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 [Add new features here]
14 Deprecated and removed features, and other changes affecting compatibility:
16 * The stdio.h functions 'getc' and 'putc' are no longer defined as macros.
17 This was never required by the C standard, and the macros just expanded
18 to call alternative names for the same functions. If you hoped getc and
19 putc would provide performance improvements over fgetc and fputc, instead
20 investigate using (f)getc_unlocked and (f)putc_unlocked, and, if
21 necessary, flockfile and funlockfile.
23 Changes to build and runtime requirements:
25 [Add changes to build and runtime requirements here]
27 Security related changes:
29 [Add security related changes here]
31 The following bugs are resolved with this release:
33 [The release manager will add the list generated by
34 scripts/list-fixed-bugs.py just before the release.]
41 * The GNU C Library can now be compiled with support for building static
42 PIE executables (See --enable-static-pie in INSTALL). These static PIE
43 executables are like static executables but can be loaded at any address
44 and provide additional security hardening benefits at the cost of some
45 memory and performance. When the library is built with --enable-static-pie
46 the resulting libc.a is usable with GCC 8 and above to create static PIE
47 executables using the GCC option '-static-pie'. This feature is currently
48 supported on i386, x86_64 and x32 with binutils 2.29 or later, and on
49 aarch64 with binutils 2.30 or later.
51 * Optimized x86-64 asin, atan2, exp, expf, log, pow, atan, sin, cosf,
52 sinf, sincosf and tan with FMA, contributed by Arjan van de Ven and
55 * Optimized x86-64 trunc and truncf for processors with SSE4.1.
57 * Optimized generic expf, exp2f, logf, log2f, powf, sinf, cosf and sincosf.
59 * In order to support faster and safer process termination the malloc API
60 family of functions will no longer print a failure address and stack
61 backtrace after detecting heap corruption. The goal is to minimize the
62 amount of work done after corruption is detected and to avoid potential
63 security issues in continued process execution. Reducing shutdown time
64 leads to lower overall process restart latency, so there is benefit both
65 from a security and performance perspective.
67 * The abort function terminates the process immediately, without flushing
68 stdio streams. Previous glibc versions used to flush streams, resulting
69 in deadlocks and further data corruption. This change also affects
70 process aborts as the result of assertion failures.
72 * On platforms where long double has the IEEE binary128 format (aarch64,
73 alpha, mips64, riscv, s390 and sparc), the math library now implements
74 _Float128 interfaces for that type, as defined by ISO/IEC TS 18661-3:2015.
75 These are the same interfaces added in version 2.26 for some platforms where
76 this format is supported but is not the format of long double.
78 * On platforms with support for _Float64x (aarch64, alpha, i386, ia64,
79 mips64, powerpc64le, riscv, s390, sparc and x86_64), the math library now
80 implements interfaces for that type, as defined by ISO/IEC TS
81 18661-3:2015. These are corresponding interfaces to those supported for
84 * The math library now implements interfaces for the _Float32, _Float64 and
85 _Float32x types, as defined by ISO/IEC TS 18661-3:2015. These are
86 corresponding interfaces to those supported for _Float128.
88 * glibc now implements the memfd_create and mlock2 functions on Linux.
90 * Support for memory protection keys was added. The <sys/mman.h> header now
91 declares the functions pkey_alloc, pkey_free, pkey_mprotect, pkey_set,
94 * The copy_file_range function was added.
96 * Optimized memcpy, mempcpy, memmove, and memset for sparc M7.
98 * The ldconfig utility now processes `include' directives using the C/POSIX
99 collation ordering. Previous glibc versions used locale-specific
100 ordering, the change might break systems that relied on that.
102 * Support for two grammatical forms of month names has been added.
103 In a call to strftime, the "%B" and "%b" format specifiers will now
104 produce the grammatical form required when the month is used as part
105 of a complete date. New "%OB" and "%Ob" specifiers produce the form
106 required when the month is named by itself. For instance, in Greek
107 and in many Slavic and Baltic languages, "%B" will produce the month
108 in genitive case, and "%OB" will produce the month in nominative case.
110 In a call to strptime, "%B", "%b", "%h", "%OB", "%Ob", and "%Oh"
111 are all valid and will all accept any known form of month
112 name---standalone or complete, abbreviated or full. In a call to
113 nl_langinfo, the query constants MON_1..12 and ABMON_1..12 return
114 the strings used by "%B" and "%b", respectively. New query
115 constants ALTMON_1..12 and _NL_ABALTMON_1..12 return the strings
116 used by "%OB" and "%Ob", respectively.
118 In a locale definition file, use "alt_mon" and "ab_alt_mon" to
119 define the strings for %OB and %Ob, respectively; these have the
120 same syntax as "mon" and "abmon". These arrays are optional; if they
121 are not provided then they have the same content as "mon" and "abmon",
124 These features are provided for locales which define "alt_mon" and/or
125 "ab_alt_mon" in their locale source data. This release includes such
126 alternative month name data for the following languages: Belarusian,
127 Croatian, Greek, Lithuanian, Polish, Russian, and Ukrainian.
129 This feature is currently a GNU extension, but it is expected to
130 be added to the next revision of POSIX, and it is also already
131 available on some BSD-derived operating systems.
133 This feature will cause existing statically compiled applications
134 to fail to load locales and fall back to the builtin C/POSIX locales.
135 See notes below for other changes affecting compatibility.
137 * Support for the RISC-V ISA running on Linux has been added. This port
138 requires at least binutils-2.30, gcc-7.3.0, and linux-4.15; and is supported
139 for the following ISA and ABI pairs:
145 Deprecated and removed features, and other changes affecting compatibility:
147 * Statically compiled applications attempting to load locales compiled for the
148 GNU C Library version 2.27 will fail and fall back to the builtin C/POSIX
149 locale. The reason for this is that the addition of the new "%OB" and "%Ob",
150 support for two grammatical forms of the month names, also extends the locale
151 data binary format. Static applications needing locale support must be
152 recompiled to match the runtime and data they are deployed with. In some
153 distributions there is an upgrade window where dynamically linked applications
154 may use a new library but the old locale data and also fall back to the
155 builtin C/POSIX locales; restarting the application process is sufficient to
158 * Support for statically linked applications which call dlopen is deprecated
159 and will be removed in a future version of glibc. Applications which call
160 dlopen need to be linked dynamically instead.
162 * Support for old programs which use internal stdio data structures and
163 functions is deprecated. This includes programs which use the C++ streams
164 provided by libstdc++ in GCC 2.95. Programs which use the internal
165 symbols _IO_adjust_wcolumn, _IO_default_doallocate, _IO_default_finish,
166 _IO_default_pbackfail, _IO_default_uflow, _IO_default_xsgetn,
167 _IO_default_xsputn, _IO_doallocbuf, _IO_do_write, _IO_file_attach,
168 _IO_file_close, _IO_file_close_it, _IO_file_doallocate, _IO_file_fopen,
169 _IO_file_init, _IO_file_jumps, _IO_fileno, _IO_file_open,
170 _IO_file_overflow, _IO_file_read, _IO_file_seek, _IO_file_seekoff,
171 _IO_file_setbuf, _IO_file_stat, _IO_file_sync, _IO_file_underflow,
172 _IO_file_write, _IO_file_xsputn, _IO_flockfile, _IO_flush_all,
173 _IO_flush_all_linebuffered, _IO_free_backup_area, _IO_free_wbackup_area,
174 _IO_init, _IO_init_marker, _IO_init_wmarker, _IO_iter_begin, _IO_iter_end,
175 _IO_iter_file, _IO_iter_next, _IO_least_wmarker, _IO_link_in,
176 _IO_list_all, _IO_list_lock, _IO_list_resetlock, _IO_list_unlock,
177 _IO_marker_delta, _IO_marker_difference, _IO_remove_marker, _IO_seekmark,
178 _IO_seekwmark, _IO_str_init_readonly, _IO_str_init_static,
179 _IO_str_overflow, _IO_str_pbackfail, _IO_str_seekoff, _IO_str_underflow,
180 _IO_switch_to_main_wget_area, _IO_switch_to_wget_mode,
181 _IO_unsave_wmarkers, _IO_wdefault_doallocate, _IO_wdefault_finish,
182 _IO_wdefault_pbackfail, _IO_wdefault_setbuf, _IO_wdefault_uflow,
183 _IO_wdefault_xsgetn, _IO_wdefault_xsputn, _IO_wdoallocbuf, _IO_wdo_write,
184 _IO_wfile_jumps, _IO_wfile_overflow, _IO_wfile_sync, _IO_wfile_underflow,
185 _IO_wfile_xsputn, _IO_wmarker_delta, or _IO_wsetb may stop working with a
186 future version of glibc. Unlike other symbol removals, these old
187 applications will not be supported using compatibility symbols.
189 * On GNU/Linux, the obsolete Linux constant PTRACE_SEIZE_DEVEL is no longer
190 defined by <sys/ptrace.h>.
192 * libm no longer supports SVID error handling (calling a user-provided
193 matherr function on error) or the _LIB_VERSION variable to control error
194 handling. (SVID error handling and the _LIB_VERSION variable still work
195 for binaries linked against older versions of the GNU C Library.) The
196 libieee.a library is no longer provided. math.h no longer defines struct
197 exception, or the macros X_TLOSS, DOMAIN, SING, OVERFLOW, UNDERFLOW,
198 TLOSS, PLOSS and HUGE.
200 * The libm functions pow10, pow10f and pow10l are no longer supported for
201 new programs. Programs should use the standard names exp10, exp10f and
202 exp10l for these functions instead.
204 * The mcontext_t type is no longer the same as struct sigcontext. On
205 platforms where it was previously the same, this changes the C++ name
206 mangling for interfaces involving this type.
208 * The add-ons mechanism for building additional packages at the same time as
209 glibc has been removed. The --enable-add-ons configure option is now
212 * The --without-fp configure option is now ignored. Whether hardware
213 floating-point instructions are used is now configured based on whether
214 the compiler used at configure time (without any options implied by a
215 --with-cpu= configure option) uses such instructions.
217 * The res_hnok, res_dnok, res_mailok and res_ownok functions now check that
218 the specified string can be parsed as a domain name.
220 * In the malloc_info output, the <heap> element may contain another <aspace>
221 element, "subheaps", which contains the number of sub-heaps.
223 * The libresolv function p_secstodate is no longer supported for new
226 * The tilepro-*-linux-gnu configuration is no longer supported.
228 * The nonstandard header files <libio.h> and <_G_config.h> are deprecated
229 and will be removed in a future release. Software that is still using
230 either header should be updated to use standard <stdio.h> interfaces
233 libio.h was originally the header for a set of supported GNU extensions,
234 but they have not been maintained as such in many years, they are now
235 standing in the way of improvements to stdio, and we don't think there are
236 any remaining external users. _G_config.h was never intended for public
237 use, but predates the bits convention.
239 Changes to build and runtime requirements:
241 * bison version 2.7 or later is required to generate code in the 'intl'
244 Security related changes:
246 CVE-2009-5064: The ldd script would sometimes run the program under
247 examination directly, without preventing code execution through the
248 dynamic linker. (The glibc project disputes that this is a security
249 vulnerability; only trusted binaries must be examined using the ldd
252 CVE-2017-15670: The glob function, when invoked with GLOB_TILDE,
253 suffered from a one-byte overflow during ~ operator processing (either
254 on the stack or the heap, depending on the length of the user name).
255 Reported by Tim Rühsen.
257 CVE-2017-15671: The glob function, when invoked with GLOB_TILDE,
258 would sometimes fail to free memory allocated during ~ operator
259 processing, leading to a memory leak and, potentially, to a denial
262 CVE-2017-15804: The glob function, when invoked with GLOB_TILDE and
263 without GLOB_NOESCAPE, could write past the end of a buffer while
264 unescaping user names. Reported by Tim Rühsen.
266 CVE-2017-17426: The malloc function, when called with an object size near
267 the value SIZE_MAX, would return a pointer to a buffer which is too small,
268 instead of NULL. This was a regression introduced with the new malloc
269 thread cache in glibc 2.26. Reported by Iain Buclaw.
271 CVE-2017-1000408: Incorrect array size computation in _dl_init_paths leads
272 to the allocation of too much memory. (This is not a security bug per se,
273 it is mentioned here only because of the CVE assignment.) Reported by
276 CVE-2017-1000409: Buffer overflow in _dl_init_paths due to miscomputation
277 of the number of search path components. (This is not a security
278 vulnerability per se because no trust boundary is crossed if the fix for
279 CVE-2017-1000366 has been applied, but it is mentioned here only because
280 of the CVE assignment.) Reported by Qualys.
282 CVE-2017-16997: Incorrect handling of RPATH or RUNPATH containing $ORIGIN
283 for AT_SECURE or SUID binaries could be used to load libraries from the
286 CVE-2018-1000001: Buffer underflow in realpath function when getcwd function
287 succeeds without returning an absolute path due to unexpected behaviour
288 of the Linux kernel getcwd syscall. Reported by halfdog.
290 CVE-2018-6485: The posix_memalign and memalign functions, when called with
291 an object size near the value of SIZE_MAX, would return a pointer to a
292 buffer which is too small, instead of NULL. Reported by Jakub Wilk.
294 The following bugs are resolved with this release:
296 [866] glob: glob should match dangling symlinks
297 [1062] glob: sysdeps/generic/glob.c merge from gnulib (part 3 of 3)
298 [2522] localedata: ca_ES@valencia: new Valencian (meridional Catalan)
300 [5997] math: Very slow execution of sinf function
301 [10580] localedata: hr_HR: updated locale
302 [10871] locale: 'mon' array should contain both nominative and genitive
304 [12349] localedata: eu_ES: incorrect thousands separator
305 [13605] localedata: shn_MM: new Shan locale
306 [13805] localedata: ru_RU: currency should use ',' as radix point
307 [13953] localedata: km_KH: locale update
308 [13994] localedata: mjw_IN: new locale
309 [14121] build: make writes .mo files in po directory
310 [14333] libc: Fix the race between atexit() and exit()
311 [14681] dynamic-link: _dl_get_origin leaks memory via executable link map.
312 [14925] localedata: bn_*: LC_IDENTIFICATION.language key should be
314 [15260] localedata: LC_MESSAGES.{yes,no}{str,expr}: various errors
315 [15261] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of full-
316 width Latin characters
317 [15332] localedata: es_CU: locale update
318 [15436] stdio: Don't close or flush stdio streams on abort
319 [15537] localedata: lv_LV: invalid collation for Latvian diacritical
321 [16148] localedata: ca_ES: incorrect thousands separator
322 [16750] dynamic-link: ldd should not try to execute the binaries
324 [16777] localedata: pl_PL: incorrect thousands separator in locale
325 [16905] localedata: hanzi: new collation
326 [17563] localedata: cmn_TW: add hanzi collation
327 [17750] localedata: wrong collation order of diacritics in most locales
328 [17804] libc: scandirat fails with ENOMEM because it checks for errno even
330 [17956] build: Build fails on missing definitions from header file
331 nss/nss.h when Mozilla NSS is used for cryptography
332 [18203] libc: realpath() does not handle unreachable paths correctly
333 [18572] dynamic-link: [arm] Lazy TLSDESC relocation has data race
334 [18812] localedata: kab_DZ: new Kabyle Algeria locale
335 [18822] libc: Internal functions are called via PLT
336 [18858] string: _HAVE_STRING_ARCH_xxx aren't defined for i386 nor x86_64
337 [19170] libc: __gmon_start__ defined in hppa in crtn.S
338 [19574] libc: glibc should support building static PIE binaries
339 [19852] localedata: charmaps/UTF-8: incorrect wcwidth for U+3099 and
341 [19971] glob: glob: Do not skip entries with zero d_ino values
342 [19982] localedata: fr.po: spelling mistake for error code EXDEV
343 [20008] localedata: km_KH: convert to translit_neutral
344 [20009] localedata: tr_TR: convert LC_CTYPE to i18n
345 [20142] math: [x86_64] Add SSE4.1 trunc, truncf
346 [20204] dynamic-link: _dl_open_hook and _dlfcn_hook hardening
347 [20482] localedata: de_CH: abbreviated weekdays should be two letters
348 [20498] localedata: miq_NI: new Mískitu / Miskito (miq) language locale
350 [20532] nss: getaddrinfo uses errno and h_errno without guaranteeing
351 they're set, wrong errors returned by gaih_inet when lookup functions
353 [20756] localedata: [PATCH] Use Unicode wise thousands separator
354 [20826] network: posix/tst-getaddrinfo5 fails on hosts without network
356 [20952] localedata: yuw_PG: new locale
357 [21084] localedata: charmaps/IBM858: new codepage
358 [21161] manual: [PATCH] fix typo in manual/arith.texi on strtoul prototype
359 [21242] libc: assert gives pedantic warning in old gcc versions
360 [21265] dynamic-link: _dl_runtime_resolve isn't compatible with Intel C++
361 __regcall calling convention
362 [21309] math: signed integer overflow in sysdeps/ieee754/dbl-64/e_pow.c
363 [21326] libc: C99 functions are not declared for C++11 and later if
364 _GNU_SOURCE is not predefined by g++
365 [21457] libc: sys/ucontext.h namespace
366 [21530] libc: tmpfile() should be implemented using O_TMPFILE
367 [21660] math: GCC fails to compile a formula with tgmath.h
368 [21672] nptl: sys-libs/glibc on ia64 crashes on thread exit: signal
369 SIGSEGV, Segmentation fault: pthread_create.c:432: __madvise
370 (pd->stackblock, freesize - PTHREAD_STACK_MIN, MADV_DONTNEED);
371 [21684] math: tgmath.h handling of complex integers
372 [21685] math: tgmath.h handling of bit-fields
373 [21686] math: tgmath.h handling of __int128
374 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
376 [21745] libc: [powerpc64le] Extra PLT reference with --enable-stack-
378 [21750] localedata: column width of characters incompatible with classical
380 [21754] malloc: malloc: Perform as little work as possible after heap
381 consistency check failures
382 [21780] libc: hppa: p{read,write}v2 does not set ENOSUP on invalid flag
383 [21790] libc: Missing __memset_zero_constant_len_parameter in libc.so
384 [21791] string: Unused XXX_chk_XXX functions in libc.a
385 [21815] dynamic-link: FAIL: elf/tst-prelink-cmp with GCC is defaulted to
387 [21836] localedata: Removed redundant data (LC_MONETARY) in various Indian
389 [21845] localedata: Added new Locale bho_NP
390 [21853] localedata: Fix abday Which looks same as day in zh_SG
391 [21854] localedata: Added New Locale en_SC
392 [21864] libc: xmalloc.o is compiled with -DMODULE_NAME=libc
393 [21871] dynamic-link: _dl_runtime_resolve_avx_opt is slower than
394 _dl_runtime_resolve_avx_slow
395 [21885] network: getaddrinfo: gethosts does not release resolver context
396 on memory allocation failure
397 [21899] libc: XPG4.2 sigaction namespace
398 [21908] dynamic-link: dynamic linker broke on ia64 (mmap2 consolidation is
400 [21913] libc: static binaries SIGSEGV in __brk when host's gcc is pie-by-
402 [21915] nss: nss_files can return with NSS_STATUS_SUCCESS and a clobbered
403 errno value, causing getaddrinfo to fail
404 [21920] localedata: Fix p_cs_precedes/n_cs_precedes for mt_MT
405 [21922] network: getaddrinfo with AF_INET/AF_INET6 returns EAI_NONAME
406 instead of EAI_NODATA
407 [21928] libc: sys/ptrace.h: remove obsolete temporary development Linux
408 constant PTRACE_SEIZE_DEVEL
409 [21930] math: C-only gcc builtins used in <math.h> isinf
410 [21932] network: Unpaired __resolv_context_get in generic get*_r
412 [21941] math: powerpc: Wrong register constraint for xssqrtqp in sqrtf128
413 [21944] libc: sigval namespace
414 [21951] localedata: Update hanzi collation by stroke
415 [21955] math: Wrong alignment of L(SP_RANGE)/L(SP_INF_0) in
416 sysdeps/x86_64/fpu/e_expf.S
417 [21956] libc: Stack allocation in MIPS syscall impl (ubounded stack
418 allocation in syscall loops)
419 [21959] localedata: Fix Country name for xh_ZA
420 [21960] localedata: Fix abmon for bem_ZM
421 [21966] math: AVX2 mathvec functions use FMA without checking
422 [21967] math: When 512-bit AVX2 wrapper functions in mathvec are used?
423 [21971] localedata: Added New Locale for mfe_MU
424 [21972] libc: assert macro requires operator== (int) for its argument type
425 [21973] math: [sparc] libm missing sqrtl compat symbol
426 [21974] libc: Remove __bb_init_func and __bb_exit_func
427 [21982] string: stratcliff.c: error: assuming signed overflow does not
429 [21986] stdio: __guess_grouping is called incorrectly
430 [21987] math: [sparc32] wrong bits/long-double.h installed
431 [22019] localedata: Wrong placement of monetary symbol in el_GR (negative
433 [22022] localedata: Missing country_name for mni_IN
434 [22023] localedata: Removed redundant data (LC_TIME and LC_MESSAGES) for
436 [22025] locale: iconv: Inconsistency between pointer mangling and NULL
438 [22026] locale: iconv_open: heap overflow on gconv_init failure
439 [22028] math: bits/math-finite.h _MSUF_ expansion namespace
440 [22035] math: [m68k] bits/math-inline.h macro namespace
441 [22038] localedata: Fix abbreviated weeks and months for Somali
442 [22044] localedata: Remove redundant data for Limburgish Language
443 [22050] malloc: Linking with -lmcheck does not hook
444 __malloc_initialize_hook correctly
445 [22051] libc: zero terminator in the middle of glibc's .eh_frame
446 [22052] malloc: malloc failed to compile with GCC 7 and -O3
447 [22070] localedata: charmaps/UTF-8: wcwidth for
448 Prepended_Concatenation_Mark codepoints set to 0 (should be 1)
449 [22074] localedata: charmaps/UTF-8: wcwidth for U+1160-U+11FF (Hangul
450 Jungseong and Jongseong) should be 0
451 [22078] nss: nss_files performance issue in multi mode
452 [22082] math: bits/math-finite.h exp10 condition
453 [22086] libc: pcprofiledump incorrect cross-endian condition
454 [22093] dynamic-link: ld.so no longer searches in .../x86_64
455 [22095] network: Name server address allocation memory leak in resolv.conf
457 [22096] network: __resolv_conf_attach can incorrectly free passed conf
459 [22100] localedata: om_KE: LC_TIME: copy redundant data from om_ET
460 [22101] dynamic-link: Dynamic loader must ignore "debug" shared objects
462 [22111] malloc: malloc: per thread cache is not returned when thread exits
463 [22112] localedata: Fix LC_TELEPHONE/LC_NAME for az_AZ
464 [22134] libc: [linux] implement fexecve with execveat
465 [22142] libc: [powerpc] printf oupts a wrong value of DBL_MAX on ppc64 and
467 [22145] libc: ttyname() gives up too early in the face of namespaces
468 [22146] math: C++ build issue with float128 on x86_64
469 [22153] nptl: nptl: save error code before process termination
470 [22156] libc: [hppa,ia64,microblaze] Executable stack default
471 [22159] malloc: malloc: MALLOC_CHECK_ broken with --enable-tunables=no
472 [22161] nscd: nscd cache prune for netgroups hangs after timeout bump
473 [22165] libc: [hppa] Text relocations in libc.so
474 [22180] libc: destructor registered via __cxa_atexit is called twice
475 [22183] glob: commit 5554304f0ddd ("posix: Allow glob to match dangling
476 symlinks") cause "make" segfaults
477 [22189] math: [powerpc] math_private.h definitions of math_opt_barrier and
479 [22207] libc: FAIL: stdlib/test-atexit-race
480 [22225] math: nearbyint arithmetic moved before feholdexcept
481 [22229] math: [sparc32] missing copysignl, fabsl, fmal compat symbols
482 [22235] math: iscanonical in C++ and float128
483 [22243] math: log2(0) and log10(0) are wrong in downward rounding without
484 the svid compat wrapper
485 [22244] math: ynf and yn are wrong without the svid compat wrapper
486 [22273] libc: Improper assert in Linux posix_spawn implementation
487 [22284] libc: -pg -pie doesn't work
488 [22292] locale: localedef exits with error 4 when it should be error 1
489 [22294] locale: Allow "" for int_currency_symbol definition in locales.
490 [22295] locale: Don't warn on non-symbolic characters in locale sources in
492 [22296] math: glibc 2.26: signbit build issue with Gcc 5.5.0 on x86_64
493 [22298] nptl: x32: lockups on recursive pthread_mutex_lock after upgrade
495 [22299] dynamic-link: Problem with $PLATFORM on x86_64 platform
496 [22320] glob: Buffer overflow in glob with GLOB_TILDE (CVE-2017-15670)
497 [22321] libc: sysconf(_SC_IOV_MAX) returns -1 on Linux
498 [22322] libc: [mips64] wrong bits/long-double.h installed
499 [22325] glob: Memory leak in glob with GLOB_TILDE (CVE-2017-15671)
500 [22332] glob: Buffer overflow in glob with GLOB_TILDE in unescaping
502 [22336] localedata: cs_CZ LC_COLLATE does not use i18n
503 [22343] malloc: Integer overflow in posix_memalign
504 [22347] libc: getrandom() returns the number of bytes that were copied to
505 the buffer even though the comments say "Return 0 on success and -1 on
507 [22353] string: sysdeps/i386/i586/strcpy.S isn't maintainable
508 [22362] libc: Installed crt1.o, crti,.o and crtn.o files are used with
510 [22370] dynamic-link: Incorrect note padding check
511 [22375] libc: malloc returns pointer from tcache_get when should return
512 NULL (CVE-2017-17426)
513 [22377] math: iseqsig, float128 and C++
514 [22382] localedata: Error in tpi_PG locale
515 [22387] localedata: Replace unicode sequences <Uxxxx> for characters
516 inside the ASCII printable range
517 [22402] math: [powerpc64le] __MATH_TG does not support _Float128 for
519 [22403] localedata: Slash needs escaping in some locales
520 [22408] malloc: malloc_info access heaps without arena lock, ignores heaps
521 [22409] network: res_hnok does not accept some host names used on the
523 [22412] network: res_dnok, res_hnok should perform syntax checks
524 [22413] network: ns_name_pton ignores syntactically invalid trailing
526 [22415] stdio: setvbuf can lead to invalid free/segfault
527 [22432] build: Non-deterministic build
528 [22439] malloc: malloc_info should compute summary statistics for all sub-
530 [22442] network: if_nametoindex could report index for the wrong
532 [22446] build: aliasing violation calling readlink in handle_request
533 [22447] build: unsafe call to strlen with a non-string in getlogin_r.c
534 [22457] libc: Generic preadv/pwritev incorrectly calls __posix_memalign
535 [22459] libc: FAIL: elf/check-localplt with __stack_chk_fail related to
536 __nscd_hash/__nss_hash
537 [22463] network: p_secstodate overflow handling
538 [22469] localedata: pl_PL LC_COLLATE does not use i18n
539 [22478] libc: sigwait can fail with EINTR
540 [22505] libc: ldconfig processes include directive in locale-specific
542 [22515] localedata: hsb_DE LC_COLLATE does not use copy "iso14651_t1"
543 [22517] localedata: et_EE LC_COLLATE does not use copy "iso14651_t1"
544 [22519] localedata: is_IS LC_COLLATE does not use copy "iso14651_t1"
545 [22524] localedata: lt_LT LC_COLLATE does not use copy "iso14651_t1"
546 [22527] localedata: tr_TR LC_COLLATE does not use copy "iso14651_t1"
547 [22534] localedata: Collation rules for Serbian and Bosnian should be the
549 [22561] math: [DR#471] cacosh (0 + iNaN) should return NaN +/- i pi/2
550 [22568] math: [DR#471] ctanh (0 + iNaN), ctanh (0 + i Inf)
551 [22577] libc: missing newline after "cannot allocate TLS data structures
553 [22588] manual: manual/conf.texi: missing underscore in front of
555 [22593] math: nextafter and nexttoward are declared with const attribute
556 [22596] manual: manual: finite(nan) wrongly described as returning nonzero
557 [22603] string: ia64 memchr overflows internal pointer check
558 [22605] libc: SH clone does not set the exit code correctly
559 [22606] dynamic-link: Incorrect array size computation in _dl_init_paths
561 [22607] dynamic-link: Buffer Overflow in _dl_init_paths (CVE-2017-1000409)
562 [22611] malloc: malloc/tst-realloc wrongly assumes that errno must not be
563 modified in case of success
564 [22614] build: gcc: error: unrecognized command line option ‘-no-pie’
565 [22615] manual: manual: ambiguous wording about errno value in case of
567 [22624] libc: MIPS setjmp() saves incorrect 'o0' register in --enable-
569 [22625] dynamic-link: RPATH $ORIGIN replaced by PWD for AT_SECURE/SUID
570 binaries or if /proc is not mounted (CVE-2017-16997)
571 [22627] dynamic-link: $ORIGIN in $LD_LIBRARY_PATH is substituted twice
572 [22630] build: $(no-pie-ldflag) is no longer effective
573 [22631] math: [m68k] Bad const attributes in bits/mathinline.h
574 [22635] nptl: pthread_self returns NULL before libpthread is loaded
575 [22636] nptl: PTHREAD_STACK_MIN is too small on x86-64
576 [22637] nptl: guard size is subtracted from thread stack size instead of
578 [22648] libc: getrlimit/setrlimit with RLIM_INFINITY broken on alpha
579 [22657] localedata: hu_HU: Avoid double space in date
580 [22660] math: fmax, fmin sNaN handling on alpha
581 [22664] libc: New warning of GCC8
582 [22665] math: alpha: ceil and floor raise inexact exceptions
583 [22666] math: alpha: trunc raise inexact exceptions
584 [22667] libc: makecontext lacks stack alignment on i386
585 [22678] libc: prlimit fails for RLIM_INFINITY values on 32-bit machines
586 [22679] libc: getcwd(3) can succeed without returning an absolute path
588 [22685] libc: PowerPC: Static AT_SECURE binaries segfault with lock-
590 [22687] math: [powerpc-nofpu] complex long double functions spurious
592 [22688] math: [powerpc-nofpu] remainderl wrong sign of zero result
593 [22690] math: [ldbl-128ibm] lrintl, lroundl missing "invalid" exceptions
594 [22691] math: [powerpc-nofpu] fmaxmagl, fminmagl spurious "invalid"
596 [22693] math: [ldbl-128ibm] log1pl (-qNaN) spurious "invalid" exception
597 [22697] math: [powerpc] llround spurious "inexact" exceptions on 32-bit
599 [22701] nis: Incomplete removal of libnsl
600 [22702] math: [powerpc-nofpu] nearbyintl traps with trapping "inexact"
601 [22707] libc: Missing defines in elf.h for DF_1_STUB and DF_1_PIE.
602 [22715] dynamic-link: FAIL: elf/tst-audit10
603 [22719] libc: Backtrace tests fail on hppa
604 [22742] libc: [aarch64] mcontext_t __reserved field got renamed
605 [22743] nptl: __pthread_register_cancel corrupts stack after f81ddabffd
606 [22765] crypt: (struct crypt_data *data)->initialized is not set to zero
607 before the first call to crypt_r () in crypt/badsalttest.c
614 * A per-thread cache has been added to malloc. Access to the cache requires
615 no locks and therefore significantly accelerates the fast path to allocate
616 and free small amounts of memory. Refilling an empty cache requires locking
617 the underlying arena. Performance measurements show significant gains in a
618 wide variety of user workloads. Workloads were captured using a special
619 instrumented malloc and analyzed with a malloc simulator. Contributed by
620 DJ Delorie with the help of Florian Weimer, and Carlos O'Donell.
622 * Unicode 10.0.0 Support: Character encoding, character type info, and
623 transliteration tables are all updated to Unicode 10.0.0, using
624 generator scripts contributed by Mike FABIAN (Red Hat).
625 These updates cause user visible changes, especially the changes in
626 wcwidth for many emoji characters cause problems when emoji sequences
627 are rendered with pango, see for example:
628 https://bugzilla.gnome.org/show_bug.cgi?id=780669#c5
630 * Collation of Hungarian has been overhauled and is now consistent with "The
631 Rules of Hungarian Orthography, 12th edition" (Bug 18934). Contributed by
634 * Improvements to the DNS stub resolver, contributed by Florian Weimer:
636 - The GNU C Library will now detect when /etc/resolv.conf has been
637 modified and reload the changed configuration. The new resolver option
638 “no-reload” (RES_NORELOAD) disables this behavior.
640 - The GNU C Library now supports an arbitrary number of search domains
641 (configured using the “search” directive in /etc/resolv.conf);
642 previously, there was a hard limit of six domains. For backward
643 compatibility, applications that directly modify the ‘_res’ global
644 object are still limited to six search domains.
646 - When the “rotate” (RES_ROTATE) resolver option is active, the GNU C
647 Library will now randomly pick a name server from the configuration as a
648 starting point. (Previously, the second name server was always used.)
650 * The tunables feature is now enabled by default. This allows users to tweak
651 behavior of the GNU C Library using the GLIBC_TUNABLES environment variable.
653 * New function reallocarray, which resizes an allocated block (like realloc)
654 to the product of two sizes, with a guaranteed clean failure upon integer
655 overflow in the multiplication. Originally from OpenBSD, contributed by
656 Dennis Wölfing and Rüdiger Sonderfeld.
658 * New wrappers for the Linux-specific system calls preadv2 and pwritev2.
659 These are extended versions of preadv and pwritev, respectively, taking an
660 additional flags argument. The set of supported flags depends on the
661 running kernel; full support currently requires kernel 4.7 or later.
663 * posix_spawnattr_setflags now supports the flag POSIX_SPAWN_SETSID, to
664 create a new session ID for the spawned process. This feature is
665 scheduled to be added to the next major revision of POSIX; for the time
666 being, it is available under _GNU_SOURCE.
668 * errno.h is now safe to use from C-preprocessed assembly language on all
669 supported operating systems. In this context, it will only define the
670 Exxxx constants, as preprocessor macros expanding to integer literals.
672 * On ia64, powerpc64le, x86-32, and x86-64, the math library now implements
673 128-bit floating point as defined by ISO/IEC/IEEE 60559:2011 (IEEE
674 754-2008) and ISO/IEC TS 18661-3:2015. Contributed by Paul E. Murphy,
675 Gabriel F. T. Gomes, Tulio Magno Quites Machado Filho, and Joseph Myers.
677 To compile programs that use this feature, the compiler must support
678 128-bit floating point with the type name _Float128 (as defined by TS
679 18661-3) or __float128 (the nonstandard name used by GCC for C++, and for
680 C prior to version 7). _GNU_SOURCE or __STDC_WANT_IEC_60559_TYPES_EXT__
681 must be defined to make the new interfaces visible.
683 The new functions and macros correspond to those present for other
684 floating-point types (except for a few obsolescent interfaces not
685 supported for the new type), with F128 or f128 suffixes; for example,
686 strtof128, HUGE_VAL_F128 and cosf128. Following TS 18661-3, there are no
687 printf or scanf formats for the new type; the strfromf128 and strtof128
688 interfaces should be used instead.
690 Deprecated and removed features, and other changes affecting compatibility:
692 * The synchronization that pthread_spin_unlock performs has been changed to
693 now be equivalent to a C11 atomic store with release memory order to the
694 spin lock's memory location. Previously, several (but not all)
695 architectures used stronger synchronization (e.g., containing what is
696 often called a full barrier). This change can improve performance, but
697 may affect odd fringe uses of spin locks that depend on the previous
698 behavior (e.g., using spin locks as atomic variables to try to implement
699 Dekker's mutual exclusion algorithm).
701 * The port to Native Client running on ARMv7-A (--host=arm-nacl) has been
704 * Sun RPC is deprecated. The rpcgen program, librpcsvc, and Sun RPC headers
705 will only be built and installed when the GNU C Library is configured with
706 --enable-obsolete-rpc. This allows alternative RPC implementations, such
707 as TIRPC or rpcsvc-proto, to be used.
709 * The NIS(+) name service modules, libnss_nis, libnss_nisplus, and
710 libnss_compat, are deprecated, and will not be built or installed by
713 The NIS(+) support library, libnsl, is also deprecated. By default, a
714 compatibility shared library will be built and installed, but not headers
715 or development libraries. Only a few NIS-related programs require this
716 library. (In particular, the GNU C Library has never required programs
717 that use 'gethostbyname' to be linked with libnsl.)
719 Replacement implementations based on TIRPC, which additionally support
720 IPv6, are available from <https://github.com/thkukuk/>. The configure
721 option --enable-obsolete-nsl will cause libnsl's headers, and the NIS(+)
722 name service modules, to be built and installed.
724 * The DNS stub resolver no longer performs EDNS fallback. If EDNS or DNSSEC
725 support is enabled, the configured recursive resolver must support EDNS.
726 (Responding to EDNS-enabled queries with responses which are not
727 EDNS-enabled is fine, but FORMERR responses are not.)
729 * res_mkquery and res_nmkquery no longer support the IQUERY opcode. DNS
730 servers have not supported this opcode for a long time.
732 * The _res_opcodes variable has been removed from libresolv. It had been
733 exported by accident.
735 * <string.h> no longer includes inline versions of any string functions,
736 as this kind of optimization is better done by the compiler. The macros
737 __USE_STRING_INLINES and __NO_STRING_INLINES no longer have any effect.
739 * The nonstandard header <xlocale.h> has been removed. Most programs should
740 use <locale.h> instead. If you have a specific need for the definition of
741 locale_t with no other declarations, please contact
742 libc-alpha@sourceware.org and explain.
744 * The obsolete header <sys/ultrasound.h> has been removed.
746 * The obsolete signal constant SIGUNUSED is no longer defined by <signal.h>.
748 * The obsolete function cfree has been removed. Applications should use
751 * The stack_t type no longer has the name struct sigaltstack. This changes
752 the C++ name mangling for interfaces involving this type.
754 * The ucontext_t type no longer has the name struct ucontext. This changes
755 the C++ name mangling for interfaces involving this type.
757 * On M68k GNU/Linux and MIPS GNU/Linux, the fpregset_t type no longer has
758 the name struct fpregset. On Nios II GNU/Linux, the mcontext_t type no
759 longer has the name struct mcontext. On SPARC GNU/Linux, the struct
760 mc_fq, struct rwindow, struct fpq and struct fq types are no longer
761 defined in sys/ucontext.h, the mc_fpu_t type no longer has the name struct
762 mc_fpu, the gwindows_t type no longer has the name struct gwindows and the
763 fpregset_t type no longer has the name struct fpu. This changes the C++
764 name mangling for interfaces involving those types.
766 * On S/390 GNU/Linux, the constants defined by <sys/ptrace.h> have been
767 synced with the kernel:
769 - PTRACE_GETREGS, PTRACE_SETREGS, PTRACE_GETFPREGS and PTRACE_SETFPREGS
770 are not supported on this architecture and have been removed.
772 - PTRACE_SINGLEBLOCK, PTRACE_SECCOMP_GET_FILTER, PTRACE_PEEKUSR_AREA,
773 PTRACE_POKEUSR_AREA, PTRACE_GET_LAST_BREAK, PTRACE_ENABLE_TE,
774 PTRACE_DISABLE_TE and PTRACE_TE_ABORT_RAND have been added.
776 Programs that assume the GET/SETREGS ptrace requests are universally
777 available will now fail to build, instead of malfunctioning at runtime.
779 Changes to build and runtime requirements:
781 * Linux kernel 3.2 or later is required at runtime, on all architectures
782 supported by that kernel. (This is a change from version 2.25 only for
785 * GNU Binutils 2.25 or later is now required to build the GNU C Library.
787 * On most architectures, GCC 4.9 or later is required to build the GNU C
788 Library. On powerpc64le, GCC 6.2 or later is required.
790 Older GCC versions and non-GNU compilers are still supported when
791 compiling programs that use the GNU C Library. (We do not know exactly
792 how old, and some GNU extensions to C may be _de facto_ required. If you
793 are interested in helping us make this statement less vague, please
794 contact libc-alpha@sourceware.org.)
796 Security related changes:
798 * The DNS stub resolver limits the advertised UDP buffer size to 1200 bytes,
799 to avoid fragmentation-based spoofing attacks (CVE-2017-12132).
801 * LD_LIBRARY_PATH is now ignored in binaries running in privileged AT_SECURE
802 mode to guard against local privilege escalation attacks (CVE-2017-1000366).
804 * Avoid printing a backtrace from the __stack_chk_fail function since it is
805 called on a corrupt stack and a backtrace is unreliable on a corrupt stack
808 * A use-after-free vulnerability in clntudp_call in the Sun RPC system has been
809 fixed (CVE-2017-12133).
811 The following bugs are resolved with this release:
813 [984] network: Respond to changed resolv.conf in gethostbyname
814 [5010] network: sunrpc service cleanup causes unwanted port mapper traffic
815 [12068] localedata: sc_IT: misspelled yesexpr/day/abday/mon/abmon/date_fmt
817 [12189] libc: __stack_chk_fail should not attempt a backtrace
819 [14096] time: Race condition on timezone/tst-timezone.out
820 [14172] localedata: az_IR: new locale
821 [14995] build: glibc fails to build if gold is the default linker, even if
823 [15998] build: [powerpc] Set arch_minimum_kernel for powerpc LE
824 [16637] network: inet_pton function is accepting IPv6 with bad format
825 [16640] string: string/strtok.c: undefined behaviour inconsistent between
826 x86 and other generic code
827 [16875] localedata: ko_KR: fix lang_name
828 [17225] localedata: ar_SY: localized month names for May and June are
830 [17297] localedata: da_DK: wrong date_fmt string
831 [18907] stdio: Incorrect order of __wur __THROW in <printf.h>
832 [18934] localedata: hu_HU: collate: fix multiple bugs and add tests
833 [18988] nptl: pthread wastes memory with mlockall(MCL_FUTURE)
834 [19066] localedata: ar_SA abbreviated day and month names are in English
835 [19569] network: resolv: Support an arbitrary number of search domains
836 [19570] network: Implement random DNS server selection in the stub
838 [19838] locale: localedef fails on PA-RISC
839 [19919] localedata: iso14651_t1_common: Correct the Malayalam sorting
840 order of 0D36 and 0D37
841 [19922] localedata: iso14651_t1_common: Define collation for Malayalam
843 [20098] libc: FAIL: debug/backtrace-tst on hppa
844 [20257] network: sunrpc: clntudp_call does not enforce timeout when
846 [20275] localedata: locale day/abday/mon/abmon should not have trailing
848 [20313] localedata: Update locale data to Unicode 9.0
849 [20424] manual: Document how to provide a malloc replacement
850 [20496] localedata: agr_PE: new language locale Awajún / Aguaruna (agr)
852 [20686] locale: Add el_GR@euro to SUPPORTED.
853 [20831] dynamic-link: _dl_map_segments does not test for __mprotect
854 failures consistently
855 [21015] dynamic-link: Document and fix --enable-bind-now
856 [21016] nptl: pthread_cond support is broken on hppa
857 [21029] libc: glibc-2.23 (and later) fails to compile with -fno-omit-
858 frame-pointer on i386
859 [21049] libc: segfault in longjmp_chk() due to clobbered processor
861 [21075] libc: unused assigment to %g4 in sparc/sparc{64,32}/clone.S
862 [21088] libc: Build fails with --enable-static-nss
863 [21094] math: cosf(1.57079697) has 3 ulp error on targets where the
864 generic c code is used
865 [21109] libc: Tunables broken on big-endian
866 [21112] math: powf has large ulp errors with base close to 1 and exponent
868 [21115] network: sunrpc: Use-after-free in error path in clntudp_call
870 [21120] malloc: glibc malloc is incompatible with GCC 7
871 [21130] math: Incorrect return from y0l (-inf) and y1l (-inf) when linking
873 [21134] math: Exception (divide by zero) not set for y0/y1 (0.0) and y0/y1
874 (-0.0) when linking with -lieee
875 [21171] math: log10, log2 and lgamma return incorrect results
876 [21179] libc: handle R_SPARC_DISP64 and R_SPARC_REGISTER relocs
877 [21182] libc: __memchr_sse2: regression in glibc-2.25 on i686
878 [21207] localedata: ce_RU: update weekdays from CLDR
879 [21209] dynamic-link: LD_HWCAP_MASK read in setuid binaries
880 [21217] localedata: Update months from CLDR-31
881 [21232] libc: miss posix_fadvise64 on MIPS64 when static linking
882 [21243] libc: support_delete_temp_file should issue warning for failed
884 [21244] libc: support resolv_test_start() socket fd close should be
886 [21253] libc: localedef randomly segfaults when using -fstack-check due to
887 new posix_spawn implementation
888 [21258] dynamic-link: Branch predication in _dl_runtime_resolve_avx512_opt
889 leads to lower CPU frequency
890 [21259] libc: [alpha] termios.h missing IXANY for POSIX
891 [21261] libc: [sparc64] bits/setjmp.h namespace
892 [21267] network: [mips] bits/socket.h IOC* namespace
893 [21268] libc: [alpha] termios.h NL2, NL3 namespace
894 [21270] libc: mmap64 silently truncates large offset values
895 [21275] libc: posix_spawn always crashes on ia64 now
896 [21277] libc: [alpha] termios.h missing IUCLC for UNIX98 and older
897 [21280] math: [powerpc] logbl for POWER7 return incorrect results
898 [21289] libc: Incorrect declaration for 32-bit platforms with
899 _FILE_OFFSET_BITS=64 causes build error
900 [21295] network: GETAI(AF_UNSPEC) drops IPv6 addresses if nss module does
901 not support gethostbyname4_r
902 [21298] nptl: rwlock can deadlock on frequent reader/writer phase
904 [21338] malloc: mallopt M_ARENA_MAX doesn't set the maximum number of
906 [21340] libc: Support POSIX_SPAWN_SETSID
907 [21357] libc: unwind-dw2-fde deadlock when using AddressSanitizer
908 [21359] network: ns_name_pack needs additional byte in destination buffer
909 [21361] network: resolv: Reduce advertised EDNS0 buffer size to guard
910 against fragmentation attacks (CVE-2017-12132)
911 [21369] network: resolv: Remove EDNS fallback
912 [21371] libc: Missing timespec definition when compiled with _XOPEN_SOURCE
914 [21386] nptl: Assertion in fork for distinct parent PID is incorrect
915 [21391] dynamic-link: x86: Set dl_platform and dl_hwcap from CPU features
916 [21393] stdio: Missing dup3 error check in freopen, freopen64
917 [21396] libc: Use AVX2 memcpy/memset on Skylake server
918 [21399] localedata: Bad description for U00EC in
919 localedata/charmaps/CP1254
920 [21411] malloc: realloc documentation error
921 [21426] network: sys/socket.h uio.h namespace
922 [21428] libc: [aarch64] tst-backtrace5 testsuite failure
923 [21445] libc: signal.h bsd_signal namespace
924 [21455] network: Network headers stdint.h namespace
925 [21474] network: resolv: res_init does not use RES_DFLRETRY (2) but 4 for
927 [21475] network: resolv: Overlong search path is truncated mid-label
928 [21511] libc: sigstack namespace
929 [21512] libc: clone() ends up calling exit_group() through _exit() wrapper
930 [21514] libc: sysdeps/unix/sysv/linux/sys/syscall.h:31:27: fatal error:
931 bits/syscall.h: No such file or directory
932 [21517] libc: struct sigaltstack namespace
933 [21528] dynamic-link: Duplicated minimal strtoul implementations in ld.so
934 [21533] localedata: Update locale data to Unicode 10.0
936 ../sysdeps/unix/sysv/linux/s390/s390-32/__makecontext_ret.S:44: Error:
937 junk at end of line, first unrecognized character is `@'
938 [21538] libc: SIG_HOLD missing for XPG4
939 [21539] libc: S390: Mismatch between kernel and glibc ptrace.h with
940 request 12: PTRACE_SINGLEBLOCK vs PTRACE_GETREGS.
941 [21542] libc: Use conservative default for sysconf (_SC_NPROCESSORS_ONLN)
942 [21543] libc: sigevent namespace
943 [21548] libc: [mips] get/set/make/swap context for MIPS O32 assume wrong
944 size for general purpose registers in mcontext_t structure
945 [21550] libc: sigwait namespace
946 [21552] libc: XPG4 bsd_signal namespace
947 [21554] libc: sigpause namespace
948 [21560] libc: sys/wait.h signal.h namespace
949 [21561] libc: waitid namespace
950 [21573] nptl: GCC 7: /usr/bin/install: cannot remove
951 '/usr/include/stdlib.h': Permission denied
952 [21575] libc: sys/wait.h missing struct rusage definition
953 [21584] libc: sigaltstack etc namespace
954 [21597] libc: siginterrupt namespace
955 [21607] math: hppa: FAIL: math/test-tgmath
956 [21609] dynamic-link: Incomplete workaround for GCC __tls_get_addr ABI
958 [21622] libc: [tile] missing SA_* for POSIX.1:2008
959 [21624] dynamic-link: ld.so: Unsafe alloca allows local attackers to alias
960 stack and heap (CVE-2017-1000366)
961 [21625] libc: wait3 namespace
962 [21654] nss: Incorrect pointer alignment in NSS group merge result
964 [21657] network: Parse interface zone id for node-local multicast
965 [21662] string: memcmp-avx2-movbe.S lacks saturating subtraction for
967 [21666] libc: .symver is used on common symbol
968 [21668] network: resolv: res_init cross-thread broadcast introduces race
970 [21687] math: tgmath.h totalorder, totalordermag return type
971 [21694] locale: Current Glibc Locale Does Not Support Tok-Pisin and Fiji
973 [21696] libc: Incorrect assumption of of __cpu_mask in
974 posix/sched_cpucount.c
975 [21697] libc: sysdeps/posix/spawni.c: 2 * suspicious condition ?
976 [21706] localedata: yesstr and nostr are missing for Breton [LC_MESSAGES]
978 [21707] math: ppc64le: Invalid IFUNC resolver from libgcc calls getauxval,
979 leading to relocation crash
980 [21709] libc: resolv_conf.c:552: update_from_conf: Assertion
981 `resolv_conf_matches (resp, conf)' failed.
982 [21710] localedata: Added Samoan language locale for Samoa
983 [21711] localedata: Pashto yesstr/nostr locale are missing
984 [21715] nptl: sysdeps/nptl/bits/pthreadtypes.h: typedef guard
985 __have_pthread_attr_t can cause redefinition of typedef ‘pthread_attr_t’
986 [21721] localedata: Incorrect Full Weekday names for ks_IN@devanagari
987 [21723] localedata: yesstr/nostr missing for Chinese language locale
988 [21724] localedata: yesstr and nostr are missing for Xhosa [LC_MESSAGES]
990 [21727] localedata: yesstr and nostr are missing for Tsonga [LC_MESSAGES]
992 [21728] localedata: New Locale for Tongan language
993 [21729] localedata: incorrect LC_NAME fields for hi_IN
994 [21733] localedata: yesstr and nostr are missing for zh_HK
995 [21734] localedata: Missing yesstr and nostr are for kw_GB
996 [21738] libc: misc/tst-preadvwritev2 and misc/tst-preadvwritev64v2 fail
997 [21741] libc: Undefined __memmove_chk_XXX and __memset_chk_XXX in libc.a
998 [21742] libc: _dl_num_cache_relocations is undefined in libc.a
999 [21743] localedata: ks_IN@devanagari: abday strings mismatch the day
1001 [21744] libc: Tests failing on --enable-tunables --enable-stack-
1003 [21749] localedata: Wrong abbreviated day name (“abday”) for
1005 [21756] localedata: missing yesstr, nostr for nds_DE and nds_NL
1006 [21757] localedata: missing yesstr, nostr for pap_AW and pap_CW
1007 [21759] localedata: missing yesstr and nostr for Tigrinya
1008 [21760] localedata: Fix LC_MESSAGES and LC_ADDRESS for anp_IN
1009 [21766] localedata: Wrong LC_MESSAGES for om_ET Locale
1010 [21767] localedata: Missing Bislama locales
1011 [21768] localedata: Missing yesstr and nostr for aa_ET
1012 [21770] localedata: Missing Field in li_NL
1013 [21778] nptl: Robust mutex may deadlock
1014 [21779] libc: MicroBlaze segfaults when loading libpthread
1015 [21783] localedata: Fix int_select international_call_prefixes
1016 [21784] localedata: Inconsistency in country_isbn
1017 [21788] localedata: Missing Country Postal Abbreviations
1018 [21794] localedata: Added-country_isbn-for-Italy
1019 [21795] localedata: Add/Fix country_isbn for France
1020 [21796] localedata: Added country_isbn for Republic of Korea
1021 [21797] localedata: Fix inconsistency in country_isbn and missing prefixes
1022 [21799] localedata: Added int_select international_call_prefixes
1023 [21801] localedata: Added int_select international_call_prefixes
1024 [21804] nptl: Double semicolon in thread-shared-types.h
1025 [21807] localedata: LC_ADDRESS fix for pap_CW
1026 [21808] localedata: Fix LC_ADDRESS for pap_AW
1027 [21821] localedata: Added country_name in mai_IN
1028 [21822] localedata: Fix LC_TIME for mai_IN
1029 [21823] localedata: missing yesstr, nostr for sa_IN
1030 [21825] localedata: Fix name_mrs for mag_IN
1031 [21828] localedata: 2.26 changelog should mention user visible changes
1033 [21835] localedata: Added Maithili language locale for Nepal
1034 [21838] localedata: Removed redundant data for the_NP
1035 [21839] localedata: Fix LC_MONETARY for ta_LK
1036 [21844] localedata: Fix Latin characters and Months Sequence.
1037 [21848] localedata: Fix mai_NP Title Name
1042 * The feature test macro __STDC_WANT_LIB_EXT2__, from ISO/IEC TR
1043 24731-2:2010, is supported to enable declarations of functions from that
1044 TR. Note that not all functions from that TR are supported by the GNU C
1047 * The feature test macro __STDC_WANT_IEC_60559_BFP_EXT__, from ISO/IEC TS
1048 18661-1:2014, is supported to enable declarations of functions and macros
1049 from that TS. Note that not all features from that TS are supported by
1052 * The feature test macro __STDC_WANT_IEC_60559_FUNCS_EXT__, from ISO/IEC TS
1053 18661-4:2015, is supported to enable declarations of functions and macros
1054 from that TS. Note that most features from that TS are not supported by
1057 * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are
1058 now treated as compatibility synonyms for _POSIX_C_SOURCE=199506L.
1059 Since the GNU C Library defaults to a much newer revision of POSIX, this
1060 will only affect programs that specifically request an old conformance
1061 mode. For instance, a program compiled with -std=c89 -D_REENTRANT will
1062 see a change in the visible declarations, but a program compiled with
1063 just -D_REENTRANT, or -std=c99 -D_POSIX_C_SOURCE=200809L -D_REENTRANT,
1066 Some C libraries once required _REENTRANT and/or _THREAD_SAFE to be
1067 defined by all multithreaded code, but glibc has not required this for
1070 * The inclusion of <sys/sysmacros.h> by <sys/types.h> is deprecated. This
1071 means that in a future release, the macros “major”, “minor”, and “makedev”
1072 will only be available from <sys/sysmacros.h>.
1074 These macros are not part of POSIX nor XSI, and their names frequently
1075 collide with user code; see for instance glibc bug 19239 and Red Hat bug
1076 130601. <stdlib.h> includes <sys/types.h> under _GNU_SOURCE, and C++ code
1077 presently cannot avoid being compiled under _GNU_SOURCE, exacerbating the
1080 * New <fenv.h> features from TS 18661-1:2014 are added to libm: the
1081 fesetexcept, fetestexceptflag, fegetmode and fesetmode functions, the
1082 femode_t type and the FE_DFL_MODE and FE_SNANS_ALWAYS_SIGNAL macros.
1084 * Integer width macros from TS 18661-1:2014 are added to <limits.h>:
1085 CHAR_WIDTH, SCHAR_WIDTH, UCHAR_WIDTH, SHRT_WIDTH, USHRT_WIDTH, INT_WIDTH,
1086 UINT_WIDTH, LONG_WIDTH, ULONG_WIDTH, LLONG_WIDTH, ULLONG_WIDTH; and to
1087 <stdint.h>: INT8_WIDTH, UINT8_WIDTH, INT16_WIDTH, UINT16_WIDTH,
1088 INT32_WIDTH, UINT32_WIDTH, INT64_WIDTH, UINT64_WIDTH, INT_LEAST8_WIDTH,
1089 UINT_LEAST8_WIDTH, INT_LEAST16_WIDTH, UINT_LEAST16_WIDTH,
1090 INT_LEAST32_WIDTH, UINT_LEAST32_WIDTH, INT_LEAST64_WIDTH,
1091 UINT_LEAST64_WIDTH, INT_FAST8_WIDTH, UINT_FAST8_WIDTH, INT_FAST16_WIDTH,
1092 UINT_FAST16_WIDTH, INT_FAST32_WIDTH, UINT_FAST32_WIDTH, INT_FAST64_WIDTH,
1093 UINT_FAST64_WIDTH, INTPTR_WIDTH, UINTPTR_WIDTH, INTMAX_WIDTH,
1094 UINTMAX_WIDTH, PTRDIFF_WIDTH, SIG_ATOMIC_WIDTH, SIZE_WIDTH, WCHAR_WIDTH,
1097 * New <math.h> features are added from TS 18661-1:2014:
1099 - Signaling NaN macros: SNANF, SNAN, SNANL.
1101 - Nearest integer functions: roundeven, roundevenf, roundevenl, fromfp,
1102 fromfpf, fromfpl, ufromfp, ufromfpf, ufromfpl, fromfpx, fromfpxf,
1103 fromfpxl, ufromfpx, ufromfpxf, ufromfpxl.
1105 - llogb functions: the llogb, llogbf and llogbl functions, and the
1106 FP_LLOGB0 and FP_LLOGBNAN macros.
1108 - Max-min magnitude functions: fmaxmag, fmaxmagf, fmaxmagl, fminmag,
1111 - Comparison macros: iseqsig.
1113 - Classification macros: iscanonical, issubnormal, iszero.
1115 - Total order functions: totalorder, totalorderf, totalorderl,
1116 totalordermag, totalordermagf, totalordermagl.
1118 - Canonicalize functions: canonicalize, canonicalizef, canonicalizel.
1120 - NaN functions: getpayload, getpayloadf, getpayloadl, setpayload,
1121 setpayloadf, setpayloadl, setpayloadsig, setpayloadsigf, setpayloadsigl.
1123 * The functions strfromd, strfromf, and strfroml, from ISO/IEC TS 18661-1:2014,
1124 are added to libc. They convert a floating-point number into string.
1126 * Most of glibc can now be built with the stack smashing protector enabled.
1127 It is recommended to build glibc with --enable-stack-protector=strong.
1128 Implemented by Nick Alcock (Oracle).
1130 * The function explicit_bzero, from OpenBSD, has been added to libc. It is
1131 intended to be used instead of memset() to erase sensitive data after use;
1132 the compiler will not optimize out calls to explicit_bzero even if they
1133 are "unnecessary" (in the sense that no _correct_ program can observe the
1134 effects of the memory clear).
1136 * On ColdFire, MicroBlaze, Nios II and SH3, the float_t type is now defined
1137 to float instead of double. This does not affect the ABI of any libraries
1138 that are part of the GNU C Library, but may affect the ABI of other
1139 libraries that use this type in their interfaces.
1141 * On x86_64, when compiling with -mfpmath=387 or -mfpmath=sse+387, the
1142 float_t and double_t types are now defined to long double instead of float
1143 and double. These options are not the default, and this does not affect
1144 the ABI of any libraries that are part of the GNU C Library, but it may
1145 affect the ABI of other libraries that use this type in their interfaces,
1146 if they are compiled or used with those options.
1148 * The getentropy and getrandom functions, and the <sys/random.h> header file
1151 * The buffer size for byte-oriented stdio streams is now limited to 8192
1152 bytes by default. Previously, on Linux, the default buffer size on most
1153 file systems was 4096 bytes (and thus remains unchanged), except on
1154 network file systems, where the buffer size was unpredictable and could be
1155 as large as several megabytes.
1157 * The <sys/quota.h> header now includes the <linux/quota.h> header. Support
1158 for the Linux quota interface which predates kernel version 2.4.22 has
1161 * The malloc_get_state and malloc_set_state functions have been removed.
1162 Already-existing binaries that dynamically link to these functions will
1163 get a hidden implementation in which malloc_get_state is a stub. As far
1164 as we know, these functions are used only by GNU Emacs and this change
1165 will not adversely affect already-built Emacs executables. Any undumped
1166 Emacs executables, which normally exist only during an Emacs build, should
1167 be rebuilt by re-running “./configure; make” in the Emacs build tree.
1169 * The “ip6-dotint” and “no-ip6-dotint” resolver options, and the
1170 corresponding RES_NOIP6DOTINT flag from <resolv.h> have been removed.
1171 “no-ip6-dotint” had already been the default, and support for the
1172 “ip6-dotint” option was removed from the Internet in 2006.
1174 * The "ip6-bytestring" resolver option and the corresponding RES_USEBSTRING
1175 flag from <resolv.h> have been removed. The option relied on a
1176 backwards-incompatible DNS extension which was never deployed on the
1179 * The flags RES_AAONLY, RES_PRIMARY, RES_NOCHECKNAME, RES_KEEPTSIG,
1180 RES_BLAST defined in the <resolv.h> header file have been deprecated.
1181 They were already unimplemented.
1183 * The "inet6" option in /etc/resolv.conf and the RES_USE_INET6 flag for
1184 _res.flags are deprecated. The flag was standardized in RFC 2133, but
1185 removed again from the IETF name lookup interface specification in RFC
1186 2553. Applications should use getaddrinfo instead.
1188 * DNSSEC-related declarations and definitions have been removed from the
1189 <arpa/nameser.h> header file, and libresolv will no longer attempt to
1190 decode the data part of DNSSEC record types. Previous versions of glibc
1191 only implemented minimal support for the previous version of DNSSEC, which
1192 is incompatible with the currently deployed version.
1194 * The resource record type classification macros ns_t_qt_p, ns_t_mrr_p,
1195 ns_t_rr_p, ns_t_udp_p, ns_t_xfr_p have been removed from the
1196 <arpa/nameser.h> header file because the distinction between RR types and
1197 meta-RR types is not officially standardized, subject to revision, and
1198 thus not suitable for encoding in a macro.
1200 * The types res_sendhookact, res_send_qhook, re_send_rhook, and the qhook
1201 and rhook members of the res_state type in <resolv.h> have been removed.
1202 The glibc stub resolver did not support these hooks, but the header file
1203 did not reflect that.
1205 * For multi-arch support it is recommended to use a GCC which has
1206 been built with support for GNU indirect functions. This ensures
1207 that correct debugging information is generated for functions
1208 selected by IFUNC resolvers. This support can either be enabled by
1209 configuring GCC with '--enable-gnu-indirect-function', or by
1210 enabling it by default by setting 'default_gnu_indirect_function'
1211 variable for a particular architecture in the GCC source file
1214 * GDB pretty printers have been added for mutex and condition variable
1215 structures in POSIX Threads. When installed and loaded in gdb these pretty
1216 printers show various pthread variables in human-readable form when read
1217 using the 'print' or 'display' commands in gdb.
1219 * Tunables feature added to allow tweaking of the runtime for an application
1220 program. This feature can be enabled with the '--enable-tunables' configure
1221 flag. The GNU C Library manual has details on usage and README.tunables has
1222 instructions on adding new tunables to the library.
1224 * A new version of condition variables functions have been implemented in
1225 the NPTL implementation of POSIX Threads to provide stronger ordering
1228 * A new version of pthread_rwlock functions have been implemented to use a more
1229 scalable algorithm primarily through not using a critical section anymore to
1232 Security related changes:
1234 * On ARM EABI (32-bit), generating a backtrace for execution contexts which
1235 have been created with makecontext could fail to terminate due to a
1236 missing .cantunwind annotation. This has been observed to lead to a hang
1237 (denial of service) in some Go applications compiled with gccgo. Reported
1238 by Andreas Schwab. (CVE-2016-6323)
1240 * The DNS stub resolver functions would crash due to a NULL pointer
1241 dereference when processing a query with a valid DNS question type which
1242 was used internally in the implementation. The stub resolver now uses a
1243 question type which is outside the range of valid question type values.
1246 The following bugs are resolved with this release:
1248 [4099] stdio: Overly agressive caching by stream i/o functions.
1249 [7065] build: Support building glibc with -fstack-protector or -fstack-
1251 [9842] localedata: en_CA: incorrect date format
1252 [13165] nptl: pthread_cond_wait() can consume a signal that was sent
1253 before it started waiting
1254 [14139] manual: Do not hardcode platform names in manual/libm-err-tab.pl
1255 [15765] nptl: sem_open is wrongly a cancellation point
1256 [16421] network: IN6_IS_ADDR_UNSPECIFIED can use undefined s6_addr32
1257 [16458] libc: endian.h and netinet/in.h byte order macros return values of
1259 [16628] dynamic-link: Segfault after a binary without pthread dlopen()s a
1260 library linked with pthread
1261 [16630] nptl: Use SYSENTER for pthread_cond_broadcast/signal() (i.e. fix
1262 "FIXME: Ingo" issue)
1263 [16907] libc: <argp.h> compiled with --std=cXX disables __attribute__
1264 [17252] libc: getrandom and getentropy syscall
1265 [17730] malloc: thread-local storage is sometimes improperly free()'d
1266 after being __libc_memalign()'d
1267 [18241] stdio: failed fseek on memstream does not set errno and can fail
1269 [18243] nptl: sem_wait, sem_timedwait are cancellation points shm_open is
1271 [18463] nptl: pthread_cond_broadcast issue when surrounded by
1272 PTHREAD_PRIO_INHERIT mutex on ARM
1273 [18784] network: res_query and related function crash for special record
1274 type queries (CVE-2015-5180)
1275 [19380] math: strtod does not raise "inexact"
1276 [19387] string: Integer overflow in memchr
1277 [19390] string: Integer overflow in strncat
1278 [19398] build: linknamespace tests fail with massively parallel build
1279 [19402] nptl: Deadlock with robust shared mutex and asynchronous
1281 [19469] malloc: M_PERTURB in test-skeleton.c invalidates malloc tests
1282 [19473] malloc: Turn malloc_get_state etc. in compatibility symbols
1283 [19514] libc: [PATCH] Fix spelling errors in spelling
1284 "implement"/"implementation" in several places
1285 [19582] network: Deprecate RES_USE_INET6
1286 [19673] manual: clog10 docs appear to be erroneous
1287 [19810] dynamic-link: dlopen with both RTLD_NOLOAD and RTLD_NODELETE
1288 causes a segmentation fault
1289 [19826] libc: invalid pointer returned from __tls_get_addr with static
1291 [20016] network: resolv: Remove hooks support from the API
1292 [20019] dynamic-link: NULL pointer dereference in libc.so.6 IFUNC due to
1294 [20033] math: [x86_64] vectorized math function don't call the __finite
1296 [20116] nptl: use after free in pthread_create
1297 [20181] stdio: open_memstream(): writes not at end of stream corrupt data
1298 [20292] dynamic-link: Comparison in elf/dl-open.c _dl_addr_inside_object
1300 [20311] nptl: please install proc_service.h
1301 [20366] libc: Compilation errors in installed headers in strict-compliance
1303 [20370] malloc: malloc: Arena free list management is still racy
1304 (incorrect fix in bug 19243)
1305 [20386] libc: assert (X = 0) does not result in GCC warning
1306 [20432] malloc: malloc: Minimize interface required for interposition
1307 [20435] libc: Missing unwind info in __startcontext causes infinite loop
1308 in _Unwind_Backtrace (CVE-2016-6323)
1309 [20444] hurd: recvmsg: PF_LOCAL sockets and msg_name lead to SIGLOST
1310 [20452] nptl: Addition of sysdep.o to libpthread.a breaks relinking
1312 [20455] math: [powerpc] fesetexceptflag fails to clear FE_INVALID
1313 [20459] localedata: et_EE: locale has wrong {p,n}_cs_precedes value
1314 [20477] network: resolv: incorrect double-checked locking related to
1316 [20478] libc: libc_ifunc macro and similar usages leads to false debug-
1318 [20495] math: x86_64 performance degradation due to AVX/SSE transition
1320 [20497] localedata: lt_LT: LC_TIME d_fmt used is obsolete
1321 [20508] dynamic-link: _dl_runtime_resolve_avx/_dl_runtime_profile_avx512
1322 cause transition penalty
1323 [20517] math: sparcv9 missing fdiml compat symbol
1324 [20524] manual: strverscmp is inconsistent
1325 [20525] libc: <sys/quota.h> should be based on kernel headers
1326 [20539] math: GCC 7 -static -lm fails to link at -O3
1327 [20554] libc: ld: bss-plt forced due to /usr/lib/libc_nonshared.a(ppc-
1329 [20558] string: POSIX bcopy/bzero decls do not implement Object Size
1331 [20591] network: Remove obsolete DNSSEC support
1332 [20592] network: DNS resource record type classification macros in
1333 <arpa/nameser.h> are incorrect
1334 [20593] network: Update DNS RR type definitions
1335 [20611] network: getaddrinfo accepts invalid numeric scope IDs
1336 [20615] build: glibc build fails when using --with-cpu=power9 --enable-
1338 [20629] network: libresolv: Remove support for bitstring labels
1340 [20647] libc: GLIBC quitting every program - glibc on Pentium-S leads to
1341 assertion: "maxidx >= 2"
1342 [20660] math: [arm] Use VSQRT
1343 [20662] libc: checking whether x86_64-pc-linux-gnu-gcc implicitly enables
1344 -fstack-protector no (32bit gcc 6.2.0 pie and ssp enable)
1345 [20689] libc: Test for FMA should also check for AVX.
1346 [20707] glob: gl_pathv entries not set to NULL with GLOB_DOOFFS
1347 [20715] math: iszero macro breaks existing code
1348 [20718] math: [powerpc] copysignl raises "invalid" for sNaN
1349 [20728] libc: powerpc: Missing TOC stub in clone
1350 [20729] build: glibc-2.24 fails to build for i486 with -Os
1351 [20750] build: Build fails with default PIE enabled gcc-6.x
1352 [20768] math: [alpha] sqrt fegetenv namespace
1353 [20785] libc: binutils 2.28 fails to assemble power6/memset.S file in
1355 [20787] math: float_t is defined as float incorrectly on x86_64 even if
1356 __FLT_EVAL_METHOD__ is 2
1357 [20790] network: rpcgen buffer overrun in get_prog_declaration
1358 [20822] nptl: race condition in __lll_unlock_elision on powerpc
1359 [20829] libc: crypt snprintf namespace
1360 [20847] libc: tst-vfork3 failure
1361 [20855] math: Default bits/mathdef.h has inappropriate float_t
1362 [20859] math: [sh4] FP_ILOGB0 invalid
1363 [20864] localedata: iconv: cp936 missing single-byte euro sign (0x80,
1364 U+20AC), not same as GBK
1365 [20915] dynamic-link: global-dynamic TLS broken on aarch64
1366 [20916] math: pow handling of sNaN arguments
1367 [20918] build: Building with --enable-nss-crypt fails tst-linkall-static.
1368 [20919] math: Bad pow (qNaN, 0) result with -lieee
1369 [20940] math: hypot sNaN handling
1370 [20947] math: fmax, fmin sNaN handling
1371 [20956] libc: debug/tst-backtrace3-6 don't work with -O3 anymore
1372 [20964] network: sunrpc: Stack-based buffer overflow in getrpcport with
1374 [20971] string: powerpc64/power7 memchr overflows internal pointer check
1375 [20973] nptl: robust mutexes: Lost wake-ups
1376 [20974] locale: bs_BA: yesexpr/noexpr regexes accept any character
1377 [20978] nis: strlen on null pointer in nss_nisplus
1378 [20985] nptl: robust mutexes: lowlevelrobustlock assembly on x86 blocks on
1380 [21014] string: i686 memchr overflows internal pointer check
1381 [21019] libc: [mips] n32 lseek incorrect on overflow
1382 [21022] libc: [microblaze] __backtrace get_frame_size namespace
1383 [21026] libc: [MIPS] readahead syscall is broken on n64
1384 [21028] math: Fallback fesetexceptflag should always succeed
1385 [21045] libc: [powerpc-nofpu] swapcontext does not restore signal mask
1386 [21047] math: arm: fpu_control.h: _FPU_GETCW/_FPU_SETCW is rejected by
1388 [21053] libc: [SH] Namespace pollution from sys/ucontext.h
1389 [21061] librt: [microblaze] librt lost clock_* exports
1390 [21073] libc: tunables: insecure environment variables passed to
1391 subprocesses with AT_SECURE
1392 [21081] string: Missing vzeroupper in memset-vec-unaligned-erms.S
1396 * The minimum Linux kernel version that this version of the GNU C Library
1397 can be used with is 3.2, except on i[4567]86 and x86_64, where Linux
1398 kernel version 2.6.32 or later suffices (on architectures that already
1399 required kernel versions more recent than 3.2, those requirements remain
1400 unchanged). Linux 3.2 or later kernel headers are required on all
1403 * The pap_AN locale has been deleted. This has been deprecated for a long
1404 time. It has been replaced by pap_AW & pap_CW, both of which have long
1405 been included in previous releases.
1407 * The readdir_r and readdir64_r functions have been deprecated. It is
1408 recommended to use readdir and readdir64 instead.
1410 * The type “union wait” has been removed. It was deprecated in the early
1411 1990s and never part of POSIX. Application code should use the int type
1412 instead of “union wait”.
1414 * A new NSS action is added to facilitate large distributed system
1415 administration. The action, MERGE, allows remote user stores like LDAP
1416 to be merged into local user stores like /etc/groups in order to provide
1417 easy to use, updated, and managed sets of merged credentials. The new
1418 action can be used by configuring it in /etc/nsswitch.conf:
1419 group: files [SUCCESS=merge] nis
1420 Implemented by Stephen Gallagher (Red Hat).
1422 * The deprecated __malloc_initialize_hook variable has been removed from the
1425 * The long unused localedef --old-style option has been removed. It hasn't
1426 done anything in over 16 years. Scripts using this option can safely
1429 * nextupl, nextup, nextupf, nextdownl, nextdown and nextdownf are added to
1430 libm. They are defined by TS 18661 and IEEE754-2008. The nextup functions
1431 return the next representable value in the direction of positive infinity
1432 and the nextdown functions return the next representable value in the
1433 direction of negative infinity. These are currently enabled as GNU
1436 Security related changes:
1438 * An unnecessary stack copy in _nss_dns_getnetbyname_r was removed. It
1439 could result in a stack overflow when getnetbyname was called with an
1440 overly long name. (CVE-2016-3075)
1442 * Previously, getaddrinfo copied large amounts of address data to the stack,
1443 even after the fix for CVE-2013-4458 has been applied, potentially
1444 resulting in a stack overflow. getaddrinfo now uses a heap allocation
1445 instead. Reported by Michael Petlan. (CVE-2016-3706)
1447 * The glob function suffered from a stack-based buffer overflow when it was
1448 called with the GLOB_ALTDIRFUNC flag and encountered a long file name.
1449 Reported by Alexander Cherepanov. (CVE-2016-1234)
1451 * The Sun RPC UDP client could exhaust all available stack space when
1452 flooded with crafted ICMP and UDP messages. Reported by Aldy Hernandez'
1453 alloca plugin for GCC. (CVE-2016-4429)
1455 * The IPv6 name server management code in libresolv could result in a memory
1456 leak for each thread which is created, performs a failing naming lookup,
1457 and exits. Over time, this could result in a denial of service due to
1458 memory exhaustion. Reported by Matthias Schiffer. (CVE-2016-5417)
1460 The following bugs are resolved with this release:
1462 [1170] localedata: ne_NP: update Nepali locale definition file
1463 [3629] manual: stpcpy description in string.texi refers to MS-DOG instead
1465 [6527] malloc: [powerpc] Malloc alignment insufficient for PowerPC
1466 [6796] math: fdim() does not set errno on overflow
1467 [10354] libc: posix_spawn should use vfork() in more cases than presently
1468 [11213] localedata: localedata: add copyright disclaimer to locale files
1469 [12143] localedata: chr_US: new Cherokee locale
1470 [12450] localedata: sgs_LT: new locale
1471 [12676] localedata: ln_CD: new locale
1472 [13237] localedata: LC_ADDRESS.country_name: update all locales w/latest
1474 [13304] math: fma, fmaf, fmal produce wrong results
1475 [14259] build: --localedir arg to configure is ignored
1476 [14499] nptl: Does posix_spawn invoke atfork handlers / use vfork?
1477 [14750] libc: Race condition in posix_spawn vfork usage vs signal handlers
1478 [14934] localedata: es_CL: wrong first weekday chilean locale
1479 [15262] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of
1481 [15263] localedata: LC_MESSAGES.yesexpr/noexpr: inconsistent use of 1/0
1483 [15264] localedata: LC_MESSAGES.yesstr/nostr: lacking in many locales
1484 [15368] nptl: raise() is not async-signal-safe
1485 [15479] math: ceil, floor, round and trunc raise inexact exception
1486 [15578] localedata: kk_KZ: various updates
1487 [16003] localedata: pap_AN: punt old locale
1488 [16137] localedata: iw_IL: punt old locale
1489 [16190] localedata: eo: new esperanto locale
1490 [16374] localedata: lv_LV: change currency symbol in LC_MONETARY to euro
1491 [16742] malloc: race condition: pthread_atfork() called before first
1492 malloc() results in unexpected locking behaviour/deadlocks
1493 [16975] localedata: LC_MESSAGES.yesexpr/noexpr: revisit capitalization in
1495 [16983] localedata: postal_fmt does not allow %l and %n modifiers
1496 [17565] localedata: pt_PT: wrong (work-)week start
1497 [17899] math: [powerpc] floorl returns negative zero with FE_DOWNWARD
1498 [17950] build: Build fails with -msse
1499 [18205] localedata: be_BY*: wrong first_weekday and first_workday
1500 [18433] libc: posix_spawn does not return correctly upon failure to
1502 [18453] localedata: charmaps/IBM875: incorrect codes
1503 [18712] string: bits/string2.h incompatible with -O2 -Werror=packed
1505 [18896] localedata: he_IL: improvements for currency
1506 [18911] localedata: ro_RO: Correcting week day name for "Tuesday" in
1507 Romanian locale data
1508 [18960] locale: s390: _nl_locale_subfreeres uses larl opcode on misaligned
1510 [19056] libc: Deprecate readdir_r
1511 [19133] localedata: pt_*: days & months should be lowercase in Portuguese
1513 [19198] localedata: nl_NL: small improvements for Dutch locales
1514 [19257] network: Per-thread memory leak in __res_vinit with IPv6
1515 nameservers (CVE-2016-5417)
1516 [19269] build: tst-audit4 and tst-audit10 failures with gcc-6 on non avx
1518 [19400] locale: Language missing in "iso-639.def", trivial fix in
1520 [19431] malloc: Deadlock between fflush, getdelim, and fork
1521 [19505] libc: Incorrect file descriptor validity checks in
1522 posix_spawn_file_actions_add{open,close,dup2}
1523 [19509] dynamic-link: dlsym, dlvsym do not report errors through dlerror
1524 when using RTLD_NEXT
1525 [19512] locale: Stale `#ifndef HAVE_BUILTIN_EXPECT' in
1526 `intl/{gettextP,loadinfo}.h'
1527 [19534] libc: execle, execlp may use malloc
1528 [19568] localedata: *_CH: Swiss locales have inconsistent start of week
1529 [19573] network: res_nclose and __res_maybe_init disagree about name
1530 server initialization, breaking Hesiod
1531 [19575] localedata: Status of GB18030 tables
1532 [19581] localedata: sr_* date_fmt string contains additional newline
1533 [19583] string: SSSE3_Fast_Copy_Backward flag needs to be enabled for AMD
1535 [19592] math: [ldbl-128ibm] ceill incorrect in non-default rounding modes
1536 [19593] math: [ldbl-128ibm] truncl incorrect in non-default rounding modes
1537 [19594] math: [ldbl-128ibm] roundl incorrect in non-default rounding modes
1538 [19595] math: [ldbl-128ibm] fmodl incorrect for results in subnormal
1540 [19602] math: [ldbl-128ibm] fmodl handling of equal arguments with low
1542 [19603] math: [ldbl-128ibm] remainderl, remquol incorrect sign handling in
1544 [19610] dynamic-link: ldconfig -X removes stale symbolic links
1545 [19613] libc: s390x (64 bit) macro expansion WCOREDUMP and others
1546 [19633] locale: strfmon_l applies global locale to number formatting
1547 [19642] network: Memory leak in getnameinfo
1548 [19648] libc: test-skeleton.c: Do not set RLIMIT_DATA
1549 [19653] libc: Potential for NULL pointer dereference (CWE-476) in
1551 [19654] math: [x86_64] Need testcase for BZ #19590 fix
1552 [19671] localedata: Missing Sanity Check for malloc() in 'tst-fmon.c' &
1554 [19674] math: [ldbl-128ibm] powl incorrect overflow handling
1555 [19677] math: [ldbl-128ibm] remainderl equality test incorrect for zero
1557 [19678] math: [ldbl-128ibm] nextafterl, nexttowardl incorrect sign of zero
1559 [19679] dynamic-link: gcc-4.9.3 C++ exception handling broken due to
1561 [19726] locale: Converting UCS4LE to INTERNAL with iconv() does not update
1562 pointers and lengths in error-case.
1563 [19727] locale: Converting from/to UTF-xx with iconv() does not always
1564 report errors on UTF-16 surrogates values.
1565 [19755] nscd: nscd assertion failure in gc
1566 [19758] dynamic-link: Typo in EXTRA_LD_ENVVARS for x86-64
1567 [19759] libc: mempcpy shouldn't be inlined
1568 [19762] dynamic-link: HAS_CPU_FEATURE/HAS_ARCH_FEATURE are easy to misuse
1569 [19765] libc: s390 needs an optimized mempcpy
1570 [19779] glob: glob: buffer overflow with GLOB_ALTDIRFUNC due to incorrect
1571 NAME_MAX limit assumption (CVE-2016-1234)
1572 [19783] build: benchtests don't support --enable-hardcoded-path-in-tests
1573 [19787] network: Missing and incorrect truncation checks in getnameinfo
1574 [19790] math: [ldbl-128ibm] nearbyintl incorrect in non-default rounding
1576 [19791] network: Assertion failure in res_query.c with un-connectable name
1578 [19792] libc: MIPS: backtrace yields infinite backtrace with makecontext
1579 [19822] math: libm.so install clobbers old version
1580 [19825] network: resolv: send_vc can return uninitialized data in second
1581 response to getaddrinfo
1582 [19830] network: nss_dns: should check RDATA length against buffer length
1583 [19831] network: nss_dns: getaddrinfo returns uninitialized data when
1584 confronted with A/AAAA records of invalid size
1585 [19837] nss: nss_db: No retries for some long lines with a larger buffer
1586 [19848] math: powl(10,n) for n=-4,-5,-6,-7 is off by more than 1 ULP
1587 [19853] stdio: Printing IBM long double in decimal with high precision is
1589 [19860] build: x86_64: compile errors for tst-audit10 and tst-auditmod10b
1590 [19861] nptl: libpthread IFUNC resolver for fork can lead to crash
1591 [19862] network: resolv, nss_dns: Remove remaining logging of unexpected
1593 [19865] network: Assertion failure or memory leak in
1594 _nss_dns_getcanonname_r
1595 [19868] network: nss_dns: netent code does not skip over non-PTR records
1596 [19879] network: nss_dns: Stack overflow in getnetbyname implementation
1598 [19881] string: Improve x86-64 memset
1599 [19907] string: Incorrect memcpy tests
1600 [19916] dynamic-link: S390: fprs/vrs are not saved/restored while
1602 [19925] libc: termios.h XCASE namespace
1603 [19928] string: memmove-vec-unaligned-erms.S is slow with large data size
1604 [19929] libc: limits.h NL_NMAX namespace
1605 [19931] stdio: Memory leak in vfprintf
1606 [19957] libc: clone(CLONE_VM) access invalid parent memory
1607 [19963] localedata: en_IL: New locale
1608 [19989] stdio: stdio.h cuserid namespace
1609 [19994] network: getaddrinfo does not restore RES_USE_INET6 flag in
1611 [19996] locale: langinfo.h nl_langinfo_l namespace
1612 [20005] stdio: fflush on a file opened with fmemopen resets position to 0
1613 [20010] network: getaddrinfo: Stack overflow in hostent translation
1615 [20012] stdio: libio: fmemopen append mode failure
1616 [20014] stdio: stdio.h namespace for pre-threads POSIX
1617 [20017] network: resolv: Use gmtime_r instead of gmtime in p_secstodate
1618 [20023] libc: fcntl.h timespec namespace
1619 [20024] math: [x86_64] vectorized sincos trashes the stack
1620 [20031] network: nss_hesiod: Heap overflow in get_txt_records
1621 [20041] time: sys/time.h timespec namespace
1622 [20043] libc: unistd.h missing cuserid for UNIX98 and before
1623 [20044] libc: unistd.h missing pthread_atfork for UNIX98
1624 [20051] libc: ttyslot in wrong header under wrong conditions
1625 [20054] libc: gethostname not declared for XPG4
1626 [20055] libc: termios.h missing tcgetsid for XPG4
1627 [20072] dynamic-link: x86 init_cpu_features is called twice in static
1629 [20073] libc: sys/stat.h fchmod namespace
1630 [20074] libc: stdlib.h rand_r namespace
1631 [20076] libc: sys/stat.h missing S_IFSOCK, S_ISSOCK for XPG4
1632 [20094] libc: stdlib.h should not declare grantpt, ptsname, unlockpt for
1634 [20111] libc: struct sockaddr_storage cannot be aggregate-copied
1635 [20112] network: sunrpc: stack (frame) overflow in Sun RPC clntudp_call
1637 [20115] string: Extra alignment in memset-vec-unaligned-erms.S
1638 [20119] libc: Wrong mask for processors level type from CPUID
1639 [20139] dynamic-link: Upper part of zmm is zeroed if Glibc is built with
1640 AS not supporting AVX512
1641 [20151] math: [ldbl-128/ldbl-128ibm] j0l, j1l, y0l, y1l return sNaN for
1643 [20153] math: [ldbl-128ibm] sqrtl (sNaN) returns sNaN
1644 [20156] math: [ldbl-128ibm] ceill, rintl etc. return sNaN for sNaN
1646 [20157] math: [powerpc] fabsl (sNaN) wrongly raises "invalid"
1647 [20160] math: [powerpc] ceil, rint etc. return sNaN for sNaN input
1648 [20178] libc: posix_spawn{p} should not call exit
1649 [20191] stdio: libio: vtables hardening
1650 [20195] string: FMA4 detection requires CPUID execution with register
1652 [20198] libc: quick_exit incorrectly destroys C++11 thread objects.
1653 [20205] math: [i386/x86_64] nextafterl incorrect incrementing negative
1655 [20212] math: acos (sNaN) returns sNaN
1656 [20213] math: asin (sNaN) returns sNaN
1657 [20214] network: Linux header sync with linux/in6.h and ipv6.h again.
1658 [20218] math: [i386] asinhl (sNaN) returns sNaN
1659 [20219] math: [i386] atanhl (sNaN) returns sNaN
1660 [20222] stdio: fopencookie: Mangle function pointers
1661 [20224] math: [i386] cbrtl (sNaN) returns sNaN
1662 [20225] math: ldexp, scalbn, scalbln return sNaN for sNaN input
1663 [20226] math: [i386/x86_64] expl, exp10l, expm1l return sNaN for sNaN
1665 [20227] math: [i386/x86_64] logl (sNaN) returns sNaN
1666 [20228] math: [i386/x86_64] log10l (sNaN) returns sNaN
1667 [20229] math: [i386/x86_64] log1pl (sNaN) returns sNaN
1668 [20232] math: [ldbl-128] expm1l (sNaN) returns sNaN
1669 [20233] math: [ldbl-128ibm] expm1l (sNaN) returns sNaN
1670 [20234] math: [ldbl-128ibm] log1pl (sNaN) returns sNaN
1671 [20235] math: [i386/x86_64] log2l (sNaN) returns sNaN
1672 [20237] nss: nss_db: get*ent segfaults without preceding set*ent
1673 [20240] math: modf (sNaN) returns sNaN
1674 [20248] libc: debug/tst-longjump_chk2 calls printf from a signal handler
1675 [20250] math: frexp (sNaN) returns sNaN
1676 [20252] math: atan2 (sNaN, qNaN) fails to raise "invalid"
1677 [20255] math: [i386] fdim, fdimf return with excess range and precision /
1679 [20256] math: [i386/x86_64] fdiml returns sNaN for sNaN input
1680 [20260] string: ../sysdeps/x86/bits/string.h:1092:3: error: array
1681 subscript is below array bounds [-Werror=array-bounds]
1682 [20262] nis: _nss_nis_initgroups_dyn always returns NSS_STATUS_NOTFOUND
1683 [20263] nptl: robust mutex deadlocks if other thread requests timedlock
1685 [20277] libc: $dp is not initialized correctly in sysdeps/hppa/start.S
1686 [20284] malloc: malloc: Corrupt arena avoidance causes unnecessary mmap
1688 [20296] math: [i386/x86_64] scalbl returns sNaN for sNaN input, missing
1689 "invalid" exceptions
1690 [20314] nptl: make[4]: *** [/usr/include/stdlib.h] Error 1
1691 [20316] localedata: id_ID: Februari instead of Pebruari
1692 [20327] string: POWER8 strcasecmp returns incorrect result
1693 [20347] math: Failure: Test: j0_downward (0xap+0)
1694 [20348] libc: FAIL: misc/tst-preadvwritev64
1695 [20349] libc: 64-bit value is passed differently in p{readv,writev}{64}
1696 [20350] libc: There is no test for p{read,write}64
1697 [20357] math: Incorrect cos result for 1.5174239687223976
1698 [20384] build: Don't run libmvec-sincos-avx* tests on non avx machines
1702 * Unicode 8.0.0 Support: Character encoding, character type info, and
1703 transliteration tables are all updated to Unicode 8.0.0, using new
1704 and/or improved generator scripts contributed by Mike FABIAN (Red Hat).
1705 These updates cause user visible changes, such as the fixes for bugs
1706 89, 16061, and 18568.
1708 * sched_setaffinity, pthread_setaffinity_np no longer attempt to guess the
1709 kernel-internal CPU set size. This means that requests that change the
1710 CPU affinity which failed before (for example, an all-ones CPU mask) will
1711 now succeed. Applications that need to determine the effective CPU
1712 affinities need to call sched_getaffinity or pthread_getaffinity_np after
1713 setting it because the kernel can adjust it (and the previous size check
1714 would not detect this in the majority of cases).
1716 * The fts.h header can now be used with -D_FILE_OFFSET_BITS=64. With LFS
1717 the following new symbols are used: fts64_children, fts64_close,
1718 fts64_open, fts64_read and fts64_set.
1720 * getaddrinfo now detects certain invalid responses on an internal netlink
1721 socket. If such responses are received, an affected process will
1722 terminate with an error message of "Unexpected error <number> on netlink
1723 descriptor <number>" or "Unexpected netlink response of size <number> on
1724 descriptor <number>". The most likely cause for these errors is a
1725 multi-threaded application which erroneously closes and reuses the netlink
1726 file descriptor while it is used by getaddrinfo.
1728 * A defect in the malloc implementation, present since glibc 2.15 (2012) or
1729 glibc 2.10 via --enable-experimental-malloc (2009), could result in the
1730 unnecessary serialization of memory allocation requests across threads.
1731 The defect is now corrected. Users should see a substantial increase in
1732 the concurent throughput of allocation requests for applications which
1733 trigger this bug. Affected applications typically create and
1734 destroy threads frequently. (Bug 19048 was reported and analyzed by
1737 * There is now a --disable-timezone-tools configure option for disabling the
1738 building and installing of the timezone related utilities (zic, zdump, and
1739 tzselect). This is useful for people who build the timezone data and code
1740 independent of the GNU C Library.
1742 * The obsolete header <regexp.h> has been removed. Programs that require
1743 this header must be updated to use <regex.h> instead.
1745 * The obsolete functions bdflush, create_module, get_kernel_syms,
1746 query_module and uselib are no longer available to newly linked binaries;
1747 the header <sys/kdaemon.h> has been removed. These functions and header
1748 were specific to systems using the Linux kernel and could not usefully be
1749 used with the GNU C Library on systems with version 2.6 or later of the
1752 * Optimized string, wcsmbs and memory functions for IBM z13.
1753 Implemented by Stefan Liebler.
1755 * Newly linked programs that define a variable called signgam will no longer
1756 have it set by the lgamma, lgammaf and lgammal functions. Programs that
1757 require signgam to be set by those functions must ensure that they use the
1758 variable provided by the GNU C Library and declared in <math.h>, without
1759 defining their own copy.
1761 * The minimum GCC version that can be used to build this version of the GNU
1762 C Library is GCC 4.7. Older GCC versions, and non-GNU compilers, can
1763 still be used to compile programs using the GNU C Library.
1765 Security related changes:
1767 * An out-of-bounds value in a broken-out struct tm argument to strftime no
1768 longer causes a crash. Reported by Adam Nielsen. (CVE-2015-8776)
1770 * The LD_POINTER_GUARD environment variable can no longer be used to disable
1771 the pointer guard feature. It is always enabled. Previously,
1772 LD_POINTER_GUARD could be used to disable security hardening in binaries
1773 running in privileged AT_SECURE mode. Reported by Hector Marco-Gisbert.
1776 * An integer overflow in hcreate and hcreate_r could lead to an
1777 out-of-bounds memory access. Reported by Szabolcs Nagy. (CVE-2015-8778)
1779 * The catopen function no longer has unbounded stack usage. Reported by
1780 Max. (CVE-2015-8779)
1782 * The nan, nanf and nanl functions no longer have unbounded stack usage
1783 depending on the length of the string passed as an argument to the
1784 functions. Reported by Joseph Myers. (CVE-2014-9761)
1786 * A stack-based buffer overflow was found in libresolv when invoked from
1787 libnss_dns, allowing specially crafted DNS responses to seize control
1788 of execution flow in the DNS client. The buffer overflow occurs in
1789 the functions send_dg (send datagram) and send_vc (send TCP) for the
1790 NSS module libnss_dns.so.2 when calling getaddrinfo with AF_UNSPEC
1791 family. The use of AF_UNSPEC triggers the low-level resolver code to
1792 send out two parallel queries for A and AAAA. A mismanagement of the
1793 buffers used for those queries could result in the response of a query
1794 writing beyond the alloca allocated buffer created by
1795 _nss_dns_gethostbyname4_r. Buffer management is simplified to remove
1796 the overflow. Thanks to the Google Security Team and Red Hat for
1797 reporting the security impact of this issue, and Robert Holiday of
1798 Ciena for reporting the related bug 18665. (CVE-2015-7547)
1800 The following bugs are resolved with this release:
1802 [89] localedata: Locales nb_NO and nn_NO should transliterate æøå
1803 [887] math: Math library function "logb" and "nextafter" inconsistent
1804 [2542] math: Incorrect return from float gamma (-0X1.FA471547C2FE5P+1)
1805 [2543] math: Incorrect return from float gamma (-0X1.9260DCP+1)
1806 [2558] math: Incorrect return from double gamma (-0X1.FA471547C2FE5P+1)
1807 [2898] libc: [improve] warning: the use of `mktemp' is dangerous, better
1809 [4404] localedata: German translation of "Alarm clock" is misleading
1810 [6799] math: nextafter() and nexttoward() doen't set errno on
1811 overflow/underflow errors
1812 [6803] math: scalb(), scalbln(), scalbn() do not set errno on
1814 [10432] nis: _nss_nis_setnetgrent assertion failure
1815 [11460] libc: fts has no LFS support
1816 [12926] network: getaddrinfo()/make_request() may spin forever
1817 [13065] nptl: Race condition in pthread barriers
1818 [13690] nptl: pthread_mutex_unlock potentially cause invalid access
1819 [14341] dynamic-link: Dynamic linker crash when DT_JMPREL and DT_REL{,A}
1821 [14551] math: [ldbl-128ibm] strtold overflow handling for IBM long double
1822 [14912] libc: Rename non-installed bits/*.h headers
1823 [15002] libc: Avoid undefined behavior in posix_fallocate overflow check
1824 [15367] math: Let gcc use __builtin_isinf
1825 [15384] math: One constant fewer in ieee754/dbl-64/wordsize-64/s_finite.c
1826 [15421] math: lgamma wrongly sets signgam for ISO C
1827 [15470] math: [arm] On ARM llrintl() and llroundl() do not raise
1828 FE_INVALID with argument out of range
1829 [15491] math: [i386/x86_64] x86 nearbyint implementations wrongly clear
1831 [15786] dynamic-link: ifunc resolver functions can smash function
1833 [15918] math: Unnecessary check for equality in hypotf()
1834 [16061] localedata: Review / update transliteration data
1835 [16068] math: [i386/x86_64] x86 and x86_64 fesetenv exclude state they
1837 [16141] time: strptime %z offset restriction
1838 [16171] math: drem should be alias of remainder
1839 [16296] math: fegetround is pure?
1840 [16347] math: [ldbl-128ibm] ldbl-128/e_lgammal_r.c may not be suitable.
1841 [16364] libc: sleep may leave SIGCHLD blocked on sync cancellation on
1843 [16399] math: [mips] lrint / llrint / lround / llround missing exceptions
1844 [16415] math: Clean up ldbl-128 / ldbl-128ibm expm1l for large positive
1846 [16422] math: [powerpc] math-float, math-double failing llrint tests with
1847 "Exception "Inexact" set" on ppc32
1848 [16495] localedata: nl_NL: date_fmt: shuffle year/month around
1849 [16517] math: Missing underflow exception from tanf/tan/tanl
1850 [16519] math: Missing underflow exception from sinhf
1851 [16520] math: Missing underflow exception from tanhf
1852 [16521] math: Missing underflow exception from exp2
1853 [16620] math: [ldbl-128ibm] exp10l spurious overflows / bad directed
1855 [16734] stdio: fopen calls mmap to allocate its buffer
1856 [16961] math: nan function incorrect handling of bad sequences
1857 [16962] math: nan function unbounded stack allocation (CVE-2014-9761)
1858 [16973] localedata: Fix lang_lib/lang_term as per ISO 639-2
1859 [16985] locale: localedef: confusing error message when opening output
1861 [17118] math: ctanh(INFINITY + 2 * I) returns incorrect value
1862 [17197] locale: Redundant shift character in iconv conversion output at
1864 [17243] libc: trunk/posix/execl.c:53: va_args problem ?
1865 [17244] libc: trunk/sysdeps/unix/sysv/linux/semctl.c:116: va_args muxup ?
1866 [17250] dynamic-link: static linking breaks nss loading
1867 (getaddrinfo/getpwnam/etc...)
1868 [17404] libc: atomic_exchange_rel lacking a barrier on MIPS16, GCC before
1870 [17441] math: isnan() should use __builtin_isnan() in GCC
1871 [17514] nptl: Assert failure unlocking ERRORCHECK mutex after timedlock
1872 (related to lock elision)
1873 [17787] manual: Exponent on page 324 of the PDF ends prematurely
1874 [17886] time: strptime should be able to parse "Z" as a timezone with %z
1875 [17887] time: strptime should be able to parse "+01:00" style timezones
1876 [17905] libc: catopen() Multiple unbounded stack allocations
1878 [18084] libc: backtrace (..., 0) dumps core on x86
1879 [18086] libc: nice() sets errno to 0 on success
1880 [18240] libc: hcreate, hcreate_r should fail with ENOMEM if element count
1881 is too large (CVE-2015-8778)
1882 [18251] dynamic-link: SONAME missing when audit modules provides path
1883 [18265] libc: add attributes for wchar string and memory functions
1884 [18370] math: csqrt missing underflows
1885 [18421] libc: [hppa] read-only segment has dynamic relocations
1886 [18472] libc: Obsolete syscall wrappers should be compat symbols
1887 [18480] libc: hppa glibc miscompilation in sched_setaffinity()
1888 [18491] localedata: Update tr_TR LC_CTYPE as part of Unicode updates
1889 [18525] localedata: Remove locale timezone information
1890 [18560] libc: [powerpc] spurious bits/ipc.h definitions
1891 [18568] localedata: Update locale data to Unicode 8.0
1892 [18589] locale: sort-test.sh fails at random
1893 [18595] math: ctan, ctanh missing underflows
1894 [18604] libc: assert macro-expands its argument
1895 [18610] math: S390: fetestexcept() reports any exception if DXC-code
1896 contains a vector instruction exception.
1897 [18611] math: j1, jn missing errno setting on underflow
1898 [18618] localedata: sync Chechen locale definitions with other *_RU
1900 [18647] math: powf(-0x1.000002p0, 0x1p30) returns 0 instead of +inf
1901 [18661] libc: Some x86-64 assembly codes don't align stack to 16 bytes
1902 [18665] network: In send_dg, the recvfrom function is NOT always using the
1903 buffer size of a newly created buffer (CVE-2015-7547)
1904 [18674] libc: [i386] trunk/sysdeps/i386/tst-auditmod3b.c:84: possible
1906 [18675] libc: fpathconf(_PC_NAME_MAX) fails against large filesystems for
1908 [18681] libc: regexp.h is obsolete and buggy, and should be desupported
1909 [18699] math: tilegx cproj() for various complex infinities does not yield
1911 [18724] libc: Harden put*ent functions against data injection
1912 [18743] nptl: PowerPC: findutils testcase fails with --enable-lock-elision
1913 [18755] build: build errors with -DNDEBUG
1914 [18757] stdio: fmemopen fails to set errno on failure
1915 [18778] dynamic-link: ld.so crashes if failed dlopen causes libpthread to
1917 [18781] libc: openat64 lacks O_LARGEFILE
1918 [18787] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/atomic.h:71:6:
1919 error: can’t find a register in class ‘R1_REGS’ while reloading ‘asm’
1920 [18789] math: [ldbl-128ibm] sinhl inaccurate near 0
1921 [18790] math: [ldbl-128ibm] tanhl inaccurate
1922 [18795] libc: stpncpy fortification misses buffer lengths that are
1923 statically too large
1924 [18796] build: build fails for --disable-mathvec
1925 [18803] math: hypot missing underflows
1926 [18820] stdio: fmemopen may leak memory on failure
1927 [18823] math: csqrt spurious underflows
1928 [18824] math: fma spurious underflows
1929 [18825] math: pow missing underflows
1930 [18857] math: [ldbl-128ibm] nearbyintl wrongly uses signaling comparisons
1931 [18868] nptl: pthread_barrier_init typo has in-theory-undefined behavior
1932 [18870] build: sem_open.c fails to compile with missing symbol
1934 [18872] stdio: Fix memory leak in printf_positional
1935 [18873] libc: posix_fallocate overflow check ineffective
1936 [18875] math: Excess precision leads incorrect libm
1937 [18877] libc: arm: mmap offset regression
1938 [18887] libc: memory corruption when using getmntent on blank lines
1939 [18918] localedata: hu_HU: change time to HH:MM:SS format
1940 [18921] libc: Regression: extraneous stat() and fstat() performed by
1942 [18928] dynamic-link: LD_POINTER_GUARD is not ignored for privileged
1943 binaries (CVE-2015-8777)
1944 [18951] math: tgamma missing underflows
1945 [18952] math: [ldbl-128/ldbl-128ibm] lgammal spurious "invalid", incorrect
1947 [18953] localedata: lt_LT: change currency symbol to the euro
1948 [18956] math: powf inaccuracy
1949 [18961] math: [i386] exp missing underflows
1950 [18966] math: [i386] exp10 missing underflows
1951 [18967] math: math.h XSI POSIX namespace (gamma, isnan, scalb)
1952 [18969] build: multiple string test failures due to missing locale
1954 [18970] libc: Reference of pthread_setcancelstate in libc.a
1955 [18977] math: float / long double Bessel functions not in XSI POSIX
1956 [18980] math: i386 libm functions return with excess range and precision
1957 [18981] math: i386 scalb*, ldexp return with excess range and precision
1958 [18982] stdio: va_list and vprintf
1959 [18985] time: Passing out of range data to strftime() causes a segfault
1961 [19003] math: [x86_64] fma4 version of pow inappropriate contraction
1962 [19007] libc: FAIL: elf/check-localplt with -z now and binutils 2.26
1963 [19012] locale: iconv_open leaks memory on error path
1964 [19016] math: clog, clog10 inaccuracy
1965 [19018] nptl: Mangle function pointers in tls_dtor_list
1966 [19032] math: [i386] acosh (-qNaN) spurious "invalid" exception
1967 [19046] math: ldbl-128 / ldbl-128ibm lgamma bad overflow handling
1968 [19048] malloc: malloc: arena free list can become cyclic, increasing
1970 [19049] math: [powerpc] erfc incorrect zero sign
1971 [19050] math: [powerpc] log* incorrect zero sign
1972 [19058] math: [x86_64] Link fail with -fopenmp and -flto
1973 [19059] math: nexttoward overflow incorrect in non-default rounding modes
1974 [19071] math: ldbl-96 lroundl incorrect just below powers of 2
1975 [19074] network: Data race in _res_hconf_reorder_addrs
1976 [19076] math: [ldbl-128ibm] log1pl (-1) wrong sign of infinity
1977 [19077] math: [ldbl-128ibm] logl (1) incorrect sign of zero result
1978 [19078] math: [ldbl-128ibm] expl overflow incorrect in non-default
1980 [19079] math: dbl-64/wordsize-64 lround based on llround incorrect for
1982 [19085] math: ldbl-128 lrintl, lroundl missing exceptions for 32-bit long
1983 [19086] manual: posix_fallocate64 documented argument order is wrong.
1984 [19088] math: lround, llround missing exceptions close to overflow
1986 [19094] math: lrint, llrint missing exceptions close to overflow threshold
1987 [19095] math: dbl-64 lrint incorrect for 64-bit long
1988 [19122] dynamic-link: Unnecessary PLT relocations in librtld.os
1989 [19124] dynamic-link: ld.so failed to build with older assmebler
1990 [19125] math: [powerpc32] llroundf, llround incorrect exceptions
1991 [19129] dynamic-link: [arm] Concurrent lazy TLSDESC resolution can crash
1992 [19134] math: [powerpc32] lround, lroundf spurious exceptions
1993 [19137] libc: i386/epoll_pwait.S doesn't support cancellation
1994 [19143] nptl: Remove CPU set size checking from sched_setaffinity,
1995 pthread_setaffinity_np
1996 [19156] math: [ldbl-128] j0l spurious underflows
1997 [19164] nptl: tst-getcpu fails with many possible CPUs
1998 [19168] math: math/test-ildoubl and math/test-ldouble failure
1999 [19174] nptl: PowerPC: TLE enabled pthread mutex performs poorly.
2000 [19178] dynamic-link: ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA confuses
2002 [19181] math: [i386/x86_64] fesetenv (FE_DFL_ENV), fesetenv
2003 (FE_NOMASK_ENV) do not clear SSE exceptions
2004 [19182] malloc: malloc deadlock between ptmalloc_lock_all and
2005 _int_new_arena/reused_arena
2006 [19189] math: [ldbl-128] log1pl (-qNaN) spurious "invalid" exception
2007 [19201] math: dbl-64 remainder incorrect sign of zero result
2008 [19205] math: bits/math-finite.h conditions do not match math.h and
2010 [19209] math: bits/math-finite.h wrongly maps ldexp to scalbn
2011 [19211] math: lgamma functions do not set signgam for -ffinite-math-only
2012 for C99-based standards
2013 [19212] libc: features.h not -Wundef clean
2014 [19213] math: [i386/x86_64] log* (1) incorrect zero sign for -ffinite-
2016 [19214] libc: Family and model identification for AMD CPU's are incorrect.
2017 [19219] libc: GLIBC build fails for ia64 with missing __nearbyintl
2018 [19228] math: [powerpc] nearbyint wrongly clears "inexact", leaves traps
2020 [19235] math: [powerpc64] lround, lroundf, llround, llroundf spurious
2021 "inexact" exceptions
2022 [19238] math: [powerpc] round, roundf spurious "inexact" for integer
2024 [19242] libc: strtol incorrect in Turkish locales
2025 [19243] malloc: reused_arena can pick an arena on the free list, leading
2026 to an assertion failure and reference count corruption
2027 [19253] time: tzset() ineffective when temporary TZ did not include DST
2029 [19266] math: strtod ("NAN(I)") incorrect in Turkish locales
2030 [19270] math: [hppa] Shared libm missing __isnanl
2031 [19285] libc: [hppa] sysdeps/unix/sysv/linux/hppa/bits/mman.h: missing
2032 MAP_HUGETLB and MAP_STACK defines
2033 [19313] nptl: Wrong __cpu_mask for x32
2034 [19347] libc: grantpt: try to force a specific gid even without pt_chown
2035 [19349] math: [ldbl-128ibm] tanhl inaccurate for small arguments
2036 [19350] math: [ldbl-128ibm] sinhl spurious overflows
2037 [19351] math: [ldbl-128ibm] logl inaccurate near 1
2038 [19363] time: x32: times() return value wrongly truncates/sign extends
2040 [19367] dynamic-link: Improve branch prediction on Silvermont
2041 [19369] network: Default domain name not reset by res_ninit when "search"
2042 / "domain" entry is removed from resolv.conf
2043 [19375] math: powerpc: incorrect results for POWER7 logb with negative
2045 [19385] localedata: bg_BG: time separator should be colon, not comma
2046 [19408] libc: linux personality syscall wrapper may erroneously return an
2047 error on 32-bit architectures
2048 [19415] libc: dladdr returns wrong names on hppa
2049 [19432] libc: iconv rejects redundant escape sequences in IBM900, IBM903,
2050 IBM905, IBM907, and IBM909
2051 [19439] math: Unix98 isinf and isnan functions conflict with C++11
2052 [19443] build: build failures with -DDEBUG
2053 [19451] build: Make check fails on test-double-vlen2
2054 [19462] libc: Glibc failed to build with -Os
2055 [19465] math: Wrong code with -Os
2056 [19466] time: time/tst-mktime2.c is compiled into an infinite loop with
2058 [19467] string: Fast_Unaligned_Load needs to be enabled for Excavator core
2060 [19475] libc: Glibc 2.22 doesn't build on sparc [PATCH]
2061 [19486] math: S390: Math tests fail with "Exception Inexact set".
2062 [19529] libc: [ARM]: FAIL: stdlib/tst-makecontext
2063 [19550] libc: [mips] mmap negative offset handling inconsistent with other
2065 [19590] math: Fail to build shared objects that use libmvec.so functions.
2069 * The following bugs are resolved with this release:
2071 438, 2981, 4719, 6544, 6792, 11216, 12836, 13028, 13064, 13151, 13152,
2072 14094, 14113, 14292, 14841, 14906, 14958, 15319, 15467, 15790, 15969,
2073 16159, 16339, 16350, 16351, 16352, 16353, 16361, 16512, 16526, 16538,
2074 16559, 16560, 16704, 16783, 16850, 17053, 17090, 17195, 17269, 17293,
2075 17322, 17403, 17475, 17523, 17542, 17569, 17581, 17588, 17596, 17620,
2076 17621, 17628, 17631, 17692, 17711, 17715, 17776, 17779, 17792, 17833,
2077 17836, 17841, 17912, 17916, 17930, 17932, 17944, 17949, 17964, 17965,
2078 17967, 17969, 17977, 17978, 17987, 17991, 17996, 17998, 17999, 18007,
2079 18019, 18020, 18029, 18030, 18032, 18034, 18036, 18038, 18039, 18042,
2080 18043, 18046, 18047, 18049, 18068, 18078, 18080, 18093, 18100, 18104,
2081 18110, 18111, 18116, 18125, 18128, 18134, 18138, 18185, 18196, 18197,
2082 18206, 18210, 18211, 18217, 18219, 18220, 18221, 18234, 18244, 18245,
2083 18247, 18287, 18319, 18324, 18333, 18346, 18371, 18383, 18397, 18400,
2084 18409, 18410, 18412, 18418, 18422, 18434, 18444, 18457, 18468, 18469,
2085 18470, 18479, 18483, 18495, 18496, 18497, 18498, 18502, 18507, 18508,
2086 18512, 18513, 18519, 18520, 18522, 18527, 18528, 18529, 18530, 18532,
2087 18533, 18534, 18536, 18539, 18540, 18542, 18544, 18545, 18546, 18547,
2088 18549, 18553, 18557, 18558, 18569, 18583, 18585, 18586, 18592, 18593,
2089 18594, 18602, 18612, 18613, 18619, 18633, 18635, 18641, 18643, 18648,
2090 18657, 18676, 18694, 18696, 18887.
2092 * Cache information can be queried via sysconf() function on s390 e.g. with
2093 _SC_LEVEL1_ICACHE_SIZE as argument.
2095 * A buffer overflow in gethostbyname_r and related functions performing DNS
2096 requests has been fixed. If the NSS functions were called with a
2097 misaligned buffer, the buffer length change due to pointer alignment was
2098 not taken into account. This could result in application crashes or,
2099 potentially arbitrary code execution, using crafted, but syntactically
2100 valid DNS responses. (CVE-2015-1781)
2102 * The time zone file parser has been made more robust against crafted time
2103 zone files, avoiding heap buffer overflows related to the processing of
2104 the tzh_ttisstdcnt and tzh_ttisgmtcnt fields, and a stack overflow due to
2105 large time zone data files. Overly long time zone specifiers in the TZ
2106 variable no longer result in stack overflows and crashes.
2108 * A powerpc and powerpc64 optimization for TLS, similar to TLS descriptors
2109 for LD and GD on x86 and x86-64, has been implemented. You will need
2110 binutils-2.24 or later to enable this optimization.
2112 * Character encoding and ctype tables were updated to Unicode 7.0.0, using
2113 new generator scripts contributed by Pravin Satpute and Mike FABIAN (Red
2114 Hat). These updates cause user visible changes, such as the fix for bug
2117 * CVE-2014-8121 The NSS backends shared internal state between the getXXent
2118 and getXXbyYY NSS calls for the same database, causing a denial-of-service
2119 condition in some applications.
2121 * Added vector math library named libmvec with the following vectorized x86_64
2122 implementations: cos, cosf, sin, sinf, sincos, sincosf, log, logf, exp, expf,
2124 The library can be disabled with --disable-mathvec. Use of the functions is
2125 enabled with -fopenmp -ffast-math starting from -O1 for GCC version >= 4.9.0.
2126 Shared library libmvec.so is linked in as needed when using -lm (no need to
2127 specify -lmvec explicitly for not static builds).
2128 Visit <https://sourceware.org/glibc/wiki/libmvec> for detailed information.
2130 * A new fmemopen implementation has been added with the goal of POSIX
2131 compliance. The new implementation fixes the following long-standing
2132 issues: BZ#6544, BZ#11216, BZ#12836, BZ#13151, BZ#13152, and BZ#14292. The
2133 old implementation is still present for use be by existing binaries.
2135 * The 32-bit sparc sigaction ABI was inadvertently broken in the 2.20 and 2.21
2136 releases. It has been fixed to match 2.19 and older, but binaries built
2137 against 2.20 and 2.21 might need to be recompiled. See BZ#18694.
2139 * Port to Native Client running on ARMv7-A (--host=arm-nacl).
2140 Contributed by Roland McGrath (Google).
2142 * The header <regexp.h> is deprecated, and will be removed in a future
2143 release. Use of this header will trigger a deprecation warning.
2144 Application developers should update their code to use <regex.h> instead.
2146 This header was formerly part of SUS, but was deprecated in 1994 and
2147 removed from the standard in 2001. Also, the glibc implementation
2148 leaks memory. See BZ#18681 for more details.
2152 * The following bugs are resolved with this release:
2154 6652, 10672, 12674, 12847, 12926, 13862, 14132, 14138, 14171, 14498,
2155 15215, 15378, 15884, 16009, 16418, 16191, 16469, 16576, 16617, 16618,
2156 16619, 16657, 16740, 16857, 17192, 17266, 17273, 17344, 17363, 17370,
2157 17371, 17411, 17460, 17475, 17485, 17501, 17506, 17508, 17522, 17555,
2158 17570, 17571, 17572, 17573, 17574, 17582, 17583, 17584, 17585, 17589,
2159 17594, 17601, 17608, 17616, 17625, 17630, 17633, 17634, 17635, 17647,
2160 17653, 17657, 17658, 17664, 17665, 17668, 17682, 17702, 17717, 17719,
2161 17722, 17723, 17724, 17725, 17732, 17733, 17744, 17745, 17746, 17747,
2162 17748, 17775, 17777, 17780, 17781, 17782, 17791, 17793, 17796, 17797,
2163 17801, 17803, 17806, 17834, 17844, 17848, 17868, 17869, 17870, 17885,
2166 * CVE-2015-1472 CVE-2015-1473 Under certain conditions wscanf can allocate
2167 too little memory for the to-be-scanned arguments and overflow the
2168 allocated buffer. The implementation now correctly computes the required
2169 buffer size when using malloc, and switches to malloc from alloca as
2172 * A new semaphore algorithm has been implemented in generic C code for all
2173 machines. Previous custom assembly implementations of semaphore were
2174 difficult to reason about or ensure that they were safe. The new version
2175 of semaphore supports machines with 64-bit or 32-bit atomic operations.
2176 The new semaphore algorithm is used by sem_init, sem_open, sem_post,
2177 sem_wait, sem_timedwait, sem_trywait, and sem_getvalue.
2179 * Port to Altera Nios II has been contributed by Mentor Graphics.
2181 * Optimized strcpy, stpcpy, strncpy, stpncpy, strcmp, and strncmp
2182 implementations for powerpc64/powerpc64le.
2183 Implemented by Adhemerval Zanella (IBM).
2185 * Added support for TSX lock elision of pthread mutexes on powerpc32, powerpc64
2186 and powerpc64le. This may improve lock scaling of existing programs on
2187 HTM capable systems. The lock elision code is only enabled with
2188 --enable-lock-elision=yes. Also, the TSX lock elision implementation for
2189 powerpc will issue a transaction abort on every syscall to avoid side
2190 effects being visible outside transactions.
2192 * Optimized strcpy, stpcpy, strchrnul and strrchr implementations for
2193 AArch64. Contributed by ARM Ltd.
2195 * i386 memcpy functions optimized with SSE2 unaligned load/store.
2197 * CVE-2104-7817 The wordexp function could ignore the WRDE_NOCMD flag
2198 under certain input conditions resulting in the execution of a shell for
2199 command substitution when the application did not request it. The
2200 implementation now checks WRDE_NOCMD immediately before executing the
2201 shell and returns the error WRDE_CMDSUB as expected.
2203 * CVE-2012-3406 printf-style functions could run into a stack overflow when
2204 processing format strings with a large number of format specifiers.
2206 * CVE-2014-9402 The nss_dns implementation of getnetbyname could run into an
2207 infinite loop if the DNS response contained a PTR record of an unexpected
2210 * The minimum GCC version that can be used to build this version of the GNU
2211 C Library is GCC 4.6. Older GCC versions, and non-GNU compilers, can
2212 still be used to compile programs using the GNU C Library.
2214 * The GNU C Library is now built with -Werror by default. This can be
2215 disabled by configuring with --disable-werror.
2217 * New locales: tu_IN, bh_IN, raj_IN, ce_RU.
2219 * The obsolete sigvec function has been removed. This was the original
2220 4.2BSD interface that inspired the POSIX.1 sigaction interface, which
2221 programs have been using instead for about 25 years. Of course, ABI
2222 compatibility for old binaries using sigvec remains intact.
2224 * Merged gettext 0.19.3 into the intl subdirectory. This fixes building
2225 with newer versions of bison.
2227 * Support for MIPS o32 FPXX, FP64A and FP64 ABI Extensions.
2228 The original MIPS o32 hard-float ABI requires an FPU where double-precision
2229 registers overlay two consecutive single-precision registers. MIPS32R2
2230 introduced a new FPU mode (FR=1) where double-precision registers extend the
2231 corresponding single-precision registers which is incompatible with the
2232 o32 hard-float ABI. The MIPS SIMD ASE and the MIPSR6 architecture both
2233 require the use of FR=1 making a transition necessary. New o32 ABI
2234 extensions enable users to migrate over time from the original o32 ABI
2235 through to the updated o32 FP64 ABI. To achieve this the dynamic linker now
2236 tracks the ABI of any loaded object and verifies that new objects are
2237 compatible. Mode transitions will also be requested as required and
2238 unsupportable objects will be rejected. The ABI checks include both soft and
2239 hard float ABIs for o32, n32 and n64.
2241 GCC 5 with GNU binutils 2.25 onwards:
2242 It is strongly recommended that all o32 system libraries are built using the
2243 new o32 FPXX ABI (-mfpxx) to facilitate the transition as this is compatible
2244 with the original and all new o32 ABI extensions. Configure a MIPS GCC
2245 compiler using --with-fp-32=xx to set this by default.
2249 * The following bugs are resolved with this release:
2251 6804, 9894, 11505, 12994, 13347, 13651, 14308, 14770, 15119, 15132, 15347,
2252 15514, 15698, 15804, 15894, 15946, 16002, 16064, 16095, 16194, 16198,
2253 16275, 16284, 16287, 16315, 16348, 16349, 16354, 16357, 16362, 16447,
2254 16516, 16532, 16539, 16545, 16561, 16562, 16564, 16574, 16599, 16600,
2255 16609, 16610, 16611, 16613, 16619, 16623, 16629, 16632, 16634, 16639,
2256 16642, 16648, 16649, 16670, 16674, 16677, 16680, 16681, 16683, 16689,
2257 16695, 16701, 16706, 16707, 16712, 16713, 16714, 16724, 16731, 16739,
2258 16740, 16743, 16754, 16758, 16759, 16760, 16770, 16786, 16789, 16791,
2259 16796, 16799, 16800, 16815, 16823, 16824, 16831, 16838, 16839, 16849,
2260 16854, 16876, 16877, 16878, 16882, 16885, 16888, 16890, 16892, 16912,
2261 16915, 16916, 16917, 16918, 16922, 16927, 16928, 16932, 16943, 16958,
2262 16965, 16966, 16967, 16977, 16978, 16984, 16990, 16996, 17009, 17022,
2263 17031, 17042, 17048, 17050, 17058, 17061, 17062, 17069, 17075, 17078,
2264 17079, 17084, 17086, 17088, 17092, 17097, 17125, 17135, 17137, 17150,
2265 17153, 17187, 17213, 17259, 17261, 17262, 17263, 17319, 17325, 17354.
2267 * Reverted change of ABI data structures for s390 and s390x:
2268 On s390 and s390x the size of struct ucontext and jmp_buf was increased in
2269 2.19. This change is reverted in 2.20. The introduced 2.19 symbol versions
2270 of getcontext, setjmp, _setjmp, __sigsetjmp, longjmp, _longjmp, siglongjmp
2271 are preserved pointing straight to the same implementation as the old ones.
2272 Given that, new callers will simply provide a too-big buffer to these
2273 functions. Any applications/libraries out there that embed jmp_buf or
2274 ucontext_t in an ABI-relevant data structure that have already been rebuilt
2275 against 2.19 headers will have to rebuilt again. This is necessary in any
2276 case to revert the breakage in their ABI caused by the glibc change.
2278 * Support for file description locks is added to systems running the
2279 Linux kernel. The standard file locking interfaces are extended to
2280 operate on file descriptions, not file descriptors, via the use of
2281 F_OFD_GETLK, F_OFD_SETLK, and F_OFD_SETLKW. File description locks
2282 are associated with an open file instead of a process.
2284 * Optimized strchr implementation for AArch64. Contributed by ARM Ltd.
2286 * The minimum Linux kernel version that this version of the GNU C Library
2287 can be used with is 2.6.32.
2289 * Running the testsuite no longer terminates as soon as a test fails.
2290 Instead, a file tests.sum (xtests.sum from "make xcheck") is generated,
2291 with PASS or FAIL lines for individual tests. A summary of the results is
2292 printed, including a list of failing lists, and "make check" exits with
2293 error status if there were any unexpected failures. "make check
2294 stop-on-test-failure=y" may be used to keep the old behavior.
2296 * The am33 port, which had not worked for several years, has been removed
2299 * The _BSD_SOURCE and _SVID_SOURCE feature test macros are no longer
2300 supported; they now act the same as _DEFAULT_SOURCE (but generate a
2301 warning). Except for cases where _BSD_SOURCE enabled BSD interfaces that
2302 conflicted with POSIX (support for which was removed in 2.19), the
2303 interfaces those macros enabled remain available when compiling with
2304 _GNU_SOURCE defined, with _DEFAULT_SOURCE defined, or without any feature
2305 test macros defined.
2307 * Optimized strcmp implementation for ARMv7. Contributed by ARM Ltd.
2309 * Added support for TX lock elision of pthread mutexes on s390 and s390x.
2310 This may improve lock scaling of existing programs on TX capable systems.
2311 The lock elision code is only built with --enable-lock-elision=yes and
2312 then requires a GCC version supporting the TX builtins. With lock elision
2313 default mutexes are elided via __builtin_tbegin, if the cpu supports
2314 transactions. By default lock elision is not enabled and the elision code
2317 * CVE-2014-4043 The posix_spawn_file_actions_addopen implementation did not
2318 copy the path argument. This allowed programs to cause posix_spawn to
2319 deference a dangling pointer, or use an unexpected pathname argument if
2320 the string was modified after the posix_spawn_file_actions_addopen
2323 * All supported architectures now use the main glibc sysdeps directory
2324 instead of some being in a separate "ports" directory (which was
2325 distributed separately before glibc 2.17).
2327 * The NPTL implementation of POSIX pthreads is no longer an "add-on".
2328 On configurations that support it (all Linux configurations), it's now
2329 used regardless of the --enable-add-ons switch to configure. It is no
2330 longer possible to build such configurations without pthreads support.
2332 * Locale names, including those obtained from environment variables (LANG
2333 and the LC_* variables), are more tightly checked for proper syntax.
2334 setlocale will now fail (with EINVAL) for locale names that are overly
2335 long, contain slashes without starting with a slash, or contain ".." path
2336 components. (CVE-2014-0475) Previously, some valid locale names were
2337 silently replaced with the "C" locale when running in AT_SECURE mode
2338 (e.g., in a SUID program). This is no longer necessary because of the
2341 * On x86-64, the dynamic linker's lazy-binding support is now compatible
2342 with application code using Intel MPX instructions. (With all previous
2343 versions, the MPX register state could be clobbered when making calls
2344 into or out of a shared library.) Note that while the new dynamic
2345 linker is compatible with all known x86 hardware whether or not it
2346 supports Intel MPX, some x86 instruction-set emulators might fail to
2347 handle the new instruction encodings. This is known to affect Valgrind
2348 versions up through 3.9 (but will be fixed in the forthcoming 3.10
2349 release), and might affect other tools that do instruction emulation.
2351 * Support for loadable gconv transliteration modules has been removed.
2352 The support for transliteration modules has been non-functional for
2353 over a decade, and the removal is prompted by security defects. The
2354 normal gconv conversion modules are still supported. Transliteration
2355 with //TRANSLIT is still possible, and the //IGNORE specifier
2356 continues to be supported. (CVE-2014-5119)
2358 * Decoding a crafted input sequence in the character sets IBM933, IBM935,
2359 IBM937, IBM939, IBM1364 could result in an out-of-bounds array read,
2360 resulting a denial-of-service security vulnerability in applications which
2361 use functions related to iconv. (CVE-2014-6040)
2365 * The following bugs are resolved with this release:
2367 156, 387, 431, 762, 832, 926, 2801, 4772, 6786, 6787, 6807, 6810, 6981,
2368 7003, 9721, 9954, 10253, 10278, 11087, 11157, 11214, 12100, 12486, 12751,
2369 12986, 13028, 13982, 13985, 14029, 14032, 14120, 14143, 14155, 14286,
2370 14547, 14699, 14752, 14782, 14876, 14910, 15004, 15048, 15073, 15089,
2371 15128, 15218, 15268, 15277, 15308, 15362, 15374, 15400, 15425, 15427,
2372 15483, 15522, 15531, 15532, 15593, 15601, 15608, 15609, 15610, 15632,
2373 15640, 15670, 15672, 15680, 15681, 15723, 15734, 15735, 15736, 15748,
2374 15749, 15754, 15760, 15763, 15764, 15797, 15799, 15825, 15843, 15844,
2375 15846, 15847, 15849, 15850, 15855, 15856, 15857, 15859, 15867, 15886,
2376 15887, 15890, 15892, 15893, 15895, 15897, 15901, 15905, 15909, 15915,
2377 15917, 15919, 15921, 15923, 15939, 15941, 15948, 15963, 15966, 15968,
2378 15985, 15988, 15997, 16032, 16034, 16036, 16037, 16038, 16041, 16046,
2379 16055, 16071, 16072, 16074, 16077, 16078, 16103, 16112, 16143, 16144,
2380 16146, 16150, 16151, 16153, 16167, 16169, 16172, 16195, 16214, 16245,
2381 16271, 16274, 16283, 16289, 16293, 16314, 16316, 16330, 16337, 16338,
2382 16356, 16365, 16366, 16369, 16372, 16375, 16379, 16381, 16384, 16385,
2383 16386, 16387, 16390, 16394, 16398, 16400, 16407, 16408, 16414, 16430,
2384 16431, 16453, 16474, 16506, 16510, 16529
2386 * Slovenian translations for glibc messages have been contributed by the
2387 Translation Project's Slovenian team of translators.
2389 * The public headers no longer use __unused nor __block. This change is to
2390 support compiling programs that are derived from BSD sources and use
2391 __unused internally, and to support compiling with Clang's -fblock
2392 extension which uses __block.
2394 * CVE-2012-4412 The strcoll implementation caches indices and rules for
2395 large collation sequences to optimize multiple passes. This cache
2396 computation may overflow for large collation sequences and may cause a
2397 stack or buffer overflow. This is now fixed to use a slower algorithm
2398 which does not use a cache if there is an integer overflow.
2400 * CVE-2012-4424 The strcoll implementation uses malloc to cache indices and
2401 rules for large collation sequences to optimize multiple passes and falls
2402 back to alloca if malloc fails, resulting in a possible stack overflow.
2403 The implementation now falls back to an uncached collation sequence lookup
2406 * CVE-2013-4788 The pointer guard used for pointer mangling was not
2407 initialized for static applications resulting in the security feature
2408 being disabled. The pointer guard is now correctly initialized to a
2409 random value for static applications. Existing static applications need
2410 to be recompiled to take advantage of the fix (bug 15754).
2412 * CVE-2013-4237 The readdir_r function could write more than NAME_MAX bytes
2413 to the d_name member of struct dirent, or omit the terminating NUL
2414 character. (Bugzilla #14699).
2416 * CVE-2013-4332 The pvalloc, valloc, memalign, posix_memalign and
2417 aligned_alloc functions could allocate too few bytes or corrupt the
2418 heap when passed very large allocation size values (Bugzilla #15855,
2421 * CVE-2013-4458 Stack overflow in getaddrinfo with large number of results
2422 for AF_INET6 has been fixed (Bugzilla #16072).
2424 * New locales: ak_GH, anp_IN, ar_SS, cmn_TW, hak_TW, lzh_TW, nan_TW, pap_AW,
2425 pap_CW, quz_PE, the_NP.
2427 * Substantially revised locales: gd_GB, ht_HT
2429 * The LC_ADDRESS field was updated to support country_car for almost all
2432 * ISO 1427 definitions were updated.
2434 * ISO 3166 definitions were updated.
2436 * The localedef utility now supports --big-endian and --little-endian
2437 command-line options to generate locales for a different system from that
2438 for which the C library was built.
2440 * Binary locale files now only depend on the endianness of the system for
2441 which they are generated and not on other properties of that system. As a
2442 consequence, binary files generated with new localedef may be incompatible
2443 with old versions of the GNU C Library, and binary files generated with
2444 old localedef may be incompatible with this version of the GNU C Library,
2445 in the following circumstances:
2447 + Locale files may be incompatible on m68k systems.
2449 + Locale archive files (but not separate files for individual locales) may
2450 be incompatible on systems where plain "char" is signed.
2452 * The configure option --disable-versioning has been removed. Builds with
2453 --disable-versioning had not worked for several years.
2455 * ISO 639 definitions were updated for Chiga (cgg) and Chinese (gan, hak, czh,
2456 cjy, lzh, cmn, mnp, cdo, czo, cpx, wuu, hsn, yue).
2458 * SystemTap probes for malloc have been introduced.
2460 * SystemTap probes for slow multiple precision fallback paths of
2461 transcendental functions have been introduced.
2463 * Support for powerpc64le has been added.
2465 * The soft-float powerpc port now supports e500 processors.
2467 * Support for STT_GNU_IFUNC symbols added for ppc32/power4+ and ppc64.
2469 * A new feature test macro _DEFAULT_SOURCE is available to enable the same
2470 set of header declarations that are enabled by default, even when other
2471 feature test macros or compiler options such as -std=c99 would otherwise
2472 disable some of those declarations.
2474 * The _BSD_SOURCE feature test macro no longer enables BSD interfaces that
2475 conflict with POSIX. The libbsd-compat library (which was a dummy library
2476 that did nothing) has also been removed.
2478 * Preliminary documentation about Multi-Thread, Async-Signal and
2479 Async-Cancel Safety has been added.
2481 * Change of ABI data structures for s390 and s390x:
2482 On s390 and s390x the size of struct ucontext and jmp_buf was increased to
2483 allow for future hardware extensions. All library functions that accept or
2484 return these structures were versioned in order to provide backwards
2485 compatibility with existing code. However, not all cases can be handled
2486 with interface versioning. If an application embeds either structure into
2487 another structure and that structure is passed to another compilation unit
2488 which expects the newer larger-sized object then any access to the new
2489 fields will result in undefined behaviour. Similarly any access to fields
2490 that were moved by the enlarging of the structures will cause undefined
2491 behaviour. To fix the undefined behaviour all objects that are part of the
2492 final in-memory image must agree on the size of structures, and this may
2493 require recompilation.
2497 * The following bugs are resolved with this release:
2499 2546, 2560, 5159, 6809, 7006, 10060, 10062, 10283, 10357, 10686, 11120,
2500 11561, 12310, 12387, 12492, 12515, 12723, 13550, 13889, 13951, 13988,
2501 14142, 14176, 14200, 14256, 14280, 14293, 14317, 14327, 14478, 14496,
2502 14582, 14686, 14812, 14888, 14894, 14907, 14908, 14909, 14920, 14941,
2503 14952, 14964, 14981, 14982, 14985, 14991, 14994, 14996, 15000, 15003,
2504 15006, 15007, 15014, 15020, 15022, 15023, 15036, 15054, 15055, 15062,
2505 15078, 15084, 15085, 15086, 15100, 15160, 15214, 15221, 15232, 15234,
2506 15283, 15285, 15287, 15304, 15305, 15307, 15309, 15327, 15330, 15335,
2507 15336, 15337, 15339, 15342, 15346, 15359, 15361, 15366, 15380, 15381,
2508 15394, 15395, 15405, 15406, 15409, 15416, 15418, 15419, 15423, 15424,
2509 15426, 15429, 15431, 15432, 15441, 15442, 15448, 15465, 15480, 15485,
2510 15488, 15490, 15492, 15493, 15497, 15506, 15529, 15536, 15553, 15577,
2511 15583, 15618, 15627, 15631, 15654, 15655, 15666, 15667, 15674, 15711,
2514 * CVE-2013-2207 Incorrectly granting access to another user's pseudo-terminal
2515 has been fixed by disabling the use of pt_chown (Bugzilla #15755).
2516 Distributions can re-enable building and using pt_chown via the new configure
2517 option `--enable-pt_chown'. Enabling the use of pt_chown carries with it
2518 considerable security risks and should only be used if the distribution
2519 understands and accepts the risks.
2521 * CVE-2013-0242 Buffer overrun in regexp matcher has been fixed (Bugzilla
2524 * CVE-2013-1914 Stack overflow in getaddrinfo with many results has been
2525 fixed (Bugzilla #15330).
2527 * Add support for calling C++11 thread_local object destructors on thread
2528 and program exit. This needs compiler support for offloading C++11
2529 destructor calls to glibc.
2531 * Improved worst case performance of libm functions with double inputs and
2534 * Support for priority inherited mutexes in pthread condition variables on
2535 non-x86 architectures.
2537 * Port to Xilinx MicroBlaze contributed by David Holsgrove.
2539 * Optimized string functions for AArch64. Implemented by Marcus Shawcroft.
2541 * Optimized string functions for ARM. Implemented by Will Newton and
2544 * Optimized libm functions for SPARC. Implemented by David S. Miller.
2546 * Support for building more of ARM glibc as Thumb-2 code. Implemented by
2549 * Support for building most of MIPS glibc as MIPS16 code. Implemented by
2550 Chung-Lin Tang, Maciej W. Rozycki and Maxim Kuvyrkov.
2552 * Added a benchmark framework to track performance of functions in glibc.
2554 * New <math.h> macro named issignaling to check for a signaling NaN (sNaN).
2555 It is based on draft TS 18661 and currently enabled as a GNU extension.
2557 * On Linux, the clock function now uses the clock_gettime system call
2558 for improved precision, rather than old times system call.
2560 * Added support for version-3 tz format files. This is needed when using
2561 the tz database release 2013e or later, and affects a few unusual cases --
2562 currently only TZ='America/Godthab' for time stamps after 2037.
2564 * Added new API functions pthread_getattr_default_np and
2565 pthread_setattr_default_np to get and set the default pthread
2566 attributes of a process.
2568 * Added support for TSX lock elision for pthread mutexes on i386 and x86-64.
2569 This may improve lock scaling of existing programs on TSX capable systems.
2570 When the --enable-lock-elision=yes parameter is specified at configure
2571 time lock elision will be enabled by default for all PTHREAD_MUTEX_DEFAULT
2574 * Support added for AT_HWCAP2 (to coincide with Linux kernel 3.10
2575 availability). Implemented by Ryan S. Arnold.
2577 * Support added for POWER8 platform. Implemented by Ryan S. Arnold.
2581 * The following bugs are resolved with this release:
2583 1349, 3439, 3479, 3665, 5044, 5246, 5298, 5400, 6530, 6677, 6778, 6808,
2584 9685, 9914, 10014, 10038, 10114, 10191, 10631, 10873, 11438, 11607, 11638,
2585 11741, 12140, 13013, 13412, 13542, 13601, 13603, 13604, 13629, 13679,
2586 13696, 13698, 13717, 13741, 13759, 13761, 13763, 13881, 13939, 13950,
2587 13952, 13966, 14042, 14047, 14090, 14150, 14151, 14152, 14154, 14157,
2588 14166, 14173, 14195, 14197, 14237, 14246, 14251, 14252, 14283, 14298,
2589 14303, 14307, 14328, 14331, 14336, 14337, 14347, 14349, 14368, 14376,
2590 14417, 14447, 14459, 14476, 14477, 14501, 14505, 14510, 14516, 14518,
2591 14519, 14530, 14532, 14538, 14543, 14544, 14545, 14557, 14562, 14568,
2592 14576, 14579, 14583, 14587, 14595, 14602, 14610, 14621, 14638, 14645,
2593 14648, 14652, 14660, 14661, 14669, 14672, 14683, 14694, 14716, 14719,
2594 14743, 14767, 14783, 14784, 14785, 14793, 14796, 14797, 14801, 14803,
2595 14805, 14807, 14811, 14815, 14821, 14822, 14824, 14828, 14831, 14833,
2596 14835, 14838, 14856, 14863, 14865, 14866, 14868, 14869, 14871, 14872,
2597 14879, 14889, 14893, 14898, 14914.
2599 * Optimization of memcpy for MIPS.
2601 * CVE-2011-4609 svc_run() produces high cpu usage when accept fails with
2602 EMFILE has been fixed (Bugzilla #14889).
2604 * The add-on ports collection is now distributed in the "ports" subdirectory
2605 of the main GNU C Library distribution, rather than separately.
2607 * Port to ARM AArch64 contributed by Linaro.
2609 * Support for STT_GNU_IFUNC symbols added for s390 and s390x.
2610 Optimized versions of memcpy, memset, and memcmp added for System z10 and
2612 Implemented by Andreas Krebbel.
2614 * The new function secure_getenv allows secure access to the environment,
2615 returning NULL if running in a SUID/SGID process. This function replaces
2616 the internal function __secure_getenv.
2618 * SystemTap static probes have been added into the dynamic linker.
2619 Implemented by Gary Benson.
2621 * Optimizations of string functions strstr, strcasestr and memmem.
2622 Implemented by Maxim Kuvyrkov.
2624 * The minimum Linux kernel version that this version of the GNU C Library
2625 can be used with is 2.6.16.
2627 * Optimizations of string functions memchr, wcschr, wcscpy, and wcsrchr for
2628 powerpc POWER7. Implemented by Will Schmidt.
2630 * New configure option --disable-nscd builds the C library such that it
2631 never attempts to contact the Name Service Caching Daemon (nscd).
2632 New configure option --disable-build-nscd avoids building nscd itself;
2633 this is the default if --disable-nscd is used.
2635 * Improved support for cross-compilation, including cross-testing and
2636 bootstrap builds without a previously built glibc.
2638 * Several testsuite tests are now able to test multiple IFUNC variants of an
2639 interface, rather than just testing the one that would be chooen by
2642 * New configure options --with-bugurl and --with-pkgversion, for
2643 distributors to use to embed their bug-reporting and package version
2644 information in --help and --version output.
2646 * The ttyname and ttyname_r functions on Linux now fall back to searching for
2647 the tty file descriptor in /dev/pts or /dev if /proc is not available. This
2648 allows creation of chroots without the procfs mounted on /proc.
2650 * The `crypt' function now fails if passed salt bytes that violate the
2651 specification for those values. On Linux, the `crypt' function will
2652 consult /proc/sys/crypto/fips_enabled to determine if "FIPS mode" is
2653 enabled, and fail on encrypted strings using the MD5 or DES algorithm
2654 when the mode is enabled.
2656 * The `clock_*' suite of functions (declared in <time.h>) is now available
2657 directly in the main C library. Previously it was necessary to link with
2658 -lrt to use these functions. This change has the effect that a
2659 single-threaded program that uses a function such as `clock_gettime' (and
2660 is not linked with -lrt) will no longer implicitly load the pthreads
2661 library at runtime and so will not suffer the overheads associated with
2662 multi-thread support in other code such as the C++ runtime library.
2664 * New locales: ayc_PE, doi_IN, ia_FR, mni_IN, nhn_MX, niu_NU, niu_NZ,
2669 * The following bugs are resolved with this release:
2671 174, 208, 350, 369, 411, 706, 766, 2074, 2541, 2547, 2548, 2550, 2551,
2672 2552, 2553, 2554, 2562, 2563, 2565, 2566, 2570, 2576, 2636, 2678, 3335,
2673 3440, 3748, 3768, 3866, 3868, 3906, 3976, 3992, 4026, 4108, 4596, 4822,
2674 5077, 5461, 5805, 5993, 6471, 6486, 6578, 6649, 6730, 6770, 6794, 6884,
2675 6890, 6894, 6895, 6907, 6911, 6959, 7064, 9739, 9902, 10110, 10135, 10140,
2676 10153, 10210, 10254, 10346, 10375, 10545, 10716, 10846, 10882, 11174,
2677 11261, 11322, 11365, 11451, 11494, 11521, 11677, 11837, 11959, 12047,
2678 12097, 12193, 12194, 12297, 12298, 12301, 12340, 12354, 12416, 12495,
2679 13058, 13223, 13361, 13525, 13526, 13527, 13528, 13529, 13530, 13531,
2680 13532, 13533, 13547, 13551, 13552, 13553, 13555, 13556, 13559, 13563,
2681 13566, 13576, 13579, 13583, 13592, 13594, 13613, 13618, 13637, 13656,
2682 13658, 13673, 13691, 13695, 13704, 13705, 13706, 13718, 13726, 13738,
2683 13739, 13743, 13750, 13758, 13760, 13761, 13775, 13786, 13787, 13792,
2684 13806, 13824, 13840, 13841, 13844, 13846, 13848, 13851, 13852, 13854,
2685 13871, 13872, 13873, 13879, 13882, 13883, 13884, 13885, 13886, 13892,
2686 13895, 13908, 13910, 13911, 13912, 13913, 13914, 13915, 13916, 13917,
2687 13918, 13919, 13920, 13921, 13922, 13923, 13924, 13926, 13927, 13928,
2688 13938, 13941, 13942, 13954, 13955, 13956, 13963, 13967, 13968, 13970,
2689 13973, 13979, 13983, 13986, 13996, 14012, 14027, 14033, 14034, 14036,
2690 14040, 14043, 14044, 14048, 14049, 14050, 14053, 14055, 14059, 14064,
2691 14075, 14080, 14083, 14103, 14104, 14109, 14112, 14117, 14122, 14123,
2692 14134, 14153, 14183, 14188, 14199, 14210, 14218, 14229, 14241, 14273,
2695 * Support for the x32 ABI on x86-64 added. The x32 target is selected by
2696 configuring glibc with:
2697 BUILD_CC='gcc' CC='gcc -mx32' CXX='g++ -mx32'
2698 Visit <http://sites.google.com/site/x32abi/> for more x32 ABI info.
2699 Implemented by H.J. Lu.
2703 + define static_assert
2705 + do not declare gets
2707 + declare at_quick_exit and quick_exit also for ISO C11
2709 + aligned_alloc. NB: The code is deliberately allows the size parameter
2710 to not be a multiple of the alignment. This is a moronic requirement
2711 in the standard but it is only a requirement on the caller, not the
2714 + timespec_get added
2716 + uchar.h support added
2718 + CMPLX, CMPLXF, CMPLXL added
2720 Implemented by Ulrich Drepper.
2722 * Support for the IA-64 has been moved to ports.
2724 * Remove support for anything but ELF binary format
2726 * Checking versions of poll, ppoll added.
2727 Implemented by Ulrich Drepper.
2729 * More generic and 64-bit performance optimizations to math functions.
2730 Implemented by Ulrich Drepper.
2732 * New configure option --enable-obsolete-rpc makes the deprecated RPC
2733 headers and functions available at compile time as they were before
2734 version 2.14. This option will be removed at some time in the future
2735 after the TI-RPC library becomes fully sufficient for the needs of
2736 existing applications.
2738 * Compatibility code for Linux kernel versions before 2.4 has been removed.
2739 Note that glibc is not expected to work with any Linux kernel version
2742 * New header <sys/auxv.h> and function getauxval allowing easy access to
2743 the AT_* key-value pairs passed from the Linux kernel. The header also
2744 defines the HWCAP_* bits associated with the AT_HWCAP key.
2746 * New locales: mag_IN
2748 * New configure option --enable-systemtap builds SystemTap static probes
2749 into libc for setjmp and longjmp and into libpthread for various operations.
2750 So far the setjmp/longjmp probes and some of the libpthread probes are
2751 provided only for i*86 and x86_64.
2752 Implemented by Roland McGrath and Rayson Ho.
2754 * Optimized expf for x86-32 and x86-64. Implemented by Liubov Dmitrieva.
2756 * More optimized functions for PowerPC. Implemented by Adhemerval Zanella
2759 * More optimized functions for SPARC. Implemented by David S. Miller.
2761 * Improved support for cross-compilation, especially bootstrap builds
2762 without a previously built glibc.
2764 * Ports for the TILE-Gx and TILEPro families of processors. Contributed by
2765 Chris Metcalf from Tilera.
2767 * Support for the old ARM ABI has been removed from ports. Only the EABI is
2768 now supported for ARM processors.
2770 * The hard-float variant of the ARM EABI now uses /lib/ld-linux-armhf.so.3
2771 as the name of the dynamic linker, to distinguish it from the
2772 /lib/ld-linux.so.3 dynamic linker used for the base EABI.
2774 * Support for CRIS has been removed from ports.
2776 * A new class of installed header has been documented for low-level
2777 platform-specific functionality. PowerPC added the first instance with a
2778 function to provide time base register access. Contributed by Tulio
2779 Magno Quites Machado Filho.
2781 * ix86 configurations now install header files that are consistent with
2782 what x86-64 configurations install. These same header files can be used
2783 for -m32, -m64, or -mx32 builds.
2784 Contributed by H.J. Lu.
2786 * Math library bug fixes. A thorough audit of all open math library bugs was
2787 conducted by Joseph Myers. Significant progress was made on many math
2788 library bugs resulting in more accurate exceptions and function results.
2789 Many thanks to all those that contributed including Andreas Jaeger for his
2790 patch review and work on the x87 trigonometric instruction issues.
2792 * Timezone data is no longer installed. Timezone-related binaries and scripts
2793 will continue to be installed. Users should obtain their timezone data from
2794 their distribution provider or from the tzdata package at
2795 <ftp://munnari.oz.au/pub/>.
2799 * The following bugs are resolved with this release:
2801 6779, 6783, 9696, 10103, 10709, 11589, 11929, 12403, 12786, 12840, 12847,
2802 12868, 12852, 12874, 12885, 12892, 12906, 12907, 12922, 12935, 12962,
2803 13007, 13021, 13061, 13062, 13067, 13068, 13085, 13088, 13090, 13092,
2804 13096, 13114, 13118, 13123, 13134, 13138, 13147, 13150, 13166, 13179,
2805 13185, 13189, 13192, 13268, 13276, 13282, 13291, 13305, 13328, 13335,
2806 13337, 13344, 13358, 13367, 13413, 13416, 13423, 13439, 13446, 13472,
2807 13484, 13506, 13515, 13523, 13524, 13538, 13540
2809 * New program pldd to list loaded object of a process
2810 Implemented by Ulrich Drepper.
2812 * Add nss_db support back to glibc. No more dependency on Berkeley db
2813 and support for initgroups lookups.
2814 Implemented by Ulrich Drepper.
2816 * Optimized strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-32.
2817 Contributed by HJ Lu.
2819 * Improved strcpy, strncpy, stpcpy, stpncpy for SSE2 and SSSE3 on x86-64.
2820 Contributed by HJ Lu.
2822 * Optimized strcat, strncat on x86-64 and optimized wcscmp, wcslen, strnlen
2823 on x86-32 and x86-64.
2824 Contributed by Liubov Dmitrieva.
2826 * Optimized strchr and strrchr for SSE on x86-32.
2827 Contributed by Liubov Dmitrieva.
2829 * Optimized memchr, memrchr, rawmemchr, memcmp, wmemcmp, wcschr, wcscpy
2830 for x86-64 and x86-32.
2831 Contributed by Liubov Dmitrieva.
2833 * New interfaces: scandirat, scandirat64
2834 Implemented by Ulrich Drepper.
2836 * Checking versions of FD_SET, FD_CLR, and FD_ISSET added.
2837 Implemented by Ulrich Drepper.
2839 * nscd now also caches the netgroup database.
2840 Implemented by Ulrich Drepper.
2842 * Integrate libm with gcc's -ffinite-math-only option.
2843 Implemented by Ulrich Drepper.
2845 * Lots of generic, 64-bit, and x86-64-specific performance optimizations
2846 to math functions. Implemented by Ulrich Drepper.
2848 * Optimized strcasecmp and strncasecmp for AVX on x86-64.
2849 Implemented by Ulrich Drepper.
2851 * New Linux interfaces: process_vm_readv, process_vm_writev
2853 * Optimized strcasecmp and strncasecmp for SSSE3 and SSE4.2 on x86-32.
2854 Implemented by Ulrich Drepper.
2856 * Optimized nearbyint and strcasecmp for PPC.
2857 Implemented by Adhemerval Zanella.
2859 * New locales: bho_IN, unm_US, es_CU, ta_LK
2863 * The following bugs are resolved with this release:
2865 386, 6420, 7101, 9730, 9732, 9809, 10138, 10149, 10157, 11099, 11257,
2866 11258, 11487, 11532, 11558, 11578, 11634, 11653, 11668, 11697, 11724,
2867 11781, 11799, 11820, 11837, 11857, 11884, 11892, 11895, 11901, 11945,
2868 11947, 11952, 11987, 12052, 12083, 12158, 12178, 12200, 12346, 12350,
2869 12393, 12420, 12432, 12445, 12449, 12453, 12454, 12460, 12469, 12489,
2870 12509, 12510, 12511, 12518, 12527, 12541, 12545, 12551, 12582, 12583,
2871 12587, 12597, 12601, 12611, 12625, 12626, 12631, 12650, 12653, 12655,
2872 12660, 12671, 12681, 12684, 12685, 12711, 12713, 12714, 12717, 12723,
2873 12724, 12734, 12738, 12746, 12766, 12775, 12777, 12782, 12788, 12792,
2874 12795, 12811, 12813, 12814, 12841
2876 * The RPC implementation in libc is obsoleted. Old programs keep working
2877 but new programs cannot be linked with the routines in libc anymore.
2878 Programs in need of RPC functionality must be linked against TI-RPC.
2879 The TI-RPC implementation is IPv6 enabled and there are other benefits.
2881 Visible changes of this change include (obviously) the inability to link
2882 programs using RPC functions without referencing the TI-RPC library and the
2883 removal of the RPC headers from the glibc headers.
2884 Implemented by Ulrich Drepper.
2886 * New Linux interfaces: clock_adjtime, name_to_handle_at, open_by_handle_at,
2887 syncfs, setns, sendmmsg
2889 * New locales: os_RU, bem_ZA, en_ZA, ff_SN, sw_KE, sw_TZ, lb_LU, wae_CH,
2890 yue_HK, lij_IT, mhr_RU
2892 * New encodings: CP770, CP771, CP772, CP773, CP774
2894 * New program sotruss to trace calls through PLTs
2895 Implemented by Ulrich Drepper.
2897 * The malloc hook implementation is marked deprecated and will be removed
2898 from the default implementation in the next version. The design never
2899 worked ever since the introduction of threads. Even programs which do
2900 not create threads themselves can use multiple threads created internally.
2904 * The following bugs are resolved with this release:
2906 3268, 6812, 7066, 10085, 10484, 10563, 10851, 11149, 11155, 11577, 11611,
2907 11640, 11655, 11701, 11840, 11856, 11883, 11903, 11904, 11968, 11979,
2908 12005, 12037, 12067, 12077, 12078, 12092, 12093, 12107, 12108, 12113,
2909 12140, 12159, 12167, 12191, 12194, 12201, 12204, 12205, 12207, 12348,
2912 * New Linux interfaces: prlimit, prlimit64, fanotify_init, fanotify_mark
2914 * POWER7 optimizations: memset, memcmp, strncmp
2916 * New optimized string functions for x86-64: strnlen (SSE2),
2917 strcasecmp (SSE2, SSSE3, SSE4.2), strncasecmp (SSE2, SSSE3, SSE4.2)
2918 Implemented by Ulrich Drepper.
2922 * The following bugs are resolved with this release:
2924 3662, 4457, 5553, 10162, 10401, 10414, 10554, 10824, 10864, 10915,
2925 10918, 10936, 10939, 10958, 10968, 10969, 10972, 10992, 11000, 11001,
2926 11007, 11010, 11027, 11039, 11040, 11041, 11043, 11046, 11056, 11070,
2927 11093, 11115, 11125, 11126, 11127, 11134, 11141, 11149, 11183, 11184,
2928 11185, 11186, 11187, 11188, 11189, 11190, 11191, 11192, 11193, 11194,
2929 11200, 11230, 11235, 11242, 11254, 11258, 11271, 11272, 11276, 11279,
2930 11287, 11292, 11319, 11332, 11333, 11387, 11389, 11390, 11394, 11397,
2931 11410, 11438, 11449, 11470, 11471, 11520, 11537, 11538, 11571
2933 * New interfaces: pthread_getname_np, pthread_setname_np
2935 * New Linux interface: recvmmsg
2937 * STT_GNU_IFUNC implemented for Sparc by David Miller.
2939 * The dynamic linker now recognizes supported ABI versions from the
2940 EI_ABIVERSION field in the ELF header.
2941 Implemented by Ulrich Drepper.
2943 * New NIS mode selector ADJUNCT_AS_SHADOW. The passwd.adjunct.byname table
2944 will not be used to fill in password fields in the passwd.byname replies.
2945 Instead it is used to synthesize the shadow.byname table, should it be
2946 missing. This is a useful mode in some installations involving Solaris.
2947 Implemented by Ulrich Drepper.
2949 * New locales: kok_IN, sq_MK, cv_RU
2953 * The following bugs are resolved with this release:
2955 3270, 7094, 9924, 9986, 10011, 10085, 10107, 10159, 10162, 10166, 10183,
2956 10195, 10196, 10200, 10203, 10207, 10211, 10213, 10217, 10221, 10229,
2957 10262, 10286, 10312, 10315, 10319, 10349, 10360, 10391, 10402, 10416,
2958 10418, 10422, 10425, 10446, 10448, 10452, 10540, 10553, 10560, 10564,
2959 10609, 10643, 10692, 10713, 10717, 10730, 10731, 10742, 10780, 10784,
2960 10789, 10817, 10823, 10840, 10847
2962 * New interfaces: execvpe, pthread_sigqueue, mkstemps, mkstemps64, mkostemps,
2964 Implemented by Ulrich Drepper.
2966 * Checking version of longjmp added that fails if an uninitialized stack
2967 frame would be created. Implemented by Ulrich Drepper.
2969 * STT_GNU_IFUNC is now supported in static executables.
2970 Implemented by H.J. Lu.
2972 * STT_GNU_IFUNC implemented for PPC by Alan Modra.
2974 * New optimized string functions for x86-64: strstr, strcasestr, memcmp,
2975 strcspn, strpbrk, strspn, strcpy, stpcpy, strncpy, strcmp (SSE2, SSE4.2),
2976 strncmp (SSE2, SSE4.2), strchr (SSE4.2), strrchr (SSE4.2).
2977 Contributed by H.J. Lu.
2979 strlen, rawmemchr, strcmp (SSSE3), strncmp (SSSE3).
2980 Implemented by Ulrich Drepper.
2982 * New optimized string functions for x86: strlen, strcspn, strspn, strpbrk,
2984 Contributed by H.J. Lu.
2986 * Support for fma instruction in AVX on x86-64.
2987 Implemented by H.J. Lu and Ulrich Drepper.
2989 * AVX support in x86-64 auditing support in ld.so.
2990 Implemented by H.J. Lu.
2992 * STB_GNU_UNIQUE support added.
2993 Implemented by Ulrich Drepper.
2995 * Implement second fallback mode for optimized DNS lookup for even more
2996 broken environments. If two requests from the same port are not
2997 handled correctly close the socket and open a new one before sending
2998 the second request. The 'single-request-reopen' option in /etc/resolv.conf
2999 can be used to select this mode right away, instead of rediscovering the
3000 necessity is every process again.
3001 Implemented by Ulrich Drepper.
3003 * New resolver flag RES_USE_DNSSEC to enable use of verified lookup.
3004 Implemented by Adam Tkac.
3006 * Optimized iconv conversions for S390x.
3007 Implemented by Andreas Krebbel.
3009 * Using condvars with PI mutexes is now more efficient due to kernel
3010 support for requeueing to PI futexes. NPTL support added for x86-64.
3011 Implemented by Ulrich Drepper.
3013 * New locales: ps_AF, my_MM
3017 * The following bugs are resolved with this release:
3019 697, 5381, 5807, 6411, 6545, 7040, 7058, 7067, 7080, 7083, 7095, 7098,
3020 9677, 9694, 9697, 9701, 9704, 9705, 9706, 9720, 9726, 9730, 9733, 9736,
3021 9741, 9750, 9753, 9759, 9781, 9791, 9793, 9823, 9833, 9844, 9880, 9881,
3022 9891, 9893, 9895, 9913, 9916, 9920, 9948, 9955, 9957, 9963, 9985, 10052,
3023 10069, 10086, 10087, 10090, 10092, 10093, 10100, 10118, 10128
3025 * New interfaces: preadv, preadv64, pwritev, pwritev64, malloc_info
3026 Implemented by Ulrich Drepper.
3028 * New Linux interfaces: accept4, fallocate, fallocate64.
3029 Implemented by Ulrich Drepper.
3031 * Correct declarations of string function when used in C++ code. This
3032 could lead to compile errors for invalid C++ code.
3034 * XPG7/POSIX 2008 compilation environment. Many GNU-specific functions are
3037 * New POSIX 2008 interface: psiginfo
3038 Implemented by Ulrich Drepper.
3040 * New ISO C++1x interfaces: quick_exit, at_quick_exit
3041 Implemented by Ulrich Drepper.
3043 * Support for selecting between multiple function definitions at runtime
3044 using STT_GNU_IFUNC symbols. Implemented by Ulrich Drepper.
3046 * The libcrypt library can now use the hash function implementations in
3047 NSS. Implemented by Ulrich Drepper.
3049 * The malloc implementation can be compiled to be less memory efficient
3050 but higher performing in multi-threaded programs.
3051 Implemented by Ulrich Drepper.
3053 * New locales: nan_TW@latin, ks_IN
3055 * Faster strlen, strchr, strchrnul, strrchr, memchr, and rawmemchr for x86-64.
3056 Implemented by Ulrich Drepper.
3058 * Extended printf hook support. It is possible to use user-defined types
3059 and extend existing format specifiers.
3060 Implemented by Ulrich Drepper.
3062 * Handling for group shadow files has been added.
3063 Implemented by Ulrich Drepper.
3065 * DNS IPv4-IPv6 parallel lookup now deals better with broken DNS
3066 servers (the case, e.g., for some people using the built-in DNS
3067 server in ADSL modems/routers). There is a once-per-process timeout
3068 in case of a broken server. To avoid it, users can run nscd or put
3069 'options single-request' in /etc/resolv.conf.
3070 Implemented by Ulrich Drepper.
3074 * The following bugs are resolved with this release:
3076 3406, 5209, 5210, 5381, 5794, 5814, 5911, 6428, 6438, 6442, 6447, 6461,
3077 6472, 6505, 6544, 6589, 6612, 6634, 6653, 6654, 6657, 6698, 6712, 6713,
3078 6719, 6723, 6724, 6740, 6763, 6771, 6790, 6791, 6817, 6824, 6839, 6843,
3079 6845, 6856, 6857, 6860, 6867, 6875, 6919, 6920, 6942, 6947, 6955, 6966,
3080 6968, 6974, 6980, 6995, 7008, 7009, 7029
3082 * Unified lookup for getaddrinfo: IPv4 and IPv6 addresses are now looked
3083 up at the same time. Implemented by Ulrich Drepper.
3085 * TLS descriptors for LD and GD on x86 and x86-64.
3086 Implemented by Alexandre Oliva.
3088 * getaddrinfo now handles DCCP and UDPlite.
3089 Implemented by Ulrich Drepper.
3091 * New fixed-size conversion macros: htobe16, htole16, be16toh, le16toh,
3092 htobe32, htole32, be32toh, le32toh, htobe64, htole64, be64toh, le64toh.
3093 Implemented by Ulrich Drepper.
3095 * New implementation of memmem, strstr, and strcasestr which is O(n).
3096 Implemented by Eric Blake.
3098 * New Linux interfaces: inotify_init1, dup3, epoll_create1, pipe2
3100 * Implement "e" option for popen to open file descriptor with the
3101 close-on-exec flag set. Implemented by Ulrich Drepper.
3103 * Implement "b" mode for fmemopen. In this mode writes writes don't
3104 implicitly add a NUL byte and seeks from the end of the buffer really
3105 use the buffer end, not the string length as the basis.
3106 Implemented by Ulrich Drepper.
3108 * Many functions, exported and internal, now atomically set the close-on-exec
3109 flag when run on a sufficiently new kernel. Implemented by Ulrich Drepper.
3111 * Sorting rules for some Indian languages (Kannada, Gurumukhi, Telugu,
3113 Implemented by Pravin Satpute.
3115 * New locales: sd_IN, sd_IN@devanagari, ks_IN@devanagari, ht_HT, en_AG, nl_AW.
3119 * The following bugs are resolved with this release:
3121 2549, 3112, 4314, 4407, 4997, 5012, 5040, 5112, 5204, 5208, 5209, 5220,
3122 5222, 5224, 5225, 5237, 5238, 5240, 5245, 5277, 5346, 5375, 5378, 5382,
3123 5424, 5427, 5428, 5435, 5436, 5439, 5441, 5442, 5443, 5451, 5452, 5454,
3124 5455, 5463, 5464, 5465, 5475, 5477, 5478, 5520, 5541, 5545, 5553, 5600,
3125 5602, 5607, 5614, 5627, 5628, 5736, 5737, 5741, 5753, 5760, 5762, 5768,
3126 5774, 5777, 5778, 5779, 5786, 5790, 5818, 5854, 5857, 5903, 5939, 5979,
3127 5983, 5995, 5998, 6004, 6007, 6020, 6021, 6024, 6040, 6042
3129 * New locales: bo_CN, bo_IN, shs_CA.
3131 * New encoding: HP-ROMAN9, HP-GREEK8, HP-THAI8, HP-TURKISH8.
3133 * Sorting rules for some Indian languages (Devanagari and Gujarati).
3134 Implemented by Pravin Satpute.
3136 * IPV6 addresses in /etc/resolv.conf can now have a scope ID
3138 * nscd caches now all timeouts for DNS entries
3139 Implemented by Ulrich Drepper.
3141 * nscd is more efficient and wakes up less often.
3142 Implemented by Ulrich Drepper.
3144 * More checking functions: asprintf, dprintf, obstack_printf, vasprintf,
3145 vdprintf, and obstack_vprintf.
3146 Implemented by Jakub Jelinek.
3148 * Faster memset for x86-64.
3149 Implemented by Harsha Jagasia and H.J. Lu.
3151 * Faster memcpy on x86.
3152 Implemented by Ulrich Drepper.
3154 * ARG_MAX is not anymore constant on Linux. Use sysconf(_SC_ARG_MAX).
3155 Implemented by Ulrich Drepper.
3157 * Faster sqrt and sqrtf implemention for some PPC variants.
3158 Implemented by Steven Munroe.
3162 * The following bugs are resolved with this release:
3164 4125, 4126, 4178, 4359, 4407, 4512, 4514, 4525, 4554, 4556, 4557, 4566,
3165 4582, 4586, 4588, 4599, 4610, 4647, 4702, 4705, 4726, 4745, 4772, 4773,
3166 4775, 4776, 4792, 4813, 4814, 4816, 4833, 4858, 4860, 4896, 4905, 4925,
3167 4936, 4937, 4938, 4941, 4946, 4963, 4972, 5010, 5028, 5043, 5058, 5063,
3168 5071, 5103, 5104, 5112, 5113, 5184, 5186
3170 * More checking functions: fread, fread_unlocked, open*, mq_open.
3171 Implemented by Jakub Jelinek and Ulrich Drepper.
3173 * Extend fortification to C++. Implemented by Jakub Jelinek.
3175 * Implement 'm' modifier for scanf. Add stricter C99/SUS compliance
3176 by not recognizing 'a' as a modifier when those specs are requested.
3177 Implemented by Jakub Jelinek.
3179 * PPC optimizations to math and string functions.
3180 Implemented by Steven Munroe.
3182 * New interfaces: mkostemp, mkostemp64. Like mkstemp* but allow additional
3183 options to be passed. Implemented by Ulrich Drepper.
3185 * More CPU set manipulation functions. Implemented by Ulrich Drepper.
3187 * New Linux interfaces: signalfd, eventfd, eventfd_read, and eventfd_write.
3188 Implemented by Ulrich Drepper.
3190 * Handle private futexes in the NPTL implementation.
3191 Implemented by Jakub Jelinek and Ulrich Drepper.
3193 * Add support for O_CLOEXEC. Implement in Hurd. Use throughout libc.
3194 Implemented by Roland McGrath and Ulrich Drepper.
3196 * Linux/x86-64 vDSO support. Implemented by Ulrich Drepper.
3198 * SHA-256 and SHA-512 based password encryption.
3199 Implemented by Ulrich Drepper.
3201 * New locales: ber_DZ, ber_MA, en_NG, fil_PH, fur_IT, fy_DE, ha_NG, ig_NG,
3202 ik_CA, iu_CA, li_BE, li_NL, nds_DE, nds_NL, pap_AN, sc_IT, tk_TM, ug_CN,
3205 + New iconv modules: MAC-CENTRALEUROPE, ISO-8859-9E, KOI8-RU.
3206 Implemented by Ulrich Drepper.
3210 * The following bugs are resolved with this release:
3212 3156, 3213, 3285, 3291, 3306, 3313, 3320, 3322, 3325, 3326, 3334, 3348,
3213 3352, 3363, 3369, 3426, 3427, 3429, 3451, 3458, 3465, 3480, 3483, 3493,
3214 3514, 3515, 3559, 3632, 3664, 3673, 3674, 3745, 3747, 3818, 3840, 3842,
3215 3851, 3855, 3884, 3885, 3902, 3905, 3919, 3944, 3954, 3955, 3957, 3991,
3216 3995, 3996, 4024, 4040, 4069, 4070, 4074, 4076, 4096, 4101, 4102, 4114,
3217 4130, 4131, 4181, 4276, 4306, 4342, 4344, 4349, 4364, 4368, 4381, 4392,
3218 4403, 4405, 4406, 4411, 4438, 4439, 4455, 4456, 4465, 4512, 4514, 4586,
3221 * New Linux interfaces: epoll_pwait, sched_getcpu.
3223 * New generic interfaces: strerror_l.
3225 * nscd can now cache the services database. Implemented by Ulrich Drepper.
3229 * The following bugs are resolved with this release:
3231 39, 68, 192, 352, 388, 531, 935, 1006, 1201, 1203, 1386, 1782, 1783, 1784,
3232 1785, 1786, 1787, 1951, 1974, 1996, 2072, 2098, 2125, 2134, 2158, 2182,
3233 2349, 2376, 2386, 2389, 2415, 2418, 2420, 2423, 2450, 2451, 2466, 2472,
3234 2473, 2477, 2498, 2499, 2501, 2502, 2505, 2507, 2508, 2509, 2517, 2526,
3235 2569, 2571, 2592, 2611, 2625, 2632, 2644, 2662, 2680, 2683, 2684, 2693,
3236 2695, 2703, 2734, 2751, 2764, 2766, 2775, 2792, 2821, 2823, 2832, 2841,
3237 2843, 2883, 2892, 2908, 2914, 2926, 2961, 2978, 2980, 2987, 2997, 2998,
3238 3013, 3018, 3034, 3040, 3044, 3123, 3124, 3137, 3138, 3143, 3155, 3189,
3239 3225, 3251, 3252, 3253, 3273
3241 * For Linux, the sorting of addresses returned by getaddrinfo now also
3242 handles rules 3, 4, and 7 from RFC 3484. I.e., all rules are handled.
3243 Implemented by Ulrich Drepper.
3245 * Allow system admin to configure getaddrinfo with the /etc/gai.conf file.
3246 Implemented by Ulrich Drepper.
3248 * New Linux interfaces: splice, tee, sync_file_range, vmsplice.
3250 * New iconv module for MIK. Contributed by Alexander Shopov.
3252 * For sites with broken group and/or passwd database, the auto-propagate
3253 option of nscd can prevent creating ID lookup entries from the results
3254 of a name lookup and vice versa. This usually is no problem but some
3255 site might have problems with the default behavior.
3256 Implemented by Ulrich Drepper.
3258 * Iterating over entire database in NIS can be slow. With the
3259 SETENT_BATCH_READ option in /etc/default/nss a system admin can decide
3260 to trade time for memory. The entire database will be read at once.
3261 Implemented by Ulrich Drepper.
3263 * The interfaces introduced in RFC 3542 have been implemented by
3266 * Support for the new ELF hash table format was added by Ulrich Drepper.
3268 * Support for priority inheritance mutexes added by Jakub Jelinek and
3271 * Support for priority protected mutexes added by Jakub Jelinek.
3273 * New locales: nr_ZA, pa_PK, ca_AD, ca_FR, ca_IT, el_CY, tr_CY, as_IN,
3274 or_IN, csb_PL, fy_NL, sr_ME.
3278 * More overflow detection functions.
3280 * New iconv converters for IBM1025, IBM1122, IBM1137, IBM1153,
3281 IBM1154, IBM1155, IBM1156, IBM1157, and IBM1158 contributed by Jiro Sekiba.
3283 More converters for IBM803, IBM901, IBM902, IBM921, IBM1008, IBM1097,
3284 IBM1112, IBM1123, IBM1130, IBM1140, IBM1141, IBM1142, IBM1143, IBM1144,
3285 IBM1145, IBM1146, IBM1147, IBM1148, IBM1149, IBM1166, IBM1167, IBM4517,
3286 IBM4899, IBM4909, IBM4971, IBM5347, IBM9030, IBM9066, IBM9448, IBM12712,
3287 IBM16804, IBM1364, IBM1371, IBM1388, IBM1390, and IBM1399 contributed
3288 by Masahide Washizawa.
3290 * It is now possible to install audit libraries for the dynamic linker using
3291 LD_AUDIT. Implemented by Ulrich Drepper.
3293 * The LinuxThreads add-on, providing pthreads on Linux 2.4 kernels, is no
3294 longer supported. The new NPTL implementation requires Linux 2.6 kernels.
3295 For a libc and libpthread that works well on Linux 2.4 kernels, we
3296 recommend using the stable 2.3 branch.
3298 * The new function open_wmemstream acts like open_memstream,
3299 but uses a wchar_t wide character buffer.
3301 * The new function ppoll is an improved version of poll, similar to pselect.
3303 * New interfaces: fdopendir, openat, fstatat, fchmodat, fchownat,
3304 futimesat, faccessat, mkdirat, mkfifoat, mknodat,
3305 renameat, unlinkat, linkat, symlinkat, readlinkat.
3307 * New Linux kernel system calls: unshare,
3308 inotify_init, inotify_add_watch, inotify_rm_watch.
3310 * The euidaccess function is now also known by the alias eaccess,
3311 for compatibility with some other systems.
3313 * Timezone data updated to 2006b version.
3317 * The following bugs are resolved with this release:
3319 38, 253, 549, 622, 653, 721, 758, 851, 877, 915, 934, 955, 961,
3320 1016, 1037, 1076, 1079, 1080, 1081, 1082, 1083, 1084, 1085, 1086,
3321 1087, 1088, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098,
3322 1099, 1100, 1101, 1102, 1103, 1104, 1105, 1106, 1107, 1108, 1109,
3323 1110, 1111, 1112, 1113, 1125, 1137, 1138, 1249, 1250, 1251, 1252,
3324 1253, 1254, 1350, 1358, 1394, 1438, 1498, 1534
3326 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3328 * As of this release, GCC 4 can be used to compile the C Library.
3330 * Timezone data updated to 2005m version.
3334 * The following bugs are resolved with this release:
3336 284, 592, 602, 626, 633, 640, 650, 661, 671, 681, 693, 700, 710, 719,
3337 722, 723, 725, 726, 727, 728, 729, 730, 731, 732, 733, 734, 735, 736,
3338 737, 738, 739, 740, 741, 742, 743, 744, 745, 765, 767, 768, 769, 776,
3339 777, 787, 821, 822, 823, 825
3341 Visit <http://sources.redhat.com/bugzilla/> for the details of each bug.
3345 * Support for RFC 3678. Real implementations exist only for Linux so far.
3346 Implemented by Ulrich Drepper.
3348 * nscd can now cache entries persistently. Expiring entries are reloaded.
3349 For speedups the cache can be shared in memory with client processes.
3350 Implemented by Ulrich Drepper.
3352 * nscd can now perform SELinux checks.
3353 Implemented by Matthew Rickard <mjricka@epoch.ncsc.mil>.
3355 * getaddrinfo queries are now cached. Canonical name lookup is performed
3357 Implemented by Ulrich Drepper.
3359 * The nothrow function attribute is used when headers are used by gcc when
3360 compiling C code. This can avoid significant amounts of exception
3363 * The malloc functions perform more error checking and are stricter when
3364 it comes to reacting on errors. The default action is to terminate
3365 the process after showing an error message. Implemented by Ulrich Drepper.
3367 * Reverse lookups of IPv6 addresses does not use bit string or .ip6.int
3368 lookups anymore unless explicitly requested. Implemented by Ulrich Drepper.
3370 * Namespaces in ld.so are implemented. DSOs can be loaded in separate
3371 namespaces using the new function dlmopen(). This feature is of course,
3372 like most other dynamic loading functionality, not available in statically
3373 linked applications. Implemented by Ulrich Drepper.
3375 * Low-overhead boundary checking variants of string and some stdio functions
3376 were added. These are to be used in conjunction with a gcc patch by
3377 Jakub Jelinek which adds calls to these functions if possible.
3378 Implemented by Jakub Jelinek and Ulrich Drepper.
3380 * Old code for several operating systems and machine architectures that
3381 have not been in working condition in a long time have been removed from
3382 the main source tree maintained by the GNU C Library's maintainers.
3383 These files are now reside in the separate `ports' source module
3384 that is usable as an add-on when building the library.
3388 * New functions `dladdr1' and `dlinfo' in <dlfcn.h> provide more ways to
3389 interrogate the dynamic linker, compatible with the Solaris interface.
3391 * ELF thread-local storage support (TLS) now works on PowerPC and PowerPC64;
3392 implemented by Paul Mackerras, Steven Munroe, and Roland McGrath.
3394 * getifaddrs now uses the netlink interface on Linux to get its information.
3395 Implemented by Thorsten Kukuk.
3397 * getaddrinfo now implements AI_V4MAPPED, AI_ALL, and AI_ADDRCONF.
3398 Implemented by Ulrich Drepper.
3400 * support for non-executable stacks on x86 has been added. Changes mostly
3403 * regex is now much faster for multibyte locales. Changes by Jakub Jelinek
3406 * getaddrinfo now performs destination address selection according to
3411 * Thread-safe interfaces for many functions that access locale data
3412 were added in version 2.3, but these features were omitted from NEWS.
3413 Many functions have variants with an `_l' suffix that take a `locale_t'
3414 object as a parameter rather than consulting the current locale.
3415 The new functions `newlocale', `duplocale', and `freelocale' in <locale.h>
3416 create and maintain `locale_t' objects. Additionally, the new function
3417 `uselocale' sets "the current locale" (as used by functions not so
3418 parameterized) set for an individual thread. These features were added
3419 in version 2.3, implemented by Ulrich Drepper and Roland McGrath.
3421 * The functions getresuid, getresgid, setresuid, and setresgid, which
3422 have long been available on Linux, are now declared in <unistd.h>
3423 and are now also available on the Hurd.
3425 * ELF thread-local storage support (TLS) now works on x86-64.
3427 * The new dynamic string token $LIB is expanded in shared library names.
3428 This normally expands to lib, but on some 64-bit platforms to lib64 instead.
3430 * Aldy Hernandez contributed complete software floating point support for
3431 PowerPC machines with no FPU.
3433 * fexecve is implemented on Linux.
3435 * The `btowc' function should work at least twice as fast due to
3436 specialized callbacks in the iconv modules. Implemented by Bruno Haible.
3438 * With approriate thread add-ons cancelable functions are now implemented
3439 in libc.so as well. No need to call the function in libpthread. This
3440 change allowed to finally disable the incorrect and expensive handling
3441 of weak definition in ld.so.
3443 * Yet more PLT entries in libc.so have been removed. We finally arrived
3444 at the bare minimum. Startup times improved appropriately.
3446 * Support for the new Linux/x86 system call interface was added. The
3447 AT_SYSINFO auxiliary vector entry is recognized and handled.
3451 * Masahide Washizawa contributed iconv modules for IBM1163 and IBM1164
3454 * iconv (the program and the interface) now accepts empty names (excluding
3455 options like //TRANSLIT) to mean "use charset of current locale".
3457 * localedef can now transliterate characters in strings which are not in
3458 the provided charmap. The information from the input locale is used.
3460 * Prelinking support was added for ELF targets. This requires additional
3461 tools and recent versions of the GNU binutils. Contributed by Jakub Jelinek.
3463 * Read-only stdio streams now use mmap to speed up operation by eliminating
3464 copying and buffer underflows. To use add 'm' to the mode string of
3465 the fopen/fdopen/freopen call. Implemented by Ulrich Drepper.
3467 * The malloc functions were completely rewritten by Wolfram Gloger based
3468 on Doug Lea's malloc-2.7.0.c.
3470 * Isamu Hasegawa contributed a completely new and POSIX-conformant
3471 implementation of regex.
3473 * Bruno Haible upgraded the iconv and locale implementation to support
3476 * Contents of the LC_* and LANG environment variables in the CEN style are
3477 not recognized anymore. It never was used. Change by Ulrich Drepper.
3479 * The runtime (ld.so, libc, libpthread for Linux) now can handle the ELF
3480 thread-local storage (TLS) ABI on some platforms.
3481 Changes by Ulrich Drepper. SH support by Kaz Kojima.
3483 * Bruno Haible contributed iconv converters for ISO-2022-JP-3, SHIFT JIS-X0213,
3484 EUC-JISX0213, and TSCII.
3486 * New header <ifaddrs.h> with functions `getifaddrs' and `freeifaddrs':
3487 BSD-compatible interface for getting all network interface addresses.
3488 Implementation for IPv4 by Roland McGrath.
3490 * Loading of locale data is faster due to the introduction of a locale
3491 archive. Implemented by Roland McGrath and Ulrich Drepper.
3493 * Startup times are significantly reduced by not using exported functions
3494 inside the library itself. Changes by Jakub Jelinek, Roland McGrath,
3497 * Steven Munroe contributed a port to PowerPC64/Linux.
3501 * The Hurd now uses the GNU libio implementation of stdio rather than the
3502 old GNU stdio implementation, and uses a new ABI (libc.so.0.3).
3504 * The Hurd on x86 now has the `ioperm' function and <sys/io.h> header file
3505 with the same behavior as the Linux system call of the same name.
3509 * Stephen Moshier implemented log2, log10, powl and cbrtl for the
3510 128-bit long double format.
3512 * Masahide Washizawa contributed iconv modules for IBM1132, IBM1133, IBM1160,
3513 IBM1161, and IBM1162 charsets.
3515 * Andreas Jaeger contributed a port to x86-64/Linux.
3517 * Peter Bruin contributed a port to PowerPC/Hurd.
3519 * libc functions using I/O streams now can handle wide-oriented streams
3522 * optimizations in the dynamic linker. Binaries created by recent binutils
3523 versions start up quicker due to reduced time spent on relocations.
3525 * Support for use of gcc3 added by Jakub Jelinek and HJ Lu.
3529 * Stephen Moshier implemented cosh, expm1, log1p, acos, sinh, tanh,
3530 asinh, atanh, j0 for the 128-bit long double format.
3532 * Bruno Haible updated all the code handling Unicode in some form to
3533 support Unicode 3.1.
3535 * Speed of regex for single-byte locales is back to previous levels.
3536 Patch by Isamu Hasegawa.
3538 * Alpha, SPARC, and IA-64 now also using floating stacks.
3540 * Startup time of internationalized applications greatly improved through
3541 iconv cache. Use iconvconfig to generate the cache file.
3542 Contributed by Ulrich Drepper.
3544 * The IA-64 specific part of ld.so was rewritten to eliminate some pretty
3545 severe performance problems. Patch by David Mosberger.
3547 * The Hurd port got a lot more functionality like AIO, various stdio
3548 extensions, etc. Mainly done by Roland McGrath.
3550 * mtrace can now lookup symbols in shared libraries.
3554 * Intel's IA-64 math library is largely integrated. It provides fast and
3555 accurate implementatations for most basic and standard math functions
3556 in float, double, and long double format.
3558 * Stephen Moshier implemented j0, j1, jn, y0, y1, yn, lgamma, erf, erfc,
3559 and asin for the 96-bit long double format and asin, log, tan for the
3560 128-bit long double format.
3562 * The beginning of a last-bit accurate math library by IBM Haifa were added.
3563 The basic double functions exist today. Contributed by Abraham Ziv
3564 <ziv@il.ibm.com>, Moshe Olshansky <olshansk@il.ibm.com>, Ealan Henis
3565 <ealan@il.ibm.com>, and Anna Reitman <reitman@il.ibm.com>.
3567 * An asynchronous name lookup library was added. The interface is designed
3568 after POSIX AIO. The proposal was circulated beforehand to get comments.
3569 No negative ones came in. Implemented by Ulrich Drepper.
3571 * Port to S390/64bit contributed by Martin Schwidefsky
3572 <schwidefsky@de.ibm.com>.
3574 * David Mosberger <davidm@hpl.hp.com> implemented the setcontext family
3575 of functions for Linux/IA-64.
3577 * The RPC code is now thread safe. Threads can now use the same service
3578 of different services at the same time. Patch by Eric Norum
3579 <eric.norum@usask.ca> with some help by Ulrich Drepper.
3581 * Martin Schwidefsky <schwidefsky@de.ibm.com> implemented the setcontext
3582 family of functions for Linux/S390.
3584 * Ulrich Drepper <drepper@redhat.com> implemented the setcontext family
3585 of functions for Linux/x86.
3587 * Port to Linux/CRIS contributed by Axis Communications.
3591 * Lots of headers were cleaned up. Using the tool in the conform/ subdir
3592 we can now check for namespace violations and missing declarations. The
3593 result is that almost all headers are now Unix-compliant (as defined in
3594 the upcoming XPG6). The negative side is that some programs might need
3595 corrections, too, if they depend on the incorrect form of the headers in
3596 previous versions which defined too many symbols and included too many
3599 * regex now handles multibyte character sets correctly.
3600 Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>.
3602 * iconv (the program) does now conform to the upcoming XPG6 and handles
3603 charmaps. Instead of the charset names the path of charmaps can be
3604 provided and the conversion happens based on this data.
3605 Contributed by Ulrich Drepper.
3607 * The locale program now provides detailed information about the installed
3612 only lists the names of the supported locales
3616 provides details such as country, language, and codeset name.
3617 Contributed by Ulrich Drepper.
3621 * The gencat program now parses the input file according to the charset
3622 selected by the LC_CTYPE category. This is important for stateful
3623 character sets. To make generating catalogs easier there is a way
3624 to overwrite the charset selected by the locale: before the first
3625 message or $ quote line the catalog can contain a line like
3627 $ codeset=ISO-8859-2
3629 to select the charset (ISO-8859-2 in this case).
3631 Implemented by Shinya Hanataka and Ulrich Drepper.
3633 * New codeset conversion modules: IBM-922 (Estonia PC codeset),
3634 IBM-1124 (Ukraine PC codeset), IBM-1129 (Vietnamese PC codeset).
3635 Contributed by Masahide Washizawa <washi@jp.ibm.com>.
3637 * Optimized string functions for Alpha ev6 and ev67 by Richard Henderson
3638 <rth@redhat.com> and Rick Gorton <rick.gorton@alpha-processor.com>.
3640 * The LANGUAGE environment variable is now ignored unless the locale is
3641 changed from the default "C" locale.
3643 * The usual bug fixes.
3647 * Greg McGary added runtime support for bounds checking using gcc's
3648 new -fbounded-pointers option. ix86 target is complete. PowerPC
3651 * Thorsten Kukuk added secure mode support to nscd.
3653 * The Berkeley db libraries have been removed.
3655 Related, the nss_db module is now in a separate package since it
3656 obviously requires a database library being available.
3658 * Wide character I/O streams implemented by Ulrich Drepper.
3660 * Functions from the extended socket API added by Ulrich Drepper.
3662 * Functions feenableexcept and fedisableexcept to control the
3663 behaviour of individual exceptions have been added by Andreas Jaeger.
3665 * ldconfig program added by Andreas Jaeger and Jakub Jelinek.
3667 * The resolver code has been updated from bind 8.2.3-T5B which supports
3668 threads. The integration was done by Andreas Jaeger, Adam D. Bradley,
3671 This change could in some situations effect backward compatibility. Since
3672 now `_res' is a thread-local instead of a global variable, modifying it
3673 in one thread does not have any effect in other threads.
3675 The resolver library was also extended to allow IPv6 as the transport
3676 protocol for the requests. This work was done by Stig Venaas.
3678 * Compatibility code for K&R C compilers has been removed from the
3679 header files. A ISO C compiler is needed to use the library
3680 (conforming to either C89 or C99 standard).
3682 * Complete rewrite of the localedef program to support multibyte character
3683 sets. Implement handling of ISO 14651 and ISO 14652. Rewrite strcoll,
3684 strxfrm, wcscoll, and wcsxfrm functions. Make isw*() functions work.
3685 Implemented by Ulrich Drepper.
3687 Bruno Haible significantly improved the generation and use of the data
3688 structures for the wide character tables.
3690 * Plural handling in gettext implemented by Ulrich Drepper.
3692 * The utmp daemon has been removed.
3694 * The port to MIPS-Linux has been finished by Andreas Jaeger.
3696 * A port to Hitachi SH3 and SH4 has been contributed by Kazumoto Kojima
3699 * POSIX clocks and timers implemented by Kaz Kylheku and Ulrich Drepper.
3701 * POSIX barriers implemented by Kaz Kylheku.
3703 * POSIX spawn function family implemented by Ulrich Drepper.
3705 * POSIX spinlocks are now available.
3707 * Timed wait functions for mutex, rwlock, and semaphores are implemented.
3709 * the configure option --enable-kernel=X.Y.Z allows to strip out
3710 compatibility for kernel versions before X.Y.Z. This is currently only
3711 implemented for Linux.
3713 * the sockaddr_in6 structure changed. The IPv6 working group added a new
3714 field sin6_scope_id. This means that all programs using IPv6 should be
3715 recompiled. Don't expect binary compatibility with previous glibc
3718 * various conversion modules for IBM character sets contributed by
3721 * IA-64 port by Jes Sorensen and HJ Lu.
3735 * New ISO C 9x function _Exit, imaxabs, and imaxdiv are added.
3737 * New xdr functions are added; some rpc functions are now 64bit clean.
3739 * Fixed a number of bugs and memory leaks (especially in NIS+ code).
3741 * Fixed known incompatibilities with glibc 2.0.
3743 * New functions lock64, strchrnul, rawmemchr, getutmp and getutmpx.
3745 * Optimized a number of functions (especially the ELF dynamic loader).
3747 * Update timezone data files.
3749 * lots of charmaps corrections
3751 * some new locale definitions and charmaps
3756 * Richard Henderson corrected size of struct timeval on Linux/Alpha to
3757 conform to POSIX member type requirements. Symbol versions have been
3758 adjusted as needed within the library, and for direct use by applications,
3759 but there is potential for problems if third-party libraries use
3760 struct timeval as part of their interface. This does not present
3761 a problem for X and other "essential" system libraries.
3763 * An additional locale model to support C++ Standard Library locale
3764 model and probably more was implemented by Ulrich Drepper.
3766 * Eric Youngdale and Ulrich Drepper implemented versioning of objects on
3769 * Miles Bader provided the `argp' function family to support hierachical
3770 command line argument parsing, layered on top of getopt.
3772 * strtod accepts new hexadecimal floating-point format from ISO C 9X.
3774 * printf has two new specifiers %a and %A to print hexadecimal flaoting-point
3777 * scanf recognizes the %a and %A format for scanning floating point numbers.
3779 * the new headers <stdint.h> and <inttypes.h> from ISO C 9X provides
3780 information and interfaces for the available integer types.
3782 * about 130 new math functions were added to implement the ISO C9x math
3785 * the new header <complex.h> contains definitions of the complex math
3786 functions from ISO C 9X.
3788 * the new header <tgmath.h> defines generic macros to use complex or
3789 real valued functions.
3791 * Thorsten Kukuk provided an implementation for NIS+, securelevel 0, 1 and 2.
3793 * Andreas Jaeger provided a test suite for the math library.
3795 * Mark Kettenis implemented the utmpx interface and an utmp daemon.
3797 * Ulrich Drepper added character set conversion functions (iconv).
3799 * Optimized string functions have been added.
3801 * The localedata addon is now part of glibc.
3803 * An implementation of profiling shared libraries was added by Ulrich Drepper.
3805 * Thorsten Kukuk and Ulrich Drepper provided an implementation for a caching
3806 daemon for NSS (nscd).
3808 Missing a better place here are some numbers on improvements. Under
3809 Linux 2.1.125 un-tar-ing the kernel sources takes
3813 using local files 12.19s 6.88s 22.91s
3815 using NIS 13.92s 8.91s 26.34s
3817 using NIS & nscd 10.37s 7.34s 25.30s
3819 using NIS+ 27.57s 30.37s 640.46s
3821 using NIS+ & nscd 10.25s 7.83s 26.51s
3823 using NIS & old nscd [1] 13.83s 8.32s 29.60s
3825 Keep in mind that non-namelookup related operations dominate above times.
3826 It was just a common complain that using NIS+ unpacking the kernel is
3829 [1] The old nscd implementation is not available anymore since it was
3830 distributed with glibc up to version 2.0.98 and thus is now replaced.
3832 * Tim Waugh provided an implementation of the POSIX.2 wordexp function family.
3834 * Mark Kettenis provided a Hesiod NSS module.
3836 * The ELF dynamic loader knows how to process dynamic string tokens ($ORIGIN
3837 and $PLATFORM) in RPATHs and similar strings (Ulrich Drepper).
3839 * rcmd can now handle netgroups (Dick Streefland).
3841 * A port to the ARM was contributed by Phil Blundell, Pat Beirne and Scott
3844 * Support for the IPv6 protocol has been added to the socket API, as per the
3845 latest draft standards.
3847 * Support for Linux 2.2 has been added.
3849 * Interface changes relative to the latest 2.0.x release:
3850 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3851 addseverity NEW: Unix98
3852 alphasort64 NEW: LFS
3853 argp_err_exit_status NEW: argp, GNU ext
3854 argp_error NEW: argp, GNU ext
3855 argp_failure NEW: argp, GNU ext
3856 argp_help NEW: argp, GNU ext
3857 argp_parse NEW: argp, GNU ext
3858 argp_program_bug_address NEW: argp, GNU ext
3859 argp_program_version NEW: argp, GNU ext
3860 argp_program_version_hook NEW: argp, GNU ext
3861 argp_state_help NEW: argp, GNU ext
3862 argp_usage NEW: argp, GNU ext
3863 authdes_create NEW: Secure RPC
3864 authdes_getucred NEW: Secure RPC
3865 authdes_pk_create NEW: Secure RPC
3866 backtrace NEW: GNU ext.
3867 backtrace_symbols NEW: GNU ext.
3868 backtrace_symbols_fd NEW: GNU ext.
3870 cacosf NEW: ISO C 9x
3871 cacosh NEW: ISO C 9x
3872 cacoshf NEW: ISO C 9x
3873 cacoshl NEW: ISO C 9x
3874 cacosl NEW: ISO C 9x
3881 casinf NEW: ISO C 9x
3882 casinh NEW: ISO C 9x
3883 casinhf NEW: ISO C 9x
3884 casinhl NEW: ISO C 9x
3885 casinl NEW: ISO C 9x
3887 catanf NEW: ISO C 9x
3888 catanh NEW: ISO C 9x
3889 catanhf NEW: ISO C 9x
3890 catanhl NEW: ISO C 9x
3891 catanl NEW: ISO C 9x
3892 cbc_crypt NEW: Secure RPC
3896 ccoshf NEW: ISO C 9x
3897 ccoshl NEW: ISO C 9x
3903 cimagf NEW: ISO C 9x
3904 cimagl NEW: ISO C 9x
3905 clearerr_locked REMOVED
3906 clntunix_create NEW: sunrpc ext
3908 clog10 NEW: ISO C 9x
3909 clog10f NEW: ISO C 9x
3910 clog10l NEW: ISO C 9x
3920 cprojf NEW: ISO C 9x
3921 cprojl NEW: ISO C 9x
3923 crealf NEW: ISO C 9x
3924 creall NEW: ISO C 9x
3929 csinhf NEW: ISO C 9x
3930 csinhl NEW: ISO C 9x
3933 csqrtf NEW: ISO C 9x
3934 csqrtl NEW: ISO C 9x
3938 ctanhf NEW: ISO C 9x
3939 ctanhl NEW: ISO C 9x
3941 des_setparity NEW: Secure RPC
3942 ecb_crypt NEW: Secure RPC
3943 endutxent NEW: Unix98
3945 exp10f NEW: ISO C 9x
3946 exp10l NEW: ISO C 9x
3950 fattach NEW: STREAMS
3951 fdetach NEW: STREAMS
3955 feclearexcept NEW: ISO C 9x
3956 fegetenv NEW: ISO C 9x
3957 fegetexceptflag NEW: ISO C 9x
3958 fegetround NEW: ISO C 9x
3959 feholdexcept NEW: ISO C 9x
3961 feraiseexcept NEW: ISO C 9x
3962 ferror_locked REMOVED
3963 fesetenv NEW: ISO C 9x
3964 fesetexceptflag NEW: ISO C 9x
3965 fesetround NEW: ISO C 9x
3966 fetestexcept NEW: ISO C 9x
3967 feupdateenv NEW: ISO C 9x
3968 fflush_locked REMOVED
3972 fgets_unlocked NEW: GNU ext.
3973 fileno_locked REMOVED
3985 fputc_locked REMOVED
3986 fputs_unlocked NEW: GNU ext.
3987 fread_unlocked NEW: GNU ext.
3992 fstatvfs NEW: Unix98
3996 ftruncate64 NEW: LFS
3998 fwrite_unlocked NEW: GNU ext.
3999 gai_strerror NEW: IPv6
4003 getchar_locked REMOVED
4005 getdate_err NEW: Unix98
4006 getdate_r NEW: GNU ext.
4008 getnameinfo NEW: IPv6
4009 getnetname NEW: Secure RPC
4010 getpmsg NEW: STREAMS
4011 getpt NEW: Unix98 PTY
4012 getrlimit64 NEW: LFS
4013 getutxent NEW: Unix98
4014 getutxid NEW: Unix98
4015 getutxline NEW: Unix98
4016 glob64 NEW: GNU ext.
4017 globfree64 NEW: GNU ext.
4018 gnu_get_libc_release NEW: GNU ext.
4019 gnu_get_libc_version NEW: GNU ext.
4020 grantpt NEW: Unix98 PTY
4021 host2netname NEW: Secure RPC
4023 iconv_close NEW: iconv
4024 iconv_open NEW: iconv
4025 if_freenameindex NEW: IPv6
4026 if_indextoname NEW: IPv6
4027 if_nameindex NEW: IPv6
4028 if_nametoindex NEW: IPv6
4029 in6addr_any NEW: IPv6
4030 in6addr_loopback NEW: IPv6
4031 inet6_isipv4mapped NEW: IPv6
4032 isastream NEW: STREAMS
4033 iswblank NEW: GNU ext.
4034 key_decryptsession NEW: Secure RPC
4035 key_decryptsession_pk NEW: Secure RPC
4036 key_encryptsession NEW: Secure RPC
4037 key_encryptsession_pk NEW: Secure RPC
4038 key_gendes NEW: Secure RPC
4039 key_get_conv NEW: Secure RPC
4040 key_secretkey_is_set NEW: Secure RPC
4041 key_setnet NEW: Secure RPC
4042 key_setsecret NEW: Secure RPC
4043 llrint NEW: ISO C 9x
4044 llrintf NEW: ISO C 9x
4045 llrintl NEW: ISO C 9x
4046 llround NEW: ISO C 9x
4047 llroundf NEW: ISO C 9x
4048 llroundl NEW: ISO C 9x
4053 lrintf NEW: ISO C 9x
4054 lrintl NEW: ISO C 9x
4055 lround NEW: ISO C 9x
4056 lroundf NEW: ISO C 9x
4057 lroundl NEW: ISO C 9x
4059 makecontext NEW: Unix98
4060 mempcpy NEW: GNU ext.
4063 modify_ldt NEW: kernel
4067 nearbyint NEW: ISO C 9x
4068 nearbyintf NEW: ISO C 9x
4069 nearbyintl NEW: ISO C 9x
4070 netname2host NEW: Secure RPC
4071 netname2user NEW: Secure RPC
4072 nexttoward NEW: ISO C 9x
4073 nexttowardf NEW: ISO C 9x
4074 nexttowardl NEW: ISO C 9x
4078 passwd2des NEW: Secure RPC
4080 pow10f NEW: GNU ext.
4081 pow10l NEW: GNU ext.
4084 printf_size NEW: GNU ext.
4085 printf_size_info NEW: GNU ext.
4086 profil_counter REMOVED
4087 pthread_mutexattr_getkind_np REPLACED
4088 pthread_mutexattr_setkind_np REPLACED
4089 ptsname NEW: Unix98 PTY
4090 ptsname_r NEW: Unix98 PTY
4092 putchar_locked REMOVED
4093 putgrent NEW: GNU ext.
4095 putpmsg NEW: STREAMS
4096 pututxline NEW: Unix98
4100 readdir64_r NEW: LFS
4101 remquo NEW: ISO C 9x
4102 remquof NEW: ISO C 9x
4103 remquol NEW: ISO C 9x
4105 roundf NEW: ISO C 9x
4106 roundl NEW: ISO C 9x
4108 scalbln NEW: ISO C 9x
4109 scalblnf NEW: ISO C 9x
4110 scalblnl NEW: ISO C 9x
4112 sendfile NEW: kernel
4113 setcontext NEW: Unix98
4114 setrlimit64 NEW: LFS
4115 setutxent NEW: Unix98
4117 sigignore NEW: Unix98
4118 sigqueue NEW: POSIX.1b
4119 sigrelse NEW: Unix98
4120 sigset NEW: POSIX.1b
4121 sigtimedwait NEW: POSIX.1b
4122 sigwaitinfo NEW: POSIX.1b
4123 sincos NEW: GNU ext.
4124 sincosf NEW: GNU ext.
4125 sincosl NEW: GNU ext.
4129 strcasestr NEW: GNU ext.
4130 strtoimax NEW: ISO C 9x
4131 strtoumax NEW: ISO C 9x
4132 strverscmp NEW: GNU ext.
4133 svcauthdes_stats NEW: Secure RPC
4134 svcunix_create NEW: sunrpc etx
4135 svcunixfd_create NEW: sunrpc ext
4136 swapcontext NEW: Unix98
4137 tcgetsid NEW: Unix98 PTY
4138 tdestroy NEW: GNU ext.
4139 tgamma NEW: ISO C 9x
4140 tgammaf NEW: ISO C 9x
4141 tgammal NEW: ISO C 9x
4145 truncf NEW: ISO C 9x
4146 truncl NEW: ISO C 9x
4148 unlockpt NEW: Unix98 PTY
4149 updwtmpx NEW: Unix98
4150 user2netname NEW: Secure RPC
4151 utmpxname NEW: Unix98
4152 versionsort NEW: GNU ext.
4153 versionsort64 NEW: GNU ext.
4155 wcscasecmp NEW: GNU ext.
4156 wcsncasecmp NEW: GNU ext.
4157 wcsnlen NEW: GNU ext.
4158 wcstoimax NEW: ISO C 9x
4159 wcstoll NEW: ISO C 9x
4160 wcstoull NEW: ISO C 9x
4161 wcstoumax NEW: ISO C 9x
4163 wordexp NEW: POSIX.2
4164 wordfree NEW: POSIX.2
4165 write_profiling REMOVED
4166 xdecrypt NEW: Secure RPC
4167 xdr_authdes_cred NEW: Secure RPC
4168 xdr_authdes_verf NEW: Secure RPC
4169 xdr_cryptkeyarg NEW: Secure RPC
4170 xdr_cryptkeyarg2 NEW: Secure RPC
4171 xdr_cryptkeyres NEW: Secure RPC
4172 xdr_getcredres NEW: Secure RPC
4173 xdr_key_netstarg NEW: Secure RPC
4174 xdr_key_netstres NEW: Secure RPC
4175 xdr_keybuf NEW: Secure RPC
4176 xdr_keystatus NEW: Secure RPC
4177 xdr_netnamestr NEW: Secure RPC
4178 xdr_sizeof NEW: Secure RPC
4179 xdr_unixcred NEW: sunrpc ext
4180 xencrypt NEW: Secure RPC
4181 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4192 * inet_ntoa is thread-safe
4194 * updwtmp is moved from libutil to libc
4196 * rewrite of cbrt function
4198 * update of timezone data
4212 * add atoll function
4214 * fix complex problems in Berkeley DB code
4216 * fix math functions
4220 * fixed lots of header problems (especially Linux/GNU specific)
4222 * dynamic loader preserves all registers
4224 * Roland McGrath provided support for handling of auxiliary objects in
4225 the ELF dynamic loader.
4227 * support for parallel builds is improved
4231 * GNU extensions are no longer declared by default. To enable them you
4232 must define the macro `_GNU_SOURCE' in your program or compile with
4235 * The library has changed from using GNU ld symbol aliases to using weak
4236 symbols where available. The ELF object file format supports weak
4237 symbols; GNU ld also supports weak symbols in the a.out format. (There
4238 is also now support for other GNU ld extensions in ELF. Use the
4239 `--with-elf' option to configure to indicate you have ELF, and
4240 `--with-gnu-ld' if using GNU ld.) This change resulted in the deletion
4241 of many files which contained only symbol aliases, reducing the size of
4242 the source and the compiled library; many other files were renamed to
4243 less cryptic names previously occupied by the symbol alias files.
4244 There is a new header file <elf.h> for programs which operate on
4245 files in the ELF format.
4247 * Converted to Autoconf version 2, so `configure' has more options.
4248 Run `configure --help' to see the details.
4250 * The library can now be configured to build profiling, highly-optimized
4251 (but undebuggable), and/or shared libraries (ELF with GNU ld only). The
4252 `--enable-profile', `--enable-omitfp', and `--enable-shared' options to
4253 `configure' enable building these extra libraries. The shared library is
4254 built by default when using both ELF and GNU ld. When shared libraries
4255 are enabled, the new library `-ldl' is available for arbitrary run-time
4256 loading of shared objects; its interface is defined in <dlfcn.h>. The
4257 new header file <link.h> gives access to the internals of the run-time
4258 dynamic linker, `ld.so'. The shell script `ldd' is similar to the
4259 application of same name on other systems and it provides information
4260 about dynamically linked binaries.
4262 * The C library now provides the run-time support code for profiling
4263 executables compiled with `-pg'. Programs can control the profiling code
4264 through the interface in <sys/gmon.h>. The `gmon.out' files written by
4265 the GNU C library can be read only by GNU `gprof' (from GNU binutils);
4266 the support for this file format was contributed by David Mosberger-Tang.
4268 * The math code has been replaced with a math library based on fdlibm from
4269 Sun, and modified by JT Conklin and Ulrich Drepper with i387 support, by
4270 Ian Taylor with `float' functions and by Ulrich Drepper with `long double'
4271 functions. The math functions now reside in a separate library, so
4272 programs using them will need to use `-lm' their linking commands.
4274 * John C. Bowman contributed optimized ix87 assembler inline functions.
4276 * Ulrich Drepper has contributed support for an `/etc/nsswitch.conf'
4277 mechanism similar to that found in Solaris 2. This is now used for the
4278 group, passwd, hosts, networks, services, protocols, rpc, ethers,
4279 shadow, netgroup, publickey, and alias databases. The `nsswitch.conf'
4280 file controls what services are used for each individual database. This
4281 works by loading shared libraries with names specified in `nsswitch.conf',
4282 so service modules can be changed or added at any time without even
4283 relinking any program. Currently there are the file, db, and NIS based
4284 NSS services available.
4286 * The new functions `strtoq' and `strtouq' parse integer values from
4287 strings, like `strtol' and `strtoul', but they return `long long int' and
4288 `unsigned long long int' values, respectively (64-bit quantities).
4290 * The new functions `strtof' and `strtold' parse floating-point values from
4291 strings, like `strtod', but they return `float' and `long double' values,
4292 respectively (on some machines `double' and `long double' are the same).
4294 * Ulrich Drepper has contributed new implementations of the floating-point
4295 printing and reading code used in the `printf' family of functions and
4296 `strtod', `strtof', and `strtold'. These new functions are perfectly
4297 accurate, and much faster than the old ones.
4299 * The implementation of the POSIX locale model was completely rewritten by
4300 Ulrich Drepper. This includes the new programs `localedef' and `locale'
4301 to compile the POSIX locale definition.
4303 * The former dummy implementations of the strcoll and strxfrm function are
4304 now replaced by fully functional code contributed by Ulrich Drepper. The
4305 collation information comes from the POSIX locale definitions.
4307 * The new header <langinfo.h> defines an interface for accessing
4308 various locale-dependent data (using the locale chosen with `setlocale').
4310 * Ulrich Drepper has contributed a new suite of functions for operation on
4311 wide-character and multibyte-character strings, in <wchar.h>;
4312 and classification and case conversion of wide characters, in <wctype.h>.
4313 These new functions are conforming to the ISO C, Amendement 1 specification.
4315 * There is now a second implementation of the standard I/O library available.
4316 It comes from GNU libg++ as was written by Per Bothner, heavily modified
4317 by Hongjiu Lu and made thread safe by Ulrich Drepper.
4319 * You can now use positional parameter specifications in format strings
4320 for the `printf' and `scanf' families of functions. For example,
4321 `printf ("Number %2$d, Mr %1$s\n", "Jones", 6);'' prints
4322 ``Number 6, Mr Jones''. This is mainly useful when providing different
4323 format strings for different languages, whose grammars may dictate
4324 different orderings of the values being printed. To support this
4325 feature, the interface for `register_printf_handler' has changed; see
4326 the header file <printf.h> for details.
4328 * The `printf' and `scanf' families of functions now understand a new
4329 formatting flag for numeric conversions: the ' flag (e.g. %'d or %'f) says
4330 to group numbers as indicated by the locale; for `scanf' and friends, this
4331 says to accept as valid only a number with all the proper grouping
4332 separators in the right places. In the default "C" locale, numbers are
4333 not grouped; but locales for specific countries will define the usual
4334 conventions (i.e. separate thousands with `,' in the US locale).
4336 * The pgrp functions have been regularized, slightly incompatibly but much
4337 less confusingly. The core functions are now `getpgid' and `setpgid',
4338 which take arguments for the PID to operate on; the POSIX.1 `getpgrp' (no
4339 argument) and BSD `setpgrp' (identical to `setpgid') functions are
4340 provided for compatibility. There is no longer an incompatible `getpgrp'
4341 with an argument declared under _BSD_SOURCE; no BSD code uses it.
4343 * The new header file <fts.h> and suite of functions simplify programs that
4344 operate on directory trees. This code comes from 4.4 BSD.
4346 * The resolver code has been updated from the BIND 4.9.5-P1 release.
4347 Parts of the code were heavily modified by Ulrich Drepper to fit in the
4348 NSS scheme used in glibc.
4350 * There is a new malloc debugging hook `__memalign_hook'.
4352 * There are new typedefs `ushort' for `unsigned short int' and `uint' for
4353 `unsigned int' in <sys/types.h>. These are for compatibility only and
4354 their use is discouraged.
4356 * The `-lmcheck' library to enable standard malloc debugging hooks is now
4357 done differently, so that it works even without GNU ld.
4359 * New function `euidaccess' checks allowed access to a file like `access',
4360 but using the effective IDs instead of the real IDs.
4362 * The time zone data files have been updated for the latest and greatest
4363 local time conventions of the countries of the world.
4365 * The new function `dirfd' extracts the file descriptor used by a DIR stream;
4368 * The new functions `ecvt', `fcvt', and `gcvt' provide an obsolete interface
4369 for formatting floating-point numbers. They are provided only for
4370 compatibility; new programs should use `sprintf' instead. There are
4371 also equivalent function for the `long double' floating-point type and
4372 all functions also exist in a reentrant form.
4374 * The new auxiliary library `-lutil' from 4.4 BSD contains various
4375 functions for maintaining the login-record files (primarily of use to
4376 system programs such as `login'), and convenient functions for
4377 allocating and initializing a pseudo-terminal (pty) device.
4379 * Ulrich Drepper has contributed new support for System V style
4380 shared memory and IPC on systems that support it.
4382 * Ulrich Drepper has contributed several miscellaneous new functions found
4383 in System V: The `hsearch' family of functions provide an effective
4384 implementation of hash tables; `a64l' and `l64a' provide a very simple
4385 binary to ASCII mapping; `drand48' and friends provide a 48-bit random
4388 * Ulrich Drepper has contributed new reentrant counterparts for the
4389 `random' and `hsearch' families of functions; `random_r', `hsearch_r', etc.
4391 * Ulrich Drepper has contributed new, highly-optimized versions of several
4392 string functions for the i486/Pentium family of processors.
4394 * Ulrich Drepper has updated the Linux-specific code, based largely
4395 on work done in Hongjiu Lu's version of GNU libc for Linux.
4396 The GNU library now supports Linux versions 2.0.10 and later,
4397 using the ELF object file format (i[3456]86-*-linux).
4399 * Andreas Schwab has ported the C library to Linux/m68k (m68k-*-linux).
4401 * David Mosberger-Tang and Richard Henderson have ported the C library
4402 to Linux/Alpha (alpha-*-linux). Richard Henderson contributed the
4403 dynamic linking support for ELF/Alpha.
4405 * Richard Henderson contributed several Alpha optimized assembler function
4406 for arithmetic and string handling.
4408 * Ulrich Drepper has contributed a new set of message catalog functions to
4409 support multiple languages using the <libintl.h> interface, for use with
4410 his new package GNU gettext. Translation volunteers have contributed
4411 catalogs of the library's messages in Spanish, German, and Korean.
4413 * For compatibility with XPG4, Ulrich Drepper has contributed the `gencat'
4414 program and the `catgets' function for reading the catalog files it
4415 creates. (The <libintl.h> interface is preferred; we include the
4416 <nl_types.h> interface using `catgets' only for source compatibility with
4417 programs already written to use it.)
4419 * New header file <values.h> gives SVID-compatible names for <limits.h>
4422 * Various new macros, declarations, and small header files for compatibility
4425 * New function `group_member' is a convenient way to check if a process has
4426 a given effective group ID.
4428 * When using GCC 2.7 and later, the socket functions are now declared in a
4429 special way so that passing an argument of type `struct sockaddr_in *',
4430 `struct sockaddr_ns *', or `struct sockaddr_un *' instead of the generic
4431 `struct sockaddr *' type, does not generate a type-clash warning.
4433 * New function `error' declared in header file <error.h> is a convenient
4434 function for printing error messages and optionally exiting; this is the
4435 canonical function used in GNU programs. The new functions `err', `warn',
4436 and friends in header file <err.h> are the canonical 4.4 BSD interface for
4437 doing the same thing.
4439 * The <glob.h> interface has several new flags from 4.4 BSD that extend the
4440 POSIX.2 `glob' function to do ~ and {...} expansion.
4442 * New function `unsetenv' complements `setenv' for compatibility with 4.4 BSD.
4443 `clearenv' which is used in POSIX.9 is also available.
4445 * New function `getsid' returns session ID number on systems that support it.
4447 * We have incorporated the 4.4 BSD `db' library (version 1.85). New header
4448 files <db.h> and <mpool.h> provide a rich set of functions for several
4449 types of simple databases stored in memory and in files, and <ndbm.h> is
4450 an old `ndbm'-compatible interface using the `db' functions. Link with
4451 `-ldb' to get these functions.
4453 * New macro `strdupa' copies a string like `strdup', but uses local stack
4454 space from `alloca' instead of dynamic heap space from `malloc'.
4456 * New function `strnlen' is like `strlen' but searches only a given maximum
4457 number of characters for the null terminator. `stpncpy', `strndup' and
4458 `strndupa' are similar variants for the `stpcpy', `strdup' and `strdupa'
4461 * New function `statfs' in header <sys/statfs.h>.
4463 * The new <argz.h> and <envz.h> interfaces contributed by Miles Bader
4464 provide convenient functions for operating on blocks of null-terminated
4467 * A new suite of functions in <utmp.h> handle all the details of reading
4468 and writing the utmp file.
4470 * An implementation of the NIS/YP(tm) based NSS service was contributed by
4473 * Paul Eggert and Ulrich Drepper modified the `strftime' function to be
4474 completely POSIX compliant and also implemented the extended functionality
4475 to handle alternate digit representation and alternate era date formats.
4477 * Ulrich Drepper provided an implementation of the `strptime' function
4478 defined in XPG4.2 which transforms a string into a `struct tm' value.
4480 * Paul Eggert provided the tzselect shell script as part of the timezone
4481 code. The shell script makes it easy to select the correct timezone
4484 * The implementation of the malloc family of functions is completely replaced
4485 by a new implementation by Doug Lea with many improvements by Wolfram Gloger.
4486 The implementation uses the mmap function (if available) and it is
4487 optimized for the use in multi threaded programs.
4489 * Ulrich Drepper contributed a MD5 "encryption" for the crypt family of
4490 functions. This new functionality is usable by specifying a special
4491 salt string and it is compatible with implementation on *BSD systems.
4493 * Lots of functions from the XPG4.2 standard were added by Ulrich Drepper:
4494 `getsubopt' to handle second level command line options, `bsd_signal'
4495 to access BSD style `signal' functionality, the obsolete `regexp' style
4498 * the `lchown' function is available on system which support this
4501 * The implementation of the shadow password handling function was contributed
4504 * David Mosberger-Tang changed the SunRPC implementation to be 64bit safe.
4506 * POSIX.1g support was added. The <sys/select.h> header is available,
4507 `isfdtype' and `pselect' are implemented. Craig Metz contributed an
4508 implementation of `getaddrinfo'.
4512 * For cross-compilation you should now set `BUILD_CC' instead of `HOST_CC'.
4514 * New header file <fstab.h> and new functions `getfsspec', `getfsent' and
4515 friends, for parsing /etc/fstab. This code comes from 4.4 BSD.
4517 * The new function `daemon' from 4.4 BSD is useful for server programs that
4518 want to put themselves in the background.
4520 * Joel Sherrill has contributed support for several standalone boards that
4521 run without an operating system.
4523 * `printf', `scanf' and friends now accept a `q' type modifier for long
4524 long int as well as `ll'. Formats using these might be `%qu' or `%lld'.
4526 * All of the code taken from BSD (notably most of the math and networking
4527 routines) has been updated from the BSD 4.4-Lite release.
4529 * The resolver code has been updated from the BIND-4.9.3-BETA9 release.
4531 * The new functions `getdomainname' and `setdomainname' fetch or change the
4532 YP/NIS domain name. These are system calls which exist on systems which
4535 * The time zone data files have been updated for the latest international
4538 * The SunRPC programs `portmap' and `rpcinfo' are now installed in
4539 $(sbindir) (usually /usr/local/sbin) instead of $(bindir).
4543 * The C library now includes support for Sun RPC, from Sun's free
4544 RPCSRC-4.0 distribution. The `portmap', `rpcinfo', and `rpcgen' programs
4545 are included. (There is still no support for YP.)
4547 * Tom Quinn has contributed a port of the C library to SGI machines running
4548 Irix 4 (mips-sgi-irix4).
4550 * The new `lockf' function is a simplified interface to the locking
4551 facilities of `fcntl', included for compatibility.
4553 * New time functions `timegm', `timelocal', and `dysize' for compatibility.
4555 * New header file <sys/timeb.h> and new function `ftime' for compatibility.
4557 * New header files <poll.h> and <sys/poll.h> and new function `poll' for
4560 * The error message printed by `assert' for a failed assertion now includes
4561 the name of the program (if using GNU ld) and the name of the calling
4562 function (with versions of GCC that support this).
4564 * The `psignal' function is now declared in <signal.h>, not <stdio.h>.
4566 * The library now includes the <sys/mman.h> header file and memory
4567 management functions `mmap', `munmap', `mprotect', `msync', and
4568 `madvise', on systems that support those facilities.
4570 * The interface for `mcheck' has changed slightly: the function called to
4571 abort the program when an allocation inconsistency is detected now takes
4572 an argument that indicates the type of failure. The new function
4573 `mprobe' lets you request a consistency check for a particular block at
4574 any time (checks are normally done only when you call `free' or `realloc'
4577 * It is now possible to easily cross-compile the C library, building on one
4578 system a library to run on another machine and/or operating system. All
4579 you need to do is set the variable `HOST_CC' in `configparms' to the
4580 native compiler for programs to run on the machine you are building on (a
4581 few generator programs are used on Unix systems); set `CC' to the
4584 * The new function `fexecve' (only implemented on the GNU system) executes
4585 a program file given a file descriptor already open on the file.
4589 * Brendan Kehoe has contributed most of a port to the DEC Alpha
4590 running OSF/1 (alpha-dec-osf1). He says it is 75% complete.
4592 * You can set the variable `libprefix' in `configparms' to specify a prefix
4593 to be prepended to installed library files; this makes it easy to install
4594 the GNU C library to be linked as `-lgnuc' or whatever.
4596 * The new `stpncpy' is a cross between `stpcpy' and `strncpy': It
4597 copies a limited number of characters from a string, and returns the
4598 address of the last character written.
4600 * You no longer need to check for whether the installed `stddef.h' is
4601 compatible with the GNU C library. configure now checks for you.
4603 * You can now define a per-stream `fileno' function to convert the
4604 stream's cookie into an integral file descriptor.
4606 * ``malloc (0)'' no longer returns a null pointer. Instead, it
4607 allocates zero bytes of storage, and returns a unique pointer which
4608 you can pass to `realloc' or `free'. The behavior is undefined if
4609 you dereference this pointer.
4611 * The C library now runs on Sony NEWS m68k machines running either
4612 NewsOS 3 or NewsOS 4.
4614 * The new `syscall' function is a system-dependent primitive function
4615 for invoking system calls. It has the canonical behavior on Unix
4616 systems, including unreliable return values for some calls (such as
4617 `pipe', `fork' and `getppid').
4619 * The error code `EWOULDBLOCK' is now obsolete; it is always defined
4620 to `EAGAIN', which is the preferred name. On systems whose kernels
4621 use two distinct codes, the C library now translates EWOULDBLOCK to
4622 EAGAIN in every system call function.
4626 * The GNU C Library Reference Manual is now distributed with the library.
4627 `make dvi' will produce a DVI file of the printed manual.
4628 `make info' will produce Info files that you can read on line using C-h i
4629 in Emacs or the `info' program.
4630 Please send comments on the manual to bug-glibc-manual@gnu.org.
4632 * The library now supports SVR4 on i386s (i386-unknown-sysv4).
4634 * Brendan Kehoe has contributed a port to Sun SPARCs running Solaris 2.
4636 * Jason Merrill has contributed a port to the Sequent Symmetry running
4637 Dynix version 3 (i386-sequent-dynix).
4639 * The library has been ported to i386s running SCO 3.2.4 (also known as SCO
4640 ODT 2.0; i386-unknown-sco3.2.4) or SCO 3.2 (i386-unknown-sco3.2).
4642 * New function `memory_warnings' lets you arrange to get warnings when
4643 malloc is running out of memory to allocate, like Emacs gives you.
4645 * The C library now contains the relocating allocator used in Emacs 19 for
4646 its editing buffers. This allocator (ralloc) minimizes allocation
4647 overhead and fragmentation by moving allocated regions around whenever it
4648 needs to. You always refer to a ralloc'd region with a "handle" (a
4649 pointer to a pointer--an object of type `void **').
4651 * There is a new `printf' format: `%m' gives you the string corresponding
4652 to the error code in `errno'.
4654 * In `scanf' formats, you can now use `%as' or `%a[' to do the normal `%s'
4655 or `%[' conversion, but instead of filling in a fixed-sized buffer you
4656 pass, the `a' modifier says to fill in a `char **' you pass with a
4659 * The `fnmatch' function supports the new flag bits `FNM_LEADING_DIR' and
4660 `FNM_CASEFOLD'. `FNM_LEADING_DIR' lets a pattern like `foo*' match a
4661 name like `foo/bar'. `FNM_CASEFOLD' says to ignore case in matching.
4663 * `mkstemp' is a traditional Unix function to atomically create and open a
4664 uniquely-named temporary file.
4668 * The standard location for the file that says what the local timezone is
4669 has changed again. It is now `/usr/local/etc/localtime' (or more
4670 precisely, `${prefix}/etc/localtime') rather than `/etc/localtime'.
4672 * The distribution no longer contains any files with names longer than 14
4675 * `struct ttyent' has two new flag bits: TTY_TRUSTED and TTY_CONSOLE.
4676 These are set by the new `trusted' and `console' keywords in `/etc/ttys'.
4678 * New functions `ttyslot' and `syslog' from 4.4 BSD.
4682 * The configuration process has changed quite a bit. The `configure'
4683 script is now used just like the configuration scripts for other GNU
4684 packages. The `sysdeps' directory hierarchy is much rearranged.
4685 The file `INSTALL' explains the new scheme in detail.
4687 * The header files no longer need to be processed into ANSI C and
4688 traditional C versions. There is just one set of files to install, and
4689 it will work with ANSI or old C compilers (including `gcc -traditional').
4691 * Brendan Kehoe and Ian Lance Taylor have ported the library to the
4692 MIPS DECStation running Ultrix 4.
4694 * The Sun 4 startup code (crt0) can now properly load SunOS 4 shared libraries.
4695 Tom Quinn contributed the initial code. The GNU C library can NOT yet be
4696 made itself into a shared library.
4698 * Yet further improved support for the i386, running 4.3 BSD-like systems
4699 (such as Mach 3 with the Unix single-server), or System V.
4701 * New function `strncasecmp' to do case-insensitive string comparison
4702 with limited length.
4704 * New function `strsep' is a reentrant alternative to `strtok'.
4706 * New functions `scandir' and `alphasort' for searching directories.
4708 * New function `setenv' is a better interface to `putenv'.
4710 * Ian Lance Taylor has contributed an implementation of the SVID `ftw'
4711 function for traversing a directory tree.
4713 * The GNU obstack package is now also part of the C library.
4714 The new function `open_obstack_stream' creates a stdio stream that
4715 writes onto an obstack; `obstack_printf' and `obstack_vprintf' do
4716 formatted output directly to an obstack.
4718 * Miscellaneous new functions: reboot, nice, sigaltstack (4.4 BSD only),
4719 cfmakeraw, getusershell, getpass, swab, getttyent, seteuid, setegid.
4721 * `FNM_FILE_NAME' is another name for `FNM_PATHNAME', used with `fnmatch'.
4723 * The new functions `strfry' and `memfrob' do mysterious and wonderful
4724 things to your strings.
4726 * There are some new test programs: test-fseek, testmb, and testrand.
4728 * Some work has been done to begin porting the library to 4.4 BSD and Linux.
4729 These ports are not finished, but are a good starting place for really
4730 supporting those systems.
4732 * `/etc/localtime' is now the standard location for the file that says what
4733 the local timezone is, rather than `/usr/local/lib/zoneinfo/localtime'.
4734 This follows the general principle that `/etc' is the place for all local
4735 configuration files.
4737 * The C library header files now use `extern "C"' when used by the C++
4738 compiler, so the C library should now work with C++ code.
4740 * The header file <bstring.h> is gone. <string.h> now declares bcopy,
4741 bcmp, bzero, and ffs. (Update: nowadays these functions are declared
4744 * Mike Haertel (of GNU e?grep and malloc fame) has written a new sorting
4745 function which uses the `merge sort' algorithm, and is said to be
4746 significantly faster than the old GNU `qsort' function. Merge sort is
4747 now the standard `qsort' function. The new algorithm can require a lot
4748 of temporary storage; so, the old sorting function is called when the
4749 required storage is not available.
4751 * The C library now includes Michael Glad's Ultra Fast Crypt, which
4752 provides the Unix `crypt' function, plus some other entry points.
4754 * The code and header files taken from 4.4 BSD have been updated with the
4755 latest files released from Berkeley.
4757 ----------------------------------------------------------------------
4758 Copyright information:
4760 Copyright (C) 1992-2018 Free Software Foundation, Inc.
4762 Permission is granted to anyone to make or distribute verbatim copies
4763 of this document as received, in any medium, provided that the
4764 copyright notice and this permission notice are preserved,
4765 thus giving the recipient permission to redistribute in turn.
4767 Permission is granted to distribute modified versions
4768 of this document, or of portions of it,
4769 under the above conditions, provided also that they
4770 carry prominent notices stating who last changed them.
4773 version-control: never