
2733 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
2734 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
2735 Implemented by Ulrich Drepper.
2737 * New implementation of memmem, strstr, and strcasestr which is O(n).
2738 Implemented by Eric Blake.
2740 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
2742 * Implement "e" option for popen to open file descriptor with the
2743 close-on-exec flag set. Implemented by Ulrich Drepper.
2745 * Implement "b" mode for fmemopen. In this mode writes writes don't
2746 implicitly add a NUL byte and seeks from the end of the buffer really
2747 use the buffer end, not the string length as the basis.
2748 Implemented by Ulrich Drepper.
2750 * Many functions, exported and internal, now atomically set the close-on-exec
2751 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
2753 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
2755 Implemented by Pravin Satpute.
2757 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
2761 * The following bugs are resolved with this release:
2763 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
2764 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
2765 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
2766 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
2767 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
2768 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
2769 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
2771 * New locales: bo_CN, bo_IN, shs_CA.
2773 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
2775 * Sorting rules for some Indian languages (Devanagari and Gujarati).
2776 Implemented by Pravin Satpute.
2778 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
2780 * nscd caches now all timeouts for DNS entries
2781 Implemented by Ulrich Drepper.
2783 * nscd is more efficient and wakes up less often.
2784 Implemented by Ulrich Drepper.
2786 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
2787 vdprintf, and obstack_vprintf.
2788 Implemented by Jakub Jelinek.
2790 * Faster memset for x86-64.
2791 Implemented by Harsha Jagasia and H.J. Lu.
2793 * Faster memcpy on x86.
2794 Implemented by Ulrich Drepper.
2796 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
2797 Implemented by Ulrich Drepper.
2799 * Faster sqrt and sqrtf implemention for some PPC variants.
2800 Implemented by Steven Munroe.
2804 * The following bugs are resolved with this release:
2806 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
2807 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
2808 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
2809 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
2810 5071, 5103, 5104, 5112, 5113, 5184, 5186
2812 * More checking functions: fread, fread_unlocked, open*, mq_open.
2813 Implemented by Jakub Jelinek and Ulrich Drepper.
2815 * Extend fortification to C++. Implemented by Jakub Jelinek.
2817 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
2818 by not recognizing 'a' as a modifier when those specs are requested.
2819 Implemented by Jakub Jelinek.
2821 * PPC optimizations to math and string functions.
2822 Implemented by Steven Munroe.
2824 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
2825 options to be passed. Implemented by Ulrich Drepper.
2827 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
2829 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
2830 Implemented by Ulrich Drepper.
2832 * Handle private futexes in the NPTL implementation.
2833 Implemented by Jakub Jelinek and Ulrich Drepper.
2835 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
2836 Implemented by Roland McGrath and Ulrich Drepper.
2838 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
2840 * SHA-256 and SHA-512 based password encryption.
2841 Implemented by Ulrich Drepper.
2843 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
2844 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
2847 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
2848 Implemented by Ulrich Drepper.
2852 * The following bugs are resolved with this release:
2854 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
2855 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
2856 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
2857 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
2858 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
2859 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
2860 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
2863 * New Linux interfaces: epoll_pwait, sched_getcpu.
2865 * New generic interfaces: strerror_l.
2867 * nscd can now cache the services database. Implemented by Ulrich Drepper.
2871 * The following bugs are resolved with this release:
2873 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
2874 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
2875 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
2876 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
2877 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
2878 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
2879 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
2880 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
2881 3225, 3251, 3252, 3253, 3273
2883 * For Linux, the sorting of addresses returned by getaddrinfo now also
2884 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
2885 Implemented by Ulrich Drepper.
2887 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
2888 Implemented by Ulrich Drepper.
2890 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
2892 * New iconv module for MIK. Contributed by Alexander Shopov.
2894 * For sites with broken group and/or passwd database, the auto-propagate
2895 option of nscd can prevent creating ID lookup entries from the results
2896 of a name lookup and vice versa. This usually is no problem but some
2897 site might have problems with the default behavior.
2898 Implemented by Ulrich Drepper.
2900 * Iterating over entire database in NIS can be slow. With the
2901 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
2902 to trade time for memory. The entire database will be read at once.
2903 Implemented by Ulrich Drepper.
2905 * The interfaces introduced in RFC 3542 have been implemented by
2908 * Support for the new ELF hash table format was added by Ulrich Drepper.
2910 * Support for priority inheritance mutexes added by Jakub Jelinek and
2913 * Support for priority protected mutexes added by Jakub Jelinek.
2915 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
2916 or_IN, csb_PL, fy_NL, sr_ME.
2920 * More overflow detection functions.
2922 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
2923 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
2925 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
2926 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
2927 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
2928 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
2929 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
2930 by Masahide Washizawa.
2932 * It is now possible to install audit libraries for the dynamic linker using
2933 LD_AUDIT. Implemented by Ulrich Drepper.
2935 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
2936 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
2937 For a libc and libpthread that works well on Linux 2.4 kernels, we
2938 recommend using the stable 2.3 branch.
2940 * The new function open_wmemstream acts like open_memstream,
2941 but uses a wchar_t wide character buffer.
2943 * The new function ppoll is an improved version of poll, similar to pselect.
2945 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
2946 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
2947 renameat, unlinkat, linkat, symlinkat, readlinkat.
2949 * New Linux kernel system calls: unshare,
2950 inotify_init, inotify_add_watch, inotify_rm_watch.
2952 * The euidaccess function is now also known by the alias eaccess,
2953 for compatibility with some other systems.
2955 * Timezone data updated to 2006b version.
2959 * The following bugs are resolved with this release:
2961 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
2962 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
2963 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
2964 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
2965 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
2966 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
2968 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2970 * As of this release, GCC 4 can be used to compile the C Library.
2972 * Timezone data updated to 2005m version.
2976 * The following bugs are resolved with this release:
2978 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
2979 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
2980 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
2981 777, 787, 821, 822, 823, 825
2983 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
2987 * Support for RFC 3678. Real implementations exist only for Linux so far.
2988 Implemented by Ulrich Drepper.
2990 * nscd can now cache entries persistently. Expiring entries are reloaded.
2991 For speedups the cache can be shared in memory with client processes.
2992 Implemented by Ulrich Drepper.
2994 * nscd can now perform SELinux checks.
2995 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
2997 * getaddrinfo queries are now cached. Canonical name lookup is performed
2999 Implemented by Ulrich Drepper.
3001 * The nothrow function attribute is used when headers are used by gcc when
3002 compiling C code. This can avoid significant amounts of exception
3005 * The malloc functions perform more error checking and are stricter when
3006 it comes to reacting on errors. The default action is to terminate
3007 the process after showing an error message. Implemented by Ulrich Drepper.
3009 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3010 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3012 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3013 namespaces using the new function dlmopen(). This feature is of course,
3014 like most other dynamic loading functionality, not available in statically
3015 linked applications. Implemented by Ulrich Drepper.
3017 * Low-overhead boundary checking variants of string and some stdio functions
3018 were added. These are to be used in conjunction with a gcc patch by
3019 Jakub Jelinek which adds calls to these functions if possible.
3020 Implemented by Jakub Jelinek and Ulrich Drepper.
3022 * Old code for several operating systems and machine architectures that
3023 have not been in working condition in a long time have been removed from
3024 the main source tree maintained by the GNU C Library's maintainers.
3025 These files are now reside in the separate `ports' source module
3026 that is usable as an add-on when building the library.
3030 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3031 interrogate the dynamic linker, compatible with the Solaris interface.
3033 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3034 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3036 * getifaddrs now uses the netlink interface on Linux to get its information.
3037 Implemented by Thorsten Kukuk.
3039 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3040 Implemented by Ulrich Drepper.
3042 * support for non-executable stacks on x86 has been added. Changes mostly
3045 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3048 * getaddrinfo now performs destination address selection according to
3053 * Thread-safe interfaces for many functions that access locale data
3054 were added in version 2.3, but these features were omitted from NEWS.
3055 Many functions have variants with an `_l' suffix that take a `locale_t'
3056 object as a parameter rather than consulting the current locale.
3057 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3058 create and maintain `locale_t' objects. Additionally, the new function
3059 `uselocale' sets "the current locale" (as used by functions not so
3060 parameterized) set for an individual thread. These features were added
3061 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3063 * The functions getresuid, getresgid, setresuid, and setresgid, which
3064 have long been available on Linux, are now declared in <unistd.h>
3065 and are now also available on the Hurd.
3067 * ELF thread-local storage support (TLS) now works on x86-64.
3069 * The new dynamic string token $LIB is expanded in shared library names.
3070 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3072 * Aldy Hernandez contributed complete software floating point support for
3073 PowerPC machines with no FPU.
3075 * fexecve is implemented on Linux.
3077 * The `btowc' function should work at least twice as fast due to
3078 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3080 * With approriate thread add-ons cancelable functions are now implemented
3081 in libc.so as well. No need to call the function in libpthread. This
3082 change allowed to finally disable the incorrect and expensive handling
3083 of weak definition in ld.so.
3085 * Yet more PLT entries in libc.so have been removed. We finally arrived
3086 at the bare minimum. Startup times improved appropriately.
3088 * Support for the new Linux/x86 system call interface was added. The
3089 AT_SYSINFO auxiliary vector entry is recognized and handled.
3093 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3096 * iconv (the program and the interface) now accepts empty names (excluding
3097 options like //TRANSLIT) to mean "use charset of current locale".
3099 * localedef can now transliterate characters in strings which are not in
3100 the provided charmap. The information from the input locale is used.
3102 * Prelinking support was added for ELF targets. This requires additional
3103 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3105 * Read-only stdio streams now use mmap to speed up operation by eliminating
3106 copying and buffer underflows. To use add 'm' to the mode string of
3107 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3109 * The malloc functions were completely rewritten by Wolfram Gloger based
3110 on Doug Lea's malloc-2.7.0.c.
3112 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3113 implementation of regex.
3115 * Bruno Haible upgraded the iconv and locale implementation to support
3118 * Contents of the LC_* and LANG environment variables in the CEN style are
3119 not recognized anymore. It never was used. Change by Ulrich Drepper.
3121 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3122 thread-local storage (TLS) ABI on some platforms.
3123 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3125 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3126 EUC-JISX0213, and TSCII.
3128 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3129 BSD-compatible interface for getting all network interface addresses.
3130 Implementation for IPv4 by Roland McGrath.
3132 * Loading of locale data is faster due to the introduction of a locale
3133 archive. Implemented by Roland McGrath and Ulrich Drepper.
3135 * Startup times are significantly reduced by not using exported functions
3136 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3139 * Steven Munroe contributed a port to PowerPC64/Linux.
3143 * The Hurd now uses the GNU libio implementation of stdio rather than the
3144 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3146 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3147 with the same behavior as the Linux system call of the same name.
3151 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3152 128-bit long double format.
3154 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3155 IBM1161, and IBM1162 charsets.
3157 * Andreas Jaeger contributed a port to x86-64/Linux.
3159 * Peter Bruin contributed a port to PowerPC/Hurd.
3161 * libc functions using I/O streams now can handle wide-oriented streams
3164 * optimizations in the dynamic linker. Binaries created by recent binutils
3165 versions start up quicker due to reduced time spent on relocations.
3167 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3171 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3172 asinh, atanh, j0 for the 128-bit long double format.
3174 * Bruno Haible updated all the code handling Unicode in some form to
3175 support Unicode 3.1.
3177 * Speed of regex for single-byte locales is back to previous levels.
3178 Patch by Isamu Hasegawa.
3180 * Alpha, SPARC, and IA-64 now also using floating stacks.
3182 * Startup time of internationalized applications greatly improved through
3183 iconv cache. Use iconvconfig to generate the cache file.
3184 Contributed by Ulrich Drepper.
3186 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3187 severe performance problems. Patch by David Mosberger.
3189 * The Hurd port got a lot more functionality like AIO, various stdio
3190 extensions, etc. Mainly done by Roland McGrath.
3192 * mtrace can now lookup symbols in shared libraries.
3196 * Intel's IA-64 math library is largely integrated. It provides fast and
3197 accurate implementatations for most basic and standard math functions
3198 in float, double, and long double format.
3200 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3201 and asin for the 96-bit long double format and asin, log, tan for the
3202 128-bit long double format.
3204 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3205 The basic double functions exist today. Contributed by Abraham Ziv
3206 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3207 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3209 * An asynchronous name lookup library was added. The interface is designed
3210 after POSIX AIO. The proposal was circulated beforehand to get comments.
3211 No negative ones came in. Implemented by Ulrich Drepper.
3213 * Port to S390/64bit contributed by Martin Schwidefsky
3214 <schwidefsky@de.ibm.com>.
3216 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3217 of functions for Linux/IA-64.
3219 * The RPC code is now thread safe. Threads can now use the same service
3220 of different services at the same time. Patch by Eric Norum
3221 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3223 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3224 family of functions for Linux/S390.
3226 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3227 of functions for Linux/x86.
3229 * Port to Linux/CRIS contributed by Axis Communications.
3233 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3234 we can now check for namespace violations and missing declarations. The
3235 result is that almost all headers are now Unix-compliant (as defined in
3236 the upcoming XPG6). The negative side is that some programs might need
3237 corrections, too, if they depend on the incorrect form of the headers in
3238 previous versions which defined too many symbols and included too many
3241 * regex now handles multibyte character sets correctly.
3242 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3244 * iconv (the program) does now conform to the upcoming XPG6 and handles
3245 charmaps. Instead of the charset names the path of charmaps can be
3246 provided and the conversion happens based on this data.
3247 Contributed by Ulrich Drepper.
3249 * The locale program now provides detailed information about the installed
3254 only lists the names of the supported locales
3258 provides details such as country, language, and codeset name.
3259 Contributed by Ulrich Drepper.
3263 * The gencat program now parses the input file according to the charset
3264 selected by the LC_CTYPE category. This is important for stateful
3265 character sets. To make generating catalogs easier there is a way
3266 to overwrite the charset selected by the locale: before the first
3267 message or $ quote line the catalog can contain a line like
3269 $ codeset=ISO-8859-2
3271 to select the charset (ISO-8859-2 in this case).
3273 Implemented by Shinya Hanataka and Ulrich Drepper.
3275 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3276 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3277 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3279 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3280 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3282 * The LANGUAGE environment variable is now ignored unless the locale is
3283 changed from the default "C" locale.
3285 * The usual bug fixes.
3289 * Greg McGary added runtime support for bounds checking using gcc's
3290 new -fbounded-pointers option. ix86 target is complete. PowerPC
3293 * Thorsten Kukuk added secure mode support to nscd.
3295 * The Berkeley db libraries have been removed.
3297 Related, the nss_db module is now in a separate package since it
3298 obviously requires a database library being available.
3300 * Wide character I/O streams implemented by Ulrich Drepper.
3302 * Functions from the extended socket API added by Ulrich Drepper.
3304 * Functions feenableexcept and fedisableexcept to control the
3305 behaviour of individual exceptions have been added by Andreas Jaeger.
3307 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3309 * The resolver code has been updated from bind 8.2.3-T5B which supports
3310 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3313 This change could in some situations effect backward compatibility. Since
3314 now `_res' is a thread-local instead of a global variable, modifying it
3315 in one thread does not have any effect in other threads.
3317 The resolver library was also extended to allow IPv6 as the transport
3318 protocol for the requests. This work was done by Stig Venaas.
3320 * Compatibility code for K&R C compilers has been removed from the
3321 header files. A ISO C compiler is needed to use the library
3322 (conforming to either C89 or C99 standard).
3324 * Complete rewrite of the localedef program to support multibyte character
3325 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3326 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3327 Implemented by Ulrich Drepper.
3329 Bruno Haible significantly improved the generation and use of the data
3330 structures for the wide character tables.
3332 * Plural handling in gettext implemented by Ulrich Drepper.
3334 * The utmp daemon has been removed.
3336 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3338 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3341 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3343 * POSIX barriers implemented by Kaz Kylheku.
3345 * POSIX spawn function family implemented by Ulrich Drepper.
3347 * POSIX spinlocks are now available.
3349 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3351 * the configure option --enable-kernel=X.Y.Z allows to strip out
3352 compatibility for kernel versions before X.Y.Z. This is currently only
3353 implemented for Linux.
3355 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3356 field sin6_scope_id. This means that all programs using IPv6 should be
3357 recompiled. Don't expect binary compatibility with previous glibc
3360 * various conversion modules for IBM character sets contributed by
3363 * IA-64 port by Jes Sorensen and HJ Lu.
3377 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3379 * New xdr functions are added; some rpc functions are now 64bit clean.
3381 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3383 * Fixed known incompatibilities with glibc 2.0.
3385 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3387 * Optimized a number of functions (especially the ELF dynamic loader).
3389 * Update timezone data files.
3391 * lots of charmaps corrections
3393 * some new locale definitions and charmaps
3398 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3399 conform to POSIX member type requirements. Symbol versions have been
3400 adjusted as needed within the library, and for direct use by applications,
3401 but there is potential for problems if third-party libraries use
3402 struct timeval as part of their interface. This does not present
3403 a problem for X and other "essential" system libraries.
3405 * An additional locale model to support C++ Standard Library locale
3406 model and probably more was implemented by Ulrich Drepper.
3408 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3411 * Miles Bader provided the `argp' function family to support hierachical
3412 command line argument parsing, layered on top of getopt.
3414 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3416 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3419 * scanf recognizes the %a and %A format for scanning floating point numbers.
3421 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3422 information and interfaces for the available integer types.
3424 * about 130 new math functions were added to implement the ISO C9x math
3427 * the new header <complex.h> contains definitions of the complex math
3428 functions from ISO C 9X.
3430 * the new header <tgmath.h> defines generic macros to use complex or
3431 real valued functions.
3433 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3435 * Andreas Jaeger provided a test suite for the math library.
3437 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3439 * Ulrich Drepper added character set conversion functions (iconv).
3441 * Optimized string functions have been added.
3443 * The localedata addon is now part of glibc.
3445 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3447 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3448 daemon for NSS (nscd).
3450 Missing a better place here are some numbers on improvements. Under
3451 Linux 2.1.125 un-tar-ing the kernel sources takes
3455 using local files 12.19s 6.88s 22.91s
3457 using NIS 13.92s 8.91s 26.34s
3459 using NIS & nscd 10.37s 7.34s 25.30s
3461 using NIS+ 27.57s 30.37s 640.46s
3463 using NIS+ & nscd 10.25s 7.83s 26.51s
3465 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3467 Keep in mind that non-namelookup related operations dominate above times.
3468 It was just a common complain that using NIS+ unpacking the kernel is
3471 [1] The old nscd implementation is not available anymore since it was
3472 distributed with glibc up to version 2.0.98 and thus is now replaced.
3474 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3476 * Mark Kettenis provided a Hesiod NSS module.
3478 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3479 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3481 * rcmd can now handle netgroups (Dick Streefland).
3483 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3486 * Support for the IPv6 protocol has been added to the socket API, as per the
3487 latest draft standards.
3489 * Support for Linux 2.2 has been added.
3491 * Interface changes relative to the latest 2.0.x release:
3492 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3493 addseverity NEW: Unix98
3494 alphasort64 NEW: LFS
3495 argp_err_exit_status NEW: argp, GNU ext
3496 argp_error NEW: argp, GNU ext
3497 argp_failure NEW: argp, GNU ext
3498 argp_help NEW: argp, GNU ext
3499 argp_parse NEW: argp, GNU ext
3500 argp_program_bug_address NEW: argp, GNU ext
3501 argp_program_version NEW: argp, GNU ext
3502 argp_program_version_hook NEW: argp, GNU ext
3503 argp_state_help NEW: argp, GNU ext
3504 argp_usage NEW: argp, GNU ext
3505 authdes_create NEW: Secure RPC
3506 authdes_getucred NEW: Secure RPC
3507 authdes_pk_create NEW: Secure RPC
3508 backtrace NEW: GNU ext.
3509 backtrace_symbols NEW: GNU ext.
3510 backtrace_symbols_fd NEW: GNU ext.
3512 cacosf NEW: ISO C 9x
3513 cacosh NEW: ISO C 9x
3514 cacoshf NEW: ISO C 9x
3515 cacoshl NEW: ISO C 9x
3516 cacosl NEW: ISO C 9x
3523 casinf NEW: ISO C 9x
3524 casinh NEW: ISO C 9x
3525 casinhf NEW: ISO C 9x
3526 casinhl NEW: ISO C 9x
3527 casinl NEW: ISO C 9x
3529 catanf NEW: ISO C 9x
3530 catanh NEW: ISO C 9x
3531 catanhf NEW: ISO C 9x
3532 catanhl NEW: ISO C 9x
3533 catanl NEW: ISO C 9x
3534 cbc_crypt NEW: Secure RPC
3538 ccoshf NEW: ISO C 9x
3539 ccoshl NEW: ISO C 9x
3545 cimagf NEW: ISO C 9x
3546 cimagl NEW: ISO C 9x
3547 clearerr_locked REMOVED
3548 clntunix_create NEW: sunrpc ext
3550 clog10 NEW: ISO C 9x
3551 clog10f NEW: ISO C 9x
3552 clog10l NEW: ISO C 9x
3562 cprojf NEW: ISO C 9x
3563 cprojl NEW: ISO C 9x
3565 crealf NEW: ISO C 9x
3566 creall NEW: ISO C 9x
3571 csinhf NEW: ISO C 9x
3572 csinhl NEW: ISO C 9x
3575 csqrtf NEW: ISO C 9x
3576 csqrtl NEW: ISO C 9x
3580 ctanhf NEW: ISO C 9x
3581 ctanhl NEW: ISO C 9x
3583 des_setparity NEW: Secure RPC
3584 ecb_crypt NEW: Secure RPC
3585 endutxent NEW: Unix98
3587 exp10f NEW: ISO C 9x
3588 exp10l NEW: ISO C 9x
3592 fattach NEW: STREAMS
3593 fdetach NEW: STREAMS
3597 feclearexcept NEW: ISO C 9x
3598 fegetenv NEW: ISO C 9x
3599 fegetexceptflag NEW: ISO C 9x
3600 fegetround NEW: ISO C 9x
3601 feholdexcept NEW: ISO C 9x
3603 feraiseexcept NEW: ISO C 9x
3604 ferror_locked REMOVED
3605 fesetenv NEW: ISO C 9x
3606 fesetexceptflag NEW: ISO C 9x
3607 fesetround NEW: ISO C 9x
3608 fetestexcept NEW: ISO C 9x
3609 feupdateenv NEW: ISO C 9x
3610 fflush_locked REMOVED
3614 fgets_unlocked NEW: GNU ext.
3615 fileno_locked REMOVED
3627 fputc_locked REMOVED
3628 fputs_unlocked NEW: GNU ext.
3629 fread_unlocked NEW: GNU ext.
3634 fstatvfs NEW: Unix98
3638 ftruncate64 NEW: LFS
3640 fwrite_unlocked NEW: GNU ext.
3641 gai_strerror NEW: IPv6
3645 getchar_locked REMOVED
3647 getdate_err NEW: Unix98
3648 getdate_r NEW: GNU ext.
3650 getnameinfo NEW: IPv6
3651 getnetname NEW: Secure RPC
3652 getpmsg NEW: STREAMS
3653 getpt NEW: Unix98 PTY
3654 getrlimit64 NEW: LFS
3655 getutxent NEW: Unix98
3656 getutxid NEW: Unix98
3657 getutxline NEW: Unix98
3658 glob64 NEW: GNU ext.
3659 globfree64 NEW: GNU ext.
3660 gnu_get_libc_release NEW: GNU ext.
3661 gnu_get_libc_version NEW: GNU ext.
3662 grantpt NEW: Unix98 PTY
3663 host2netname NEW: Secure RPC
3665 iconv_close NEW: iconv
3666 iconv_open NEW: iconv
3667 if_freenameindex NEW: IPv6
3668 if_indextoname NEW: IPv6
3669 if_nameindex NEW: IPv6
3670 if_nametoindex NEW: IPv6
3671 in6addr_any NEW: IPv6
3672 in6addr_loopback NEW: IPv6
3673 inet6_isipv4mapped NEW: IPv6
3674 isastream NEW: STREAMS
3675 iswblank NEW: GNU ext.
3676 key_decryptsession NEW: Secure RPC
3677 key_decryptsession_pk NEW: Secure RPC
3678 key_encryptsession NEW: Secure RPC
3679 key_encryptsession_pk NEW: Secure RPC
3680 key_gendes NEW: Secure RPC
3681 key_get_conv NEW: Secure RPC
3682 key_secretkey_is_set NEW: Secure RPC
3683 key_setnet NEW: Secure RPC
3684 key_setsecret NEW: Secure RPC
3685 llrint NEW: ISO C 9x
3686 llrintf NEW: ISO C 9x
3687 llrintl NEW: ISO C 9x
3688 llround NEW: ISO C 9x
3689 llroundf NEW: ISO C 9x
3690 llroundl NEW: ISO C 9x
3695 lrintf NEW: ISO C 9x
3696 lrintl NEW: ISO C 9x
3697 lround NEW: ISO C 9x
3698 lroundf NEW: ISO C 9x
3699 lroundl NEW: ISO C 9x
3701 makecontext NEW: Unix98
3702 mempcpy NEW: GNU ext.
3705 modify_ldt NEW: kernel
3709 nearbyint NEW: ISO C 9x
3710 nearbyintf NEW: ISO C 9x
3711 nearbyintl NEW: ISO C 9x
3712 netname2host NEW: Secure RPC
3713 netname2user NEW: Secure RPC
3714 nexttoward NEW: ISO C 9x
3715 nexttowardf NEW: ISO C 9x
3716 nexttowardl NEW: ISO C 9x
3720 passwd2des NEW: Secure RPC
3722 pow10f NEW: GNU ext.
3723 pow10l NEW: GNU ext.
3726 printf_size NEW: GNU ext.
3727 printf_size_info NEW: GNU ext.
3728 profil_counter REMOVED
3729 pthread_mutexattr_getkind_np REPLACED
3730 pthread_mutexattr_setkind_np REPLACED
3731 ptsname NEW: Unix98 PTY
3732 ptsname_r NEW: Unix98 PTY
3734 putchar_locked REMOVED
3735 putgrent NEW: GNU ext.
3737 putpmsg NEW: STREAMS
3738 pututxline NEW: Unix98
3742 readdir64_r NEW: LFS
3743 remquo NEW: ISO C 9x
3744 remquof NEW: ISO C 9x
3745 remquol NEW: ISO C 9x
3747 roundf NEW: ISO C 9x
3748 roundl NEW: ISO C 9x
3750 scalbln NEW: ISO C 9x
3751 scalblnf NEW: ISO C 9x
3752 scalblnl NEW: ISO C 9x
3754 sendfile NEW: kernel
3755 setcontext NEW: Unix98
3756 setrlimit64 NEW: LFS
3757 setutxent NEW: Unix98
3759 sigignore NEW: Unix98
3760 sigqueue NEW: POSIX.1b
3761 sigrelse NEW: Unix98
3762 sigset NEW: POSIX.1b
3763 sigtimedwait NEW: POSIX.1b
3764 sigwaitinfo NEW: POSIX.1b
3765 sincos NEW: GNU ext.
3766 sincosf NEW: GNU ext.
3767 sincosl NEW: GNU ext.
3771 strcasestr NEW: GNU ext.
3772 strtoimax NEW: ISO C 9x
3773 strtoumax NEW: ISO C 9x
3774 strverscmp NEW: GNU ext.
3775 svcauthdes_stats NEW: Secure RPC
3776 svcunix_create NEW: sunrpc etx
3777 svcunixfd_create NEW: sunrpc ext
3778 swapcontext NEW: Unix98
3779 tcgetsid NEW: Unix98 PTY
3780 tdestroy NEW: GNU ext.
3781 tgamma NEW: ISO C 9x
3782 tgammaf NEW: ISO C 9x
3783 tgammal NEW: ISO C 9x
3787 truncf NEW: ISO C 9x
3788 truncl NEW: ISO C 9x
3790 unlockpt NEW: Unix98 PTY
3791 updwtmpx NEW: Unix98
3792 user2netname NEW: Secure RPC
3793 utmpxname NEW: Unix98
3794 versionsort NEW: GNU ext.
3795 versionsort64 NEW: GNU ext.
3797 wcscasecmp NEW: GNU ext.
3798 wcsncasecmp NEW: GNU ext.
3799 wcsnlen NEW: GNU ext.
3800 wcstoimax NEW: ISO C 9x
3801 wcstoll NEW: ISO C 9x
3802 wcstoull NEW: ISO C 9x
3803 wcstoumax NEW: ISO C 9x
3805 wordexp NEW: POSIX.2
3806 wordfree NEW: POSIX.2
3807 write_profiling REMOVED
3808 xdecrypt NEW: Secure RPC
3809 xdr_authdes_cred NEW: Secure RPC
3810 xdr_authdes_verf NEW: Secure RPC
3811 xdr_cryptkeyarg NEW: Secure RPC
3812 xdr_cryptkeyarg2 NEW: Secure RPC
3813 xdr_cryptkeyres NEW: Secure RPC
3814 xdr_getcredres NEW: Secure RPC
3815 xdr_key_netstarg NEW: Secure RPC
3816 xdr_key_netstres NEW: Secure RPC
3817 xdr_keybuf NEW: Secure RPC
3818 xdr_keystatus NEW: Secure RPC
3819 xdr_netnamestr NEW: Secure RPC
3820 xdr_sizeof NEW: Secure RPC
3821 xdr_unixcred NEW: sunrpc ext
3822 xencrypt NEW: Secure RPC
3823 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3834 * inet_ntoa is thread-safe
3836 * updwtmp is moved from libutil to libc
3838 * rewrite of cbrt function
3840 * update of timezone data
3854 * add atoll function
3856 * fix complex problems in Berkeley DB code
3858 * fix math functions
3862 * fixed lots of header problems (especially Linux/GNU specific)
3864 * dynamic loader preserves all registers
3866 * Roland McGrath provided support for handling of auxiliary objects in
3867 the ELF dynamic loader.
3869 * support for parallel builds is improved
3873 * GNU extensions are no longer declared by default. To enable them you
3874 must define the macro `_GNU_SOURCE' in your program or compile with
3877 * The library has changed from using GNU ld symbol aliases to using weak
3878 symbols where available. The ELF object file format supports weak
3879 symbols; GNU ld also supports weak symbols in the a.out format. (There
3880 is also now support for other GNU ld extensions in ELF. Use the
3881 `--with-elf' option to configure to indicate you have ELF, and
3882 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
3883 of many files which contained only symbol aliases, reducing the size of
3884 the source and the compiled library; many other files were renamed to
3885 less cryptic names previously occupied by the symbol alias files.
3886 There is a new header file <elf.h> for programs which operate on
3887 files in the ELF format.
3889 * Converted to Autoconf version 2, so `configure' has more options.
3890 Run `configure --help' to see the details.
3892 * The library can now be configured to build profiling, highly-optimized
3893 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
3894 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
3895 `configure' enable building these extra libraries. The shared library is
3896 built by default when using both ELF and GNU ld. When shared libraries
3897 are enabled, the new library `-ldl' is available for arbitrary run-time
3898 loading of shared objects; its interface is defined in <dlfcn.h>. The
3899 new header file <link.h> gives access to the internals of the run-time
3900 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
3901 application of same name on other systems and it provides information
3902 about dynamically linked binaries.
3904 * The C library now provides the run-time support code for profiling
3905 executables compiled with `-pg'. Programs can control the profiling code
3906 through the interface in <sys/gmon.h>. The `gmon.out' files written by
3907 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
3908 the support for this file format was contributed by David Mosberger-Tang.
3910 * The math code has been replaced with a math library based on fdlibm from
3911 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
3912 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
3913 functions. The math functions now reside in a separate library, so
3914 programs using them will need to use `-lm' their linking commands.
3916 * John C. Bowman contributed optimized ix87 assembler inline functions.
3918 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
3919 mechanism similar to that found in Solaris 2. This is now used for the
3920 group, passwd, hosts, networks, services, protocols, rpc, ethers,
3921 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
3922 file controls what services are used for each individual database. This
3923 works by loading shared libraries with names specified in `nsswitch.conf',
3924 so service modules can be changed or added at any time without even
3925 relinking any program. Currently there are the file, db, and NIS based
3926 NSS services available.
3928 * The new functions `strtoq' and `strtouq' parse integer values from
3929 strings, like `strtol' and `strtoul', but they return `long long int' and
3930 `unsigned long long int' values, respectively (64-bit quantities).
3932 * The new functions `strtof' and `strtold' parse floating-point values from
3933 strings, like `strtod', but they return `float' and `long double' values,
3934 respectively (on some machines `double' and `long double' are the same).
3936 * Ulrich Drepper has contributed new implementations of the floating-point
3937 printing and reading code used in the `printf' family of functions and
3938 `strtod', `strtof', and `strtold'. These new functions are perfectly
3939 accurate, and much faster than the old ones.
3941 * The implementation of the POSIX locale model was completely rewritten by
3942 Ulrich Drepper. This includes the new programs `localedef' and `locale'
3943 to compile the POSIX locale definition.
3945 * The former dummy implementations of the strcoll and strxfrm function are
3946 now replaced by fully functional code contributed by Ulrich Drepper. The
3947 collation information comes from the POSIX locale definitions.
3949 * The new header <langinfo.h> defines an interface for accessing
3950 various locale-dependent data (using the locale chosen with `setlocale').
3952 * Ulrich Drepper has contributed a new suite of functions for operation on
3953 wide-character and multibyte-character strings, in <wchar.h>;
3954 and classification and case conversion of wide characters, in <wctype.h>.
3955 These new functions are conforming to the ISO C, Amendement 1 specification.
3957 * There is now a second implementation of the standard I/O library available.
3958 It comes from GNU libg++ as was written by Per Bothner, heavily modified
3959 by Hongjiu Lu and made thread safe by Ulrich Drepper.
3961 * You can now use positional parameter specifications in format strings
3962 for the `printf' and `scanf' families of functions. For example,
3963 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
3964 ``Number 6, Mr Jones''. This is mainly useful when providing different
3965 format strings for different languages, whose grammars may dictate
3966 different orderings of the values being printed. To support this
3967 feature, the interface for `register_printf_handler' has changed; see
3968 the header file <printf.h> for details.
3970 * The `printf' and `scanf' families of functions now understand a new
3971 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
3972 to group numbers as indicated by the locale; for `scanf' and friends, this
3973 says to accept as valid only a number with all the proper grouping
3974 separators in the right places. In the default "C" locale, numbers are
3975 not grouped; but locales for specific countries will define the usual
3976 conventions (i.e. separate thousands with `,' in the US locale).
3978 * The pgrp functions have been regularized, slightly incompatibly but much
3979 less confusingly. The core functions are now `getpgid' and `setpgid',
3980 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
3981 argument) and BSD `setpgrp' (identical to `setpgid') functions are
3982 provided for compatibility. There is no longer an incompatible `getpgrp'
3983 with an argument declared under _BSD_SOURCE; no BSD code uses it.
3985 * The new header file <fts.h> and suite of functions simplify programs that
3986 operate on directory trees. This code comes from 4.4 BSD.
3988 * The resolver code has been updated from the BIND 4.9.5-P1 release.
3989 Parts of the code were heavily modified by Ulrich Drepper to fit in the
3990 NSS scheme used in glibc.
3992 * There is a new malloc debugging hook `__memalign_hook'.
3994 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
3995 `unsigned int' in <sys/types.h>. These are for compatibility only and
3996 their use is discouraged.
3998 * The `-lmcheck' library to enable standard malloc debugging hooks is now
3999 done differently, so that it works even without GNU ld.
4001 * New function `euidaccess' checks allowed access to a file like `access',
4002 but using the effective IDs instead of the real IDs.
4004 * The time zone data files have been updated for the latest and greatest
4005 local time conventions of the countries of the world.
4007 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4010 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4011 for formatting floating-point numbers. They are provided only for
4012 compatibility; new programs should use `sprintf' instead. There are
4013 also equivalent function for the `long double' floating-point type and
4014 all functions also exist in a reentrant form.
4016 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4017 functions for maintaining the login-record files (primarily of use to
4018 system programs such as `login'), and convenient functions for
4019 allocating and initializing a pseudo-terminal (pty) device.
4021 * Ulrich Drepper has contributed new support for System V style
4022 shared memory and IPC on systems that support it.
4024 * Ulrich Drepper has contributed several miscellaneous new functions found
4025 in System V: The `hsearch' family of functions provide an effective
4026 implementation of hash tables; `a64l' and `l64a' provide a very simple
4027 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4030 * Ulrich Drepper has contributed new reentrant counterparts for the
4031 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4033 * Ulrich Drepper has contributed new, highly-optimized versions of several
4034 string functions for the i486/Pentium family of processors.
4036 * Ulrich Drepper has updated the Linux-specific code, based largely
4037 on work done in Hongjiu Lu's version of GNU libc for Linux.
4038 The GNU library now supports Linux versions 2.0.10 and later,
4039 using the ELF object file format (i[3456]86-*-linux).
4041 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4043 * David Mosberger-Tang and Richard Henderson have ported the C library
4044 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4045 dynamic linking support for ELF/Alpha.
4047 * Richard Henderson contributed several Alpha optimized assembler function
4048 for arithmetic and string handling.
4050 * Ulrich Drepper has contributed a new set of message catalog functions to
4051 support multiple languages using the <libintl.h> interface, for use with
4052 his new package GNU gettext. Translation volunteers have contributed
4053 catalogs of the library's messages in Spanish, German, and Korean.
4055 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4056 program and the `catgets' function for reading the catalog files it
4057 creates. (The <libintl.h> interface is preferred; we include the
4058 <nl_types.h> interface using `catgets' only for source compatibility with
4059 programs already written to use it.)
4061 * New header file <values.h> gives SVID-compatible names for <limits.h>
4064 * Various new macros, declarations, and small header files for compatibility
4067 * New function `group_member' is a convenient way to check if a process has
4068 a given effective group ID.
4070 * When using GCC 2.7 and later, the socket functions are now declared in a
4071 special way so that passing an argument of type `struct sockaddr_in *',
4072 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4073 `struct sockaddr *' type, does not generate a type-clash warning.
4075 * New function `error' declared in header file <error.h> is a convenient
4076 function for printing error messages and optionally exiting; this is the
4077 canonical function used in GNU programs. The new functions `err', `warn',
4078 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4079 doing the same thing.
4081 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4082 POSIX.2 `glob' function to do ~ and {...} expansion.
4084 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4085 `clearenv' which is used in POSIX.9 is also available.
4087 * New function `getsid' returns session ID number on systems that support it.
4089 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4090 files <db.h> and <mpool.h> provide a rich set of functions for several
4091 types of simple databases stored in memory and in files, and <ndbm.h> is
4092 an old `ndbm'-compatible interface using the `db' functions. Link with
4093 `-ldb' to get these functions.
4095 * New macro `strdupa' copies a string like `strdup', but uses local stack
4096 space from `alloca' instead of dynamic heap space from `malloc'.
4098 * New function `strnlen' is like `strlen' but searches only a given maximum
4099 number of characters for the null terminator. `stpncpy', `strndup' and
4100 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4103 * New function `statfs' in header <sys/statfs.h>.
4105 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4106 provide convenient functions for operating on blocks of null-terminated
4109 * A new suite of functions in <utmp.h> handle all the details of reading
4110 and writing the utmp file.
4112 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4115 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4116 completely POSIX compliant and also implemented the extended functionality
4117 to handle alternate digit representation and alternate era date formats.
4119 * Ulrich Drepper provided an implementation of the `strptime' function
4120 defined in XPG4.2 which transforms a string into a `struct tm' value.
4122 * Paul Eggert provided the tzselect shell script as part of the timezone
4123 code. The shell script makes it easy to select the correct timezone
4126 * The implementation of the malloc family of functions is completely replaced
4127 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4128 The implementation uses the mmap function (if available) and it is
4129 optimized for the use in multi threaded programs.
4131 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4132 functions. This new functionality is usable by specifying a special
4133 salt string and it is compatible with implementation on *BSD systems.
4135 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4136 `getsubopt' to handle second level command line options, `bsd_signal'
4137 to access BSD style `signal' functionality, the obsolete `regexp' style
4140 * the `lchown' function is available on system which support this
4143 * The implementation of the shadow password handling function was contributed
4146 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4148 * POSIX.1g support was added. The <sys/select.h> header is available,
4149 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4150 implementation of `getaddrinfo'.
4154 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4156 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4157 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4159 * The new function `daemon' from 4.4 BSD is useful for server programs that
4160 want to put themselves in the background.
4162 * Joel Sherrill has contributed support for several standalone boards that
4163 run without an operating system.
4165 * `printf', `scanf' and friends now accept a `q' type modifier for long
4166 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4168 * All of the code taken from BSD (notably most of the math and networking
4169 routines) has been updated from the BSD 4.4-Lite release.
4171 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4173 * The new functions `getdomainname' and `setdomainname' fetch or change the
4174 YP/NIS domain name. These are system calls which exist on systems which
4177 * The time zone data files have been updated for the latest international
4180 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4181 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4185 * The C library now includes support for Sun RPC, from Sun's free
4186 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4187 are included. (There is still no support for YP.)
4189 * Tom Quinn has contributed a port of the C library to SGI machines running
4190 Irix 4 (mips-sgi-irix4).
4192 * The new `lockf' function is a simplified interface to the locking
4193 facilities of `fcntl', included for compatibility.
4195 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4197 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4199 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4202 * The error message printed by `assert' for a failed assertion now includes
4203 the name of the program (if using GNU ld) and the name of the calling
4204 function (with versions of GCC that support this).
4206 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4208 * The library now includes the <sys/mman.h> header file and memory
4209 management functions `mmap', `munmap', `mprotect', `msync', and
4210 `madvise', on systems that support those facilities.
4212 * The interface for `mcheck' has changed slightly: the function called to
4213 abort the program when an allocation inconsistency is detected now takes
4214 an argument that indicates the type of failure. The new function
4215 `mprobe' lets you request a consistency check for a particular block at
4216 any time (checks are normally done only when you call `free' or `realloc'
4219 * It is now possible to easily cross-compile the C library, building on one
4220 system a library to run on another machine and/or operating system. All
4221 you need to do is set the variable `HOST_CC' in `configparms' to the
4222 native compiler for programs to run on the machine you are building on (a
4223 few generator programs are used on Unix systems); set `CC' to the
4226 * The new function `fexecve' (only implemented on the GNU system) executes
4227 a program file given a file descriptor already open on the file.
4231 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4232 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4234 * You can set the variable `libprefix' in `configparms' to specify a prefix
4235 to be prepended to installed library files; this makes it easy to install
4236 the GNU C library to be linked as `-lgnuc' or whatever.
4238 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4239 copies a limited number of characters from a string, and returns the
4240 address of the last character written.
4242 * You no longer need to check for whether the installed `stddef.h' is
4243 compatible with the GNU C library. configure now checks for you.
4245 * You can now define a per-stream `fileno' function to convert the
4246 stream's cookie into an integral file descriptor.
4248 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4249 allocates zero bytes of storage, and returns a unique pointer which
4250 you can pass to `realloc' or `free'. The behavior is undefined if
4251 you dereference this pointer.
4253 * The C library now runs on Sony NEWS m68k machines running either
4254 NewsOS 3 or NewsOS 4.
4256 * The new `syscall' function is a system-dependent primitive function
4257 for invoking system calls. It has the canonical behavior on Unix
4258 systems, including unreliable return values for some calls (such as
4259 `pipe', `fork' and `getppid').
4261 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4262 to `EAGAIN', which is the preferred name. On systems whose kernels
4263 use two distinct codes, the C library now translates EWOULDBLOCK to
4264 EAGAIN in every system call function.
4268 * The GNU C Library Reference Manual is now distributed with the library.
4269 `make dvi' will produce a DVI file of the printed manual.
4270 `make info' will produce Info files that you can read on line using C-h i
4271 in Emacs or the `info' program.
4272 Please send comments on the manual to bug-glibc-manual@gnu.org.
4274 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4276 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4278 * Jason Merrill has contributed a port to the Sequent Symmetry running
4279 Dynix version 3 (i386-sequent-dynix).
4281 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4282 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4284 * New function `memory_warnings' lets you arrange to get warnings when
4285 malloc is running out of memory to allocate, like Emacs gives you.
4287 * The C library now contains the relocating allocator used in Emacs 19 for
4288 its editing buffers. This allocator (ralloc) minimizes allocation
4289 overhead and fragmentation by moving allocated regions around whenever it
4290 needs to. You always refer to a ralloc'd region with a "handle" (a
4291 pointer to a pointer--an object of type `void **').
4293 * There is a new `printf' format: `%m' gives you the string corresponding
4294 to the error code in `errno'.
4296 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4297 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4298 pass, the `a' modifier says to fill in a `char **' you pass with a
4301 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4302 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4303 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4305 * `mkstemp' is a traditional Unix function to atomically create and open a
4306 uniquely-named temporary file.
4310 * The standard location for the file that says what the local timezone is
4311 has changed again. It is now `/usr/local/etc/localtime' (or more
4312 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4314 * The distribution no longer contains any files with names longer than 14
4317 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4318 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4320 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4324 * The configuration process has changed quite a bit. The `configure'
4325 script is now used just like the configuration scripts for other GNU
4326 packages. The `sysdeps' directory hierarchy is much rearranged.
4327 The file `INSTALL' explains the new scheme in detail.
4329 * The header files no longer need to be processed into ANSI C and
4330 traditional C versions. There is just one set of files to install, and
4331 it will work with ANSI or old C compilers (including `gcc -traditional').
4333 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4334 MIPS DECStation running Ultrix 4.
4336 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4337 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4338 made itself into a shared library.
4340 * Yet further improved support for the i386, running 4.3 BSD-like systems
4341 (such as Mach 3 with the Unix single-server), or System V.
4343 * New function `strncasecmp' to do case-insensitive string comparison
4344 with limited length.
4346 * New function `strsep' is a reentrant alternative to `strtok'.
4348 * New functions `scandir' and `alphasort' for searching directories.
4350 * New function `setenv' is a better interface to `putenv'.
4352 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4353 function for traversing a directory tree.
4355 * The GNU obstack package is now also part of the C library.
4356 The new function `open_obstack_stream' creates a stdio stream that
4357 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4358 formatted output directly to an obstack.
4360 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4361 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4363 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4365 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4366 things to your strings.
4368 * There are some new test programs: test-fseek, testmb, and testrand.
4370 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4371 These ports are not finished, but are a good starting place for really
4372 supporting those systems.
4374 * `/etc/localtime' is now the standard location for the file that says what
4375 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4376 This follows the general principle that `/etc' is the place for all local
4377 configuration files.
4379 * The C library header files now use `extern "C"' when used by the C++
4380 compiler, so the C library should now work with C++ code.
4382 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4383 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4386 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4387 function which uses the `merge sort' algorithm, and is said to be
4388 significantly faster than the old GNU `qsort' function. Merge sort is
4389 now the standard `qsort' function. The new algorithm can require a lot
4390 of temporary storage; so, the old sorting function is called when the
4391 required storage is not available.
4393 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4394 provides the Unix `crypt' function, plus some other entry points.
4396 * The code and header files taken from 4.4 BSD have been updated with the
4397 latest files released from Berkeley.
4399 ----------------------------------------------------------------------
4400 Copyright information:
4402 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4404 Permission is granted to anyone to make or distribute verbatim copies
4405 of this document as received, in any medium, provided that the
4406 copyright notice and this permission notice are preserved,
4407 thus giving the recipient permission to redistribute in turn.
4409 Permission is granted to distribute modified versions
4410 of this document, or of portions of it,
4411 under the above conditions, provided also that they
4412 carry prominent notices stating who last changed them.
4415 version-control: never